Pfkfb3 is transcriptionally upregulated in diabetic mouse liver through proliferative signals Joan Duran 1, *, Merce ` Obach 1, *, Aurea Navarro-Sabate 1 , Anna Manzano 1 , Marta Go ´ mez 1 , Jose L. Rosa 1 , Francesc Ventura 1 , Jose C. Perales 2 and Ramon Bartrons 1 1 Unitat Bioquı ´ mica i Biologia Molecular, Universitat de Barcelona, Spain 2 Unitat de Biofı ´ sica, Departament de Cie ` ncies Fisiolo ` giques, IDIBELL, Universitat de Barcelona, Spain Introduction Diabetes is a common metabolic disorder in humans, associated with significant morbidity and mortality. In this pathological situation, the liver, one of the major targets of insulin action, develops biochemical and functional abnormalities, which include alterations in carbohydrate, lipid and protein metabolism and changes in antioxidant status [1]. Insulin-dependent diabetes mellitus is currently modelled by the injection of streptozotocin (STZ) in rodents, which degenerates pancreatic insulin-producing b-cells [2]. This model is characterized by decreased plasma insulin levels, severe hyperglycaemia and alterations in insulin-dependent signal transduction [3]. STZ-induced diabetes in rats is also associated with hepatomegaly as a result of the Keywords 6-phosophofructo-2-kinase ⁄ fructose-2,6- bisphosphatase; diabetes; fructose-2,6- bisphosphate; liver; streptozotocin Correspondence R. Bartrons, Unitat Bioquı ´ mica i Biologia Molecular, Universitat de Barcelona, Feixa Llarga s ⁄ n, E-08907 L’Hospitalet, Barcelona, Spain Fax: 34934024268 Tel: 34934024252 E-mail: rbartrons@ub.edu *These authors contributed equally to this work (Received 3 April 2009, revised 12 June 2009, accepted 17 June 2009) doi:10.1111/j.1742-4658.2009.07161.x The ubiquitous isoform of 6-phosphofructo-2-kinase ⁄ fructose-2,6-bisphos- phatase (uPFK-2), a product of the Pfkfb3 gene, plays a crucial role in the control of glycolytic flux. In this study, we demonstrate that Pfkfb3 gene expression is increased in streptozotocin-induced diabetic mouse liver. The Pfkfb3 ⁄ -3566 promoter construct linked to the luciferase reporter gene was delivered to the liver via hydrodynamic gene transfer. This promoter was upregulated in streptozotocin-induced diabetic mouse liver compared with transfected healthy cohorts. In addition, increases were observed in Pfkfb3 mRNA and uPFK-2 protein levels, and intrahepatic fructose-2,6-bisphos- phate concentration. During streptozotocin-induced diabetes, phosphoryla- tion of both p38 mitogen-activated protein kinase and Akt was detected, together with the overexpression of the proliferative markers cyclin D and E2F. These findings indicate that uPFK-2 induction is coupled to enhanced hepatocyte proliferation in streptozotocin-induced diabetic mouse liver. Expression decreased when hepatocytes were treated with either rapamycin or LY 294002. This shows that uPFK-2 regulation is phosphoinositide 3-kinase–Akt–mammalian target of rapamycin dependent. These results indicate that fructose-2,6-bisphosphate is essential to the maintenance of the glycolytic flux necessary for providing energy and biosynthetic precursors to dividing cells. Abbreviations C ⁄ EBP, CCAAT ⁄ enhancer-binding protein; CDK, cyclin-dependent kinase; EGF, epidermal growth factor; EMSA, electrophoresis mobility shift assay; ERK, extracellular signal-regulated kinase; Fru-2,6-P 2 , fructose-2,6-bisphosphate; GFP, green fluorescent protein; iNOS, inducible nitric oxide synthase; LAP, liver activation protein; LPS, lipopolysaccharide; mTOR, mammalian target of rapamycin; mTORC 1 ⁄ 2, mTOR complex 1 ⁄ 2; NF jB, nuclear factor kappa-light-chain-enhancer of activated B cells; PCNA, proliferating cell nuclear antigen; PEPCK, phosphoenolpyruvate carboxykinase; PFK-2, 6-phosphofructo-2-kinase ⁄ fructose-2,6-bisphosphatase (EC 2.7.1.105 ⁄ EC 3.1.3.46); PI3K, phosphoinositide 3-kinase; Rb, retinoblastoma; ROS, reactive oxygen species; STZ, streptozotocin; TBARS, thiobarbituric acid reactive substances; uPFK-2, ubiquitous PFK-2. FEBS Journal 276 (2009) 4555–4568 ª 2009 The Authors Journal compilation ª 2009 FEBS 4555 high cell proliferation rates and decreased apoptosis [1,3,4]. In addition, the mechanisms that regulate cell division are upregulated in STZ-induced diabetic mice. This observation is consistent with the robust repair of tissue damage caused by hepatotoxicants observed in diabetic mouse liver [4]. On days 5 and 10 after STZ treatment, significantly higher numbers of G2 cells were found in diabetic liver compared with controls [3,4]. Cell proliferation and tumour growth are supported by high glycolytic flux. This is mainly controlled by 6-phosphofructo-1-kinase, which is potently activated by the regulatory metabolite fructose-2,6-bisphosphate (Fru-2,6-P 2 ) [5,6]. 6-Phosphofructo-2-kinase ⁄ fructose- 2,6-bisphosphatase (PFK-2) is a homodimeric enzyme that catalyses the synthesis and degradation of Fru- 2,6-P 2 [6–9]. Since the discovery of this system in the liver, other mammalian isozymes have been identified with a range of expression profiles and kinetic responses to allosteric effectors, hormonal and growth factor signals [7–10]. These isozymes are generated by alternative splicing from four independent genes, desig- nated Pfkfb1–4 [11]. The Pfkfb3 gene encodes a ubiq- uitous PFK-2 (uPFK-2) isozyme [12], which is induced by progesterone [13], inflammatory stimuli [14] and hypoxia [15,16], and is degraded through the ubiqu- itin–proteasome proteolytic pathway [17]. The Pfkfb3 gene product has the highest kinase to bisphosphatase activity ratio and thus maintains elevated Fru-2,6-P 2 levels, which, in turn, sustain high glycolytic rates in the cell [18]. This gene has been implicated in cell pro- liferation as it is ubiquitously expressed in proliferating tissues, transformed cell lines and in various tumours [13,14,19–22]. Recently, in order to determine the effects of uPFK-2 overexpression in mouse liver and to examine its involvement in metabolic disturbances, we designed a transgenic mouse model that overexpresses