The role of non-parenchymal cells in liver metabolism

Protein phosphatases regulate the liver microenvironment in the development of hepatocellular carcinoma

The liver is a complicated heterogeneous organ composed of different cells. Parenchymal cells called hepatocytes and various nonparenchymal cells, including immune cells and stromal cells, are distributed in liver lobules with hepatic architecture. They interact with each other to compose the liver microenvironment and determine its characteristics. Although the liver microenvironment maintains liver homeostasis and function under healthy conditions, it also shows proinflammatory and profibrogenic characteristics that can induce the progression of hepatitis and hepatic fibrosis, eventually changing to a protumoral microenvironment that contributes to the development of hepatocellular carcinoma (HCC). According to recent studies, phosphatases are involved in liver diseases and HCC development by regulating protein phosphorylation in intracellular signaling pathways and changing the activities and characteristics of liver cells. Therefore, this review aims to highlight the importance of protein phosphatases in HCC development and in the regulation of the cellular components in the liver microenvironment and to show their significance as therapeutic targets.

Mechanisms determining phenotypic heterogeneity of hepatocytes

This review summarizes results of biochemical and immunohistochemical studies indicating the existence of functional heterogeneity of hepatocytes depending on their localization in the hepatic acinus; this determines characteristic features of metabolism of carbohydrates, lipids, and xenobiotics. The physiological significance of hepatocyte heterogeneity is discussed. According to the proposed model of intercellular communication, the metabolic specialization of hepatocytes is determined by secretory activity of hepatic resident macrophages (Kupffer cells) localized mainly in the periportal zone of the liver acinus. Macrophages participate in secretion of a wide spectrum of intercellular mediators (cytokines, prostaglandins, growth factors) and also in metabolism of numerous blood metabolites and biologically active substances (hormones, lipoproteins, etc.). In the sinusoid and in the space of Disse (also known as perisinusoidal space) they form a concentration gradient of regulatory factors and metabolites inducing the phenotypic differences between hepatocytes.

Evaluation of inhibitory guanine nucleotide regulatory protein Gi function in hepatocyte and liver membranes from obese Zucker (fa/fa) rats and their lean (Fa/?) littermates

Previous studies have shown that hepatocyte and liver membranes from insulin resistant animals exhibit an impairment of inhibitory guanine nucleotide binding regulatory protein, Gi function, such that a Gi defect may contribute towards the diabetic syndrome. In the current studies, it is shown that the demonstration of Gi activity in liver and hepatocyte membranes is dependent critically on the membrane preparation technique. A technique is defined that allows functional Gi activity to be demonstrated in liver and hepatocyte membranes from both lean (Fa/?) and obese (fa/fa) Zucker rats. Consequently, previous reports on the loss of Gi function in insulin resistant states require revaluation.

Diabetes-induced changes in guanine-nucleotide-regulatory-protein mRNA detected using synthetic oligonucleotide probes

Synthetic oligonucleotide probes were designed to detect the alpha-subunits of the guanine-nucleotide-regulatory proteins (G-proteins) Gi-1, Gi-2, Gi-3 and Gs (Gi is inhibitory and Gs is stimulatory). Each probe detected a single major mRNA species in Northern blots of RNA extracted from a variety of tissues. A probe was designed to identify the two forms of G-protein beta-subunits, beta 1 and beta 2. This probe hybridised with a single 1.8-kb transcript (beta 2) in RNA from all tissues studied except for brain, where a less-abundant 3.4-kb transcript (beta 1) was also detected. These probes were used to assess whether the induction of diabetes, using streptozotocin, altered the levels of mRNA coding for specific G-protein components. In hepatocytes, diabetes caused a significant reduction in the number of transcripts coding for alpha-Gs, alpha-Gi-2 and alpha-Gi-3; mRNA for alpha-Gi-1 was undectable. In adipocytes, diabetes increased dramatically the mRNA coding for alpha-Gi-1 and alpha-Gi-3, whilst no significant changes occurred in the fractions coding for alpha-Gi-2 and alpha-Gs. No significant changes in the mRNA coding for G-protein alpha-subunits were observed in either brain, heart, skeletal muscle or kidney. Diabetes did not cause any significant changes in the mRNA coding for beta 2 in any tissue or cell population studied. Such results on the relative levels of mRNA encoding G-protein components was obtained by comparing equal amounts of total RNA from tissues of control and diabetic animals. G-protein mRNA levels were expressed relative to ribosomal 28S RNA levels and, in some instances, relative to transcripts for a structural protein called CHO-B. The total cellular levels of both RNA and DNA were assessed in the various tissues and cells studied. Major falls in RNA levels/cell appeared to occur in hepatocytes and to a lesser extent in adipocytes and skeletal muscle. Thus major reductions in G-protein transcripts occurred in hepatocytes. The detected changes in G-protein mRNA are discussed in relation to the available evidence on G-protein expression. We suggest that diabetes causes tissue-specific changes in the levels of mRNA for particular G-protein species; this may have consequences for the functioning of cellular signal-transduction mechanisms in the affected tissues.

