Advertisement

Differential Expression of E Prostanoid Receptors in Murine and Human Non-Melanoma Skin Cancer

      Enhanced prostaglandin production via upregulated cyclooxygenase-2 (COX-2) expression is a likely contributing factor in ultraviolet B (UVB)-induced non-melanoma skin cancer (NMSC), which consists primarily of squamous cell carcinoma (SCC) and basal cell carcinoma (BCC). The four E prostanoid (EP) receptors, designated EP1 through EP4, are known to bind prostaglandin E2 (PGE2), the major prostaglandin present in the skin. We used murine models of UVB-induced SCC and BCC, as well as human NMSC from sun-exposed sites, to investigate the expression of EP receptors during UVB-induced tumorigenesis. We observed that UVB-induced murine SCC are associated with markedly altered expression patterns of the EP receptors when compared with non-irradiated skin. In contrast, expression of all EP receptors was largely absent in UVB-induced murine BCC. We also observed expression of all four EP receptors in human SCC, with altered expression of their mRNA levels as compared with adjacent tumor-free skin. Consistent with our murine studies, no EP receptor expression was detected in human BCC, and their mRNA expression levels showed no change from the adjacent non-tumor-bearing skin. These data suggest that altered EP receptor expression may play a differential role in the development of UVB-induced SCC and BCC in murine and human skin.

      Keywords

      Abbreviations

      AK
      actinic keratosis
      BCC
      basal cell carcinoma
      BD
      Bowar's Disease
      cAMP
      cyclic AMP
      COX-1
      cyclooxygenase-1
      COX-2
      cyclooxygenase-2
      EP
      E prostanoid
      NMSC
      non-melanoma skin cancer
      PGE2
      prostaglandin E2
      SCC
      squamous cell carcinoma
      UV
      ultraviolet
      UVB
      ultraviolet B
      Cyclooxygenase-2 (COX-2) is the inducible isoform of the cyclooxygenase enzyme, and its activity leads to the formation of eicosanoids including prostaglandins (
      • Marnett L.J.
      • Rowlinson S.W.
      • Goodwin D.C.
      • Kalgutkar A.S.
      • Lanzo C.A.
      Arachidonic acid oxygenation by COX-1 and COX-2. Mechanisms of catalysis and inhibition.
      ). COX-2 is upregulated in a number of epithelial cancers (
      • Dannenberg A.J.
      • Altorki N.K.
      • Boyle J.O.
      • Dang C.
      • Howe L.R.
      • Weksler B.B.
      • Subbaramaiah K.
      Cyclo-oxygenase 2: A pharmacological target for the prevention of cancer.
      ), and inhibition of the enzyme has developed as a strategy for chemoprevention of various epithelial neoplasms (
      • Harris R.E.
      • Kasbari S.
      • Farrar W.B.
      Prospective study of nonsteroidal anti-inflammatory drugs and breast cancer.
      ,
      • Harris R.E.
      • Beebe-Donk J.
      • Schuller H.M.
      Chemoprevention of lung cancer by non-steroidal anti-inflammatory drugs among cigarette smokers.
      ;
      • Schapira D.V.
      • Theodossiou C.
      • Lyman G.H.
      The effects of NSAIDs on breast cancer prognostic factors.
      ;
      • Smalley W.
      • Ray W.A.
      • Daugherty J.
      • Griffin M.R.
      Use of nonsteroidal anti-inflammatory drugs and incidence of colorectal cancer: A population-based study.
      ).
      Non-melanoma skin cancers (NMSC) including basal cell carcinomas (BCC) and squamous cell carcinomas (SCC) are the most common type of human malignancy (
      • Marks R.
      An overview of skin cancers. Incidence and causation.
      ). The major etiologic factor involved in the development of NMSC is solar ultraviolet B (UVB) irradiation (
      • Marks R.
      An overview of skin cancers. Incidence and causation.
      ). In both human and murine skin, COX-2 is upregulated following acute UVB exposure and in UVB-induced SCC and SCC precursors (
      • Buckman S.Y.
      • Gresham A.
      • Hale P.
      • Hruza G.
      • Anast J.
      • Masferrer J.
      • Pentland A.P.
      COX-2 expression is induced by UVB exposure in human skin: Implications for the development of skin cancer.
      ;
      • Athar M.
      • An K.P.
      • Morel K.D.
      • et al.
      Ultraviolet B (UVB)-induced cox-2 expression in murine skin: An immunohistochemical study.
      ;
      • An K.P.
      • Athar M.
      • Tang X.
      • et al.
      Cyclooxygenase-2 expression in murine and human nonmelanoma skin cancers: Implications for therapeutic approaches.
      ). COX-2 deficiency in mice reduces susceptibility to skin tumorigenesis, whereas epidermal overexpression augments mouse skin carcinogenesis (
      • Neufang G.
      • Furstenberger G.
      • Heidt M.
      • Marks F.
      • Muller-Decker K.
      Abnormal differentiation of epidermis in transgenic mice constitutively expressing cyclooxygenase-2 in skin.
      ;
      • Muller-Decker K.
      • Neufang G.
      • Berger I.
      • Neumann M.
      • Marks F.
      • Furstenberger G.
      Transgenic cyclooxygenase-2 overexpression sensitizes mouse skin for carcinogenesis.
      ;
      • Tiano H.F.
      • Loftin C.D.
      • Akunda J.
      • et al.
      Deficiency of either cyclooxygenase (COX)-1 or COX-2 alters epidermal differentiation and reduces mouse skin tumorigenesis.
      ). Pharmacologic administration of COX-2 inhibitors dramatically reduces UVB-induced SCC in murine skin (
      • Fischer S.M.
      • Lo H.H.
      • Gordon G.B.
      • Seibert K.
      • Kelloff G.
      • Lubet R.A.
      • Conti C.J.
      Chemopreventive activity of celecoxib, a specific cyclooxygenase-2 inhibitor, and indomethacin against ultraviolet light-induced skin carcinogenesis.
      ;
      • Pentland A.P.
      • Schoggins J.W.
      • Scott G.A.
      • Khan K.N.
      • Han R.
      Reduction of UV-induced skin tumors in hairless mice by selective COX-2 inhibition.
      ,
      • Chun K.S.
      • Kim S.H.
      • Song Y.S.
      • Surh Y.J.
      Celecoxib inhibits phorbol ester-induced expression of COX-2 and activation of AP-1 and p38 MAP kinase in mouse skin.
      ). These results indicate that COX-2 inhibition is a promising approach to the chemoprevention of cutaneous SCC.
      Whereas SCC are tumors of epithelial origin characterized by the accumulation of UV-induced signature mutations in the tumor suppressor gene p53, BCC arise predominantly in hair follicles and manifest mutations in the tumor suppressor gene ptch, leading to activation of the sonic hedgehog signaling pathway (
      • Gailani M.R.
      • Stahle-Backdahl M.
      • Leffell D.J.
      • et al.
      The role of the human homologue of Drosophila patched in sporadic basal cell carcinomas.
      ;
      • Bonifas J.M.
      • Pennypacker S.
      • Chuang P.T.
      • et al.
      Activation of expression of hedgehog target genes in basal cell carcinomas.
      ). Note that although COX-2 expression is a characteristic feature of SCC, it is undetectable in both human and murine BCC (
      • An K.P.
      • Athar M.
      • Tang X.
      • et al.
      Cyclooxygenase-2 expression in murine and human nonmelanoma skin cancers: Implications for therapeutic approaches.
      ). Furthermore, COX-2 inhibitors afford less protection against UVB-induced BCC development in Ptch1+/- mice, a murine model of BCC carcinogenesis.
      Our unpublished observations and E. Epstein, personal communication.
      1Our unpublished observations and E. Epstein, personal communication.
      The differential role of COX-2 in the progression of SCC versus BCC, two NMSC that share UVB irradiation as their major etiologic factor and yet have quite divergent phenotypes, remains ill-defined.
      UVB-mediated increases in COX-2 activity augment levels of the major prostaglandin, prostaglandin E2 (PGE2), in skin keratinocytes (
      • Buckman S.Y.
      • Gresham A.
      • Hale P.
      • Hruza G.
      • Anast J.
      • Masferrer J.
      • Pentland A.P.
      COX-2 expression is induced by UVB exposure in human skin: Implications for the development of skin cancer.
      ). PGE2 acts via four cell-surface seven-transmembrane G-protein-coupled receptors known as E prostanoid (EP) receptors, designated EP1 through EP4. Little information exists about the expression and function of the EP receptors in the skin, including the effect of UV radiation on their expression or their role in the pathogenesis of NMSC. In this study, we assessed the expression pattern of the four EP receptors at both the mRNA and protein levels following acute and chronic UVB irradiation in murine skin. We further assessed their expression during the development of UVB-induced SCC and BCC in relevant murine models. Finally, to assess whether the changes observed during murine skin carcinogenesis also occur in human skin, we have investigated the expression of EP receptors in human actinic keratoses (AK), Bowen's disease, SCC, and BCC.

