Structure-based design of a new class of anti-inflammatory drugs: secretory phospholipase A(2) inhibitors, SPI.

Human non-pancreatic secretory phospholipase A(2) (hnps-PLA(2)) is a group IIA enzyme that is massively over-expressed in a variety of severe inflammatory diseases. The enzyme degrades membrane phospholipids and it has been hypothesized that this activity can lead to a loss of tissue and organ integrity and function. This report overviews efforts directed toward the identification and clinical evaluation of a new class of anti-inflammatory drugs that specifically targets and inhibits the catalytic site of this hydrolytic enzyme. To achieve this goal, structure-based drug design was applied to a lead molecule identified by random high volume screening. Through an iterative process consisting of X-ray structure determination followed by inhibitor modification and testing, the lead compound was improved more than 6000-fold. Detailed information learned from earlier X-ray studies of stable substrate mimics aided this inhibitor improvement process. The optimized drug candidate, LY315920/S-5920, is currently undergoing phase II clinical evaluation. The outcome of studies such as these will define with greater clarity the pathological role of hnps-PLA(2) in human inflammatory diseases.

Indole inhibitors of human nonpancreatic secretory phospholipase A2. 2. Indole-3-acetamides with additional functionality.

As reported in our previous paper, a series of indole-3-acetamides which possessed potency and selectivity as inhibitors of human nonpancreatic secretory phospholipase A2(hnps-PLA2) was developed. The design of these compounds was based on information derived from x-ray crystal structures determined for complexes between the enzyme and its inhibitors. We describe here the further implementation of this structure-based design strategy and continued SAR development to produce indole-3-acetamides with additional functionalities which provide increased interaction with important residues within the enzyme active site. These efforts led to inhibitors with substantially enhanced potency and selectivity.

Indole inhibitors of human nonpancreatic secretory phospholipase A2. 3. Indole-3-glyoxamides.

The preceding papers of this series detail the development of functionalized indole-3-acetamides as inhibitors of hnps-PLA2. We describe here the extension of the structure-activity relationship to include a series of indole-3-glyoxamide derivatives. Functionalized indole-3-glyoxamides with an acidic substituent appended to the 4- or 5-position of the indole ring were prepared and tested as inhibitors of hnps-PLA2. It was found that the indole-3-glyoxamides with a 4-oxyacetic acid substituent had optimal inhibitory activity. These inhibitors exhibited an improvement in potency over the best of the indole-3-acetamides, and LY315920 (6m) was selected for evaluation clinically as an hnps-PLA2 inhibitor.

Indole inhibitors of human nonpancreatic secretory phospholipase A2. 1. Indole-3-acetamides.

Phospholipases (PLAs) produce rate-limiting precursors in the biosynthesis of various types of biologically active lipids involved in inflammatory processes. Increased levels of human nonpancreatic secretory phospholipase A2 (hnps-PLA2) have been detected in several pathological conditions. An inhibitor of this enzyme could have therapeutic utility. A broad screening program was carried out to identify chemical structures which could inhibit hnps-PLA2. One of the lead compounds generated by the screening program was 5-methoxy-2-methyl-1-(phenylmethyl)-1H-indole-3-acetic acid (13a). We describe the syntheses, structure--activity relationships, and pharmacological activities of a series of indole-3-acetamides and related compounds derived from this lead. This SAR was undertaken with the aid of X-ray crystal structures of complexes between the inhibitors and hnps-PLA2 which were of great value in directing the SAR.

Recombinant human secretory phospholipase A2 released thromboxane from guinea pig bronchoalveolar lavage cells: in vitro and ex vivo evaluation of a novel secretory phospholipase A2 inhibitor.

