|Home | About | Journals | Submit | Contact Us | Français|
Prostaglandins belong to a class of cyclic lipid-derived mediators synthesized from arachidonic acid via COX-1, COX-2 and various prostaglandin synthases. Members of this family include prostaglandins such as PGE2, PGF2α, PGD2 and PGI2 (prostacyclin) as well as thromboxane. In the present studies we analyzed the effects of UVB on prostaglandin production and prostaglandin synthase expression in primary cultures of undifferentiated and calcium-differentiated mouse keratinocytes. Both cell types were found to constitutively synthesize PGE2, PGD2 and the PGD2 metabolite PGJ2. Twenty-four hr after treatment with UVB (25 mJ/cm2), production of PGE2 and PGJ2 increased, while PGD2 production decreased. This was associated with increased expression of COX-2 mRNA and protein. UVB (2.5 – 25 mJ/cm2) also caused marked increases in mRNA expression for the prostanoid synthases PGDS, mPGES-1, mPGES-2, PGFS and PGIS, as well as expression of receptors for PGE2 (EP1 and EP2), PGD2 (DP and CRTH2) and prostacyclin (IP). UVB was more effective in inducing COX-2 and DP in differentiated cells and EP1 and IP in undifferentiated cells. UVB readily activated keratinocyte PI-3-kinase (PI3K)/Akt, JNK and p38 MAP signaling pathways which are known to regulate COX-2 expression. While inhibition of PI3K suppressed UVB-induced mPGES-1 and CRTH2 expression, JNK inhibition suppressed mPGES-1, PGIS, EP2 and CRTH2, and p38 kinase inhibition only suppressed EP1 and EP2. These data indicate that UVB modulates expression of prostaglandin synthases and receptors by distinct mechanisms. Moreover, both the capacity of keratinocytes to generate prostaglandins and their ability to respond to these lipid mediators are stimulated by exposure to UVB.
Eicosanoids are derived from arachidonic acid released from cell membranes through the action of phospholipases. Three major classes of enzymes are involved in generating eicosanoids: cyclooxygenases (COX), which produce prostaglandins, lipoxygenases, which generate leukotrienes and hydroxyeicosatetraenoic acids, and cytochrome P450 enzymes which mediate the synthesis of 19- and 20-hydroxyeicosatetraenoic acids (HETEs), epoxyeicosatrienoic acids (EETs) and diHETEs (see Fig 1 for a summary of eicosanoid metabolism) (Funk, 2001). In the skin, COX-derived metabolites are known to control many aspects of inflammation, cell growth control, apoptosis and tumor development (Trifan and Hia, 2003). Two key COX enzymes have been extensively characterized, COX-1, a constitutive enzyme, and COX-2, an enzyme induced by many irritants and inflammatory agents (Fitzpatrick, 2004). These enzymes mediate the synthesis of the prostanoid precursor, prostaglandin H2 (PGH2) (Fitzpatrick, 2004). Metabolism of PGH2 leads to the generation of PGD2, PGE2, PGF2, PGI2 and thromboxane A2 which are synthesized by PGD2 synthase, PGE2 synthase, PGF2 synthase, PGI2 synthase and thromboxane A2 synthase, respectively (Helliwell et al., 2004). Exposure to UVB light, which is known to cause cutaneous inflammation and cancer (Hruza and Pentland, 1993), has been reported to upregulate COX-2 in human and mouse skin (Buckman et al., 1998; Isoherranen et al., 1999; Chen et al., 2001; Bachelor et al., 2005), a process linked to many of its biological effects (Fitzpatrick, 2004). Increased concentrations of PGE2, PGF2α and PGD2 have been detected in UVB-exposed skin (Black et al., 1980a) and, in cultured keratinocytes, PGE2 is the predominant prostaglandin identified following UVB exposure (Miller et al., 1994).
