PGE2 induces apoptosis of hepatic stellate cells and attenuates liver fibrosis in mice by downregulating miR-23a-5p and miR-28a-5p.

MicroRNAs (miRNAs), small noncoding RNAs modulating messenger RNA (mRNA) and protein expression, have emerged as key regulatory molecules in chronic liver diseases, whose end stage is hepatic fibrosis, a major global health burden. Pharmacological strategies for prevention or treatment of hepatic fibrosis are still limited, what makes it necessary to establish a better understanding of the molecular mechanisms underlying its pathogenesis. In this context, we have recently shown that cyclooxygenase-2 (COX-2) expression in hepatocytes restricts activation of hepatic stellate cells (HSCs), a pivotal event in the initiation and progression of hepatic fibrosis. Here, we evaluated the role of COX-2 in the regulation of a specific set of miRNAs on a mouse model of CCl4 and bile duct ligation (BDL)-induced liver fibrosis. Our results provide evidence that COX-2 represses miR-23a-5p and miR-28-5p expression in HSC. The decrease of miR-23a-5p and miR-28-5p expression promotes protection against fibrosis by decreasing the levels of pro-fibrogenic markers α-SMA and COL1A1 and increasing apoptosis of HSC. Moreover, we demonstrate that serum levels of miR-28-5p are decreased in patients with chronic liver disease. These results suggest a protective effect exerted by COX-2-derived prostanoids in the process of hepatofibrogenesis.

Inhibition of IkappaB kinase and IkappaB phosphorylation by 15-deoxy-Delta(12,14)-prostaglandin J(2) in activated murine macrophages.

Activation of the macrophage cell line RAW 264.7 with lipopolysaccharide (LPS) and gamma interferon (IFN-gamma) induces the expression of gene products involved in host defense, among them type 2 nitric oxide synthase. Treatment of cells with 15-deoxy-Delta(12,14)-prostaglandin J(2) (15dPGJ(2)) inhibited the LPS- and IFN-gamma-dependent synthesis of NO, a process that was not antagonized by similar concentrations of prostaglandin J(2), prostaglandin E(2), or rosiglitazone, a peroxisomal proliferator-activated receptor gamma ligand. Incubation of activated macrophages with 15dPGJ(2) inhibited the degradation of IkappaBalpha and IkappaBbeta and increased their levels in the nuclei. NF-kappaB activity, as well as the transcription of NF-kappaB-dependent genes, such as those encoding type 2 nitric oxide synthase and cyclooxygenase 2, was impaired under these conditions. Analysis of the steps leading to IkappaB phosphorylation showed an inhibition of IkappaB kinase by 15dPGJ(2) in cells treated with LPS and IFN-gamma, resulting in an impaired phosphorylation of IkappaBalpha, at least in the serine 32 residue required for targeting and degradation of this protein. Incubation of partially purified activated IkappaB kinase with 2 microM 15dPGJ(2) reduced by 83% the phosphorylation in serine 32 of IkappaBalpha, suggesting that this prostaglandin exerts direct inhibitory effects on the activity of the IkappaB kinase complex. These results show rapid actions of 15dPGJ(2), independent of peroxisomal proliferator receptor gamma activation, in macrophages challenged with low doses of LPS and IFN-gamma.

Relationship between the displacement of phosphatidate phosphohydrolase from the membrane-associated compartment by chlorpromazine and the inhibition of the synthesis of triacylglycerol and phosphatidylcholine in rat hepatocytes.

