Curcumin inhibits the expression of COX-2 in UVB-irradiated human keratinocytes (HaCaT) by inhibiting activation of AP-1: p38 MAP kinase and JNK as potential upstream targets

Ultraviolet B (UVB) irradiation of skin induces an acute inflammation. Cyclooxygenase-2 (COX-2) protein plays key roles in acute inflammation in UVB-irradiated keratinocyte cell line HaCaT. Recently, curcumin has been regarded as a promising anti-inflammatory agent due to its ability to inhibit COX-2 expression. However, it remains largely unknown whether curcumin inhibits the UVB-induced COX-2 expression in HaCaT cells. This study was undertaken to clarify the effect of curcumin on the expression of COX-2 in UVB- irradiated HaCaT cells and further determined the molecular mechanisms associated with this process. In this study, we have found that the expression of COX-2 mRNA and protein were up-regulated in UVB-irradiated HaCaT cells in a dose- and time-dependent manner. Interestingly, treatment with curcumin strongly inhibited COX-2 mRNA and protein expressions in UVB-irradiated HaCaT cells. Notably, there was effective inhibition by curcumin on UVB-induced activations of p38 MAPK and JNK in HaCaT cells. The DNA binding activity of AP-1 transcription factor was also markedly decreased with curcumin treatment in UVB-irradiated HaCaT cells. These results collectively suggest that curcumin may inhibit COX- 2 expression by suppressing p38 MAPK and JNK activities in UVB-irradiated HaCaT cells. We propose that curcumin may be applied as an effective and novel sunscreen drug for the protection of photoinflammation.


Introduction
Exposure of excessive sunlight is an important etiologic factor in the development of acute inflammation, characterized by erythema, edema, and immunosuppression, and thus consequently linked to the progression of skin cancer (Granstein and Matsui, 2004;Matsumura and Ananthaswamy, 2004). Ultraviolet B (UVB) is a well-known major risk factor for the development of acute inflammation as well as non-melanoma skin cancer in epidermis (De Fabo et al., 2004;Ramos et al., 2004). Accumulating data indicate that UVB exerts its detrimental effect mainly through the induction of direct DNA damage or the production of reactive oxygen species (ROS) (de Gruijl, 2002;Kulms et al., 2002;Heck et al, 2003). Direct DNA damage or ROS often triggers some signaling pathw ays such as m itogen-activated protein kinases (MAPKs) which are known to be involved in proliferation and survival of the cells (Rhee, 1999;Torres and Forman, 2003). The MAPKs are a family of proline-directed Ser/Thr kinases composed of extracellular signal-regulated kinase (ERK), c-Jun NH2-terminal kinase (JNK), and p38 MAPK. Recent studies have shown that activation of ERK, JNK, and p38

Curcum in inhibits the expression of COX-2 in UVB-irradiated hum an keratinocytes (HaCaT) by inhibiting activation of AP-1: p38 MAP kinase and JNK as potential upstream targets
MAPK is tightly correlated with acute inflammation and development of skin cancer through increased expression of cyclooxygenase-2 (COX-2) (Chen et al., 2001;Lin et al., 2004;Mahns et al., 2004).
Curcumin is a yellow pigment present in the rhizomes of turmeric (C.longa Linn) and related species, and has a wide array of pharmacological and biological activities including anti-inflammatory, anti-oxidant, and chemopreventive properties (Dorai and Aggarwal, 2004;Zhu et al., 2004). Recently, it has been reported that curcumin-treated cells show the decreased expressions of COX-2 in various cell lines by inhibition of MAPKs signaling pathways (Chun et al., 2003;Hong et al., 2004;Kang et al., 2004). Currently, it is unknown whether curcumin modulates the expression of COX-2 in UVB-irradiated HaCaT cells. Therefore, in this study, we investigated the effects of curcumin on the expression of COX-2 in UVB-irradiated HaCaT cells and further determined the molecular mechanisms of anti-inflammatory and anti-tumor promoting activities of curcumin.

C ell culture
Human keratinocyte cell line, HaCaT cell, was maintained at 37 o C in a humidified atmosphere of 95% air and 5% CO2 in Eagle's minimum essential medium supplemented with 10% heat inactivated fetal bovine serum, 2 mM glutamine, and 100 U/ml penicillin and 100 µg/ml streptomycin. For experiments, cells (5 × 10 4 cells/ml) were seeded in culture dish, and maintained in a tissue culture incubator.

