Effect of rapeseed oil and dietary n-3 fatty acids on triacylglycerol synthesis and secretion in Atlantic salmon hepatocytes.

Fish oil (FO) has traditionally been used as the dominating lipid component in fish feed. However, FO is a limited resource and the price varies considerably, which has led to an interest in using alternative oils, such as vegetable oils (VOs), in fish diets. It is far from clear how these VOs affect liver lipid secretion and fish health. The polyunsaturated fatty acids (PUFAs), eicosapentanoic acid (EPA) and docosahexanioc acid (DHA), reduce the secretion of lipoproteins rich in triacylglycerols (TAGs) in Atlantic salmon, as they do in humans. The mechanism by which n-3 fatty acids (FAs) in the diet reduce TAG secretion is not known. We have therefore investigated the effects of rapeseed oil (RO) and n-3 rich diets on the accumulation and secretion of (3)H-glycerolipids by salmon hepatocytes. Salmon, of approximately 90 g were fed for 17 weeks on one of four diets supplemented with either 13.5% FO, RO, EPA-enriched oil or DHA-enriched oil until a final average weight of 310 g. Our results show that the dietary FA composition markedly influences the endogenous FA composition and lipid content of the hepatocytes. The intracellular lipid level in hepatocytes from fish fed RO diet and DHA diet were higher, and the expressions of the genes for microsomal transfer protein (MTP) and apolipoprotein A1 (Apo A1) were lower, than those in fish fed the two other diets. Secretion of hepatocyte glycerolipids was lower in fish fed the EPA diet and DHA diet than it was in fish fed the RO diet. Our results indicate that EPA and DHA possess different hypolipidemic properties. Both EPA and DHA inhibit TAG synthesis and secretion, but only EPA induces mitochondrial proliferation and reduce intracellular lipid. Expression of the gene for peroxisome proliferator-activated receptor alpha (PPARalpha) was higher in the DHA dietary group than it was in the other groups.

Sesamin increases alpha-linolenic acid conversion to docosahexaenoic acid in atlantic salmon (Salmo salar L.) hepatocytes: role of altered gene expression.

In vitro cultivated Atlantic salmon (Salmo salar L.), hepatocytes were incubated without or with a mixture of sesamin and episesamin in order to test for possible effects on lipid metabolism. Sesamin/episesamin exposure (0.05 mM, final concentration) led to increased elongation and desaturation of (14)C 18:3n-3 to docosahexaenoic acid ((14)C 22:6n-3, DHA, P < 0.01) and down regulated gene expression of Delta6 and Delta5 desaturases compared to control treatment. Sesamin/episesamin further increased the hepatocytes capacity for fatty acid beta-oxidation of (14)C 18:3n-3 (P < 0.01) to the (14)C acid soluble products, acetate, malate and oxaloacetate, in agreement with an increased gene expression of carnitine palmitoyltransferase I. Also the gene expression of cluster of differentiation 36 was upregulated and the expression of scavenger receptor type B, peroxisome proliferator-activated receptors alpha and gamma were downregulated. The amount of triacylglycerols secreted by the cells tended to be lower in the sesamin/episesamin incubated hepatocytes than the control cells. This study shows that sesamin has favourable effects on lipid metabolism leading to increased level of DHA, which may be of interest for aquaculture use.

Sesamin supplementation increases white muscle docosahexaenoic acid (DHA) levels in rainbow trout (Oncorhynchus mykiss) fed high alpha-linolenic acid (ALA) containing vegetable oil: metabolic actions.

