Pericentral expression pattern of glucokinase mRNA in the rat liver lobulus

Rapid translocation of hepatic glucokinase in response to intraduodenal glucose infusion and changes in plasma glucose and insulin in conscious rats

The rate of liver glucokinase (GK) translocation from the nucleus to the cytoplasm in response to intraduodenal glucose infusion and the effect of physiological rises of plasma glucose and/or insulin on GK translocation were examined in 6-h-fasted conscious rats. Intraduodenal glucose infusion (28 mg.kg(-1).min(-1) after a priming dose at 500 mg/kg) elevated blood glucose levels (mg/dl) in the artery and portal vein from 90 +/- 3 and 87 +/- 3 to 154 +/- 4 and 185 +/- 4, respectively, at 10 min. At 120 min, the levels had decreased to 133 +/- 6 and 156 +/- 5, respectively. Plasma insulin levels (ng/ml) in the artery and the portal vein rose from 0.7 +/- 0.1 and 1.8 +/- 0.3 to 11.8 +/- 1.5 and 20.2 +/- 2.0 at 10 min, respectively, and 12.4 +/- 3.1 and 18.0 +/- 4.8 at 30 min, respectively. GK was rapidly exported from the nucleus as determined by measuring the ratio of the nuclear to the cytoplasmic immunofluorescence (N/C) of GK (2.9 +/- 0.3 at 0 min to 1.7 +/- 0.2 at 10 min, 1.5 +/- 0.1 at 20 min, 1.3 +/- 0.1 at 30 min, and 1.3 +/- 0.1 at 120 min). When plasma glucose (arterial; mg/dl) and insulin (arterial; ng/ml) levels were clamped for 30 min at 93 +/- 7 and 0.7 +/- 0.1, 81 +/- 5 and 8.9 +/- 1.3, 175 +/- 5 and 0.7 +/- 0.1, or 162 +/- 5 and 9.2 +/- 1.5, the N/C of GK was 3.0 +/- 0.5, 1.8 +/- 0.1, 1.5 +/- 0.1, and 1.2 +/- 0.1, respectively. The N/C of GK regulatory protein (GKRP) did not change in response to the intraduodenal glucose infusion or the rise in plasma glucose and/or insulin levels. The results suggest that GK but not GKRP translocates rapidly in a manner that corresponds with changes in the hepatic glucose balance in response to glucose ingestion in vivo. Additionally, the translocation of GK is induced by the postprandial rise in plasma glucose and insulin.

Fetal programming of perivenous glucose uptake reveals a regulatory mechanism governing hepatic glucose output during refeeding

Increased hepatic gluconeogenesis maintains glycemia during fasting and has been considered responsible for elevated hepatic glucose output in type 2 diabetes. Glucose derived periportally via gluconeogenesis is partially taken up perivenously in perfused liver but not in adult rats whose mothers were protein-restricted during gestation (MLP rats)-an environmental model of fetal programming of adult glucose intolerance exhibiting diminished perivenous glucokinase (GK) activity. We now show that perivenous glucose uptake rises with increasing glucose concentration (0-8 mmol/l) in control but not MLP liver, indicating that GK is flux-generating. The data demonstrate that acute control of hepatic glucose output is principally achieved by increasing perivenous glucose uptake, with rising glucose concentration during refeeding, rather than by downregulation of gluconeogenesis, which occurs in different hepatocytes. Consistent with these observations, glycogen synthesis in vivo commenced in the perivenous cells during refeeding, MLP livers accumulating less glycogen than controls. GK gene transcription was unchanged in MLP liver, the data supporting a recently proposed posttranscriptional model of GK regulation involving nuclear-cytoplasmic transport. The results are pertinent to impaired regulation of hepatic glucose output in type 2 diabetes, which could arise from diminished GK-mediated glucose uptake rather than increased gluconeogenesis.

