Roles of keratinocyte inflammation in oral cancer: regulating the prostaglandin E2, interleukin-6 and TNF-alpha production of oral epithelial cells by areca nut extract and arecoline

Betel quid (BQ) chewing is an etiologic factor of oral cancer and submucus fibrosis (OSF). Keratinocyte inflammation is crucial for the pathogenesis of cancer and tissue fibrosis. We found that areca nut (AN) extract (100-400 micro g/ml) induced PGE2 production by KB cells by 2.34- to 23.1-fold and also TNF-alpha production by gingival keratinocytes (GK). Arecoline (0.2-1.2 mM) elevated PGE2 production by KB cells by 2.5- to 6.1-fold. AN extract (200-400 micro g/ml) also induced IL-6 production by GK (7.5- to 8.4-fold) and KB cells. In contrast, arecoline (0.1-1.2 mM) suppressed IL-6 production by GK and KB cells, with 42-81 and 41-63% inhibition, respectively. A 48 h exposure of GK to 800-1200 micro g/ml AN extract led to 37-69% cell death. Arecoline cytotoxicity to GK was noted at concentrations of 0.8-1.2 mM, which led to 28-38% cell death. AN extract (400-800 micro g/ml) induced Cox-2 and IL-6 mRNA expression and also COX-2 protein production by KB cells. IL-6 (5-100 ng/ml) suppressed GK growth by 20-33%, but enhanced oral fibroblast (OMF) and KB cell growth. PGE2 (0.05-5 micro g/ml) and anti-IL-6 antibody (ab) (50-1000 ng/ml) showed little effect on GK and KB cell growth. Incubation of GK and KB cells with aspirin, anti-IL-6 ab and anti-TNF-alpha ab showed little effect on arecoline- and AN-induced cytotoxicity, cell cycle arrest and apoptosis. Exposure to anti-TNF-alpha ab slightly affected arecoline- and AN-modulated PGE2 and IL-6 production by GK and KB cells. Arecoline- and AN-conditioned medium decreased phytohemagglutinin-mediated CD4+ and CD8+ T cell activation. These results indicate that BQ chewing contributes to the pathogenesis of cancer and OSF by impairing T cell activation and by induction of PGE2, TNF-alpha and IL-6 production, which affect oral mucosal inflammation and growth of OMF and oral epithelial cells.     

Areca nut extract and arecoline induced the cell cycle arrest but not apoptosis of cultured oral KB epithelial cells: association of glutathione, reactive oxygen species and mitochondrial membrane potential

There are 600 million betel quid (BQ) chewers in the world. BQ chewing is a major etiologic factor of oral cancer. Areca nut (AN) and arecoline may inhibit the growth of oral mucosal fibroblasts (OMF) and keratinocytes. In this study, AN extract (100-800 microg/ml) and arecoline (20-120 microM) inhibited the growth of oral KB cells by 36-90 and 15-75%, respectively. Exposure to arecoline (> 0.2 mM) for 24 h induced G(2)/M cell cycle arrest of OMF and KB cells. Areca nut extract (> 400 microg/ml) also induced G(2)/M arrest of KB cells, being preceded by S-phase arrest at 7-h of exposure. No evident sub-G(0)/G(1) peak was noted. Marked retraction and intracellular vacuoles formation of OMF and KB cells were observed. Glutathione (GSH) level, mitochondrial membrane potential (Deltabetam) and H(2)O(2) production of KB cells were measured by flow cytometry. GSH level [indicated by 5-chloromethyl-fluorescein (CMF) fluorescence] was depleted by 24-h exposure of KB cells to arecoline (0.4-1.2 mM) and AN extract (800-1200 microg/ml), with increasing the percentage of cells in low CMF fluorescence. By contrast, arecoline (0.1-1.2 mM) and AN extract (800-1200 microg/ml) induced decreasing and increasing H(2)O(2) production (by 2',7'-dichloro- fluorescein fluorescence), respectively. Hyperpolarization of Deltabetam (increasing of rhodamine uptake) was noted by 24-h exposure of KB cells to arecoline (0.4-1.2 mM) and AN extract (800-1200 microg/ml). AN extract (100- 1200 microg/ml) and arecoline (0.1-1.2 mM) induced little DNA fragmentation on KB cells within 24 h. These results indicate that AN ingredients are crucial in the pathogenesis of oral submucous fibrosis (OSF) and oral cancer by differentially inducing the dysregulation of cell cycle control, Deltabetam, GSH level and intracellular H(2)O(2) production, these events being not coupled with cellular apoptosis.     

