Search

Colon cancer preferentially metastasizes to the liver. To determine cellular backgrounds of this preference, we generated an enhanced green fluorescent protein (eGFP)-expressing rat adenocarcinoma cell line (CC531s) that forms metastases in rat liver after administration to the portal vein. Intravital videomicroscopy (IVVM) was used to visualize early events in the development of tumors in livers of live animals from the time of injection of the cancer cells up to 4 days afterward. Based on information obtained with IVVM, tissue areas were selected for further analysis using confocal laser scanning microscopy (CLSM), electron microscopy (EM), and electron tomography. It was shown that initial arrest of colon cancer cells in sinusoids of the liver was due to size restriction. Adhesion of cancer cells to endothelial cells was never found. Instead, endothelial cells retracted rapidly and interactions were observed only between cancer cells and hepatocytes. Tumors developed exclusively intravascularly during the first 4 days. In conclusion, initial steps in the classic metastatic cascade such as adhesion to endothelium and extravasation are not essential for colon cancer metastasis in liver.

Activity of xanthine oxidoreductase (total xanthine dehydrogenase plus xanthine oxidase) and xanthine oxidase was determined cytophotometrically in periportal and pericentral areas of livers of rats under various (patho)physiological conditions that are known to affect the content of reduced glutathione. For this purpose, rats were either normally fed or fasted for 24 hours, fasted for 24 hours, and treated with diethylmaleate that depleted glutathione or treated by in vivo ischemia for 2 hours in the livers. Xanthine oxidoreductase activity was shown histochemically with the use of a tetrazolium salt procedure, and xanthine oxidase activity was localized with a cerium- diaminobenzidine-cobalt-hydrogen peroxide technique in unfixed cryostat sections of the livers. Cytophotometric measurements showed that total xanthine oxidoreductase activity was decreased after fasting and ischemia, whereas only ischemia caused reduced xanthine oxidase activity. Moreover, the percentage of xanthine oxidase of total xanthine oxidoreductase activity was constant in both periportal and pericentral areas at the level of approximately 4% in normally fed and 24-hour fasted and diethylmaleate-treated rats. Ischemia reduced this percentage in both areas of the liver to 2%. It was concluded that the amount of endogenous reduced glutathione did not affect the percentage of xanthine oxidase. The low percentage of xanthine oxidase as determined in the present in situ histochemical study indicates that in vivo the percentage oxidase in rat liver is lower than is assumed on the basis of biochemical assays in liver homogenates even after strictly controlled homogenization procedures. Apparently, conversion of xanthine dehydrogenase into xanthine oxidase may occur in vitro to yield percentages of xanthine oxidase of 10%-20% as are reported in the literature. The latter increase in the percentage of xanthine oxidase may be caused by changes in the local environment of the enzymes, which is left completely intact in histochemical assays. The finding of this low percentage of xanthine oxidase further stresses that the main function of xanthine oxidoreductase in the liver is not the production of superoxide anion radicals and/or hydrogen peroxide but rather the metabolism of xanthine to uric acid, which can act as a potent antioxidant. (Hepatology 1996 Nov;24(5):1179-84)

The aim of this study was to test whether conversion of xanthine dehydrogenase into xanthine oxidase as induced by fasting, ischemia of the liver or both is an in vivo process or only occurs in vitro in homogenates. For this purpose, the conversion rate of xanthine dehydrogenase into xanthine oxidase was studied in liver homogenates obtained from rats after normal feeding or 24 hr of fasting followed or not by 2 hr of ischemia of the liver. In fed rats, the conversion rate of xanthine dehydrogenase into xanthine oxidase was studied as well in liver homogenates after different periods of reperfusion after 2 hr of ischemia. Homogenization was carried out under strictly controlled conditions, after which the supernatants were incubated at 37 degrees C in buffer for 0 to 5 hr. Enzyme activities were assayed spectrophotometrically by measuring urate production at 295 nm. Conversion started only after 2 to 3 hr of incubation of supernatants of control fed livers, whereas conversion started immediately after 24 hr of fasting. The percentage oxidase activity of total xanthine oxidoreductase activity in ischemic livers from fed animals was slightly higher (26.7% +/- 1.7%; p < 0.05) than in control livers (19.3% +/- 1.4%), whereas the percent oxidase activity in ischemic livers from fasted animals (16.7% +/- 1.0%) was not different from that in control animals (16.8% +/- 1.1%). Ischemia for 2 hr caused in vitro a substantial increase in the conversion rate in supernatants of livers of fed and fasted rats as compared with their controls.(ABSTRACT TRUNCATED AT 250 WORDS)

