Cell cycle-dependent expression and subcellular localization of fructose 1,6-bisphosphatase

Recently a gluconeogenic enzyme was discovered—fructose 1,6-bisphosphatase (FBPase)—that localizes in the nucleus of a proliferating cell, but its physiological role in this compartment remains unclear. Here, we demonstrate the link between nuclear localization of FBPase and the cell cycle progression. Results of our studies indicate that in human and mouse squamous cell lung cancer, as well as in the HL-1 cardiomyocytes, FBPase nuclear localization correlates with nuclear localization of S and G2 phase cyclins. Additionally, activity and expression of the enzyme depends on cell cycle stages. Identification of FBPase interacting partners with mass spectrometry reveals a set of nuclear proteins involved in cell cycle regulation, mRNA processing and in stabilization of genomic DNA structure. To our knowledge, this is the first experimental evidence that muscle FBPase is involved in cell cycle events.     

Calcium inhibits muscle FBPase and affects its intracellular localization in cardiomyocytes

As our recent investigation revealed, in mammalian heart muscle, fructose 1,6-bisphosphatase (FBPase)--a key enzyme of glyconeogenesis--is located around the Z-line, inside cells' nuclei and, as we demonstrate here for the first time, it associates with intercalated discs. Since the degree of association of numerous enzymes with subcellular structures depends on the metabolic state of the cell, we studied the effect of elevated Ca2+ concentration on localization of FBPase in cardiomyocytes. In such conditions, FBPase dissociated from the Z-line, but no visible effect on FBPase associated with intercalated discs or on the nuclear localization of the enzyme was observed. Additionally, Ca2+ appeared to be a strong inhibitor of muscle FBPase.     

Glucagon induces translocation of glucokinase from the cytoplasm to the nucleus of hepatocytes by transfer between 6-phosphofructo 2-kinase/fructose 2,6-bisphosphatase-2 and the glucokinase regulatory protein

Glucokinase activity is a major determinant of hepatic glucose metabolism and blood glucose homeostasis. Liver glucokinase activity is regulated acutely by adaptive translocation between the nucleus and the cytoplasm through binding and dissociation from its regulatory protein (GKRP) in the nucleus. Whilst the effect of glucose on this mechanism is well established, the role of hormones in regulating glucokinase location and its interaction with binding proteins remains unsettled. Here we show that treatment of rat hepatocytes with 25 mM glucose caused decreased binding of glucokinase to GKRP, translocation from the nucleus and increased binding to 6-phosphofructo 2-kinase/fructose 2,6 bisphosphatase-2 (PFK2/FBPase2) in the cytoplasm. Glucagon caused dissociation of glucokinase from PFK2/FBPase2, concomitant with phosphorylation of PFK2/FBPase2 on Ser-32, uptake of glucokinase into the nucleus and increased interaction with GKRP. Two novel glucagon receptor antagonists attenuated the action of glucagon. This establishes an unequivocal role for hormonal control of glucokinase translocation. Given that glucagon excess contributes to the pathogenesis of diabetes, glucagon may play a role in the defect in glucokinase translocation and activity evident in animal models and human diabetes.     

Nuclear accumulation of fructose 1,6-bisphosphatase is impaired in diabetic rat liver

Using a streptozotocin-induced type 1 diabetic rat model, we analyzed and separated the effects of hyperglycemia and hyperinsulinemia over the in vivo expression and subcellular localization of hepatic fructose 1,6-bisphosphatase (FBPase) in the multicellular context of the liver. Our data showed that FBPase subcellular localization was modulated by the nutritional state in normal but not in diabetic rats. By contrast, the liver zonation was not affected in any condition. In healthy starved rats, FBPase was localized in the cytoplasm of hepatocytes, whereas in healthy re-fed rats it was concentrated in the nucleus and the cell periphery. Interestingly, despite the hyperglycemia, FBPase was unable to accumulate in the nucleus in hepatocytes from streptozotocin-induced diabetic rats, suggesting that insulin is a critical in vivo modulator. This idea was confirmed by exogenous insulin supplementation to diabetic rats, where insulin was able to induce the rapid accumulation of FBPase within the hepatocyte nucleus. Besides, hepatic FBPase was found phosphorylated only in the cytoplasm, suggesting that the phosphorylation state is involved in the nuclear translocation. In conclusion, insulin and not hyperglycemia plays a crucial role in the nuclear accumulation of FBPase in vivo and may be an important regulatory mechanism that could account for the increased endogenous glucose production of liver of diabetic rodents.     