In vivo binding and uptake of low-density lipoprotein-gold- and albumin-gold conjugates by parenchymal and sinusoidal cells of the fetal rat liver

To elucidate the participation of fetal rat liver cells in the receptor-mediated internalization of low-density lipoproteins (LDL), rat fetuses were injected with either LDL-gold or albumin-gold conjugates. The degree of binding and uptake of LDL-gold and albumin-gold by parenchymal and sinusoidal cells of the fetal rat liver differs markedly. Endothelial cells exhibit low LDL-gold uptake. In contrast, parenchymal cells internalize LDL-gold more actively (45 +/- 8 LDL conjugates/100 micrometers2 cytoplasm within 60 min). Kupffer cells exceed this value by a factor of 20. The uptake of albumin-gold by endothelial and Kupffer cells is high, whereas it is extremely low in parenchymal cells. Estradiol pretreatment causes a significant doubling (p less than 0.05) of the LDL-gold particle density/100 micrometers2 cytoplasm both in parenchymal and Kupffer cells, whereas estradiol has no effect on the albumin uptake. The results strongly indicate that LDL uptake by parenchymal and Kupffer cells in the fetal rat liver is mediated by estrogen-inducible receptors, which may correspond to B, E receptors in the adult liver.

Liver parenchymal cells differ from the fat-storing cells in their lipid composition

The neutral lipid and phospholipid compositions of purified sinusoidal (fat-storing, endothelial and Kupffer) cells, parenchymal cells and liver homogenates were determined by thin layer chromatography. In addition, the retinoid content of the same purified cell populations was determined by high performance liquid chromatography. From each cell type, both a lipid droplet fraction and a pellet fraction (containing the majority of the remaining cell organelles) were prepared by differential centrifugation. Electron microscopic analysis showed that lipid droplets isolated from fat-storing cells were larger (up to 8 microns) than those isolated from parenchymal cells (up to 2.5 microns). Moreover, the parenchymal lipid droplets seemed to be surrounded by a membranous structure, while the fat-storing lipid droplets seemed not to be. Both fat-storing and parenchymal cells contained high concentrations of neutral lipids, 57.9 micrograms and 71.0 micrograms/10(6) cells, respectively, while endothelial and Kupffer cells contained only 8.6 micrograms and 13.8 micrograms/10(6) cells of neutral lipids, respectively. Sixty-five percent of fat-storing cell lipid droplet fractions comprised esters of retinol and cholesterol. This combined ester fraction contained mainly retinyl esters. In addition, considerable quantities (20%) of triglycerides were present. Parenchymal cell lipid droplet fractions comprised triglycerides (62%) and cholesteryl esters (up to 30%). The pellet fractions prepared from all four cell types consisted mainly of cholesterol (41-67%) and free fatty acids (20-28%). The phospholipid content was much higher in parenchymal cells than in the sinusoidal liver cell types. The relative proportions of the four major phospholipid classes were comparable in all liver cell types analyzed.(ABSTRACT TRUNCATED AT 250 WORDS)

Isolation and culture of biliary epithelial cells from the biliary tract fraction of normal rats

A method has been developed for the isolation of a population of cells enriched in epithelial lining cells from the bile ducts of normal rats. The procedure utilized digestion by pronase of the white strands of biliary and connective tissue which remained after hepatocytes had been mechanically removed from collagenase-perfused liver. The resulting cell population was enriched in cells whose ultrastructure resembled that of the epithelial cells of intrahepatic bile ducts. Contamination with hepatocytes, hepatocyte nuclei and erythrocytes was less than 2%. The cells have been maintained in short-term culture. The major morphological change during the first 2 days of culture was proliferation of microvilli, but cell protein composition was unchanged when analysed by polyacrylamide gel electrophoresis. A rabbit antiserum against bovine hoof prekeratin was used to immunohistochemically stain the intermediate filaments of biliary epithelium and was shown to stain more than 90% of the cells in the isolated cell population.

