Liver cell heterogeneity. The distribution of pyruvate kinase and phosphoenolpyruvate carboxykinase (GTP) in the liver lobule of fed and starved rats.

Pyruvate kinase and phosphoenolpyruvate carboxykinase activities were determined in microdissected freeze-dried liver cells from the periportal and pericentral area of the liver lobule. Pyruvate kinase activity was measured by a microfluorimetric procedure adapted to 20-200 ng tissue dry weight. In livers from fed rats, its activity was twice as high in the central zone as in the periportal cells; starvation reduced this gradient by decreasing central activities. Phosphoenolpyruvate carboxykinase activity was measured by a microradiochemical technique in 100-300 ng tissue dry weight. In livers from fed rats, this enzyme was nearly 3 times more active in the periportal cells than in the central area. Starvation increased this enzyme in both zones with a more pronounced change in the central cells. The results indicate a heterogeneous distribution of enzymes of carbohydrate metabolism in the liver lobule. Gluconegenesis seems to be localized preferentially in periportal hepatocytes, whereas the glycolytic enzyme was found to be more active in cells surrounding the pericentral liver cells.     

Regulation of glycogen metabolism in primary cultures of rat hepatocytes. Restoration of acute effects of insulin and glucose in cells from diabetic rats

Defects in the deposition of glycogen and the regulation of glycogen synthesis in the livers of severely insulin-deficient rats can be reversed, in vivo, within hours of insulin administration. Using primary cultures of hepatocytes isolated from normal and diabetic rats in a serum-free chemically defined medium, the present study addresses the chronic action of insulin to facilitate the direct effects of insulin and glucose on the short term regulation of the enzymes controlling glycogen metabolism. Primary cultures were maintained in the presence of insulin, triiodothyronine, and cortisol for 1-3 days. On day 1 in alloxan diabetic cultures, 10(-7) M insulin did not acutely activate glycogen synthase over a period of 15 min or 1 h, whereas insulin acutely activated synthase in cultures of normal hepatocytes. By day 3 in hepatocytes isolated from alloxan diabetic rats, insulin effected an approximate 30% increase in per cent synthase I within 15 min as was also the case for normal cells. The acute effect of insulin on synthase activation was independent of changes in phosphorylase alpha. Whereas glycogen synthase phosphatase activity could not be shown to be acutely affected by insulin, the total activity in diabetic cells was restored to normal control values over the 3-day culture period. The acute effect of 30 mM glucose to activate glycogen synthase in cultured hepatocytes from normal rats after 1 day of culture was missing in hepatocytes isolated from either alloxan or spontaneously diabetic (BB/W) rats. After 3 days in culture, glucose produced a 50% increase in glycogen synthase activity during a 10-min period under the same conditions. These studies clearly demonstrate that insulin acts in a chronic manner in concert with thyroid hormones and steroids to facilitate acute regulation of hepatic glycogen synthesis by both insulin and glucose.     

Variations in immunoreactivity for phenobarbital- and 3-methylcholanthrene-inducible cytochromes P-450, and NADPH-cytochrome P-450 reductase in rat liver over twenty-four hours.

To reveal distribution patterns of phenobarbital- and 3-methylcholanthrene-inducible cytochromes P-450 (PB and MC) and NADPH-cytochrome P-450 reductase (P-450red) within the liver acinus of untreated rats, and their variations over 24 h, hepatic samples were examined by immunohistochemistry and image-analyzer at evenly spaced six time points over 24 h. When examined in semi-thin sections obtained from Epon-embedded, freeze-dried, and paraformaldehyde vapor-phase fixed materials, the immunoreactivity for these enzymes showed different distribution patterns within the liver acinus. Immunodeposits for PB were predominantly distributed in perivenous hepatocytes, whereas those for MC and P-450red were slightly more intense in periportal hepatocytes at each time point. The immunoreactivity for PB and MC in both perivenous and periportal hepatocytes increased during the dark period, peaking early in the light period. These variations coincide well with our previous morphometric results (Uchiyama and Asari, 1984); the volume and surface densities of rough endoplasmic reticulum (rER) in hepatocytes increased during the dark period. On the other hand, weak fluctuation was demonstrated in the immunoreactivity for P-450red in hepatocytes of both zones. These results suggest that PB and MC are retained in rER rather than smooth endoplasmic reticulum (sER) of hepatocytes obtained from untreated rats. These enzymes in sER may be short in their half-life spans.     

