cis-Urocanic Acid Enhances Prostaglandin E2

ORIGINAL ARTICLE

cis-Urocanic Acid Enhances Prostaglandin E2 Release and Apoptotic Cell Death via Reactive Oxygen Species in Human Keratinocytes Kazuyo Kaneko1, Susan L. Walker1, Joey Lai-Cheong1, Mary S. Matsui2, Mary Norval3 and Antony R. Young1 Urocanic acid (UCA) is a major UVR-absorbing skin molecule that undergoes trans to cis photoisomerization in the epidermis following UVR exposure. Murine studies have established that cis-UCA is an important mediator of UVR-induced immune suppression, but little is known about its signaling pathway. We have previously demonstrated that treatment of normal human epidermal keratinocytes with cis-UCA resulted in increased synthesis of prostaglandin E2 (PGE2) and cell death. Here, using immortalized human keratinocytes, we report that cis-UCA but not trans-UCA generates reactive oxygen species (ROS) in a dose-dependent manner and that the natural antioxidant a-tocopherol can reduce this ROS generation, subsequent PGE2 release, and apoptotic cell death. Western blot analysis revealed that cis-UCA leads to a transient phosphorylation of EGFR as well as downstream mitogen-activated protein kinases (MAPKs), extracellular signal-regulated kinase (ERK) and p38. Pharmacological inhibition of their activity attenuated PGE2 release induced by cis-UCA. After transient activation, cis-UCA downregulated EGFR protein expression that corresponded to activation of caspase-3. In addition, pretreatment with a-tocopherol inhibited EGFR downregulation and caspase-3 activation induced by cis-UCA. These results suggest that cis-UCA exerts its effects on human keratinocytes via intracellular ROS generation that modulates EGFR signaling and subsequently induces PGE2 synthesis and apoptotic cell death. Journal of Investigative Dermatology (2011) 131, 1262–1271; doi:10.1038/jid.2011.37; published online 17 March 2011

INTRODUCTION Exposure to solar UVR suppresses cutaneous cell-mediated immunity. UVR-induced immune suppression has been implicated in diverse pathologies including skin cancers, infectious diseases, failure of vaccination, and some autoimmune diseases as well as phototherapy (Norval, 2006). The pathway leading to immunosuppression following UVR is complex but is initiated by chromophores located in the epidermis. Evidence from mouse models indicate that DNA and urocanic acid (UCA) are critical chromophores (De Fabo and Noonan, 1983; Kripke et al., 1992). Upon exposure to UVR, UCA undergoes a trans to cis isomerization until equilibrium is reached with B60–70% cis-UCA. Topical or systemic application of cis-UCA mimics many aspects of UVR-induced immune suppression in mice 1

St John’s Institute of Dermatology, King’s College London School of Medicine, London, UK; 2The Estee Lauder Companies, Melville, New York, USA and 3Biomedical Sciences, University of Edinburgh Medical School, Edinburgh, UK Correspondence: Antony R. Young, King’s College London, St John’s Institute of Dermatology, Floor 9, Tower Wing, Guy’s Hospital, London SE1 9RT, UK. E-mail: [email protected] Abbreviations: AAPH, 2,20 -azobis (2-amidinopropane) di-hydrochloride; COX-2, cyclooxygenase-2; ERK, extracellular signal-regulated kinase; JNK, c-Jun N-terminal kinase; MAPK, mitogen-activated protein kinase; NADPH, nicotinamide adenine dinucleotide phosphate; PBS, phosphate-buffered saline; PGE2, prostaglandin E2; ROS, reactive oxygen species; UCA, urocanic acid Received 28 July 2010; revised 4 December 2010; accepted 12 January 2011; published online 17 March 2011

1262 Journal of Investigative Dermatology (2011), Volume 131

(Gibbs et al., 2008). Furthermore, treatment of mice with a cisUCA mAb restores many of the downregulatory changes in local and systemic immunity that follow UVR (El-Ghorr and Norval, 1995; Kondo et al., 1995; Moodycliffe et al., 1996). However, the pathways involved in cis-UCA-induced immune suppression remain poorly understood. To date, the target cell for cis-UCA-mediated effects has not been identified, but as the photoisomerization to cis-UCA takes place in the epidermis, keratinocytes may be a potential target for cis-UCA. Our previous microarray analysis using primary cultures of human keratinocytes demonstrated that cis-UCA but not trans-UCA upregulates the expression of UVR-inducible genes associated with oxidative stress, cell growth arrest, apoptosis, and immunomodulatory mediators. Among them, cyclooxygenase-2 (COX-2), the rate-limiting enzyme in prostanoid biosynthesis, was most dramatically upregulated by cis-UCA, resulting in an enhanced secretion of prostaglandin E2 (PGE2) (Kaneko et al., 2008). Jaksic et al. (1995) have also shown a synergistic effect of cis-UCA and histamine on increased production of PGE2 by human keratinocytes. Release of immunomodulatory mediators by keratinocytes promotes immune suppression locally and systemically following UVR exposure. PGE2 is considered a particularly important mediator as it can initiate production of the immunosuppressive cytokine IL-4, followed by IL-10 (Shreedhar et al., 1998). In addition, the recent demonstration that mice deficient in proapoptotic BH3-interacting death domain protein showed resistance to UVR-induced suppression & 2011 The Society for Investigative Dermatology