Changes in subcellular localization of fructose 1,6-bisphosphatase during differentiation of isolated muscle satellite cells

Subcellular localization of FBPase, a regulatory enzyme of glyconeogenesis, was examined inside dividing and differentiating satellite cells from rat muscle. In dividing myoblasts, FBPase was located in cytosol and nuclei. When divisions ceased, FBPase became restricted to the cytosolic compartment and finally was found to associate with the Z-lines, as in adult muscle. Moreover, a 12-fold decrease was observed in the number of FBPase-positive nuclei associated with muscle fibres of adult rat, as compared with young muscle, possibly reflecting the reduction in number of active satellite cells during muscle maturation. The data might suggest that FBPase participates in some nuclear processes during development and regeneration of skeletal muscle.     

Cloning,structure and expression of a pea cDNA clone coding for a photosynthetic fructose-1,6-bisphosphatase with some features different from those of the leaf chloroplast enzyme

A positive clone against pea (Pisum sativum L.) chloroplast fructose-1,6-bisphosphatase (FBPase; EC 3.1.3.11) antibodies was obtained from a copy DNA (cDNA) library in λgt11. The insert was 1261 nucleotides long, and had an open reading frame of 1143 base pairs with coding capability for the whole FBPase subunit and a fragment of a putative processing peptide. An additional 115 base pairs corresponding to a 3′-untranslated region coding for an mRNA poly(A)⁺ tail were also found in the clone. The deduced sequence for the FBPase subunit was a 357-amino-acid protein of molecular mass 39253 daltons (Da), showing 82–88% absolute homology with four chloroplastic FBPases sequenced earlier. The 3.1-kilobase (kb)KpnI-SacI fragment of the λgt11 derivative was subcloned between theKpnI-SacI restriction sites of pTZ18R to yield plasmid pAMC100. Lysates ofEscherichia coli (pAMC100) showed FBPase activity; this was purified as a 170-kDa protein which, upon sodium dodecyl sulfate-polyacrylamide gel electrophoresis, displayed a 44-kDa band. As occurs with native FBPases, this indicates a homotetrameric structure for the expressed FBPase. When assayed under excess Mg²⁺ (10 mM), the expressed enzyme had a higher affinity for the substrate than the native pea leaf FBPase; this parameter appears to be substantiated by a tenfold higher specific activity than that of the native enzyme. However, when activated with dithiothreitol plus saturating concentrations of pea thioredoxin (Td) f, both FBPase had similar activities, with a 4:1 Td f-FBPase stoichiometry. In contrast to the native pea chloroplast FBPase, theE. coli-expressed enzyme did not react with the monoclonal antibody GR-PB5. It also had a higher heat sensitivity, with 42% residual activity after heating for 30 min at 60°C, conditions which preserved the native enzyme in a fully active state. These results show the existence of some difference(s) in the conformation of the two FBPases; this could be a consequence of a different expression of the genomic and cDNA clones, or be due to the need for some factor for the correct assembly of the oligomeric structure of the native chloroplast enzyme. Accession number for pea chloroplast FBPase coding sequence: X68826 in the European Molecular Biology Laboratory (EMBL)     

Activation of NADP-Malate Dehydrogenase, Pyruvate,Pi Dikinase, and Fructose 1,6-Bisphosphatase in Relation to Photosynthetic Rate in Maize

The activity and extent of light activation of three photosynthetic enzymes, pyruvate,Pi dikinase, NADP-malate dehydrogenase (NADP-MDH), and fructose 1,6-bisphosphatase (FBPase), were examined in maize (Zea mays var Royal Crest) leaves relative to the rate of photosynthesis during induction and under varying light intensities. There was a strong light activation of NADP-MDH and pyruvate,Pi dikinase, and light also activated FBPase 2- to 4-fold. During the induction period for whole leaf photosynthesis at 30°C under high light, the time required to reach half-maximum activation for all three enzymes was only 1 minute or less. After 2.5 minutes of illumination the enzymes were fully activated, while the photosynthetic rate was only at half-maximum activity, indicating that factors other than enzyme activation limit photosynthesis during the induction period in C₄ plants.