Prevention of the areca nut extract-induced unscheduled DNA synthesis of gingival keratinocytes by vitamin C and thiol compounds

There are about 600 million betel quid (BQ) chewers in the world. BQ chewing is the major risk factor of oral cancer in India, Taiwan, South Africa and numerous other countries. Areca nut (AN) extract, the main component of BQ, exerts cytotoxicity and genotoxicity to several types of cells. In the present study, AN extract induced the unscheduled DNA synthesis (UDS) of gingival keratinocytes (GK). Vitamin C, at concentration of 50 and 200 microg/ml prevented the AN-induced UDS by 41 and 56%, respectively. Glutathione (GSH, 1-3 mM) and N-acetyl-L-cysteine (NAC, 1-3 mM) also protected the AN-induced UDS by 89-100 and 76-90%. These preventive effects were not due to cytotoxicity as analyzed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide (MTT) assay. Deferoxamine (20 and 30 mM), an iron chelator and a free radical scavenger, also prevented AN extract induced UDS of GK by 30-55%. On the contrary, banthocuproine (50-200 microM, a copper chelator) and 1,10-phenanthroline (50, 100 microM, a lipid permeable iron chelator), lacked preventive effects. Specific reactive oxygen species scavengers such as dimethyl-sulfoxide (2%), mannitol (10-20 mM), dimethylthiourea (10-20 mM), pyruvate (10 mM), catalase (200 and 400 U/ml), and superoxide dismutase (50 and 200 U/ml) also lacked these preventive effects. Moreover, higher concentrations of H(2)O(2) (0.5-1 mM) inhibited the basal levels of UDS by 19-37%. Interestingly, NAC, GSH, Vitamin C and deferoxamine cannot prevent the AN-induced morphological changes of GK at similar concentrations. These results reveal that AN extract-induced UDS of GK is associated with free radical reactions. Possibly different ingredients of AN is responsible for genotoxicity and cytotoxicity. Vitamin C, GSH and NAC may be potentially used in the future for chemoprevention of BQ chewing related oral mucosal lesions.     

Inhibitory effects of caffeic acid phenethyl ester on the activity and expression of cyclooxygenase-2 in human oral epithelial cells and in a rat model of inflammation.

We investigated the mechanisms by which caffeic acid phenethyl ester (CAPE), a phenolic antioxidant, inhibited the stimulation of prostaglandin (PG) synthesis in cultured human oral epithelial cells and in an animal model of acute inflammation. Treatment of cells with CAPE (2.5 microg/ml) suppressed phorbol ester (12-O-tetradecanoylphorbol-13-acetate; TPA) and calcium ionophore (A23187)-mediated induction of PGE2 synthesis. This relatively low concentration of CAPE did not affect amounts of cyclooxygenase (COX) enzymes. CAPE nonselectively inhibited the activities of baculovirus-expressed hCOX-1 and hCOX-2 enzymes. TPA- and A23187-stimulated release of arachidonic acid from membrane phospholipids was also suppressed by CAPE (4-8 microg/ml). Higher concentrations of CAPE (10-20 microg/ml) suppressed the induction of COX-2 mRNA and protein mediated by TPA. Transient transfections using human COX-2 promoter deletion constructs were performed; the effects of TPA and CAPE were localized to a 124-bp region of the COX-2 promoter. In the rat carrageenan air pouch model of inflammation, CAPE (10-100 mg/kg) caused dose-dependent suppression of PG synthesis. Amounts of COX-2 in the pouch were markedly suppressed by 100 mg/kg CAPE but were unaffected by indomethacin. These data are important for understanding the anticancer and anti-inflammatory properties of CAPE.     

