Receptors allow the cells to recognize specific ligands and to receive extracellular messages. They can be classified into five families: 1) receptors for lipidic or lipophilic ligands; 2) the seven transmembrane receptors which mediate their messages by transduction through the activation of G-proteins, effectors and second messengers to amplify the response; 3) receptors which present an enzymatic activity on their transmembrane domains; 4) channel-receptors, transmembrane oligomeric molecules which let ions flow into the cell and 5) receptors which role is to internalize ligands, whatever their various functions. In parallel a concept of membrane plasticity was developed: vesicles are constantly formed from the plasma membrane, addressing complexes of ligand-receptors to specific intracellular compartments. This receptor-mediated endocytosis of ligand plays a critical role in regulating the number of a given receptor at the plasma membrane and in the cellular uptake of nutrients, growth factors and hormones. Many pathways exist for these transports but little is known about the signals which select the ligands or the receptors and direct them to their appropriate intracellular destination.

The hepatic asialoglycoprotein receptor is the first studied mammalian lectin. Modulations in vivo by diabetes and in vitro by the carboxylic ionophore monensin gave rise to similar apparent alterations on its biosynthesis, structure and ligand binding capacity. In normal rats, the receptor (whether purified by ligand or antibody-affinity chromatography) presented a similar pattern in SDS-PAGE analysis, with a major 42-kDa band and two minor ones (49 and 52-54 kDa). In diabetic rats, a new 38-kDa band appeared, but only after antibody-affinity purification. In vitro biosynthesis of the receptor by normal hepatocytes in the presence of 35S-methionine showed that this 38-kDa band was present at the end of a 30-min pulse but decreased during a 180-min chase, in association with an increase in the major 42-kDa band. In diabetic cells, this evolution was retarded. Using a 30-min pulse followed by a 120-min chase in the presence of 100 microM monensin, we showed that this carboxylic ionophore had similar effects on diabetes, leading to a delay in the maturation process of the 42-kDa band and the persistent emergence of the 38-kDa species. Allowing incubation in the presence of 25 or 100 microM monensin, we observed a decrease in the number of ligand binding sites both at the surface (40%) and within the cell (28%). In hepatocytes from diabetic rats, monensin showed no additional effect on the partial diabetes-induced inactivation.

Both in vivo and in vitro models have certain disadvantages for the study of the chronic hepatotoxicity of drugs. The aim of this work was to evaluate a new approach based on an in vivo/in vitro model. After chronic in vivo treatment of rats with Vincamine and Vindeburnol (an eburnamenine derivative which exhibits hepatotoxic properties in man) liver cells were isolated, and functional and metabolic disorders (metabolic utilization of fructose and protein biosynthesis) were studied to determine injury. The results showed no modification of blood parameters, but a direct relationship between the dose of Vindeburnol administered in vivo and the metabolic disorders observed in vitro, evidencing the high sensitivity and reliability of this model.

The hepatic asialoglycoprotein receptor is a membrane glycoprotein used as a model to study receptormediated endocytosis. In order to examine the ability of second messengers to modulate intracellular trafficking, we performed a comparative study on normal and diabetic rat hepatocytes exploring the effects of an in vivo modulation, streptozotocin-diabetes, and an in vitro modulator, vasopressin, which transduces signals via the phosphoinositide pathway. We studied three main experimental aspects: (1) constitutive endocytosis, (2) continuous ligand flux, and (3) a synchronous wave of ligand. In normal cells, vasopressin decreased ligand-binding capacity by 20%, without altering the mechanism of internalization, and decreased the level of degradation, without affecting the distribution of degradation products. Diabetic cells were characterized by a 50% decrease in cell-surface and intracellular receptor ligand-binding capacity, slowed internalization of a synchronous wave of ligand, and markedly reduced degradation with an altered distribution of degraded products. Vasopressin had no additive effect on the modification induced by diabetes. These results suggest that second messengers generated by hormones play a role in the regulation of receptor-mediated endocytosis. They also confirm that receptors are subdivided into those susceptible to modulation of any kind and those insensitive to modulation, although the boundary between the two subsets is variable.

We have studied the effects of two modulations — streptozotocin-induced diabetes in vivo, and the presence of the carboxylic proton ionophore monensin in vitro — on the degradation of3H-asialoorosomucoid ligand in isolated rat hepatocytes. The ligand was internalized by means of a synchronous wave procedure. Diabetes was associated with a marked decrease in the amount of total degraded radioactive ligand compared to that in normal cells (3.6% and 37.3% of internalized ligand respectively, at 60 min), together with increased secretion of degradation products into the incubation medium (87% and 46.3% of the total degraded ligand was secreted by diabetic and normal cells, respectively). Monensin induced similar effects in normal cells, but had no apparent effect in diabetic cells.