Topographic distribution of dividing hepatocytes in the regenerating liver lobule in the period of maximum mitotic activity

Topographic distribution of dividing hepatocytes was studied in the liver lobule after hepatectomy performed at 10-11.30 a. m. The studies were conducted 20-32 h after surgery during the first increase in mitotic activity. It has been established that dividing hepatocytes appear simultaneously in all lobule areas, i. e. that all the hepatocytes are capable of responding to regenerating stimuli. However, further behaviour of dividing hepatocytes becomes different. It is concluded that hepatocytes from outer and intermediate lobule areas possess a more pronounced proliferative potential during peak mitotic activity after partial hepatectomy.     

Synthesis of alpha-fetoprotein in primary cultures of hepatocytes from adult mice

Alpha-fetoprotein (AFP) and albumin synthesis in primary hepatocyte cultures of adult mice was studied by the immunoperoxidase techniques. Albumin was detectable in all hepatocytes during the cultivation period (from day 1 to day 8). AFP could be found regularly beginning from days 3-4 up to day 8. Attempts to obtain hepatocytes surviving for a longer period of time ended in failure. The number of AFP-containing hepatocytes in the culture ranged within single cells to about half of all the hepatocytes. The proliferating hepatocytes that contained both AFP and albumin were found in some of the experiments.     

Variations in immunoreactivity for phenobarbital- and 3-methylcholanthrene-inducible cytochromes P-450, and NADPH-cytochrome P-450 reductase in rat liver over twenty-four hours

Summary To reveal distribution patterns of phenobarbital-and 3-methylcholanthrene-inducible cytochromes P-450 (PB and MC) and NADPH-cytochrome P-450 reductase (P-450red) within the liver acinus of untreated rats, and their variations over 24 h, hepatic samples were examined by immunohistochemistry and image-analyzer at evenly spaced six time points over 24 h. When examined in semi-thin sections obtained from Epon-embedded, freeze-dried, and paraformaldehyde vapor-phase fixed materials, the immunoreactivity for these enzymes showed different distribution patterns within the liver acinus. Immunodeposits for PB were predominantly distributed in perivenous hepatocytes, whereas those for MC and P-450red were slightly more intense in periportal hepatocytes at each time point. The immunoreactivity for PB and MC in both perivenous and periportal hepatocytes increased during the dark period, peaking early in the light period. These variations coincide well with our previous morphometric results (Uchiyama and Asari, 1984); the volume and surface densities of rough endoplasmic reticulum (rER) in hepatocytes increased during the dark period. On the other hand, weak fluctuation was demonstrated in the immunoreactivity for P-450red in hepatocytes of both zones. These results suggest that PB and MC are retained in rER rather than smooth endoplasmic reticulum (sER) of hepatocytes obtained from untreated rats. These enzymes in sER may be short in their half-life spans.     

Deposition of amyloid in the liver of hamsters: an enzyme-histochemical and electron-microscopical study.

In induced amyloidosis the amyloid was first deposited in the portal areas, then in the centre of the lobule and disseminated through it in the spaces of Disse. No intracellular amyloid was found in the predeposit phase. In the hepatic lobule both mononuclear phagocytes and hepatocytes showed a topographic relationship to the first deposits of amyloid. Some macrophages showed invaginations or vacuoles containing amyloid fibrils. Between the microvilli of hepatocytes, parallel amyloid fibrils occurred. Between the amyloid fibrils were 30-50 nm membrane-bounded spherical particles which may have been lipoprotein aggregates. There was no large-scale phagocytosis of amyloid during the induction period or after survival without casein treatment up to 18 weeks. Lysosomal enzyme activity was seen in the deposits of extracellular amyloid.     

Influence of culture time on the expression of drug-metabolizing enzymes in primary human hepatocytes and hepatoma cell line HepG2

Primary cultures of human hepatocytes and hepatoma cell line HepG2 are frequently used to evaluate the hepatic disposition of drugs and other xenobiotics. To check the variability of the expression of drug-metabolizing enzymes in these in vitro models, expression of genes coding for several cytochrome P450 isoforms and phase II enzymes was quantified during culture time by real-time RT-PCR. Gene expression was determined daily for primary hepatocytes maintained in a sandwich culture over 1 week and for HepG2, during the first 10 passages. In primary hepatocytes characteristic expression trends were observed which could be abstracted into three major classes of time curves. Genes of the first and the second class had an expression maximum around day 6 and day 4 in culture, respectively. The third class of genes had two expression peaks: at day 1 and 5 in culture. Surprisingly, also the cell line HepG2 showed significant expression changes during passages. For example, gene expression of cytochrome 1A1 varied 8-fold, that of cytochrome 2B6 30-fold, and that of NADP-quinone reductase 1 more than 200-fold within the first 10 passages. In conclusion, neither primary hepatocytes nor HepG2 cell line display a model for constant expression of drug-metabolizing enzymes.     

