Intracellular transport of endocytosed proteins in rat liver endothelial cells.

1. Receptor-mediated endocytosis of mannose-terminated glycoproteins in rat liver endothelial cells has been followed by means of subcellular fractionation and by immunocytochemical labelling of ultrathin cryosections after intravenous injection of ovalbumin. For subcellular-fractionation studies the ligand was labelled with 125-tyramine-cellobiose adduct, which leads to labelled degradation products being trapped intracellularly in the organelle where the degradation takes place. 2. Isopycnic centrifugation in sucrose gradients of a whole liver homogenate showed that the ligand is sequentially associated with three organelles with increasing buoyant densities. The ligand was, 1 min after injection, recovered in a light, slowly sedimenting vesicle and subsequently (6 min) in larger endosomes. After 24 min the ligand was recovered in dense organelles, where also acid-soluble degradation products accumulated. 3. Immunocytochemical labelling of ultrathin cryosections showed that the ligand appeared rapidly after internalization in coated vesicles and subsequently in two larger types of endosomes. In the 'early' endosomes (1 min after injection) the labelling was seen closely associated with the membrane of the vesicle; after 6 min the ligand was evenly distributed in the lumen. At 24 min after injection the ligand was found in the lysosomes. 4. A bimodal distribution of endothelial cell lysosomes with different buoyant densities was revealed by centrifugation in iso-osmotic Nycodenz gradients, suggesting that two types of lysosomes are involved in the degradation of mannose-terminated glycoproteins in liver endothelial cells. Two populations of lysosomes were also revealed by sucrose-density-gradient centrifugation after injection of large amounts of yeast invertase. 5. In conclusion, ovalbumin is transferred rapidly through three endosomal compartments before delivering to the lysosomes. The degradation seems to take place in two populations of lysosomes.     

Mannose-receptor-mediated clearance of lysosomal alpha-mannosidase in scavenger endothelium of cod endocardium.

Mannose-receptor-mediated clearance of circulating glycoproteins was studied in Atlantic cod (Gadus morhua). Distribution studies with radioiodinated and fluorescently labelled ligands showed that cod liver lysosomal alpha-mannosidase and yeast invertase were rapidly eliminated from blood via a mannose specific pathway in liver parenchymal cells and endocardial endothelial cells of atrium and ventricle. Asialo-orosomucoid, a galactose-terminated glycoprotein, was cleared by liver only. In vitro studies were performed with primary cultures of atrial-endocardial endothelial cells (AEC), incubated at 12 degrees C in a serum free medium. Cod AEC endocytosed mannose-terminated glycoproteins (125I-alpha-mannosidase, 125I-invertase, 125I-mannan, 125I-ovalbumin and unlabelled lysosomal alpha-mannosidase), whereas 125I-asialo-orosomucoid was not recognised. Uptake of radiolabelled mannose-terminated ligands was inhibited 80-100% in the presence of excess amounts of mannan, invertase, D-mannose, L-fucose or EGTA. Our results suggest that the cod endocardial endothelial cells express a specific Ca(2+)-dependent mannose receptor, analogous to the mannose receptor on mammalian macrophages and liver sinusoidal endothelial cells.     

Intracellular localization and endocytosis of brush border enzymes in the enterocyte-like cell line Caco-2.