Pfkfb3. These transgenic animals sustained high Fru- 2,6-P 2 levels in the liver and increased weight gain [23]. In the liver of STZ-induced diabetic rats, the levels of Fru-2,6-P 2 and 6-phosphofructo-2-kinase activity decreased and the phosphorylation of the bifunctional enzyme increased, correlating with a fall in hepatic Fru-2,6-P 2 , ketonaemia and glycaemia [24–26]. Similar results have been reported in diabetic mouse liver, underscoring the role played by Fru-2,6-P 2 in the con- trol of fuel metabolism [27]. In the present study, we demonstrate that Pfkfb3 gene expression increases pro- gressively in STZ-induced diabetic mouse liver, leading to progressive and partial recovery of Fru-2,6-P 2 levels, and implicating this gene in liver metabolism. In addi- tion, we developed an in vivo promoter assay method based on a hydrodynamic gene delivery technique in order to determine whether the increased Pfkfb3 expression in diabetic liver was a result of transcrip- tional upregulation via promoter activation. The rela- tionship between hepatocyte proliferation and Pfkfb3 gene induction in STZ-induced diabetic mouse liver was also studied. Our results strongly support the hypothesis that this gene is transcriptionally upregulat- ed through cell proliferation pathways, involving Akt phosphorylation and cyclin D and E2F transcription factor transactivation in the liver. Results Pfkfb3 gene expression and Fru-2,6-P 2 concentra- tion in STZ-induced diabetic mouse liver Fifteen days after STZ injection, C57 ⁄ BL6 mice showed significantly higher plasma glucose levels (257.4 ± 29.2 versus 61.3 ± 8.9 mgÆdL )1 in noninject- ed controls) and almost nondetectable plasma insulin levels (< 0.15 lgÆL )1 ) after 16 h of starvation (Fig. 1). In these conditions, we analysed Pfkfb3 mRNA expression and protein levels. As shown in Fig. 2A, Pfkfb3 mRNA expression increased significantly between day 8 and day 15 after STZ injection to a peak on day 15. UPFK-2 protein expression also rose progressively during the time course of the experiment (Fig. 2B). To assess the functionality of the overex- pressed uPFK-2 isozyme, we next analysed the Fru- 2,6-P 2 concentration in liver. The concentration of hepatic Fru-2,6-P 2 decreased after fasting, recovering slowly in STZ diabetes, and reaching 30% of fed values on day 15 after injection (Fig. 2C). In order to assess the overall contribution of uPFK- 2 compared with the other isozymes, we also measured the mRNA and protein levels of the other isozymes at day 0 and 15 of STZ-induced diabetes. Significant vari- ation in the levels of uPFK-2 expression and protein 0 100 200 300 400 500 600 700 0 2 4 6 8 10 15 Days after STZ Fed Fasted > 600 mg·dL –1 Glycaemia (mg·dL –1 ) Fig. 1. Blood glucose levels during the STZ-induced diabetes time course in fed and fasting conditions (n = 10 animals per group). Pfkfb3 upregulation in STZ-induced diabetic mouse liver J. Duran et al. 4556 FEBS Journal 276 (2009) 4555–4568 ª 2009 The Authors Journal compilation ª 2009 FEBS were found after STZ treatment (Fig. 3A,B). The mRNA expression of the other isozymes either did not change significantly (PFKFB1) or decreased (PFKFB4). In addition, we measured the ‘total’ and ‘active’ PFK-2 activities. In the conditions of the assay, the ‘total’ and ‘active’ forms correspond to the V max activity and to the activity of the nonphosphory- lated form of the enzyme, respectively [28,29]. Both the ‘total’ and ‘active’ forms increased after STZ treatment (Fig. 4). Compared with the ‘total’ activity, the ‘active’ form was low in the liver of starved animals (day 0), suggesting that the enzyme present is inhibited by phosphorylation (PFKFB1). In contrast, the activity of the ‘active’ form on day 15 increased, in spite of the fact that the animals were starved and diabetic, suggesting an isoenzymatic change. 0 2 4 6 8 10 12 14 Day 0 Day 2 Day 4 Day 6 Day 8 Day 10 Day 15 ** ** ** Pfkfb3 expression (fold change) 0 0.4 0.8 1.2 1.6 2.0 2.4 2.8 Liver Fru-2,6-P 2 (nmol·g –1 ) 5.6 6.0 Fed ** ** ** * * Days after STZ 0 6 8 0 10 150 2 4 uPFK-2 Loading control Day 0 Day 2 Day 4 Day 6 Day 8 Day 10 Day 15 A B C Fig. 2. Pfkfb3 gene expression analysis in livers from STZ-induced diabetic mice. (A) Quantitative real-time PCR analysis of Pfkfb3 expression was performed using RNA extracts from mouse livers 0, 2, 4, 6, 8, 10 and 15 days after STZ injection (n = 10 animals per group). The data represent the fold change versus the lowest day 0 value, and are normalized to 18S cDNA. Statistically significant dif- ferences (**P < 0.01) in diabetic mouse livers at 8, 10 and 15 days after STZ injection were observed compared with controls (day 0). (B) Western blot against uPFK-2 was performed with 50 lg of total cell extract from the same animals. Protein was used as loading control. (C) Liver Fru-2,6-P 2 values in fasted control (day 0) and 2, 4, 6, 8, 10 and 15-days after STZ injection. All points and bars rep- resent the mean ± standard error of the mean (SEM) of the data obtained (n = 10 animals per group). Statistically significant differ- ences (*P < 0.05; **P < 0.01) were found on 2, 6, 8, 10 and 15 days after STZ versus control (day 0). Fed control value (in grey) is indicated as a reference. CT A B STZ uPFK-2 (PFKFB3) (day = 15) LPFK-2 (PFKFB1) tPFK-2 (PFKFB4) Loading control 1.5 0 0.5 1 Pfkfb1 expression (fold change) Pfkfb3 expression (fold change) Pfkfb4 expression (fold change) 10 ** 0 2 4 6 8 1.5 CT STZ 0 0.5 1 ** (Day = 15) Fig. 3. Expression of the PFKFB isozymes in livers from STZ- induced diabetic mice. Western blot against LPFK-2, uPFK-2 and tPFK-2 (A) and quantitative real-time PCR using specific primers for Pfkfb1, Pfkfb3 and Pfkfb4 genes (B). For western blot, 50 lgof total liver extracts were used. Diabetic mice in the fasting condition (16 h) and 15 days after STZ injection were compared with con- trols. Protein was used as loading control. For Pfkfb3 mRNA quanti- tative analysis, total liver RNA from control (day 0) and STZ-induced diabetic (day 15) mice was