Glycerolipid synthesis was studied in isolated hepatocytes by using 177 microM [14C]oleate and 1 mM [3H]glycerol. Chlorpromazine (25-400 microM) inhibited the synthesis of phosphatidylcholine and triacylglycerol. This was accompanied by an average increase of 12-fold in the accumulation of the labelled precursors in phosphatidate at 200 microM chlorpromazine and a decrease in the conversion of phosphatidate to diacylglycerol of 76%. These results indicate that part of the inhibition of the synthesis of phosphatidylcholine and triacylglycerol occurs at the level of phosphatidate phosphohydrolase. The relative rate of triacylglycerol synthesis at different concentrations of chlorpromazine was approximately proportional to the rate of conversion of phosphatidate to diacylglycerol. Phosphatidylcholine synthesis increased at higher rates of conversion of phosphatidate to diacylglycerol, but it was relatively independent of the latter rate when this was inhibited by more than about 30% with chlorpromazine. The addition of oleate to the hepatocytes caused a translocation of phosphatidate phosphohydrolase from the cytosol to the membrane-associated compartment. Chlorpromazine had the opposite effect and displaced the phosphohydrolase from the membranes in the presence or absence of oleate. There was a highly significant correlation between the activity of phosphatidate phosphohydrolase that was associated with the membranes of the hepatocytes and the calculated conversion of [3H]phosphatidate to diacylglycerol. Chlorpromazine also antagonized the association of the phosphohydrolase with microsomal membranes when cell-free preparations were incubated with combinations of oleate and spermine. Furthermore, it inhibited the transfer of the soluble phosphohydrolase to microsomal membranes that were labelled with [14C]phosphatidate and thereby decreased diacylglycerol production. It is concluded that part of the action of chlorpromazine in inhibiting the synthesis of triacylglycerol and phosphatidylcholine occurs because it prevents the interaction of the soluble phosphatidate phosphohydrolase with the membranes on which glycerolipid synthesis occurs. This in turn prevents the conversion of phosphatidate to diacylglycerol.

Rat liver messenger ribonucleic acid and enzyme activity of 6-phosphofructo 2-kinase/fructose 2,6-bisphosphatase impairment during the late period of pregnancy.

Fructose 2,6-bisphosphate concentration, 6-phosphofructo 2-kinase/fructose 2,6-bisphosphatase (PFK-2/FBPase2) activity, and messenger RNA decreased in maternal rat liver during the last days of gestation, and the recovery started after delivery. Phospho(enol)pyruvate carboxykinase activity and messenger RNA increased in contrast to PFK-2 changes. Measurement of the glycolytic capacity in isolated hepatocytes prepared from rats 1 h after parturition showed a low glucose consumption and an impaired capacity to metabolize glucose. These results stress the relevance of the PFK-2/fructose 2,6-bisphosphate system in the control of the glycolytic flux in liver, and these changes are intended to prevent glucose consumption by maternal liver and contribute to allow gluconeogenesis to proceed at the end of gestation. The physiological basis of this adaptation may lay on the diversion of glucose from maternal to fetal metabolism.

Fructose 2,6-bisphosphate in isolated foetal hepatocytes.

Fru 2,6-P2 was present in isolated foetal hepatocytes at a concentration of 1.6 nmol per g cells. When foetal hepatocytes were exposed to glucagon no changes were observed either in the concentration of Fru 2,6-P2 and lactate release or in the activities of 6-phosphofructo-2-kinase and pyruvate kinase. Incubation of purified 6-phosphofructo-2-kinase with the catalytic subunit of protein kinase did not change the enzyme activity. The inhibition by sn-glycerol 3-phosphate was much lower for the foetal than for adult enzyme. These results suggest that an isoenzyme of 6-phosphofructo-2-kinase in foetal hepatocytes different from that of adult hepatocytes may be present.

Long-chain fatty acids and their acyl-CoA esters cause the translocation of phosphatidate phosphohydrolase from the cytosolic to the microsomal fraction of rat liver.

A translocation of phosphatidate phosphohydrolase from the cytosolic to the microsomal fraction was promoted in cell-free extracts of rat liver by oleate and palmitate and their CoA esters. Oleate was more potent in this respect than palmitate and the CoA esters were more effective than the unesterified acids. Octanoate, octanoyl-CoA and CoA did not cause the translocation. It is proposed that the interaction of phosphatidate phosphohydrolase with the membranes that synthesize glycerolipids causes it to become metabolically active. This enables the liver to increase its capacity for triacylglycerol synthesis in response to an increased supply of fatty acids.

Spermine promotes the translocation of phosphatidate phosphohydrolase from the cytosol to the microsomal fraction of rat liver and it enhances the effects of oleate in this respect.

