Resveratrol prevents inflammation-dependent hepatic melanoma metastasis by inhibiting the secretion and effects of interleukin-18

B16M cells (B16F10 subline) were cultured in DMEM supplemented with 10% FCS and penicillin-streptomycin (Sigma Chemicals Co., St. Louis, MO) [10]. B16M-conditioned medium (B16M-CM) was obtained from subconfluent cells cultured for 12 hours as previously described [13].

Syngeneic C57BL/6J mice (male, 6-8 weeks old) were obtained from IFFA Credo (L'Arbreole, France). Animal housing, care, and experimental conditions were conducted in conformity with institutional guidelines, in compliance with the relevant national and international laws and policies. Hepatic metastases were produced by intrasplenic injection of B16 cells as previously described [14]. Mice were killed on the twelfth day afterwards. Liver tissue was processed for histology [14]. Fifteen 4 μm-thick tissue sections of formaldehyde-fixed liver (five groups, separated 500 μm) were stained with H&E. An integrated image analysis system (Olympus Microimage 4.0 capture kit) connected to an Olympus BX51TF microscope was used to quantify the number, average diameter, and position coordinates of metastases. Percentage of liver volume occupied by metastases and metastasis density (foci number/100 mm³) were also determined [14]. In order to study the effect of RVL (Sigma-Aldrich Chemicals Co, St Louis, MO) on hepatic metastasis development, some mice (10 per group) received 1 mg/kg/day RVL dissolved in ethanol (5%), via intragastric tube, from day 1 to 12. Control mice received the same volume of vehicle. Each experiment was carried out three times.

B16M cells were stably transfected with a construct containing the Photinus pyralis luciferase gene coding sequence under transcriptional control of the cytomegalovirus promoter and the neomycin resistance gene [16]. Three hundred thousand viable luciferase-transfected B16M cells were intrasplenically injected into C57BL/6J mice (n = 30). Some mice received 1 mg/kg/day RVL five days prior to B16M-Luc cell injection. All mice were killed 18 hours later, and livers were analyzed for luciferase activity (Promega Co., Madison, WI) as described previously [16].

HSE cells were isolated from syngeneic C57BL/6J mice (male, 6-8 weeks old) identified, and cultured as described previously [16]. Briefly, isolated mouse liver cells were obtained by two-step collagenase perfusion. A non-parenchymal liver cell fraction was further purified by centrifugation in a Percoll^® gradient (Amersham; Uppsala, Sweden). Kupffer cells were then removed by selective adherence to plastic substrate, HSE cells were phenotypically characterized by flow cytometry analysis with specific antibodies against CD31 (PECAM-1 Sigma, St Louis), HLA class II and CD40, (both from BD Biosciences); CD106 (VCAM-1) and CD14 (both fromBD Pharmingen, San Diego, CA), smooth muscle alpha actin (ASMA, Sigma-Aldrich, St Louis, MO). HSE cells were seeded at 2 × 10⁵ cells/well in RPMI-1640 culture medium (Sigma Chemicals, St. Louis, MO) supplemented with 10% FCS (Life Technologies, Gaithersburg, MD) onto 24-well tissue culture plates pre-coated with type I collagen solution (0.03 mg/ml) (Collagen Biomaterials, Palo Alto, CA) and allowed to spread for 45 min at 37°C and 5% CO₂.

Serum samples were obtained from hepatic (suprahepatic vein) blood of adult male C57/B1/6J mice 18 hours after intrasplenic injection of B16M cells. Some mice received (1 mg/kg/day) RVL from day 1 to 5 via intragastric tube prior to melanoma cells injection. Primary cultured HSE cells were incubated in the presence of absence of 2.5 μM RVL or recombinant VEGF antibody for 30 min, after which B16M-CM or 10 ng/ml of recombinant VEGF were added. Eight hours later, the supernatant from the treated HSE cells were collected. IL-18 concentration in serum from hepatic blood or in culture cell media was detected using a competitive enzyme immunoassay (R&D Systems, Minneapolis, MN).

B16M cells (1 × 10⁴ cells/well) were grown on 8 μm-chamber glass slides. Cells were serum-starved for 24 h and treated with IL-18 (100 ng/ml) for 30, 60 and 120 min. In some experiments, cells received 2.5 μM RVL 30 min prior to IL-18 treatment. Once treatment time had finished, cells were fixed in 4% formaldehyde (in PBS) for 30 min at room temperature and permeabilized in 1% SDS for 10 min. Non-specific binding was blocked for 1 hour with 10% bovine serum in PBS buffer. Cells were incubated with 1.5 μg/ml rabbit anti-p65 polyclonal antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA) for 1 h at room temperature. Further wash steps removed unbound antibody prior to the addition of the secondary Alexa 594 goat anti-rabbit antibody. Images were acquired on a BD Pathway™ Bioimager.