At present, little is known about the effects of UVB on expression of the prostanoid synthases in keratinocytes and this represents the focus of the present studies. We found that UVB was highly effective in inducing not only COX-2, but also mPGES-1, mPGES-2, PGDS, PGFS and PGIS in primary cultures of undifferentiated and calcium-differentiated mouse keratinocytes. UVB also induced expression of receptors for PGE2, PGD2 and PGI2. These responses were regulated by Akt and MAP kinases, two signal transduction pathways known to be activated by UVB light. Coordinate upregulation of the prostaglandin biosynthetic enzymes and prostaglandin receptors in keratinocytes may be an important mechanism contributing to the biological actions of UVB in the skin.
Rabbit polyclonal antibodies to p38, phospho-p38, JNK, phospho-JNK, ERK 1/2, phospho-ERK 1/2, Akt and phospho-Akt were from Cell Signaling Technology (Beverly, MA), and rabbit anti-keratin-1, keratin-10 and filaggrin from Covance Research Products (Berkley, CA). Goat polyclonal anti-COX-2 and donkey anti-goat secondary antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Horseradish peroxidase-labeled goat anti-rabbit secondary antibodies, Silver Stain Plus and detergent-compatible protein assay reagents were obtained from Bio-Rad Laboratories (Hercules, CA). M-MLV Reverse Transcriptase was from Promega (Madison, WI) and collagen IV from BD Biosciences (San Diego, CA). The Western Lightning enhanced chemiluminescence kit (ECL) was from Perkin Elmer Life Sciences, Inc. (Boston, MA). HPLC prostaglandin standards were purchased from Cayman Chemical (Ann Arbor, MI). SYBR Green Master Mix and other PCR reagents were from Applied Biosystems (Foster City, CA). Cell culture reagents and 2-(2, 3-naphthalimino)ethyl-trifluoromethanesulphonate (NE-OTf) were purchased from Invitrogen Corp (Carlsbad, CA). SP600125 and Wortmannin were from Calbiochem (La Jolla, CA). SB203580, protease inhibitor cocktail, arachidonic acid and all other chemicals were from Sigma (St. Louis, MO).
Primary keratinocytes were isolated from the epidermis of C57BL neonatal mice and cultured as previously described (Hager et al., 1999). Cells, grown in six-well collagen IV-coated plates, were placed in serum-free low calcium (0.05 mM) Keratinocyte Growth Medium (Cambrex, Walkersville, MD) to maintain their undifferentiated phenotype. Differentiation was induced by the addition of calcium (0.15 mM) to the medium (Yuspa et al., 1988), and was confirmed by morphological changes (Figure 2) and expression of differentiation markers including keratin 1, keratin 10 and filaggrin (Black et al., 2008), as analyzed by Western blotting (see below). For some experiments, primary keratinocytes were obtained from the Yale University Cell Culture Facility (New Haven, CT).
UVB was generated from a bank of 2 FS40BL bulbs calibrated using an International IL-1700 UV-radiometer. According to the manufacturer, the spectrum of the bulbs was approximately 85% UVB (290–320 nm), 1% UVA (320–400 nm) and less than 1% UVC (< 290 nm) with the remainder composed of visible light. Undifferentiated and differentiated cells were grown to 90% confluence and exposed to UVB (2.5–25 mJ/cm2) in phosphate-buffered saline (PBS) as previously described (Black et al., 2008). Following UVB treatment, the PBS was removed and the cells refed with Keratinocyte Growth Medium containing 0.05 mM (low) or 0.15 mM (high) calcium. In some experiments, the p38 MAP kinase inhibitor SB203580 (10 µM), the JNK kinase inhibitor SP600125 (20 µM), the PI3-kinase inhibitor wortmannin (0.1 µM) or DMSO control was added to the medium, and the cells incubated at 37°C for 3 hr prior to UVB treatment as previously described (Black et al., 2008). When the cells were refed after UVB exposure, the medium contained the same concentrations of inhibitors. The cytotoxicity of UVB treatment was assayed using trypan blue exclusion when the cells were collected for mRNA and protein. The viability (SEM, n = 4) was 97% ± 2% for control cells and 94% ± 3% for cells treated with the highest dose of UVB (25 mJ/cm2).