Adenylate cyclase and phosphodiesterase activities in rat hepatocytes cultured in the presence and absence of dexamethasone

The effects of glucagon and dexamethasone on the activities of the enzymes involved in cyclic adenosine 3':5'-monophosphate (cyclic AMP) metabolism in primary monolayer cell cultures of adult rat hepatocytes were examined. Short-term experiments indicated that the magnitude of the cultured cells' response to glucagon, as measured by production of cyclic AMP, was essentially the same as that for freshly isolated hepatocytes. However, the time course of this response was markedly different. Although the activity of adenylate cyclase is maintained throughout the culture period at a level similar to that of the freshly isolated hepatocytes, the activity of both low and high Km forms of phosphodiesterase decreases rapidly with length of time in vitro. This is reflected by an increase in cyclic AMP produced in response to glucagon and theophylline by cells of different ages. Dexamethasone caused an increased loss of phosphodiesterase activity, as well as increased cyclic AMP accumulation in the presence or absence of theophylline. Various agents failed to restore the lost phosphodiesterase activity. These results may indicate that phosphodiesterase activity is more sensitive to the inevitable inadequacies of the in vitro environment of cultured hepatocytes than adenylate cyclase. It was also found that a modification of the method of Seglen (1) for the preparation of isolated hepatocytes yielded cells that had less phosphodiesterase activity than those prepared by the method of Berry and Friend (2).     

Changes in activity and intra-acinar distribution of glucose-6-phosphate dehydrogenase and malic enzyme during pregnancy in rat liver

Summary Using techniques of microdissection and microassay as well as qualitative histochemistry the activity and intra-acinar distribution of G6PDH and ME were studied on selected days of pregnancy in the rat. Both enzymes show distinct fluctuations during the course of pregnancy in keeping with changes in hepatic lipogenesis. Marked increases in activity are seen as early as the 4th day, while highest values are attained on day 20, with a predominant perivenous induction. On day 22, just before parturition a sharp decrease of both enzyme activities with a flattening of the periportal/perivenous gradient was detected. G6PDH shows proportionally considerably larger increases and more distinct changes in zonation. The perivenous fluctuations in G6PDH activity of late gestation are supposed to be caused primarily by insulin. Although estrogen is known to induce both enzymes, the temporal changes in enzyme activity in pregnancy cannot be related to the action of estrogen alone. The changes in enzyme activity, however, correspond well to those of progesterone, and although no direct action of progesterone on these enzymes has yet been proposed, further work on its effects on enzyme activity and distribution is indicated.     

Intralobular distribution of UDP-glucuronosyltransferase in livers from untreated, 3-methylcholanthrene- and phenobarbital-treated rats

A 3-methylcholanthrene-inducible enzyme form of UDP-glucuronosyltransferase has been localized within the liver lobule both immunohistochemically and enzymatically in microdissected centrilobular and periportal liver tissue. Livers of untreated, 3-methylcholanthrene- and phenobarbital-treated rats have been compared. The enzyme was detected in hepatocytes throughout the liver. However both immunohistochemical determination of the enzyme level and biochemical determination of its activity towards 1-naphthol revealed a heterogeneous distribution of the enzyme. In untreated controls and 3-methylcholanthrene-treated rats both enzyme activity and histochemical staining was highest in centrilobular hepatocytes. However, after phenobarbital-treatment enzyme staining and activity was highest in periportal hepatocytes, suggesting that the differentially inducible enzyme activities may be localized in different zones of the liver lobule. The results demonstrate that the 3-methylcholanthrene-inducible UDP-glucuronosyltransferase is preferentially expressed in centrilobular hepatocytes.     

Adenylate cyclase and phosphodiesterase activities in rat hepatocytes cultured in the presence and absence of dexamethasone