The primary objective of this study was to develop a functional assay that could provide rapid and reliable information on some pharmacologic characteristics of a novel inhibitor of human secretory phospholipase A2 (sPLA2). Guinea pig bronchoalveolar lavage (BAL) fluid, containing predominantly macrophages, eosinophils and epithelial cells, released thromboxane A2, as measured by thromboxane B2, in a concentration-dependent manner on exposure to recombinant human sPLA2 (rh-sPLA2). Similarly, n-formyl-L-methionyl-L-leucyl-L-phenylalanine (n-F-Met-Leu-Phe) or arachidonic acid also released this lipid mediator. Indomethacin, a cyclooxygenase inhibitor, blocked synthesis of thromboxane in response to these agents. p-Bromophenacylbromide-inactivated rh-sPLA2 was substantially less effective than the untreated enzyme in causing release of thromboxane. LY311727 is a potent indole-derived inhibitor of the isolated enzyme (IC50 = 23 nM). Incubation of this agent with BAL cells, just before addition of rh-sPLA2, reduced release of thromboxane with an IC50 = 1.8 x 10(-6) M. Specificity for sPLA2 was demonstrated in that LY311727, unlike indomethacin, did not reduce synthesis and subsequent release of thromboxane A2 in response to arachidonic acid. Using this technique as a basis, we determined whether LY311727 could sufficiently accumulate in lung after i.v. administration to inhibit rh-sPLA2-induced thromboxane A2 release from BAL cells. The compound, given i.v. to guinea pigs 5 min before collecting BAL fluid, produced a dose-dependent inhibition of rh-sPLA2 with an ED50 = 50 mg/kg. Thus, new in vitro and ex vivo assays were developed that permit functional evaluation of novel sPLA2 inhibitors. These techniques should serve as secondary assays for evaluation of human sPLA2 inhibitory activity from a chemical series and in addition provide initial data related to metabolic stability and distribution to the lung.

Transgenic model for the discovery of novel human secretory non-pancreatic phospholipase A2 inhibitors.

Transgenic mice were created which overexpress human secretory non-pancreatic phospholipase A2 (sPLA2) pansomatically as a potential disease and drug-testing model. The mice were produced using a DNA construct in which the inducible mouse metallothionein gene promoter drives expression of a human sPLA2 minigene. High levels of sPLA2 were detected in several tissues by immunofluorescence localization. Expression in the testes caused hypospermia and male infertility. Circulating catalytically active sPLA2 could be induced to levels observed in patients undergoing a systemic inflammatory response but had no detectable effect on the mice. Therefore, these results suggest that sPLA2 hyperphospholipasemia alone may have only limited pathophysiological consequences. We further show that 3-[3-acetamide-1-benzyl-2-ethylindolyl-5-oxy]propane phosphonic acid LY311727), a potent new inhibitor of phospholipase A2 catalysis developed by our group, dramatically suppresses the circulating enzyme activity in these animals whereas 3-[3-acetamide-1-benzyl-2-propylindolyl-5-oxy]propane phosphonic acid (LY314024), a substantially less potent LY311727 analog, is without effect. These later results thus motivate the further development of this compound as a potential new therapeutic agent and valuable research tool.

Structure-based design of the first potent and selective inhibitor of human non-pancreatic secretory phospholipase A2.

A lead compound obtained from a high volume human non-pancreatic secretory phospholipase A2 (hnps-PLA2) screen has been developed into a potent inhibitor using detailed structural knowledge of inhibitor binding to the enzyme active site. Four crystal structures of hnps-PLA2 complexed with a series of increasingly potent indole inhibitors were determined and used as the structural basis for both understanding this binding and providing valuable insights for further development. The application of structure-based drug design has made possible improvements in the binding of this screening lead to the enzyme by nearly three orders of magnitude. Furthermore, the optimized structure (LY311727) displayed 1,500-fold selectivity when assayed against porcine pancreatic s-PLA2.

Characterization of the contractile effects of human recombinant nonpancreatic secretory phospholipase A2 (PLA2) and other PLA2s on guinea pig lung pleural strips.