Cyclooxygenase-2 pathway correlates with vascular endothelial growth factor expression and tumor angiogenesis in hepatitis B virus-associated hepatocellular carcinoma

Evidence indicates that cyclooxygenase (COX)-2-derived prostaglandins (PGs) contribute to tumor growth by inducing angiogenesis. We investigated the role of COX-2 in hepatitis B virus (HBV)-associated hepatocellular carcinoma (HCC). COX-2 and vascular endothelial growth factor (VEGF) expressions were examined by immunohistochemistry in 24 HBV-associated HCC. Tumor micro-vessel density (MVD) was assessed using CD34 immunohistochemistry. Hep3B HCC cell line, which carries integrated HBV genome, was stably transfected with human COX-2 cDNA. COX-2 and VEGF expressions were determined by Western blot while PG level was determined by ELISA. The effects of PGs on VEGF expression were also investigated. Expression of COX-2 and VEGF in HCC cells were observed in 19 (79%) and 16 (67%) cases, respectively. Well-differentiated HCC expressed COX-2 more strongly than less-differentiated HCC (p<0.001). COX-2 expression was found to correlate with VEGF expression and MVD (p=0.003 and 0.004, respectively). COX-2 overexpressing Hep3B clone had higher VEGF expression as compared to non-COX-2 expressing clone and parental cells. Treatment of the COX-2 overexpressing cells with a COX-2-selective inhibitor, NS-398 (10 microM), decreased PGE2 level and attenuated VEGF expression. Addition of PGE2 (10 microM) and the stable analog of PGI2, carbaprostacyclin (5 microM), to Hep3B cells also increased VEGF expression. Up-regulation of COX-2 correlates with VEGF expression and tumor angiogenesis in HBV-associated HCC. Moreover, COX-2 up-regulates VEGF expression in HCC cells, possibly via PGs production. Selective inhibition of COX-2 may block HCC associated angiogenesis and thus provides a rational approach for treatment of this malignancy.     

Expression of cyclooxygenase 2 in human malignant melanoma

Cyclooxygenase (COX)-2 is an inducible enzyme involved in production of prostaglandins in inflammatory processes. There is now increasing evidence that a constitutive expression of COX-2 plays a role in development and progression of malignant epithelial tumors. In the present study we investigated expression and function of COX-2 in malignant melanoma. Expression of COX-2 was determined by immunohistochemistry in 28 cases of primary skin melanoma and 4 benign nevi. We show that COX-2 was expressed in 26 cases (93%) of melanomas, with a moderate to strong expression in 19 cases (68%). Benign nevi as well as normal epithelium were negative in all cases. A constitutive expression of COX-2 mRNA and protein was found in five melanoma cell lines (A375, MeWo, SK-Mel-13, SK-Mel-28, and IGR-37) by using Northern blot as well as immunoblotting. All melanoma cell lines produced prostaglandin (PG) E2 between 468 and 3500 pg/ml as determined by ELISA. Treatment with NS-398 (50 microM), a specific inhibitor of COX-2, suppressed PGE2 production of all melanoma cell lines by 50-96%. The IC50 for inhibition of PGE2 production by NS-398 was determined as 4 microM, indicating that NS-398 acts via inhibition of the COX-2 isoenzyme. We could show that proliferation of melanoma cell lines was not influenced by treatment with NS-398 in concentrations up to 100 microM. However, NS-398 reduced Matrigel invasion of all five malignant melanoma cell lines by 50-68%. Our results indicate that COX-2 is expressed in malignant melanomas and may be involved in regulation of melanoma invasion. It remains to be investigated whether selective inhibitors of COX-2 might be useful for prevention or treatment of malignant melanoma.     