Western blotting was performed as previously described (Black et al., 2008). Briefly, cell lysates were prepared using an SDS-lysis buffer (10 mM Tris-base and 1% SDS, pH 7.6 supplemented with a protease inhibitor cocktail consisting of 4-(2-aminoethyl)benzenesulfonyl fluoride, aprotinin, bestatin hydrochloride, N-(trans-epoxysuccinyl)-L-leucine 4-guanidinobutylamide, EDTA and leupeptin). Proteins (20 µg) from lysates were separated on 7.5% or 10% SDS-polyacrylamide gels and then transferred to nitrocellulose membranes. After incubating the membranes in blocking buffer (5% dry milk Tris-buffered saline containing 0.1% Tween 20) for either 1 hr at room temperature or overnight at 4°C, the membranes were then incubated either 1 to 2 hr at room temperature or overnight at 4°C with primary antibodies followed by horseradish peroxidase-conjugated secondary antibodies for 1 hr at room temperature. Protein expression was visualized using enhanced chemiluminescence (ECL) reagents. Equal protein loading was confirmed through silver staining of the gels.
Confluent monolayers of undifferentiated cells were exposed to UVB (25 mJ/cm2) and refed with 1 ml culture medium. Arachidonic acid (10 µM) was added to the medium 24 hr later. After an additional 20 min at 37°C, prostaglandin metabolites were extracted from the cells and medium by the addition of 1 ml of ice-cold methanol. Both cells and medium were collected, centrifuged (12,000 × g for 10 min at 4°C) and aliquots of the supernatants evaporated to dryness under helium. Prostaglandin metabolites were then derivatized as described previously (Yue et al., 2004). Briefly, the dried extracts were solubilized in anhydrous acetonitrile, supplemented with N,N-diiospropylethylamine catalyst (dried with 5 Å molecular sieves) and NE-OTf (2 mg/ml in anhydrous acetonitrile). After incubation in a dessicator for 30 min at 4°C, the samples were evaporated to dryness under helium. The derivatized samples were then dissolved in methanol and analyzed using a Shimadzu HPLC 10A system (Kyoto, Japan) fitted with a Beckman Ultrasphere C18 column (4.6 mm × 250 mm) (Fullerton, CA) and a Shimadzu RF-551 fluorescence detector. The excitation and emission wavelengths of the fluorescence detector were set at 260 and 396 nm, respectively. Derivatized prostaglandins were separated at a flow rate of 1.0 ml/min using a step gradient consisting of 85% mobile phase A (acetonitrile:water:acetic acid, 25:75:0.1): 15% mobile phase B (100 % acetonitrile) for 25 min, 50% A: 50% B for 15 min; and 15% B for 10 min.
RNA was isolated from the cells using Tri Reagent following the protocol provided by the manufacturer. The RNA was converted to cDNA using M-MLV reverse transcriptase. For each gene to be tested, a standard curve composed of serial dilutions of a pool of the cDNA from the samples was used as a reference. All values were normalized to β-actin (n = 3, ± SEM). The undifferentiated control was assigned a value of 1, and the values of both the undifferentiated and differentiated samples were calculated relative to this control. Real-time PCR was performed on an ABI Prism 7900 Sequence Detection System using 96- well optical reaction plates. SYBR-Green was used for detection of fluorescent signal and the standard curve method was used for relative quantitative analysis. The primer sequences for the genes were generated using Primer Express software (Applied Biosystems) and the oligonucleotides were synthesized by Integrated DNA Technologies, Inc. (Coralville, IA). The forward and reverse sequences used are listed in Table 1.
Data are expressed as mean ± SEM. Statistical differences between the means were determined using two-way analysis of variance (ANOVA) and were considered statistically significant at p < 0.05.