Contractile activities of nPLA2, pPLA2 and hPLA2 were characterized on pleural strips of guinea pig lung. The rank order of potency for these PLA2s was nPLA2 > pPLA2 > hPLA2. The concentration-related contractions induced by nPLA2 (0.0002-0.67 micrograms/ml), pPLA2 (0.006-20 micrograms/ml) and hPLA2 (0.1-30 micrograms/ml) appear to be mediated primarily by the formation of cyclooxygenase products and to a lesser extent by 5-lipoxygenase products, as revealed by experiments using indomethacin and BW A4C. To further support a PLA2-related mechanism, the selectivity and inhibitory effects of two irreversible PLA2 inhibitors, parabromophenacyl bromide (pBPB) and manoalogue, were evaluated against the contractile responses induced by each PLA2. Various concentrations of manoalogue and pBPB were incubated with individual PLA2s for 24 hr before initiating experiments. Both agents suppressed each PLA2-induced contractile activity in a concentration-related manner. The inhibitory effects of pBPB were similar at the highest concentration, whereas manoalogue was more effective in blocking contractions induced by pPLA2 and hPLA2. Conversely, methylated manoalogue, an inactive analog, failed to reduce the PLA2-induced contractions. These results demonstrate that hPLA2 has the ability to catalytically induce the release of arachidonic acid and the formation of proinflammatory eicosanoids that contract the pleural strips. This tissue bath preparation may be a useful model for the evaluation of novel PLA2 inhibitors as potentially useful therapeutic agents.

Evidence that phospholipase A2 (PLA2) inhibitors can selectively block PLA2-mediated contractions of guinea pig lung pleural strips.

We have demonstrated previously that porcine pancreatic phospholipase A2 (PLA2)-induced contractions of guinea pig lung pleural strips can be abated by inhibitors of cyclooxygenase and 5-lipoxygenase pathways, suggesting the liberation of arachidonic acid. To validate further the involvement of a PLA2-related mechanism, the effects of three known inhibitors of PLA2 were evaluated. Manoalogue, an irreversible inhibitor of PLA2, parabromophenacyl bromide, an irreversible, active site directed inhibitor and a transition-state analog, a competitive inhibitor of PLA2 were used. Transition-state analog (3-30 microM), added to the tissues for 30 min before PLA2, shifted the PLA2 curves to the right in a concentration-related manner. In contrast, the reported inactive enantiomer of transition-state analog failed to alter the PLA2 curves. Manoalogue and para-bromophenacyl bromide, at concentrations up to 40- and 50-fold higher than the enzyme concentration, respectively, were incubated in Krebs' buffer with the enzyme for 24 hr before challenging the tissues. Under these conditions, the PLA2-induced contractions were suppressed markedly. In contrast, neither reduced nor methylated manoalogue, incubated at a 20-fold molar excess with PLA2 for 24 hr, suppressed the maximal PLA2-induced responses. These results demonstrate that inhibitors of secretory PLA2, acting by different mechanisms, can alter the contractile responses induced by PLA2 on pleural strips from guinea pig lung.

Relative potencies of 5-lipoxygenase inhibitors on antigen-induced contractions of guinea pig tracheal strips.

A quantitative method to assess relative potencies (IC50) of 5-lipoxygenase (5-LO) enzyme inhibitors was established in antigen-induced contractions of tracheas isolated from actively sensitized guinea pigs (Schultz-Dale model). The relative potencies of four purported 5-LO inhibitors determined in this tissue assay were compared with those from a crude enzyme preparation isolated from guinea pig neutrophils. All compounds suppressed ovalbumin (OA)-induced tracheal contractions in a concentration-related manner in the presence of indomethacin and pyrilamine. IC50 Values, determined from the percent inhibition values obtained from responses at 30 ng/mL OA of these compounds ranged from 0.56-15 microM. A similar rank order of potency for inhibition of 5-HETE formation from a crude enzyme preparation was observed. This suggested that these agents had a common mechanism of action in the two assay systems and further validated the IC50 values determined in trachea assay. LY171883, an LTD4/LTE4 receptor antagonist, also suppressed OA-induced contractions concentration dependently with an IC50 of 4.9 microM determined by this method. LTD4 concentration-response curves were not altered by any of the four 5-LO inhibitors, ruling out the possibility that these agents were acting as LT receptor antagonists. Results of this study demonstrated that relative potencies of 5-LO inhibitors can be quantitatively assessed using this airway tissue model, which helps in identifying potential therapeutic agents for asthma.

Kinetic characterization of phospholipase A2 modified by manoalogue.