Immunoelectron microscopy was used to localize the brush border hydrolases sucrase-isomaltase (SI) and dipeptidylpeptidase IV (DPPIV) in the human colon carcinoma cell line Caco-2. Both enzymes were detected at the microvillar membrane, in small vesicles and multivesicular bodies (MVBs), and in lysosomal bodies. In addition, DPPIV was found in the Golgi apparatus, a variety of apical vesicles and tubules, and at the basolateral membrane. To investigate whether the hydrolases present in the lysosomal bodies were endocytosed from the apical membrane, endocytic compartments were marked with the endocytic tracer cationized ferritin (CF). After internalization from the apical membrane through coated pits, CF was first recovered in apical vesicles and tubules, and larger electronlucent vesicles (early endosomes), and later accumulated in MVBs (late endosomes) and lysosomal bodies. DPPIV was localized in a subpopulation of both early and late endocytic vesicles, which contained CF after 3 and 15 min of uptake, respectively. Also, internalization of the specific antibody against DPPIV and gold labeling on cryosections showed endocytosed DPPIV in both early and late endosomes. However, unlike CF, no accumulation of DPPIV was seen in MVBs or lysosomal bodies after longer chase times. The results indicate that in Caco-2 cells the majority of brush border hydrolases present in lysosomal bodies are not endocytosed from the brush border membrane. Furthermore, the labeling patterns obtained, suggest that late endosomes may be involved in the recycling of endocytosed DPPIV to the microvilli.     

Binding, uptake, and transcytosis of ligands for mannose-specific receptors in rat liver: an electron microscopic study

We have investigated the initial distribution of mannose-specific binding sites in rat liver as well as the uptake and transcytosis pathways of ligands for this receptor in in situ and in vivo experiments. As ligands we used mannan adsorbed onto colloidal gold particles of sizes 5, 17, and 35 nm (Man-Au5, Man-Au17, or Man-Au35). The in situ binding pattern of Man-Au5 in the prefixed liver is identical to the one described earlier for galactose-exposing ligands in the same organ. With the exception of the binding by hepatocytes, where only scarce binding of Man-Au5 was observed, ligands were found adhering in a preclustered pattern all over the cell surface of liver macrophages and binding in aggregates over the coated pits of endothelial cells. In double-labeling experiments different particle sizes were used for glycoproteins with terminal mannosyl or galactosyl residues. This simultaneous localization of the two binding activities revealed that on endothelial cells the two activities are always found to be present in the same coated pit. On liver macrophages the clustered binding occurred at different membrane areas. Uptake and transcytosis of Man-Au5, 17, 35 were studied after their injection into the tail vein. Three and fifteen minutes after injection most of the Man-Au5 and all of Man-Au17 or Man-Au35 was found in sinusoidal liver cells, i.e., macrophages and endothelial cells. One hour after injection, endocytosed ligand is redistributed from large--presumably lysosomal--vacuoles to small noncoated vesicles that are localized predominantly near the space of Dissé. Between 1 and 40 h after injection, ligands of all sizes are transcytosed and found in the hepatocytes. No ligand accumulation is observed in hepatocytes as an indirect indication for secretion into bile. With this investigation we give evidence for transcytotic activity not only of liver endothelium but also of the resident liver macrophages.     

Characterization of the interaction in vivo of tissue-type plasminogen activator with liver cells

The interaction in vivo of 125I-labeled tissue-type plasminogen activator (t-PA) with the rat liver and the various liver cell types was characterized. Intravenously injected 125I-t-PA was rapidly cleared from the plasma (t1/2 = 1 min), and 80% of the injected dose associated with the liver. After uptake, t-PA was rapidly degraded in the lysosomes. The interaction of 125I-t-PA with the liver could be inhibited by preinjection of the rats with ovalbumin or unlabeled t-PA. The intrahepatic recognition site(s) for t-PA were determined by subfractionation of the liver in parenchymal, endothelial, and Kupffer cells. It can be calculated that parenchymal cells are responsible for 54.5% of the interaction of t-PA with the liver, endothelial cells for 39.5%, and Kupffer cells for only 6%. The association of t-PA with parenchymal cells was not mediated by a carbohydrate-specific receptor and could only be inhibited by an excess of unlabeled t-PA, indicating involvement of a specific t-PA recognition site. The association of t-PA with endothelial cells could be inhibited 80% by the mannose-terminated glycoprotein ovalbumin, suggesting that the mannose receptor plays a major role in the recognition of t-PA by endothelial liver cells. An excess of unlabeled t-PA inhibited the association of 125I-t-PA to endothelial liver cells 95%, indicating that an additional specific t-PA recognition site may be responsible for 15% of the high affinity interaction of t-PA with this liver cell type. It is concluded that the uptake of t-PA by the liver is mainly mediated by two recognition systems: a specific t-PA site on parenchymal cells and the mannose receptor on endothelial liver cells. It is suggested that for the development of strategies to prolong the half-life of t-PA in the blood, the presence of both types of recognition systems has to be taken into account.     