Summary The effects of glucagon and dexamethasone on the activities of the enzymes involved in cyclic adenosine 3′∶5′-monophosphate (cyclic AMP) metabolism in primary monolayer cell cultures of adult rat hepatocytes were examined. Short-term experiments indicated that the magnitude of the cultured cells' response to glucagon, as measured by production of cyclic AMP, was essentially the same as that for freshly isolated hepatocytes. However, the time course of this response was markedly different. Although the activity of adenylate cyclase is maintained throughout the culture period at a level similar to that of the freshly isolated hepatocytes, the activity of both low and highK _m forms of phosphodiesterase decreases rapidly with length of time in vitro. This is reflected by an increase in cyclic AMP produced in response to glucagon and theophylline by cells of different ages. Dexamethasone caused an increased loss of phosphodiesterase activity, as well as increased cyclic AMP accumulation in the presence or absence of theophylline. Various agents failed to restore the lost phosphodiesterase activity. These results may indicate that phosphodiesterase activity is more sensitive to the inevitable inadequacies of the in vitro environment of cultured hepatocytes than adenylate cyclase. It was also found that a modification of the method of Seglen (1) for the preparation of isolated hepatocytes yielded cells that had less phosphodiesterase activity than those prepared by the method of Berry and Friend (2). This work was supported by grants from the Medical Research Council of New Zealand and the Medical Research Distribution Committe.     

Mathematical modelling of liver regeneration after intoxication with CCl(4)

Liver regeneration is a complex process, having evolved to protect animals from the consequences of liver loss caused by food toxins. In this study, we established a mathematical spatial-temporal model of the liver lobule regenerating after CCl(4) intoxication. The aim of modelling the regeneration process by matching experimental observations with those from a mathematical model is to gain a better understanding of the process and to recognize which parameters are relevant for specific phenomena. In order to set up a realistic minimal model, we first reconstructed a schematised liver lobule after determination of: (i) the mean number of hepatocytes between the central vein and the periphery of the lobule, (ii) the mean size of the hepatocytes and (iii) the mean number of hepatocyte columns in the inner, midzonal and peripheral ring of the lobule. In a next step, we determined the time course of cell death and BrdU incorporation after intoxication of male Sprague Dawley rats with CCl(4), thereby differentiating between inner, midzonal and peripheral hepatocytes. These parameters were used to construct a model. The basic unit of this model is the individual cell. The detailed behaviour of the cells is studied, controlled by the model parameters: (1) probability of cell division at defined positions of the lobule at a given time, (2) "coordinated cell orientation", i.e., the ability of the cells to align during the regeneration process into columns towards the central vein of a liver lobule, (3) cell cycle duration, (4) the migration activity and (5) the polarity of the hepatocytes resulting in polar cell-cell adhesion between them. In a schematised lobule, the model shows that CCl(4) initially induced cell death of a pericentral ring of hepatocytes, followed by a wave of proliferation that starts in the surviving hepatocytes next to the inner ring of dead cells and continues to the peripheral hepatocytes, finally restoring the characteristic micro-architecture of the lobule in a 7-day process. This model was used to systematically analyze the influence of parameters 1-5. Interestingly, coordinated cell orientation and cell polarity were identified to be the most critical parameters. Elimination led to destruction of the characteristic micro-architecture of the lobule and to a high degree of disorder characterized by hexagonal cell structures. Our model suggests that the ability of hepatocytes to realign after cell division by a process of coordinated cell orientation (model parameter 2) in combination with cell polarity (model parameter 5) may be at least as critical as hepatocyte proliferation (model parameter 1) itself.     

Quantitative topochemistry of rat liver enzymes during postnatal development in relation to activity-rest cycle

A quantitative histochemical method (Trident) has been adapted to measure the activities of 4 enzymes, succinate dehydrogenase (SD), isocitrate dehydrogenase (ICD), glucose-6-phosphate dehydrogenase (G6PD) and 6-phosphogluconate dehydrogenase (6-PGD), within the liver acini of the rat during the postnatal developmental period. Quantitative changes of these enzymes in livers of rats of 25 g and 50 g body weight were studied, with particular emphasis on the activity-rest cycle. The results indicate a time-dependent heterogeneous distribution of enzymes along the acinar zones and the pattern of localization is age-dependent. When the mean enzyme activity from each group in relation to the time of the day are compared, a mirror image of each other could be seen. In general, a high enzyme activity has been observed during the resting phase in 25-g rats and low in 50-g rats. During the developmental period, the mean ICD activity is diminished, whereas G6PD and 6-PGD are augmented, and SD activity remains unchanged.     