      Results

      Immunohistochemical distribution of EP receptors in UVB-irradiated murine skin

      To assess the effects of UVB irradiation on the expression and distribution of EP receptors in murine skin, immunohistochemical analysis was performed on non-tumor-bearing skin from SKH-1 hairless mice exposed to 180 mJ per cm2 UVB twice weekly for 40 weeks. The expression pattern of EP receptors in UVB-irradiated non-tumor-bearing skin was compared with that of age-matched non-irradiated control mice. Figure 1 shows the epidermal distribution of EP receptors detected with antibodies specific for each receptor. The EP1 receptor was expressed at low levels in non-irradiated control skin (Figure 1a), but was strongly increased in UVB-irradiated hyperplastic skin, predominantly in the suprabasal layers of the epidermis (Figure 1b). The EP2 receptor demonstrated patchy expression throughout the epidermis in both non-irradiated and chronic UVB-irradiated skin (Figure 1c and d). The EP3 receptor was present at moderate levels in control skin but was undetectable in chronic UVB-irradiated skin (Figure 1e and f). Finally, the EP4 receptor was virtually undetectable in both non-irradiated skin and in UVB-irradiated skin (Figure 1g and h). Similar immunohistochemical analysis following acute UVB exposure (240 mJ per cm2, single exposure) on EP receptor expression in the skin of SKH-1 mice demonstrated that the alterations in the expression of the EP1 and EP3 receptors were detectable as early as eight hours after UVB irradiation (data not shown). These data indicate that the four EP receptors are differentially expressed in normal skin, and that UVB irradiation alters the expression of the EP1 and EP3 receptors, with little overall change in the expression of the EP2 and EP4 receptors.
      Figure thumbnail gr1
      Figure 1Differential expression of E prostanoid receptors in normal and chronic ultraviolet B (UVB)-irradiated murine skin. Immunohistochemical analysis was performed in skin obtained from SKH-1 mice. Chronic UVB-irradiated skin was obtained from mice exposed to 180 mJ per cm2 UVB twice weekly for 30 wk or greater, whereas control skin was obtained from non-irradiated, age-matched controls.

