Species specificity of amidine-based urokinase inhibitors.

Inhibition of the proteolytic activity of urokinase has been shown to inhibit the progression of tumors in rodent models and is being investigated for use in human disease. Understanding the rodent/human species-specificity of urokinase inhibitors is therefore critical for interpretation of rodent cancer progression models that use these inhibitors. We report here studies with a panel of 11 diverse urokinase inhibitors in both human and mouse enzymatic assays. Inhibitors such as amiloride, B428, and naphthamidine, that occupy only the S1 subsite pocket were found to be nearly equipotent between the human and the murine enzymes. Inhibitors that access additional, more distal, pockets were significantly more potent against the human enzyme but there was no corresponding potency increase against the murine enzyme. X-ray crystallographic structures of these compounds bound to the serine protease domain of human urokinase were solved and examined in order to explain the human/mouse potency differences. The differences in inhibitor potency could be attributed to four amino acid residues that differ between murine and human urokinases: 60, 99, 146, and 192. These residues are Asp, His, Ser, and Gln in human and Gln, Tyr, Glu, and Lys in mouse, respectively. Compounds bearing a cationic group that interacts with residue 60 will preferentially bind to the human enzyme because of favorable electrostatic interactions. The hydrogen bonding to residue 192 and steric considerations with residues 99 and 146 also contribute to the species specificity. The nonparallel human/mouse enzyme inhibition observations were extended to a cell-culture assay of urokinase-activated plasminogen-mediated fibronectin degradation with analogous results. These studies will aid the interpretation of in vivo evaluation of urokinase inhibitors.

Thiazole analogues of the NSAID indomethacin as selective COX-2 inhibitors.

The carboxyl group of the NSAID indomethacin was replaced with a variety of substituted thiazoles to obtain a series of potent, selective inhibitors of COX-2. Additional substitutions were made at the 1-position and 5-position of the indole of indomethacin.

Symmetrical bis(heteroarylmethoxyphenyl)alkylcarboxylic acids as inhibitors of leukotriene biosynthesis.

Symmetrical bis(quinolylmethoxyphenyl)alkylcarboxylic acids were investigated as inhibitors of leukotriene biosynthesis and 4, 4-bis(4-(2-quinolylmethoxy)phenyl)pentanoic acid sodium salt (47.Na) met our design parameters for a drug candidate (ABT-080). This compound was readily synthesized in three steps from commercially available diphenolic acid. Against intact human neutrophils, 47.Na inhibited ionophore-stimulated LTB(4) formation with an IC(50) = 20 nM. In zymosan-stimulated mouse peritoneal macrophages producing both LTC(4) and PGE(2), 47.Na showed 9000-fold selectivity for inhibition of LTC(4) (IC(50) = 0.16 nM) over PGE(2) (IC(50) = 1500 nM). Preliminary pharmacokinetic evaluation in rat and cynomolgus monkey demonstrated good oral bioavailability and elimination half-lives of 9 and 5 h, respectively. Pharmacological evaluation of leukotriene inhibition with oral dosing was demonstrated in a rat pleural inflammation model (ED(50) = 3 mg/kg) and a rat peritoneal passive anaphylaxis model (LTB(4), ED(50) = 2.5 mg/kg; LTE(4), ED(50) = 1.0 mg/kg). In a model of airway constriction induced by antigen challenge in actively sensitized guinea pigs, 47.Na dosed orally blocked bronchoconstriction with an ED(50) = 0.4 mg/kg, the most potent activity we have observed for any leukotriene inhibitor in this model. The mode of inhibitory action of 47.Na occurs at the stage of 5-lipoxygenase biosynthesis as it blocks both leukotriene pathways leading to LTB(4) and LTC(4) but not PGH(2) biosynthesis. However, 47.Na does not inhibit 5-lipoxygenase catalysis in a broken cell enzyme assay; therefore it is likely that 47.Na acts as a FLAP inhibitor.

Fluorescent microplate assay for cancer cell-associated cathepsin B.