The effects of including an equi-mixture of sesamin and episesamin in fish diets based on vegetable oils of different fatty acid composition were examined. Sesamin/episesamin (hereafter named sesamin) was included at 0.58 g/100 g diet. The oil used in the feed was either a mixture of linseed and sunflower oils (6:4, by vol) or 100% linseed oil. Addition of sesamin increased the percentages of docosahexaenoic acid (DHA) in white muscle phospholipid and triacylglycerol fraction by up to 37% but the fatty acids in red muscle and liver were not affected. The expression of the peroxisome proliferator-activated receptor PPARalpha was significantly down regulated in the liver of the fish fed sesamin and mixed oil diet (P < 0.05). Sesamin and episesamin were detected in liver and muscle tissues of the fish that had been fed sesamin. Fish fed sesamin had elevated levels of total cytochrome P450 (CYP) enzymes and EROD activity in the liver, indicating an induction of CYP1A in this tissue. Our conclusion was that supplementation of fish feed with sesamin increased the proportions of DHA in the white muscle.

Characterization of scavenger receptor class B, type I in Atlantic salmon (Salmo salar L.).

The scavenger receptor class B, type I (SR-BI) is an important player in regulation of mammalian lipid homeostasis. We therefore wanted to study this receptor in Atlantic salmon (Salmo salar L.), which requires a diet with particular high lipid content. We have for the first time cloned and characterized SR-BI from a salmonid fish. The predicted 494 amino acid protein contained two transmembrane domains, several putative N-glycosylation sites, and showed 72% sequence identity with the predicted homolog from zebrafish. SR-BI expression was analyzed by reverse transcription Real-Time PCR in several tissues, and a high relative expression in salmon midgut was detected, which may suggest that SR-BI has a role in uptake of lipids from the diet. We also expressed a construct of salmon myc-tagged SR-BI in salmon TO cells and HeLa cells, which gave a protein of approximately 80 kDa on reducing SDS-PAGE using an antibody against the myc-epitope. Immunofluorescence microscopy analyses of the salmon SR-BI protein in transiently transfected HeLa cells revealed staining in the cell periphery and in some intracellular membranes, but not in the nucleus, which indicated that the salmon protein may be a functional membrane protein. We also observed a high degree of co-localization using an anti-peptide SR-BI antiserum. We found that 20 microg mL(-1) insulin up-regulated the SR-BI mRNA levels in primary cultures of salmon hepatocytes relative to untreated cells. Oleic acid, EPA, DHA, or dexamethasone did not affect the relative expression of SR-BI in this liver model system. In conclusion, the salmon SR-BI cDNA encoded a protein with several features common to those of mammalian species. SR-BI gene expression was high in the intestine, which leads us to propose that SR-BI may contribute to the uptake of lipids from the diet.

Metabolism of high-density lipoproteins in rainbow trout.

Trout high-density lipoproteins have been labelled with residualizing tracers for the lipid and protein moieties ([3H]cholesteryloleyl ether and 125I-tyramine-cellobiose, respectively). Plasma kinetics and tissue site of catabolism were determined for both tracers. The lipid tracer was cleared about twice as fast from the blood as the protein tracer (half lifes were 63.5 and 125.3 h, respectively). This selective removal of lipid from the lipoprotein was mainly accomplished by the higher liver uptake of the cholesteryl ether. The main catabolic site for HDL protein was kidney tissue. This data established the existence of differential HDL catabolism in a lower vertebrate, in which HDL is the dominant plasma lipoprotein. In addition, the findings confirm the importance of fish kidney as a major site of endocytosis of macromolecules, of both exogenous and endogenous origin.

Hepatic uptake and intracellular processing of LDL in rainbow trout.

The process of receptor-mediated endocytosis in poikilothermic vertebrates such as salmonid fish have not been subjected to much research, compared to the detailed studies done in mammalian systems. We have investigated the hepatic uptake and intracellular processing of low-density lipoprotein (LDL) in rainbow trout liver. After intravenous injection of the [125I]tyramine-cellobiose ([125I]TC) -labelled lipoprotein, the liver was perfused and cells isolated or fractionated by differential centrifugation and isopycnic centrifugation in Nycodenz gradients. We found that LDL was mainly endocytosed by parenchymal cells of the liver. Cell fractionation experiments showed that LDL was localized sequentially in three groups of organelles of increasing density. Initially, LDL was localized in small, slowly sedimenting endosomes before being transferred to denser endosomes (prelysosomes) and finally to dense lysosomes. The lysosomes were identified by three lysosomal marker enzymes. Degradation products formed from [125I]TC-labelled LDL could also be detected in prelysosomal vesicles. In vitro experiments with cultured trout hepatocytes demonstrated that intracellular processing of [125I]TC-LDL in these cells could be suppressed by endocytic and lysosomal inhibitors. The catabolism of LDL in rainbow trout therefore follows the endocytic-lysosomal pathway described for many macromolecules in mammalian cells.