      Differential expression of EP receptors in UVB-induced murine SCC and BCC

      In order to further define alterations in EP receptor expression during the ensuing stages of UVB-induced carcinogenesis, we examined the expression of EP receptors in UVB-induced benign papillomas and SCC of SKH-1 mice (Table I, Figure 2a and b). We found that UVB-induced benign papillomas and SCC, like non-tumor-bearing UVB-irradiated skin, demonstrated intense EP1 receptor expression. In addition, the patchy staining of EP2 seen in both normal and UVB-irradiated skin was also seen in UVB-induced benign papillomas and SCC. Furthermore, very little expression of the EP3 receptor was seen in either benign papillomas or SCC, again similar to the low level of EP3 receptor expression seen in non-tumor-bearing UVB-irradiated skin. We however, observed, moderate levels of EP4 receptor expression in both benign papillomas and SCC, whereas expression of this receptor was nearly undetectable in both unirradiated and non-tumor-bearing UVB-irradiated skin.
      Table IDifferential expression of EP receptors in murine and human NMSC
      EP1EP2EP3EP4
      Murine NMSC
      Papilloma4/4 (100)3/4 (75)2/4 (50)3/4 (75)
      SCC5/5 (100)4/5 (80)1/5 (20)2/5 (40)
      BCC0/4 (0)0/4 (0)0/4 (0)0/4 (0)
      Human NMSC
      AK11/12 (92)8/12 (66)7/12 (58)8/12 (67)
      BD9/10 (90)6/10 (60)7/10 (70)8/10 (80)
      SCC9/12 (75)3/12 (33)5/12 (42)3/12 (25)
      BCC1/14 (7)1/14 (7)2/14 (14)2/14 (14)
      EP receptor expression in murine and human NMSC was assessed using immunohistochemical analysis. Number of tumors with positive staining over total number of tumors analyzed are indicated. Number in parentheses represents percentage with positive staining.EP, E prostanoid; NMSC, non-melanoma skin cancer; SCC, squamous cell carcinoma; BCC, basal cell carcinioma; AK, actinic keratonisis.
      Figure thumbnail gr2
      Figure 2E prostanoid receptors are expressed in murine ultraviolet B (UVB)-induced benign papillomas and squamous cell carcinomas (SCC) but not in UVB-induced basal cell carcinomas (BCC). Immunohistochemical analysis was performed on benign papillomas and SCC from SKH-1 mice and BCC from Ptch1+/- mice subjected to a standard photocarcinogenesis protocol. Areas designated as T represent BCC tumor islands, whereas red arrows indicate positive staining in overlying epidermis. Scale bars (A, B)=100 μm and (C)=50 μm.
      In order to determine whether a similar pattern of EP receptor expression might be seen in UVB-induced murine BCC, we assessed EP receptor expression in UVB-induced BCC in Ptch1+/- mice. In Ptch1+/- mice, all four EP receptors were easily detectable in both normal and chronically UVB-irradiated epidermis (Figure 2c, arrows), but very little receptor expression was detectable within BCC tumor islands (Table I, Figure 2c, areas designated T). The frequency of positive EP receptor expression observed in murine papillomas, SCC, and BCC is presented in Table I. No staining for EP receptors was detected in murine BCC (Table I). The absence of EP receptor expression in murine BCC was also confirmed by indirect immunofluorescence staining with identical results (data not shown). This pattern of EP receptor expression in murine NMSC is in agreement with our prior observations of murine COX-2 expression, in which we observed upregulation of COX-2 in SCC but absence of COX-2 in BCC. These observations as a whole support a differential role for prostaglandins and EP receptors in the pathogenesis of UVB-induced SCC and BCC (
      • An K.P.
      • Athar M.
      • Tang X.
      • et al.
      Cyclooxygenase-2 expression in murine and human nonmelanoma skin cancers: Implications for therapeutic approaches.
      ).

      Quantitative analysis of EP receptor mRNA in UVB-induced murine papillomas and SCC

      To assess whether alterations in EP receptor expression correlate with mRNA levels in UVB-induced SCC, we performed semi-quantitative RT-PCR on total RNA isolated from non-irradiated, age-matched control skin and UVB-induced papillomas and SCC in SKH-1 mice. Five tissue samples from each group were analyzed. Consistent with the immunohistochemical data, EP1 receptor mRNA was undetecTable In non-irradiated skin, but increased sequentially in papillomas and SCC (Figure 3a). Four of five SCC analyzed showed a substantial increase in EP1 mRNA level, with an overall 4.5-fold induction as analyzed by densitometric scanning (Figure 3b). EP2 receptor mRNA was present in non-irradiated control skin and further increased in papillomas and SCC, showing a similar but less dramatic pattern of induction as EP1 receptor mRNA. EP3 receptor mRNA, in parallel to the expression pattern seen using immunohistochemical analysis, was present in normal skin but became virtually undetecTable In UVB-induced papillomas and SCC. Finally, we found that EP4 receptor mRNA level were variable but remained largely unchanged throughout UVB-induced tumorigenesis. Densitometric analysis revealed that these alterations in mRNA expression levels were statistically significant for the EP1, EP2, and EP3 receptors, but not for the EP4 receptor (Figure 3b). These results are physiologically consistent with our immunohistochemical results in UVB-induced papillomas and SCC (Figure 2a and b).
      Figure thumbnail gr3
      Figure 3Differential expression of E prostanoid receptor mRNA in ultraviolet B (UVB)-induced benign papillomas and squamous cell carcinomas (SCC) in SKH-1 mice. (A) RT-PCR analysis was performed on five samples each of control skin, benign papillomas, and SCC harvested from SKH-1 mice subjected to a standard photocarcinogenesis protocol. Unirradiated mice served as controls. (B) Densitometric analysis demonstrated that these changes were statistically significant. *p-value <0.05 when compared with normal skin using Student's t test.

      Immunohistochemical analysis of EP receptor expression in human NMSC

      We next sought to determine whether the alterations in EP receptor expression observed during the growth of UVB-induced murine NMSC also occur in human NMSC. In humans, AK are precursor lesions in the development of SCC, whereas Bowen's disease is thought to represent SCC in situ. We analyzed the immunohistochemical distribution of EP receptors in human NMSC tissue samples, including AK, Bowen's disease, SCC, and BCC. All tumors examined were sporadic tumors occurring on sun-exposed sites.
      We found that all four EP receptors were expressed in both AK and Bowen's disease, with expression of the EP4 receptor quite patchy and often limited to the suprabasal layers in both lesions (Table I, Figure 4a and b). As shown in Figure 4c, we also found that all four receptors are expressed in human SCC. This was in contrast to our observations in murine skin, in which we observed very little EP3 receptor expression in UVB-induced benign papillomas and SCC. Consistent with our murine studies, however, very little EP receptor expression was detecTable In human BCC tumor islands (Table I, Figure 4d, areas designated T), despite intense staining in the epidermal layer of the overlying skin (Figure 4d, arrows). We obtained the confirmed paraffin-embedded sections of human AK, BD, SCC, and BCC, and assessed them for immunohistochemical staining for the EP receptors (Table I). Consistent with the lack of EP receptor staining observed in murine BCC, only one or two out of fourteen human BCC were positive for any EP receptor (Table I). The low frequency of EP receptor staining observed in all human BCC tissue specimens analyzed confirms the near absence of EP receptor expression in human BCC specimens.
      Figure thumbnail gr4
      Figure 4E prostanoid receptors are expressed in human actinic keratosis (AK), Bowen's disease, and squamous cell carcinomas (SCC), but not in basal cell carcinomas (BCC). Immunohistochemical analysis was performed on histologic sections of sporadic human AK, Bowen's disease, SCC, and BCC from sun-exposed sites. Areas designated as T represent BCC tumor islands, whereas red arrows indicate positive staining in overlying epidermis. Scale bars (A, D)=100 μm and (B, C)=50 μm.