Cathepsin B and in particular cell-surface and secreted cathepsin B has been implicated in the invasive and metastatic phenotype of numerous types of cancer. We describe here a method to easily survey cancer cell lines for cathepsin B activity using the highly selective substrate Z-Arg-Arg-AMC. Intact human U87 glioma cells hydrolyze Z-Arg-Arg-AMC with a Km of 460 microM at pH 7.0 and 37 degrees C. This is nearly the same as the Km of 430 microM obtained with purified cathepsin B assayed under the same conditions. The pericellular (i.e. both cell-surface and released) cathepsin B activity was inhibited by the cysteine protease inhibitors E-64, leupeptin, Mu-Np2-HphVS-2Np, Mu-Leu-HpHVSPh and the cathepsin B selective inhibitor Mu-Tyr(3,5 I2)-HphVSPh with IC50 values similar to those observed for the inhibition of purified human liver cathepsin B. Other human cancer cell lines with measurable pericellular cathepsin B activity included HT-1080 fibrosarcoma, MiaPaCa pancreatic, PC-3 prostate and HCT-116 colon. Cathepsin B activity correlated with protein levels of cathepsin B as determined by immunoblot analysis. Pericellular cathepsin B activity was also detected in the rat cell lines MatLyLu prostate and Mat B III adenocarcinoma and in the murine lines B16a melanoma and Lewis lung carcinoma. The ability to determine pericellular cathepsin B activity will be useful in selecting appropriate cell lines for use in vivo when analyzing the effects of inhibiting cathepsin B activity on tumor growth and metastasis.

Heteroarylmethoxyphenylalkoxyiminoalkylcarboxylic acids as leukotriene biosynthesis inhibitors.

A novel series of heteroarylmethoxyphenylalkoxyiminoalkylcarboxylic acids was studied as leukotriene biosynthesis inhibitors. A hypothesis of structure-activity optimization by insertion of an oxime moiety was investigated using REV-5901 as a starting point. A systematic structure-activity optimization showed that the spatial arrangement and stereochemistry of the oxime insertion unit proved to be important for inhibitory activity. The promising lead, S-(E)-11, inhibited LTB(4) biosynthesis in the intact human neutrophil with IC(50) of 8 nM and had superior oral activity in vivo, in a rat pleurisy model (ED(50) = 0.14 mg/kg) and rat anaphylaxis model (ED(50) = 0.13 mg/kg). In a model of lung inflammation, S-(E)-11 blocked LTE(4) biosynthesis (ED(50) of 0.1 mg/kg) and eosinophil influx (ED(50) of 0.2 mg/kg). S-(E)-11 (A-93178) was selected for further preclinical evaluation.

Discovery of a new cyclooxygenase-2 lead compound through 3-D database searching and combinatorial chemistry.

Using a combination of computational and combinatorial chemistry methodologies, a phenothiazine compound was discovered that is a selective inhibitor of cyclooxygenase-2 and serves as a lead compound for a potentially novel series of anti-inflammatory compounds.

Stimulus dependence of non-steroidal antiinflammatory drug potency in a cellular assay of prostaglandin H synthase-2.

The prostaglandin H synthase-2 selective non-steroidal antiinflammatory drugs nimesulide, NS-398 (N-(2-cyclohexyloxy-4-nitrophenyl)methanesulfonamide), flosulide and SC 58125 (5-(4-fluorophenyl)-1-[4-(methylsulfonyl)phenyl]-3-(trifluoromethyl)-1H- pyrazole) as well as the non-selective non-steroidal antiinflammatory drugs indomethacin, meclofenamate and ibuprofen were compared in a WISH (human amnionic epithelial cell) cellular assay of prostaglandin H synthase-2 activity. Varying amounts of prostaglandin E2 were induced in WISH cells using either interleukin-1beta, tumor necrosis factor-alpha or phorbol 12-myristate 13-acetate, alone or in combination, or with okadaic acid as stimulants. The results from these studies demonstrated that under conditions which generate greater amounts of prostaglandin E2, the potency of both prostaglandin H synthase-2 selective and non-selective non-steroidal antiinflammatory drugs may be reduced. Dexamethasone, a transcriptional inhibitor of prostaglandin H synthase-2, also became progressively less effective in cells activated by combinations of stimuli or with okadaic acid. We conclude that decreases in potency under conditions of high levels of prostaglandin H synthase-2 expression and prostaglandin E2 production are observed equally with prostaglandin H synthase-2 selective and non-selective inhibitors.

