Conferring drug resistance by MDR1 gene transfection increases susceptibility to irradiation and lipid peroxidation in 3T3 cell line.

This study was performed to test the hypothesis that conferring multiple drug resistance reduces cell susceptibility to irradiation and iron-stimulated lipid peroxidation. Multidrug resistant (PN1A) and parental drug sensitive (PSI-2) cell lines were exposed to ADP-Fe or Ascorbate-Fe complexes at 37 degrees C and to irradiation. Lipid peroxidation was estimated by the TBA test, whereas x-ray effect was estimated by clonogenic assay. Cell glutathione-S-transferase (GST), total and Se-dependent glutathione peroxidase (GSH-Px) activities, and glutathione and vitamin E were measured. PN1A produced more peroxides than PSI-2 after exposure to iron complexes and formed fewer colonies after irradiation. Higher activities of GST and total and Se-GSH-Px were observed in PN1A. Vitamin E and total glutathione did not differ in the two cell subclones. These data show that the induction of the mdr1 phenotype by transfection of mdr1 gene in 3T3 cells increases susceptibility to irradiation and iron stimulated lipid peroxidation.

Haeme oxygenase-1 and cardiac anaphylaxis.

1. Haeme oxygenase (HO) is an enzyme mainly localized in the smooth endoplasmic reticulum and involved in haeme degradation and in the generation of carbon monoxide (CO). Here we investigate (1) whether the inducible isoform of HO (HO-1) is expressed in the isolated heart of the guinea-pig and (2) the functional significance of HO-1 on the response to antigen in isolated hearts taken from actively sensitized guinea-pigs. 2. Both the HO-1 expression and activity are consistently increased in hearts from guinea-pigs pretreated with hemin, an HO-1 inducer (4 mg kg(-1) i.p., 18 h before antigen challenge). The administration of the HO-1 inhibitor zinc-protoporphyrin IX (ZnPP-IX, 50 micromol kg(-1), i.p., 6 h before hemin) abolished the increase of both the HO-1 expression and activity. 3. In vitro challenge with the specific antigen of hearts from actively sensitized animals evokes a positive inotropic and chronotropic effect, a coronary constriction followed by dilation and an increase in the amount of histamine in the perfusates. In hearts from hemin-pretreated animals, antigen challenge did not modify the heart rate and the force of contraction; the coronary outflow was significantly increased and a diminution of the release of histamine was observed. The patterns of cardiac anaphylaxis were fully restored in hearts from animals treated with ZnPP-IX 6 h before hemin. 4. In isolated hearts perfused with a Tyrode solution gassed with 100% CO for 5 min and successively reoxygenated, the response to antigen was similar to that observed in hearts from hemin-pretreated animals. 5. Pretreatment with hemin or the exposure to exogenous CO were linked to an increase in cardiac cyclic GMP levels and to a decrease of tissue Ca(2+) levels. 6. The study demonstrates that overexpression of HO-1 inhibits cardiac anaphylaxis through the generation of CO which, in turn, decreases the release of histamine through a cyclic GMP- and Ca(2+)-dependent mechanism.

Heme oxygenase-1 and the ischemia-reperfusion injury in the rat heart.

Carbon monoxide (CO) is a signaling gas produced intracellularly by heme oxygenase (HO) enzymes using heme as a substrate. During heme breakdown, HO-1 and HO-2 release CO, biliverdin, and Fe(2+). In this study, we investigated the effects of manipulation of the HO-1 system in an in vivo model of focal ischemia-reperfusion (FIR) in the rat heart. Male Wistar albino rats, under general anesthesia and artificial ventilation, underwent thoracotomy, the pericardium was opened, and a silk suture was placed around the left descending coronary artery; ischemia was induced by tightening the suture and was monitored for 30 min. Subsequently, the ligature was released to allow reperfusion lasting for 60 min. The first group of rats was sham operated and injected intraperitoneally (i.p.) with saline. The second group underwent FIR. The third group was treated ip 18 hr before FIR with hemin (4 mg/kg). The fourth group was pretreated ip 24 hr before FIR and 6 hr before hemin with zinc protoporphyrin IX (ZnPP-IX, 50 microg/kg). Specimens of the left ventricle were taken for determination of HO expression and activity, infarct size, malonyldialdehyde (MDA) production, and tissue calcium content. FIR led to a significant increase in the generation of MDA and notably raised tissue calcium levels. Induction of HO-1 by hemin significantly decreased infarct size, incidence of reperfusion arrhythmias, MDA generation, and calcium overload induced by FIR. These effects were prevented by the HO-1 inhibitor ZnPP-IX. The present experiments show that the concerted actions of CO, iron, and biliverdin/bilirubin modulate the FIR-induced myocardial injury.