Spermine (0.5-2 mM) promoted the translocation of phosphatidate phosphohydrolase from the soluble to the microsomal fraction in a cell-free system derived from rat liver. By contrast, spermidine (1 mM) and putrescine (1 mM) had no significant effect on the translocation when added alone. Spermine, and to a lesser extent, spermidine, enhanced the translocating action of oleate and increased its effectiveness in transferring the phosphohydrolase from the soluble to the microsomal fraction. It is proposed that the phosphohydrolase becomes metabolically active when it combines with membranes and that polyamines might help to regulate this interaction. This could facilitate the action of fatty acids and enable cells to increase their capacity for triacylglycerol synthesis to match an increased availability of fatty acids.

Epidermal growth factor inhibits cytokine-dependent nitric oxide synthase expression in hepatocytes.

Liver cells express a wide range of extracellular receptors involved in the control of cell growth and arrest that can be studied ex vivo. Incubation of primary cultures of hepatocytes with IL-1 beta, TNF-alpha or lipopolysaccharide promotes the expression of the inducible form of nitric oxide synthase and NO release, a process that is inhibited to a different extent by incubation of the cells with EGF. In addition to this growth factor, IL-6 and TGF-beta also inhibited NO synthesis. Therefore, EGF by itself or in combination with other cytokines may be involved in the down-regulation of the NO synthesis that occurs in the early steps of liver regeneration.

Expression of cyclooxygenase-2 in foetal rat hepatocytes stimulated with lipopolysaccharide and pro-inflammatory cytokines.

Cyclooxygenase-2 (COX-2) is involved in the biosynthesis of prostanoids in the course of inflammatory reactions. This isoenzyme is regulated at the transcription level and many cells express COX-2 upon challenge with lipopolysaccharide (LPS) or pro-inflammatory cytokines. Since hepatocytes respond to LPS and pro-inflammatory stimuli, we investigated the expression of COX-2 in foetal and adult hepatocytes upon challenge with these substances. COX-2 was expressed in foetal hepatocytes incubated with LPS, tumour necrosis factor-alpha and interleukin-1beta. This response rapidly decreased after birth and was absent in hepatocytes from animals aged 2 days or more and treated under identical conditions. The expression of COX-2 was determined at the mRNA, protein and enzyme activity levels using Northern and Western blot, and following the synthesis of prostaglandin E2, respectively. The use of NS 398, a specific pharmacological inhibitor of COX-2, confirmed the expression of this isoenzyme in activated foetal hepatocytes. Synergism in COX-2 expression was observed between LPS, tumour necrosis factor-alpha and interleukin-1beta. Interleukin-6 and permeant analogues of cyclic AMP failed to induce COX-2 or to synergize with LPS. Also, transforming growth factor-beta inhibited the LPS- and pro-inflammatory cytokines-dependent expression of COX-2. These results indicate that foetal hepatocytes are competent to express COX-2 upon challenge with pro-inflammatory stimuli, a process lost completely in hepatocytes isolated from animals aged 2 days.

Effects of an antitumoural rhodium complex on thioacetamide-induced liver tumor in rats. Changes in the activities of ornithine decarboxylase, tyrosine aminotransferase and of enzymes involved in fatty acid and glycerolipid synthesis.

Rats were injected daily for 8 weeks with 50 mg of thioacetamide per kg to produce liver tumours. Some of these rats were given three doses of 50 mg of an antitumoural Rh(III) complex/kg at 14, 9 and 5 days before the end of the thioacetamide treatment. Thioacetamide decreased the rate of weight gain of the rats and the Rh(III) complex partly restored it. The activities of ATP citrate lyase, acetyl-CoA carboxylase and fatty acid synthetase in the livers were decreased by thioacetamide treatment and the Rh(III) complex partly reversed this effect. By contrast the activity of malic enzyme was increased by both thioacetamide and the Rh(III) complex and this effect probably relates to NADPH production for detoxification rather than for lipogenesis. Treatment with thioacetamide increased the rate of synthesis of di- and triacylglycerols from glycerol phosphate by liver homogenates, the activity of phosphatidate phosphohydrolase and the incorporation of [3H]glycerol into liver triacylglycerol in vivo. The Rh(III) complex did not produce a significant reversal of these effects of thioacetamide on glycerolipid synthesis. The total uptake of intraportally injected [3H]glycerol by the livers of thioacetamide treated rats was decreased and this was associated with a lowered activity of glycerol kinase. Thioacetamide increased the activity of hepatic ornithine decarboxylase by about 40-fold, but the Rh(III) complex did not reverse this effect. However, the decrease in tyrosine aminotransferase activity that was produced by thioacetamide was partly reversed by the Rh(III) complex. These results are discussed in relation to the tumour-promoting effects of thioacetamide and the antitumoural action of the Rh(III) complex.