Under steady state conditions, the light intensity required to reach half-maximum activation of the three enzymes was similar (300-400 microEinsteins per square meter per second), while the light intensity required for half-maximum rates of photosynthesis was about 550 microEinsteins per square meter per second. The light activated levels of NADP-MDH and FBPase were well in excess of the in vivo activities which would be required during photosynthesis, while maximum activities of pyruvate,Pi dikinase were generally just sufficient to accommodate photosynthesis, suggesting the latter may be a rate limiting enzyme.

There was a large (5-fold) light activation of FBPase in isolated bundle sheath strands of maize, whereas there was little light activation of the enzyme in isolated mesophyll protoplasts. In mesophyll protoplasts the enzyme was largely located in the cytoplasm, although there was a low amount of light-activated enzyme in the mesophyll chloroplasts. The results suggest the chloroplastic FBPase in maize is primarily located in the bundle sheath cells.

     

Evolutionary conserved N-terminal region of human muscle fructose 1,6-bisphosphatase regulates its activity and the interaction with aldolase

N-terminal residues of muscle fructose 1,6-bisphosphatase (FBPase) are highly conserved among vertebrates. In this article, we present evidence that the conservation is responsible for the unique properties of the muscle FBPase isozyme: high sensitivity to AMP and Ca(2+) inhibition and the high affinity to muscle aldolase, which is a factor desensitizing muscle FBPase toward AMP and Ca(2+). The first N-terminal residue affecting the affinity of muscle FBPase to aldolase is arginine 3. On the other hand, the first residue significantly influencing the kinetics of muscle FBPase is proline 5. Truncation from 5-7 N-terminal residues of the enzyme not only decreases its affinity to aldolase but also reduces its k-(cat) and activation by Mg(2+), and desensitizes FBPase to inhibition by AMP and calcium ions. Deletion of the first 10 amino acids of muscle FBPase abolishes cooperativity of Mg(2+) activation and results in biphasic inhibition of the enzyme by AMP. Moreover, this truncation lowers affinity of muscle FBPase to aldolase about 14 times, making it resemble the liver isozyme. We suggest that the existence of highly AMP-sensitive muscle-like FBPase, activity of which is regulated by metabolite-dependent interaction with aldolase enables the precise regulation of muscle energy expenditures and might contributed to the evolutionary success of vertebrates.     

Rabbit muscle fructose-1,6-bisphosphatase is phosphorylatedin vivo

Phosphorylated fructose-1,6-bisphosphatase (FBPase) was isolated from rabbit muscle in an SDS/PAGE homogeneous form. Its dephosphorylation with alkaline phosphatase revealed 2.8 moles of inorganic phosphate per mole of FBPase. The phosphorylated FBPase (P-FBPase) differs from the dephosphorylated enzyme in terms of its kinetic properties like K(m) and k(cat), which are two times higher for the phosphorylated FBPase, and in the affinity for aldolase, which is three times lower for the dephosphorylated enzyme. Dephosphorylated FBPase can be a substrate for protein kinase A and the amount of phosphate incorporated per FBPase monomer can reach 2-3 molecules. Since interaction of muscle aldolase with muscle FBPase results in desensitisation of the latter toward AMP inhibition (Rakus & Dzugaj, 2000, Biochem. Biophys. Res. Commun. 275, 611-616), phosphorylation may be considered as a way of muscle FBPase activity regulation.     

Identification of Novel Vesicles in the Cytosol to Vacuole Protein Degradation Pathway

The key gluconeogenic enzyme, fructose1,6-bisphosphatase (FBPase), is induced when Saccharomyces cerevisiae are starved of glucose. FBPase is targeted from the cytosol to the yeast vacuole for degradation when glucose-starved cells are replenished with fresh glucose. Several vid mutants defective in the glucose-induced degradation of FBPase in the vacuole have been isolated. In some vid mutants, FBPase is found in punctate structures in the cytoplasm. When extracts from these cells are fractionated, a substantial amount of FBPase is sedimentable in the high speed pellet, suggesting that FBPase is associated with intracellular structures in these vid mutants. In this paper we investigated whether FBPase association with intracellular structures also existed in wild-type cells. We report the purification of novel FBPase-associated vesicles from wild-type cells to near homogeneity. Kinetic studies indicate that FBPase association with these vesicles is stimulated by glucose and occurs only transiently, suggesting that these vesicles are intermediate in the FBPase degradation pathway. Fractionation analysis demonstrates that these vesicles are distinct from known organelles such as the vacuole, ER, Golgi, mitochondria, peroxisomes, endosomes, COPI, or COPII vesicles. Under EM, these vesicles are 30–40 nm in diam. Proteinase K experiments indicate that the majority of FBPase is sequestered inside the vesicles. We propose that FBPase is imported into these vesicles before entering the vacuole.     