In initial experiments we analyzed prostaglandin production by primary cultures of undifferentiated and differentiated murine keratinocytes. These cells were found to constitutively synthesize several prostaglandins important in regulating inflammation including PGE2, PGD2, PGJ2 and PGA2 (Fig 3 and not shown). PGA2 and PGJ2 are cyclopentenone prostaglandins formed through the spontaneous dehydration of PGE2 and PGD2, respectively (Straus and Glass, 2001). Treatment of the cells with UVB (25 mJ/cm2) markedly increased production of PGE2 and PGJ2, but decreased production of PGD2. No alterations were evident in the production of PGA2 and there were no major differences between undifferentiated and differentiated cells (not shown).
To analyze mechanisms mediating prostaglandin production in keratinocytes, we measured mRNA expression of enzymes controlling the release of arachidonic acid from cell membranes. These included cytosolic phospholipase A2 (cPLA2), which acts directly on phospholipid membranes to release arachidonic acid (Balsinde et al., 2002), as well as phospholipase Cβ1 (PLC β1), which acts indirectly by generating diacylglycerol (DAG), a metabolite that is hydrolyzed by diacylglycerol lipase (DAGL) to monoacylglycerol (MAG) and then to arachidonic acid by monoacylglycerol lipase (MAGL) (Rebecchi and Pentyala, 2000). Real time PCR analysis revealed that keratinocytes express mRNA for cPLA2, PLCβ1, DAGL and MAGL (Fig. 4). While UVB had no major effects on cPLA2 or MAGL expression, expression of PLCβ1 increased 8-fold and DAGL decreased 2–3 fold. Although the differentiation status of keratinocytes has been reported to be important in regulating biological responses to UVB light (Tron et al., 1999), we found no major differences in expression of arachidonic acid mobilizing enzymes between undifferentiated and calcium-differentiated keratinocytes (Fig. 4).
Two isoforms of COX have been identified; COX-1 and COX-2, which catalyze the two-step oxidation of arachidonic acid to PGH2, the precursor for PGE2, PGF2α and PGD2 as well as prostacyclin and thromboxane (see Fig. 1). UVB was found to stimulate expression of COX-2 mRNA and protein in both undifferentiated and differentiated cells (Fig. 5). Differentiated cells appeared to be more sensitive to UVB when compared to undifferentiated cells. In contrast, no major changes in COX-1 expression were noted in either cell type. PGE2 is the major prostaglandin generated in the skin and is produced by the action of the microsomal PGE2 synthases, mPGES-1 and mPGES-2 (Murakami et al., 2003). We found that mRNA expression of these enzymes increased 4–6-fold in both undifferentiated and differentiated keratinocytes following exposure to UVB (Fig 6). Interestingly, UVB also upregulated expression of PGFS, PGDS and PGIS but not TXAS, the enzymes responsible for the formation of PGF2α, PGD2, PGI2 and TXA2, respectively (Fig. 6). In general, no major differences were noted between the undifferentiated and differentiated cell types. These data are consistent with our findings that PGE2 and PGJ2 production is increased in keratinocytes in response to UVB (Fig. 3). We also found that differentiated cells contained approximately two times more constitutive PGFS when compared to undifferentiated cells. Notably, PGFS expression in these cells was approximately two-fold higher across the range of UVB doses tested (2.5 – 25 mJ/cm2) (Fig 6).
Prostaglandins initiate their biological activity by binding to specific membrane receptors (Narumiya et al., 1999). To date, four PGE2 receptors have been identified (EP1, EP2, EP3 and EP4). We found that expression of EP1 and EP2 was induced by UVB light (approximately 4- and 6-fold, respectively) in both undifferentiated and differentiated keratinocytes (Fig. 7). In contrast, no major changes were noted in expression of EP3 or EP4. Interestingly, at 10 and 15 mJ/cm2 UVB, greater increases in EP1 and EP2 expression were observed in undifferentiated keratinocytes when compared to differentiated cells. Expression of the two known receptors for PGD2, DP and CRTH2, were also increased following UVB exposure. Although differentiated cells were more responsive to UVB in terms of DP receptor expression, no major differences were observed in CRTH2 expression between the cell types. UVB also induced expression of the prostacyclin receptor, IP, and the maximal responses were greater in undifferentiated cells (16-fold) when compared to the differentiated cells (8-fold). No significant alterations were detected in the expression of the PGF2α receptor FP or the thromboxane A2 receptor TP.