Characterization of nitric oxide dependent changes in carbohydrate hepatic metabolism during septic shock.

The role of nitric oxide in the alterations of liver carbohydrate metabolism during septic shock has been studied in fed and starved animals injected with bacterial lipopolysaccharide (LPS). One h after LPS injection an hyperglycemic peak was observed followed by hypoglycemia when the plasma nitric oxide concentration increased. However, in animals pharmacologically treated with nitric oxide donors only hypoglycemia was observed. In isolated hepatocytes from LPS treated rats an impairment of the gluconeogenic flux was observed accompanied by a decrease in the mRNA levels of the glucose transporter GLUT-2 and phosphoenolpyruvate carboxykinase, at the time that increased the mRNA levels of the inducible form of nitric oxide synthase. These results suggest that part of the effects observed in response to LPS challenge are due to early signaling molecules (cytokines and other factors molecules) whereas other effects can be attributed to nitric oxide synthesis which in turn has specific effects on hepatic metabolism.

Translocation of phosphatidate phosphohydrolase from the cytosol to microsomal membranes in thioacetamide-induced liver tumours in rats.

The translocation of phosphatidate phosphohydrolase induced by oleate was higher (two-fold) in liver homogenates obtained from long-term thioacetamide-treated rats than from control rats. These differences between thioacetamide-treated and control livers were noticeably higher (four-fold) in the presence of physiological concentrations of salt (0.15 M KCl). In homogenates from control rats, there was a lack of response when physiological concentrations of the salt were present. The enhanced response to translocate phosphatidate phosphohydrolase activity in liver homogenates from thioacetamide-treated rats was due to an increased binding ability of microsomal membranes.

Effect of rhodium (III) complex on experimental carcinogenesis in the livers of rats treated with thioacetamide.

Enzyme activities and protein content were determined in the cytosolic and mitochondrial fractions of liver homogenates obtained from Rh(III) complex-, thioacetamide- and thioacetamide + Rh(III) complex-treated rats.The Rh(III) complex administered to nonthioacetamide-treated rats produced no significant changes either in the enzymatic activities assayed or in the protein concentration.The Rh(III) complex administered to thioacetamide-treated rats produced significant restoration of the following altered values: cytosolic and mitochondrial aspartate aminotransferase, glutamate dehydrogenase, NADP-isocitrate dehydrogenase, and protein concentration. However, a further increase was produced in the activities of glucose-6-phosphate dehydrogenase and malic enzyme. These increases can be interpreted in terms of an enhancement of the NADPH-dependent detoxifying processes and of nucleic acid synthesis and repair.

Caveolin-1 is required for TGF-ß-induced transactivation of the EGF receptor pathway in hepatocytes through the activation of the metalloprotease TACE/ADAM17.

Transforming growth factor-beta (TGF-ß) plays a dual role in hepatocytes, inducing both pro- and anti-apoptotic responses, whose balance decides cell fate. Survival signals are mediated by the epidermal growth factor receptor (EGFR) pathway, which is activated by TGF-ß in these cells. Caveolin-1 (Cav1) is a structural protein of caveolae linked to TGF-ß receptors trafficking and signaling. Previous results have indicated that in hepatocytes, Cav1 is required for TGF-ß-induced anti-apoptotic signals, but the molecular mechanism is not fully understood yet. In this work, we show that immortalized Cav1(-/-) hepatocytes were more sensitive to the pro-apoptotic effects induced by TGF-ß, showing a higher activation of caspase-3, higher decrease in cell viability and prolonged increase through time of intracellular reactive oxygen species (ROS). These results were coincident with attenuation of TGF-ß-induced survival signals in Cav1(-/-) hepatocytes, such as AKT and ERK1/2 phosphorylation and NFκ-B activation. Transactivation of the EGFR pathway by TGF-ß was impaired in Cav1(-/-) hepatocytes, which correlated with lack of activation of TACE/ADAM17, the metalloprotease responsible for the shedding of EGFR ligands. Reconstitution of Cav1 in Cav1(-/-) hepatocytes rescued wild-type phenotype features, both in terms of EGFR transactivation and TACE/ADAM17 activation. TACE/ADAM17 was localized in detergent-resistant membrane (DRM) fractions in Cav1(+/+) cells, which was not the case in Cav1(-/-) cells. Disorganization of lipid rafts after treatment with cholesterol-binding agents caused loss of TACE/ADAM17 activation after TGF-ß treatment. In conclusion, in hepatocytes, Cav1 is required for TGF-ß-mediated activation of the metalloprotease TACE/ADAM17 that is responsible for shedding of EGFR ligands and activation of the EGFR pathway, which counteracts the TGF-ß pro-apoptotic effects. Therefore, Cav1 contributes to the pro-tumorigenic effects of TGF-ß in liver cancer cells.