      Quantitative analysis of EP receptor mRNA expression in human NMSC

      In order to determine whether the differential expression of EP receptors in human NMSC might be attributable to differences at the mRNA expression level, we obtained fresh frozen tissue specimens for the analysis of relative EP receptor mRNA expression. mRNA levels of all four EP receptors were measured by real-time PCR in human SCC and BCC and compared with that in adjacent non-tumor-bearing skin from the same site. All tumor samples were obtained from sun-exposed areas of the head and neck. We observed significantly enhanced expression of the EP1, EP2, and EP4 receptor mRNA in human SCC as compared with adjacent non-tumor-bearing skin (Figure 5). mRNA levels of the EP3 receptor, however were largely unchanged as compared with the adjacent non-tumor-bearing skin. In addition, mRNA levels for the four EP receptors in human BCC were not significantly different from that observed in the adjacent skin. These results suggest that, in part, alterations of EP1, EP2, and EP4 receptor expression during NMSC development may occur at the mRNA expression level, and further substantiate a differential role for these receptors in the progression of UVB-induced SCC and BCC.
      Figure thumbnail gr5
      Figure 5E prostanoid (EP) receptor mRNA levels are significantly increased in human squamous cell carcinomas (SCC) but not basal cell carcinomas (BCC) as compared with adjacent non-tumor skin. Quantitative assessment of EP receptor mRNA levels using real-time PCR in five human SCC and BCC tumor samples as compared with adjacent non-tumor-bearing skin from the same patient, with non-tumor-bearing skin normalized to value 1. *p-value <0.05 when compared with adjacent tumor-free skin using Student's paired t test.

      Discussion

      Although COX-2 over-expression and enhanced PGE2 production are known to play a critical role in augmenting the growth of various epithelial neoplasms (
      • Dannenberg A.J.
      • Altorki N.K.
      • Boyle J.O.
      • Dang C.
      • Howe L.R.
      • Weksler B.B.
      • Subbaramaiah K.
      Cyclo-oxygenase 2: A pharmacological target for the prevention of cancer.
      ;
      • Lee J.L.
      • Mukhtar H.
      • Bickers D.R.
      • Kopelovich L.
      • Athar M.
      Cyclooxygenases in the skin: Pharmacological and toxicological implications.
      ), the events downstream of enhanced PGE2 secretion that drive the carcinogenic process remain largely undefined. In this study, we assessed the expression of EP receptors during the progression of UVB-induced SCC and BCC tumorigenesis. In our murine studies, we observed an increase in the level of EP1 receptor expression and concurrent decrease in the EP3 receptor expression level in UVB-induced benign papillomas and SCC in murine skin. These alterations were observed at both the immunohistochemical and mRNA level, and these results suggest that the expression of these receptors is an important consequence of the enhanced COX-2 signaling that characterizes the progressive growth of these epithelial neoplasms.
      The expression profile of EP receptors, like COX-2 molecules, is also altered during development of other epithelial tumors, including those of the breast, cervix, and colon (
      • Sales K.J.
      • Katz A.A.
      • Davis M.
      • et al.
      Cyclooxygenase-2 expression and prostaglandin E(2) synthesis are up-regulated in carcinomas of the cervix: A possible autocrine/paracrine regulation of neoplastic cell function via EP2/EP4 receptors.
      ;
      • Takafuji V.
      • Lublin D.
      • Lynch K.
      • Roche J.K.
      Mucosal prostanoid receptors and synthesis in familial adenomatous polyposis.
      ,
      • Takafuji V.A.
      • Evans A.
      • Lynch K.R.
      • Roche J.K.
      PGE(2) receptors and synthesis in human gastric mucosa: Perturbation in cancer.
      ;
      • Chang S.H.
      • Liu C.H.
      • Conway R.
      • et al.
      Role of prostaglandin E2-dependent angiogenic switch in cyclooxygenase 2-induced breast cancer progression.
      ). During development of SCC of the cervix, increased expression of EP2 and EP4 receptor mRNA is observed in addition to COX-2 upregulation (
      • Sales K.J.
      • Katz A.A.
      • Davis M.
      • et al.
      Cyclooxygenase-2 expression and prostaglandin E(2) synthesis are up-regulated in carcinomas of the cervix: A possible autocrine/paracrine regulation of neoplastic cell function via EP2/EP4 receptors.
      ). Similarly, during murine gastrointestinal tumorigenesis, increased expression of EP2 and EP4 receptor mRNA and decreased expression of EP3 receptor mRNA have been observed (
      • Sonoshita M.
      • Takaku K.
      • Sasaki N.
      • et al.
      Acceleration of intestinal polyposis through prostaglandin receptor EP2 in Apc(Delta 716) knockout mice.
      ).
      Our observation that EP receptors are differentially expressed during the development of UVB induced SCC in both experimental animals and humans suggests that these receptors are important in mediating the effects of prostaglandins in UVB-induced SCC. EP2 and EP4 receptor-dependent signaling leads to an increase in adenylyl cyclase activity and resultant cyclic AMP (cAMP) levels, whereas EP3 receptor-dependent signaling is associated with a decrease in cAMP levels (
      • Narumiya S.
      • Sugimoto Y.
      • Ushikubi F.
      Prostanoid receptors: Structures, properties, and functions.
      ). Given that signaling downstream of the EP3 receptor acts to oppose EP2- and EP4-mediated increases in cAMP activity, the observation that the EP3 receptor alone is significantly decreased in murine SCC and gastrointestinal tumors, along with the observed increase in EP2 and EP4 receptors in cervical and gastrointestinal cancers, suggests the possibility that altered cAMP signaling may be a downstream mediator of the COX-2 overexpression that characterizes SCC progression. Indeed, Sales et al have demonstrated that application of PGE2 to cervical SCC tissue results in a 10-fold increase in cAMP signaling as compared with non-malignant cervical tissue (
      • Sales K.J.
      • Katz A.A.
      • Davis M.
      • et al.
      Cyclooxygenase-2 expression and prostaglandin E(2) synthesis are up-regulated in carcinomas of the cervix: A possible autocrine/paracrine regulation of neoplastic cell function via EP2/EP4 receptors.
      ).
      EP receptors are known to undergo internalization with agonist-induced alterations of subcellular localization (
      • Desai S.
      • April H.
      • Nwaneshiudu C.
      • Ashby B.
      Comparison of agonist-induced internalization of the human EP2 and EP4 prostaglandin receptors: Role of the carboxyl terminus in EP4 receptor sequestration.
      ;
      • Hasegawa H.
      • Katoh H.
      • Fujita H.
      • Mori K.
      • Negishi M.
      Receptor isoform-specific interaction of prostaglandin EP3 receptor with muskelin.
      ;
      • Desai S.
      • Ashby B.
      Agonist-induced internalization and mitogen-activated protein kinase activation of the human prostaglandin EP4 receptor.
      ). We observed variable subcellular staining patterns in various lesions throughout NMSC development in both murine and human skin. Pronounced cytoplasmic or membrane localization patterns may indicate different activation states of these receptors but the functional implications of these variations in the carcinogenic process remain to be defined.
      In contrast to our observations in murine and human SCC and SCC precursors, we noted an absence of EP receptor expression in murine and human BCC, and a lack of significant alteration in any EP receptor mRNA expression level in human BCC as compared with adjacent non-tumor-bearing skin. The low level of EP receptor expression in murine BCC correlates well with our prior observation that these BCC lack COX-2 expression (
      • An K.P.
      • Athar M.
      • Tang X.
      • et al.
      Cyclooxygenase-2 expression in murine and human nonmelanoma skin cancers: Implications for therapeutic approaches.
      ). This result emphasizes the known differences in signaling pathways and phenotypes of these two types of NMSC, and suggests that the COX-2 signaling pathway may be less pivotal in the development of these BCC than in SCC.
      In summary, our results demonstrate differential EP receptor expression during the development of UVB-induced SCC in both murine and human skin, whereas these receptors are largely absent during the growth of BCC. These results suggest differential roles for PGE2 signaling in SCC and BCC, and further substantiate the divergent pathogenesis of these neoplasms. A differential role for prostaglandins in these two types of UVB-induced NMSC may also explain why COX-2 inhibitors, although potent as chemopreventive agents in SCC development, are largely ineffective in inhibiting the growth of UVB-induced BCC in Ptch1+/- mice, with clinical implications regarding the potential of COX-2 inhibition in the chemoprevention of these two NMSC. Further studies are needed to define how PGE2 levels and EP receptor expression may influence the pathogenesis and the invasiveness of NMSC.