K Kaneko et al. cis-UCA Generates Reactive Oxygen Species

of local and systemic contact hypersensitivity responses as well as reduced level of apoptosis in epidermal keratinocytes and Langerhans cells following UVR exposure implicates a link between induction of apoptosis and UVR-induced immune suppression (Pradhan et al., 2008). It is well established that UVR generates reactive oxygen species (ROS) within cells and the ROS activates EGFR in a ligand-independent manner. This further activates downstream mitogen-activated protein kinase (MAPK) signaling that leads to COX-2 expression and subsequent PGE2 production (Rundhaug and Fischer, 2008). Intracellular ROS generation is also intimately associated with apoptosis and the activation of caspase-3 has been suggested to be a downstream event (Mates et al., 2008). As cis-UCA was demonstrated to upregulate many oxidative stress-related genes in primary human keratinocytes (Kaneko et al., 2008), in this study we examined the involvement of intracellular ROS in cis-UCA-induced PGE2 synthesis and apoptotic cell death using immortalized human keratinocytes. RESULTS cis-UCA enhances PGE2 synthesis via a COX-2-dependent pathway

trans-UCA 100 μg ml –1

cis-UCA induces cell growth arrest and apoptotic cell death

Treatment of primary human keratinocytes with cis-UCA resulted in the release of mono- and oligonucleosomes into the cytoplasm, and distinctive morphological changes, such as spurring and flattening (Figure 2a–c). Similar morphological changes were observed in nTERT keratinocytes (Figure 2c). cis-UCA-induced cell growth arrest and/or death was further evaluated by the 3-(4,5-dimethylthiazol-2-yl)-2,5dimethyl tetrazolium bromide (MTT) cell proliferation and 4

cis-UCA 100 μg ml –1 0

COX-2

COX-2

Actin

Actin

1

2

6 10 24 Hours

COX-2 (fold change)

PBS

cis-UCA 100 μg ml –1

That cis-UCA induces a dose-dependent increase in COX-2 mRNA and PGE2 release has been demonstrated in primary

human keratinocytes (Kaneko et al., 2008). To further test if cis-UCA induces PGE2 production in a COX-2-dependent manner, we used nTERT keratinocytes derived from human primary keratinocytes immortalized with human telomerase reverse transcriptase (Dickson et al., 2000). nTERT keratinocytes showed a similar dose-dependent increase in PGE2 release 24 hours after cis-UCA treatment (Figure 1d). cis-UCA but not trans-UCA also significantly upregulated the expression of COX-2 protein after 2 and 6 hours (Figure 1a–c). An increase in PGE2 release induced by cis-UCA was observed after 6 hours (Figure 1e). In addition, the COX-inhibitor indomethacin abrogated the increase in PGE2 release induced by cis-UCA (Figure 1f), suggesting that cis-UCA-induced PGE2 synthesis is mediated by a COX-2-dependent pathway.

3

** 2

**

1 cis-UCA 100 μg ml –1

0 6

0

12

18

24

Time (hours)

**

50

**

3

PGE2 (pg ml –1)

100

150

4 PGE2 (fold change)

PGE2 (pg ml –1)

150

2

* * 1

* 100

50


0

0 0

6 12 18 Time (hours)

24

Vehicle Indo PBS cis-UCA 100 μg ml –1

ci

sU C A PB ci 10 sS U μg C ci s- A m 5 U tra C 0 μ l –1 ns A g m -U 10 0 l –1 C μg A 10 m –1 l 0 μg m l –1

0

Figure 1. cis-Urocanic acid (UCA) increases cyclooxygenase-2 (COX-2) protein expression and prostaglandin E2 (PGE2) release, and the COX inhibitor indomethacin abrogates PGE2 synthesis induced by cis-UCA in nTERT keratinocytes. (a) COX-2 protein expression 6 hours after treatment, (b) kinetics of COX-2 protein expression, and (c) optical density values for COX-2 relative to actin in response to cis-UCA. (d–f) PGE2 release into the culture medium; (d) 24 hours after treatment, (e) kinetics in response to cis-UCA, and (f) pretreatment with 0.5 mM indomethacin for 1 hour before treatment with phosphate-buffered saline (PBS) or cis-UCA for 24 hours. Results are expressed as mean±SE of three independent experiments. *Po0.05, **Po0.01 versus PBS control (d, f) or before treatment (c, e). The blots are representative of three independent experiments.

www.jidonline.org 1263


viability assay in nTERT keratinocytes. As shown in Figure 2d, cis-UCA significantly decreased MTT cell proliferation and viability of keratinocytes in a dose-dependent manner. Also, 100 mg ml–1 trans-UCA slightly decreased MTT cell proliferation and viability. To determine if this was due to apoptosis, cells were stained with annexin V-FITC and propidium iodide following UCA treatment and analyzed by flow cytometry. As shown in Figure 2e, 50 and 100 mg ml–1 cis-UCA clearly induced early apoptosis after 24 hours, with 25 and 37% of early apoptotic (annexin positive) cells in the 50 and 100 mg ml–1 cis-UCA treatment groups, respectively, and only 6% in the phosphate-buffered saline (PBS) control. This cell population shifted toward the late stage of apoptosis (annexin and propidium iodide positive) 48 hours after treatment with 100 mg ml–1 cis-UCA. In contrast, 10 mg ml–1 cis-UCA and 100 mg ml–1 trans-UCA did not increase either early or late apoptotic cells at 24 hours. As shown in Figure 2f, the electron microscopy analysis also showed that 100 mg ml–1 cis-UCA-treated nTERT cells exhibited hallmark changes of apoptosis such as irregular plasma membrane and nuclear envelope shape, condensation of nuclear chromatin, blebbing, and loss of nucleoli at 24 hours, indicating that the cells had left the latent phase of apoptosis, but had not reached the execution phase with cytoplasmic spillage. These changes were not observed in PBS or trans-UCA-treated cells. Taken together, these observations suggest that keratinocytes exposed to cis-UCA undergo growth arrest and apoptotic cell death. cis-UCA induces the generation of intracellular ROS that mediate PGE2 synthesis and apoptotic cell death