Endocytosis via galactose receptors in vivo : Ligand size directs uptake by hepatocytes and/or liver macrophages

We followed the intrahepatic binding and uptake of variously sized ligands with terminal galactosyl residues in rat livers. The ligands were administered to prefixed livers in binding studies and in vivo and in situ (serum-free perfused livers) in uptake studies. Gold sols with different particle diameters were prepared: 5 nm (Au5), 17 nm (Au17), 50 nm (Au50) and coated with galactose exposing glycoproteins (asialofetuin (ASF) or lactosylated BSA (LacBSA)). Electron microscopy of mildly prefixed livers perfused with LacBSA-Au5 in serum-free medium showed ligand binding to liver macrophages, hepatocytes and endothelial cells. Ligands bound to prefixed cell surfaces reflect the initial distribution of receptor activity: pre-aggregated clusters of ligands are found on liver macrophages, single particles statistically distributed on hepatocytes and pre-aggregated clusters of particles restricted to coated pits on endothelial cells. Ligand binding is prevented in the presence of 80 mM N-acetylgalactosamine (GalNAc), while N-acetylglucosamine (GlcNAc) is without effect. Electron microscopy of livers after ligand injection into the tail vein shows that in vivo uptake of electron-dense galactose particles by liver cells is size-dependent. Using a LacBSA-Au preparation with heterogeneous particle diameter (2.2-11.7 nm) we found that hepatocytes take up only ligands up to the size of 7.8 nm, whereas particles of all sizes available in this experiment are found in liver macrophages and endothelial cells. ASF-Au17 and LacBSA-Au17 are endocytosed by liver macrophages and endothelial cells, but not by hepatocytes. ASF-Au50 is taken up by liver macrophages only. In vivo uptake by liver macrophages is mediated by galactose-specific recognition as shown by inhibition with GalNAc. Some 52-65% inhibition was measured in in vivo experiments and 78% inhibition in in situ experiments. GlNAc showed no inhibitory effect. Furthermore, we measured uptake of [125J]ASF and of [125J]ASF adsorbed to Au17 by the different cell populations of rat livers in vivo. While the bulk of the molecular ligand is found in the hepatocyte fraction, the particulate ligand is located in the sinusoidal fraction.     

Visualization of the interaction of native and modified low density lipoproteins with isolated rat liver cells

Morphological characteristics of the interaction of low density lipoproteins (LDL) and acetylated low density lipoproteins (AcLDL) with rat liver cells are described. These liver cell types are mainly responsible for the catabolism of these lipoproteins in vivo. Isolated rat liver Kupffer, endothelial, and parenchymal cells were incubated with LDL or AcLDL conjugated to 20 nm colloidal gold. LDL was mainly internalized by Kupffer cells, whereas AcLDL was predominantly found in endothelial cells. Kupffer and endothelial cells displayed different morphological characteristics in the processing of these lipoproteins. Kupffer cells bound LDL at uncoated regions of the plasma membrane often at the base of pseudopodia, and internalized the particles via small smooth vesicles. These uptake characteristics differ from the classical LDL uptake pathway, as described for other cell types, and may be related to the unique recognition properties of the receptor of Kupffer cells as observed in biochemical studies. Liver endothelial cells bound AcLDL in coated pits, followed by rapid uptake. Uptake proceeded through small coated vesicles, and after 5 min of incubation large (600-1200 nm) electron-lucent vacuoles (endosomes) with AcLDL-gold particles arranged along the membrane region were present. The endosomes were often associated closely with the cell membrane which might enable direct recycling of AcLDL receptors. These observations might explain the high efficiency of these cells in the processing of modified LDL in vivo.     