UVB irradiation
As previously described (Lee et al., 2003), UVB was supplied by a closely spaced array of seven Westinghouse FS-40 sunlamps, which delivered uniform irradiation at a distance of 38 cm. The energy output of UVB (290-320 nm) at 38 cm was measured with a UVB photometer (IL1350 photometer, International Light, Newburyport, MA). Cells were exposed for 0, 45, 90, and 180 sec of UVB, corresponding to doses of 0, 50, 100, and 200 mJ/cm 2 . To prevent light absorption by tissue-culture medium, the culture medium was removed just prior to irradiation and replaced with a thin layer of phosphate-buffered saline (PBS) to cover the cells. Tissue culture medium was replaced in dishes immediately after the last UVB dose had been administered.

Effect of UVB on expression of CO X-2
To study the effect of UVB irradiation on the expression of COX-2, HaCaT cells were exposed to UVB at doses ranging from 50 to 200 mJ/cm 2 , and then cells were harvested 12 h after irradiation for Western. As shown in Figure 1A, COX-2 expressions were dramatically increased in a dose-dependent manner. Increased expression of COX-2 was clearly visualized at 6 h, following the markedly increased expression at 12 h after UVB (200 mJ/cm 2 ) irradiation ( Figure 1B). To determine whether up-regulation of COX-2 is regulated at the level of transcription, we performed RT-PCR analysis using specific COX-2 primers. In agreement with Western, increased expression of COX-2 mRNA was clearly visualized at 6 h, following the increased expression levels were maintained at 12 h after UVB (200 mJ/cm 2 ) irradiation ( Figure 1C). These results suggest that the up-regulation of COX-2 by UVB in HaCaT cells is largely due to increased synthesis of COX-2 mRNA.

Inhibitory effect of curcum in on UVB-induced expression of CO X-2
To investigate whether curcumin inhibits the UVB-induced COX-2 expression, HaCaT cells were exposed to UVB (200 mJ/cm 2 ) with or without curcumin. As shown in Figure 2A, up-regulation of COX-2 by UVB was markedly decreased in a concentration-dependent manner. To investigate whether curcumin attenuates COX-2 expression by UVB at transcriptional level, we performed RT-PCR analysis using COX-2 specific primers. As shown in Figure 2B, increased The cells were cultured to 90 % confluence in DMEM supplemented with 10% fetal bovine serum at 37 o C and in 5% CO2. The cells were then serum-starved for 24 h. Before UVB irradiation, the cells were washed with PBS and were exposed to 200 mJ/cm 2 of UVB. After UVB irradiation cells were continuously cultured in serum-free media. The cells were harvested at indicated times and the cell lysates were prepared for Western blot (A, B) and RT-PCR analysis (C). Similar results were shown in two different experiments. expression of COX-2 mRNA by UVB was clearly inhibited in the presence of curcumin, indicating that COX-2 expression is regulated at transcriptional level by curcumin.

Effect of curcum in on UVB-stim ulated activation of M APKs
Accumulating data suggest that UVB-irradiated HaCaT cells show an increased expression of COX-2 by MAPKs-dependent pathways, such as p38 MAPK and JNK. However it is largely unknown whether curcumin modulates the expression of COX-2 by inhibition of these MAPKs pathways in UVB-irradiated HaCaT cells. We examined the effect of UVB on activations of p38 MAPK and JNK in HaCaT cells. As shown in Figure 3A, irradiation of UVB (200 mJ/cm 2 ) resulted in phosphorylation (activation) of both kinases. Activation of both kinases became apparent at 5 min following irradiation of UVB. Stripping and reprobing the same membrane with antibodies against p38 MAPK and JNK revealed no change in total protein levels of each kinase, indicating that UVB induced activation of preexisting p38 MAPK and JNK. To confirm the p38 MAPK-dependent and JNK-dependent COX-2 up-regulation, HaCaT cells were pretreated with SB203580 (20 µM) and SP600125 (20 µM) and then the cells were irradiated by UVB. p38 inhibitor SB203580 and JNK inhibitor SP600125 specifically inhibited the activation of p38 and JNK, respectively ( Figure 3B). We next investigated whether curcumin (10 µM) inhibits the activation of p38 MAPK and JNK in UVB-irradiated HaCaT cells. As shown in Figure 3C, phosphorylation of p38 MAPK and JNK by UVB was markedly decreased in curcumin-treated HaCaT cells. Furthermore UVB-induced COX-2 expressions were dramatically attenuated by pretreatment of SB203580, SP600125, or curcumin ( Figure 3D), indicating that curcumin attenuates the UVB-induced COX-2 expression through inhibition of p38 MAPK and JNK activation. However, curcumin  did not completely inhibit the activation of p38 MAPK or JNK after UVB irradiation, curcumin showed a partial inhibitory effect on the expression of COX-2 compared to SB203580 or SP600125.
Effect of curcum in on UVB-induced activation of AP-1 AP-1, a down-stream molecule of p38 and JNK, plays a key role as a transcription factor involved in UVB-induced COX-2 expression. To investigate whether curcumin decreases the AP-1 binding activity in UVB-irradiated HaCaT cells, we performed an EMSA using 21 bp consensus oligonucleotides. Exposure of HaCaT cells to UVB (200 mJ/cm 2 ) led to an activation of AP-1 dramatically as compared to control ( Figure  4A). However, in the presence of curcumin, UVBinduced AP-1 activation was dramatically inhibited, evidenced by decreased AP-1 binding activity ( Figure  4B). These results suggest that UVB-induced p38 and JNK activations are inhibited by curcumin treatment, and subsequently resulting in decreased activation of AP-1 in nucleus, thereby leading the down-regulation of COX-2 expression in HaCaT cells.