We have previously shown that cis-UCA significantly increased the release of 8-isoprostane, a biomarker for photo-oxidative stress, in primary human keratinocytes (Kaneko et al., 2008). To determine whether cis-UCA induces generation of ROS in nTERT keratinocytes, intracellular ROS were visualized with the carboxydichlorofluorescein, which is oxidized by ROS within cells and emits green fluorescence. As shown in Figure 3a, cis-UCA induced an increase in intracellular ROS in a dose-dependent manner, whereas only a background level of fluorescence was observed after treatment with trans-UCA. NADPH (nicotinamide adenine dinucleotide phosphate) oxidase is a potential cellular source for the extended production and enhanced level of ROS after exposure of skin to UVR. To assess whether this enzyme is involved in the production of ROS induced by cis-UCA, cells were pretreated with the NADPH oxidase inhibitor, diphenyleneiodonium chloride, and analyzed for ROS generation. As shown in Figure 3b, diphenyleneiodonium chloride reduced ROS generation induced by cis-UCA, suggesting that NADPH

oxidase is at least in part involved in cis-UCA-induced ROS generation. The natural antioxidant a-tocopherol effectively scavenges UVR-induced ROS (Jin et al., 2007); therefore, further studies assessed its ability to inhibit the cis-UCA-induced increase in PGE2 release and apoptotic cell death. As shown in Figure 3c, a-tocopherol reduced intracellular ROS generation in a dosedependent manner. Pretreatment with 10 mM a-tocopherol resulted in reduced production of PGE2 induced by cis-UCA (Figure 3d). To inhibit apoptotic cell death, a higher concentration of a-tocopherol was required, and 100 mM atocopherol significantly inhibited cis-UCA-induced MTT reduction and early apoptosis (Figure 3e and f). These results suggest that cis-UCA-induced PGE2 synthesis and apoptotic cell death can be attributed to enhanced production of intracellular ROS in human keratinocytes. cis-UCA increases PGE2 release mediated through activation of EGFR and downstream MAPKs

We further examined the involvement of the EGFR/MAPK signaling pathway in cis-UCA-induced PGE2 release. As shown in Figure 4a and b, cis-UCA but not trans-UCA induced a transient increase in phosphorylation of EGFR 2 hours after treatment. The expression of phosphorylated EGFR then returned to the basal level at 6 hours and was not detected at 24 hours. cis-UCA also decreased the constitutive expression of EGFR after 24 hours. In addition, as shown in Figure 4d and e, cis-UCA but not trans-UCA induced a transient increase in phosphorylation of the downstream MAPKs, extracellular signal-regulated kinase (ERK) and p38. The phosphorylation of ERK peaked 2 hours after treatment with cis-UCA, returned to the basal level after 6 hours, and then diminished, as did the EGFR phosphorylation. The kinetics of the phosphorylation of p38 were slower and prolonged compared with ERK, with peaking at 6 hours. In contrast, the phosphorylation of c-Jun N-terminal kinase (JNK) by cis-UCA was not observed over this time period. Consistent with these observations, pretreatment with the EGFR kinase inhibitor AG1478, ERK inhibitor PD98059, and p38 inhibitor SB202190 inhibited the increase in PGE2 release induced by cis-UCA. The JNK inhibitor SP600125 did not have a similar inhibitory effect (Figure 4c and f). These results indicate that cis-UCA-induced PGE2 synthesis requires activation of EGFR and downstream ERK and p38 MAPK signaling. cis-UCA-induced apoptotic cell death involves a caspase-3mediated pathway

To clarify the involvement of caspase-3 in cis-UCA-induced apoptotic cell death, keratinocytes were treated with cis-UCA and the expression of caspase-3 was analyzed by immuno-

Figure 2. cis-Urocanic acid (UCA) increases apoptotic cells in primary and nTERT keratinocytes. (a) Mono- and oligonucleosome release 24 hours after treatment, (b, c) light microscope images 24 hours after treatment (scale bars ¼ 100 mm in b and 50 mm in c), (d) MTT cell proliferation/viability 24 hours after treatment, and (e) flow cytometric analysis of apoptotic cells. The values indicate the percentage of living (lower left quadrant), early apoptotic (lower right quadrant), late apoptotic (upper right quadrant), and dead (upper left quadrant) cells. Similar results were obtained in at least three independent experiments. (f) Electron microscope images 24 hours after treatment (scale bars ¼ 2 or 10 mm as indicated). (a–c) Primary human keratinocytes and (c–f) nTERT keratinocytes. (a, d) Results are expressed as mean±SE of three independent experiments. *Po0.05, ***Po0.001 versus phosphate-buffered saline (PBS) control.



6

–1

*

5

Enrichment factor

cis -UCA 10 μg ml

PBS

4 3 2

cis -UCA 50 μg ml

–1

cis -UCA 100 μg ml

–1

1

ci 10 PB sU μg S C ci m s- A 5 l –1 tra UC 0 μ ns A 1 g m -U 0 C 0 μ l –1 A g 10 m l –1 0 μg m l –1

0

cis -UCA 100 μg ml –1

100

***

75

*** 50 25 0 μg

105

105

104

104

103

103

102

102

C

U

ci

s-

C

A

10 trans -UCA 100 μg ml –1, 24 hours

ci

U ci

s-

50 μm

PBS, 24 hours PI

*

***

PB

nTERT keratinocytes Primary keratinocytes

MTT (% of PBS control)