UVB was found to upregulate p38 and JNK MAP kinases in a dose-dependent manner in both undifferentiated and differentiated keratinocytes (Fig 8) and the effects of UVB on JNK were more pronounced. In contrast, only small increases in the ERK1/2 MAP kinase were evident in undifferentiated cells. The undifferentiated cells also responded to UVB with activation of Akt. This appeared to be due to increased expression of total Akt protein. In differentiated cells, high constitutive levels of activated Akt were noted and this did not change with UVB exposure.
We next used inhibitors of p38, JNK and Akt to evaluate the role of these kinases in regulating expression of COX-2, mPGES-1 and PGIS, three genes that were highly responsive to UVB. SB203580, a p38 kinase inhibitor, and SP600125, a JNK kinase inhibitor, were found to markedly suppress UVBinduced COX-2 expression in both undifferentiated and differentiated keratinocytes (Fig 9 and not shown). In contrast, mPGES-1 expression decreased in response to the JNK inhibitor and the p38 inhibitor was effective in suppressing mPGES-1 only at the highest dose of UVB (25 mJ/cm2). Whereas UVB-induced expression of PGIS was unchanged following p38 inhibition, JNK inhibition effectively suppressed expression of this enzyme (Fig. 9). The role of PI3K/Akt in regulating expression of these genes was assessed using Wortmannin, an inhibitor of PI3K. As observed with the MAP kinases, expression of COX-2 and mPGES-1, but not PGIS, was markedly reduced by wortmannin. No major differences were noted in the effects of MAP kinase or Akt kinase inhibition on UVB-induced gene expression in undifferentiated and differentiated keratinocytes (data not shown).
We also examined the effects of p38, JNK and Akt inhibitors on keratinocyte expression of EP1, EP2, CRTH2, DP and IP, the prostanoid receptors upregulated in response to UVB. UVB-induced EP1 expression was significantly reduced by p38, but not JNK inhibition (Fig 10). In contrast, expression of EP2 was decreased by JNK, but only slightly with p38 inhibition. Kinase inhibition also differentially affected UVB-induced expression of the two PGD2 receptors; CRTH2 was reduced by JNK inhibition, DP was unaffected. No major alterations were detected in IP expression with the kinase inhibitors. PI3K inhibition suppressed UVB-induced expression of CRTH2, while no changes were observed in expression of EP1, EP2, DP or IP.
Previous work has demonstrated that exposure of mouse and human skin to ultraviolet light leads to alterations in the production of lipid-derived mediators including PGE2, PGF2α and PGD2 (Black et al., 1980a; Miller et al., 1994). These eicosanoids are known to mediate many skin reactions and a number of drugs that inhibit their production effectively suppress skin inflammation (Fogh and Kragballe, 2000; Wilgus et al., 2003). The present studies demonstrate that primary cultures of mouse keratinocytes constitutively produce these prostaglandins, as well as low levels of PGJ2. Each of these prostaglandins or their metabolites is known to be important in regulating cell growth, and their constitutive production in keratinocytes is presumably important in skin homeostasis (Fogh and Kragballe, 2000). As previously observed in mouse and human skin (Black et al., 1980b; Ruzicka et al., 1983), UVB stimulated production of PGE2 in undifferentiated and differentiated cultures of primary mouse keratinocytes. PGE2 is a potent proinflammatory mediator and it has been shown to play a key role in the development of UVB-induced skin cancer (Furstenberger et al., 1989). UVB was also found to cause marked increases in PGJ2 in keratinocytes, but to reduce PGD2 production. Decreased production of PGD2 is consistent with an earlier report demonstrating that UVB suppressed production of this arachidonic acid metabolite in hairless mouse skin (Ruzicka et al., 1983). PGD2 is relatively unstable and degrades by spontaneous dehydration and isomerization reactions to PGJ2 which is metabolized to a variety of biological active metabolites (Kikawa et al., 1984). Increased levels of PGJ2 in keratinocytes after UVB may be due to increased PGD2 biosynthesis (see Figure 1). PGJ2, and the PGE2 dehydration product PGA2, contain a reactive unsaturated carbonyl prostane which forms Michael adducts with cellular thiols including glutathione and cysteine residues in cellular proteins (Straus and Glass, 2001). A variety of signaling molecules important in mediating inflammation in response to UVB are modulated by these cyclopentenones including NF-κB (Rossi et al., 2000) and PPARγ (Shibata et al., 2002). Moreover, 15-deoxy-Δ12,14-PGJ2, a highly active anti-inflammatory metabolite of PGJ2, is known to potentiate tumor formation in a mouse skin carcinogenesis model (Millan et al., 2006). These data suggest that cyclopentenones including PGJ2 and its metabolites may be important in mediating, at least in part, the biological actions of UVB in the skin.
