Tumor necrosis factor alpha and glutathione interplay in chronic heart failure.

The pro-inflammatory cytokine, tumor necrosis factor alpha (TNF alpha), in concert with neurohormones, contributes to chronic heart failure (CHF) progression. This implies that TNF a antagonism may constitute an important target for CHF therapy. However, clinical trials in CHF patients using compounds that trap TNF alpha, comprising infliximab, an antibody directed to TNF alpha, and etanercept, a soluble recombinant receptor of TNF alpha, gave disappointing results bringing back to light the dual, short-term beneficial and long-term harmful effect of TNF alpha. This review focuses on the dual, concentration- and time-related effects of TNF alpha, the yin and yang action of TNF alpha in cardiac ischemia/reperfusion and contraction. Importantly, the harmful effects of TNF a are related to glutathione deficiency, a common hallmark to several other chronic inflammatory diseases. Recently, in rat models of CHF, oral administration of the glutathione precursor, N-acetylcysteine (NAC), was shown to hinder pathways of TNF alpha harmful signalling and to rescue cardiac structure and function. These results suggest that glutathione deficiency in association with TNF alpha activation may play a role in the pathophysiology of CHF and that NAC may represent a potential therapy in CHF.

Differential alterations in cardiac adrenergic signaling in chronic hypoxia or norepinephrine infusion.

Norepinephrine (NE)-induced desensitization of the adrenergic receptor pathway may mimic the effects of hypoxia on cardiac adrenoceptors. The mechanisms involved in this desensitization were evaluated in male Wistar rats kept in a hypobaric chamber (380 Torr) and in rats infused with NE (0.3 mg. kg(-1). h(-1)) for 21 days. Because NE treatment resulted in left ventricular (LV) hypertrophy, whereas hypoxia resulted in right (RV) hypertrophy, the selective hypertrophic response of hypoxia and NE was also evaluated. In hypoxia, alpha(1)-adrenergic receptors (AR) density increased by 35%, only in the LV. In NE, alpha(1)-AR density decreased by 43% in the RV. Both hypoxia and NE decreased beta-AR density. No difference was found in receptor apparent affinity. Stimulated maximal activity of adenylate cyclase decreased in both ventricles with hypoxia (LV, 41%; RV, 36%) but only in LV with NE infusion (42%). The functional activities of G(i) and G(s) proteins in cardiac membranes were assessed by incubation with pertussis toxin (PT) and cholera toxin (CT). PT had an important effect in abolishing the decrease in isoproterenol-induced stimulation of adenylate cyclase in hypoxia; however, pretreatment of the NE ventricle cells with PT failed to restore this stimulation. Although CT attenuates the basal activity of adenylate cyclase in the RV and the isoproterenol-stimulated activity in the LV, pretreatment of NE or hypoxic cardiac membranes with CT has a less clear effect on the adenylate cyclase pathway. The present study has demonstrated that 1) NE does not mimic the effects of hypoxia at the cellular level, i.e., hypoxia has specific effects on cardiac adrenergic signaling, and 2) changes in alpha- and beta-adrenergic pathways are chamber specific and may depend on the type of stimulation (hypoxia or adrenergic).

Specific activation of the mu opioid receptor (MOR) by endomorphin 1 and endomorphin 2.

The recently discovered endomorphin 1 (Tyr-Pro-Trp-Phe-NH2) and endomorphin 2 (Tyr-Pro-Phe-Phe-NH2) were investigated with respect to their direct receptor-binding properties, and to their ability to activate G proteins and to inhibit adenylyl cyclase in both cellular and animal models. Both tetrapeptides activated G proteins and inhibited adenylyl cyclase activity in membrane preparations from cells stably expressing the mu opioid receptor, an effect reversed by the mu receptor antagonist CTAP (D-Phe-Cys-Tyr-D-Trp-Arg-Thr-Pen-Thr-NH2), but they had no influence on cells stably expressing the delta opioid receptor. To further establish the selectivity of these peptides for the mu opioid receptor, brain preparations of mice lacking the mu opioid receptor gene were used to study their binding and signalling properties. Endomorphin 2, tritiated by a dehalotritiation method resulting in a specific radioactivity of 1.98 TBq/mmol (53.4 Ci/mmol), labelled the brain membranes of wild-type mice with a Kd value of 1.77 nM and a Bmax of 63.33 fmol/mg protein. In membranes of mice lacking the mu receptor gene, no binding was observed, and both endomorphins failed to stimulate [35S]guanosine-5'-O-(3-thio)triphosphate ([35S]GTPgammaS) binding and to inhibit adenylyl cyclase. These data show that endomorphins are capable of activating G proteins and inhibiting adenylyl cyclase activity, and all these effects are mediated by the mu opioid receptors.