Cross-talk between the signals hypoxia and glucose at the glucose response element of the L-type pyruvate kinase gene

The signals oxygen and glucose play an important role in metabolism, angiogenesis, tumorigenesis, and embryonic development. Little is known about an interaction of these two signals. We demonstrate here the cross-talk between oxygen and glucose in the regulation of L-type pyruvate kinase (L-PK) gene expression in the liver. In the liver the periportal to perivenous drop in O(2) tension was proposed to be an endocrine key regulator for the zonated gene expression. In primary rat hepatocyte cultures the expression of the L-PK gene on mRNA and on protein level was induced by venous pO(2), whereas its glucose-dependent induction occurred predominantly under arterial pO(2). It was shown by transient transfection of L-PK promoter luciferase and glucose response element (Glc(PK)RE) SV40 promoter luciferase gene constructs that the modulation by O(2) of the glucose-dependent induction occurred at the Glc(PK)RE in the L-PK gene promoter. The reduction of the glucose-dependent induction of the L-PK gene expression under venous pO(2) appeared to be mediated via an interference between hypoxia inducible factor-1 (HIF-1) and upstream stimulating factor at the Glc(PK)RE. The glucose response element also functioned as an hypoxia response element which was confirmed in cotransfection assays with Glc(PK)RE luciferase gene constructs and HIF-1alpha expression vectors. Furthermore, it was found by gel shift and supershift assay that HIF-1alpha and USF-1 or USF-2 could bind to the Glc(PK)RE. Our findings implicate that the cross-talk between oxygen and glucose might have a fundamental role in the regulation of several physiological and pathophysiological processes.

Identification of upstream stimulatory factor as transcriptional activator of the liver promoter of the glucokinase gene

A functionally important cis-acting element termed P2 was identified in the liver promoter of the glucokinase gene. Element P2 was delineated by footprinting in vitro with nuclear proteins from rat liver and spleen. Its core sequence in the rat gene is a canonical CACGTG E-box. In the electrophoretic mobility-shift assay with nuclear proteins from rat liver, hepatocytes and hepatoma cells, an oligonucleotide with P2 in the context of the glucokinase promoter sequence gave rise to a DNA-protein complex shown to contain the upstream stimulatory factor (USF) by specific competition experiments and by reactivity with anti-USF antibodies. Transient transfection of hepatoma HepG2 cells, combined with site-directed mutagenesis, demonstrated that the P2 element was important for liver glucokinase promoter activity. Co-transfection of an expression plasmid coding for USF1 activated reporter gene expression in a manner dependent on an intact P2 element, whereas an expression plasmid for c-Myc was ineffective. Expression of a truncated form of USF1 lacking the transcription activation domain and the basic region decreased reporter activity by a dominant-negative effect. The functional significance of the P2 element was also demonstrated in transient transfection of primary hepatocytes.

Cell-specific expression and regulation of a glucokinase gene locus transgene

Transgenic mice containing one or more extra copies of the entire glucokinase (GK) gene locus were generated and characterized. The GK transgene, an 83-kilobase pair mouse genomic DNA fragment containing both promoter regions, was expressed and regulated in a cell-specific manner, and rescued GK null lethality when crossed into mice bearing a targeted mutation of the endogenous GK gene. Livers from the transgenic mice had elevated GK mRNA, protein, and activity levels, compared with controls, and the transgene was regulated in liver by dietary manipulations. The amount of GK immunoreactivity in hepatocyte nuclei, where GK binds to the GK regulatory protein, was also increased. Pancreatic islets displayed increased GK immunoreactivity and NAD(P)H responses to glucose, but only when isolated and cultured in 20 mM glucose, as a result of the hypoglycemic phenotype of these mice (Niswender, K. D., Shiota, M., Postic, C., Cherrington, A. D., and Magnuson, M. A. (1997) J. Biol. Chem. 272, 22604-22609). Together, these results indicate that the region of the gene from -55 to +28 kilobase pairs (relative to the liver GK transcription start site) contains all the regulatory sequences necessary for expression of both GK isoforms, thereby placing an upper limit on the size of the GK gene locus.

Predominant periportal expression of the fructose 1,6-bisphosphatase gene in rat liver: dynamics during the daily feeding rhythm and starvation-refeeding cycle

Expression of the gene of the key gluconeogenic enzyme fructose 1,6-bisphosphatase (FBPase) was studied in rat liver during the daily feeding cycle and during refeeding after starvation. Total abundance of FBPase mRNA could be quantified by Northern blotting analysis with a digoxigenin-labelled 40-mer oligonucleotide probe. The zonal localization could not be demonstrated by in situ hybridization under several varied conditions with the 32P-end-labelled oligonucleotide probably due to insufficient sensitivity but was demonstrated with a 35S-labelled cRNA probe; the latter was synthesized from a polymerase chain reaction (PCR)-amplified 751 bp cDNA fragment inserted into a pBluescript. During a normal 12:12 h day/night rhythm (darkness with feeding from 1900 to 0700 hours), the total amount of FBPase mRNA stayed almost the same throughout the whole day. After 60 h of starvation the FBPase mRNA level decreased from a maximum at 1800 hours by approximately one-third at the end of refeeding at 0700 hours. Both during the normal feeding rhythm, after 60 h of starvation and during refeeding, i.e. under all conditions, FBPase mRNA was predominantly distributed in the periportal zone. The results clearly show that the preferentially periportal distribution of the FBPase enzyme activity is controlled mainly at a pretranslational level.