PBS

S m s- A – l U 1 tra C 50 μ ns A 1 g m -U 0 C 0 μ l –1 A g 10 m l –1 0 μg m l –1

ci

s-

U

C

A

100 μm

PBS

2 103 104 105 10 cis -UCA 10 μg ml –1, 24 hours

102 103 104 105 –1 cis -UCA 50 μg ml , 24 hours

105

105

104

104

103

103

102

10 μm

cis -UCA 100 μg ml

102 102

103

104

cis -UCA 100 μg ml

–1

105

102

105

105

104

104

3

103

2

102

10

102

103

104

105

103

104

10 μm –1

cis -UCA 100 μg ml –1

105

cis -UCA 100 μg ml –1, 48 hours

, 24 hours

10

trans -UCA 100 μg ml –1

10 μm

102

103

104

2 μm

105

Annexin V-FITC



blotting. As shown in Figure 5a and b, cis-UCA markedly increased the expression of cleaved caspase-3 at 24 hours in a dose-dependent manner, which was consistent with the downregulation of the EGFR protein expression (Figure 4b). In addition, pretreatment with 100 mM a-tocopherol clearly inhibited the cleavage of caspase-3 and the degradation of EGFR induced by cis-UCA (Figure 5c). DISCUSSION This study provides the evidence that cis-UCA can induce generation of ROS within human keratinocytes (Figure 3a). Previously, NADPH oxidase had been reported as the major source of UVR-induced ROS in human keratinocytes (Beak et al., 2004; Valencia and Kochevar, 2008). In agreement with this, our results suggest that NADPH oxidase is likely to play a role, at least in part, in cis-UCA-induced ROS generation (Figure 3b). This activation may involve signaling through the membrane receptor for cis-UCA, although the identity of such a receptor remains elusive and controversial (Walterscheid et al., 2006; Woodward et al., 2006; Kaneko et al., 2009). Recent evidence suggests that cis-UCA has the capacity to decrease the intracellular pH in a mildly acidic extracellular environment (pH o6.7) (Laihia et al., 2010). However, this is unlikely to be the underlying mechanism in this study as the pH of the medium containing cis-UCA was between 7.0 and 7.4 before and 24 hours after incubation (data not shown). The mitochondrial electron transport chain is another potential cellular source for ROS after exposure of skin to UVR (Gniadecki et al., 2000), and products of membrane lipid peroxidation induce mitochondrial ROS formation (Landar et al., 2006). As our earlier study indicated that cis-UCA induces lipid peroxidation in primary human keratinoytes (Kaneko et al., 2008), it is possible that the mitochondrial pathway also mediates cis-UCA-induced generation of ROS. In contrast to our finding, Rinaldi et al. (2006) have shown that cis-UCA inhibits the generation of extracellular superoxide while not affecting the generation of intracellular superoxide or other ROS in bovine neutrophils. Although they did not test trans-UCA, this inhibitory effect is unlikely to be specific to the cis-isomer because it has previously been shown that either trans- or cis-UCA inhibits the respiratory burst activity of human neutrophils (Kivisto et al., 1996). UCA is an imidazole compound and, as well as other imidazole derivatives including histidine and histamine, both UCA isomers are known to act as hydroxyl radical scavengers in aqueous solution (Kammeyer et al., 1999). The effect observed in their in vitro system may derive from the radical scavenging activity of exogenous UCA. Whether cis-UCA can penetrate cell membranes remains to be determined and,

if a membrane receptor mediates in its activity, cis-UCA may act as a radical initiator rather than as a scavenger. Phosphorylation of EGFR via UVR-generated ROS has been well documented (Huang et al., 1996; Peus et al., 1998). This phosphorylation provides a necessary signal to induce COX-2 expression and subsequent PGE2 production in keratinocytes (Ashida et al., 2003; Rundhaug and Fischer, 2008). Similar to UVR, cis-UCA activated the EGFR and downstream MAPK signaling pathways that are required for PGE2 synthesis in human keratinocytes (Figure 4). It has been suggested that COX-2 expression and PGE2 levels are regulated by different MAPK signaling pathways depending on the cell type and stimulus (Chen et al., 2001; Ashida et al., 2003; Cui et al., 2004; Kim and Kim, 2004). Our results showed that cis-UCA induces phosphorylation of ERK and p38 but not JNK (Figure 4e). ERK and p38 inhibitors but not a JNK inhibitor also suppressed cis-UCA-induced PGE2 release (Figure 4f). In contrast, other groups have demonstrated that phosphorylation of JNK as well as ERK and p38 is induced by UVB and 2,20 -azobis (2-amidinopropane) di-hydrochloride (AAPH), a water-soluble free radical initiator, in HaCaT keratinocytes. However, these inhibition studies showed that the induction of COX-2 expression by UVB or AAPH was not blocked by the inhibition of JNK but by the inhibition of ERK and p38 (Ashida et al., 2003; Cui et al., 2004), suggesting that UVB and AAPH induce COX-2 induction independently of the JNK signaling. This supports our finding that ERK and p38 signaling are the major pathways involved in cis-UCAinduced PGE2 synthesis in keratinocytes. The mechanism by which cis-UCA differentially activates MAPKs remains elusive, but different ROS elicit different MAPK activation patterns (Baas and Berk, 1995). As well as PGE2 synthesis, keratinocytes exposed to cisUCA underwent cell growth arrest and apoptotic cell death (Figure 2). This observation is consistent with recent reports showing that dermal injection of cis-UCA results in a high number of TUNEL-positive apoptotic cells in mouse skin (Sreevidya et al., 2010) and also that cis-UCA induces the irreversible termination of proliferation of two human bladder carcinoma and one human metastatic melanoma cell lines (Laihia et al., 2009, 2010). We further demonstrated that cisUCA induces apoptotic cell death via intracellular ROS generation (Figure 3). Accumulating evidence indicates that intracellular ROS activates signaling pathways, leading to the activation of caspase-3 and subsequent apoptosis (Ueda et al., 2002). In human keratinocytes, it has been reported that treatment with H2O2 or singlet oxygen induces cleavage of caspase-3 as well as apoptosis (Zhuang et al., 2003; Zuliani et al., 2005). We also detected dose-dependent upregulation of the cleaved active form of caspase-3 at

Figure 3. cis-Urocanic acid (UCA) generates reactive oxygen species (ROS) mediated through nicotinamide adenine dinucleotide phosphateoxidase (NADPH), and the natural antioxidant a-tocopherol reduces prostaglandin E2 (PGE2) increase and apoptotic cell death induced by cis-UCA in nTERT keratinocytes. (a–c) ROS generation at 2 hours after treatment. Cells were pretreated with 10 mM diphenyleneiodonium chloride (DPI; b) or 10–100 mM a-tocopherol (c) before cis-UCA treatment (scale bar ¼ 200 mm). (d) PGE2 release into the culture medium for 24 hours, (e) MTT cell proliferation/viability, and (f) flow cytometric analysis of apoptotic cells after 24 hours. Cells were pretreated with 10 (d) or 100 mM (e, f) a-tocopherol before cis-UCA treatment. (d, e) Results are expressed as mean±SE of three independent experiments. *Po0.05, **Po0.01. Similar results were obtained in at least three independent experiments.