The anti-inflammatory potential of berberine in vitro and in vivo

Berberine, an isoquinoline alkaloid, has a wide range of pharmacological effects, including anti-inflammation, yet the exact mechanism is unknown. Because cyclooxygenase-2 (COX-2) plays a key role in prostaglandins (PGs) synthesis, which is elevated in inflammation, we examined whether the anti-inflammatory mechanism of berberine is mediated through COX-2 regulation. In oral cancer cell line OC2 and KB cells, a 12 h berberine treatment (1, 10, and 100 microM) reduced prostaglandin E2 (PGE2) production dose-dependently with or without 12-O-tetradecanoylphorbol-13-acetate (TPA, 10 nM) induction. This berberine induced effect occurred rapidly (3 h) as a result of reduced COX-2 protein, but not enzyme activity. The electrophoretic mobility shift assay revealed that activator protein 1 (AP-1) binding was decreased in oral cancer cells treated with berberine for 2 h. Further analysis showed that berberine inhibited AP-1 binding directly. These anti-inflammatory effects paralleled to the in vivo results where berberine pretreatment of Wistar rat inhibited the production of exudates and PGE2 in carrageenan induced air pouch.     

A proposed COX-2 and PGE(2) receptor interaction in UV-exposed mouse skin

The cyclooxygenases (COX-1 and COX-2) and the prostaglandins (PGs) they generate play a major role in the skin's response to sunlight. Sunlight, especially the ultraviolet B (UVB) component, induces COX-2 and increases PG levels. However, PGs can have both beneficial and adverse cutaneous effects. To elucidate the roles of the COXs and the PGs they generate in response to UVB exposure, experiments with the COX-1- and COX-2-deficient mice have provided insight into the specific roles of each isoform. Furthermore, because PGE(2) is the major PG produced following UV exposure and PGE(2) manifests its biological activity via four membrane receptors (EP1, EP2, EP3, EP4), elucidating contributions of these receptors is essential for understanding the roles of PGs in UVB-induced effects. In this review, we summarize recent findings from the COX-deficient mice showing how COX-2 generated PGE(2) acting via the receptors EP2 and EP4 could contribute to short term beneficial, but also contribute to long-term carcinogenic effects in response to UVB exposure.     

Role of areca nut in betel quid-associated chemical carcinogenesis: current awareness and future perspectives.

Betel quid (BQ)-chewing is a popular oral habit with potential links to the occurrence of oral cancer. Many of the literature-based studies reveal that areca nut (AN) extract may demonstrate mutagenic and genotoxic effects, in addition to inducing preneoplastic as well as neoplastic lesions in experimental animals. Areca nut should, thus, be highly suspected as a human carcinogen. Toxicity studies relating to AN-contained polyphenols and tannins are not conclusive, with both carcinogenic and anti-carcinogenic effects being reported. The mutagenicity and genotoxicity of areca alkaloids has been detected by many short-term assays. However, their genotoxicity to oral fibroblasts and keratinocytes, the target cells of BQ, has not been identified. It would thus appear that AN toxicity is not completely due to its polyphenol, tannin and alkaloid content. The single agent which is responsible for AN carcinogenicity awaits further clarification. Reactive oxygen species produced during auto-oxidation of AN polyphenols in the BQ-chewer's saliva, are crucial in the initiation and promotion of oral cancer. Nitrosation of areca alkaloids also produces AN-specific nitrosamines, that have been demonstrated to be mutagenic, genotoxic and are capable of inducing tumors in experimental animals. Arecaidine and AN extract are further suggested to be tumor promoters. Antioxidants such as glutathione and N-acetyl-L-cysteine can potentially prevent such AN-elicited cytotoxicity. Further studies are needed to delineate the metabolism of AN ingredient and their roles in the multi-step chemical carcinogenesis, in order to enhance the success of the future chemoprevention of oral cancer and oral submucous fibrosis.     