Decreased type VI adenylyl cyclase mRNA concentration and Mg(2+)-dependent adenylyl cyclase activities and unchanged type V adenylyl cyclase mRNA concentration and Mn(2+)-dependent adenylyl cyclase activities in the left ventricle of rats with myocardial infarction and longstanding heart failure.

OBJECTIVE: To address the effect of longstanding left ventricular (LV) hypertrophy and failure on LV adenylyl cyclase (AC) gene expression, mRNA concentrations of the main cardiac AC isoforms were measured in the non-infarcted area of LV from rats with myocardial infarction (MI), without (H) or with (F) LV failure, and in control (C) rats. Basal, GTP- and forskolin-stimulated Mg(2+)- and Mn(2+)-dependent AC activities were also measured in F and C rats. METHODS: Two- and six months after MI, steady-state AC mRNA concentrations were assessed by Northern blot analysis and RNase protection assay with isoform-specific cDNA and cRNA probes, respectively. AC activities were assessed on LV microsomal fractions using standard procedures. RESULTS: Types V and VI, and types IV and VII were the major and minor AC mRNA isoforms in both the LVs of F and C rats. Two months after MI, no difference in LV type V or VI mRNA to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA ratios was observed in rats with H or F compared to C. Six months after MI, no difference in LV type V mRNA concentration was observed between the three rat groups, whether this level was normalized to GAPDH, poly-(A+) or 18S RNAs. In contrast, a 35% decrease in the type VI mRNA to poly-(A+) RNA ratio and a 29% decrease in the type VI mRNA to 18S RNA ratio was observed only in rats with F compared to C (p < 0.05 vs. C for the two comparisons). Two- and six months after MI, basal and forskolin-stimulated Mg(2+)-dependent AC activities were decreased by 30-35% in F rats compared to C (p < 0.05), whereas Mn(2+)-dependent activities were unchanged. CONCLUSION: Longstanding LV hypertrophy and failure resulting from MI in rats is not associated with altered expression of the most abundant, type V, AC mRNA isoform, whereas that of type VI is decreased. The lack of change in Mn(2+)-dependent AC activities in the LV of F rats suggests that this decrease has no functional consequence on overall AC activity and that decreased Mg(2+)-dependent activities are related to alterations occurring upstream.

Lack of acidic fibroblast growth factor activation by heparan sulfate species from diabetic rat skin.

The glucosaminoglycans isolated from the skin of control and streptozotocin-diabetic rats were fractionated on ion-exchange chromatography into a heparan sulfate (HS)-like and a heparin-like species. In addition, a low sulfated fraction was isolated from the diabetics. The HS and heparin-like fractions isolated from the diabetics (in contrast to the low sulfated fractions) retained high affinity for the acidic (FGF-1) and basic (FGF-2) fibroblast growth factors. In culture, the fractions purified from the control rats and the heparin-like material isolated from the diabetics mediated the biological activity of both FGFs in a dose-dependent manner. By contrast, the diabetic HS-like fractions promoted the biological activity of FGF-2 but not of FGF-1. The results support the idea that the structural motives in HS required for FGF-1 and FGF-2 mediated receptor signalling are different. They may be relevant to the impaired wound healing observed in the disease.

Mesoglycan and sulodexide act as stabilizers and protectors of fibroblast growth factors (FGFs).

Heparin and heparan sulfate proteoglycans (HSPGs) stabilize FGFs which belong to heparin-binding growth factors (HBGFs) on active conformation. They also strongly potentiate their mitogenic activity on many cell types, and protect them against thermal denaturation and enzymatic degradation. In the present work we have tested two heparin-like substances named mesoglycan and sulodexide obtained from bovine intestinal mucosal extracts. These products are used as heparin, in various of therapeutic fields such as atherosclerosis or antithrombotic therapy. The compositions of mesoglycan and sulodexide are partially known and include chondroitin, dermatan and heparan sulfate. We have shown that mesoglycan and sulodexide potentiated the mitogenic activity of FGF1 and FGF2. The magnitude of this effect was identical with that of heparin used as a control substance but at double concentration. Mesoglycan and sulodexide also exerted stabilizing and protective effects on FGFs for heat denaturation and enzymatic degradation. The suppression of the protective properties after heparinase treatment of mesoglycan and sulodexide indirectly demonstrated the presence of heparan sulfate which was shown to represent about 60% of the commercial products.