D iscussion
COX-2 plays important roles in the development of carcinogenesis as well as inflammation in UVBirradiated skin (Grandjean-Laquerriere A et al., 2002;Surh YJ, 2003;Wilgus et al., 2003;Cui et al., 2004;Nijsten et al., 2004). Curcumin is known to exert its anti-inflammatory action in various cell lines through inhibition of COX-2 expression (Chun et al., 2003;Kang et al., 2004). Thus, it is important to develop an efficient strategy using curcumin to down-regulate the expression of COX-2 in UVB-irradiated skin (Wolf et al., 2003).
Accumulating evidences suggest that curcumin inhibits the expression of COX-2 in several experimental models (Zhang et al., 1999;Chun et al., 2003;Surh, 2003;Kang et al., 2004). Consistent with these, through the present study, we have demonstrated that curcumin also effectively inhibits the expressions of COX-2 mRNA and protein induced by UVB irradiation in a human keratinocyte cell line HaCaT, which seems to be, to our knowledge, the first report. Accordingly, the expression of COX-2 has been shown to be affected by various intracellular signaling proteins such as p38 MAPK and JNK (Goel et al., 2001;Surh et al., 2001;Liu et al., 2003;Lin et al., 2004;Kang et al., 2004). Interestingly, it has been recently shown that UVB induces the expression of COX-2 by activation of p38 MAPK in HaCaT cells (Chen et al., 2001) or by activations of p38 MAPK and JNK in artificial epidermis (Mahns et al., 2004). In this study, we have observed that p38 MAPK and JNK were dramatically activated in UVB-irradiated HaCaT cells. Of importance, we have found in this study that pretreatment with curcumin effectively suppresses UVB-induced activation of p38 MAPK and JNK as well as expression of COX-2, thus suggesting that curcumin may exert its inhibitory effect on UVB-induced COX-2 expression by inhibiting p38 MAPK and JNK activities. In most cases, p38 MAPK and JNK activations in response to various extracellular stimuli are linked to activation of AP-1 (Minden et al., 1994;Davis, 1995;Minden and Karin, 1997). AP-1 is known to regulate the expression of COX-2 in many systems (Dori and Aggarwal, 2004;Lin et al., 2004;Kang et al., 2004). In this study, we have observed that UVB is able to increase AP-1 DNA binding activity in HaCaT cells that is attenuated by treatment with curcumin. These results may suggest that curcumin may suppress AP-1 activity pro- 2 ), nuclear extracts were prepared and AP-1 DNA binding activity was determined by EMSA. (B) The cells were also pretreated with curcumin (10 µM) and were exposed to UVB (200 mJ/cm 2 ). Nuclear extracts were prepared and AP-1 DNA binding activity was determined by EMSA. bably by inhibition of p38 MAPK and JNK activation in HaCaT cells. Taken together, our data suggest that suppressions of p38 MAPK, JNK, and AP-1 activity are an important molecular mechanism underlying curcumin-mediated down-regulation of COX-2 in UVB-irradiated HaCaT cells. It should be noted that curcumin can inhibit phorbol ester-induced expression of COX-2 in mouse skin through suppression of ERK activity (Chun et al., 2003). However, in this study, we have observed that though UVB is able to activate ERK in HaCaT cells, curcumin does not prevent the activation of ERK in response to UVB irradiation in these cells (data not shown). Thus, molecular mechanisms underlying suppression of COX-2 expression by curcumin may be dependent on the contexts of cells or kinds of stimuli treated to cells.
In conclusion, findings of the present study demonstrate for the first time that curcumin inhibits the expression of COX-2 mRNA and protein in UVBirradiated HaCaT cells, and indicate that the inhibitory effect of curcumin on UVB-induced COX-2 expression is likely to be, at least in part, associated with suppression of JNK, p38 MAPK, and AP-1, thereby suggesting that curcumin may be used as a promising sunscreen substance.