Manoalogue, a synthetic analogue of the sea sponge-derived manoalide, has been previously shown to partially inactivate the phospholipase A2 from cobra venom (Reynolds, L. J., Morgan, B. P., Hite, E. D., Mihelich, E. D., & Dennis, E. A. (1988) J. Am. Chem. Soc. 110, 5172) by reacting with enzyme lysine residues. In the present study, the inactivation of the phospholipases A2 from pig pancreas, bee venom, and cobra (Naja naja naja) venom by manoalogue was studied in detail. Manoalogue-treated enzymes were examined in the scooting mode on vesicles of 1,2-dimyristoyl-sn-glycero-3-phosphomethanol. Here the native enzymes bound irreversibly to the vesicles and hydrolyzed all of the phospholipids in the outer monolayer without leaving the surface of the interface. All three manoalogue-treated enzymes showed reduced catalytic turnover for substrate hydrolysis in the scooting mode, and the modified enzymes did not hop from one vesicle to another. Thus, inactivation by manoalogue is not due to the decrease in the fraction of enzyme bound to the substrate interface. This result was also confirmed by fluorescence studies that directly monitored the binding of phospholipase A2 to vesicles. A chemically modified form of the pig pancreatic phospholipase A2 in which all of the lysine epsilon-amino groups have been amidinated was not inactivated by manoalogue, indicating that the modification of lysine residues and not the amino-terminus is required for the inactivation. Several studies indicated that the manoalogue-modified enzymes contain a functional active site. For example, studies that monitored the protection by ligands of the active site from attack by a alkylating agent showed that manoalogue-modified pig phospholipase A2 was capable of binding calcium, a substrate analogue, lipolysis products, and a competitive inhibitor. Furthermore, relative to native enzymes, manoalogue-modified enzymes retained significantly higher catalytic activities when acting on water-soluble substrates than when acting on vesicles in the scooting mode. Intact manoalogue had no affinity for the catalytic site on the enzyme as it did not inhibit the enzyme in the scooting mode and it did not protect the active site from alkylation. Pig pancreatic phospholipase A2 bound to micelles of 2-hexadecyl-sn-glycero-3-phosphocholine was resistant to inactivation by manoalogue, suggesting that the modification of lysine residues on the interfacial recognition surface of the enzyme was required for inactivation.(ABSTRACT TRUNCATED AT 400 WORDS)

Inhibition of venom phospholipases A2 by manoalide and manoalogue. Stoichiometry of incorporation.

We have previously described the irreversible inhibition of cobra venom phospholipase A2 (PLA2) by the marine natural product manoalide (MLD) (Lombardo, D., and Dennis, E. A. (1985) J. Biol. Chem. 260, 7234-7240) and by its synthetic analog, manoalogue (MLG) (Reynolds L. J., Morgan, B. P., Hite, G. A., Mihelich, E. D., and Dennis, E. A. (1988) J. Am. Chem. Soc. 110, 5172-5177). We have now made a direct comparison of the action of these two inhibitors on PLA2 from cobra, bee, and rattlesnake venoms and have found that MLG behaves kinetically similarly to MLD in all cases with only minor differences. The time courses of inactivation differ significantly between the three enzymes, however, with the inactivation of bee and rattlesnake PLAs2, occurring much faster than does the inactivation of the cobra venom enzyme. The enzymes also differ in their sensitivity to the presence of Ca2+ during the inactivation. Of the three enzymes, the most Ca(2+)-sensitive is the rattlesnake enzyme, which shows a much faster rate of inactivation in the presence of Ca2+ than in the presence of EGTA. However, the same rate of inactivation was also observed when the inhibitor Ba2+ was substituted for Ca2+, indicating that catalytic activity is not required for inactivation of the enzyme. To probe the mechanism of inactivation and to determine the stoichiometry of incorporation, we have synthesized 3H-labeled MLG and have found that inactivation of cobra PLA2 is accompanied by an incorporation of 3.8 mol of [3H]MLG/mol of enzyme. The same amount of 3H incorporation was observed when p-bromophenacyl bromide-inactivated PLA2 was incubated with [3H]MLG, again indicating that catalytic activity is not required for the reaction of PLA2 with MLG. All together, these results suggest that MLD and MLG are not suicide inhibitors of PLA2. A portion of the incorporated radioactivity was acid-labile, and dialysis of the radiolabeled PLA2 under acidic conditions resulted in a loss of about one-third of the enzyme-associated radioactivity, leaving 2.4 mol of [3H]MLG/mol of PLA2. In previous studies, amino acid analysis, which also included acid treatment, indicated that MLG-modified cobra phospholipase A2 contained 2.8 mol of Lys less than the native enzyme. Thus, 1 mol of [3H]MLG is incorporated per mol of Lys lost. The implications of this 1:1 stoichiometry of MLG to Lys on the mechanism of reaction of these inhibitors is discussed.