The influence of experimental conditions was investigated in pulse and pulse-chase experiments involving L-[35s]methionine incorporation by isolated rat hepatocytes. The incorporation of the labelled amino acid must be linear as a function of dose and time, and similar hepatocyte densities should be used to compare radioactive uptake. High hepatocyte densities give greater precision. In pulse-chase experiments, it was shown that a pool of free radioactivity moves progressively during the chase phase from the intracellular compartment. Protein-associated radioactivity must therefore be expressed as a function of both total radioactive uptake and total labelled protein.

We have characterized a binding site for galactosyl terminal glycoproteins in hepatocytes isolated from human biopsies. The binding of asialoorosomucoid on hepatocytes previously treated by Triton X-100 was saturable, calcium-dependent and highly affine (K(a) = 1.11 +/- 0.87.10(9) M-1) thus corresponding to a ligand-receptor binding. The total number of receptors in the normal human liver was 140 000 +/- 65 000 sites per cell. This corresponded to the value obtained in the human hepatoma cell line HepG2, but was significantly lower than for isolated rat hepatocytes. Furthermore, in hepatocytes isolated from livers with histological features of either fibrosis, cirrhosis, hepatocarcinoma with cirrhosis or nodular regenerative hyperplasia, the number of asialoglycoprotein receptors per cell was increased, while the binding affinity was unchanged.

alpha 1-Acid glycoprotein concentrations and reactivity to concanavalin A were measured in maternal and fetal serum and amniotic fluid obtained from 24 women undergoing diagnostic cordocentesis at 20 to 33 wk gestation and in 30 additional fetal sera (19 to 34 weeks gestation). Maternal alpha 1-acid glycoprotein serum levels were five to ten times higher than fetal and amniotic levels. Fetal alpha 1-acid glycoprotein levels were found to increase with advancing gestational age. Using crossed immunoaffino electrophoresis with concanavalin A, alpha 1-acid glycoprotein patterns were identical in maternal serum and amniotic fluid but totally different in fetal serum. The fetal concanavalin A pattern changed progressively during fetal life towards that of the newborn. These data confirm earlier assumptions of fetal synthesis of alpha 1-acid glycoprotein and provide normal reference values for alpha 1-acid glycoprotein in fetal serum. In addition, the specific fetal concanavalin A pattern indicates that the alpha 1-acid glycoprotein glycosylation process during fetal life differs from that in post-natal life.

The ability of second messengers to modulate receptor-mediated endocytosis was studied on isolated rat hepatocytes. A 20-min preincubation with vasopressin was used as a modulation. We observed a 20% inactivation of both surface and intracellular receptors, with no change in the affinity of those remaining active. The internalization and dissociation of a synchronous wave of ligand was not affected, but its degradation was partially inhibited. Our observations suggest that second messengers such as intracellular calcium and diacylglycerol play a complex role in the intracellular trafficking associated with endocytosis.

This is a noncompetitive enzyme-linked immunosorbent assay for measuring low concentrations (2 to 100 micrograms/L) of human alpha 1-acid glycoprotein (AGP; orosomucoid). The method is based on a simple "sandwich" technique involving polyclonal rabbit antisera against AGP. Mean within-run and total (between-run) CVs were 6.2% and 9.7%, respectively. Analytical recovery, tested in various biological fluids, averaged 101%. The technique has been successfully applied to diluted biological fluids such as bronchoalveolar lavage, cerebrospinal and amniotic fluids, and hepatocyte-culture supernates. Because of its analytical validity and the commercial availability of the reagents, this assay is suitable for large-scale determinations of AGP concentrations in those biological fluids in which its concentration is relatively low.

Based on the affinity for concanavalin A (Con A), human alpha 1-acid glycoprotein (AGP) can be separated by chromatography on Con A-Sepharose gel into three variants: Con A unreactive AGP, Con A weakly reactive AGP, and Con A strongly reactive AGP. When exposed to native AGP or to its glycan variants, murine peritoneal macrophages released a factor that inhibited the interleukin-1 (IL-1) proliferative activity as measured in terms of the thymocyte comitogenic assay. Con A unreactive AGP, which contains tri- and tetraantennary glycans and no biantennae, proved to be more effective than Con A weakly and Con A strongly reactive variants, which contain one and two diantennary glycans, respectively. The inhibitory effect was not a function of the negative charge related to the sialyl residues and was not mediated by the mannosyl-fucosyl receptor.