Subconfluent cultures of B16M cells were treated as above described for 60 min. Then, they were harvested and incubated in lysis buffer (10 mM HEPES (pH 7.9), 10 mM KCl, 0.1 mM EDTA, 0.1 mM DTT, 0.1% Nonidet P-40 and 0.5 mM PMSF) for 20 minutes on ice. The crude nuclei were collected by centrifugation, further incubated in 20 mM HEPES (pH 7,9), 0.4 mM NaCl, 1 mM EDTA, 1 mM DTT, 1 mM phenylmethylsulfonyl fluoride (PMSF) for 20 minutes on ice and clarified by micro-centrifugation and frozen. Fifty micrograms of nuclear extracts were resolved on 12% of SDS-PAGE and p65 protein was analyzed by Western blot using rabbit polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA). Lamin-B expression was used as loading control.

Freshly isolated HSE cells were seeded onto 24-well plates for 24 hours. Then, they were cultured for 12 hours in the presence of basal medium, B16M-CM or 1 ng/ml IL-18. In some experiments, HSE cells received 2.5 μM RVL 30 min prior to B16M-CM or IL-18 treatment. HSE cells were disrupted in 50 mM Tris (pH 7.5), 150 mM NaCl, 1% NP40, 0.5% deoxycholic acid, 0.1% SDS, 2 mM EDTA, 10 mM NaF, 10 μg/ml leupeptin, 20 μg/ml aprotinin, and 1 mM PMSF. Then, 40 μg of protein from cell lysates were separated by 12% SDS-PAGE followed by Western analysis using rat anti-mouse VCAM-1 monoclonal antibody and β-tubulin (both from Santa Cruz Biotechnology, Santa Cruz, CA) and the appropriate secondary antibodies.

Blot was imaged using a Syngene G-Box gel imaging system (Synoptics Ltd., Cambridge) and band density analyzed using Gene Tools analysis software.

B16M cells were cultured in DMEM without phenol red with 20 μmol/L 2', 7'-dichlorofluorescein-diacetate (DCFH-DA) as described [12]. H₂O₂ produced by incubated cells oxidizes DCFH to the highly fluorescent DCF so that fluorescence intensity is directly proportional to the amount of H₂O₂ produced by the cells. DCF fluorescence was recorded at 485/22-nm excitation and 530/25-nm emission filters. Non-DCFH-DA-incubated cells were used to determine basal autofluorescence. B16M cells were treated with 1 ng/ml IL-18, for 2 hours. In some experiments B16M cells were incubated with 2.5 μM of RVL 30 minutes prior to IL-18 addition. H₂O₂ production per well was quantified in arbitrary fluorescence units after subtracting basal autofluorescence.

B16M cells were cultured overnight in DMEM plus 10% FCS. Then, they were incubated for 72 hours in the presence of 0.1 ng/ml IL-18 supplemented with 0.5% FCS. Control cells received the same culture medium without the cytokine. In some experiments, 2.5 μM RVL was added to control and IL-18-treated B16M cells. After 48 hours incubation, B16M cell proliferation was measured using sulforhodamine 101 protein assay, as described previously [15].

B16M cells were labeled with 2', 7'-bis-(2-carboxyethyl)-5,6-carboxyfluoresceinacetoxymethylester solution (Molecular Probes, Eugene, OR) and added to primary culture of HSE cells (2 × 10⁵ cells/well). Eight minutes later, wells were washed three times with fresh medium. The number of adhering cells was determined using a quantitative method based on a previously described fluorescence measurement system [14]. HSE were treated with B16M-CM for 6 hours prior to the addition of B16M cells. In some experiments, HSE cells received 2.5 μM RVL or vehicle 30 minutes prior to B16M-CM. In other experiments, B16M cells were pretreated with IL-18 for 6 hours prior to the adhesion assay. In this case, B16M cells received 2.5 μM RVL or vehicle 30 minutes before IL-18.

RVL given orally at 1 mg/kg from day 1 to 12 after B16M cell injection reduced hepatic metastasis volume by 75% as compared to control mice treated with vehicle (Figure 1A-1E). Antimetastatic activities of RVL did not involve any secondary effect that might compromise animal survival. The same treatment schedule also significantly (P < 0.01) decreased metastasis number by 50% (Figure 1E). Moreover, majority of metastases developed in RVL-treated mice were on average significantly smaller (by 60%) than those developed in vehicle-treated mice. Therefore, B16M cells predominantly colonized hepatic tissue through RVL-sensitive mechanisms. Consistent with RVL-dependent decrease in hepatic metastasis number, oral administration of the same daily dose of RVL to animals during 5 days prior to cancer cell injection, led to a statistically significant (P < 0.01) hepatic retention decrease by 45% of luciferase-transfected B16M cells (Figure 2.). Moreover, although anti-metastatic effects of RVL affected 50% of metastases, hepatic metastasis volume decreased by 75%, indicating that either RVL-resistant metastasis had a slower growth rate than RVL-sensitive ones, or that metastasis implantation by RVL-resistant mechanism still had a RVL-sensitive growth mechanism.