Formaldehyde treated albumin contains monomeric and polymeric forms that are differently cleared by endothelial and Kupffer cells of the liver: evidence for scavenger receptor heterogeneity.

Formaldehyde treated albumin (F-HSA) was found to consist of a monomeric and a polymeric fraction. Both fractions were primarily endocytosed by rat liver sinusoidal cells. However, immunohistochemical staining of endocytosed material showed that the relative contribution of the endothelial and Kupffer cells in uptake of the monomer and the polymer differed significantly, with the monomer mainly having an endothelial cell- and the polymer predominantly having a Kupffer cell pattern of distribution. To directly confirm these heterogeneous patterns, we injected in vivo the 125I-labeled F-HSA fractions and isolated the endothelial and Kupffer cells by centrifugal elutriation. 73.7% of the monomeric F-HSA was found in endothelial cells and only 14.9% was found in Kupffer cells. In contrast, the polymeric F-HSA (1500 kD) was mainly endocytosed by Kupffer cells (71%), whereas the endothelial cells contributed only for 24% in hepatic uptake. In vivo studies and isolated perfused rat liver experiments showed that endocytosis of both monomer and polymer was inhibited by co-administration of polyinosinic acid, a well known inhibitor for scavenger receptors, indicating that these receptors on endothelial and Kupffer cells are mainly involved in this uptake process.     

Intracellular transport of formaldehyde-treated serum albumin in liver endothelial cells after uptake via scavenger receptors.

Endocytosis of formaldehyde-treated serum albumin (FSA) mediated by the scavenger receptor was studied in rat liver endothelial cells. Suspended cells had about 8000 receptors/cell, whereas cultured cells had about 19,000 receptors/cell. Kd was 10(-8) M in both systems. Cell-surface scavenger receptors were found exclusively in coated pits by electron microscopy, by using ligand labelled with colloidal gold. Cell-surface-bound FSA could be released by decreasing the pH to 6.0; it was therefore possible to assess the rate of internalization of surface-bound ligand. This rate was very high: t1/2 for internalization of ligand prebound at 4 degrees C was 24 s. The endocytic rate constant at 37 degrees C, Ke, measured as described by Wiley & Cunningham [(1982) J. Biol. Chem. 257, 4222-4229], was 2.44 min-1, corresponding to t1/2 = 12 s. Uptake of FSA at 37 degrees C after destruction of one cell-surface pool of receptors by Pronase was decreased to 60%. This finding is compatible with a relatively large intracellular pool of receptors. The intracellular handling of 125I-tyramine-cellobiose-labelled FSA (125I-TC-FSA) was studied by subcellular fractionation in sucrose gradients, Nycodenz gradients or by differential centrifugation. The density distributions of degraded and undegraded 125I-TC-FSA after fractionation of isolated non-parenchymal cells and whole liver were similar, when studied in Nycodenz and sucrose gradients, suggesting that the subcellular distribution of the ligand was not influenced by the huge excess of non-endothelial material in a whole liver homogenate. Fractionation in sucrose gradients showed that the ligand was sequentially associated with organelles banding at 1.14, 1.17 and 1.21 g/ml. At 9-12 min after intravenous injection the ligand was in a degradative compartment, as indicated by the accumulation of acid-soluble radioactivity at 1.21 g/ml. A rapid transfer of ligand to the lysosomes was also indicated by the finding that a substantial proportion of the ligand could be degraded by incubating mitochondrial fractions prepared 12 min after intravenous injection of the ligand. The results indicate that FSA is very rapidly internalized and transferred through an endosomal compartment to the lysosomes. The endosomes are gradually converted into lysosomes between 9 and 12 min after injection of FSA. The rate-limiting step in the intracellular handling of 125I-TC-FSA is the degradation in the lysosomes.     