TBHP 100 μM


Vehicle

DPI


PBS

PBS

200 μm

10 μg

50 μg

ml –1

100 μg

ml –1

200 μm

ml –1

cis-UCA 200 α-Tocopherol 10 μM

Vehicle

*

100 μM


PBS

PGE2 (pg ml –1)

150

PBS

100

cis-UCA 100 μg ml

50 200 μm

0

Vehicle

α-Toc α-Tocopherol

PI

Vehicle

PBS

100

**

105

105

104

104

103

103

102

102

75

50

102

25

0 cis-UCA α-Toc

– –

+ –

+ +


MTT (% of control)

–1

103

104

105

105

105

104

104

103

103

102

102

102

103

104

105

102

103

104

105

102

103

104

105

Annexin V-FITC




provides a mechanism to exacerbate cis-UCA-induced apoptotic cell death, but, to prove this, additional studies using genetic or pharmacological inhibition of EGFR and caspase-3 are necessary. Our data and that of others have clearly demonstrated the capacity of cis-UCA to induce apoptosis. The quantity of cisUCA used in this study (10–100 mg ml–1, equivalent to 70–700 mM) approximates the concentration found in human skin (2-62 nmol cm–2, equivalent to 40–1,230 mg ml–1; Laihia et al., 1998). However, apoptosis was not observed at the lowest concentration of cis-UCA tested (10 mg ml–1), although PGE2 release was increased (Figures 1 and 2), suggesting that a higher level of ROS is required to induce apoptosis than that required for phosphorylation of EGFR to produce PGE2. One study reports that there is no correlation between total UCA concentration and minimal erythema dose (De Fine Olivarius et al., 1997). This may be explained by recent evidence in histidinemic mice suggesting that the formation of cyclobutane pyrimidine dimers is more important than cis-UCA for the induction of apoptosis (Barresi et al., 2011). Further investigations are needed to understand the role of cis-UCA on apoptosis induction and its relationship with immune suppression in vivo. Keratinocytes are an important candidate for the site of action of cis-UCA. The data presented here suggest that the

200

** cis-UCA 100 μg ml –1 0

1

2

6

10 24 Hours

PGE2 ( pg ml –1)


PBS

24 hours after cis-UCA treatment (Figure 5a and b) and this activation was blocked by pretreatment with the antioxidant a-tocopherol (Figure 5c). Similar activation of caspase-3 by cis-UCA in human melanoma cells has been observed in vitro and in vivo (Laihia et al., 2010). Moreover, expression of EGFR protein was downregulated by cis-UCA after 24 hours (Figures 4b and 5b), and a-tocopherol inhibited this downregulation (Figure 5c). EGFR signaling is known to regulate cell survival, proliferation, and migration (Pastore et al., 2008). Downregulation of EGFR protein expression has been seen during apoptosis in response to various stimuli including UVR, singlet oxygen, and tumor necrosis factor-a (Bae et al., 2001; He et al., 2003, 2006; Zhuang et al., 2003). Overexpression of cleavage-deficient EGFR mutant also delayed the apoptosis progression in HaCaT keratinocytes (He et al., 2006). Using purified EGFR, Bae et al. (2001) have demonstrated that caspase-3 exhibits the ability to cleave EGFR. This was further extended by other groups showing that UVA and singlet oxygen produced by Rose Bengal photosensitization induce downregulation of EGFR protein expression and concurrent caspase-3 activation during apoptosis, and that inhibition of caspase-3 reduces EGFR degradation in human keratinocytes (He et al., 2003; Zhuang et al., 2003). Taken together, it is possible that inactivation of EGFR-mediated antiapoptotic signaling via active caspase-3

150

* 100

pEGFR

pEGFR

EGFR

EGFR

50

Actin

Actin

0

0

1

2

6

AG1478

10 24 Hours 250

pERK

*

PBS cis-UCA 100 μg ml –1

*

200 ERK

pERK

pp38

ERK

p38

pp38

pJNK

p38

JNK

PGE2 (pg ml –1)

trans-UCA 100 μg ml

Vehicle

PBS cis-UCA 100 μg ml –1


–1 –1

cis-UCA 100 μg ml

PBS

**

150 100 50

0 SB202190 PD98059 SP600125

– – –

+ – –

– + –

– – +

Figure 4. cis-Urocanic acid (UCA) activates EGFR and mitogen-activated protein kinase (MAPK), and inhibition of EGFR or MAPK blocks prostaglandin E2 (PGE2) increase induced by cis-UCA in nTERT keratinocytes. (a) Expression of phosphorylated and total EGFR after 2 hours, (b) kinetics of EGFR expression in response to cis-UCA, (d) expression of phosphorylated and total extracellular signal-regulated kinase (ERK; after 2 hours) and p38 (after 6 hours), and (e) kinetics of ERK, p38, and c-Jun N-terminal kinase (JNK) expression in response to cis-UCA. Similar results were obtained in at least three independent experiments. (c, f) PGE2 release into the culture medium for 24 hours. Cells were pretreated with 0.5 mM EGFR or MAPK inhibitors for 1 hour before cis-UCA treatment. Results are expressed as mean±SE of three independent experiments. *Po0.05, **Po0.01.



Apoptosis

cis-UCA 100 μg ml –1 0

1

2

6

10

24 Hours Cleavage

Cleaved caspase-3

cis-UCA Receptor?

Cleaved caspase-3

EGFR

EGFR


Cell membrane

ROS

EGF receptor p

NADPH oxidase?

α-Toc

PBS

PBS

Cleaved caspase-3


Vehicle





PBS

Actin

?

p

MAPK

ERK p p38

p

Caspase-3 COX-2

PGE2 Nucleus

Actin

Actin

Figure 5. cis-Urocanic acid (UCA) induces activation of caspase-3 and downregulation of EGFR in nTERT keratinocytes. (a) Kinetics of cleaved caspase-3 expression in response to cis-UCA, (b) expression of cleaved caspase-3 and EGFR 24 hours after treatment, and (c) expression of cleaved caspase-3 and EGFR 24 hours after phosphate-buffered saline (PBS) or cis-UCA treatment with or without pretreatment with 100 mM a-tocopherol (a-Toc) for 24 hours. The blots are representative of four independent experiments.