      Materials and Methods

      Tissue samples

      Animals

      Female SKH-1 hairless mice were obtained at the age of 4–6 wk from Charles River Laboratories (Kingston, New York). They were allowed to acclimate to the animal facility for one week and were housed under standard conditions. All animals were fed ad libitum with water and Purina Laboratory Chow 5001 diet (Ralston-Purina, St Louis, Missouri). Ptch1+/- mice were obtained from our colony at Columbia University. All animal experiments were conducted with approval from the Columbia University Institutional Animal Care and Use Committee, and were in accordance with their guidelines.

      UV light source

      UV irradiation unit (Daavlin, Bryan, Ohio) equipped with an electronic controller to regulate dosage was used. This UV source consisted of eight FS72T12-UVB-HO lamps that emit UVB (290–320 nm, 75%–80% of total energy). We used a Kodacel cellulose film (Kodacel TA401/407 Rochester, New York) to eliminate UVC radiation. The dose of UVB was quantified with a UVB Spectra 305 Dosimeter (Daavlin). The radiation was further calibrated with an IL1700 Research Radiometer/Photometer (International Light, Newburyport, Massachusetts). The radiation source to target distance was maintained at 30 cm. No measurable increase in ambient temperature was detected during the irradiation procedure.

      Photocarcinogenesis protocol

      SKH-1 hairless mice were irradiated with 180 mJ per cm2 of UVB twice weekly for 40 wk, in accordance with our photocarcinogenesis protocol, which has been previously described for this mouse strain (
      • Athar M.
      • An K.P.
      • Morel K.D.
      • et al.
      Ultraviolet B (UVB)-induced cox-2 expression in murine skin: An immunohistochemical study.
      ;
      • An K.P.
      • Athar M.
      • Tang X.
      • et al.
      Cyclooxygenase-2 expression in murine and human nonmelanoma skin cancers: Implications for therapeutic approaches.
      ). At that time, the animals were sacrificed, their dorsal skin was removed, and non-tumor skin and tumors were harvested. Control skin samples were obtained from age-matched, non-irradiated SKH-1 mice, whereas chronic UVB skin samples were harvested from non-tumor-bearing areas of the irradiated animals. Ptch1+/- heterozygous knockout mice have been developed by deleting exons 1 and 2 and inserting the LacZ gene at the deletion site as described previously (
      • Aszterbaum M.
      • Epstein J.
      • Oro A.
      • Douglas V.
      • LeBoit P.E.
      • Scott M.P.
      • Epstein Jr, E.H.
      Ultraviolet and ionizing radiation enhance the growth of BCCs and trichoblastomas in patched heterozygous knockout mice.
      ), and were obtained from our colony. In accordance with the photocarcinogenesis protocol previously developed for this mouse strain (
      • Aszterbaum M.
      • Epstein J.
      • Oro A.
      • Douglas V.
      • LeBoit P.E.
      • Scott M.P.
      • Epstein Jr, E.H.
      Ultraviolet and ionizing radiation enhance the growth of BCCs and trichoblastomas in patched heterozygous knockout mice.
      ;
      • Athar M.
      • Li C.
      • Tang X.
      • et al.
      Inhibition of smoothened signaling prevents ultraviolet B-induced basal cell carcinomas through regulation of Fas expression and apoptosis.
      ;
      • Tang X.
      • Kim A.L.
      • Feith D.J.
      • et al.
      Ornithine decarboxylase is a target for chemoprevention of basal and squamous cell carcinomas in Ptch1+/- mice.
      ), Ptch1+/- mice were UVB irradiated with 240 mJ per cm2 UVB three times per week for 52 wk, after which the animals were sacrificed, their dorsal skin was removed, and non-tumor-bearing skin and tumors were harvested.