[Thrombomodulin: a new proteoglycan. Structure-function relation]. / La thrombomoduline: un nouveau protéoglycanne. Relation structure-fonction.

The endothelial cell surface receptor thrombomodulin (TM) displays various anticoagulant functions: it acts as a cofactor for the activation of protein C (PC) by thrombin, prevents the activation of fibrinogen, platelets and Factor V by thrombin. TM was also shown to accelerate the inhibition of thrombin by its physiological inhibitor antithrombin III (ATIII). The studies performed on rabbit lung TM were undertaken in order to provide better understanding, along with the identification and the characterization of functional domains, to the mechanism of action of TM. On the basis of the physical and chemical properties of TM, which were compatible with those of a proteoglycan, the presence of a sulfated polysaccharide chain covalently bound to TM, constituting an acidic domain independent of the protein C activation cofactor site, was suggested. Further enzymatic and chemical characterization showed that rabbit TM was in fact a chondroitin sulfate proteoglycan. Monoclonal antibodies raised against rabbit TM and proteins known for their ability to neutralize the activity of heparin, as well as TM submitted to chondroitinase digestion were used in order to identify the role of the different structural domains of TM. Binding of thrombin to TM at a primary site on the protein part is a prerequisite for all the biological activities of TM. However, while this binding is sufficient for TM to promote the activation of PC by thrombin, the inhibition by TM of thrombin-induced fibrinogen clotting and factor V activation requires the interaction of thrombin at a secondary site with the polysaccharide chain of TM. This interaction with the polysaccharide chain (which carries a highly sulfated trisaccharide at the non-reducing terminus) leads to the inhibition of the procoagulant functions of TM-bound thrombin towards fibrinogen and factor V as well as an increased reactivity of the enzyme with ATIII. These results were rationalized in the functional model proposed for the rabbit TM-proteoglycan. An original aspect of the TM-proteoglycan resides in the fact that the chondroitin sulfate side chain brings new anticoagulant activities, in addition to the PC activation cofactor activity, to the molecule. TM is a new type of proteoglycan with important regulatory function in hemostasis, which anticoagulant properties depend on both the protein core and the polysaccharide chain.

Isolation and characterization of the glycosaminoglycan component of rabbit thrombomodulin proteoglycan.

Previous studies on rabbit thrombomodulin (TM) revealed that certain anticoagulant activities expressed by TM depend on the presence of an acidic domain tentatively identified as a sulfated galactosaminoglycan (Bourin, M.-C., Ohlin, A.-K., Lane, D., Stenflo, J., and Lindahl, U. (1988) J. Biol. Chem. 263, 8044-8052). The glycan was released by alkaline beta-elimination, isolated by ion-exchange chromatography, and radiolabeled by partial N-deacetylation (hydrazinolysis) followed by re-N-[3H]acetylation. The labeled product behaved like standard chondroitin sulfate on ion-exchange chromatography, exhibited a Mr of 10-12 x 10(3) on gel chromatography, and was susceptible to degradation by chondroitinase and testicular hyaluronidase. The major labeled degradation products following digestion of the glycosaminoglycan with chondroitinase were identified, depending on the incubation conditions, either as 4/6-mono-O-sulfated, 4,5-unsaturated disaccharides (delta HexA-GalNAc(S] and N-acetylgalactosamine 4,6-di-O-sulfate (GalcNAc (diS], the latter component accounting for approximately 25% of the total label, or as a major fraction of labeled trisaccharide, with the predominant structure GalNAc(diS)-GlcA-GalNAc(diS). The terminal GalNAc(diS) unit (not substituted at C3) was shown to be more susceptible to N-deacetylation during hydrazinolysis than were the internal GalNAc units (substituted at C3), and thus was more extensively labeled, resulting in over-representation of this unit. It is concluded that rabbit TM is a chondroitin sulfate proteoglycan, which carries a single glycan side chain characterized by an unusual accumulation of sulfate groups at the nonreducing terminus. Metabolically 35S-labeled TM was isolated from cultured rabbit heart endothelial cells and characterized as a chondroitin sulfate proteoglycan which accounted for 1-2% of the total 35S-labeled cell-associated macromolecules. The isolated chondroitin sulfate showed weaker antithrombin-dependent anticoagulant activity, on a molar basis, than the intact TM proteoglycan. The anticoagulant action of TM thus depends on a unique form of functional collaboration between the different constituents of a glycoconjugate.

