Leflunomide: an immunomodulatory drug for the treatment of rheumatoid arthritis and other autoimmune diseases.

Leflunomide (Arava) has recently been approved by the Food and Drug Administration for the treatment of rheumatoid arthritis (RA). The drug, due to its protective effects on structural joint damage, has been classified as a disease modifying anti-rheumatic drug (DMARD). Leflunomide is structurally dissimilar from other drugs currently used to treat RA and exhibits a different mechanism of action. It has shown to be protective in a variety of animal models of arthritis and autoimmunity based on its immunomodulatory activity. Leflunomide is rapidly converted in vivo to its pharmacologically active metabolite A77 1726. This metabolite is a potent non-cytotoxic inhibitor of the enzyme dihydroorotate dehydrogenase (DHODH), a key enzyme in the de novo synthesis of uridine monophosphate (UMP). Activated lymphocytes depend on the pyrimidine de novo syntheses to fulfill their metabolic needs for clonal expansion and terminal differentiation into effector cells. De novo synthesis of pyrimidines is not only essential to provide precursors for new RNA and DNA synthesis, but also for phospholipid synthesis and the pyrimidine sugars necessary for protein glycosylation, which support the massive expansion in membrane biosynthesis to form daughter cells. This mechanism likely contributes to leflunomide's action as a DMARD in RA and other autoimmune diseases. This review is a summary of current in vivo and in vitro data, focussing primarily on the mechanism of action of leflunomide in RA.

Mechanism of action for leflunomide in rheumatoid arthritis.

Leflunomide (Arava) has recently been approved by the Food and Drug Administration for the treatment of rheumatoid arthritis (RA). This approval was based on data from a double-blind, multicenter trials in the United States (leflunomide versus methotrexate versus placebo) in which leflunomide was superior to placebo and similar to methotrexate (Strand et al., Arch. Intern. Med., in press, 1999). In a multicenter European trial, leflunomide was similar to sulfasalazine in efficacy and side effects (Smolen et al., Lancet 353, 259-266, 1999). Both methotrexate and leflunomide retarded the rate of radiolographic progression, entitling them to qualify as disease-modifying agents (Strand et al., Arch. Intern. Med., in press, 1999). Leflunomide is an immunomodulatory drug that may exert its effects by inhibiting the mitochondrial enzyme dihydroorotate dehydrogenase (DHODH), which plays a key role in the de novo synthesis of the pyrimidine ribonucleotide uridine monophosphate (rUMP). The inhibition of human DHODH by A77 1726, the active metabolite of leflunomide, occurs at levels (approximately 600 nM) that are achieved during treatment of RA. We propose that leflunomide prevents the expansion of activated and autoimmune lymphocytes by interfering with the cell cycle progression due to inadequate production of rUMP and utilizing mechanisms involving p53. The relative lack of toxicity of A77 1726 on nonlymphoid cells may be due to the ability of these cells to fulfill their ribonucleotide requirements by use of salvage pyrimidine pathway, which makes them less dependent on de novo synthesis.

Comparative gene-expression analysis.

The study of differences in gene-expression patterns is one of the most promising approaches for understanding mechanisms of differentiation and development. In addition, the identification of disease-related target molecules opens new avenues for rational pharmaceutical intervention. Recent technical advances and improvements are accelerating the analysis of gene-expression profiles at the transcript level. The knowledge and comprehension of currently applied methods is one of the central criteria for an efficient and successful gene-screening approach.

How does leflunomide modulate the immune response in rheumatoid arthritis?