Pivotal role of protein tyrosine phosphatase 1B (PTP1B) in the macrophage response to pro-inflammatory and anti-inflammatory challenge.

Inhibition of protein tyrosine phosphatase 1B (PTP1B) has been suggested as an attractive target to improve insulin sensitivity in different cell types. In the present work, we have investigated the effect of PTP1B deficiency on the response of human and murine macrophages. Using in vitro and in vivo approaches in mice and silencing PTP1B in human macrophages with specific siRNAs, we have demonstrated that PTP1B deficiency increases the effects of pro-inflammatory stimuli in both human and rodent macrophages at the time that decreases the response to alternative stimulation. Moreover, the absence of PTP1B induces a loss of viability in resting macrophages and mainly after activation through the classic pathway. Analysis of early gene expression in macrophages treated with pro-inflammatory stimuli confirmed this exacerbated inflammatory response in PTP1B-deficient macrophages. Microarray analysis in samples from wild-type and PTP1B-deficient macrophages obtained after 24 h of pro-inflammatory stimulation showed an activation of the p53 pathway, including the excision base repair pathway and the insulin signaling pathway in the absence of PTP1B. In animal models of lipopolysaccharide (LPS) and D-galactosamine challenge as a way to reveal in vivo inflammatory responses, animals lacking PTP1B exhibited a higher rate of death. Moreover, these animals showed an enhanced response to irradiation, in agreement with the data obtained in the microarray analysis. In summary, these results indicate that, although inhibition of PTP1B has potential benefits for the treatment of diabetes, it accentuates pro-inflammatory responses compromising at least macrophage viability.

Impaired autophagic flux is associated with increased endoplasmic reticulum stress during the development of NAFLD.

The pathogenic mechanisms underlying the progression of non-alcoholic fatty liver disease (NAFLD) are not fully understood. In this study, we aimed to assess the relationship between endoplasmic reticulum (ER) stress and autophagy in human and mouse hepatocytes during NAFLD. ER stress and autophagy markers were analyzed in livers from patients with biopsy-proven non-alcoholic steatosis (NAS) or non-alcoholic steatohepatitis (NASH) compared with livers from subjects with histologically normal liver, in livers from mice fed with chow diet (CHD) compared with mice fed with high fat diet (HFD) or methionine-choline-deficient (MCD) diet and in primary and Huh7 human hepatocytes loaded with palmitic acid (PA). In NASH patients, significant increases in hepatic messenger RNA levels of markers of ER stress (activating transcription factor 4 (ATF4), glucose-regulated protein 78 (GRP78) and C/EBP homologous protein (CHOP)) and autophagy (BCN1) were found compared with NAS patients. Likewise, protein levels of GRP78, CHOP and p62/SQSTM1 (p62) autophagic substrate were significantly elevated in NASH compared with NAS patients. In livers from mice fed with HFD or MCD, ER stress-mediated signaling was parallel to the blockade of the autophagic flux assessed by increases in p62, microtubule-associated protein 2 light chain 3 (LC3-II)/LC3-I ratio and accumulation of autophagosomes compared with CHD fed mice. In Huh7 hepatic cells, treatment with PA for 8 h triggered activation of both unfolding protein response and the autophagic flux. Conversely, prolonged treatment with PA (24 h) induced ER stress and cell death together with a blockade of the autophagic flux. Under these conditions, cotreatment with rapamycin or CHOP silencing ameliorated these effects and decreased apoptosis. Our results demonstrated that the autophagic flux is impaired in the liver from both NAFLD patients and murine models of NAFLD, as well as in lipid-overloaded human hepatocytes, and it could be due to elevated ER stress leading to apoptosis. Consequently, therapies aimed to restore the autophagic flux might attenuate or prevent the progression of NAFLD.