Intracellular transport and degradation of chylomicron remnants in rat liver cells after in vivo endocytosis

The intracellular transport and degradation of in vivo endocytosed chylomicron remnants labelled with 125I in the protein moiety was studied in rat liver cells by means of subcellular fractionation in Nycodenz and sucrose density gradients. Initially, the radioactivity was located in low-density endosomes and was sequentially transferred to light and dense lysosomes. Data from gel filtration of the light and dense lysosomal fractions showed radioactive material with a molecular weight of about 1000-2000, representing short peptide fragments or amino acids which remain attached to iodinated tyramine cellobiose. In addition, undegraded apoproteins accumulated in both types of lysosome. Our data suggest that endocytosed chylomicron remnant apoproteins are first located in low-density endosomes and are sequentially transferred to light and dense lysosomes. Furthermore, the degradation process starts in the light lysosomes.     

Role of apical tubules in endocytosis in nonciliated cells of the ductuli efferentes of the rat: a kinetic analysis

The apical region of nonciliated cells of the ductuli efferentes of the rat contains tubular coated pits (TCP) connected to the apical plasma membrane, apical tubules (AT) which occasionally show a partial coat, and endosomes which are often continuous with one or more apical tubules. To investigate the formation and fate of TCP and AT, a quantitative analysis was performed on the labeling indices of these structures at various time intervals (0.5-120 min) after a single injection of a tracer, cationic ferritin (CF), into the lumen of the rete testis. The labeling indices of both TCP and AT exhibited similar cyclical patterns, first reaching a peak at 25 min, then dropping to a minimum at 35 min, then rising to a second peak at 60 min. Since TCP were well labeled at 30 sec while AT were not, the tracer must rapidly enter TCP and thence AT. However, since tracer was virtually absent from the lumen by 30 min, it was not possible to reconcile the second peak of labeling index of TCP and AT by this mechanism. In another experiment, rats were injected once as before, injected again at 30 min, and then sacrificed at 30 min following the second injection. The results from this experiment showed that the labeling index of TCP and AT did not drop but was similar to that of the 60-min peak after a single injection. The interpretation is that there was recycling of tracer, which had already migrated from TCP to AT to endosomes, back to the apical plasma membrane via apical tubules. Moreover, when rats were injected once, injected again at 30 min, and sacrificed 3 min following the second injection, the labeling index for TCP and AT was significantly higher (P less than .05) than at the 30-min time interval after a single injection, indicating that recycled apical tubules were functionally capable of binding further CF. Morphological observations on images of transition between TCP and AT and the fact that AT were often found connected to endosomes suggest that TCP detach from the cell surface to give rise to AT, which in turn fuse to form endosomes. The kinetic analysis demonstrates in quantitative terms that a portion of the AT, which fuses to form endosomes, recycles back to the apical plasma membrane and contributes to the formation of new TCP.     

Cation-independent mannose 6-phosphate and 78 kDa receptors for lysosomal enzyme targeting are located in different cell compartments

The distribution of the cation-independent mannose 6-phosphate and 78 kDa receptors was studied in postnuclear subcellular fractions from two rat liver cell lines. ELISA assays revealed that the mannose 6-phosphate receptor is enriched in the light buoyant Percoll fractions that contain Golgi structures and early endosomes. Most of the 78 kDa receptor is localized in a heavy fraction at the bottom of the Percoll gradient and smaller amounts in the endosomal fractions. The high-density compartment is denser than lysosomes, contains LAMP2 but not LIMPII or acid hydrolases, and is not disrupted with glycyl-l-phenylalanine 2-naphthylamide, a substrate for cathepsin C that selectively disrupts lysosomes. Immunofluorescence microscopy studies indicate no colocalization of the 78 kDa receptor with the mannose 6-phosphate receptor or LIMPII. Mannose 6-phosphate-independent endocytosed beta-glucuronidase was found in the lysosomal, the early and late endosomal fractions. These fractions were immunoadsorbed in columns containing antibodies against the 78 kDa receptor. Only the endocytosed beta-glucuronidase present in the early and late endosomal fractions is associated to immunoadsorbed vesicles. In these vesicles, LAMP2 was detected but no LIMPII or the mannose 6-phosphate receptor. Results obtained suggest that the 78 kDa receptor is found along the endocytic pathway, but in vesicles different from the cation-independent mannose 6-phosphate receptor.     