Prostaglandin E(2) stimulates progression-related gene expression in early colorectal adenoma cells

Upregulation of cyclooxygenase-2 (COX-2) and prostaglandin-dependent vascularisation in small adenomatous polyps is an essential part of colon carcinogenesis. To study the underlying cellular mechanisms, LT97 and Caco2 human colorectal tumour cells not expressing endogenous COX-2 were exposed to 1 microM prostaglandin E(2) (PGE(2)) in their medium. At 30 min after addition, expression of c-fos was stimulated 5-fold and 1.3-fold, respectively, depending on the activation of both extracellular signal-regulated kinase and p38. The amount of c-jun in nuclear extracts was increased 20% in LT97 cells. Expression of COX-2 was upregulated 1.7-fold in LT97 cells and 1.5-fold in Caco2 2 h after prostaglandin (PG) addition by a p38-mediated pathway. The known PGE(2) target gene vascular endothelial growth factor (VEGF) was not modulated. Effects of sustained PGE(2) production were studied in VACO235 cells that have high endogenous COX-2 and in LT97 cells infected with an adenovirus expressing COX-2. Prostaglandin E(2) secretion into the medium was 1-2 nM and 250 pM, respectively. Expression of both VEGF and c-fos was high in VACO235 cells. In LT97 cells, COX-2 upregulated c-fos expression and c-jun content in nuclear extracts 1.7- and 1.2-fold, respectively, in a PG-dependent way. This shows that exogenous PGE(2) as well as COX-2 overexpression affect signalling and gene expression in a way that enhances tumour progression.     

Cyclooxygenase-2 inhibitors demonstrate anti-proliferative effects in oesophageal cancer cells by prostaglandin E(2)-independent mechanisms

The incidence of oesophageal cancer (OC) has risen in recent decades, with survival rates remaining poor despite surgical treatment and adjuvant chemotherapy. Studies have reported cyclooxygenase-2 (COX-2) overexpression in OC and current evidence suggests NSAIDs have major potential for chemoprevention through COX-2 inhibition. However, several reports have questioned the specificity of these inhibitors, suggesting they may act through mechanisms other than COX-2. We evaluated the effects of specific COX-2 inhibitors, NS-398 and nimesulide, on cell lines of both histological types of OC. COX-2 protein expression varied in the cell lines and corresponded with levels of prostaglandin E(2) (PGE(2)) production. Following treatment with low concentrations of NS-398 (0.1 microM), PGE(2) production was reduced dramatically, indicating inhibition of COX-2 activity. Examination of cellular morphology, caspase-3 activity and mitochondrial membrane integrity found no major induction of apoptotic cell death at concentrations below 100 microM. Tumour cell proliferation was significantly reduced at high concentrations (50-100 microM) of both inhibitors over 6 days. Cellular responses were more evident in NS-398-treated adenocarcinoma cells. However, concentrations required to inhibit proliferation were up to 1000-fold higher than those needed to inhibit enzyme activity. Addition of exogenous PGE(2) to NS-398-treated adenocarcinoma cells failed to reverse the inhibitory effects, indicating PG and COX-2 independence. It remains possible that in vivo COX-2 is the primary target, as enzyme inhibition can be achieved at low concentrations, however, inhibition of proliferation is not the primary mechanism of their anti-tumour activity.     

Inhibition of cyclooxygenase-2 activity by celecoxib does not lead to radiosensitization of human prostate cancer cells in vitro.

PURPOSE: To evaluate the potential radiosensitizing effect of the specific COX-2 inhibitor celecoxib (Celebrex) on prostate carcinoma cells in vitro. MATERIALS AND METHODS: The influence of celecoxib (concentration range 5 to 75 microM) on radiation-induced cellular and clonogenic survival was investigated in prostate carcinoma cell lines PC-3, DU145, LNCaP and normal prostate epithelial cells (PrEC). Western blot analysis and ELISA were used to determine the impact of radiation alone or radiation combined with celecoxib treatment on COX-2 expression and prostaglandin E2 synthesis. To evaluate induction of celecoxib-induced apoptosis cell cycle analysis has been performed. RESULTS: Celecoxib (5, 10 and 25 microM) in combination with single-dose irradiation of 2 Gy induced a significant radiosensitization in normal prostate epithelial cells which could not be observed for any of the prostate carcinoma cell lines investigated. Increased COX-2 protein expression in PC-3 cells was obvious only after IR with 15 Gy, while PGE2 production was elevated following irradiation (2-15 Gy) in a dose-dependent manner. Treatment with celecoxib alone or in combination with IR led to a dose-dependent increase in COX-2 protein expression. Nevertheless pre-treatment with celecoxib caused a marked reduction of radiation-induced enzyme activity as tested at the level of PGE2 production, both in PC-3 and DU145 cells. Following fractionated irradiation with single doses of 2 Gy, elevated COX-2 protein expression as well as enhanced PGE2 production was observed already after the second fraction in PC-3 cells. Pre-treatment with celecoxib reduced the amount of PGE(2) significantly, but not of COX-2 protein. CONCLUSIONS: Our data obtained for the human prostate cancer cell lines do not indicate that a marked inhibition of prostaglandin E2 synthesis by celecoxib leads to enhanced radiosensitization. Thus, in terms of radiosensitization the analysed prostate cancer cells can be classified as non-responders to celecoxib treatment.     