Peroxisome proliferator activated receptors in Atlantic salmon (Salmo salar): effects on PPAR transcription and acyl-CoA oxidase activity in hepatocytes by peroxisome proliferators and fatty acids.

A cDNA fragment which encodes salmon peroxisome proliferator activated receptor y (sPPARgamma) was amplified by PCR from the liver of Atlantic salmon (Salmo salar L.). The fragment was 627 bp long. The sequence of the amplified PCR product was similar to the PPARgamma of mouse and hamster. 59% of the bases were identical. Northern blot analysis of salmon liver mRNA showed that the amplified sPPARgamma fragment hybridised to three specific transcripts of lengths 1.6, 2.4 and 3.3 kb. Clofibric acid and bezafibrate, administered to salmon hepatocytes in culture, resulted in a 1.7-fold increase of the 1.6 kb sPPARgamma transcript. The activity of acyl-CoA oxidase also increased approx. 1.7-fold after administration of fibrates. These results indicate that PPAR is an important factor in mediating enzymatic response to fibrates in fish.

Binding of metastatic colon carcinoma cells to liver macrophages.

The liver is frequently colonized by metastatic tumor cells despite its dense population of macrophages (Kupffer cells). We have studied the interactions between metastatic colon carcinoma cells (DHD) and syngeneic Kupffer cells under different experimental conditions in vitro. In an adhesion assay the binding of DHD cells to Kupffer cell monolayers was shown to be time and temperature dependent, reaching a maximum level after about 90 min of incubation at 37 degrees C. In contrast, only a low level of binding could be observed at 4 degrees C. The level of binding could be increased by pretreatment of the Kupffer cells with phorbol 12-myristate 13-acetate. A firm interaction between the two cell types was shown to be dependent on the presence of calcium- and trypsin-sensitive structures on the surface of the Kupffer cells. Pretreatment of the macrophages with the cytoskeletal inhibitors colchicine and cytochalasin B was also found to reduce significantly the binding of tumor cells. This binding was also inhibited to a large extent by D-mannose and N-acetyl-D-galactosamine. The Kupffer cells were not cytotoxic against the colon carcinoma cells.

Effect of 18:1n-9, 20:5n-3, and 22:6n-3 on lipid accumulation and secretion by Atlantic salmon hepatocytes.

We have studied the effects of dietary FA on the accumulation and secretion of [3H]glycerolipids by salmon hepatocytes in culture. Atlantic salmon were fed diets supplemented with either 100% soybean oil (SO) or 100% fish oil (FO), and grew from an initial weight of 113 +/- 5 g to a final weight of 338 +/- 19 g. Hepatocytes were isolated from both dietary groups and incubated with [3H]glycerol in an FA-free medium; a medium supplemented with 0.75 mM of one of three FA-18:1 n-9, 20:5n-3, or 22:6n-3--or a medium supplemented with 0.75 mM of the sulfur-substituted FA analog tetradecylthioacetic acid (TTA), which cannot undergo beta-oxidation. Incubations were allowed to proceed for 1, 2, 6, or 24 h. The rate of the secretion of radioactive glycerolipids with no FA added was 36% lower from hepatocytes isolated from fish fed the FO diet than it was from hepatocytes isolated from fish fed the SO diet. Hepatocytes incubated with 18:1 n-9 secreted more [3H]TAG than when incubated with no FA, whereas hepatocytes incubated with 20:5n-3 or TTA secreted less labeled TAG than when incubated with no FA. This observation was independent of the feeding group. Hepatocytes incubated with 22:6n-3 secreted the highest amounts of total [3H]glycerolipids compared with the other treatments, owing to increased secretion of phospholipids and mono- and diacylglycerols (MDG). In contrast, the same amounts of [3H]TAG were secreted from these cells as from cells incubated in an FA-free medium. The lipid-lowering effect of FO is thus independent of 22:6n-3, showing that 20:5n-3 is the FA that is responsible for the lipid-lowering effect. The ratio of TAG to MDG in lipids secreted from hepatocytes to which 20:5n-3 or TTA had been added was lower than that in lipids secreted from hepatocytes incubated with 18:1 n-9 or 22:6n-3, suggesting that the last step in TAG synthesis was inhibited. Morphometric measurements revealed that hepatocytes incubated with 20:5n-3 accumulated significantly more cellular lipid than cells treated with 18:1n-9, 22:6n-3, TTA, or no treatment. The area occupied by mitochondria was also greater in these cells. The present study shows that dietary FO reduces TAG secretion from salmon hepatocytes and that 20:5n-3 mediates this effect.