Distribution of albumin, alpha 1-inhibitor 3 and their respective mRNAs in periportal and perivenous rat hepatocytes isolated by the digitonin-collagenase technique

The expression of albumin and alpha 1-inhibitor 3 genes was investigated in rat cell suspensions enriched in periportal (n = 10) and perivenous (n = 10) hepatocytes obtained by the digitonin-collagenase technique. The degree of enrichment of the cell suspensions was assessed: (1) by enzymic assays for the periportal marker alanine aminotransferase and for the perivenous marker glutamine synthetase; and (2) by their content of mRNAs for the periportal marker hepatic glutaminase and for glutamine synthetase. The existence of an antegrade intra-lobular gradient for albumin and alpha 1-inhibitor 3 mRNAs was demonstrated, with periportal:perivenous ratios of 2.33 and 3.80, respectively. However, no gradient was demonstrated for the respective protein contents with corresponding ratios of 0.98 and 1.21. A certain degree of overlap existed between periportal and perivenous suspensions for their content in albumin and alpha 1-inhibitor 3 mRNAs. A morphometrical analysis of the surface of digitonin-permeabilized hepatic tissue revealed that this overlap could be explained by a variable extent of permeabilization of the mediolobular zone from one rat to another and from one lobule to another in a given animal. These results suggest that while the digitonin-collagenase technique is well suited for studies in vitro of proteins expressed in sharp intra-lobular gradients or restricted to an intra-lobular compartment, it is not completely reliable for proteins distributed in continuous moderate intra-lobular gradients, such as albumin and alpha 1-inhibitor 3.

Transcriptional control of genes that regulate glycolysis and gluconeogenesis in adult liver

Images

Application of dual-digitonin-pulse perfusion to the study of hepatic mRNA zonation

Heterogeneous zonation of hepatic protein expression over the liver lobule has been recognized by using several analytical techniques, including microdissection, selective cell isolation, immunohistochemistry and hybridization of mRNA in situ. We previously employed the technique of dual-digitonin-pulse perfusion for the highly selective collection and analysis of periportal and perivenous soluble protein. In the present work we have now documented the feasibility of the application of this technique to the study of zonal distribution of mRNA. By using a split-stream design, both protein and RNA fractions can be simultaneously collected from hepatic zones. High-quality RNA (average yield approximately 9-33 micrograms of total RNA per mg of eluted protein) is obtained for analysis. As analysed by immunoblotting and Northern-blot analysis, the zonal distribution of several important cytosolic metabolic enzymes and their mRNAs can be documented. This technique is also applicable to the study of mRNAs for organelle- and membrane-associated proteins that are not recoverable with this digitonin-lysis technique. The application of this experimental technique should allow further molecular insight into the mechanisms underlying zonation of hepatic function.