Functional role of the polysaccharide component of rabbit thrombomodulin proteoglycan. Effects on inactivation of thrombin by antithrombin, cleavage of fibrinogen by thrombin and thrombin-catalysed activation of factor V.

Thrombomodulin (TM), a major anticoagulant protein at the vessel wall, serves as a potent cofactor for the activation of Protein C by thrombin. Previous work has indicated that (rabbit) TM is a proteoglycan that contains a single polysaccharide chain, tentatively identified as a sulphated galactosaminoglycan, and furthermore suggested that this component may be functionally related to additional anticoagulant activities expressed by the TM molecule [Bourin, Ohlin, Lane, Stenflo & Lindahl (1988) J. Biol. Chem. 263, 8044-8052]. Results of the present study establish that (enzymic) removal of the polysaccharide chain abolishes the inhibitory effect of TM on thrombin-induced fibrinogen clotting as well as the promoting effect of TM on the inactivation of thrombin by antithrombin, but does not affect the ability of TM to serve as a cofactor in the activation of Protein C. Studies of yet another biological activity of rabbit TM, namely the ability to prevent the activation of Factor V by thrombin [Esmon, Esmon & Harris (1982) J. Biol. Chem. 257, 7944-7947], confirmed that TM markedly delays the conversion of the native 330 kDa Factor V precursor into polypeptide intermediates, and further into the 96 kDa heavy chain and 71-74 kDa light-chain components of activated Factor Va. In contrast, the activation kinetics of a similar sample of Factor V incubated with thrombin in the presence of chondroitinase ABC-digested TM did not differ from that observed in the absence of TM. It is concluded that the inhibitory effect of TM on Factor V activation also depends on the presence of the polysaccharide component on the TM molecule.

Effect of rabbit thrombomodulin on thrombin inhibition by antithrombin in the presence of heparin.

Thrombomodulin acts as a cofactor for protein C activation by thrombin (PC activation cofactor activity) and inhibits thrombin-induced fibrinogen clotting (direct anticoagulant activity). In addition, rabbit thrombomodulin has been shown to promote thrombin inactivation by antithrombin (AT-dependent anticoagulant activity). However, a non-acidic form (i.e. non-retarded on ion-exchange chromatography) of thrombomodulin generated by limited proteolysis retained only the PC activation cofactor activity. The acidic form (retarded on ion-exchange chromatography) of thrombomodulin is now shown to prevent the rapid inactivation of thrombin by antithrombin in the presence of heparin, presumably by preventing the formation of the ternary thrombin-AT-heparin complex. This effect was not observed with non-acidic thrombomodulin. When submitted to chondroitinase digestion, thrombomodulin was converted into an essentially non-acidic form that lacked both the AT-dependent and the direct anticoagulant activities but showed a PC activation cofactor function indistinguishable from that of native thrombomodulin. This chondroitinase-digested form did not prevent the catalytic effect of heparin on the inhibition of thrombin by AT. It is concluded that the acidic domain of rabbit thrombomodulin, a chondroitin (dermatan) sulfate glycosaminoglycan, interacts with a site of the thrombin molecule that is not involved in the protein C activation cofactor function, but is essential to the cleavage of fibrinogen or binding of heparin.

Relationship between anticoagulant activities and polyanionic properties of rabbit thrombomodulin.

Rabbit thrombomodulin displays three distinct blood anticoagulant activities: it promotes the activation of protein C by thrombin (protein C activation cofactor activity); it promotes the inactivation of thrombin by thrombin (direct anticoagulant activity). The effects on these activities of mouse anti-thrombomodulin monoclonal antibodies and of the heparin-neutralizing proteins, platelet factor 4, histidine-rich glycoprotein, and S-protein, were investigated. One of the antibodies, which did not influence the functional properties of thrombomodulin, was used as an immunoaffinity ligand for purification of the protein. Two other antibodies, which were found to abrogate the protein C activation cofactor activity of the purified thrombomodulin, also abolished the antithrombin-dependent and the direct anticoagulant activities. The heparin-neutralizing proteins all inhibited the two latter activities, albeit to a varying extent, but did not appreciably affect the activation of protein C. These results are interpreted in relation to our previous finding that rabbit thrombomodulin contains an acidic domain, tentatively identified as a sulfated glycosaminoglycan (Bourin, M.-C., Boffa, M.-C., Björk, I., and Lindahl, U. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 5924-5928). It is proposed that the acidic domain interacts with thrombin at the protein C activation site and that this interaction is a prerequisite to the expression of direct as well as antithrombin-dependent anticoagulant activity. The interaction is not essential to, but compatible with, the activation of protein C. Experiments involving treatment of thrombomodulin with various glycanases or with nitrous acid, followed by measurement of anticoagulant activities, indicated that the acidic domain is constituted by a sulfated galactosaminoglycan and not by a heparin-related polysaccharide as previously suggested.