used. The data represent the fold change versus the lowest day 0 value and were normalized to 18S cDNA. All graph points and bars represent the mean ± standard error of the mean (SEM) of the data obtained. Statistically signifi- cant increases (**P < 0.01) in diabetic mouse livers compared with controls were observed for Pfkfb3 gene determination. J. Duran et al. Pfkfb3 upregulation in STZ-induced diabetic mouse liver FEBS Journal 276 (2009) 4555–4568 ª 2009 The Authors Journal compilation ª 2009 FEBS 4557 To identify possible liver damage caused by STZ treatment, we measured plasma transaminase levels. Alanine aminotransferase activities increased slightly only during the first 5 days (37.8 ± 9.1 UÆL )1 on the fifth day versus 23.3 ± 5.6 UÆL )1 in controls), return- ing to control values afterwards. uPFK-2 immunohistochemical analysis in the liver uPFK-2 isozyme was overexpressed in the hepatic parenchyma of diabetic mice (Fig. 5A). The expression of proliferating cell nuclear antigen (PCNA) was also increased at day 15 (Fig. 5B). Detailed observation of uPFK-2-positive cell distribution revealed a clustering formation of these hepatocytes (Fig. 5A,C), in accor- dance with a previous report of a PCNA expression pattern in mice liver 5 and 10 days after STZ injection [4]. Next, hydrodynamic transfection of the green fluo- rescent protein (GFP) expression vector was performed to distinguish between perivenous and periportal hepatocytes [30]. UPFK-2-overexpressing hepatocytes were predominantly located in the perivenous zone [31] of the liver (Fig. 5C, merged). Mouse liver transfection of Pfkfb3/-3566 promoter construct during STZ-induced diabetes development To elucidate whether increased Pfkfb3 expression was caused by its transcriptional upregulation via promoter activation, we developed an in vivo promoter assay method based on the hydrodynamic gene delivery tech- nique. Hydrodynamic gene transfer is an efficient sys- tem that allows the DNA to distribute mainly to the liver [30]. The Pfkfb3 ⁄ -3566 promoter construct (con- 0 2 4 6 8 10 12 14 16 18 20 Da y 0 Da y 15 PFK-2 activity (µU·(mg protein) –1 ) PFK-2 activity (µU·(mg protein) –1 ) * * Total PFK-2 Act i v i ty Act i ve PFK-2 Act i v i ty 0 1 2 3 4 5 6 7 8 * Da y 0 Da y 15 Fig. 4. Hepatic PFK-2 activity. Liver ‘total’ and ‘active’ PFK-2 activi- ties in fasted control (day 0) and at day 15 after STZ-induced diabe- tes. All graph points and bars represent the mean ± standard error of the mean (SEM) of the data obtained (n = 10 animals per group). Statistically significant differences (*P < 0.05; **P < 0.01) were found in diabetic animals versus controls (day 0). Control liver A B C D Diabetic liver (day 15 after STZ injection) uPFK-2 uPFK-2 GFP Merged PCNA Control STZ (day 15) Loading control 14 ** 4 6 8 10 12 * * Pfkfb3 promoter-luciferase activity (fold induction) 0 2 Day 0 Day15 Day 10 Day 8 Day 6 Day 4 Day 2 Fig. 5. UPFK-2 immunostaining and hydrodynamic transfection analysis of Pfkfb3 ⁄ -3566 promoter construct. (A) uPFK-2 immuno- staining in control and diabetic mouse livers. Fixed liver samples included in OCT were cut and prepared for immunohistochemistry procedures. Immunostaining was performed by indirect immunoflu- orescence using uPFK-2 (1 : 10) primary antibody, followed by an rabbit IgG secondary antibody conjugated to Alexa-Fluor 568. Omis- sion of primary antibody was used as a negative control. (B) For western blot against PCNA, 50 lg of total liver extract were used and protein was employed as a loading control. (C) Animals (n =10 for each condition) were cotransfected, using hydrodynamic gene delivery, with Pfkfb3 ⁄ -3566 promoter construct, and GFP expres- sion vector was injected through the mouse tail vein in a volume of 10% of the body weight. The liver transfection efficiency was assessed using the percentage of hepatocytes expressing GFP. Clusters of hepatocytes overexpressing uPFK-2 colocalize with GFP in perivenous cells. (D) Hydrodynamic transfection analysis of Pfkfb3 ⁄ -3566 promoter construct at baseline (day 0) and 2, 4, 6, 8, 10 and 15 days after STZ injection. Statistically significant differ- ences in luciferase activity were observed in livers from mice on days 4, 10 (*P < 0.05) and 15 (**P < 0.01) after STZ injection com- pared with controls. Pfkfb3 upregulation in STZ-induced diabetic mouse liver J. Duran et al. 4558 FEBS Journal 276 (2009) 4555–4568 ª 2009 The Authors Journal compilation ª 2009 FEBS taining a 3566-nucleotide fragment of the Pfkfb3 pro- moter) linked to the luciferase reporter gene was deliv- ered into mouse liver during diabetes development. As indicated by the cotransfection of Pfkfb3 ⁄ -3566 and GFP constructs (Fig. 5C), approximately 20–40% of the liver cells were transfected. Moreover, no signifi- cant differences were found in alanine aminotransfer- ase levels 24 h after transfection between animal groups (Pfkfb3 ⁄ -3566 + GFP; GFP). Alanine amino- transferase levels were in the range of those receiving saline (data not shown), indicating that the liver was not affected after transfection treatment. Transient in vivo transfection of the Pfkfb3 ⁄ -3566 promoter construct demonstrated significant luciferase activity on day 4 (around four-fold), and large increases (8–12-fold) on days 10 and 15 after STZ injection, in comparison with basal values (Fig. 5D). Involvement of pro-inflammatory signals and oxidative stress in Pfkfb3 expression in diabetic liver Nuclear factor kappa-light-chain-enhancer of activated B cells (NF-jB) has been found to be expressed in liver epithelium, where it regulates hepatic cell prolifer- ation and survival during regeneration and develop- ment [32]. Furthermore, we have previously described various NF-jB consensus sequences in the Pfkfb3 gene promoter [16]. In the light of these data, we examined whether NF-jB might be responsible for Pfkfb3 activa- tion in our diabetic model. The presence of NF-jBin liver nuclear extracts from days 0, 4, 8 and 15 after STZ injection was studied by electrophoresis