generation of intracellular ROS plays a critical role in signal transduction induced by cis-UCA as well as by UVR in human keratinocytes (Figure 6). Whether cis-UCA induces ROS generation in other cell types remains to be determined, but our findings fit well with previous mouse studies showing that topical application of antioxidants inhibits the suppression of contact hypersensitivity induced by cis-UCA (Hemelaar and Beijersbergen van Henegouwen, 1996; Steenvoorden and Beijersbergen van Henegouwen, 1999). It is well known that ROS persist following UV irradiation despite their short lifetime (Valencia and Kochevar, 2008). The cellular source of these ROS is unclear at present but it is possible that cis-UCA, which is found in the epidermis for at least 2 weeks after a single UVR exposure (Kammeyer et al., 1997), may induce ROS over the same period. Accumulating evidence suggests that molecular mechanisms underlying UVR-induced immune suppression may be linked by ROS generation (Halliday, 2010). Our data further strengthen this theory and support the idea that antioxidants may be of some benefit against UVR-induced immune suppression and photocarcinogenesis, as proposed by other studies (Afaq and Mukhtar, 2006; Wright et al., 2006; Camouse et al., 2009; Matsui et al., 2009). MATERIALS AND METHODS Cell culture Primary cultures of epidermal keratinocytes were prepared from biopsies taken from the unexposed buttock of healthy Caucasian volunteers as described previously (Kaneko et al., 2008). The

Figure 6. A model of signaling pathways involved in cis-urocanic acid (UCA)-induced prostaglandin E2 (PGE2) synthesis and apoptosis in human keratinocytes. cis-UCA induces intracellular reactive oxygen species (ROS) generation that transiently activates EGFR. Subsequent activation of the downstream extracellular signal-regulated kinase (ERK) and p38 mitogenactivated protein kinase (MAPK) signaling pathways leads to increased transcription of cyclooxygenase-2 (COX-2) that then stimulates PGE2 release. At higher concentrations, cis-UCA activates caspase-3 via intracellular ROS generation that downregulates EGFR, resulting in apoptosis.

procedure was approved by St Thomas’ Hospital Ethics Committee, adhering to the Declaration of Helsinki Principles, and the donors gave written informed consent. The primary and immortalized nTERT human keratinocytes were grown in serum-free keratinocyte medium supplemented with 50 mg ml–1 pituitary extract, 2.5 ng ml–1 epidermal growth factor, 100 units ml–1 penicillin, and 100 mg ml–1 streptomycin (Invitrogen, Paisley, UK). The cells were maintained at 37 1C in 5% CO2 and passaged at 60% confluence. The medium was changed every 2 to 3 days. For all experiments, cells were fed 24 hours before treatment and allowed to grow to 70–80% confluence.

Urocanic acid trans-UCA (99% purity) was obtained from Sigma-Aldrich Company (Dorset, UK). cis-UCA was prepared from trans-UCA as previously described and was 499% pure as determined by HPLC (Norval et al., 1989). UCA isomers were dissolved in PBS and added to keratinocyte cultures. PBS served as a baseline control.

Antibodies and reagents Rabbit anti-human ERK, phospho-ERK, JNK, phospho-JNK and phospho-p38 mAbs and rabbit anti-human COX-2, cleaved caspase-3, EGFR, phospho-EGFR, and p38 polyclonal antibodies were purchased from Cell Signaling Technology (Danvers, MA). Polyclonal rabbit anti-human actin and goat anti-rabbit secondary antibodies were obtained from Abcam (Cambridge, UK) and Invitrogen, respectively. Diphenyleneiodonium chloride, dl-a-tocopherol, indomethacin, MTT, SB202190, SP600125, and AG 1478 were purchased from Sigma-Aldrich. PD98059 was obtained from www.jidonline.org 1269


Merck Chemicals (Hull, UK). a-Tocopherol and indomethacin were dissolved in ethanol and the rest were dissolved in DMSO (SigmaAldrich).

ROS detection For detection of ROS, the Image-iT live green ROS detection kit (Invitrogen) was used according to the manufacturer’s instructions. Cells were analyzed by fluorescence microscopy using an Axiovert 40 CFL microscope (Carl Zeiss, Welwyn Garden City, UK). Tertbutyl hydroperoxide, a short-chain lipid peroxide analog, served as a positive control. Diphenyleneiodonium chloride and a-tocopherol were added 1 and 24 hours before cis-UCA treatment, respectively.

Measurement of PGE2 release Supernatants were collected by centrifugation. The concentration of PGE2 was determined by Parameter PGE2 Assay (R&D Systems Europe, Abingdon, UK) according to the manufacturer’s instructions. EGFR (AG1478) and MAPK (SB202190, PD98059, and SP600125) inhibitors were added 1 hour before treatment and a-tocopherol was added 24 hours before treatment.

MTT assay A total of 2 104 cells per well were plated in a 96-well plate and incubated with or without a-tocopherol for 24 hours at 37 1C. cis-UCA or PBS was then added for a further 24 hours, and 3 hours before the end of treatment 0.5% MTT solution in PBS was added to each well. The culture medium was removed and 0.04 M hydrochloric acid in isopropanol was added to each well to dissolve the formazan. Following 5 minutes of incubation at 37 1C, the optical density of each well was measured at 550 nm.

Electron microscopy analysis Cells were grown on glass coverslips. At 24 hours after treatment with UCA, the coverslips were fixed with 2.5% glutaraldehyde in phosphate buffer, pH 7.3, and then treated with 1% osmium tetroxide in phosphate buffer, pH 7.3. Ultrathin sections were stained with uranyl acetate and lead citrate. Micrographs were obtained using a Hitachi H7600 transmission electron microscope (Maidenhead, UK).

Apoptosis detection Apoptotic cells were detected 24 or 48 hours after UCA treatment using the TACS Annexin V-FITC (R&D Systems) according to the manufacturer’s instructions. Samples were analyzed by flow cytometry as early apoptotic cells (annexin V-FITC positive), late apoptotic cells (annexin V and propidium iodide positive), dead cells (propidium iodide positive), and viable cells (unstained). For measurement of mono- and oligonucleosome release into the cytoplasm, cell death detection ELISAplus (Roche Diagnostics GmbH, Penzberg, Germany) was used, according to the manufacturer’s instructions. The enrichment factor was calculated using the following formula: absorbance of treated sample/absorbance of PBS control.