[Effect of rabbit thrombomodulin on the inhibition of thrombin by the antithrombin-heparin complex: role of the acid domain of thrombomodulin]. / Effet de la thrombomoduline de lapin sur l'inhibition de la thrombine par le complexe antithrombine-héparine: rôle du domaine acide de la thrombomoduline.

Acidic and non-acidic forms of rabbit thrombomodulin were studied with regard to their effects on the inhibition of thrombin by antithrombin in the presence of exogenous heparin. The non acidic form was obtained by proteolytic cleavage of a polyanionic component (presumably a sulfated polysaccharide) from the parent acidic form of thrombomodulin, and purified by ion-exchange chromatography. It was previously found that the acidic form of thrombomodulin increases the rate of thrombin inactivation by antithrombin. The present study showed that thrombin bound to acidic thrombomodulin was inactivated at a lower rate by antithrombin in the presence of exogenous heparin than was free thrombin or thrombin bound to the non-acidic form of thrombomodulin. The data suggest that the acidic component of thrombomodulin is primarily responsible for the retardation of thrombin-antithrombin complex formation in the presence of exogenous heparin. It is proposed that the polyanionic component of thrombomodulin blocks a site on thrombin required for heparin binding, thus rendering the antithrombin-heparin complex ineffective.

Functional domains of rabbit thrombomodulin.

Thrombomodulin isolated from rabbit lung was separated by ion-exchange chromatography on DEAE-cellulose into a retarded (acidic) and a nonretarded (nonacidic) fraction. Both fractions contained the cofactor required for the activation of protein C. In addition, the acidic fraction (but not the nonacidic fraction) prevented the clotting of fibrinogen by thrombin ("direct" anticoagulant activity) and accelerated the inhibition of thrombin by antithrombin (effect corresponding to 2-10 international units of heparin per mg of protein). Both of these activities were readily neutralized by the synthetic polycation Polybrene, which did not appreciably affect protein C activation. They were also eliminated by digestion of thrombomodulin with bacterial heparinase, which, in addition, converted the acidic form of the protein C activation cofactor to a nonacidic form. Similar conversion observed during storage of thrombomodulin was attributed to endogenous proteinase activity. Density-gradient centrifugation of the acidic form of thrombomodulin in CsCl/4M guanidinium chloride failed to separate either of the direct or antithrombin-dependent anticoagulant activities from the protein C activation cofactor, which showed a buoyant density of 1.31-1.34 g/ml. The nonacidic cofactor had a lower density, 1.26-1.28 g/ml. Unreduced thrombomodulin yielded two major fractions of protein C activation cofactor on NaDodSO4/PAGE, with apparent Mr of approximately 68,000 and 57,000, respectively. The larger component contained essentially all of the direct and antithrombin-dependent anticoagulant activities. We propose that these activities as well as the negative charge and the higher buoyant density of the acidic, Mr 68,000 form of thrombomodulin are due to a heparin-like polysaccharide and, further, that this component can be separated from the major portion of the molecule, which contains the protein C activation site, through the action of a proteinase.

Effect of phorbol esters on guniea pig skin in vivo.

When topically applied to guniea pig ear skin the tumor promoting phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) induced inflammation and epidermal hyperproliferation which could be inhibited by indomethacin. This inhibition could be reversed both by prostaglandins E and F. Five minutes after TPA treatment an increase in the level of prostaglandin E but not of prostaglandin F was observed in the epidermis. The non-promoting phorbol ester 4-O-methyl-TPA also stimulated epidermal cell proliferation but this stimulation was not inhibited by indomethacin. The above results are in agreement with those already reported in the mouse system with these two compounds. Ornithine decarboxylase (ODC) activity has been evaluated in the epidermis of guniea pig ear after topical application of 20 nmol of TPA. No increase was noted. This is in contrast with the well documented activation of ODC in mouse skin treated with TPA. Since TPA acts as a promoter in the mouse whereas both croton oil and TPA have no promoting action in the guinea pig, the above result supports the view that ODC activationis related to promotion, and provides a possible explanation for the resistance of this animal species to promotion. This resistance is further documented by the fact that no "dark cells" were found in guinea pig ear skin.