Differences between fluid-phase endocytosis (pinocytosis) and receptor-mediated endocytosis in isolated rat hepatocytes.

To characterize possible differences between the fluid-phase endocytosis (pinocytosis) of bovine serum albumin and the receptor-mediated endocytosis of asialo-orosomucoid (AOM) in isolated rat hepatocytes, both probes were conjugated to radioiodinated tyramine-cellobiose, [125I]TC. The use of these conjugates made it possible to measure the uptake and intracellular distribution of the intact proteins as well as of their acid-soluble, membrane-impermeant degradation products. [125I]TC-albumin was taken up at a very low rate (0.5%/h) compared to [125I]TC-AOM (45%/h), suggesting that neither membrane adsorption nor membrane permeation compromised its suitability as a fluid-phase marker. Sucrose gradient analysis indicated that both probes sequentially entered light endosomes (1.11 g/ml), dense endosomes (1.14 g/ml) and lysosomes (1.18 g/ml), but [125I]TC-albumin traversed the endocytic compartments more rapidly than [125I]TC-AOM, and was partially degraded intralysosomally already after 15 min. The microtubule inhibitor, vinblastine, had a stronger inhibitory effect on the uptake and degradation of [125I]TC-AOM (80% and 95%, respectively) than on the uptake and degradation of [125I]TC-albumin (50% and 70%, respectively). In the presence of vinblastine, [125I]TC-AOM was retained both in light and dense endosomes, whereas [125I]TC-albumin was retained in dense endosomes only, suggesting that the early steps of fluid-phase endocytosis were less critically dependent on microtubular function than the early steps of receptor-mediated endocytosis. A perturbant of vacuolar pH, propylamine, inhibited the degradation of both probes strongly (75-100%), as would be expected from its lysosomotropic effect. Propylamine also inhibited endocytic uptake, with a stronger effect on [125I]TC-AOM uptake (95% inhibition) than on [125I]TC-albumin uptake (60% inhibition), probably reflecting a reduction in endosomal acidity, reduced receptor-ligand dissociation and diminished recycling of free asialoglycoprotein receptors to the cell surface in addition to a general trapping of membrane in swollen vacuoles. A protein phosphatase inhibitor, okadaic acid, strongly (80-100%) inhibited the uptake and degradation of both [125I]TC-albumin and [125I]TC-AOM. An inhibitor of lysosomal proteinases, leupeptin, strongly suppressed the degradation of both probes and moderately reduced the uptake of [125I]TC-AOM, whereas the uptake of [125I]TC-albumin was unaffected. In contrast, an inhibitor of autophagic sequestration, 3-methyladenine, reduced both the uptake and degradation of [125I]TC-albumin markedly (55% and 75%, respectively), with considerably less effect on [125I]TC-AOM (25% and 35%, respectively). As autophagy-inhibitory amino acid mixture did not share these effects, suggesting that 3-methyladenine may suppress endocytic fluid-phase uptake by an autophagy-independent mechanism. Fluid-phase and receptor-mediated endocytosis in hepatocytes thus appear to differ with respect to uptake mechanisms as well as in the kinetics by which endocytosed material traverses the endocytic-lysosomal pathway.