Intracellular transport of endocytosed chylomicron [3H]retinyl ester in rat liver parenchymal cells. Evidence for translocation of a [3H]retinoid from endosomes to endoplasmic reticulum

The intracellular transport of chylomicron remnants labeled with [3H]retinyl ester was studied in rat liver parenchymal cells by means of subcellular fractionation in Nycodenz and sucrose density gradients. The data presented indicate that endocytosed chylomicron remnant [3H]retinyl ester initially is located in low density endosomes. Radioactivity is subsequently transferred to a denser vesicle. Equilibrium as well as rate zonal centrifugation suggest that this denser [3H] retinoid-containing vesicle may represent endoplasmic reticulum. We have compared the intracellular transport of chylomicron remnant [3H]retinyl ester and 125I-asialofetuin. The receptor-mediated endocytosis of asialoglycoproteins in rat liver parenchymal cells is a thoroughly studied system. Our results suggest that the [3H] retinoid and 125I-asialofetuin follow the same path initially to the endosomes. After transit in endosomes, the intracellular transport differs. While asialofetuin is transported to the lysosomes, the retinoid is probably transferred to the endoplasmic reticulum.     

Uptake and transport of glucagon by rat liver: evidence for transcytotic pathway

Summary The binding of¹²⁵I-glucagon to the cell surface and the pathway of intracellular transport of this hormone by rat hepatocytes in vivo were studied by light and EM autoradiography. Radiolabeled glucagon injected into the blood stream was taken up predominantly by the hepatocytes. Negligible radioactivity was found to be associated with other cell types such as endothelial or Kupffer cells. Our results indicate that at early time points after injection glucagon has been preferentially interacting with the sinusoidal domain of the hepatocytes and found to be associated with coated pits and uncoated vesicles corresponding to endosomes. At 15–20 min time intervals glucagon grains were found within hepatocyte interior. Later, at 30 min after injection glucagon grains accumulate in the Golgi-lysosomal region of hepatocyte often in close proximity to the opening of the bile canaliculi. Accordingly a portion of internalized¹²⁵I-glucagon was found to be released into the bile thereby indicating that a transcytotic pathway may be involved in this peptide's clearance process.     

Uptake of lithium carmine by sinusoidal endothelial and Kupffer cells of the rat liver: new insights into the classical vital staining and the reticulo-endothelial system

Sinusoidal cells in the rat liver were studied in vivo and in vitro using the original vital staining with lithium carmine, which has contributed much to the development of the concept of the reticulo-endothelial system. Immunohistochemical and electron-microscopic studies revealed that the dye-incorporating cells were sinusoidal endothelial cells, Kupffer cells, and monocytes. The endothelial cells took up much more dye than did the Kupffer cells and bulged largely into the sinusoidal lumen. Electron microscopy revealed that small particles of lithium carmine were associated with coated vesicles of endothelial cells and ruffled membranes of Kupffer cells. In the endothelial cells, these particles were present in various concentrations within vacuolated structures and condensed in the lysosomes forming large aggregates of lithium carmine lumps. These lumps showed crystalline structures, within which the size of the individual particle was up to 30 nm in width and 50 nm in length. A few endothelial cells containing abundant dye underwent degeneration, and some were taken up by Kupffer cells. Liver endothelial cells isolated from lithium carmine-administered rats endocytosed fluorescence-labeled collagen. Isolated endothelial cells from normal rat liver, when cultured with lithium carmine, did not take up any dye, and their endocytosis of formaldehyde-treated albumin was inhibited dose-dependently. We conclude that in the liver, endothelial cells, but not Kupffer cells, predominantly take up lithium carmine. Furthermore, we propose the existence of a generalized cell system based on its vital staining capacity.     