mobility shift assay (EMSA). No changes in phosphorylated NF-jB oligonucleotide interactions were found between the various time course samples (Fig. 6A). In addition, in order to rule out NF-jB involvement in Pfkfb3 upregulation, we used RAW wild-type and RAW IjBa dominant active (IjB aDA) cells [33]. In RAW wild-type cells, inducible nitric oxide synthase (iNOS) expression increased gradually 8, 16 and 24 h after lipopolysaccharide (LPS) treatment; at the same time, NF-jB was induced. Moreover, no expression of this pro-inflammatory marker was detected in RAW IjB aDA cells after LPS treatment. In these condi- tions, small changes in uPFK-2 protein levels were found in the presence or absence of LPS in both cell lines (Fig. 6B). Furthermore, no iNOS expression was detected in any liver sample from any day of the study (data not shown). The steady-state levels of lipoperoxi- dation product (thiobarbituric acid reactive substances, TBARS) concentration and catalase activity were determined to rule out the involvement of oxidative stress in our STZ diabetic model. No significant differ- ences were found between post-STZ injection liver samples (results not shown). Cell growth and proliferation in STZ-induced diabetic mouse liver Several reports have described a significantly larger number of G2 cells in STZ-induced diabetic mouse liver than in nondiabetic cohorts [4]. Moreover, Pfkfb3 gene expression has also been found to be increased in proliferating cells [22,34]. We studied various cell growth and proliferation markers in order to find a plausible explanation for uPFK-2 overexpression in STZ-induced diabetic mouse liver. The hepatocyte pro- liferation observed in response to growth and auto- crine factors is attempted, at least in part, via the activation of the phosphoinositide 3-kinase (PI3K) pathway and its downstream signal transduction effec- tors [35–38]. In addition, the predominant role of PI3K and the mammalian target of rapamycin (mTOR) in DNA replication and cyclin D activation has been reported [35,36]. To evaluate the involvement of this pathway in our STZ-induced diabetic model, phosphorylation of Akt on Ser473 (P-Akt Ser473) [39] and cyclin D expression were studied. Moreover, it has been speculated that, in type I diabetes mellitus, p38 Days after STZ c+ 0 48 15 Hours after LPS treatment 8816 1624 240 iNOS uPFK-2 RAW mock RAW DA Loading control A B Fig. 6. Oxidative stress analysis. (A) Fresh liver nuclear extracts from days 0, 4, 8 and 15 after STZ injection were tested for the presence of NF-jB transcription factor by EMSA. A 32 P-labelled oligonucleotide containing the NF-jB consensus binding site was used as probe. A nuclear cell extract from SH-SY-5Y cells was used as positive control (c+). (B) Western blot of RAW WT and RAW IjB aDA cells treated with LPS (1 lgÆmL )1 ) for 0, 8, 16 and 24 h. Fifty micrograms of total cell extracts were blotted using antibodies against iNOS (as positive control) and uPFK-2 enzymes. Protein was used as a loading control. J. Duran et al. Pfkfb3 upregulation in STZ-induced diabetic mouse liver FEBS Journal 276 (2009) 4555–4568 ª 2009 The Authors Journal compilation ª 2009 FEBS 4559 mitogen-activated protein kinase, a sensor of oxidative stress, may also control Akt phosphorylation [40]. Therefore, in our model, we also analysed the phos- phorylation of p38 (P-p38). Another well-known fam- ily of transcription factors involved in hepatocyte dedifferentiation and proliferation is the CCAAT⁄ en- hancer-binding protein (C ⁄ EBP). We studied the C ⁄ EBPb liver activation protein (LAP), involved in hepatocyte proliferation, and C ⁄ EBPa, involved in cell cycle arrest [41,42]. C ⁄ EBPa is necessary for hepatic growth inhibition, but this activity can be blocked in liver tumours by the PI3K ⁄ Akt signal transduct ion pathway. This pathway dephosphorylates C ⁄ EBPa on Ser193, activating cell growth and proliferation through sequestering retinoblastoma (Rb) protein, and leading to a reduction in Rb-E2F blocking complexes [38]. In addition, E2F r elease can also be caused by cyclin D activity. In the light of these data, the presence of C ⁄ EBPb (LAP), C ⁄ EBPa and E2F was investigated. Western blot analysis of STZ-induced diabetic mouse liver revealed phosphorylation of Akt at Ser473 between days 4 and 15 (Fig. 7A). However, phosphor- ylation of p38 mitogen-activated protein kinase was only detected on day 4. In addition, the results obtained for C ⁄ EBPb protein levels agree with those reported for hepatocyte proliferation, as the LAP iso- form was overexpressed on day 15 after STZ adminis- tration (Fig. 7B). However, C ⁄ EBPa expression was barely detected at day 15 after STZ injection. Cyclin D and E2F1, two final effectors of these proliferative pathways, were increased from day 4, correlating posi- tively with the uPFK-2 upregulation pattern during the time course presented above. Inhibition of PI3K ⁄ mTOR pathway reduces uPFK-2 expression in proliferating rat primary hepatocytes Rat primary hepatocytes coated on a collagen mono- layer exhibit rapid proliferation and dedifferentiation, progressing in G1 independent of growth factor stimula- tion up to the restriction point located in the mid–late G1 phase. After mitogenic stimulation, hepatocytes progress in late G1 and undergo DNA synthesis [36]. A complex network of different signalling cascades, including the PI3K, extracellular signal-regulated kinase (ERK) and p38 pathways [43] participate in the regula- tion of hepatocyte proliferation and survival. With regard to the signalling pathways that control the tran- scription of the Pfkfb3 gene, freshly isolated hepato- cytes, cultured with 10% fetal bovine serum medium in order to promote cell proliferation, were treated with various inhibitors, and several signal cascade intermediates were studied. LY 294002, a PI3K inhibi- tor, and rapamycin, a specific mTOR inhibitor, were both able to reduce uPFK-2 levels by approximately 20% compared with the serum-activated control condi- tion, whereas treatments