Romazarit: a potential disease-modifying antirheumatic drug.

The synthesis of a series of substituted heterocyclic alkoxypropionic acids is described. They were evaluated for antiinflammatory effects in two animal models of chronic inflammation; adjuvant arthritis and type II collagen arthritis in the rat. The desired profile of biological activity was characterized by the reduction of inflammation with the coincident restoration toward normal levels of the biochemical markers (acute phase proteins) associated with the inflammatory response, an effect that was not shared by classical nonsteroidal antiinflammatory agents. Romazarit, (Ro 31-3948, 7), 2-[[2-(4-chlorophenyl)-4-methyl-5-oxazolyl]methoxy]-2-methylpropio nic acid, was selected for further evaluation. In contrast to NSAIDs, romazarit was inactive in animal models of acute inflammation, and furthermore it did not inhibit the cyclooxygenase enzyme in vitro or in vivo. Inhibition of interleukin-1-mediated events in vitro has been observed.

Biologic properties of romazarit (Ro 31-3948), a potential disease-modifying antirheumatic drug.

The biologic effects of a new potential disease-modifying antirheumatic drug, romazarit (Ro 31-3948, 2-[[2-(4-chlorophenyl)-4-methyl-5-oxazolyl]methoxy]-2-methylpropio nic acid), have been investigated. In a 5-day adjuvant arthritis model, romazarit inhibited the development of hindpaw inflammation with a minimum effective dose of 30 mg kg-1. Plasma levels of the acute phase reactants seromucoid and haptoglobulin were also significantly reduced. Romazarit was equally effective in adrenalectomized animals, indicating that the compound is not acting via stimulation of the pituitary/adrenal axis. When the developing adjuvant arthritis was extended to 15 days romazarit showed dose-related improvements of all the symptoms of arthritis with a minimum effective dose of 25 mg kg-1. Romazarit caused a dose-dependent (range 20-250 mg kg-1) reduction in both the inflammatory and bony changes occurring during collagen arthritis in the rat, without any significant effect on anticollagen antibody titers except at the highest dose. Collagenase and prostaglandin E2 production in cultures of talus bones taken from rats with collagen arthritis were reduced by romazarit. In vitro romazarit was an extremely weak inhibitor of prostaglandin synthetase activity in both sheep seminal vesicle (IC50 6500 microM) and rat renal medulla (IC50 greater than 300 microM) cell-free preparations. Romazarit showed little or no activity in models of acute inflammation such as rabbit skin edema, carrageenan pleurisy or UV-induced erythema. In both acute and chronic tests romazarit displayed no ulcerogenic potential. In comparison with the structurally similar compound clobuzarit, hepatic changes such as increases in catalase and peroxisome proliferation-associated 80,000 mol.wt. protein were markedly less with romazarit. Clinical studies with romazarit are currently in progress.

Site-directed mutagenesis of citrate synthase; the role of the active-site aspartate in the binding of acetyl-CoA but not oxaloacetate.

Asp-362, a potential key catalytic residue of Escherichia coli citrate synthase (citrate oxaloacetate-lyase [pro-3S)-CH2COO- ----acetyl-CoA), EC 4.1.3.7) has been converted to Gly-362 by oligonucleotide-directed mutagenesis. The mutant gene was completely sequenced, using a series of synthetic oligodeoxynucleotides spanning the structural gene to confirm that no additional mutations had occurred during genetic manipulation. The mutant gene was expressed in M13 bacteriophage and produced a protein which migrated in an identical manner to wild-type E. coli citrate synthase on SDS-polyacrylamide gels and which cross-reacted with E. coli citrate synthase antiserum. The mutant gene was subsequently recloned into pBR322 for large scale purification of the protein, and the resulting plasmid, pCS31, used to transform the citrate synthase deletion strain, W620. The mutant enzyme purified in an analogous manner to wild-type E. coli citrate synthase and expressed less than 2% of wild-type enzyme activity. The activity of the partial reactions catalysed by citrate synthase was similarly affected suggesting that this residual activity may be due to contaminating wild-type enzyme activity. The mutant citrate synthase retains a high-affinity NADH-binding site consistent with the protein preserving its overall structural integrity. Oxaloacetate binding to the protein is unaffected by the Asp-362 to Gly-362 mutation. Binding of the acetyl-CoA analogue, carboxymethyl-CoA, could not be detected in the mutant protein indicating that the lack of catalytic competence is due primarily to the inability of the protein to bind the second substrate, acetyl-CoA.

Phosphorylation of type-L pyruvate kinase in intact hepatocytes. Localisation of the phosphorylation site in response to both glucagon and the Ca2+-linked agonist phenylephrine.

Pyruvate kinase is one of the enzymes which can be phosphorylated by stimulation of the cell with either glucagon or Ca2+-linked hormones. Whether these two classes of hormones phosphorylate the same site on the enzyme is unclear. Our results demonstrate that isolation of [32P]phosphorylated type-L pyruvate kinase from glucagon-treated hepatocytes followed by aspartyl-prolyl cleavage yields a [32P]phosphorylated peptide of Mr 17,000. This fragment is also phosphorylated in response to the Ca2+-mediated agonist phenylephrine.