To investigate mechanisms mediating prostaglandin production in keratinocytes, we characterized the effects of UVB on expression of the enzymes regulating their biosynthesis. The initial step of prostanoid biosynthesis is mobilization of arachidonic acid from phospholipid membranes through the action of phospholipases (Funk, 2001). Previous studies reported that UVB stimulates arachidonic acid release from membrane phospholipids (DeLeo et al., 1985; Punnonen et al., 1987), as well as cPLA2 synthesis and activity in skin (Kang-Rotondo et al., 1993; Gresham et al., 1996). In contrast, we found no major changes in cPLA2 mRNA expression in undifferentiated or differentiated mouse keratinocytes following UVB light treatment. An alternative pathway for arachidonic acid release from cell membranes is via activation of PLC, forming 1,2-diacylglycerol and inositol-1,4,5-triphosphate (Rebecchi and Pentyala, 2000). Inositol-1,4,5-triphosphate causes calcium release from intracellular stores (Berridge and Irvine, 1984); both 1,2-diacylglycerol and calcium activate protein kinase C, a signaling enzyme that can cause phosphorylation and activation of cPLA2 (Lin et al., 1993).We found that PLCβ1 mRNA expression increased 24 hr after UVB. On the contrary, there were no changes or small decreases in mRNA expression in undifferentiated and differentiated cells for DAG lipase and MAG lipase, the two enzymes that sequentially hydrolyze diacylglycerol to form arachidonic acid (Rebecchi and Pentyala, 2000). This is consistent with a report of increased diacylglycerol levels in keratinocytes after UVB exposure (Punnonen and Yuspa, 1992). Thus, either increases in PLCβ1 or in the activities of DAG or MAG lipase may be sufficient to stimulate arachidonic acid production in mouse keratinocytes.
It is well recognized that UVB-induced prostaglandin production in the skin is mediated by COX-2 (Fitzpatrick, 2004). Increased expression of this enzyme has been directly associated with many UVB-induced skin inflammatory reactions including edema, erythema, keratinocyte proliferation and epidermal hyperplasia (Rodriguez-Burford et al., 2005). Changes in COX-2 expression have also been linked to the accumulation of DNA damage and mutations, as well as the development of premalignant lesions and skin cancer (Muller-Decker et al., 1998; Fischer et al., 2007). We found that COX-2 mRNA and protein were increased in mouse keratinocytes following UVB exposure. These data are consistent with earlier reports showing that UVB effectively induces COX-2 in mouse and human keratinocytes (Isoherranen et al., 1999; Black et al., 2008). Our data also demonstrate that differentiated cells are more sensitive to UVB-induced upregulation of COX-2 mRNA and protein, a finding consistent with previous studies showing that UVB-induced COX-2 expression in mouse skin occurs primarily in the suprabasal, differentiated keratinocytes (Athar et al., 2001). At the present time, the mechanisms underlying the increased sensitivity of differentiated cells to UVB are not known although it should be noted that abnormal differentiation occurs in mouse epidermis following overexpression of COX-2 (Neufang et al., 2001). Differentiation related changes in mRNA expression for enzymes regulating arachidonic acid release and other prostanoid synthases and receptors were not observed following UVB treatment (see further below) indicating that this response was selective for COX-2.