Consistent with previous reports [18], our histologic study (Figure 1B and 1D) showed that hepatic melanoma metastases from vehicle-treated mice were predominantly of sinusoidal-type, containing a rich network of intratumoral microvessels supported by sinusoidal-derived myofibroblasts; while a minority were of portal-type, containing a lower density of intratumoral microvessels, mainly supported by portal tract-derived myofibroblasts. Interestingly, RVL decreased by 80% sinusoidal-type metastasis number, while it did not alter the number of portal-type metastasis, indicating that RVL-sensitive metastases were of sinusoidal-type (E Olaso, C Salado, B Arteta, A Lopategi, F Vidal-Vanaclocha: Resveratrol inhibits the proangiogenic response of hepatic sinusoidal cells to tumor-derived soluble factors during liver metastasis by colon carcinoma, submitted).

Previously, we reported that IL-18 increased in hepatic venous blood over basal level during the sinusoidal inflammation associated with liver-infiltrating cancer cells [13, 15]. A similar 5 day-RVL pretreatment schedule as above also completely abrogated the augmentation of IL-18 concentration in the hepatic blood obtained 18 hours after B16M cell injection into treated mice, without affecting the physiological levels of this cytokine in mice not injected with tumor cells (Figure 3A). In order to discard that the increase in plasma IL-18 might reflect a decreased hepatic metabolism of this cytokine subsequent to the metastasis process [19], we next determined the effect of RVL on IL-18 secretion from tumor-activated hepatic sinusoidal cells. Primary cultured HSE cells received either RVL or vehicle 30 min prior to being treated with B16M-CM for 8 hours and the level of IL-18 was measured in the supernatant by ELISA. As shown in Figure 3B, RVL abolished the augmentation of IL-18 concentration in B16-CM-treated HSE cells, while it did not affect basal IL-18 concentrations in the supernatant of untreated primary cultured HSE cells.

Previously, we showed that the pro-adhesive effect of IL-18 on tumor-activated HSE cells was VEGF-dependent [20]. Herein, we showed that anti-murine VEGF antibody completely abrogated IL-18 secretion from tumor-activated HSE cells and that RVL abolished IL-18 production from VEGF-activated HSE cells. Therefore, RVL inhibited IL-18 secretion from tumor-activated HSE cells through the specific inhibition of tumor-derived VEGF on HSE (Figure 3B).

Because IL-18 promotes the adhesion of B16 melanoma cells to the hepatic microvascular endothelium via VCAM-1-dependent mechanism [13, 15], we next studied the effect of RVL on the VCAM-1 expression level in primary cultured hepatic endothelial cells given either recombinant murine IL-18 or B16M-CM. As shown in Figure 4A, pretreatment of HSE cells with RVL abrogated VCAM-1 expression increase in HSE cells given either IL-18 or B16M-CM. Consistently, RVL also abolished the adhesion of B16M cells to HSE cells given the same B16M-CM for 6 hours prior to adhesion assays (Figure 4B).

As assessed by RT-PCR and flow cytometry, B16M cells expressed IL-18 receptor alpha under basal culture conditions [15, 17]. Therefore, together with proinflammatory effects of IL-18 on HSE cells, melanoma cell function might also be affected by HSE-derived IL-18 in the microenvironment of tumor-activated liver [13]. As shown in Figure 5A, in vitro treatment with RVL completely abrogated the hepatic microvascular arrest augmentation induced in B16M cells given 1 ng/ml IL-18 for 6 hours prior to their intrasplenic injection into normal mice. The same treatment schedule with RVL also prevented both B16M cell adhesion to primary cultured HSE cells and B16M cell proliferation induced in IL-18-pretreated B16M cells, without affecting these functional activities in basal condition-cultured B16M cells (Figure 5B and 5C). Because IL-18 regulates H₂O₂ production from B16M-CM-treated HSE cells [16], we next studied the effect of RVL on H₂O₂ production from IL-18-treated B16M cells. Interestingly, B16M cells given 1 ng/ml IL-18 for 2 hours significantly (P < 0,01) increased their H₂O₂ production. However, this cytokine-induced oxidative reaction was completely neutralized when B16M cells received RVL 30 min prior to IL-18 (Figure 5D). Moreover, NF-kB activation is a known pathway for attenuating cancer cell response to inflammatory mediators. Therefore, we examined the nuclear translocation of p65, the transcriptionally active subunit of NF-kB, in cultured B16M cells given recombinant murine IL-18 (Figure 5E and 5F). First, as shown immunofluorescence using an anti-p65 antibody (Figure 5E), p65 remained cytosolic under basal conditions and no change was observed when basal condition-cultured cells received RVL. In contrast, a 60 min stimulation with IL-18 (100 ng/mL) remarkably induced p65 translocation into the nucleus. P65 translocation persisted for 120 min (data not shown). However, addition of RVL to IL-18-treated cells completely abrogated p65 translocation (Figure 5E). Second, similar results were obtained when nuclear extracts from same experimental conditions as above were analyzed for p65 protein expression by Western blot (Figure 5F). Our data therefore indicates that IL-18 activates NF-kB in B16M cells and that RVL can efficiently inhibits this activation.