Synthesis of indolylalkoxyiminoalkylcarboxylates as leukotriene biosynthesis inhibitors.

A series of substituted indolylalkoxyiminoalkylcarboxylates were found to be potent leukotriene biosynthesis inhibitors. The structure-activity relationships were investigated. Representative potent inhibitors identified were the quinolyl 3a (A-86885) and pyridyl 3b (A-86886) congeners with in vitro IC50s of 21 and 9 nM and in vivo leukotriene inhibition in the rat with oral ED50s of 0.9 and 1.7 mg/kg, respectively.

ABT-761 attenuates bronchoconstriction and pulmonary inflammation in rodents.

Our primary goal has been to discover leukotriene biosynthesis inhibitors with characteristics that are appropriate for use as clinical agents. The success of the use of zileuton in the treatment of asthma led us to explore further the use of the N-hydroxyurea class of 5-lipoxygenase inhibitors as longer-acting compounds with good lung penetration. A variety of in vitro and in vivo methods were used to evaluate a large number of compounds, from which ABT-761 [(R)-N-(3-(5-(4-fluorophenylmethyl)thien-2-yl)-1-methyl-2-pr opynyl)-N-hydroxyurea] was selected for study. ABT-761 exhibited potent and selective inhibition of leukotriene formation both in vitro and in vivo. More importantly, the compound potently inhibited antigen-induced bronchospasm in guinea pigs when given either prophylactically or therapeutically. In addition, ABT-761 was a potent inhibitor of eosinophil influx into the lungs of Brown Norway rats. These data provide added support for the role of leukotrienes in both bronchospasm and eosinophilic inflammation and characterize ABT-761 as a particularly potent inhibitor of leukotrienes formed in pulmonary tissues. These data combined with the excellent pharmacokinetic characteristics of the compound indicate its potential use in the treatment of leukotriene-dependent human disease.

Induction of prostaglandin H synthase-2 and tumor necrosis factor-alpha in human amnionic WISH cells by various stimuli occurs through distinct intracellular mechanisms.

These studies examined the signal transduction mechanisms by which prostaglandin (PG) E2 production can occur in human amnionic WISH cells in response to the stimuli okadaic acid, interleukin (IL)-1beta, tumor necrosis factor (TNF)-alpha, phorbol-12-myristate-13-acetate (PMA) or combinations of PMA with IL-1beta or TNF-alpha. We also investigated whether WISH cells are capable of producing TNF-alpha or IL-1beta in response to stimulation, because these cytokines can be produced in an autocrine fashion to perpetuate an inflammatory response. Our data indicate that the magnitude of PGE2 production induced by a given stimulus correlated temporally with the level of PGH synthase-2 (PGHS-2) protein. PMA or IL-1beta induced PGE2 production 2 to 4 hr after treatment, whereas the combination of these agents produced the most rapid induction 2 hr after treatment. Only okadaic acid induced the production of both PGE2 and TNF-alpha, after a lag of 12 to 18 hr. PGE2 production by all stimuli was inhibited by dexamethasone, the IL-1 receptor antagonist (IL-1ra), the specific PGHS-2 inhibitor NS-398 and the protein kinase inhibitor staurosporin. In contrast, TNF-alpha production in response to okadaic acid was inhibited by the TNF-converting enzyme inhibitor GI 129471 and staurosporin but was unaffected by either IL-1ra, dexamethasone or NS-398. We conclude that WISH cells are capable of producing bioactive proinflammatory mediators such as TNF-alpha and PGE2 through separable intracellular signal transduction mechanisms. The ability of IL-1ra to reduce PGE2 production caused by all stimuli used suggests an autocrine role for IL-1 in PGHS-2 induction in these cells.

Conversion of cyclooxygenase inhibitors into hydroxythiazole 5-lipoxygenase inhibitors.

Representative examples of NSAID cyclooxygenase inhibitors such as naproxen, ibufenac, ibuprofen, and butibufen have been transformed into 5-lipoxygenase inhibitors by replacement of the carboxylic acid moiety with a 4-hydroxythiazole group.

Nonsteroidal anti-inflammatory drugs as scaffolds for the design of 5-lipoxygenase inhibitors.