Cellular distribution of lysosomal hydrolase activities in the regenerating rat liver

Cathepsins B and D, beta-galactosidase, and acid phosphatase activities were found to be decreased in the regenerating rat liver, the reduction being maximal around the peak of hepatocyte mitoses (30 h). To investigate whether these changes could be heterogeneously distributed among hepatic cells, total cell populations from control or two-thirds hepatectomized rat livers were dissociated by the collagenase perfusion technique and analysed by different procedures. Isopycnic centrifugation in a Metrizamide gradient satisfactorily resolved hepatocytes and non-parenchymal cells from control animals but was not adequate when applied to 30-h regenerating liver cells. Colchicine treatment of the hepatectomized animals, resulted in substantial accumulation of phase M-hepatocytes. Subpopulations considerably enriched in fast-sedimenting phase M-cells were obtained by sedimentation at 1 g of the total liver cell population, and subsequently analysed by isopycnic equilibration. Phase M-hepatocytes were shown to have markedly reduced levels of beta-galactosidase, acid phosphatase, and cathepsin B activities in comparison, not only with control hepatocytes, but also with those parenchymal cells which were not metaphase-arrested in the same regenerating livers. Therefore, in partially-hepatectomized rats, hepatocytes progressing up to metaphase in the first mitotic cycle exhibited a selective depletion of lysosomal enzyme activities. The mechanism(s) underlying this change remain(s) presently unknown.

Rapid transport of fatty acids from rat liver endothelial to parenchymal cells after uptake of cholesteryl ester-labeled acetylated LDL

Acetylated low-density lipoprotein (acetyl-LDL) radiolabeled in the oleate moiety of cholesteryloleate was injected into rats. Isolation of the various liver cell types at different times after acetyl-LDL injection by a low-temperature procedure allowed the intrahepatic metabolism of the oleate moiety to be followed in vivo. The cholesteryloleate radioactivity is rapidly cleared from the circulation and at 5 min after injection recovered into parenchymal and endothelial liver cells, mainly as cholesteryloleate ester. At longer time intervals after injection, the amount of cholesteryl esters associated with the endothelial cells was sharply decreased and the [14C]oleate was redistributed within the liver and mainly recovered in the parenchymal cells. The cholesteryl ester initially directly taken up by the parenchymal cells was also rapidly hydrolysed but, in contrast to the endothelial cells, the [14C]oleate remained inside the cells and was incorporated into triacylglycerols and phospholipids. The 14C radioactivity in parenchymal cells taken up between 5 and 30 min after injection of the cholesteryl [14C]oleate-labeled acetyl-LDL (transported as oleate from endothelial cells), followed a similar metabolic route as the amount which was directly associated to parenchymal cells. The data indicate that the liver and, in particular, the liver endothelial cell has the full capacity to rapidly catabolize modified lipoproteins. In this catabolism, the liver functions as an integrated organ in which fatty acids, formed from cholesteryl esters in endothelial cells, are rapidly transported to parenchymal cells, indicating the concept of metabolic cooperation between the various liver cell types.

Localization of cathepsin D in rat liver. Immunocytochemical study using post-embedding immunoenzyme and protein A-gold techniques

Light and electron microscopic localization of cathepsin D in rat liver was investigated by post-embedding immunoenzyme and protein A-gold techniques. By light microscopy, cytoplasmic granules of parenchymal cells and Kupffer cells were stained for cathepsin D. Weak staining was also noted in sinusoidal endothelial cells. In the parenchymal cells many of positive granules located around bile canaliculi. In the Kupffer cells and the endothelial cells, diffuse staining was noted in the cytoplasm in addition to granular staining. By electron microscopy, gold particles representing the antigenic sites for cathepsin D were seen in typical secondary lysosomes and some multivesicular bodies of the parenchymal cells and Kupffer cells. The lysosomes of the endothelial cells and fat-storing cells were weakly labeled. Quantitative analysis of the labeling density in the lysosomes of these three types of cells demonstrated that the lysosomes of parenchymal cells and Kupffer cells are main containers of cathepsin D in rat liver. The results suggest that cathepsin D functions in the intracellular digestive system of parenchymal cells and Kupffer cells but not so much in that of the endothelial cells.