Leflunomide has recently been approved by the US Food and Drug Administration for the treatment of rheumatoid arthritis. This approval was based on data from double-blind multicentre trials in the US (US 301; leflunomide versus methotrexate versus placebo) and multicentre European trials (leflunomide versus sulfasalazine versus placebo, and leflunomide versus methotrexate versus placebo). In these trials, leflunomide was superior to placebo and similar to methotrexate or sulfasalazine in efficacy and adverse effects. Both methotrexate and leflunomide retarded the rate of radiological progression, entitling them to qualify as disease-modifying agents (DMARDs). Leflunomide is an immunomodulatory drug that may exert its effects by inhibiting the mitochondrial enzyme dihydro-orotate dehydrogenase (DHO-DH), which plays a key role in the de novo synthesis of the pyrimidine ribonucleotide uridine monophosphate (rUMP). The inhibition of human DHO-DH by A77-1726, the active metabolite of leflunomide, occurs at concentrations (approximately 600 nmol/L) that are achieved during treatment of rheumatoid arthritis. We propose that leflunomide prevents the expansion of activated and autoimmune lymphocytes by interfering with cell cycle progression. This is mediated by inadequate production of rUMP and utilises mechanisms involving the sensor protein p53. The relative lack of toxicity of A77-1726 on nonlymphoid cells may be due to the ability of these cells to fulfil their ribonucleotide requirements by use of the salvage pyrimidine pathway, which makes them less dependent on de novo synthesis.

Novel muscle-specific enhancer sequences upstream of the cardiac actin gene.

A DNase I-hypersensitive site analysis of the 5'-flanking region of the mouse alpha-cardiac actin gene with muscle cell lines derived from C3H mice shows the presence of two such sites, at about -5 and -7 kb. When tested for activity in cultured cells with homologous and heterologous promoters, both sequences act as muscle-specific enhancers. Transcription from the proximal promoter of the alpha-cardiac actin gene is increased 100-fold with either enhancer. The activity of the distal enhancer in C2/7 myotubes is confined to an 800-bp fragment, which contains multiple E boxes. In transfection assays, this sequence does not give detectable transactivation by any of the myogenic factors even though one of the E boxes is functionally important. Bandshift assays showed that MyoD and myogenin can bind to this E box. However, additional sequences are also required for activity. We conclude that in the case of this muscle enhancer, myogenic factors alone are not sufficient to activate transcription either directly via an E box or indirectly through activation of genes encoding other muscle factors. In BALB/c mice, in which cardiac actin mRNA levels are 8- to 10-fold lower, the alpha-cardiac actin locus is perturbed by a 9.5-kb insertion (I. Garner, A. J. Minty, S. Alonso, P. J. Barton, and M. E. Buckingham, EMBO J. 5:2559-2567, 1986). This is located at -6.5 kb, between the two enhancers. The insertion therefore distances the distal enhancer from the promoter and from the proximal enhancer of the bona fide cardiac actin gene, probably thus perturbing transcriptional activity.

Definition of the transcriptional activation domain of recombinant 43-kilodalton USF.

The cellular transcription factor USF is involved in the regulation of both cellular and viral genes and consists of 43- and 44-kDa polypeptides which independently show site-specific DNA binding. Cloning of the corresponding cDNA revealed that the 43-kDa polypeptide (USF43) is a member of the basic (B)-helix-loop-helix (HLH)-leucine zipper (LZ) family of proteins and provided a means for its functional dissection. Initial structure-function studies revealed that the HLH and LZ regions are both important for USF43 oligomerization and DNA binding. The studies presented here have focused on the determination of domains that contribute to transcriptional activation in vitro and show that (i) both a small region close to the N terminus and a region between residues 93 and 156 contribute strongly to transcriptional activation, (ii) full activation depends on the presence of both domains, (iii) the B-HLH-LZ region has no intrinsic activation potential but DNA binding is absolutely required for transcriptional activation, and (iv) the B-HLH-LZ region can be replaced by the Gal4 DNA binding domain without loss of activation potential.

Antagonistic effects of chronic low frequency stimulation and thyroid hormone on myosin expression in rat fast-twitch muscle.