Protracted circulating lifetimes of mannose-terminated glycoproteins and aggregated albumin in mice infected with LDH-elevating virus.

Several macromolecular homeostasis-regulating mechanisms were tested for functional integrities in mice during acute and early chronic phases of infection with lactic dehydrogenase-elevating virus (LDV). Fractional catabolic rates of carbodiimide-aggregated albumin and immunoglobulin G were studied to evaluate glomerular filtration and hepatic Kupffer cell phagocytic activities. Several glycoproteins (fetuin, IgG antibodies, and ovalbumin) were also compared with their deglycosylated counterparts for fractional catabolic rates and organ distributions as a basis for evaluating virus-induced modifications of saccharide-binding "receptor functions" in vivo. Findings were that normal hepatic clearance of aggregated albumin and of ovalbumin is slowed from the onset of viremia. Fractional catabolic rates of amannosyl-ovalbumin and amannosyl-IgG are similar in uninfected animals to those seen with native ovalbumin or with mannose-terminated IgG in LDV-infected animals. Ovalbumin and aggregated albumin were also found to be mutually competitive for hepatic uptake in uninfected animals. It is proposed that LDV, which replicates in cells of the mononuclear phagocyte system (reticuloendothelial system), alters the clearance functional state of fixed tissue macrophage, thereby explaining in part the protracted circulatory longevity of several enzymes, aggregated albumin and mannose-terminated ovalbumin, and IgG in LDV-infected mice.     

Uptake of mannose-terminated glycoproteins in isolated rat liver cells. Evidence for receptor-mediated endocytosis in hepatocytes.

Even though most of the hepatic binding capacity for mannose-terminated glycoproteins has previously been shown to reside in the hepatocytes (not in the non-parenchymal cells), detailed evidence for the specific uptake of mannose-terminated ligands has been lacking. In the present studies, yeast invertase, a large glycoprotein (Mr 270 000) containing about 50% mannose, was shown to be taken up into hepatocytes by receptor-mediated endocytosis. The uptake was saturable and could be specifically inhibited by mannosides or by a Ca2+ chelator. The asialo-glycoprotein receptor was not involved. The low-Mr (13 000) ligand ribonuclease B, which contains a single high-mannose glycan, was not taken up by hepatocytes; however, it was taken up as fast as invertase by non-parenchymal liver cells. After injection of 131I-invertase into a rat in vivo, about one-half of the labelled protein was recovered in the hepatocytes. On a per-cell basis, each endothelial cell contained 3-4 times as much radioactivity as did the hepatocytes. On fractionation of hepatocytes in sucrose gradients, invertase showed a different intracellular distribution from that of asialo-fetuin, in that invertase moved much faster into that region of the gradient where the lysosomes were recovered. This indicates that invertase and asialo-fetuin are not transported intracellularly by identical mechanisms.     

Rapid subcellular fractionation of the rat liver endocytic compartments involved in transcytosis of polymeric immunoglobulin A and endocytosis of asialofetuin.

The distributions of two endocytosed radiolabelled ligands (polymeric immunoglobulin A and asialofetuin) in rat liver endocytic compartments were investigated by using rapid subcellular fractionation of post-mitochondrial supernatants on vertical density gradients of Ficoll or Nycodenz. Two endocytic compartments were identified, both ligands being initially associated with a light endocytic-vesicle fraction on Ficoll gradients, asialofetuin then accumulating in denser endosomes before transfer to lysosomes for degradation.     