Histochemical and immunohistochemical localization of hexokinase isoenzymes in normal rat liver

Histochemical and immunohistochemical procedures have been used to examine the localization of three of the four hexokinase isoenzymes present in the liver of fed female Wistar rats. Distinctive distribution patterns were found for hexokinase type I and glucokinase but hexokinase type II was not detectable. Hexokinase type I was identified in sinusoidal cells and in bile duct epithelia, nerves and arteries in the portal triad. Glucokinase, the major isoenzyme, was confined to parenchymal cells where it was present in much higher amounts in perivenous compared with periportal hepatocytes. Staining within these two zones was not homogeneous and each had a mosaic appearance caused by the presence of a few hepatocytes containing little or no glucokinase amongst the majority of darkly stained cells in perivenous areas and a few darkly stained cells amongst the majority of unstained cells in periportal areas. Hence, hepatocytes in situ are a strikingly heterogeneous population of cells. Their metabolic status cannot be controlled simply by the differential supply of oxygen, substrates and hormones to different regions of the liver acini as proposed in the metabolic zonation model. Phenotypic differences may exist between cells within a given metabolic zone which influence their ability to respond to different environmental conditions.

Altered distribution of hexokinase and glucokinase between parenchymal and nonparenchymal cells of rat liver after methapyrilene intoxication

The cell number as well as the hexokinase and glucokinase activity of liver parenchymal and nonparenchymal cells were studied in methapyrilene treated rats. The number of nonparenchymal cells was doubled after treatment with methapyrilene for two weeks while that of hepatocytes remained constant. The hexokinase activity was increased fourfold in the nonparenchymal cell fraction while it was unchanged in the parenchymal cells. The glucokinase activity was decreased in the hepatocytes to one third. Hence, the increased hexokinase activity was due to a proliferation of nonparenchymal cells rather than to a toxic dedifferentiation of hepatocytes.

How accurate are computed tomographic scans in assessment of changes in tumor size?

The computed tomographic scan plays an integral part in the diagnosis and management of tumors; however, its potential has not yet been fully exploited. With a computer-assisted volume determination method, the reproducibility of derived volume calculations was assessed, and radiologists' standard interpretations of interval change on serial scans were compared with the investigators' calculations of tumor volume change. Interobserver reproducibility of tumor volume calculations of the mean of two repeated volume determinations was satisfactory (mean of 3 percent, median of 1 percent). There were 29 comparisons (47 scans of 19 patients with liver tumors) of computed tumor changes with the radiologists' computed tomographic reports of consecutive scans. In only 41 percent (12 of 29) of the cases did the radiologists' interpretations and the computer-assisted volume determinations agree. It is concluded that objective computer-assisted volume determination provides a potentially more sensitive assessment of tumor change and that such precise, specific, reproducible determination of tumor volume should further clinical research and improve patient care.

Intracellular transport and degradation of 125I-Labelled denatured serum albumin in isolated nonparenchymal rat liver cells

The intracellular movement, following uptake of 125I-labelled denatured serum albumin into nonparenchymal liver cells, was followed by means of subcellular fractionation. Isolated nonparenchymal rat liver cells were prepared by means of differential centrifugation. The cells were homogenized in a sonifier and the cytoplasmic extract subjected to isopycnic centrifugation in a sucrose gradient. The intracellular movement of the labelled albumin was followed by comparing the distribution profile of radioactivity in the sucrose gradient with those of marker enzymes for plasma membrane and lysosomes. The distribution profiles for radioactivity after the cells had been exposed to the labelled denatured albumin for different time periods indicated that the radioactivity was first associated with subcellular fractions of lower modal densities than the lysosomes. With time of incubation the radioactivity moved towards higher densities. After prolonged incubations in the absence of extracellular labelled denatured albumin the radioactivity peak coincided with that of the lysosomal marker beta-acetylglucosaminidase. When the cells were treated with the lysosomal inhibitor leupeptin, degradation of the labelled albumin was decreased, resulting in a massive intracellular accumulation of radioactivity. The radioactivity peak coincided with the peak of activity for the lysosomal marker beta-acetylglucosaminidase, suggesting lysosomal degradation.

