Expression of TWEAK and its receptor Fn14 in human subcutaneous adipose tissue. Relationship with other inflammatory cytokines in obesity

TWEAK, a cytokine of the TNF family, has been found to be expressed under different inflammatory conditions but no data is available concerning the expression of this cytokine and its receptor (Fn14) in human obesity. In the present work we have evaluated the expression of many pro-inflammatory TNF system cytokines (TNF-alpha, TWEAK and their respective receptors, TNFR1, TNFR2 and Fn14) in human adipose tissue of 84 subjects some with different degree of obesity and type 2 diabetes, and its relation with inflammation by also measuring the expression of macrophage marker CD68. We detected expression of TWEAK and Fn14 in isolated mature adipocytes and in the stromovascular fraction. Additionally, we found that LPS upregulates the expression of both genes on THP-1 human monocytic cell line. TWEAK was expressed in adipose tissue of all studied subjects with no differences between obesity group, and was associated with Fn14 expression in morbid obese, mainly in women with type 2 diabetes. The data obtained here also showed that TNF-alpha and TNFR2 mRNAs were significantly more expressed in subcutaneous adipose tissue of subjects with morbid obesity compared to obese and non-obese subjects. In contrast, TNFR1 gene expression was negatively associated with BMI. Our results suggest that the expression of TNF-derived pro-inflammatory cytokines are increased in severe obesity, where macrophage infiltrate could modulate the inflammatory environment through activation of its receptors.

Glycosylated phosducin-like protein long regulates opioid receptor function in mouse brain

Phosducin (Phd), a protein that in retina regulates rhodopsin desensitization by controlling the activity of Gt beta gamma-dependent G-protein-coupled receptor kinases (GRKs), is present in very low levels in the CNS of mammals. However, this tissue contains proteins of related sequence and function. This paper reports the presence of N-glycosylated phosducin-like protein long (PhLP(L)) in all structures of mouse CNS, mainly in synaptic plasma membranes and associated with G beta subunits and 14-3-3 proteins. To analyze the role PhLP(L) in opioid receptor desensitization, its expression was reduced by the use of antisense oligodeoxynucleotides (ODNs). The antinociception induced by morphine, [D-Ala(2), N-MePhe(4),Gly-ol(5)]-enkephalin (DAMGO), beta-endorphin, [D-Ala(2)]deltorphin II, [D-Pen(2,5)]-enkephalin (DPDPE) or clonidine in the tail-flick test was reduced in PhLP(L)-knock-down mice. A single intracerebroventricular (icv)-ED(80) analgesic dose of morphine gave rise to acute tolerance that lasted for 4 days, but which was prevented or reversed by icv-injection of myristoylated (myr(+)) G(i2)alpha subunits. PhLP(L) knock-down brought about a myr(+)-G(i2)alpha subunit-insensitive acute tolerance to morphine that was still present after 8 days. It also diminished the specific binding of (125)I-Tyr(27)-beta-endorphin-(1-31) (human) to mouse periaqueductal gray matter membranes. After being exposed to chronic morphine treatment, post-dependent mice required about 10 days for complete recovery of morphine antinociception. The impairment of PhLP(L) extended this period beyond 17 days. It is concluded that PhLP(L) knock-down facilitates desensitization and uncoupling of opioid receptors.

Transport of CSF antibodies to Galpha subunits across neural membranes requires binding to the target protein and protein kinase C activity

In the light of functional studies, it has been suggested that antibodies directed to alpha subunits of G-proteins delivered into cerebrospinal fluid (CSF) reached and blocked the function of neural transducer proteins. Current understanding indicates that IgGs do not move freely across plasma membranes. Therefore, to characterize the uptake of these antibodies by neural cells, anti-Gi2alpha IgGs were labeled with 125I, fluorescein or with gold particles. The expression of Galpha subunits was also reduced by blocking their mRNA with antisense oligodeoxynucleotides (ODN). Following intracerebroventricular (icv) injection of gold-conjugated anti-Gi2alpha IgGs, electrondense particles entered and became distributed in the cytoplasm and plasma membranes of neural cells. Scattered particles were also found in dendrites and nuclei. Unlabeled IgGs diminished cerebral signals of fluorescein-labeled anti-Galpha IgGs, indicating that this uptake can be saturated. Cerebral radiostaining promoted by in vivo anti-Gi2alpha 125I-IgGs was almost absent in Gi2alpha knocked-down mice, but not after decreasing the quantity of Gzalpha subunits. The immunosignals of CSF anti-Galpha 125I-IgGs, as well as the impairment of opioid-evoked antinociception, were increased by agonist-induced activation of G protein-coupled receptors. The impairing effect of the antibodies on opioid-evoked antinociception was prevented by agents blocking the cellular uptake of proteins, i.e., cytochalasin B, BSA, DMSO, H7, and by down regulation of protein kinase Cbeta1 (PKCbeta1). In mice treated with an ODN to PKCbeta1 mRNA, 125I-IgGs to Gi2alpha subunits remained bound to periventricular structures and did not spread to deeper areas of the CNS. These results indicate that IgGs delivered into the CSF show a saturable binding to Galpha subunits that translocate to the external side of the neural membrane before being internalized by a PKCbeta1-dependent mechanism.