Hamsters chewing betel quid or areca nut directly show a decrease in body weight and survival rates with concomitant epithelial hyperplasia of cheek pouch

Betel quid (BQ) chewing is strongly associated with the occurrence of oral leukoplakia, oral submucous fibrosis, and oral cancer. There are about 200-600 million BQ chewers in the world. Previous animal studies support the potential carcinogenicity of BQ in different test systems. However, little animal experiment has let hamsters or rats to chew BQ directly, similar to that in humans. In the present study, we established a hamster model of chewing BQ or areca nut (AN). A total of 81 2-week-old hamsters were randomly divided into three groups: 25 for control group, 28 for BQ-chewing group, and 28 for AN-chewing group. These animals were fed with powdered diet with/without BQ or AN for 18 months. Although the consumption of BQ or AN showed some variations, hamsters fed with powdered diet could chew and grind AN or BQ into small pieces of coarse fibers during the entire experimental period. The survival rate of AN-chewing hamsters decreased significantly after 6 months of exposure. The mean survival time was 15.6 +/- 0.9 months for control animals, 13.6 +/- 0.98 months for AN-chewing animals, and 15.7 +/- 0.55 months for BQ-chewing animals. The body weight of BQ- or AN-chewing animals also decreased after 4-13 months. Hamsters fed with AN for 18 months showed hyperkeratosis in 80% and acanthosis in 50% of cheek pouches. Animals fed with BQ for 18 months also showed hyperkeratosis in 93% and acanthosis in 14% of cheek pouches. These results indicate that AN and BQ components may induce alterations in proliferation and differentiation of oral epithelial cells. Animal model of chewing BQ or AN can be useful for future tumor initiation, promotion and chemoprevention experiments simulating the condition of BQ chewing in humans.     

Effect of prostaglandin E1 on vascular endothelial growth factor production by human macrophages and colon cancer cells

We previously demonstrated that cyclooxygenase-2 (COX-2) was predominantly expressed in macrophages of sporadic human colonic adenomas; however, the role of COX-2-expressing cells during colon carcinogenesis has not yet been elucidated. In the present study, we showed the effect of PGE, on vascular endothelial growth factor (VEGF) production by PMA-differentiated U937 cells, a human macrophage model (H-Mac), and by human colon cancer cells T84. PGE1 dramatically induced VEGF production by H-Mac, but not that by T84. PGE1 significantly increased intracellular cAMP formation by H-Mac, but only modestly increased that by T84. 8-bromo-cAMP and cholera toxin also increased VEGF production by H-Mac. In contrast, neither of these agents modulated VEGF production by T84. EP2 and EP4 (PGE specific receptors) mRNA was expressed in both cells. PG dramatically increased VEGF production by activated macrophages, but not by cancer cells, through a specific PGE receptor-mediated process. These findings suggest that PGs produced by COX-2-expressing macrophages induce VEGF production by macrophages, but not by cancer cells, in an autocrine fashion.     

Cyclooxygenase-1, but not -2, is upregulated in NB4 leukemic cells and human primary promyelocytic blasts during differentiation.