      Human tissue samples

      Human tissue sections were collected from the Dermatopathology Laboratory of the Columbia Department of Dermatology in compliance with the Declaration of Helsinki Guidelines for use of human tissues/subjects. Tissue samples were not associated with patient names, hospital numbers, or other identifying information. The Columbia University Medical Center Institutional Review Board approved the use of this tissue under IRB exemption (#X0958). A total of 48 human tissue sections were examined including 12 AK, 10 Bowen's disease, 12 SCC, and 14 BCC. For mRNA analysis, fresh tissue specimens of tumor and adjacent non-tumor skin were obtained at the time of surgical excision, placed immediately on dry ice, and stored at -80°C until use. A total of five SCC tissue pairs (tumor and non-tumor adjacent skin) and five BCC tissue pairs were analyzed.

      Immunohistochemistry

      EP1, EP2, EP3, and EP4 receptor polyclonal rabbit antibodies and blocking peptides were obtained from Cayman Chemical (Ann Arbor, Michigan) and were used for the immunohistochemical studies. Skin tissue samples were fixed in cold formalin solution overnight at 4°C. The tissues were washed for 30 min each sequentially in 1 × PBS and 0.9% NaCl/alcohol solutions containing 30%, 50%, and 70% ethanol, and then embedded in paraffin wax and sectioned onto slides. Sections were then subjected to deparaffinization in xylene and hydrated through graded alcohols, heated to 95°C in antigen unmasking solution (Vector Laboratories, Burlingame, California), and allowed to cool for 20 min. Endogenous peroxidase activity was quenched in 0.3% hydrogen peroxide in methanol for 30 min. Staining was carried out using the Vectastain Rabbit IgG ABC kit (Vector Laboratories) according to the manufacturer's instructions with modifications as follows. Sections were incubated in blocking buffer for one hour at room temperature (1.5% goat normal serum in PBS), and then incubated overnight at 4°C in primary antibody against the EP receptors (Cayman Chemical) at a dilution of 5 μg per mL in PBS. Sections were incubated with biotinylated secondary antibody at a dilution of 7.5 μg per mL in PBS. Staining was visualized by the reaction of avidin–biotin peroxidase complex with diaminobenzidine (Sigma Chemical, St Louis, Missouri). Sections were counterstained with Harris hematoxylin (Sigma Diagnostics, St Louis, Missouri), dehydrated through graded alcohols and xylene, and mounted using Permount (Fisher Scientific, Fair Lawn, New Jersey). Sections presented are representative of a minimum of four sections. Data inTable I reflect the number of sections demonstrating positive staining when assessed by a single blinded observer.
      To confirm the specificity of the EP receptor antibodies, competing peptide assays were performed. For these studies, the primary antibody was pre-incubated with 500 μg per mL peptide in PBS prior to incubation with tissue sections. All other steps were carried out as described above. No staining was seen in the presence of peptide, confirming the specificity of these antibodies for their receptors.

      Semi-quantitative RT-PCR

      Total RNA was isolated from 20 mg of murine skin following separation of adherent subcutaneous fat, using the RNeasy Mini kit (Qiagen, Valencia, California) according to the manufacturer's instructions. Concentration and purity were assessed by measurement of absorbance at 260 and 280 nm using a Beckman DU520 Spectrophotometer (Beckman Coulter, Fullerton, California). In all, 2 μg of RNA was reverse transcribed in a total volume of 40 μL with 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 2.5 mM MgCl2, 0.5 mM dNTP, 2.5 μM oligo d(T)16 primer, 20 U RNase inhibitor, and 50 U murine leukemia virus reverse transcriptase (Roche Applied Science, Indianapolis, Indiana) under the following conditions: room temperature for 10 min, 42°C for 15 min, 99°C for 5 min, and 5°C for 5 min. The resulting cDNA was used to perform PCR using a Hybaid PCR Express machine (Thermo Electron Corporation, Waltham, Massachusetts). Reactions were performed in 25 μL volume with 2 μL of cDNA, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 2 mM MgCl2, 0.4 mM dNTP, 400 nM forward primer, 400 nM reverse primer, and 2.5 U Taq polymerase (Applied Biosystems, Foster City, CA) according to the manufacturer's specifications. PCR was performed under the following conditions:
      EP1: 95°C for 30 s, 62°C for 30 s, 70°C for 45 s, for 35 cycles, and then 70°C for 10 min. EP2 and EP4: 95°C for 30 s, 62°C for 30 s, 70°C for 45 s, for 30 cycles, and then 70°C for 10 min. EP3: 95°C for 30 s, 61°C for 30 s, 70°C for 45 s, for 40 cycles, and then 70°C for 10 min. Primer sequences were as follows: EP1 forward: 5′-TTAACCTGAGCCTAGCGGAT-3′; EP1 reverse: 5′-CGCTGAGCGTATTGCACACTA-3′; EP2 forward: 5′-GTGGCCCTGGCTCCCGAAAGTC-3′; EP2 reverse: 5′-GGCAAGGAGCATATGGCGAAGGTG-3′; EP3 forward: 5′-CCGGGCACGTGGTGCTTCAT-3′; EP3 reverse: 5′-TAGCAGCAGATAAACCCAGG-3′; EP4 forward: 5′-TTCCGCTCGTGGTGCGAGTGTTC-3′; EP4 reverse: 5′-GAGGTGGTGTCTGCTTGGGTCAG-3′; and B-actin forward: 5′-GGCATCCTCACCCTGAAGTA-3′; B-actin reverse: 5′-GCCAGATTTTCTCCATGTCG-3′. Sequence identity of each PCR product was confirmed by direct sequencing. Samples were subjected to electrophoresis on a 1% agarose gel with 0.5 μg per mL ethidium bromide, and quantitatively assessed by densitometry.