Susceptibility to lipid peroxidation of human hepatocellular carcinoma cell lines with different levels of multiple drug-resistant phenotype.

BACKGROUND: It has not been established whether the presence of intrinsic or acquired multiple drug-resistant (MDR) phenotype affects susceptibility to undergo iron-stimulated lipid peroxidation. EXPERIMENTAL DESIGN: To assess this point, human hepatocellular carcinoma cell lines with moderate, clinically relevant (P1) or elevated (P1(0.5)) MDR phenotype and their parental drug-sensitive (P5) cell line were exposed to ADP-Fe or ascorbate-Fe complexes and H2O2 in different experiments. Thiobarbituric acid-reactive substances (TBARS) were measured. Total cell glutathione, glutathione-S-transferase, total and selenium-dependent glutathione peroxidase activities, and cell alpha-tocopherol content were also determined. RESULTS: P5 and P1 cell lines showed similar and significant formation of TBAR after 1-hour incubation exposure to iron complexes, whereas P1(0.5) subclone did not. No accumulation of TBAR was observed during the exposure to H2O2 in the three cell lines. Among antioxidants, only alpha-tocopherol cell content was significantly higher in P1(0.5) in comparison with either P1 or P5. CONCLUSIONS: These data suggest that MDR phenotype development per se does not increase resistance to iron-related free radical attack in human hepatocellular carcinoma cell lines. Resistance to undergo lipid peroxidation is associated only to high degrees of drug resistance and appears more related to increased alpha-tocopherol cell content rather than an MDR phenotype.

Reflux and pH: 'alkaline' components are not neutralized by gastric pH variations.

The ability of the 'alkaline' components of reflux to cause harm in vivo is still open to debate, although these components have been shown in vitro to be capable of damaging the mucosa. The precipitation of bile acids and lysolecithin that occurs at low pH values is the main reason for questioning in vivo mucosal damage. This study was undertaken to determine the composition of gastric aspirates at different original pH values and the degree of solubility of the alkaline components when pH modifications are artificially induced. The samples for chemical analysis were collected from indwelling nasogastric tubes after surgical procedures that did not involve the upper gastrointestinal tract. Bile acid and lysolecithin concentrations were assessed by means of dedicated methods. Thirty-five samples were available for bile acid evaluation and 27 for lysolecithin evaluation. Bile acid and lysolecithin assessments were repeated after pH adjustment at 2, 3.5, 5.5 and 7. For easier assessment of the results, three ranges of the original pH were selected (pH < 2, 2 < or = pH < 5, pH > or = 5). For each pH range, results were pooled together and compared with those in the other pH ranges. Bile acid concentrations were 113+/-48, 339+/-90 and 900+/-303 (mean +/- s.e.m. micromol/L), respectively, in the three groups selected on account of the different original pH values. Differences were significant (p < 0.001). Both taurine- and glycine-conjugated bile acids were represented even at pH < 2. No major differences were observed in bile acid concentration with the artificially induced pH variations. Lysolecithin concentrations were 5.99+/-3.27, 30.80+/-8.43 and 108.37+/-22.17 (mean +/- SEM microg/ml), respectively, in the three groups selected on account of the different original pH ranges. Differences were significant (p < 0.001). No significant differences in lysolecithin concentration were detected with the artificially induced pH variations. In conclusion, both bile acids and lysolecithin are naturally represented in the gastric environment even at very low pH values, although their concentrations decrease on lowering of the naturally occurring pH. Given the concentration variability of bile acids and lysolecithin, further studies are needed to assess the minimal concentration capable of mucosal damage in vivo.

Bile acid inhibition of P-glycoprotein-mediated transport in multidrug-resistant cells and rat liver canalicular membrane vesicles.