A role for scavenger receptors in phagocytosis of protein-coated particles in rainbow trout head kidney macrophages.

In macrophages of higher vertebrates, Fc receptors and receptors for complement and other serum factors, are generally known to enhance the phagocytic process. In lower vertebrates like salmonid fishes, none of these or other phagocytic receptors have been thoroughly characterized. The purpose of this study was to elucidate to what extent these and other receptors are involved in the process of phagocytosis in rainbow trout (Oncorhynchus mykiss) head kidney macrophages. We used tosyl activated, paramagnetic dynabeads (2.8 microm in diameter), specifically coated with 125I labeled Atlantic salmon (Salmo salar) IgM or bovine serum albumin (BSA) as phagocytic probes. The effect of complement opsonization was also investigated by incubating the beads in serum. Our results indicate that neither the Fc- nor the complement-receptor(s) were important for phagocytosis of these beads. Our data support the idea that scavenger receptors are involved in phagocytosis in rainbow trout head kidney macrophages, as the use of a competitive scavenger receptor ligand extensively decreased degradation of the labeled protein coat on the beads.

Beta-oxidation, esterification, and secretion of radiolabeled fatty acids in cultivated Atlantic salmon skeletal muscle cells.

The white muscle of Atlantic salmon metabolizes FA with different chain lengths and different saturations at different rates, but few details are available on the processes involved or the products formed. We have investigated how multinucleated muscle cells (myotubes) in culture metabolize [1-(14)C]8:0, [1-(14)C]18:1n-9, and [1-(14)C]20:5n-3. The myotubes were formed by the differentiation of isolated myosatellite cells from the white skeletal muscle of salmon fry. Almost all (98%) of the [1-(14)C]8:0 substrate was oxidized to acid-soluble products (ASP) and (14)CO2 after 48 h of incubation, whereas only approximately 50% of the [1-(14)C]18:1n-9 and [1-(14)C]20:5n-3 substrates were oxidized. However, only one cycle of beta-oxidation was measured by the method used. For all three substrates, the main ASP were acetate and a combined fraction of oxaloacetate and malate. Nearly twice as much radioactivity from the [1-(14)C]20:5n-3 substrate was found in the cellular lipids as radioactivity from [1-(14)C]18:1n-9, indicating that [1-(14)C]20:5n-3 was taken up into muscle cells more rapidly than [1-(14)C]18:1n-9. Approximately 10% of the added [1-(14)C]20:5n-3 substrate and 5% of the added [1-(14)C]18:1n-9 substrate was secreted from the muscle cells into the culture media as esterified lipids. Immunocytochemical staining showed that the cells synthesized apolipoprotein A-I. Differentiated muscle cells also expressed peroxisome proliferator-activated receptor alpha (PPARalpha) and PPARbeta, two transcription factors that are involved in regulating beta-oxidation.

Influence of dietary sesamin, a bioactive compound on fatty acids and expression of some lipid regulating genes in Baltic Atlantic salmon (Salmo salar L.) juveniles.