with specific inhibitors of ERK and p38 pathways did not affect uPFK-2 expression (Fig. 8A,B). An additional experiment was performed to test the hypothesis that the Pfkfb3 gene is modulated by the PI3K pathway in proliferating liver cells. Epidermal growth factor (EGF)-stimulated and nonstimulated primary hepatocytes were treated with the PI3K inhibi- tor LY 294002. EGF promotes cell cycle progression and DNA synthesis in hepatocyte cultures [35] through an LY 294000-sensitive pathway. The results in Fig. 8C show that uPFK-2 protein was decreased by blocking PI3K activity in both the EGF-stimulated and nonstimulated hepatocytes 48 h after seeding. Cyclin D levels were also reduced in the presence of LY 294002, confirming the link between uPFK-2 expression and hepatocyte proliferative status. Quanti- tative real-time PCR performed in EGF-stimulated conditions showed a significant reduction in Pfkfb3 mRNA levels when hepatocytes were treated with LY 294002 (Fig. 8D). Discussion Even with insulin treatment, diabetic patients show disturbances in tissue growth, and these have been linked to chronic hyperglycaemia and subsequent met- abolic alterations [3]. In type I diabetes, the liver, which maintains blood glucose homeostasis, contrib- utes to hyperglycaemia. STZ-induced diabetic rats show hepatocyte proliferation, decreased apoptosis and hypertrophy in the liver [3]. Moreover, Shankar et al. [4] have demonstrated that diabetic mice exhibit milder liver injury after exposure to lethal doses of hepatotox- icants, suggesting a robust compensatory tissue repair in this experimental situation. These results suggest that liver damage is repaired as a result of hepatic cell proliferation. However, in a previous study, Rosa et al. [44] reported an increase in Fru-2,6-P 2 content during the replicative period of liver regeneration, correlating with transcriptional activation and PFK-2 mRNA accumulation. They suggested that PFK-2 was regu- lated in response to hepatic insult. Recently, we have demonstrated the upregulation of the Pfkfb3 gene and the uPFK-2 isozyme in hepatic cell growth and prolif- eration [22]. Bearing in mind that the Pfkfb3 gene is extensively involved in cell proliferation events as a result of its key role in carbohydrate metabolism [16,19,21,22,34,45,46], we analysed its physiological Pfkfb3 upregulation in STZ-induced diabetic mouse liver J. Duran et al. 4560 FEBS Journal 276 (2009) 4555–4568 ª 2009 The Authors Journal compilation ª 2009 FEBS role in liver dysfunction described in diabetes induced by STZ. The detailed diabetes induction time course pre- sented here showed both a gradual increase in uPFK-2 protein and Pfkfb3 mRNA after STZ injection, accom- panied by a progressive transcriptional upregulation of this gene. These increases correlate with the partial recovery of Fru-2,6-P 2 concentration observed in fasted diabetic liver. The contribution of the ubiqui- tous PFK-2 isozyme to Fru-2,6-P 2 synthesis in the dia- betic liver is greater than that of other isozymes, as the mRNA and protein levels of the liver isozyme (PFKFB1) do not change and the testis isoform (PFKFB4) decreases after STZ. Moreover, the hepatic activities measured suggest that the enzyme present at day 0 is inhibited by phosphorylation. In contrast, the activity on day 15 increased, in spite of the fact that the animals were starved and diabetic, suggesting an isoenzymatic change. Taken together, these results demonstrate that the Pfkfb3 gene and uPFK-2 protein are induced in STZ diabetic mouse liver, and are mainly responsible for the changes in Fru-2,6-P 2 concentration. Moreover, our results demonstrate that uPFK-2 is overexpressed in perivenous diabetic liver zones, as assessed by hydrodynamically delivered GFP as a perivenous marker. In the liver, hepatocytes are not terminally differentiated in normal conditions, and can E2F1 Cyclin D uPFK-2 04 08 15 0 2 4 6 8 10 12 0408015Days after STZ Protein expression (folds of induction) 0 2 4 6 8 10 12 Protein expression (folds of induction) 0 2 4 6 8 10 12 Protein expression (folds of induction) * ** * * ** * *** Cyclin D E2F1 uPFK-2 Loading control P-Akt ser473 P-p38 0 4 8 15 Days after A B STZ Loading control C/EBP 33 KDa C/EBP LAP Days after STZ 04 08015 Loading control Fig. 7. Western blot analysis of cell growth and proliferation markers in mouse livers during diabetes induction. Western blot anal- ysis was performed with antibodies against phosphorylated p38 and Akt Ser473 (A) and against C ⁄ EBPb,C⁄ EBPa, cyclin D, E2F1 and uPFK-2 (B). Fifty micrograms of mice liver extracts, obtained in fasting conditions, on day 0, 4, 8 and 15 after STZ injection were used. Protein was used as a loading control. (C) Cyclin D, E2F1 and uPFK-2 pro- tein expression, expressed as fold induction versus day 0, are represented. All graph points and bars represent the mean ± stan- dard error of the mean (SEM) of the data obtained by western blot densitometric scanning from different animals (n = 4–5) (*P < 0.05; **P < 0.01; ***P < 0.001). J. Duran et al. Pfkfb3 upregulation in STZ-induced diabetic mouse liver FEBS Journal 276 (2009) 4555–4568 ª 2009 The Authors Journal compilation ª 2009 FEBS 4561 rapidly enter the cell division cycle on stimulation. Proliferation starts in the periportal zone, but most hepatocytes proliferate after a few days. This phase is associated with the formation of hepatocyte accumula- tion of 8–10 cells (clusters) arranged around immature or disintegrating vascular channels [31,47]. Our results demonstrate that uPFK-2-positive hepatocytes display a cluster distribution in diabetic liver. To determine whether the Pfkfb3 gene was induced via transcriptional upregulation, we performed an in vivo promoter study using hydrodynamic gene deliv- ery. Luciferase activity was determined in liver trans- fected with the Pfkfb3 ⁄ -3566 promoter construct [16]. Our results reveal sustained Pfkfb3 promoter activa- tion in STZ-induced diabetic mouse liver compared with nondiabetic controls, demonstrating