This study investigates effects of chronic low frequency stimulation (CLFS) on myosin heavy (MHC) and light chain (MLC) expression in fast-twitch muscles in hypothyroid, euthyroid, and hyperthyroid rats. The changes at both the mRNA and protein level indicated antagonistic effects of thyroid hormone and CLFS: under euthyroid conditions, CLFS mainly elicited a MHCIIb----MCHIId----MHCIIa transition. Whereas CLFS did not induce the slow MHCI in the euthyroid state, this isoform was present in the hypothyroid state and was further enhanced with CLFS indicating the suppressive effect of thyroid hormone to be stronger than the inductive influence of CLFS. Hyperthyroidism alone suppressed the expression MHCIIa and enhanced a MHCIId to MHCIIb transition. This shift to the faster MHC isoforms was only partially counteracted by CLFS. Thus, it appeared that thyroid hormone had a graded suppressive effect on the expression of MHC isoforms in the order MHCIId less than MHCIIa less than MHCI. Elevated neuromuscular activity partially counteracted these hormone effects. Changes in MLC mRNAs were consistent with those in the MHC pattern, i.e. increases or decreases in MHCIIb led to corresponding changes in the expression of MLC3f. A similar relationship existed for the slow MHCI and the slow MLC isoforms.

Rapid and reversible changes in myosin heavy chain expression in response to increased neuromuscular activity of rat fast-twitch muscle.

Chronic 10 Hz stimulation of rat fast-twitch muscle induced rapid and reversible changes in the tissue levels of fast myosin heavy chain (HC) mRNA isoforms. These changes consisted of a rapid decrease in HCIIb mRNA and a progressive increase in HCIIa mRNA. After 15 days, the HCIIb mRNA normally amounting to approximately 80%, had decreased to less than 5% of the sum of the two HC mRNA isoforms. HCIIb mRNA was again detectable one day after cessation of stimulation and progressively increased at the expense of HCIIa mRNA with ongoing recovery. These results point to a down-regulation of the HCIIb gene by the applied stimulus pattern which, conversely, enhances the expression of the HCIIa gene.

Chronic stimulation-induced changes of myosin light chains at the mRNA and protein levels in rat fast-twitch muscle.

Transitions in the expression of the myosin light chains (LC) were investigated in fast-twitch muscles of the rat during chronic (10 h/day), low-frequency (10 Hz) stimulation. Changes were followed at the mRNA level by Northern blot analysis and in vitro translation, as well as at the protein level by electrophoresis under denaturing and nondenaturing conditions. In vivo synthesis of the light chains was assessed by measuring the incorporation of intramuscularly injected [35S]methionine. Chronic stimulation induced a transition in the isomyosin pattern with an increase of FM3, a concomitant decrease in FM1 and, after longer stimulation periods, the appearance of low concentrations of the slow isomyosin. These changes were accompanied by an elevated LC1f/LC3f ratio and increases in the amounts of both the LC1sb and, to a lesser degree, LC2s proteins. Alterations in the amounts of specific mRNAs were the same whether determined by Northern blot analysis or by in vitro translation of total RNA preparations from the same muscles. Generally, the changes in the relative concentrations of fast and slow light-chain proteins agreed with the changes detected at the mRNA level and the alterations in protein synthesis detected with the use of an in vivo labeling assay. An exception was the elevated tissue content of LC2s where no changes were detectable in the concentration of its mRNA as determined by in vitro translation or in vivo synthesis. The increase in LC2s protein may, therefore, have been due to reduced degradation. In addition, the decrease in LC3f was more pronounced at the protein level than at the mRNA level. This might indicate an increased turnover of LC3f or the existence of additional post-transcriptional regulations of LC3f expression.

Electrostimulation-induced fast-to-slow transitions of myosin light and heavy chains in rabbit fast-twitch muscle at the mRNA level.

Chronic low-frequency stimulation of rabbit fast-twitch muscle induces progressive increases in slow myosin light chain mRNAs followed by an increase in the slow myosin heavy chain HCI mRNA. Therefore, the effects of chronic stimulation are more pronounced in rabbit than in rat fast-twitch muscle. The latter responds mainly with a rearrangement of its fast isomyosin pattern.

The multiplicity of troponin T isoforms. Distribution in normal rabbit muscles and effects of chronic stimulation.