The effects of UVB on expression of the various prostanoid synthases in keratinocytes have not been investigated previously. The synthesis of PGE2 is catalyzed by the microsomal prostaglandin E synthases mPGES-1 and mPGES-2 (Murakami et al., 2002), both of which were upregulated by UVB in undifferentiated and differentiated keratinocytes. mPGES-1 is a glutathione-dependent enzyme and member of the membrane associated proteins involved in eicosanoid and glutathione metabolism (MAPEG) superfamily. This enzyme is known to be upregulated in response to proinflammatory stimuli, a process thought to be critical for PGE2 production during inflammatory responses (Claveau et al., 2003). Our findings showing that mPGES-2 is also upregulated by UVB is in contrast to earlier studies indicating that this is a constitutively expressed enzyme, and suggest that mPGES-2, like mPGES-1, is also important in regulating increases in PGE2 production in the skin during inflammation. Coordinate increases in expression of COX-2 and mPGES-1 have been previously described (Murakami et al., 2000; Murakami et al., 2003). UVB-induced increases in COX-2, mPGES-1 and mPGES-2 in the skin are likely to be important for maximal stimulation of PGE2 production. It should be noted that mPGES-1 also possesses glutathione transferase and peroxidase activity (Murakami et al., 2002), and increases in this enzyme may also be key in reducing cytotoxic lipophilic hydroperoxides generated in the skin following exposure to UVB.
Also of interest is our findings that exposure to UVB resulted in increased expression of two (EP1 and EP2) of the four known keratinocyte PGE2 receptors. This indicates that UVB not only upregulates the capacity of keratinocytes to generate PGE2, but also their responsiveness to this prostaglandin. PGE2 receptors belong to a family of G-protein coupled receptors that function via distinct signaling mechanisms (Narumiya et al., 1999). All of the known receptors, including EP1 and EP2, are expressed in mouse skin and several have been directly linked to UVB- and phorbol ester-induced skin inflammation and carcinogenesis (Thompson et al., 2001; Sung et al., 2006; Brouxhon et al., 2007; Chun et al., 2007). Thus, UVB has previously been reported to stimulate expression of EP1, EP2 and EP4 in mouse skin, but to reduce expression of EP3 (Tober et al., 2007). Moreover, modulating EP receptor expression or activity using genetic and pharmacological approaches has demonstrated that EP1, EP2 and/or EP4 are critical mediators of the actions of UV light in the skin (Tober et al., 2006). Also of note was our observation that EP1 and EP2 relative mRNA levels increased at intermediate doses of UVB (10 and/or 15 mJ/cm2), suggesting that undifferentiated cells were more responsive to UVB than differentiated cells in this parameter. Recent studies using techniques in immunohistochemistry have shown that these receptors are localized in suprabasal keratinocytes in mouse skin (Tober et al., 2007).
An important finding from our work is that UVB also stimulated PGE2 biosynthesis, as well as expression of mRNA for enzymes mediating production of PGD2, PGF2α and prostacyclin, and their respective receptors. These data are generally consistent with the results of our metabolism studies. PGD2 and its receptors, DP and CRTH2, are known to be involved in chronic allergic skin inflammation (Angeli et al., 2004; Satoh et al., 2006). The marked increases in PGFS and PGIS that we noted in response to UVB also suggest that eicosanoids produced by these enzymes may be important mediators of cutaneous inflammation (Muller et al., 2000; Takahashi et al., 2002). Increases in expression of receptors for PGD2α and prostacyclin provide further evidence that UVB stimulates both the capacity of keratinocytes to produce these prostaglandins and cellular responsiveness to these mediators. Additionally, UVB was more effective in inducing expression of IP in undifferentiated cells and DP in differentiated cells. These data suggest that the two cell types may respond distinctly to prostacyclin and PGD2, respectively.