The effects of inclusion of sesamin / episesamin in Baltic Atlantic salmon (Salmo salar L.) diets based on vegetable oils were studied. The study was designed as a dose response study with two control diets, one diet based on fish oil (FO) and one diet based on a mixture of linseed and sunflower oil (6:4 by vol.) (MO). As experimental diets three different levels of inclusion of sesamin / episesamin (hereafter named sesamin) to the MO based diet and one diet based on sesame oil and linseed oil (SesO) (1:1 by vol.) were used. The dietary oils were mirrored in the fatty acid profile of the white muscle. Sesamin significantly decreased the levels of 18:3n-3 in the white muscle phospholipid (PL) fraction of all groups fed sesamin, no significant differences were found in the triacylglycerol fraction (TAG). Slightly increased levels of docosahexaenoic acid (22:6n-3, DHA) in PL and TAG were found in some of the sesamin fed groups. Sesamin significantly affected the expression of peroxisome proliferator-activated receptor alpha, scavenger receptor type B and hormone sensitive lipase, in agreement with previous studies on rainbow trout (Oncorhynchus mykiss) and Atlantic salmon (Salmo salar L.) hepatocytes published by our group. No significant effects on toxicological response measured as ethoxyresorufin O-deethylase activity was found. The total cytochrome P450 enzymes were significantly higher in MO 0.29 and SesO group. The amount of alpha- and gamma-tocopherols in liver and the amount of gamma-tocopherol in white muscle were significantly lower in fish fed the FO diet compared to the MO diet, but no difference after inclusion of sesamin was found in this study. Increased inclusion of sesamin increased the levels of sesamin and episesamin in the liver, but did not affect the amounts in white muscle.

Metabolism of low density lipoproteins in rainbow trout.

Plasma kinetics and tissue sites of degradation for native and chemically modified low density lipoproteins have been investigated in (Oncorhynchus mykiss). Native and modified LDL labelled with(125)I-tyramine-cellobiose, (a residualizing adduct), were injected intravenously, and plasma and organ samples analyzed. Native LDL were cleared with a half life of about 30 hours, and mainly catabolized in the liver. Acetylation of LDL resulted in accelerated clearance (t1/2=2 h) and catabolism in the kidneys. Methylation of LDL had only minor effects on catabolism. The cellular localization of lipoprotein uptake was visualized in kidney by fluorescence microscopy. Native LDL were endocytozed by spheroid, parenchymal cells, supposedly steroid-producing cells. Acetylated, fluorescent LDL were found in vacuoles of flattened, sinusoidal lining endothelial cells. Our data show that catabolism of native low density lipoproteins in salmonids takes place mainly via hepatic receptors. A scavenger receptor pathway, for modified lipoproteins (mainly localized in the kidney) is also operative in trout.

Interaction of low density lipoproteins with liver cells in rainbow trout.

Liver is the main catabolic tissue for low density lipoprotein in rainbow trout (Gjøen and Berg 1992). We have investigated the interaction of LDL with isolated trout liver cells and liver membranes. (125)I-TC labelled trout LDL bound to isolated trout liver cells in a time dependent and saturable manner with an apparant Kd of 20.1 µg/ml, suggesting the existence of a specific binding site on the surface of these cells. The binding was Ca(2+) dependent assessed by the 50% reduction obtained by 5 mM EDTA. Saturable binding to isolated trout liver membranes could also be demonstrated, but with lower affinity as compared to intact cells. Degradation of (125)I-TC-LDL in hepatocytes was also saturable as degradation could be inhibited about 60% by a 100 fold surplus of unlabelled LDL. The rate of degradation increased with temperature up to 20°C. Both cell association (binding + uptake) and degradation were reduced down to 20% of control in the presence of microtubular and lysosomal inhibitors. Hepatic catabolism of trout LDL therefore seems to depend on receptormediated endocytosis, followed by lysosomal degradation.

Lysosomal and endosomal heterogeneity in the liver: a comparison of the intracellular pathways of endocytosis in rat liver cells.