that the Pkfbf3 upregulation observed in STZ-induced diabetic animals is a result of its transcriptional activation. As it has been reported that persistent hyperglyca- emia increases the production of oxygen free radicals from glucose autoxidation and protein glycosylation [48,49], we next focused on the possibility that NF-jB might be determinant for Pfkfb3 gene transcription in this model. Livers from STZ-induced diabetic mice were analysed by EMSA in order to test for the pres- ence of p65 NF-jB, but, surprisingly, we found no dif- ferences in this transcription factor among time course samples. To analyse whether NF-jB regulates Pfkfb3 gene expression in an in vitro model, we used RAW wild-type and RAW IjB aDA cells. After LPS uPFK-2 P-p70S6K P-p38 P-ERK Caspase 3 Cyclin D -tubulin P-Akt Ser473 Basal LY 294002 Rapamycin PD 98059 SB 203580 0.0 0.2 0.4 0.6 0.8 1.0 1.2 EGF LY294002 ++ –+ Pfkfb3 expression (relative values) * 0 0.5 1 1.5 uPFK-2 expression (folds of induction) Basal LY 294002 Rapamycin PD 98059 SB 203580 * * t = 48 h LY 294002 EGF + uPFK-2 Cyclin D –– + – + – + -tubulin A B C D Fig. 8. uPFK-2 expression in proliferating primary hepatocytes after treatment with different inhibitors. (A) Rat primary hepatocytes were cultured with Williams E supplemented with 10% fetal bovine serum on collagen-coated plates. Before and during culture, the medium was supplemented with 50 l M LY 294002, 50 nM rapamy- cin, 50 l M PD 98059 and 10 lM SB 203580, inhibiting pI3K, mTOR, ERK and p38, respectively. UPFK-2 expression was analysed in all circumstances, as was cell apoptosis status (caspase-3), using 50 lg of 24-h-treated hepatocyte protein extracts. Anti-P-Akt Ser473, anti-P-p70S6K, anti-P-ERK and anti-P-p38 were used to assess pathway inhibitions. c-Tubulin was used as a loading con- trol. (B) In LY 294002- and rapamycin-treated cells, uPFK-2 protein levels decreased by 20% according to densitometric analysis. All graph points and bars represent the mean ± standard error of the mean (SEM) of the data obtained by western blot densitometric scanning from different animals (n =5)(*P < 0.05). (C) Rat primary hepatocytes treated with LY 294002 and ⁄ or EGF (20 ngÆmL )1 ). UPFK-2 and cyclin D expression were analysed in all samples using 50 lg of 48-h-treated hepatocyte extracts. (D) Quantitative real-time PCR for Pfkfb3 in rat hepatocytes. Total RNA isolated after 48 h of treatment with EGF and EGF + LY 294002 was used. The data rep- resent the fold change versus the highest EGF-treated value and were normalized to 18S cDNA. Graph bars represent the mean ± SEM of three independent experiments. Pfkfb3 upregulation in STZ-induced diabetic mouse liver J. Duran et al. 4562 FEBS Journal 276 (2009) 4555–4568 ª 2009 The Authors Journal compilation ª 2009 FEBS stimulation, slight changes in uPFK-2 protein levels were observed despite NF-jB activation. All of these features also correlate with nonmodified liver TBARS levels and catalase activity, two key markers of hepatic oxidative stress. It seems clear that NF-jB is not necessary for Pfkfb3 induction in our experimental conditions. We then demonstrated that the progressive induction of uPFK-2 after STZ administration was accompanied by the activation of hepatic cell proliferation pathways, as shown by the detection of the activated markers and transcription factors involved. Some of the mark- ers initially studied can be ruled out. For example, the gradual Pfkfb3 gene upregulation is probably not caused by members of the C ⁄ EBP transcription factor family. The increase in C ⁄ EBPb contributes to gluco- neogenesis and phosphoenolpyruvate carboxykinase (PEPCK) protein expression in STZ diabetic mice, favouring hyperglycaemia [42]. Furthermore, total C ⁄ EBPa expression fell on day 8 after STZ injection. p38 involvement can also be ruled out: using the p38 inhibitor SB 203580 in cultured hepatocytes, uPFK-2 expression was maintained, whereas P-Akt Ser473 expression was totally inhibited, confirming the results of a previous study [50]. Hepatocyte proliferation in response to growth and autocrine factors, as well as in hepatocarcinogenesis, is mediated via the activation of the PI3K pathway and its downstream signal transduction effectors [35– 37,51,52]. Akt, one of these effectors, can regulate cell growth through the regulation of mTOR. The mTOR protein assembles into two functionally distinct com- plexes: mTORC1 (mTOR complex 1) and mTORC2 [39]. Activation of Akt indirectly (through phosphory- lation and inhibition of the GTPase-activating protein activity of tuberous sclerosis complex 2) stimulates mTORC1 activity and the phosphorylation of its sub- strates, S6K1 ⁄ S6K2 and 4E-BP1 ⁄ 4E-BP2, thus stimulat- ing translation and cell growth. Activation of mTORC2, the other mTOR-containing complex, by growth fac- tors stimulates the phosphorylation and activation of its substrate Akt on Ser473. Thus, mTOR through mTORC1 and mTORC2 activities participates in the signalling of Akt. These activities can be distinguished using the inhibitor rapamycin, which specifically inhib- its mTORC1, at least in short treatments [39,40,53]. Our results demonstrate that, during the development of diabetes, livers from STZ-treated animals gradually show activation of PI3K and mTORC2, corroborated by a sustained increase in P-Akt Ser473 levels from day 4 after STZ injection onwards (Fig. 7A). Akt pro- motes cell survival by phosphorylating transcription factors that control the expression of pro- and anti- apoptotic genes, and also via cell cycle progression, through several mechanisms including increased cyclin D transcription and translation, inhibitory phosphory- lation and reduced transcription of cyclin-dependent kinase (CDK) inhibitors. In agreement with this, we have also shown that, during diabetes, Akt stimulates protein synthesis by the activation of mTORC1 through the phosphorylation of its target p70-S6K1 on Thr389. The use of specific inhibitors of mTORC1 and PI3K confirmed this observation (Fig. 8A). These data