      Real-time PCR

      Total RNA was isolated from 20 mg of human skin tissue, following separation of adherent subcutaneous fat, using TRIzol reagent (Invitrogen, Carlsbad, CA) followed an RNA cleanup protocol using the Qiagen RNeasy mini-kit (Qiagen) according to the manufacturers' instructions. Concentration and purity were assessed by measurement of absorbance at 260 and 280 nm using a Beckman DU520 Spectrophotometer (Beckman Coulter, Fullerton, California). 2 μg of total RNA was used for first-strand cDNA synthesis using random hexamers. Resulting cDNA was used to perform real-time PCR using a Bio-Rad iCycler Machine (Bio-Rad Laboratories, Hercules, CA). Reactions were performed in a 96-well plate (Bio-Rad Laboratories) in a 20 μL volume with 2 μL of cDNA, 400 nM forward primer, 400 nM reverse primer, and SYBR Green PCR Master Mix at a 1 × dilution (Applied Biosystems). Each sample was amplified in duplicate. The plate was heated to 95°C for 10 min, and then amplification was carried out for 40 cycles (94°C for 30 s, 62°C for 30 s, and 72°C for 30 s). Cycling was followed by melt curve analysis (55°C–95°C) to confirm the amplification of specific PCR products and the absence of non-specific amplification or primer dimer. Primer sequences were as follows: EP1 forward: 5′-TGGGCCAGCTTGTCGGTA-3′; EP1 reverse: 5′-AGGGCCA CCAACACCAG-3′; EP2 forward: 5′-TGGGTCTTTGCCA TCCTT-3′; EP2 reverse: 5′-TCCGACAACAGAGGACTG-3′; EP3 forward: 5′-CAGCTTATGGGGATCATG-3′; EP3 reverse: 5′-TCCGTG TGTGTCTTGCAG-3′; EP4 forward: 5′-TGTGAACCCCATCCTAGA-3′; and EP4 reverse: 5′-GCAGAAGAGGCATTTGAT-3′. 18S RNA was used as an internal control (
      • Schmittgen T.D.
      • Zakrajsek B.A.
      Effect of experimental treatment on housekeeping gene expression: Validation by real-time, quantitative RT-PCR.
      ), with the primer sequence as follows: 18S forward: 5′-GTAACCCGTTGAACCCCATT-3′; 18S reverse: 5′-CCATCCAATCGGTAGTAGCG-3′. Real-time PCR for 18S was carried out as above, except using an annealing temperature of 55°C.
      Primers were confirmed to be specific based on a single peak on melting curve analysis and sequence identity was confirmed by direct sequencing. The absence of genomic DNA was confirmed by performing direct PCR of total RNA with no amplification detected. Results for each sample were calculated using the mean of duplicate samples using the formula:(EtargetCtcontrolCttreatment))/(EinternalcontrolCtcontrol-Cttreatment)), where E is the calculated efficiency of amplification and Ct is the threshold cycle of amplification (
      • Pfaffl M.W.
      A new mathematical model for relative quantification in real-time RT-PCR.
      ). Calculated efficiency for EP1, EP2, EP3, and EP4 was 1.95, 1.96, 1.98, and 1.98, respectively, using these primers.

      ACKNOWLEDGMENTS

      We are grateful to Dr Desiree Ratner for her assistance in obtaining the human tissue samples, and to Dr David N. Silvers for his expert assistance in analyzing the human dermatopathology specimens. We also appreciate the assistance of Drs Baoheng Du, Erik Cohen, and Andrew J. Dannenberg in carrying out analyses of EP receptor mRNA. This work was supported by the NIH Grants CA-10106-01, NO1-CN-35105, N01-CN-15109, N01-CN-35006-72, N01-CN-15011-72, RO1 CA-97249-01, NIH/NCI U19 CA 81888, NIAMS-P30 AR 44535, T32 CA09685 (E. G. C.), K01-AR048582 (A.L.K.) and the Doris Duke Charitable Foundation.