Immunoblotting Following UCA treatment, cells were washed with PBS, harvested using a scraper and solubilized in cold lysis buffer. Then, 10 mg of cell lysate protein was separated on a NuPAGE 4–12% Bis-Tris gel 1270 Journal of Investigative Dermatology (2011), Volume 131

(Invitrogen) and subsequently electrotransferred to a polyvinylidene difluoride membrane (Invitrogen) following the manufacturer’s instructions. After blocking, the membrane was probed with primary antibody and then secondary antibody. The protein bands on the blots were detected with enhanced chemiluminescence reagents mix (GE Healthcare UK, Little Chalfont, UK) according to the manufacturer’s instructions. The densitometric analysis of each band was performed using Image J software (National Institutes of Health, Bethesda, MD).

Statistical analyses For all analyses, Student’s unpaired t-test was performed using SPSS 11.5 software (SPSS, Chicago, IL). Values of Po0.05 were considered significant. CONFLICT OF INTEREST The authors state no conflict of interest.

ACKNOWLEDGMENTS We thank Ms Katarzyna Grys for the technical assistance with FACS analysis and Mr Ken Brady for preparing sections and the technical assistance for electron microscopy. We are grateful to Professor Gareth E Jones and Dr Juliet A Ellis for their advice on electron microscopic analysis. This work was supported by the Estee Lauder Companies, Melville, NY, St John’s Institute of Dermatology, and the Department of Health via the National Institute for Health Research (NIHR) comprehensive Biomedical Research Centre award to Guy’s and St Thomas’ NHS Foundation Trust in partnership with King’s College London and King’s College Hospital NHS Foundation Trust.

REFERENCES Afaq F, Mukhtar H (2006) Botanical antioxidants in the prevention of photocarcinogenesis and photoaging. Exp Dermatol 15:678–84 Ashida M, Bito T, Budiyanto A et al. (2003) Involvement of EGF receptor activation in the induction of cyclooxygenase-2 in HaCaT keratinocytes after UVB. Exp Dermatol 12:445–52 Baas AS, Berk BC (1995) Differential activation of mitogen-activated protein kinases by H2O2 and O2- in vascular smooth muscle cells. Circ Res 77:29–36 Bae SS, Choi JH, Oh YS et al. (2001) Proteolytic cleavage of epidermal growth factor receptor by caspases. FEBS Lett 491:16–20 Barresi C, Stremnitzer C, Mlitz V et al. (2011) Increased sensitivity of histidinemic mice to UVB radiation suggests a crucial role of endogenous urocanic acid in photoprotection. J Invest Dermatol 131:188–94 Beak SM, Lee YS, Kim JA (2004) NADPH oxidase and cyclooxygenase mediate the ultraviolet B-induced generation of reactive oxygen species and activation of nuclear factor-kB in HaCaT human keratinocytes. Biochimie 86:425–9 Camouse MM, Domingo DS, Swain FR et al. (2009) Topical application of green and white tea extracts provides protection from solar-simulated ultraviolet light in human skin. Exp Dermatol 18:522–6 Chen W, Tang Q, Gonzales MS et al. (2001) Role of p38 MAP kinases and ERK in mediating ultraviolet-B induced cyclooxygenase-2 gene expression in human keratinocytes. Oncogene 20:3921–6 Cui Y, Kim DS, Park SH et al. (2004) Involvement of ERK and p38 MAP kinase in AAPH-induced COX-2 expression in HaCaT cells. Chem Phys Lipids 129:43–52 De Fabo EC, Noonan FP (1983) Mechanism of immune suppression by ultraviolet irradiation in vivo. I. Evidence for the existence of a unique photoreceptor in skin and its role in photoimmunology. J Exp Med 158:84–98 De Fine Olivarius F, Wulf HC, Therkildsen P et al. (1997) Urocanic acid isomers: relation to body site, pigmentation, stratum corneum thickness and photosensitivity. Arch Dermatol Res 289:501–5


Dickson MA, Hahn WC, Ino Y et al. (2000) Human keratinocytes that express hTERT and also bypass a p16INK4a-enforced mechanism that limits life span become immortal yet retain normal growth and differentiation characteristics. Mol Cell Biol 20:1436–47 El-Ghorr AA, Norval M (1995) A monoclonal antibody to cis-urocanic acid prevents the ultraviolet-induced changes in Langerhans cells and delayed hypersensitivity responses in mice, although not preventing dendritic cell accumulation in lymph nodes draining the site of irradiation and contact hypersensitivity responses. J Invest Dermatol 105:264–8 Gibbs NK, Tye J, Norval M (2008) Recent advances in urocanic acid photochemistry, photobiology and photoimmunology. Photochem Photobiol Sci 7:655–67 Gniadecki R, Thorn T, Vicanova J et al. (2000) Role of mitochondria in ultraviolet-induced oxidative stress. J Cell Biochem 80:216–22 Halliday GM (2010) Common links among the pathways leading to UV-induced immunosuppression. J Invest Dermatol 130:1209–12 He YY, Huang JL, Chignell CF (2006) Cleavage of epidermal growth factor receptor by caspase during apoptosis is independent of its internalization. Oncogene 25:1521–31 He YY, Huang JL, Gentry JB et al. (2003) Epidermal growth factor receptor down-regulation induced by UVA in human keratinocytes does not require the receptor kinase activity. J Biol Chem 278:42457–65 Hemelaar PJ, Beijersbergen van Henegouwen GM (1996) The protective effect of N-acetylcysteine on UVB-induced immunosuppression by inhibition of the action of cis-urocanic acid. Photochem Photobiol 63:322–7 Huang RP, Wu JX, Fan Y et al. (1996) UV activates growth factor receptors via reactive oxygen intermediates. J Cell Biol 133:211–20 Jaksic A, Finlay-Jones JJ, Watson CJ et al. (1995) cis-Urocanic acid synergizes with histamine for increased PGE2 production by human keratinocytes: link to indomethacin-inhibitable UVB-induced immunosuppression. Photochem Photobiol 61:303–9 Jin GH, Liu Y, Jin SZ et al. (2007) UVB induced oxidative stress in human keratinocytes and protective effect of antioxidant agents. Radiat Environ Biophys 46:61–8 Kammeyer A, Eggelte TA, Bos JD et al. (1999) Urocanic acid isomers are good hydroxyl radical scavengers: a comparative study with structural analogues and with uric acid. Biochim Biophys Acta 1428:117–20 Kammeyer A, Pavel S, Asghar SS et al. (1997) Prolonged increase of cisurocanic acid levels in human skin and urine after single total-body ultraviolet exposures. Photochem Photobiol 65:593–8 Kaneko K, Smetana-Just U, Matsui M et al. (2008) cis-Urocanic acid initiates gene transcription in primary human keratinocytes. J Immunol 181:217–24 Kaneko K, Travers JB, Matsui MS et al. (2009) cis-Urocanic acid stimulates primary human keratinocytes independently of serotonin or plateletactivating factor receptors. J Invest Dermatol 129:2567–73 Kim HJ, Kim TY (2004) IGF-II-mediated COX-2 gene expression in human keratinocytes through extracellular signal-regulated kinase pathway. J Invest Dermatol 123:547–55 Kivisto K, Punnonen K, Toppari J et al. (1996) Urocanic acid suppresses the activation of human neutrophils in vitro. Inflammation 20:451–9 Kondo S, Sauder DN, McKenzie RC et al. (1995) The role of cis-urocanic acid in UVB-induced suppression of contact hypersensitivity. Immunol Lett 48:181–6