Hypolipidemic effect of polymethylenemethane thiosulfonates: inhibitors of acetoacetyl coenzyme A thiolase.

A new series of bifunctional thiosulfonates of the general formula CH3SO2S-(CH2)n-SSO2CH3(SS1 polymethylene bis methane thiosulfonate) with variable polymethylene chain lengths (n = 6, 8, 10 and 12) were evaluated for their hypolipidemic action on serum cholesterol and triglyceride levels in rats. Their action was based on their specific inhibitory effect on cytoplasmic acetoacetyl-CoA thiolase, one of the key enzymes in cholesterol biosynthesis. These compounds inhibited the enzyme in vitro and in vivo. The inhibition in vitro was in the order of n = 12 greater than n = 10 greater than n = 8 greater than n = 6 greater than, where n is the number of methylene groups inserted between the two thiosulfonate groups. In vivo, the compounds produced variable hypocholesterolemic and/or hypotriglyceridemic effects when injected into groups of newly weaned rats fed standard chow, high fat or high carbohydrate diets. When the enzyme activity was measured in isolated liver homogenates in vitro after injections of the drugs in vivo, 80% of original thiolase activity was lost. This inhibition of enzyme activity did not seem to be rate limiting for their hypolipidemic action in vivo as these effects did not correlate with the inhibition of the isolated enzyme. The lack of correlation between in vitro and in vivo activity might be due to the compounds affecting other enzyme systems and/or due to their differential disposition in vivo.

Anti-inflammatory effects of synthetic retinoids may be related to their immunomodulatory action.

The effects of retinoids have been studied in a model of delayed-type hypersensitivity using the T-cell-dependent antigen, methylated bovine serum albumin to elicit inflammation in the hind paws of mice. A number of synthetic retinoids, including etretinate and arotinoids, showed a marked anti-inflammatory action in this model. Using differential dosing schedules, the anti-inflammatory effect of retinoids was clearly distinguished from conventional cyclooxygenase inhibitors. By screening a number of synthetic retinoids, structure-activity relationships for this effect can be deduced.

Identification of subsidiary catalytic groups at the active site of beta-ketoacyl-CoA thiolase by covalent modification of the protein.

beta-Ketoacyl-CoA thiolase (acyl-CoA:acetyl-CoA C-acyltransferase, EC 2.3.1.16) is known to possess sulfhydryl groups of cysteines at the active site that are essential for its catalytic activity. Other groups at the active site that participate in the catalytic process were identified by using anhydride reagents which covalently modify the protein by specifically reacting with any amino groups potentially present at the active site. Since these reagents may also react with thiol groups, the enzyme's amino groups were modified after masking the cysteine thiols present by an alkylalkane thiosulfonate-type reagent, methyl methanethiol-sulfonate (MMTS), that selectively formed a disulfide bridge, thus generating an inactive thiolmethylated enzyme. When this procedure was followed, the enzyme could be undoubtedly modified at its amino by the anhydride reagent, leading to a doubly modified protein. The thiomethyl group could then be removed by reduction with dithiothreitol, yielding an enzyme modified solely on the amino residues. The amino group could be unblocked in turn by exposure to acidic pH. The different anhydrides inactivated thiolase, but only acetoacetyl coenzyme A (AcAcCoA) provided any protection against inactivation. When thiolmethylcitraconyl thiolase was reduced with dithiothreitol the enzyme remained inactive, but when the doubly modified enzyme was exposed to pH 5 then the reduction led to formation of an active enzyme. These results are interpreted as demonstrating a role for an amino group at the enzyme active site. A catalytic mechanism is proposed for the enzyme which involves the amino group.

The synthesis and use of oligodeoxynucleotides in plasmid DNA sequencing.

A convenient procedure for the synthesis and purification of oligonucleotides is described. 16-base long primers synthesised by this method were used to investigate DNA sequencing using plasmid DNA as a template. This allowed the further analysis of the E. coli glt A sequence coding for citrate synthase and enabled determination of the 5'-non-coding regulatory region of the aminoglycoside phosphotransferase gene.

Specific inhibition of L-type pyruvate kinase by the triazine dye Procion Blue MX-R.

Incubation of the triazine dye Procion Blue MX-R with L- and M-type pyruvate kinase resulted in rapid time- and dye-concentration-dependent loss of activity. L-type pyruvate kinase was protected only by a low concentration of Mg2+; this was not the case with the M-type enzyme. Modification of the L-type form resulted in the incorporation of 1.54 +/- 0.057 mol of dye/mol of enzyme subunit in the absence of Mg2+, but only 0.73 +/- 0.024 mol of dye/mol of enzyme subunit in the presence of Mg2+. Tryptic peptide mapping of L-type pyruvate kinase modified in the presence and in the absence of Mg2+ further indicated that there were two sites modified in the enzyme, one of which was protected by Mg2+. The pKa of the nucleophile involved in the modification was calculated to be 7.1, implicating the possible involvement of a histidine residue. L-type enzyme was bound to Sepharose-immobilized Procion Blue MX-R specifically in the presence of Mg2+, whereas binding of the M-type enzyme was Mg2+-independent. The specific interaction of L-type pyruvate kinase with the dye was exploited in the large-scale purification of the enzyme and in the isolation of the phosphorylated enzyme.