Characterization of LY233569 on 5-lipoxygenase and reperfusion injury of ischemic myocardium.

LY233569 produced concentration-dependent inhibition of isolated guinea pig 5-lipoxygenase (5-LPO) and 5-LPO activity of human polymorphonuclear leukocytes in vitro; IC50 values were 0.4 and 0.1 microM, respectively. LY233569 also inhibited (IC50 approximately 1.8 microM) zymosan-stimulated production of leukotriene B4 in canine whole blood but had little or no concomitant effect on the production of thromboxane B2. Concentrations of LY233569 as high as 10 microM did not inhibit production or scavenge superoxide from activated human neutrophils. In normal anesthetized dogs, infusion of LY233569 (0.11 mg/kg/min, i.v.) for 6 hr produced persistent inhibition (approximately 80%) of leukotriene B4 production in blood challenged ex vivo with zymosan; the plasma concentration (approximately 4 microM) of LY233569 was consistent with in vitro data illustrating selective and maximal inhibition of 5-LPO. In subsequent experiments, myocardial infarct size was measured after 1 hr of occlusion of the circumflex coronary artery and 5 hr of reperfusion. Continuous infusion of LY233569 (0.11 mg/kg/min, i.v.) had little or no effect on base-line systolic arterial pressure, cardiac rate and the pressure rate product when compared with time-related changes observed in control dogs. LY233569 infusion also did not alter the degree of ST-segment deviation or the intensity and duration of cardiac arrhythmias associated with coronary artery occlusion and reperfusion. Resultant myocardial infarct sizes were 45 +/- 5% of the left ventricle placed at risk in control dogs and 43 +/- 4% in dogs given LY233569. Myeloperoxidase activity of infarcted myocardium did not differ between groups.(ABSTRACT TRUNCATED AT 250 WORDS)

The inhibition of phospholipase A2 by manoalide and manoalide analogues.

Manoalide, a natural product from sponge, displays anti-inflammatory activity in vivo. Previous work has shown that manoalide is also a potent covalent inhibitor of the extracellular phospholipase A2 from cobra venom and that the inhibition correlated with a pH-dependent change in manoalide (Lombardo and Dennis (1985) J. Biol. Chem. 260, 7234-7240). Manoalide contains two rings and the opening of either would produce an alpha,beta-unsaturated aldehyde. The cobra venom phospholipase A2 may be able to catalyze the opening or isomerization of one of these rings, raising the possibility that manoalide is acting as a suicide substrate. To ascertain the role of the gamma-lactone ring in the inhibition, we have now investigated a synthetic manoalide analogue, 3(cis,cis-7,10)-hexadecadienyl-4-hydroxy-2-butenolide (HDHB) which contains only the alpha,beta-unsaturated gamma-lactone ring. We have found that the closed and open forms are in rapid equilibrium between pH 4 and 9 with the cyclic form being preferred at acidic pH values and the open cis form preferred at pH 9.5. When the pH is raised above 12, the alpha,beta double bond isomerizes to form trans-HDHB. Once the trans compound is formed, it is stable at all pH values and does not recyclize to the gamma-lactone ring. The observed pKa of 7.7 found for the inhibition of manoalide agrees well with the transition of the closed to the cis form of the gamma-lactone ring. Kinetic experiments with the HDHB compound show that under conditions in which the cis and closed form of the inhibitor are present in equal molar ratios, HDHB is not an irreversible inhibitor, but reversibly competes with substrate. However, the kinetics of this inhibition are complex and do not follow either pure competitive or non-competitive inhibition. The trans-HDHB exhibits similar complex kinetic but is several times more potent. The distinct differences between the behavior of manoalide and HDHB clearly indicate that while the gamma-lactone ring may play an important role in manoalide inhibition, it alone does not produce irreversible inhibition.