Drug-induced lipid peroxidation in mice--III. Glutathione content of liver, kidney and spleen after intravenous administration of free and liposomally entrapped glutathione

The half-life of extracellular glutathione was found to be 1.9 min in fed mice with a hepatic glutathione content of 44 +/- 10 nmol glutathione per mg protein. It was 4.9 min in animals that had been fed for 48 hr a liquid sucrose diet resulting in a decreased hepatic glutathione of 25 +/- 7 nmol/mg. A single intravenous injection of 16.2 mumol liposomally entrapped glutathione led to an increase in hepatic glutathione to 45 nmol/mg in the sucrose-fed mice after 2 hr and had no effect in the fed group. The spleen glutathione content reached a maximum at 30 min after injection in both groups. The maximum uptake into liver was 21% of the applied dose, into the spleen 7% and into the kidneys 2.4%. Injection of glutathione in solution led to a similar increase of hepatic glutathione as observed with GSH-containing liposomes, while liposomes filled with the constituent amino acids had only a marginal effect. The spleen took up only liposomal GSH. In contrast, the kidney glutathione content increased within 10 min up to 150% upon injection of free glutathione. The findings are consistent with a rapid hydrolysis of extracellular free glutathione followed by an interorgan turnover utilizing the constituent amino acids for resynthesis in the liver. Pretreatment of the animals with the glutathione synthesis inhibitor buthionine sulfoximine essentially abolished the hepatic glutathione increase upon treatment with GSH-liposomes or with the free compound. The finding that only liposomally entrapped glutathione protects mice against liver necrosis induced by highly dosed paracetamol is discussed with respect to differential uptake and distribution of GSH-liposomes in the liver.

Involvement of prostaglandin E and adenosine 3', 5'-monophosphate in lipopolysaccharide-stimulated collagenase release by rat Kupffer cells

Kupffer cells exposed to bacterial lipopolysaccharide in vitro synthesized collagenase and released the major portion of it into the extracellular space while the intracellular level of enzyme was not altered significantly. Cycloheximide prevented the appearance of collagenase in the medium indicating de novo synthesis. Indomethacin, an inhibitor of cyclooxygenase, also blocked collagenase synthesis. In line with this observation. Kupffer cells were found to synthesize substantial amounts of prostaglandin E2 when exposed to lipopolysaccharide; concomitantly, cellular cAMP levels were increased. Indomethacin was shown to abolish the stimulated cAMP formation. Addition to the culture medium of cAMP or dibutyryladenosine 3', 5'-monophosphate as well as of prostaglandin E2 or, to a lesser extent, prostaglandin E1 allowed indomethacin-inhibited cells to resume the production of collagenase. It is proposed that in rat Kupffer cells lipopolysaccharide-elicited collagenase synthesis and excretion is mediated sequentially by stimulated production of prostaglandin E2, enhanced adenylate cyclase activity and increased intracellular cAMP levels.

Relative contributions of parenchymal and non-parenchymal (sinusoidal) liver cells in the uptake of chylomicron remnants

The relative contributions of parenchymal cells and non-parenchymal (sinusoidal) cells to the in vivo hepatic uptake of chylomicron remnants was measured 30 min after intravenous injection into rats. The chylomicron remnants were labeled with [3H]leucine, which was almost exclusively present in apolipoprotein B. The isolated non-parenchymal cells (a mixture of Kupffer cells and endothelial cells) contained 6.7 times more apolipoprotein B radioactivity per mg cell protein than the isolated parenchymal cells. It was calculated that the contributions of non-parenchymal and parenchymal liver cells to the total hepatic uptake of chylomicron remnants are 35% and 65%, respectively.