Muscle and liver pyruvate kinases are closely related: amino acid sequence comparisons.

Previous evidence has shown that the M1 and L pyruvate kinase isozymes differ markedly in kinetic and immunological properties, amino acid compositions and peptide maps. However, the amino acid sequence results we present here for the N-terminal region and for a region of the C domain show that the M1 and L isozymes are very similar. The variable length of the N-terminal sequences also explains the difference in regulation by phosphorylation between the M1 and L isozymes. The M1 isozyme lacks the serine residue that has been shown to be phosphorylated in the L isozyme.

Citrate synthase activity in Escherichia coli harbouring hybrid plasmids containing the gltA gene.

A hybrid plasmid, pDB2, was constructed by ligating a 3.24 kb EcoRI/HindIII fragment of the Escherichia coli chromosome into pBR322. This was used to transform a gltA mutant which was devoid of citrate synthase activity. The resultant strain expressed very high citrate synthase activity and this enabled a simplified purification of the homogeneous enzyme in high yield. The subunit Mr was estimated as 47000-49000 by SDS gel electrophoresis, which closely resembles the eukaryotic form of the enzyme. Evidence for some conservation of sequence between the two proteins was revealed in the acid cleavage pattern at aspartyl-prolyl residues. In addition to coding for the structural gene for citrate synthase, the 3.24 kb EcoRI/HindIII fragment also retained the genetic structure necessary for control of enzyme synthesis since the expression of enzyme activity in the strain harbouring pDB2 was still subject to glucose repression.

The use of bacteriophage M13 carrying defined fragments of the Escherichia coli gltA gene to determine the location and structure of the citrate synthase promoter region.

The gltA gene from Escherichia coli, which encodes citrate synthase, has been located on a 3.24 Kb HindIII/EcoRl restriction fragment. This region contains one restriction site for BamHl and two for BglII. Defined restriction fragments from this region were cloned into suitably cleaved replicative form M13mp8 and M13mp9. The recombinants (M13gtlA1 leads to 10) were isolated as single stranded DNA and characterised on the basis of molecular weight and DNA sequence. The single stranded DNA was converted to the double stranded replicative form and used to transform E. coli strain JM103 from which bacteriophage were isolated. Infection of JM103 with different bacteriophage followed by measurement of expressed citrate synthase activity showed that the complete gltA gene must span the BamHl restriction site, that the control region was on the 5'-terminal side of this restriction site and that the coding region for citrate synthase protein commenced on the 3'-terminal side. Analysis of the DNA sequence of this region allowed us to confirm this model, to identify the start sequence for translation of the structural gene and a number of sequences controlling the initiation of transcription. Of special interest is the fact that there must be an extensive leader sequence (305 nucleotides) separating the predicted sites for initiation of transcription and translation.

The immunological and catalytically active form of L-type pyruvate kinase in rat liver cytosol.

The protein species precipitated from rat liver cytosol by rabbit antisera raised to pure L-type pyruvate kinase were investigated by sodium dodecyl sulphate gel electrophoresis. The primary antisera (anti-L-type pyruvate kinase) precipitated protein species with mol. wts 56,000, 41,000 and 39,000. The 41,000 mol. wt protein was identified as fructose-bis-phosphatase. Double diffusion and immunotitration experiments established that L-type pyruvate kinase and fructose-bis-phosphatase shared common antigenic determinants. This information enabled an improved antiserum (anti-LPK) to be obtained. The use of anti-LPK showed that the 56,000 mol. wt subunit was the only catalytically and immunologically active form of L-type pyruvate kinase in liver. This was confirmed by biosynthetic experiments with cultured hepatocytes. The specific activity of the enzyme in liver extracts was also determined by quantitative immunotitration with anti-LPK. Despite changes in dietary status which varied the concentration of enzyme protein, the maximum specific activity of the enzyme remained constant and essentially the same as that of pure enzyme.

Glucocorticoid hormones have a permissive role in the phosphorylation of L-type pyruvate kinase by glucagon.

The incorporation of [32P]phosphate into L-type pyruvate kinase in response to glucagon was studied in hepatocytes isolated from control and adrenalectomized rats. In control cells, pyruvate kinase phosphorylation was maximally stimulated by 210% by glucagon. Adrenalectomy reduced both the basal extent of phosphorylation and the response to glucagon to 25% of those measured in control hepatocytes. This impaired pyruvate kinase phosphorylation was associated with altered kinetic properties of L-type pyruvate kinase. Glucagon injected in vivo (1 mg/kg) into adrenalectomized rats did not increase the substrate-co-operativity of L-type pyruvate kinase. These results suggest a molecular basis for the well-recognised glucocorticoid hormone-dependence for the stimulation by glucagon of hepatic gluconeogenesis. Further effects of this defective enzyme regulation on hepatic metabolism are suggested by the observation that adrenalectomy abolished the ability of glucagon to inhibit hepatic fatty acid biosynthesis both in vivo and in isolated hepatocytes.