Subcellular localization of liver FBPase is modulated by metabolic conditions

In primary cultured hepatocytes, fructose-1,6-bisphosphatase (FBPase) localization is modulated by glucose, dihydroxyacetone (DHA) and insulin. In the absence of these substrates, FBPase was present in the cytoplasm, but the addition of glucose or DHA induced its translocation to the nucleus. As expected, we observed the opposite effect of glucose on glucokinase localization. The addition of insulin in the absence of glucose largely increased the amount of nuclear FBPase. Moreover, at high concentrations of glucose or DHA, FBPase shifted from the cytosol to the cell periphery and co-localized with GS. Interestingly, the synthesis of Glu-6-P and glycogen induced by DHA was not inhibited by insulin. These results indicate that FBPase is involved in glycogen synthesis from gluconeogenic precursors. Overall, these findings show that translocation may be a new integrative mechanism for gluconeogenesis and glyconeogenesis.     

Subcellular localization of muscle FBPase in carp (Cyprinus carpio) tissues

Subcellular localization of muscle FBPase-a regulatory enzyme of glyconeogenesis-was investigated in carp using immunohistochemistry and protein exchange method. Results of the experiments revealed that, in striated muscles, FBPase associates with alpha-actinin of the Z-line and co-localizes with aldolase. Additionally, in cardiac and smooth muscle cells FBPase is present inside the nuclei. In the light of findings on mammalian muscle FBPase, the data presented here indicates that interaction of the enzyme with specific cellular partners and nuclear presence of FBPase is a general phenomenon in contemporary vertebrates.     

Regulation of subcellular localization of muscle FBPase in cardiomyocytes. The decisive role of calcium ions

Glyconeogenesis, the synthesis of glycogen from carbohydrate precursors like lactate, seems to be an important pathway participating in replenishing glycogen in cardiomyocytes. Fructose-1,6-bisphosphatase (FBPase), an indispensible enzyme of glyconeogenesis, has been found in cardiomyocytes on the Z-line, in the nuclei and in the intercalated discs. Glyconeogenesis may proceed only when FBPase accumulates on the Z-line. Searching for the mechanism of a FBPase regulation we investigated the effects of the calcium ionophore A23187, a muscle relaxant dantrolene, glucagon, insulin and medium without glucose on the subcellular localization of this enzyme in primary culture of neonatal rat cardiomyocytes. Immunofluorescence was used for protein localization and the intracellular calcium concentration was measured with Fura. We found that the concentration of calcium ions was the decisive factor determining the localization of muscle FBPase on the Z-line. Calcium ions had no effect on the localization of the enzyme in the intercalated discs or in the nuclei, but accumulation of FBPase in the nuclei was induced by insulin.     

Colocalization of muscle FBPase and muscle aldolase on both sides of the Z-line

Previously we have reported that in vitro muscle aldolase binds to muscle FBPase [Biochem. Biophys. Res. Commun. 275 (2000) 611-616] which results in the changes of regulatory properties of the latter enzyme. In the present paper, the evidence that aldolase binds to FBPase in living cell is presented. The colocalization experiment, in which aldolase was diffused into skinned fibres that had been pre-incubated with FBPase, has shown that aldolase in the presence of FBPase binds predominantly to the Z-line. The existence of a triple aldolase-FBPase-alpha-actinin complex was confirmed through a real-time interaction analysis using the BIAcore biosensor. The colocalization of FBPase and aldolase on alpha-actinin of the Z-line indicates the existence of glyconeogenic metabolon in vertebrates' myocytes.     