Polyclonal antibodies were raised in guinea pigs against troponin-T (TnT) isoforms purified from fast- and slow-twitch rabbit muscles. With the use of these antibodies and immunoblots of one- and two-dimensional electrophoreses, the distribution of fast and slow TnT isoforms was investigated in normal and chronically stimulated hindlimb muscles of the rabbit. According to differences in their apparent molecular masses, six fast TnT isoforms (TnTcf, TnT1f, TnT2f, TnT3f, TnT4f, TnT5f) were distinguished in normal tibialis anterior and extensor digitorum longus muscles. These muscles also contained low amounts of TnT1s and TnT2s which were the predominant TnT isoforms in slow-twitch soleus muscle. Fast and slow TnT isoforms were found to exist in several charge variants, i.e. one for TnTcf, three different charge forms for TnT1f, seven for TnT2f, four for TnT3f, three for TnT4f, one for TnT5f, four for TnT1s, and three for TnT2s. Some charge variants were phosphorylated isoforms because treatment with alkaline phosphatase reduced the number of the 19 fast and 7 slow variants to 12 and 3, respectively. The stimulation-induced fast-to-slow transition caused progressive decreases in fast and increases in slow isoforms. The decrease and the disappearance of the major fast isoforms followed a sequence of TnT2f, TnTcf, TnT4f, TnT1f, and TnT3f. This decrease in fast isoforms fits well with the reduction of fast TnT mRNAs assessed by Northern blot analysis. Prolonged stimulation ultimately created a TnT isoform pattern similar to that found in normal slow-twitch muscle. Stimulation also induced changes in the tropomyosin subunit pattern with a decrease in the fast and an increase in the slow alpha-tropomyosin subunit without altering the alpha/beta-tropomyosin subunit ratio. Similar to slow-twitch soleus muscle, long-term stimulated muscles contained appreciable amounts of the fast alpha-tropomyosin subunit, but only traces of fast TnT isoforms. This combination indicated that the predominant slow TnT isoforms may be capable of interacting with fast tropomyosin in these muscles.

Neural control of gene expression in skeletal muscle. Effects of chronic stimulation on lactate dehydrogenase isoenzymes and citrate synthase.

The aim of this study was to investigate the effects of neural activity on the expression of fibre-type-specific patterns of metabolic enzymes at the levels of transcription and translation. For this purpose, changes in tissue amounts of citrate synthase (CS) and the H- and M-subunits of lactate dehydrogenase (LDH) were followed in fast-twitch rabbit muscles during low-frequency (10 Hz, 12 h/day) nerve stimulation. These stimulation-induced alterations were correlated with changes in tissue amounts of the total poly(A)+ (polyadenylated) RNA, poly(A)+ RNAs specifically translatable in vitro, yield of total ribosomes and distributions of monosomes and polysomes. The tissue contents of poly(A)+ RNAs translatable in vitro coding for CS and H- and M-LDH were quantified by immunoprecipitation of their translation products. Increases in total ribosome yields occurred after 4 days' stimulation, reaching a maximum between 14 and 21 days. Stimulation for only 1-2 days greatly increased the amount of monosomes. An increase in polysomes occurred before that in total ribosomes, suggesting that monosomes were integrated into polysomes. Total poly(A)+ RNA significantly increased in muscles stimulated for more than 6 days. A maximum increase of 2.5-fold was attained after 14-21 days. Chronic stimulation progressively induced the appearance of LDH isoenzymes containing the H-subunit, with a predominance of LDH-3. This shift corresponded to a slow decay of the M-subunit and a 2-fold steep increase in the H-subunit. These changes correlated with those of the respective poly(A)+ RNAs translatable in vitro, thus indicating that the re-arrangement of the LDH isoenzyme pattern is mainly due to qualitatively and quantitatively altered transcription. The increase in CS was biphasic and consisted of a moderate rise during the first 4 days of stimulation and a steep rise thereafter. The latter coincided with a steep increase in poly(A)+ RNA translatable in vitro coding for CS. In view of the early increase in translational capacity, it was concluded that the initial rise in CS resulted from selective post-transcriptional control and enhanced translation in vivo of existing mRNA, whereas its steep increase was due to enhanced transcription. These results indicate that the neurally regulated expression of phenotype-specific properties in muscle includes control of both transcription and translation.