are consistent with previous reports implicating mTORC1 in hepatic proliferation [35,52]. Ser486 phos- phorylation of heart PFK-2 isozyme has been shown to be regulated via PI3K ⁄ Akt signalling [54]. In agree- ment with these results, the data presented here clearly show that uPFK-2 expression decreases when hepato- cytes are treated with rapamycin and LY 294002, and indicate that PFK-2 activity may be regulated by PI3K–Akt–mTOR. A pivotal protein in the process that leads to cell cycle progression and DNA synthesis in hepatocytes is cyclin D. Coutant et al. [36] demonstrated that, in growth factor-stimulated hepatocytes, LY 294002 and also rapamycin completely prevent cyclin D1 activa- tion at the mRNA and protein levels. Our results cor- relate the phosphorylated status of Akt with a progressive induction of cyclin D in diabetic mouse liver from day 8 after STZ injection. This observation supports the hypothesis that the complex network that leads hepatocytes to proliferate is active in diabetic liver. D-type cyclin levels are high in proliferating cells, and Yamamoto et al. [55] reported that the activation of the PI3K–Akt pathway is essential for the nuclear shift of cyclin D. Sustained mitogenic signals stimulate transcriptional activation of the D-type cyclin genes, synthesis of cyclin D proteins and their assembly with cdk4 ⁄ 6. The activated cyclin D–cdk 4 ⁄ 6 complex phos- phorylates Rb protein, disrupting its association with E2F and allowing the transcriptional activation of S-phase genes [55]. One of the genes whose expression is increased at the G1 ⁄ S transition of the cell cycle is PFK-2 [56]. Furthermore, we have also found that Pfkfb3 gene silencing induces cell cycle delay, corrobo- rating its role in sustaining high glycolytic flux in pro- liferative cells and its involvement in cell cycle progression [20]. E2F transcription factors regulate the timely expres- sion of a series of genes whose products are essential for cell proliferation. Free E2F activates transcription, but, when associated with Rb family members, it func- tions as a transcriptional repressor. E2F DNA-binding sites have been identified in promoters of many genes that are central to the regulation of cell cycle progres- J. Duran et al. Pfkfb3 upregulation in STZ-induced diabetic mouse liver FEBS Journal 276 (2009) 4555–4568 ª 2009 The Authors Journal compilation ª 2009 FEBS 4563 sion [57]. Our hypothesis that diabetic mouse liver exhibits increased hepatocyte proliferation is corrobo- rated by the high levels of E2F found in these animals, in contrast with nondiabetic cohorts. In conclusion, STZ-induced diabetic mouse liver shows increased Pfkfb3 mRNA and protein levels, regulated via transcriptional promoter activation. Moreover, progressively increased concentrations of Fru-2,6-P 2 correlate with gradually overexpressed uPFK-2 protein in liver after STZ injection. UPFK-2- overexpressing hepatocytes are perivenously distributed and exhibit a cluster-forming structure that also sug- gests cellular proliferation status. All of these events correlate with the activation, in diabetic mouse liver, of cyclin D and E2F1, two downstream effectors of the PI3K, Akt and mTOR proliferation pathways. Materials and methods Chemicals Media, sera and antibiotics were obtained from Life Tech- nologies, Inc. (Grand Island, NY, USA). Poly-dI:dC was purchased from Amersham Bioscience Corp. (Piscataway, NJ, USA). STZ and EGF were obtained from Sigma– Aldrich (St Louis, MO, USA). LY 294002, rapamycin, PD 98059 and SB 203580 were purchased from Calbiochem (Darmstadt, Germany). Plasmids pEGFP containing an early cytomegalovirus promoter and an enhanced GFP was purchased from Clontech (Palo Alto, CA, USA). The Pfkfb3 ⁄ -3566 promoter construct cloned into the pGL2-basic vector (Promega, Madison, WI, USA), with the firefly luciferase gene as a reporter, has been described elsewhere [16]. Plasmid DNA was prepared using the Nucleobond MaxiPrep Kit (Macherey-Nagel, Du ¨ ren, Germany), and contained no detectable bacterial genomic DNA or RNA contamination by DNA gel electro- phoresis. Plasmid DNA preparations contained less than 20% open circular or linear DNA. Animal care and treatment Male C57 ⁄ BL6 mice purchased from Harlan Interfarma IBERICA S.L (Barcelona, Spain) were maintained under a constant 12 h light ⁄ dark cycle and fed a standard rodent chow and water ad libitum. All animal protocols were approved by the Ethics Committee at the University of Barcelona. Mice weighing 20–22 g were made diabetic with a 2-day intraperitoneal injection of 100 mgÆkg )1 STZ in 100 mm citric ⁄ citrate buffer, pH 4.5, in fasting conditions. Glycaemia was controlled for 15 days and only mice with fasting blood glucose concentrations above 250 mgÆdL )1 and a plasma insulin concentration below 0.15 lgÆL )1 were used. All animals were killed, at different treatment steps, after 16 h of fasting by cervical dislocation. Livers were dissected, snap frozen in liquid nitrogen and stored at )80 °C until analysis. For in vivo promoter activity studies, the caudate lobe was frozen in liquid nitrogen and stored at )80 °C, or used directly. Cell culture The murine macrophage cell lines RAW 264.7 and RAW IjB aDA were kindly provided by Dr A. Castrillo (De- partamento de Bioquı ´ mica y Biologı ´ a Molecular, Universi- dad de Las Palmas, Gran Canaria, Spain). They were cultured in DMEM (Biological Industries, Kibbutz Beit Haemek, Israel) supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA, USA), l-glutamine and antibiot- ics, and incubated in a humidified atmosphere of 10% CO 2 at 37 °C. Rat hepatocytes were obtained as in Bartrons et al. [29], seeded for 4 h on collagen-coated plates with Williams E medium (BioWhittaker, Cambrex Bio Sci- ence, Verviers, Belgium) supplemented with gentamicin and 10% fetal bovine serum, and cultured with or without fetal bovine serum, EGF and inhibitors. Quantitative real-time PCR RNA was isolated from diabetic and control C57 ⁄ BL6 mice liver, or from rat primary cultured