Laihia JK, Attila M, Neuvonen K et al. (1998) Urocanic acid binds to GABA but not to histamine (H1, H2, or H3) receptors. J Invest Dermatol 111:705–6 Landar A, Zmijewski JW, Dickinson DA et al. (2006) Interaction of electrophilic lipid oxidation products with mitochondria in endothelial cells and formation of reactive oxygen species. Am J Physiol Heart Circ Physiol 290:H1777–87 Mates JM, Segura JA, Alonso FJ et al. (2008) Intracellular redox status and oxidative stress: implications for cell proliferation, apoptosis, and carcinogenesis. Arch Toxicol 82:273–99 Matsui MS, Hsia A, Miller JD et al. (2009) Non-sunscreen photoprotection: antioxidants add value to a sunscreen. J Investig Dermatol Symp Proc 14:56–9 Moodycliffe AM, Bucana CD, Kripke ML et al. (1996) Differential effects of a monoclonal antibody to cis-urocanic acid on the suppression of delayed and contact hypersensitivity following ultraviolet irradiation. J Immunol 157:2891–9 Norval M (2006) The mechanisms and consequences of ultraviolet-induced immunosuppression. Prog Biophys Mol Biol 92:108–18 Norval M, Simpson TJ, Bardshiri E et al. (1989) Urocanic acid analogues and the suppression of the delayed type hypersensitivity response to Herpes simplex virus. Photochem Photobiol 49:633–9 Pastore S, Mascia F, Mariani V et al. (2008) The epidermal growth factor receptor system in skin repair and inflammation. J Invest Dermatol 128:1365–74 Peus D, Vasa RA, Meves A et al. (1998) H2O2 is an important mediator of UVB-induced EGF-receptor phosphorylation in cultured keratinocytes. J Invest Dermatol 110:966–71 Pradhan S, Kim HK, Thrash CJ et al. (2008) A critical role for the proapoptotic protein Bid in ultraviolet-induced immune suppression and cutaneous apoptosis. J Immunol 181:3077–88 Rinaldi M, Moroni P, Leino L et al. (2006) Effect of cis-urocanic acid on bovine neutrophil generation of reactive oxygen species. J Dairy Sci 89:4188–201 Rundhaug JE, Fischer SM (2008) Cyclo-oxygenase-2 plays a critical role in UV-induced skin carcinogenesis. Photochem Photobiol 84: 322–9 Shreedhar V, Giese T, Sung VW et al. (1998) A cytokine cascade including prostaglandin E2, IL-4, and IL-10 is responsible for UV-induced systemic immune suppression. J Immunol 160:3783–9 Sreevidya CS, Fukunaga A, Khaskhely NM et al. (2010) Agents that reverse UV-induced immune suppression and photocarcinogenesis affect DNA repair. J Invest Dermatol 130:1428–37 Steenvoorden DP, Beijersbergen van Henegouwen GM (1999) Protection against UV-induced systemic immunosuppression in mice by a single topical application of the antioxidant vitamins C and E. Int J Radiat Biol 75:747–55 Ueda S, Masutani H, Nakamura H et al. (2002) Redox control of cell death. Antioxid Redox Signal 4:405–14 Valencia A, Kochevar IE (2008) Nox1-based NADPH oxidase is the major source of UVA-induced reactive oxygen species in human keratinocytes. J Invest Dermatol 128:214–22 Walterscheid JP, Nghiem DX, Kazimi N et al. (2006) cis-Urocanic acid, a sunlight-induced immunosuppressive factor, activates immune suppression via the 5-HT2A receptor. Proc Natl Acad Sci USA 103:17420–5 Woodward EA, Prele CM, Finlay-Jones JJ et al. (2006) The receptor for cisurocanic acid remains elusive. J Invest Dermatol 126:1191–3

Kripke ML, Cox PA, Alas LG et al. (1992) Pyrimidine dimers in DNA initiate systemic immunosuppression in UV-irradiated mice. Proc Natl Acad Sci USA 89:7516–20

Wright TI, Spencer JM, Flowers FP (2006) Chemoprevention of nonmelanoma skin cancer. J Am Acad Dermatol 54:933–46

Laihia JK, Kallio JP, Taimen P et al. (2010) Protodynamic intracellular acidification by cis-urocanic acid promotes apoptosis of melanoma cells in vitro and in vivo. J Invest Dermatol 130:2431–9

Zhuang S, Ouedraogo GD, Kochevar IE (2003) Downregulation of epidermal growth factor receptor signaling by singlet oxygen through activation of caspase-3 and protein phosphatases. Oncogene 22:4413–24

Laihia JK, Pylkkanen L, Laato M et al. (2009) Protodynamic therapy for bladder cancer: in vitro results of a novel treatment concept. BJU Int 104:1233–8

Zuliani T, Denis V, Noblesse E et al. (2005) Hydrogen peroxide-induced cell death in normal human keratinocytes is differentiation dependent. Free Radic Biol Med 38:307–16