Zonation of glucokinase in rat liver changes during postnatal development

In the liver many metabolic pathways are preferentially localized in different zones of the acinus. It is assumed that this zonation allows an efficient adaptation to different states of nutrition, because alternative pathways can be regulated independently. It is reported that the rate limiting enzyme for the glycolytic pathway, glucokinase (EC 2.7.1.2), is predominantly located in the pericentral zone. The gene expression of glucokinase is induced to a maximum level after a carbohydrate-rich diet. In starved or diabetic rats glucokinase gene expression is barely detectable. In postnatal development glucokinase is induced to significant levels only from day 14 onwards. The distribution of the glucokinase protein in the rat liver lobule in the first 4 weeks of postnatal life was investigated by immunohistochemistry and compared to the distribution observed in adult rats. In adult rats considerably high levels of glucokinase are measureable as shown by immunoblotting utilizing a monospecific antibody and a photometric assay of glucokinase enzyme activity, respectively. Immunohistochemically the hepatic glucokinase protein is detected in the perivenous area. During postnatal development, the quantities of hepatic glucokinase protein and glucokinase enzyme activity start to increase significantly from day 15 onwards. Subsequently, glucokinase levels rise further until day 29. In contrast to the results obtained by immunoblotting, glucokinase is already detectable in some liver cells in sections from 6-day-old rats by immunohistochemistry. The liver lobule structure at this age is not completely developed, therefore it is not possible to definitely assign these cells to periportal or pericentral areas. At day 10 post partum the number of glucokinase expressing cells, which appear to be localized preferentially in the periportal zone, increases. In agreement with the immunoblotting, an immense increase in glucokinase activity was observed at day 14. The periportal zonation, clearly detectable at this time, remains stable until day 24. In sections from 29-day-old rats the periportal zonation begins to change into a more homogeneous pattern with a slight preference for periportal areas. The observed appearance of the periportal zonation of glucokinase during neonatal development is obviously in contrast to the perivenous expression of glucokinase in adult rats.

Glucose administration induces the premature expression of liver glucokinase gene in newborn rats. Relation with DNase-I-hypersensitive sites

Glucokinase first appears in the liver of the rat 2 weeks after birth and its activity rapidly increases after weaning on to a high-carbohydrate diet. The appearance of glucokinase is principally due to the increase of plasma insulin and to the decrease of plasma glucagon concentrations. Oral glucose administration to 1- or 10-day-old suckling rats induced an increase in plasma insulin and a fall in plasma glucagon and allowed a rapid accumulation of liver glucokinase mRNA, secondarily to a stimulation of gene transcription. When unrestrained late pregnant rats were infused with glucose during 36 h to induce an increase in fetal plasma insulin and a decrease in fetal plasma glucagon concentrations, glucokinase mRNA was detectable in fetal liver but the level was 100-fold lower than that observed in 1- or 10-day-old suckling rats. It is suggested that the hormonal environment did not allow glucokinase gene expression to be induced in fetal liver and that the absence of expression of glucokinase in suckling rat liver is due to the presence of low plasma insulin and high plasma glucagon levels. The chromatin structure of the glucokinase gene was examined during development by identification of DNase-I-hypersensitive sites from the region comprised between -8 kb upstream and +4 kb downstream of the cap site. Five hypersensitive sites were found: four liver-specific sites upstream of the cap site and one non-specific site in the first intron. These sites are already present in term fetus but the intensity of the two proximal sites located upstream of the cap site increase markedly after birth. This suggests that these sites could be implicated in the regulation of glucokinase gene expression by insulin and glucagon. Full DNase-I-hypersensitivity of these two proximal sites seems necessary for the mature response of glucokinase gene in response to changes in pancreatic hormones concentrations.

Mammalian glucokinase and its gene

Images

Zonal expression of the glucokinase gene in rat liver. Dynamics during the daily feeding rhythm and starvation-refeeding cycle demonstrated by in situ hybridization

The abundance and zonal distribution of glucokinase (GK) mRNA were studied in rat liver during a normal 12 h day/12 h night rhythm (dark from 1900 to 0700 hours) and during refeeding after 60 h of starvation. Zonation of GK gene expression was examined by in situ hybridization with a radiolabelled cRNA probe and GK mRNA abundance was determined by Northern blot analysis with a digoxigenin-labelled cRNA probe. GK mRNA appeared to be almost homogeneously distributed throughout the whole daily feeding cycle; yet it was predominantly localized in the perivenous and intermediate zone during refeeding after 60 h of starvation. During the daily feeding rhythm, the total amount of GK mRNA increased quickly with the beginning of the feeding period at 1900 hours reaching a maximum at midnight and then decreased continuously to a basal level at noon. Virtually no GK mRNA was detected after 60 h of starvation. Refeeding caused a rapid increase in GK mRNA to a maximum at 2400 hours followed by a decrease to approximately two-thirds of the maximum value at 0700 hours. If the homogeneous distribution of GK mRNA during the daily feeding rhythm was real rather than apparent because of too low a sensitivity of the cRNA probe, the present results suggest that during the normal circadian cycle the mainly perivenous distribution of GK enzyme activity and protein is regulated preferentially at a translational level. The findings clearly show that during refeeding after 60 h of starvation the GK distribution is controlled predominantly at a pretranslational level.