Expression of carbamoylphosphate synthetase I and glutamine synthetase in hepatic organoids reconstructed by rat small hepatocytes and hepatic nonparenchymal cells

The objective of this study was to investigate the expression of carbamoylphosphate synthetase I (CPS) and glutamine synthetase (GS) in small hepatocyte colonies and whether the heterogeneous expression of the enzymes could be induced during the maturation of small hepatocytes. Small hepatocytes isolated from an adult rat liver were cultured and proliferated to form colonies. The expression of CPS and GS was examined using immunocytochemistry and immunoblotting. In this culture more than 99% of morphologically hepatic cells were positive for CPS and all small hepatocytes were negative for GS at day 5. CPS-positive cells dramatically decreased with time in culture, whereas GS-positive ones appeared and their number increased in the colonies. Two to 3 weeks after plating, colonies with rising and piled-up cells appeared and the number of such colonies reached about 25% of all colonies at day 30. In most rising and piled-up cells in colonies both proteins were strongly expressed, whereas many small hepatocytes in monolayer colonies did not express either protein. When small hepatocytes in monolayer colonies were overlayed with Matrigel, the cells gradually piled up and both CPS and GS proteins were dramatically induced. The expression of CPS and GS in small hepatocytes may interact with the extracellular matrix because the rising and piled-up cells appear to be induced by the extracellular matrix produced by hepatic nonparenchymal cells.     

Changes in activity and intra-acinar distribution of glucose-6-phosphate dehydrogenase and malic enzyme during pregnancy in rat liver

Using techniques of microdissection and microassay as well as qualitative histochemistry the activity and intra-acinar distribution of G6PDH and ME were studied on selected days of pregnancy in the rat. Both enzymes show distinct fluctuations during the course of pregnancy in keeping with changes in hepatic lipogenesis. Marked increases in activity are seen as early as the 4th day, while highest values are attained on day 20, with a predominant perivenous induction. On day 22, just before parturition a sharp decrease of both enzyme activities with a flattening of the periportal/perivenous gradient was detected. G6PDH shows proportionally considerably larger increases and more distinct changes in zonation. The perivenous fluctuations in G6PDH activity of late gestation are supposed to be caused primarily by insulin. Although estrogen is known to induce both enzymes, the temporal changes in enzyme activity in pregnancy cannot be related to the action of estrogen alone. The changes in enzyme activity, however, correspond well to those of progesterone, and although no direct action of progesterone on these enzymes has yet been proposed, further work on its effects on enzyme activity and distribution is indicated.     

Sodium butyrate preserves aspects of the differentiated phenotype of normal adult rat hepatocytes in culture

We have determined that sodium butyrate and, to a lesser extent, dimethylsulfoxide (DMSO) and 3-aminobenzamide (3-AB) preserve aspects of the differentiated phenotype of primary cultures of adult rat hepatocytes. The histone deacetylase inhibitor, butyrate, inhibits the increase in gamma-glutamyltranspeptidase (GGT) activity and the decrease in basal tyrosine aminotransferase (TAT) activity normally observed when hepatocytes are cultured under appropriate conditions. The effects of butyrate on GGT and TAT activities are accompanied by parallel changes in GGT and TAT mRNA levels. The poly(ADP)ribose-synthetase inhibitor, 3-aminobenzamide, has effects similar to butyrate on GGT activity and mRNA levels, while both 3-AB and DMSO increase basal TAT activity in cultured hepatocytes. Under appropriate conditions all three agents--butyrate, 3-AB, and DMSO--extend the length of time cultured hepatocytes can be maintained as confluent monolayers. However, under all the conditions studied, butyrate extended the length of time hepatocytes could be maintained as monolayers more than any other treatment used. Butyrate-treated hepatocytes maintained ultrastructural features that were more similar to those of hepatocytes in vivo than hepatocytes treated with any other of the agents tested. Histone acetylation levels of primary cultures of adult rat hepatocytes declined concomitant with the loss of the differentiated phenotype of the cells. These results suggest that histone acetylation may play a role in the changes in gene expression observed when hepatocytes are placed in culture.     

Changes in carnitine-palmitoyl-transferase and carnitine-acetyl-transferase activity in rat kidney during development; effects of fasting

Developmental changes in the activities of two enzymes catalysing transfer of fatty acid across the mitochondrial membrane (carnitine-palmitoyl-transferase and carnitine-acetyl-transferase) were studied in the kidneys of developing rats from late fetal life to 10 days post-partum and were compared to cortical adult value. The activities of carnitine-palmitoyl-transferase and carnitine-acetyl-transferase increased after birth to reach a maximal value on day 5. Thereafter both activities decreased to reach adult cortical value. The cytochrome c oxidase activity (index of mitochondrial activity) increases continuously from late fetal age to adult. In kidneys of fetuses from starved mothers the carnitine-palmitoyl-transferase activity is higher than that of controls while carnitine-acetyl-transferase activity is not changed. In postmature fetuses (23 days post-co?tum) carnitine-palmitoyl-transferase activity is the same as in 21 days post-co?tum old fetuses. These results are discussed in relation to variations in nutritional and hormonal changes occurring during the perinatal period.