Fluid phase endocytosis of [125I]iodixanol in rat liver parenchymal, endothelial and Kupffer cells

Endocytosis of [125I]iodixanol was studied in vivo and in vitro in rat liver cells to determine fluid phase endocytic activity in different liver cells (hepatocytes, Kupffer cells and endothelial cells). The Kupffer cells were more active in the uptake of [l25I]iodixanol than parenchymal cells or endothelial cells. Inhibition of endocytic uptake via clathrin-coated pits (by potassium depletion and hypertonic medium) reduced uptake of [125I]iodixanol much more in Kupffer cells and endothelial cells than in hepatocytes. To gain further information about the importance of clathrin-mediated fluid phase endocytosis, the expression of proteins known to be components of the endocytic machinery was investigated. Using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting, endothelial cells and Kupffer cells were found to express approximately fourfold more rab4, rab5 and rab7 than parenchymal cells, while clathrin was expressed at a higher level in endothelial cells than in Kupffer cells and hepatocytes. Using electron microscopy it was shown that liver endothelial cells contained approximately twice as many coated pits per membrane unit than the parenchymal and Kupffer cells, thus confirming the immunoblotting results concerning clathrin expression. Electron microscopy on isolated liver cells following fluid phase uptake of horseradish peroxidase (HRP) showed that HRP-containing organelles had a different morphology in the different cell types: In the liver endothelial cells HRP was in small, tubular endosomes, while in Kupffer cells HRP was mainly found in larger structures, reminiscent of macropinosomes. Parenchymal cells contained HRP in small vacuolar endosomes with a punctuated distribution. In conclusion, we find that the Kupffer cells and the endothelial cells have a higher pinocytic activity than the hepatocytes. The hepatocytes do, however, account for most of the total hepatic uptake. The fluid phase endocytosis in liver endothelial cells depends mainly on clathrin-mediated endocytosis, while the parenchymal cells have additional clathrin-independent mechanisms that may play an important role in the uptake of plasma membrane components. In the Kupffer cells the major uptake of fluid phase markers seems to take place via a macropinocytic mechanism.     

Transferrin receptors and cation-independent mannose-6-phosphate receptors deliver their ligands to two distinct subpopulations of multivesicular endosomes

The distribution of transferrin receptors (Tf-R) was determined in Clone 9 hepatocytes and compared to that of 215 kDa, cation-independent mannose-6-phosphate receptors (M6P-R) by double labeling. Cells were allowed to take up exogenous human transferrin (Tf) for 5 to 30 min, after which Tf, Tf-R, and M6P-R were localized by immunofluorescence using specific antibodies. All these proteins were found to be concentrated in the juxtanuclear or Golgi region. When Clone 9 cells were treated with NH4Cl to trap M6P-R in endosomes (Brown, W. J., J. Goodhouse, M. G. Farquhar: J. Cell Biol. 103, 1235-1247 (1986)), the distribution of the two receptors differed: Tf-R remained the same as in controls, but M6P-R were localized in large vacuolated endosomes. To carry out double labeling experiments at the electron microscope level, transferrin gold conjugates (Tf-Au) were prepared, and M6P-R were detected by immunoperoxidase labeling. Tf-Au binding to the cell surface was specific as it was reduced approximately 70 to 79% in the presence of excess native Tf. When Clone 9 cells were incubated with Tf-Au at 37 degrees C for 5 to 30 min, or binding of Tf-Au was carried out at 4 degrees C followed by warming to 37 degrees C, Tf-Au was found within a peripheral tubulovesicular network and within multivesicular endosomes that were not labeled with anti-M6P-R. Other multivesicular endosomes of similar size and morphology were heavily labeled for M6P-R but contained little or no Tf-Au. Tf-Au and M6P-R were also found in separate endosomes in cells treated with NH4Cl. Native Tf was localized in the same compartments as Tf-Au by immunoperoxidase labeling of both Clone 9 cells and mouse myeloma cells. We conclude that in Clone 9 hepatocytes, Tf/Tf-R internalized from the cell surface and M6P-R bearing newly synthesized lysosomal enzymes from the Golgi deliver their ligands to two different subpopulations of multivesicular endosomes. The endosomal subpopulation visited by Tf/Tf-R is known to correspond kinetically to early endosomes. The endosomal subpopulation heavily labeled for M6P-R presumably represent a later endosomal compartment which serves as the junction point where endocytosed ligands and newly synthesized lysosomal enzymes enroute to lysosomes meet.