Evaluation of epigenetic modulation of cyclooxygenase-2 as a prognostic marker for hepatocellular carcinoma.

Cyclooxygenases (COX-1 and 2) catalyze the first step in prostanoid biosynthesis. They are implicated in homeostatic processes with an important role in inflammation and carcinogenesis. In the liver, COX-2 expression is restricted to proliferation or dedifferentiation situations. The COX-2 promoter contains numerous CpG islands that, when hypermethylated, result in transcriptionally silencing thus regulating the growth of carcinoma cells. In this work, we investigated whether a correlation exists between COX-2 expression and methylation signatures at the 5'region of the gene in hepatoma cell lines and human hepatocellular carcinoma (HCC). We also examined the acetylation status of the COX-2 promoter and the effects of histone deacetylase (HDAC) inhibitors on COX-2 expression. Our results suggest a significant association between reduced COX-2 expression and promoter hypermethylation of COX-2 and histone deacetylation in some hepatoma cell lines and in HCC. Treatment with demethylating agents or HDAC inhibitors restored the expression of COX-2. Moreover, in an HCC cohort, a statistically significant inverse association was observed between COX-2 mRNA levels and promoter methylation. In agreement with these data, a reduction of overall survival of the patients was observed after decreased COX-2 expression by promoter hypermethylation and histone H3 hypoacetylation.

Lipoxin A4 impairment of apoptotic signaling in macrophages: implication of the PI3K/Akt and the ERK/Nrf-2 defense pathways.

Lipoxin A(4) (LXA(4)) is an endogenous lipid mediator that requires transcellular metabolic traffic for its synthesis. The targets of LXA(4) on neutrophils are well described, contributing to attenuation of inflammation. However, effects of lipoxins on macrophage are less known, particularly the action of LXA(4) on the regulation of apoptosis of these cells. Our data show that pretreatment of human or murine macrophages with LXA(4) at the concentrations prevailing in the course of resolution of inflammation (nanomolar range) significantly inhibits the apoptosis induced by staurosporine, etoposide and S-nitrosoglutathione or by more pathophysiological stimuli, such as LPS/IFNgamma challenge. The release of mitochondrial mediators of apoptosis and the activation of caspases was abrogated in the presence of LXA(4). In addition to this, the synthesis of reactive oxygen species induced by staurosporine was attenuated and antiapoptotic proteins of the Bcl-2 family accumulated in the presence of lipoxin. Analysis of the targets of LXA(4) identified an early activation of the PI3K/Akt and ERK/Nrf-2 pathways, which was required for the observation of the antiapoptotic effects of LXA(4). These data suggest that the LXA(4), released after the recruitment of neutrophils to sites of inflammation, exerts a protective effect on macrophage viability that might contribute to a better resolution of inflammation.

Characterization of 6-phosphofructo-2-kinase from foetal-rat liver.

Foetal and adult liver 6-phosphofructo-2-kinase (PFK-2) were purified by identical protocols. The native molecular masses of both enzymes were determined by gel filtration and were 89.1 and 100.0 kDa respectively. No differences were found in SDS/PAGE in 10%-acrylamide gel (55 kDa per subunit). The kinetic properties displayed by both enzymes were similar, except for the sensitivity to inhibition by sn-glycerol 3-phosphate. Foetal PFK-2 was a good substrate for phosphorylation by cyclic AMP-dependent protein kinase and protein kinase C, whereas the adult enzyme was phosphorylated only by cyclic AMP-dependent protein kinase. However, the phosphorylation affected only the kinetic properties of the adult enzyme, suggesting the presence in both enzymes of different sites of phosphorylation by cyclic AMP-dependent protein kinase. These differences in primary structure were consistent with the distinct chromatographic profiles of the phosphopeptides after digestion of the protein with CNBr. Western-blot analysis with antibodies specific for the N-terminal region of the liver-type PFK-2 poorly recognized the foetal enzyme, suggesting that both enzymes differ at least in the N-terminal sequence.