Post-transcriptional regulation of gene expression in guinea pig tissues

The formation of individual functional mRNA sequences in higher organisms requires many steps in addition to transcription. These include RNA splicing, polyadenylation, base modification, transport from nucleus to cytoplasm and assembly into polyribosomes. Various control mechanisms must also operate. These will function on a quantitative basis to account for the differing frequency of the various classes of cytoplasmic mRNAs, and also on a qualitative basis, because in higher organisms not all the nuclear poly(A)-containing RNA molecules are found in a cytoplasmic poly(A)-containing RNA population from the same tissue. During our studies on the mechanisms controlling the accumulation of the poly(A)-containing RNA sequences which occur with high and moderately high frequency in the cytoplasm of the lactating guinea pig mammary gland, it became apparent that > 75% of the 20,000 or so poly(A)-containing nuclear RNA sequences were not found in the cytoplasmic poly(A)-containing RNA fraction. Here we demonstrate that many of the poly(A)-containing RNA sequences retained in the nucleus of the lactating guinea pig mammary gland are also present in the nucleus and cytoplasm of the liver of the male guinea pig. These observations provide new evidence for a predominant role of post-transcriptional mechanisms in the regulation of structural gene expression in guinea pig tissue.

Uptake and degradation of rat and human very low density (remnant) apolipoprotein by parenchymal and non-parenchymal rat liver cells

1. The relative contribution of parenchymal and non-parenchymal cells to the in vivo hepatic uptake of serum apolipoproteins was measured 30 min after intravenous injection of radioiodinated rat very low density lipoprotein (VLDL) remnants, rat and human VLDL, low density lipoprotein (LDL) and high density lipoprotein (HDL). Using rat VLDL, VLDL-remnants, LDL and HDL, respectively, the non-parenchymal cells contain 4.7, 4.9, 6.1 and 5.3 times the amount of trichloroacetic acid-precipitable radioactivity per mg cell protein as compared to parenchymal cells. For human VLDL, LDL and HDL these values are 5.1, 12.0 and 5.9 respectively. 2. The abilities of homogenates of human liver, rat liver parenchymal cells and rat liver non-parenchymal cells to hydrolyze human and rat iodinated VLDL apoprotein were determined by measuring the amount of trichloroacetic acid-soluble (non-iodide) radioactivity liberated upon incubation at the optimal pH of 4.2. Non-parenchymal cells possess a 8--21-fold higher maximal capacity to degrade VLDL apoprotein per mg of cell protein than parenchymal cells. This factor is 5--6 for VLDL-remnant apoprotein degradation measured at low (suboptimal) apolipoprotein concentrations. 3. It is concluded that, in addition to parenchymal cells, the non-parenchymal cells play an important role in the hepatic uptake and possibly degradation of VLDL-(remnant) apoprotein.

Characteristics of acid lipase and acid cholesteryl esterase activity in parenchymal and non-parenchymal rat liver cells

(1) Parenchymal and non-parenchymal cells were isolated from rat liver. The characteristics of acid lipase activity with 4-methylumbelliferyl oleate as substrate and acid cholesteryl esterase activity with cholesteryl[1-14C]oleate as substrate were investigated. The substrates were incorporated in egg yolk lecithin vesicles and assays for total cell homogenates were developed, which were linear with the amount of protein and time. With 4-methylumbelliferyl oleate as substrate, both parenchymal and non-parechymal cells show maximal activities at acid pH and the maximal activity for non-parenchymal cells is 2.5 times higher than for parenchymal cells. It is concluded that 4-methylumbelliferyl oleate hydrolysis is catalyzed by similar enzyme(s) in both cell types. (2) With cholesteryl[1-14C]oleate as substrate both parenchymal and non-parenchymal cells show maximal activities at acid pH and the maximal activity for non-parenchymal cells is 11.4 times higher than for parenchymal cells. It is further shown that the cholesteryl ester hydrolysis in both cell types show different properties. (3) The high activity and high affinity of acid cholesteryl esterase from non-parenchymal cells for cholesterol oleate hydrolysis as compared to parenchymal cells indicate a relative specialization of non-parenchymal cells in cholesterol ester hydrolysis. It is concluded that non-parenchymal liver cells in cholesterol ester hydrolysis. It is concluded that non-parenchymal liver cells possess the enzymic equipment to hydrolyze very efficiently internalized cholesterol esters, which supports the suggestion that these cell types are an important site for lipoprotein catabolism in liver.