Representative nonsteroidal anti-inflammatory drug (NSAID) cyclooxygenase inhibitors such as ibuprofen, naproxen, and indomethacin were used as orally bioavailable scaffolds to design selective 5-lipoxygenase (5-LO) inhibitors. Replacement of the NSAID carboxylic acid group with a N-hydroxyurea group provided congeners with selective 5-LO inhibitory activity.

Leukotrienes do not regulate nitric oxide production in RAW 264.7 macrophages.

RAW 264.7 macrophages respond to lipopolysaccharide (LPS) and interferon-gamma (IFN-gamma) by producing large amounts of nitric oxide (NO) and prostaglandin E2 (PGE2), with maximal production 18-24 h after treatment. Following stimulation with the calcium inophore A23187, cultures of RAW cells also produce modest amounts of leukotrienes. However, the capacity of these cells to produce leukotrienes is transient, beginning 2 h after vehicle or LPS/IFN-gamma treatment, peaking by 4-6 h and absent by 8 h. A-79175, (R(+) N-[3-[5-(4-Fluorophenoxy)-2-furanyl]-1-methyl-2-propynyl]-N-hydroxyurea) a specific inhibitor of 5-lipoxygenase (5-LO), abolished leukotriene production by RAW cells in a dose-dependent, non-cytotoxic fashion while having no effect on PGE2 or NO production. By contrast, nordihydroguaiaretic acid (NDGA) inhibited production of leukotrienes, PGE2 and NO only at doses that were cytotoxic to the RAW cells. Exogenous leukotriene B4 (LTB4) had no effect on either NO or PGE2 production. An inhibitor of NO production, L-N-5-(1-iminoethyl) ornithine HCl (NIO) also did not affect leukotriene or PGE2 production, while dexamethasone blocked PGE2 and NO production, but did not affect leukotriene production in these cells. Taken collectively, these results indicate that there is no interaction between the pathways for leukotriene and nitric oxide production in RAW 264.7 macrophages.

Interleukin-1 beta induces cytosolic phospholipase A2 and prostaglandin H synthase in rheumatoid synovial fibroblasts. Evidence for their roles in the production of prostaglandin E2.

OBJECTIVE: In order to investigate potential regulatory mechanisms for the increased production of prostaglandin E2 (PGE2) in interleukin-1 beta (IL-1 beta)-stimulated rheumatoid synovial fibroblasts (RSF), this study examined the induction of phospholipase A2 (PLA2) and prostaglandin H synthase (PGHS) enzymes and the correlation of these events with PGE2 production in IL-1 beta-stimulated RSF. METHODS: Protein and messenger RNA (mRNA) levels of cytosolic PLA2 (cPLA2) and PGHS-2 enzymes in IL-1 beta-stimulated RSF were measured by Western and Northern blotting, respectively, using specific antisera and complementary DNA probes. Enzymatic activity of cPLA2 was determined in cell-free reaction mixtures utilizing mixed micelles of 14C-phosphatidylcholine and Triton X-100 as the substrate. PGE2 levels were quantitated using a commercial enzyme immunoassay kit. RESULTS: Incubation of RSF with IL-1 beta increased the mRNA and protein levels for the high molecular weight cPLA2 as well as for the mitogen/growth factor-responsive PGHS (PGHS-2). The IL-1 receptor antagonist completely abolished the induction of these two enzymes and the stimulation of PGE2 production by IL-1 beta in RSF. In contrast, levels of the other known forms of these enzymes, i.e., the 14-kd secretory group II PLA2 (sPLA2) and the constitutive form of PGHS (PGHS-1), were unaffected by IL-1 beta treatment. CONCLUSION: These are the first data to demonstrate the coordinate induction by IL-1 of cPLA2 and PGHS-2 in RSF. The time-course for the induction of these enzymes suggests that their increase contributes to the increased production of PGE2 in IL-1-treated RSF, and may help explain the capacity of RSF to produce large amounts of PGE2.

Expression of CSF-1, c-fms, and MCP-1 in the central nervous system of rats with experimental allergic encephalomyelitis.