Many enzymes are distributed heterogeneously within the liver lobule. The factors that play a determining role in the establishment and maintenance of these heterogeneous expression patterns have not yet been identified. To investigate whether the composition of the afferent hepatic blood plays a crucial role in the maintenance of the heterogeneity of gene expression of the parenchymal cells within the liver lobule, we changed the source of the afferent hepatic blood by microsurgical techniques. Three different groups of experimental animals were studied: rats with livers that are perfused with portal blood only (ligation of the hepatic artery), with caval blood only (portocaval transposition and ligation of the hepatic artery) and arterial blood only (portocaval shunt, arterialization of the distal end of the portal vein and ligation of the hepatic artery). To study differences in gene expression patterns, we chose enzymes that have a heterogeneous expression pattern within the liver lobule: the periportally located enzymes carbamoylphosphate synthase, succinate dehydrogenase, phosphoenolpyruvate carboxylkinase and the pericentrally located enzymes glutamine synthase, glutamate dehydrogenase and NADPH–cytochrome P-450 reductase. To eliminate the potential interference of the long half-lives of some of these proteins on the interpretation of the results, we also studied the distribution of the mRNAs of carbamoylphosphate synthase, glutamine synthase, glutamate dehydrogenase and phosphoenolpyruvate carboxykinase. The animals were studied 2 wk after the operations. On the basis of their changes in body weight the animals were in steady state for at least a week. The patterns of gene expression of the enzymes studied did not change, regardless of the source of the altered afferent hepatic blood. The changes in gene expression that were observed in animals that did not regain their preoperative weight were shown to be caused by a limited intake of food. This study demonstrates that the physiological position of the liver within the circulation (i.e., between the gastrointestinal tract and the systemic circulation) is not as critical as is often stated and is certainly not essential for the maintenance of liver cell heterogeneity. The data suggest that the direction of the bloodstream (i.e., the existence of an upstream and a downstream compartment) is a major determinant of zonation of gene expression. (HEPATOLOGY 1993;18:1144-1153).

A quantitative histochemical method was developed for the demonstration in rat liver of the activity of phosphofructokinase, one of the enzymes assumed to be rate-limiting for glycolysis. The procedure was based on the reduction of a tetrazolium salt as final electron acceptor and a multistep reaction using the exogenous or endogenous auxiliary enzymes aldolase, triosephosphate isomerase and glyceraldehyde-3-phosphate dehydrogenase. The highest activity was found in unfixed cryostat sections of rat liver when the incubation medium contained 17% (wt/vol) polyvinyl alcohol, 100 mmol/L Tris-maleate buffer (pH 8.4), 20 mmol/L fructose-6-phosphate, 2 mmol/L ATP, 2 mmol/L MgCl2, 5.9 mmol/L NAD+, 0.47 mmol/L 1-methoxyphenazine methosulfate, 5 mmol/L sodium azide and 5 mmol/L Nitro BT. The addition of auxiliary enzymes was not necessary to demonstrate maximum activity in rat liver. The specificity of the reaction was proven by the absence of any specific (test minus control) reaction when the incubation was performed in the presence of 25 mmol/L phosphoenolpyruvate, a competitive inhibitor of phosphofructokinase. Cytophotometric analysis revealed that linear relationships exist between the amount of specific reaction product formed and incubation time and the section thickness. The Km values for fructose-6-phosphate and the Vmax values were not significantly different in periportal and pericentral areas of livers from either normally fed or 24-hr-fasted rats. The homogeneous distribution of phosphofructokinase activity in the liver acinus is in line with biochemical findings using hepatocytes isolated from the two different areas showing that these cells contained similar amounts of enzyme activity.(ABSTRACT TRUNCATED AT 250 WORDS)