      References

        • An K.P.
        • Athar M.
        • Tang X.
        • et al.
        Cyclooxygenase-2 expression in murine and human nonmelanoma skin cancers: Implications for therapeutic approaches.
        Photochem Photobiol. 2002; 76: 73-80
        • Aszterbaum M.
        • Epstein J.
        • Oro A.
        • Douglas V.
        • LeBoit P.E.
        • Scott M.P.
        • Epstein Jr, E.H.
        Ultraviolet and ionizing radiation enhance the growth of BCCs and trichoblastomas in patched heterozygous knockout mice.
        Nat Med. 1999; 5: 1285-1291
        • Athar M.
        • An K.P.
        • Morel K.D.
        • et al.
        Ultraviolet B (UVB)-induced cox-2 expression in murine skin: An immunohistochemical study.
        Biochem Biophys Res Commun. 2001; 280: 1042-1047
        • Athar M.
        • Li C.
        • Tang X.
        • et al.
        Inhibition of smoothened signaling prevents ultraviolet B-induced basal cell carcinomas through regulation of Fas expression and apoptosis.
        Cancer Res. 2004; 64: 7545-7552
        • Bonifas J.M.
        • Pennypacker S.
        • Chuang P.T.
        • et al.
        Activation of expression of hedgehog target genes in basal cell carcinomas.
        J Invest Dermatol. 2001; 116: 739-742
        • Buckman S.Y.
        • Gresham A.
        • Hale P.
        • Hruza G.
        • Anast J.
        • Masferrer J.
        • Pentland A.P.
        COX-2 expression is induced by UVB exposure in human skin: Implications for the development of skin cancer.
        Carcinogenesis. 1998; 19: 723-729
        • Chang S.H.
        • Liu C.H.
        • Conway R.
        • et al.
        Role of prostaglandin E2-dependent angiogenic switch in cyclooxygenase 2-induced breast cancer progression.
        Proc Natl Acad Sci USA. 2004; 101: 591-596
        • Chun K.S.
        • Kim S.H.
        • Song Y.S.
        • Surh Y.J.
        Celecoxib inhibits phorbol ester-induced expression of COX-2 and activation of AP-1 and p38 MAP kinase in mouse skin.
        Carcinogenesis. 2004; 25: 713-722
        • Dannenberg A.J.
        • Altorki N.K.
        • Boyle J.O.
        • Dang C.
        • Howe L.R.
        • Weksler B.B.
        • Subbaramaiah K.
        Cyclo-oxygenase 2: A pharmacological target for the prevention of cancer.
        Lancet Oncol. 2001; 2: 544-551
        • Desai S.
        • April H.
        • Nwaneshiudu C.
        • Ashby B.
        Comparison of agonist-induced internalization of the human EP2 and EP4 prostaglandin receptors: Role of the carboxyl terminus in EP4 receptor sequestration.
        Mol Pharmacol. 2000; 58: 1279-1286
        • Desai S.
        • Ashby B.
        Agonist-induced internalization and mitogen-activated protein kinase activation of the human prostaglandin EP4 receptor.
        FEBS Lett. 2001; 501: 156-160
        • Fischer S.M.
        • Lo H.H.
        • Gordon G.B.
        • Seibert K.
        • Kelloff G.
        • Lubet R.A.
        • Conti C.J.
        Chemopreventive activity of celecoxib, a specific cyclooxygenase-2 inhibitor, and indomethacin against ultraviolet light-induced skin carcinogenesis.
        Mol Carcinog. 1999; 25: 231-240
        • Gailani M.R.
        • Stahle-Backdahl M.
        • Leffell D.J.
        • et al.
        The role of the human homologue of Drosophila patched in sporadic basal cell carcinomas.
        Nat Genet. 1996; 14: 78-81
        • Harris R.E.
        • Beebe-Donk J.
        • Schuller H.M.
        Chemoprevention of lung cancer by non-steroidal anti-inflammatory drugs among cigarette smokers.
        Oncol Rep. 2002; 9: 693-695
        • Harris R.E.
        • Kasbari S.
        • Farrar W.B.
        Prospective study of nonsteroidal anti-inflammatory drugs and breast cancer.
        Oncol Rep. 1999; 6: 71-73
        • Hasegawa H.
        • Katoh H.
        • Fujita H.
        • Mori K.
        • Negishi M.
        Receptor isoform-specific interaction of prostaglandin EP3 receptor with muskelin.
        Biochem Biophys Res Commun. 2000; 276: 350-354
        • Hasegawa H.
        • Katoh H.
        • Yamaguchi Y.
        • Nakamura K.
        • Futakawa S.
        • Negishi M.
        Different membrane targeting of prostaglandin EP3 receptor isoforms dependent on their carboxy-terminal tail structures.
        FEBS Lett. 2000; 473: 76-80
        • Lee J.L.
        • Mukhtar H.
        • Bickers D.R.
        • Kopelovich L.
        • Athar M.
        Cyclooxygenases in the skin: Pharmacological and toxicological implications.
        Toxicol Appl Pharmacol. 2003; 192: 294-306
        • Marks R.
        An overview of skin cancers. Incidence and causation.
        Cancer. 1995; 75: 607-612
        • Marnett L.J.
        • Rowlinson S.W.
        • Goodwin D.C.
        • Kalgutkar A.S.
        • Lanzo C.A.
        Arachidonic acid oxygenation by COX-1 and COX-2. Mechanisms of catalysis and inhibition.
        J Biol Chem. 1999; 274: 22903-22906
        • Muller-Decker K.
        • Neufang G.
        • Berger I.
        • Neumann M.
        • Marks F.
        • Furstenberger G.
        Transgenic cyclooxygenase-2 overexpression sensitizes mouse skin for carcinogenesis.
        Proc Natl Acad Sci USA. 2002; 99: 12483-12488
        • Narumiya S.
        • Sugimoto Y.
        • Ushikubi F.
        Prostanoid receptors: Structures, properties, and functions.
        Physiol Rev. 1999; 79: 1193-1226
        • Neufang G.
        • Furstenberger G.
        • Heidt M.
        • Marks F.
        • Muller-Decker K.
        Abnormal differentiation of epidermis in transgenic mice constitutively expressing cyclooxygenase-2 in skin.
        Proc Natl Acad Sci USA. 2001; 98: 7629-7634
        • Pentland A.P.
        • Schoggins J.W.
        • Scott G.A.
        • Khan K.N.
        • Han R.
        Reduction of UV-induced skin tumors in hairless mice by selective COX-2 inhibition.
        Carcinogenesis. 1999; 20: 1939-1944
        • Pfaffl M.W.
        A new mathematical model for relative quantification in real-time RT-PCR.
        Nucleic Acids Res. 2001; e45: 2002-2007
        • Sales K.J.
        • Katz A.A.
        • Davis M.
        • et al.
        Cyclooxygenase-2 expression and prostaglandin E(2) synthesis are up-regulated in carcinomas of the cervix: A possible autocrine/paracrine regulation of neoplastic cell function via EP2/EP4 receptors.
        J Clin Endocrinol Metab. 2001; 86: 2243-2249
        • Schapira D.V.
        • Theodossiou C.
        • Lyman G.H.
        The effects of NSAIDs on breast cancer prognostic factors.
        Oncol Rep. 1999; 6: 433-435
        • Schmittgen T.D.
        • Zakrajsek B.A.
        Effect of experimental treatment on housekeeping gene expression: Validation by real-time, quantitative RT-PCR.
        J Biochem Biophys Methods. 2000; 46: 69-81
        • Smalley W.
        • Ray W.A.
        • Daugherty J.
        • Griffin M.R.
        Use of nonsteroidal anti-inflammatory drugs and incidence of colorectal cancer: A population-based study.
        Arch Intern Med. 1999; 159: 161-166
        • Sonoshita M.
        • Takaku K.
        • Sasaki N.
        • et al.
        Acceleration of intestinal polyposis through prostaglandin receptor EP2 in Apc(Delta 716) knockout mice.
        Nat Med. 2001; 7: 1048-1051
        • Takafuji V.
        • Lublin D.
        • Lynch K.
        • Roche J.K.
        Mucosal prostanoid receptors and synthesis in familial adenomatous polyposis.
        Histochem Cell Biol. 2001; 116: 171-181
        • Takafuji V.A.
        • Evans A.
        • Lynch K.R.
        • Roche J.K.
        PGE(2) receptors and synthesis in human gastric mucosa: Perturbation in cancer.
        Prostaglandins Leukot Essent Fatty Acids. 2002; 66: 71-81
        • Tang X.
        • Kim A.L.
        • Feith D.J.
        • et al.
        Ornithine decarboxylase is a target for chemoprevention of basal and squamous cell carcinomas in Ptch1+/- mice.
        J Clin Invest. 2004; 113: 867-875
        • Tiano H.F.
        • Loftin C.D.
        • Akunda J.
        • et al.
        Deficiency of either cyclooxygenase (COX)-1 or COX-2 alters epidermal differentiation and reduces mouse skin tumorigenesis.
        Cancer Res. 2002; 62: 3395-3401