After exposure to a concanavalin A (Con A)-unreactive variant of α1-acid glycoprotein (AGP), macrophages released an inhibitor of interleukin-1 (IL-1) proliferative activity in the thymocyte comitogenic assay. This effect was observed with AGP concentrations above 100 μg/ml in the macrophage supernatant and would appear to be mediated by the macrophages, since native AGP had no activity on thymocyte proliferation. Preliminary physicochemical characterization showed that the factor was partially resistant to heating, undialyzable, and eluted with an apparent molecular mass of 50–100 kDa when subjected to Sephacryl S-200 chromatography. Murine IL-1 and human (h) recombinant (r) IL-1 were affected by this factor to the same extent. IL-1 and IL-2 co-induced thymocyte proliferation, which is mitogen-independent, was also inhibited, whereas hrIL-2 activity was not suppressed when assayed in thymocytes with PHA at a submitogenic concentration or in CTLL cells. The factor did not interfere with TNFα or hrIL-6 activity when tested against their specific cell line. These data indicate that the inhibitor may act specifically against IL-1 activity and further elucidate the possible role of AGP in the modulation of IL-1 activity via the secretion of an inhibitor.

The metabolic turnover of the Hepatic Binding Protein (HBP) was investigated in streptozotocin-diabetic rats. We have already shown that diabetes induced a decreased ligand binding capacity while the immunoreactive HBP was normal. To explore the eventual modifications due to diabetic state upon the turnover of HBP, we followed the in vivo degradation of HBP and its biosynthesis in vitro. After in vivo labelling with L-[3H] leucine and purification of HBP from rat livers, we found a 20% decrease in diabetic HBP half-life. By in vitro incubations of freshly isolated hepatocytes and a 2 h-pulse in the presence of L-[35S] methionine, we showed that diabetes provokes an increased uptake of L-[35S] methionine in hepatocytes allowing an augmented synthesis of HBP although the L-[35S] methionine incorporation into total proteins was less efficient.

Various studies have shown that oligosaccharides play an important role in the intracellular transport and secretion of glycoproteins. We show here a difference in the rate of secretion of two mature glycoforms of a single protein, alpha 1-acid glycoprotein. This indicates the existence of kinetically different pathways for these two forms for transport from the medial Golgi to the extracellular medium.

The binding of asialoglycoproteins by hepatic binding protein was studied in freshly isolated hepatocytes from genetically diabetic BB Wistar rats. The number of cell surface asialoglycoprotein receptors was dramatically decreased (58,000±38,000 for diabetic rats compared to 267,000±70,000 for normal rats), while the association equilibrium constant was not changed. These results parallel those obtained with streptozotocin-diabetic rats and support the hypothesis that insulin deprivation is responsible for the decrease in the receptor number.

Most purification procedures used previously to isolate α1-acid glycoprotein (AGP) from plasma can lead to some alterations in its carbohydrate moiety. An immunoaffinity chromatographic method is proposed for purifying in one step rat plasma AGP without any detectable modification of its glycan moiety. Crossed immunoaffinoelectrophoresis with concanavalin A before and after purification showed identical patterns, suggesting no glycan selection during the purification. In the same way no desialylation occurred during the purification step. This immunoaffinity chromatographic procedure provided evidence of a decreased level of fucosyl residues in turpentine oil rat plasma AGP compared with normal rat plasma AGP.

To determine whether alterations of the carbohydrate moiety of human alpha 1-acid glycoprotein constitute a marker of hepatic damage we studied purified alpha 1-acid glycoprotein from healthy individuals and two groups of patients with benign liver diseases: alcoholic cirrhosis and acute viral hepatitis. The results indicate: (1) increased concanavalin A-non reactive forms in cirrhosis and hepatitis, (2) a markedly increased proportion of fucosyl residues in all cirrhotic and some hepatitis patients. Although hyperfucosylation is generally considered to be a tumor marker, the observation here in the two benign liver diseases indicates that an increased fucosyl content should be considered as a more general expression of pathological glycoconjugate metabolism.

Hepatocytes were isolated from adult rats at various times after subcutaneous injection of turpentine (1 ml). The affinity to concanavalin A (Con A) of alpha 1-acid glycoprotein (AGP) and the intracellular content and rate of secretion of AGP and albumin were evaluated over a period of 19 days. Inflamed hepatocytes secreted mainly the Con A-reactive form of AGP whereas control hepatocytes secreted a higher amount of the Con A-non-reactive form. The intracellular content and rate of secretion of AGP by inflamed hepatocytes increased markedly whereas those of albumin decreased. However, when the residence time (ratio of intracellular content to rate of secretion) was evaluated, it appeared that the efficiency of secretion of both proteins was higher than in control hepatocytes. The changes in the affinity of AGP to Con A and in the secretion of AGP and albumin were reversible. These findings indicate that acute inflammation leads to posttranslational alterations during the intracellular transit of these secretory proteins.