Cyclooxygenase (COX)-1 or -2 and specific prostaglandin (PG) synthases catalyze the formation of various PGs. We investigated the expression and activity of COX-1 and -2 during granulocyte-oriented maturation induced by all-trans-retinoic acid (ATRA) of NB4 cells, originated from a human acute promyelocytic leukemia (APL), and in blasts from APL patients. The expression of COX isoenzymes or prostaglandin synthases was also investigated in circulating granulocytes and human bone marrow. COX-1 was expressed and enzymatically active in NB4 cells and primary blasts. COX-1 mRNA and protein were induced by ATRA. COX-1 protein increased approximately 2-3.5-fold by culture day 3 in NB4 cells and primary blasts, while basal COX-2 expression was very low and unaffected by ATRA. COX-1-dependent PGE(2) biosynthesis increased during differentiation approx. 5-fold. Indomethacin and the selective COX-1 inhibitor SC-560, but not selective COX-2 inhibition, impaired NB4 differentiation, reducing NADPH-oxidase activity, CD11b and CD11c expression. The immunohistochemistry of granulocytes and myeloid precursors in the bone marrow showed a large prevalence of COX-1 as compared to COX-2. In conclusion, COX-1 is induced during ATRA-dependent maturation and appears to contribute to myeloid differentiation both in vitro and ex vivo, and COX-1 activity may potentiate the differentiation of human APL.Leukemia (2004) 18, 1373-1379. doi:10.1038/sj.leu.2403407 Published online 10 June 2004     

Effect of lindane and phenobarbital on cyclooxygenase-2 expression and prostanoid synthesis by Kupffer cells.

Prostaglandins (PGs) have been implicated in tumor promotion. In this study, we investigated the effect of the hepatic tumor promoters lindane and phenobarbital (PB) on the PG metabolism of Kupffer cells in vitro and in vivo, in particular on the expression of cyclooxygenase (COX), the leading enzyme in prostanoid synthesis. Exposure of primary cultures of Kupffer cells to lindane for 1 h stimulated the production of the PGs PGE(2) and PGD(2) markedly (up to 50-fold) and that of PGF(2alpha) by >3-fold. This effect was accompanied by an increase in the COX-2 protein, as demonstrated by western blotting. Similarly, PB, which shares several effects with lindane in rat liver, also clearly induced COX-2. Lindane and PB affected the PG synthesis in vitro and in vivo in Kupffer cells of rats that had been treated with the two compounds for 56 days. Kupffer cells, which were isolated at days 2, 5 and 56 of the treatment, showed a significant increase in the levels of COX-2 mRNA and protein. Total COX activity was increased approximately 2-fold and 3- to 5-fold in Kupffer cell homogenates of PB- and lindane-treated animals, respectively, compared with the untreated controls. These results suggest that paracrine mechanisms may contribute to the tumor-promoting activity of lindane and PB, stimulating the production of PGs by Kupffer cells.     

Chemoprevention of colorectal cancer by inhibition of cyclooxygenase-2 derived prostaglandin E2 signaling: Recent advances in basic biology

Experimental evidence collected over the past decade has revealed that targeting cyclooxygenase-2 (COX-2) may have some efficacy for chemoprevention of cancer. However, recent safety concerns over the long-term use of COX-2 selective inhibitors are prompting research aimed at identifying other specific targets downstream of COX-2. In this regard, several groups have found that NSAIDs and COX-2 selective inhibitors primarily reduce the production of prostaglandin (PG) E2, a biologically active lipid product of the COX-2 enzyme, which results in attenuation of PGE2 signaling in the tumor microenvironment. Therefore, a detailed understanding of the PGE2 signaling pathway in transformed cells is critical to help identify novel and safer targets for effective prevention and/or treatment of cancer. Here we review the recent advances in elucidating the molecular mechanisms by which PGE2 promotes carcinogenesis.     