Opioids binding mu and delta receptors exhibit diverse efficacy in the activation of Gi2 and G(x/z) transducer proteins in mouse periaqueductal gray matter

A nonisotopic, immunoelectrophoretic technique was used to analyze the characteristics of opioid-evoked activation of Gi2/ G(x/z) transducer proteins of mouse periaqueductal gray matter membranes. In the presence of picomolar concentrations of guanosine 5'-O-(3-thiotriphosphate), the opioid agonists promoted concentration-dependent increases of immunoreactivity associated with free Gi2alpha and G(x/z)alpha subunits. [D-Ala2,N-MePhe4, Gly-ol5]enkephalin and morphine (preferential agonists at mu opioid receptors) and beta-endorphin-(1-31) (an agonist at mu/delta opioid receptors) activated G(x/z) proteins. In contrast, the agonists of delta opioid receptors, [D-Ala2]deltorphin II and [D-Pen(2,5)]enkephalin, displayed little or no activity on this pertussis toxin resistant regulatory protein. Although exhibiting diverse efficacy, all the opioids studied activated Gi2 transducer proteins. [D-Ala2,N-MePhe4,Gly-ol5]enkephalin and [D-Ala2]-deltorphin II were more potent at Gi2alpha subunits than at G(x/z)alpha subunits. The opioid antagonist naloxone displayed a competitive profile in reducing the activation of G proteins promoted by morphine. Moreover, [D-Pen(2,5)]enkephalin antagonized the releasing effect exerted by [D-Ala2]deltorphin II on Gi2alpha and G(x/z)alpha subunits. N,N-diallyl-Tyr-Aib-Aib-Phe-Leu (ICI-174864) reduced the G alpha-related immunosignals promoted by agonists of delta opioid receptors. Therefore, it is suggested that opioids exhibit marked differences in efficacy and/or potency in the activation of Gi2 and G(x/z) transducer proteins in mouse periaqueductal gray matter.

In vivo injection of antisense oligodeoxynucleotides to G alpha subunits and supraspinal analgesia evoked by mu and delta opioid agonists

For 5 consecutive days repeated intracerebroventricular (i.c.v.) administration of antisense oligodeoxynucleotides (ODNs) to G alpha subunit mRNAs was used to impair the function of mouse Gi1, Gi2, Gi3 and Gx/z regulatory proteins. Decreases of 20 to 60% on the G alpha-like immunoreactivity could be observed in neural structures of mouse brain, an effect that was not produced by a random-sequence ODN used as a control. The ODN to Gi1 alpha subunits lacked effect on opioid-evoked analgesia. In mice injected with the ODN to Gi2 alpha subunits the antinociceptive activity of all the opioids studied appeared greatly impaired. The ODN to Gi3 alpha subunits reduced the effects of the selective agonists of delta opioid receptors, [D-Pen2,5]-enkephalin and [D-Ala2]deltorphin II. Conversely, the analgesia evoked by opioids binding mu opioid receptors, [D-Ala2, N-MePhe4,Gly-ol5]enkephalin and morphine, appeared consistently and significantly attenuated in mice injected with the ODN to Gx/z alpha. The effect of the neuropeptide beta-endorphine-(1-31) agonist at mu and delta receptors was also reduced by ODNs to Gi3 alpha or Gx/z alpha subunits. l.c.v. injection of antibodies directed to these G alpha subunits antagonized opioid-induced analgesia with a pattern similar to that observed for the ODNs. Thus, the mu and delta opiod receptors regulate different classes of G transducer proteins to mediate the analgesic effect of agonists. The in vivo antisense strategy and the use of specific antibodies to G alpha subunits gave comparable results, indicating that in the neural tissue the mRNAs and the G alpha subunits can be accessed by the corresponding ODNs and IgGs.

Oxidative inactivation of carbamoyl phosphate synthetase (ammonia). Mechanism and sites of oxidation, degradation of the oxidized enzyme, and inactivation by glycerol, EDTA, and thiol protecting agents

Acetylglutamate and ATP accelerate the oxidative inactivation of carbamoyl phosphate synthetase I by mixtures of Fe3+, ascorbate, and O2, but the mechanism of the inactivation differs with each ligand. In the presence of acetylglutamate, MgATP prevents, Mg2+, Mn2+, and catalase have no effect, and EDTA increases the inactivation, and the two phosphorylation steps of the enzyme reaction are lost simultaneously. The inactivation appears to be mediated by dehydroascorbate and is associated with the reversible oxidation of the highly reactive cysteines 1327 and 1337 and with oxidation of non-thiolic groups in the second 40-kDa domain (the enzyme consists of 4 domains of 40, 40, 60, and 20 kDa, from the amino terminus). The data are consistent with oxidation of groups at or near the site for ATPA (ATPA yields Pi; ATPB yields carbamoyl phosphate), and with the location of this site at the interphase between the second 40-kDa and the COOH-terminal domains. The oxidative inactivation promoted by ATP is inhibited by Mg2+, Mn2+, catalase, and EDTA, is not mediated by dehydroascorbate, and is not associated with oxidation of cysteines 1327 and 1337. Groups in the 60-kDa domain are oxidized. The phosphorylation step involving ATPB is lost preferentially, and the inactivation and the binding of ATPB exhibit the same dependency on the concentration of ATP. The results indicate that the oxidation is catalyzed by FeATP bound at the site for ATPB and support the binding of ATPB in the 60-kDa domain. We also demonstrate that mercaptoethanol, reducing impurities in glycerol, and dithioerythritol, in the presence of EDTA, replace ascorbate in the oxidative system. In addition, we study the influence of the oxidation on the degradation of the enzyme by rat liver lysosomes, mitochondria, and cytosol.