We have examined the expression of factors associated with the growth, differentiation, and chemotaxis of cells of the monocyte/macrophage series in the central nervous system of Lewis rats sensitized to develop experimental allergic encephalomyelitis. CSF-1 mRNA increased significantly over that found in control animals (sensitized with OVA in CFA or CFA alone). The elevation in the levels of this growth factor commenced immediately before the onset of early clinical signs and peaked immediately before maximal clinical incidence of disease. Expression of CSF-1 message declined to base-line values with resolution of the disease process. CSF-1 protein was also detected in the central nervous system at the height of clinical disease. Expression of the receptor for CSF-1, the proto-oncogene c-fms, also paralleled the early disease process. Elevated levels of c-fms mRNA were detected immediately before the onset and peaked at the height of clinical signs of disease. In contrast to CSF-1 levels, elevated c-fms message expression persisted after resolution of the acute phase of experimental allergic encephalomyelitis. Levels of macrophage chemotactic factor-1 message were also elevated immediately before the onset of clinical signs, peaked with the height of clinical disease, and declined with resolution of the disease. Unlike CSF-1 or c-fms, no endogenous macrophage chemotactic factor-1 message was detected in control animals. Macrophage chemotactic factor-1 protein was demonstrated by Western blot in the central nervous system at the height of clinical disease. The results support the conclusion that expression of factors that specifically target cells of the monocyte/macrophage series are an important component of the disease process in experimental allergic encephalomyelitis.

Interleukin-1 beta induces cytosolic PLA2 in parallel with prostaglandin E2 in rheumatoid synovial fibroblasts.

We investigated the temporal relationship between the increase in enzymatic activity and protein of a high molecular weight (100 kDa), cytosolic PLA2 (cPLA2) in interleukin-1 beta (IL-1 beta)-treated rheumatoid synovial fibroblasts (RSF). Both of these responses increased according to a similar time-course which correlates with PGE2 production by these cells. In contrast, 14 kDa, secreted PLA2 (sPLA2), which was also produced by RSF, was not affected by IL-1 beta treatment. These findings support that an augmentation of CPLA2 activity, caused by an induction of cPLA2 protein, rather than sPLA2, is temporally associated with increased PGE2 production in IL-1 beta-treated RSF.

Synovial protein kinase C and its apparent insensitivity to interleukin-1.

Lapine synovial fibroblasts produce prostaglandin E2 (PGE2) and neutral metalloproteinases in response to phorbol 12-myristate 13-acetate (PMA), human recombinant interleukin-1 (hrIL-1) and, in an autocrine fashion, in response to partially purified preparations of their own cytokines known as cell-activating factors (CAF). Here we have examined the possible role of protein kinase C (PKC) in these responses. Whereas the 80-kDa substrate for PKC could not be detected in synovial fibroblasts, these cells contained a 35-kDa protein which fulfilled the criteria for qualifying as a specific substrate of PKC. Translocation assays based upon phosphorylation of the 35-kDa protein and Western blotting techniques allowed the movement of PKC from the cytosolic to the particulate fraction in response to PMA and CAF to be detected but not in response to 4 alpha-PMA or hrIL-1. Inhibitors of PKC suppressed synovial activation by PMA, partially blocked activation by CAF but had no effect on activation by hrIL-1. There thus appear to be PKC-dependent and PKC-independent routes to synovial cell activation. Our data suggest that IL-1 uses the latter, while CAF contains cytokines which utilize both routes.

Interleukin-1 beta stimulates cytosolic phospholipase A2 in rheumatoid synovial fibroblasts.

Phospholipase A2 (PLA2) activities in rheumatoid synovial fibroblasts (RSF) stimulated with interleukin-1 beta (IL-1 beta) were investigated. RSF incubated in the presence of IL-1 beta (120 pg/ml) for 18 h secreted 35 fold more PGE2 than did those incubated without IL-1 beta. IL-1 beta treatment did not increase the level of secretory PLA2 (sPLA2) activity or sPLA2 protein in the conditioned medium or subcellular fractions of lysed RSF. In contrast, the cell-associated PLA2 activity increased 3 to 4 fold in IL-1 beta stimulated RSF when compared with the control. The IL-1 beta stimulated, cell-associated PLA2 required submicromolar concentrations of calcium for activity, a characteristic consistent with the calcium sensitivity of cytosolic PLA2 (cPLA2) activity reported in other cell types, such as U937 cells. These findings demonstrate that an elevation in a cytosolic PLA2, rather than a sPLA2, is associated with increased PGE2 production in IL-1 beta stimulated RSF.