Cyclooxygenase inhibitors retard murine mammary tumor progression by reducing tumor cell migration, invasiveness and angiogenesis

Tumor-derived prostaglandins (PGs) have been implicated in the progression of murine and human breast cancer. Chronic treatment with a non-selective PG inhibitor indomethacin was shown in this laboratory to retard the development and metastasis of spontaneous mammary tumors in C3H/HeJ female retired breeder mice. The present study examined the role of endogenous prostaglandins in the proliferation/survival, the migratory and invasive behavior and angiogenic ability of a highly metastatic murine mammary tumor cell line, C3L5, originally derived from a C3H/HeJ spontaneous mammary tumor. This cell line was shown to express high levels of cyclooxygenase (COX) -2 mRNA and protein as detected by Northern and Western blotting as well as immunostaining. PGE(2) production by C3L5 cells was primarily owing to COX-2, since this was blocked similarly with non-selective COX inhibitor indomethacin and selective COX-2 inhibitor NS-398, but unaffected with the selective COX-1 inhibitor valeryl salicylate. C3L5 cell proliferation/survival in vitro was not influenced by PGs, since their cellularity remained unaffected in the presence of PGE(2) or NS-398 or PG-receptor (EP1/EP2) antagonist AH6809; a marginal decline was noted only at high doses of indomethacin, which was not abrogated by addition of exogenous PGE(2). Migratory and invasive abilities of C3L5 cells, as quantitated with in vitro transwell migration/invasion assays, were inhibited with indomethacin or NS-398 or AH6809 in a dose-dependent manner; the indomethacin and NS-398-mediated inhibition was partially reversed upon addition of exogenous PGE(2). An in vivo angiogenesis assay that used subcutaneous implants of growth factor-reduced matrigel inclusive of tumor cells showed a significant inhibition of blood vessel formation in these implants in animals treated with indomethacin compared with animals receiving vehicle alone. These studies show that selective and nonselective COX-2 inhibitors retarded tumor progression in this COX-2-expressing murine mammary tumor model by inhibiting tumor cell migration, invasiveness and tumor-induced angiogenesis. The inhibitory effects were not entirely PG dependent; some PG-independent effects were also noted.     

Cytokines differentially regulate the synthesis of prostanoid and nitric oxide mediators in tumorigenic versus non-tumorigenic mouse lung epithelial cell lines

Studies using transgenic and knockout mice have demonstrated that particular cytokines influence lung tumor growth and identified prostaglandin E2 (PGE2), prostacyclin (PGI2) and nitric oxide (NO) as critical mediators of this process. PGE2 and NO were pro-tumorigenic while PGI2 was antitumorigenic. We describe herein an in vitro experimental approach to examine interactions among cytokines, prostaglandins (PGs) and NO. PGE2, PGI2, and NO levels were assayed in culture media from non-tumorigenic mouse lung epithelial cell lines, their spontaneous transformants and mouse lung tumor-derived cell lines, before or after exposure to the cytokines TNFalpha, IFNgamma and IL1beta, alone and in combination. More PGE2 than PGI2 was produced by neoplastic cells, while the opposite was observed in non-tumorigenic lines. Cytokine exposure magnified the extent of these differential concentrations. The PGE2 to PGI2 ratio was also greater in chemically-induced mouse lung tumors than in adjacent tissue or control lungs, supporting the physiological relevance of this in vitro model. Expression of PG biosynthetic enzymes in these cell lines correlated with production of the corresponding PGs. Cytokine treatment enhanced NO production by inducing the inflammation-associated biosynthetic enzyme, inducible NO synthase (iNOS), but this did not correlate with the neoplastic status of cells. Inhibition of iNOS or cyclooxygenase 2 activity using aminoguanidine or NS-398 respectively, demonstrated that NO did not affect PG production nor did PGs influence NO production. Since lack of iNOS inhibits mouse lung tumor formation, we propose that this is independent of any modulation of PG synthesis in epithelial cells. The similar normal/neoplastic trends in PGE2 to PGI2 ratios both in vitro and in vivo, together with an amplification of this difference upon cytokine exposure, are consistent with the hypothesis that cytokines released during inflammation exacerbate differences in the behavior of neoplastic and normal lung cells.