It is well recognized that UVB activates a number of signaling pathways leading to inflammatory gene expression, a process referred to as ‘the UV response’ (Heck et al., 2004). Initially characterized by activation of the immediate early genes, c-fos and c-jun and various transcription factors including NFκB (Heck et al., 2004), more recent studies have demonstrated that PI3K/Akt and MAP kinase signaling are also key to the response (Bode and Dong, 2003). We found that MAP kinase and Akt were readily activated by UVB in mouse keratinocytes; however, this response was differentiation-dependent. Thus, while the ERK1/2, JNK, p38 MAP kinases and Akt were activated by UVB in undifferentiated cells, in differentiated cells, only JNK and p38 kinase were activated. Differentiated cells were also found to contain high constitutive levels of activated Akt. Differential activation of Akt in differentiated cells was most likely due to the higher intracellular concentrations of calcium which have been shown to activate the PI3K signaling pathway leading to Akt phosphorylation (Yuspa et al., 1988). These later findings are also consistent with studies showing that the PI3K/Akt pathway is activated during mouse keratinocyte differentiation both in culture and in the intact epidermis (Calautti et al., 2005). It is possible that activation of Akt is important in the process of keratinocyte differentiation. Differences in UVB-induced PI3K and MAP kinase signaling in undifferentiated and differentiated keratinocytes may contribute, at least in part, to their distinct responses following UVB treatment, with respect to expression of COX-2, EP1, EP2, IP and DP.
Previous studies have demonstrated that inhibition of p38 MAP kinase effectively suppresses UVB-induced inflammation in mouse skin including expression of proinflammatory cytokines and COX-2 production (Hildesheim et al., 2004; Kim et al., 2005). Similarly, we found that UVB-induced COX-2 expression is regulated via p38 kinase, as well as JNK and Akt kinase. The role of these kinases in controlling expression of enzymes downstream of COX-2, however, is not known. We found that UVB-induced expression of mPGES-1 and PGIS were suppressed by an inhibitor of JNK. Interestingly, mPGES-1, but not PGIS, was also suppressed by PI3K/Akt inhibition indicating that the two prostanoid synthases are controlled by distinct mechanisms. Our data indicate that JNK is also important in controlling CRTH2 and EP2 expression, while p38 kinase regulates EP1 and EP2. Moreover, PI3K/Akt is selectively involved in UVB-induced CRTH2 expression. Thus, each of the prostaglandin receptors also appears to be regulated distinctly. At the present time, signaling mechanisms regulating UVB-induced expression of PGIS, DP and IP in keratinocytes remain to be determined.
Work on the role of eicosanoids in UVB-induced skin inflammation and carcinogenesis has largely been limited to the study of COX-2 expression and PGE2 signaling. Several studies have shown a link between prostaglandin production and the initiation and progression of skin carcinogenesis (Fischer et al., 1999; Wilgus et al., 2003; Chun et al., 2007). Our studies demonstrate that UVB also stimulates a number of additional synthases important in prostaglandin production, as well as prostaglandin receptors and each may be critical for mediating the biological effects of UVB in mouse skin (see Figure 11 for summary of prostaglandin synthase and prostaglandin receptor genes upregulated by UVB and those regulated by MAP kinases and/or PI3K). Whether or not the genes induced by UVB contribute to its biological actions will depend on levels of expression of functional proteins for the prostaglandin synthases and receptors as well as amounts of prostaglandins produced in keratinocytes. Further studies are needed to more clearly define the prostaglandins produced by keratinocytes and their roles in modulating the responses of the skin to UVB.
This work was supported in part by National Institutes of Health grants CA100994, CA093798, ES005022, ES004738, AR055073, GM034310 and by NJ State Commission on Cancer Research fellowship 05-2413-CCR-E0 awarded to ATB. This work was also funded in part by the National Institutes of Health CounterACT Program through the National Institute of Arthritis and Musculoskeletal and Skin Diseases (award #U54AR055073). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the federal government.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.