hepatocytes, using the RNeasy Protect Mini Kit (Qiagen, Valencia, CA, USA) fol- lowing the manufacturer’s protocol. The concentration and purity of all RNAs were determined using the A 260 nm ⁄ A 280 nm ratio and by formaldehyde gel electropho- resis. Five micrograms of total RNA were reverse tran- scribed using a Ready-To-Go You Prime First-Strand Beads Kit (GE Healthcare, Piscataway, NJ, USA), employ- ing random primer hexamers. Pfkfb3 was specifically amplified by real-time PCR using the probe ⁄ primer set Mm00504650-m1, and Pfkfb1 and Pfkfb4 using Mm01256238-m1 and Mm 01235506-m1, respectively (Applied Biosystems, Foster City, CA, USA). The relative expression of the gene was normalized to that of 18S RNA (Hs99999901-s1). Gene expression in each sample was then compared with the expression in control mouse liver. In vivo transfection and luciferase assays Transfections were performed using hydrodynamic gene transfer, as described in Go ´ mez-Valade ´ s et al. [30]. GFP plasmid (3 lg of DNA per gram of body weight) and Pfkfb3 ⁄ -3566 promoter construct (30 l g of DNA per mouse) were injected in a 1.7–2 mL volume through the tail vein. Following the hydrodynamic-based transfection proce- Pfkfb3 upregulation in STZ-induced diabetic mouse liver J. Duran et al. 4564 FEBS Journal 276 (2009) 4555–4568 ª 2009 The Authors Journal compilation ª 2009 FEBS [...]... (Clontech, Palo Alto, CA, USA) Pfkfb3 upregulation in STZ-induced diabetic mouse liver Confocal microscopy Livers were fixed in 4% paraformaldehyde for 16 h and maintained in NaCl ⁄ Pi-Sacarose 30% for 48 h Then, they were immersed in OCT (Tissue-Tek, Miles, Inc., CA, USA) and frozen by dry ice Tissues were cut into 7 lm sections using Cryostate GFP was detected using a spectral confocal microscope (Leica TCS-SL,... Over- Pfkfb3 upregulation in STZ-induced diabetic mouse liver 31 32 33 34 35 36 37 38 39 40 41 42 43 44 coming diabetes-induced hyperglycemia through inhibition of hepatic phosphoenolpyruvate carboxykinase (GTP) with RNAi Mol Ther 13, 401–410 Jungermann K (1995) Zonation of metabolism and gene expression in liver Histochem Cell Biol 103, 81–91 Arsura M & Cavin LG (2005) Nuclear factor-kappaB and liver. .. of heart 6-phosphofructo-2-kinase by protein kinase B and other protein kinases of the insulin signaling cascades J Biol Chem 272, 17269–17275 55 Yamamoto M, Tamakawa S, Yoshie M, Yaginuma Y & Ogawa K (2006) Neoplastic hepatocyte growth associated with cyclin D1 redistribution from the cytoplasm to the nucleus in mouse hepatocarcinogenesis Mol Carcinog 45, 901–913 56 Joaquin M, Salvado C, Bellosillo... anti-caspase-3 from BD Biosciences PharMingen (San Diego, CA, USA); and antic-tubulin from Sigma-Aldrich Bound antibody was visualized with horseradish peroxidase-conjugated sheep rabbit IgG or donkey mouse IgG secondary antibodies, and developed by enhanced chemiluminescence using ECL (Amersham Bioscience Corp.) Plasma insulin levels were measured using ultrasensitive Mouse Insulin ELISA (Mercodia AB, Uppsala,... glucose levels were measured using the Glucocard Memory 2 system (Menarini, Florence, Italy) Hepatic injury was evaluated by measuring transaminase levels using commercial kits from Boehringer Mannheim (Munich, Germany) Fru-2,6-P2 was determined following the method described by Van Schaftingen et al [60] ‘Total’ and ‘active’ PFK-2 activities in liver extracts were obtained following the method described... enhancer binding protein and cell replication via PI3-kinase pathway Hepatology 37, 686–695 Wang GL & Timchenko NA (2005) Dephosphorylated C ⁄ EBPalpha accelerates cell proliferation through sequestering retinoblastoma protein Mol Cell Biol 25, 1325–1338 Wullschleger S, Loewith R & Hall MN (2006) TOR signaling in growth and metabolism Cell 124, 471–484 Zdychova J & Komers R (2005) Emerging role of Akt kinase... and fructose 2,6-bisphosphate levels in the liver of diabetic rats J Biol Chem 263, 1868–1871 Miralpeix M, Decaux JF, Kahn A & Bartrons R (1991) Vanadate induction of L-type pyruvate kinase mRNA in adult rat hepatocytes in primary culture Diabetes 40, 462–464 Wu C, Okar DA, Newgard CB & Lange AJ (2001) Overexpression of 6-phosphofructo-2-kinase ⁄ fructose2,6-bisphosphatase in mouse liver lowers blood... lowers blood glucose by suppressing hepatic glucose production J Clin Invest 107, 91–98 Van Schaftingen E, Davies DR & Hers HG (1981) Inactivation of phosphofructokinase 2 by cyclic AMP-dependent protein kinase Biochem Biophys Res Commun 103, 362–368 Bartrons R, Hue L, Van Schaftingen E & Hers HG (1983) Hormonal control of fructose 2,6-bisphosphate concentration in isolated rat hepatocytes Biochem... by mixing equal amounts of complementary single-stranded DNAs in 50 mm NaCl, heating to 70 °C for 15 min, and cooling at room temperature Oligonucleotides containing the consensus binding sequence for NF-jB were purchased from Roche Diagnostics (Basel, Switzerland) (Refs 623227 and 623228) The annealed oligonucleotides were labelled with 32P in the presence of [c-32P]ATP and T4 polynucleotide kinase... factor-kappaB and liver carcinogenesis Cancer Lett 229, 157–169 Diaz-Guerra MJ, Castrillo A, Martin-Sanz P & Bosca L (1999) Negative regulation by phosphatidylinositol 3-kinase of inducible nitric oxide synthase expression in macrophages J Immunol 162, 6184–6190 Chesney J (2006) 6-Phosphofructo-2-kinase ⁄ fructose2,6-bisphosphatase and tumor cell glycolysis Curr Opin Clin Nutr Metab Care 9, 535–539 . that Pfkfb3 gene expression is increased in streptozotocin-induced diabetic mouse liver. The Pfkfb3 ⁄ -3566 promoter construct linked to the luciferase reporter gene was delivered to the liver. This promoter was upregulated in streptozotocin-induced diabetic mouse liver compared with transfected healthy cohorts. In addition, increases were observed in Pfkfb3 mRNA and uPFK-2 protein. immunostaining and hydrodynamic transfection analysis of Pfkfb3 ⁄ -3566 promoter construct. (A) uPFK-2 immuno- staining in control and diabetic mouse livers. Fixed liver samples included in OCT