To study the effect of bile acids on P-glycoprotein-mediated drug transport, we performed experiments using multidrug resistant cells and rat canalicular membrane vesicles. Cellular accumulation and efflux of rhodamine 123 were measured in drug-resistant cells by means of computerized quantitative image analysis and fluorescence microscopy. ATP-dependent [3H]daunomycin transport was studied by means of rapid filtration in canalicular membrane vesicles prepared from normal rats. Doxorubicin-sensitive (PSI-2) and -resistant (PN1A) 3T3 cells and human-derived hepatocellular carcinoma doxorubicin-sensitive and -resistant cells were used. Taurochenodeoxycholate and glycochenodeoxycholate, taurolithocholate and ursodeoxycholate (50 to 200 mumol/L) inhibited rhodamine 123 and [3H]daunomycin transport in multidrug-resistant cells and canalicular membrane vesicles, respectively, whereas taurocholate, taurodeoxycholate and tauroursodeoxycholate did not. Primary and secondary unconjugated bile acids had no effect. These results reveal that taurolithocholate, taurochenodeoxycholate and glycochenodeoxycholate and ursodeoxycholate inhibit P-glycoprotein-mediated drug transport function in multidrug resistant cell lines and in canalicular membrane vesicles. These results suggest possible interaction between P-glycoprotein function and bile acids in cholestasis and after treatment of patients with ursodeoxycholic or chenodeoxycholic acid.

Up-regulation of cyclooxygenase 2 gene expression correlates with tumor angiogenesis in human colorectal cancer.

BACKGROUND & AIMS: Recent studies have shown that cyclooxygenase (COX)-2 and its products, prostaglandins (PGs), may be involved in colorectal carcinogenesis. The aim of this study was to determine whether COX-2 expression and PGE(2) production correlate with microvessel density, vascular endothelial growth factor (VEGF) expression, and tumor metastasis in human colorectal cancer. METHODS: Tumor samples and adjacent normal mucosa were obtained from 31 surgical specimens. Immunohistochemical expression of COX-2, VEGF, and CD31 was analyzed on paraffin-embedded tissue sections. COX-2 and COX-1 proteins were determined by Western blot analysis. COX-2 and VEGF messenger RNA expressions were evaluated using Northern blot analysis. PGE(2) production was determined by specific radioimmunoassay. RESULTS: The immunohistochemical expressions of both COX-2 and VEGF were significantly correlated with microvessel density (P = 0.02 and P = 0.002, respectively). A significant correlation was found between COX-2 and VEGF expression (P = 0.004). Western analysis confirmed the up-regulation of COX-2 protein expression. COX-2 and VEGF genes were overexpressed in tumor specimens as compared with normal mucosa. PGE(2) levels were significantly higher in metastatic tumors than in nonmetastatic ones (P = 0.03). CONCLUSIONS: COX-2 is related to tumor angiogenesis in colorectal cancer. It is likely that VEGF is one of the most important mediators of the COX-2 angiogenic pathway.

Chronic viral hepatitis induced by hepatitis C but not hepatitis B virus infection correlates with increased liver angiogenesis.

Chronic hepatitis B virus (HBV) and hepatitis C virus (HCV) infections lead to cirrhosis and increase the risk for the development of hepatocellular carcinoma (HCC). Angiogenesis is an essential step in oncogenesis and contributes to tumor progression in adult organs; however, to what extent angiogenesis occurs in the liver during chronic viral hepatitis has not been studied. Ninety-nine matched patients affected by chronic hepatitis due to either HBV or HCV were studied together with 13 controls (5 patients were affected by familial hyperbilirubinemia with normal liver histology; 6 patients with stage II primary biliary cirrhosis; and 2 patients with pseudo inflammatory tumor). Microvessel density was assessed in liver biopsies by immunostaining using two different antibodies against endothelial cell antigens, QB-END/10 and Factor VIII. In addition, the liver homogenates and sera of HCV- or HBV-positive patients and controls were tested for their capacity to stimulate the migration and proliferation of freshly isolated human endothelial cells in vitro. Evidence of angiogenesis was significantly more frequent in HCV-positive patients compared with HBV-infected subjects or controls (74% vs. 39% vs. 8%) (chi2 = 20.78; P < .0001) (HCV+ vs. HBV+ vs. controls). The degree of microvessel density was also higher in HCV- than in HBV-positive patients or controls (chi2 = 12.28; P < .005). In addition, HCV-positive sera and liver homogenates stimulated a higher migration and proliferation of human endothelial cells in vitro compared with HBV-positive or control sera and liver homogenates. These observations indicate that angiogenesis is particularly linked to HCV infection, suggesting a possible contribution to HCV-related liver oncogenesis.