Air-filled albumin microspheres, asialoorosomucoid and formaldehyde-treated serum albumin are selectively taken up by endocytosis in rat liver Kupffer cells, parenchymal cells and endothelial cells, respectively. Intracellular transport and degradation of endocytosed material were studied by subcellular fractionation in sucrose and Nycodenz gradients after intravenous injection of the ligand. By using ligands labeled with 125I-tyramine-cellobiose, the subcellular distribution of labeled degradation products can be studied because they are trapped at the site of formation. The results show that the kinetics of intracellular transport are different in hepatic parenchymal, endothelial and Kupffer cells. In endothelial cells, the ligand is associated with two types of endosomes during the first minutes after internalization and then is transferred rapidly to the lysosomes. In parenchymal cells, 125I-tyramine-cellobiose-asialoorosomucoid was located in a relatively slowly sedimenting vesicle during the first minute after internalization and subsequently in denser endosomes. Degradation of 125I-tyramine-cellobiose-asialoorosomucoid in parenchymal cells started later than that of 125I-tyramine-cellobiose-formaldehyde-treated serum albumin in endothelial cells. Furthermore, the ligand seemed to be transferred relatively slowly from endosomes to lysosomes, and most of the undegraded ligand was in the endosomes. The rate-limiting step of proteolysis in parenchymal cells is probably the transport from endosomes to lysosomes. In Kupffer cells, most 125I-tyramine-cellobiose-microspheres are found as undegraded material in very dense endosomes up to 3 hr after injection. After 20 hr, most of the ligand is degraded in lysosomes distributed at a lower density than the endosomes in Nycodenz and sucrose gradients.

The expression of endosomal rab proteins correlates with endocytic rate in rat liver cells.

Hepatic endocytosis is characterized by a division of labor between the different cell types with respect to endocytosis, which is mediated by receptors expressed on their cell surface. We have investigated the expression of GTPases of the rab family in rat liver parenchymal and endothelial cells. Small GTPases of the rab protein family control distinct steps of intracellular transport both in the secretory and the endocytic pathway. As controls have been employed the normal rat kidney (NRK) cell line and brain tissue, neuronal cells are known to express high levels of components of the endocytic machinery (clathrin, adaptins, dynamin, uncoating adenosine triphosphatase, etc.). Endothelial cells were found to express four to seven times more rab4, rab5, and rab7 than parenchymal cells. A similar relationship was found between the endocytic rates in the two cell types; the rate of internalization from the plasma membrane of mannose receptors in rat liver endothelial cells was 2.3 pools/min, whereas the corresponding value for internalization of the galactose receptor in parenchymal liver cell was 0.27 pools/min (comparable with the rate of transferrin internalization in NRK cells). Both immunofluorescence and subcellular fractionation experiments showed that rab5 and rab7 were associated with compartments along the endocytic pathway. Brain tissue showed a similar high expression of endocytic components (rab4, rab5, and rab7) as liver endothelial cells, whereas lower values were found in NRK cells. We also analyzed the following proteins involved in endocytosis: clathrin, alpha-adaptin, beta-adaptin, and rabaptin-5. These proteins showed the same pattern of expression as the rab proteins. In conclusion, the results obtained with liver cells corroborate the data obtained in transfected cells and support the notion that rab proteins may be involved in controlling the endocytic rate in liver cells.

Colonization of the rat liver by syngeneic tumor cells. An experimental approach by in vivo and in situ studies.

Syngeneic colon carcinoma cells and glioma cells were injected into the portal vein of BD IX rats. After various time periods the animals were sacrificed and the livers and lungs were fixed and prepared for histology. Atypical cells were observed in the liver 4 and 7 days after the injection of tumor cells, whereas distinct colonies of both colon carcinoma and glioma cells were demonstrated after 14 days. Lung metastases of both tumor cell types were seen after 14 and 30 days. Furthermore, injection of glioma and carcinoma cells into the tail vein gave detectable lung metastases after 7 and 4 days respectively. Intraperitoneal injection of tumor cells resulted in the accumulation of large tumor masses, particularly in the mesentery. By in situ perfusions of the liver with tumor cells included in the perfusion medium it was possible to establish that all the tumor cells were arrested in the course of 4 min. In contrast, normal rat leukocytes were not trapped in the liver, whereas trypsin-treated leukocytes were, suggesting the importance of trypsin-sensitive structures for binding to hepatic tissue. The binding of both glioma and carcinoma cells to the liver and the ensuing growth of tumor nodules in this organ indicate a lack of specificity on part of the malignant cell types for metastasis to the liver in the rat. Both tumor cell types colonized the first organ encountered after injection.