Survey of fructose 1,6-bisphosphatase isoenzyme in rat organs and ontogenic expression of the enzyme in rat fetus

1. Among eleven tissues of rat, the liver type of fructose 1,6-bisphosphatase (FBPase) subunit was detected in the liver, kidney, testis, pancreas and lung by Western blot analysis using anti-(liver FBPase) or anti-(muscle FBPase) serum. 2. The muscle type of the enzyme subunit was detected only in the pancreas other than skeletal muscle. Both types of the enzyme subunit were found in the pancreas. 3. Neither anti-(liver FBPase) nor anti-(muscle FBPase) serum detected the band of enzyme subunit on the blots of the extracts of brain, heart, small intestinal mucosa, spleen and placenta. 4. FBPase is present in fetal rat liver at least as early as the 14th day of gestation. 5. In agreement with the increase in immunological staining density, the level of the enzyme activity in fetal liver increased exponentially during fetal development. 6. The muscle enzyme was not detected until the fetus reached the 19th day of gestation.     

A comparative study on the sensitivity of Cyprinus carpio muscle and liver FBPase toward AMP and calcium

The activity of fructose-1,6-bisphosphatase (FBPase; EC 3.1.3.11) isozymes is influenced by AMP, Ca2+ and by reversible interactions with subcellular structures. In contrast to mammalian and avian isozymes, the kinetic properties of FBPases from ectothermal vertebrates are not fully described. To get some insight into mechanism of glycogen resynthesis in ectothermal vertebrates we examined the features of FBPases isolated from Cyprinus carpio skeletal muscle and liver. To investigate the evolutionary origin of the sensitivity of FBPase to effectors, we performed a phylogenetic analysis of known animal amino acids sequences of the enzyme. Based on our findings, we hypothesize that the high, mammalian-like, sensitivity of FBPase to Ca2+ is not essential for controlling the stability of glyconeogenic complex in striated muscles, instead it ensures the precise regulation of mitochondrial metabolism during prolonged Ca2+ elevation in contracting muscle fibers. Comparison of the kinetic properties of vertebrate and insect FBPases suggests that the high sensitivity of muscle isozyme to inhibitors has arisen as an adaptation enabling coordination of energy metabolism in warm-blooded animals.     

Colocalization of aldolase and FBPase in cytoplasm and nucleus of cardiomyocytes

The protein exchange method, immunocytochemistry and the nuclear import of fluorophore-labeled enzymes were used to investigate the colocalisation of aldolase and FBPase in cardiomyocytes. The results indicate in vivo interaction of these two enzymes. In the cardiomyocyte cytoplasm, these enzymes were found to colocalise at the Z-line and on intercalated discs. The translocation of both enzymes through the nuclear pores was also investigated. The immunocytochemistry revealed the colocalisation of aldolase and FBPase in the heterochromatin region of cardiomyocyte nuclei. The Pearson's correlation coefficients, which represent the degree of colocalisation were 0.47, 0.52 and 0.66 in the sarcomer, the intercalated disc and the nucleus, respectively. This is the first report on aldolase and FBPase colocalisation in cardiomyocytes. Interaction of aldolase with FBPase, which results in heterologous complex formation, is necessary for glyconeogenesis to proceed. Therefore, this metabolic pathway in the sarcomer, in the intercalated disc as well as in the nucleus might be expected.     

Late steps of egg cell differentiation are accelerated by pollination in Zea mays L.

 Egg cells were analysed cytologically during the female receptivity period in maize (Zea mays L., line A 188). Three classes of egg cell were distinguished: type A – small, non-vacuolated cells with a central nucleus; type B – larger cells with small vacuoles surrounding the perinuclear cytoplasm located in the middle of the cell; type C – big cells with a large apical vacuole and the mid-basal perinuclear cytoplasm. The less-dense cytoplasm of the vacuolated egg cells usually contained numerous cup- or bell-shaped mitochondria. The three egg types appear to correspond to three late stages of egg cell differentiation. The frequencies of each of the three egg types were monitored in developing maize ears before and after pollination. In young ears, with the silks just extending out of the husks, small A-type cells were found in about 86% of ovules. Their frequency decreased to about 58% at the optimum silk length, remained unchanged in non-pollinated ears, and fell to 16% at the end of the female receptivity period. However, after pollination and before fertilisation the frequency of these cells decreased to about 33%, and the larger vacuolated egg cells (types B and C) prevailed. At various stages of the receptivity period, pollination accelerated changes in the egg population, increasing the number of ovules bearing larger, vacuolated egg cells. Experiments with silk removal demonstrated that putative pollination signals act immediately after pollen deposition and are not species-specific. Received: 5 February 1999 / Accepted: 28 August 1999