Influence of anions on the activation of carbamoyl phosphate synthetase (ammonia) by acetylglutamate: implications for the activation of the enzyme in the mitochondria

Rat liver carbamoyl phosphate synthetase is shown to be inhibited by anions competitively with acetylglutamate (the allosteric activator of the enzyme) with a potency decreasing in the order NO3- greater than SO4(2-) greater than Cl- approximately HCO3-. Inhibition by chloride accounts for most of the inhibition reported [Lund, P., and Wiggins, D. (1987) Biochem. J. 243, 273-276] in Tris buffer. Mes, acetate, and isethionate give little or no inhibition and phosphate inhibits noncompetitively. Plots of the KA value for acetylglutamate versus the concentration of chloride or nitrate are curved upward and binding assays demonstrate that the inhibitory anions displace acetylglutamate from the enzyme. Thus, the anions may compete with the carboxyls of acetylglutamate for positive charges at the binding site. Of the organic anions found in the mitochondrial matrix, alpha-ketoglutarate, malate, succinate, and citrate increase substantially the KA for acetylglutamate. Changes in the concentrations of ATP, HCO3-, NH4+, and Mg2+, and high concentrations of protein (60 mg/ml serum albumin) influence the KA value. Changes in the concentration of the enzyme have no effect. Under assay conditions approaching the ionic, buffer, and substrate concentrations expected to occur in the mitochondrial matrix, the KA value for acetylglutamate is 27 microM and the Vmax is decreased about 50%. These results indicate that physiological changes in the level of acetylglutamate significantly influence the degree of activation of carbamoyl phosphate synthetase in vivo.

A structure-reactivity study of the binding of acetylglutamate to carbamoyl phosphate synthetase I

The requirements for binding at the N-acetyl-L-glutamate binding site of carbamoyl phosphate synthetase I were studied by the displacement of the activator from the central enzyme complex by analogs. Two carboxyls are essential and the acetamido group, if linked to the alpha-carbon, enhances binding 5000-fold. The subsite for the delta-carboxyl is mobile with respect to that for the alpha-carboxyl. Mixtures of complementary fragment of acetylglutamate do not bind, indicating a strong 'chelate' effect. Substituents revealed the existence of steric constraints around the delta-carboxyl, the alpha and gamma-carbons, and the whole of the acetamido group. However, phenyl substituents at the beta-carbon did not hamper binding, indicating that substituents at the beta-carbon face the solution. This is consistent with binding of acetylglutamate as the minimum-energy conformer. All analogs binding with high affinity are activators. Some analogs that bind poorly are competitive inhibitors. They appear to bind preferentially to a low-affinity conformation adopted by the site when the products dissociate and the substrates bind. The acetamido group plays no role in the binding of the inhibitors but it is crucial for the binding of the activators, and the high- and low-affinity conformations of the site differ markedly in structural selectivity.

Changes in urea cycle-related metabolites in the mouse after combined administration of valproic acid and an amino acid load

Increased blood ammonia was induced in fasting mice by ip administration of 200 mg/kg Na-valproate followed 1 h later by 13 and 4 mmol/kg alanine and ornithine, respectively. When valproate was not used blood or liver ammonia was not increased, but increases were observed in liver glutamate (5-fold), glutamine (2-fold), aspartate (5-fold), acetylglutamate (15-fold), citrulline (35-fold), argininosuccinate (11-fold), arginine (11-fold), and urea (3-fold). The level of carbamoyl phosphate (less than 2 nmol/g) was, by far, the lowest of all urea cycle intermediates. The large increase in citrulline indicates that argininosuccinate synthesis was limiting, and that the increase in acetylglutamate induced a considerable activation of carbamoyl phosphate synthetase, which agrees with theoretical expectations, irrespective of the actual KD value for acetylglutamate. Pretreatment with valproate resulted in lower hepatic levels of glutamate, glutamine, aspartate, acetyl-CoA, and acetylglutamate. At the level found of acetylglutamate the activation of carbamoyl phosphate synthetase would be expected to be similar to that without valproate. Indeed, the levels of citrulline were similar with or without valproate. Argininosuccinate, arginine, and urea levels exhibited little if any change. Although the model used may not replicate exactly the situation in patients, from our results it appears that changes in citrullinogenesis or in other steps of the urea cycle do not account for the increase in blood ammonia induced by valproate, and it is proposed that valproate may alter glutamine metabolism.