Read an Excerpt

Anti-Inflammatory Drug Discovery

The Royal Society of Chemistry

Microsomal Prostaglandin E2 Synthase-1

ANDREAS KOEBERLE AND OLIVER WERZ

Prostaglandins (PGs) are pivotal bioactive lipid mediators that mediate inflammatory reactions but also contribute to various homeostatic biological processes. The cyclooxygenase (COX) isoenzymes namely COX-1 (constitutively expressed in numerous cell types and thought to provide PGs mainly for physiological functions) and COX-2 (an inducible isoform in inflammatory cells, primarily producing PGs relevant for inflammation, fever and pain) initiate PG biosynthesis from arachidonic acid. The non-steroidal anti-inflammatory drugs (NSAIDs) or COX-2 selective drugs (coxibs) inhibit COX activities and thus exert potent anti-inflammatory and analgesic effects but may also cause severe side-effects such as gastrointestinal and renal complications. The latter effects have been ascribed to the general suppression of constitutively formed prostanoids such as COX-1-derived cytoprotective PGE2 and prostacyclin (PGI2) in gastroduodenal epithelium. Even though coxibs have an improved gastrointestinal tolerance, recent clinical studies revealed small but significantly increased risks for cardiovascular events such as myocardial infarction, stroke, systemic and pulmonary hypertension, congestive heart failure and sudden cardiac death. Apparently, the imbalance between the anti-thrombotic and vasodilatory PGI2 on one hand and the pro-thrombotic TxA2 on the other are responsible for those side-effects. Since side-effects are particularly problematic in the therapy of chronic pathologies, such as rheumatoid arthritis, requiring long-term drug application, there is a strong need for novel potent and safe anti-inflammatory drugs lacking such toxicity.

1.1 Function of PGE2 as Bioactive Mediator

Among the PGs, PGE2 is the most prominent mediator in inflammation, fever and pain but also has physiological functions in the gastrointestinal tract, the kidney and in the immune and the central nervous system. PGE2 mediates its bioactivities essentially by four G-protein coupled PGE2 receptor subtypes (EP1–EP4) in diverse tissues, supported by experiments with knockout mice deficient (-/-) in EP receptor subtypes and selective EP receptor antagonists. For example, deletion of respective EP receptor subtypes significantly reduced exudate formation in carrageenan-induced mouse pleurisy (EP2 and EP3), diminished arachidonic acid-induced cutaneous inflammation (EP3) and inflammation and joint destruction in collagen (EP4), as well as collagen-antibody-induced arthritis (EP4 and EP2). In lipopolysaccharide-challenged mice, activation of the hypothalamo-pituitary-adrenal axis and systemic illness including the febrile response in response to PGE2 was mainly ascribed to the EP3 receptor (although other subtypes are also involved). Regarding pain sensation, all four EP receptors seemingly contribute, but the EP-receptor subtype involved depends on the nociceptive stimulus and/or the pre-treatment of the animals. For example, the EP1 receptor mediates the acute pain response in the acetic acid-induced writhing test in mice, whereas EP3 is the critical receptor for LPS-induced hyperalgesia. However, PGE2 also exerts immunosuppressive effects contributing to the resolution of inflammation, mediates protection of gastrointestinal mucosa and regulation of natriuresis and regulates renal blood flow and blood pressure.

1.2 PGE2 Biosynthesis by mPGES-1

PGE2 synthases (PGES) perform the terminal step in the biosynthesis of PGE2, that is, the isomerization of the COX-derived peroxide PGH2 to PGE2. Three terminal isoforms of PGE2 synthases have been cloned and characterized. Co-transfection of COX-1 and -2 with PGES isoenzymes in mammalian cells suggests that molecular interactions may cause preferential functional coupling. While the constitutively expressed cytosolic PGE2 synthase (cPGES) was found to be coupled to COX-1, mPGES-1 is predominantly involved in COX-2-mediated PGE2 production. mPGES-2 is also constitutively expressed, uses PGH2 supplied by COX-1 and COX-2 and contributes to basal PGE2 synthesis but its functional role in physiology and patho-physiology is still elusive.

1.3 Structure and Biochemical Properties of mPGES-1

Human mPGES-1 (16 kDa) is a member of the membrane-associated proteins involved in eicosanoid and glutathione metabolism (MAPEG) family and is characterized by a high turnover number for PGH2 (kcat = 50 s-1). The enzyme is membrane-bound and localized to the microsomal fraction after subcellular fractionation. Km values of 14–160 µM and 710–750 µM were reported for PGH2 and glutathione, respectively.

The homotrimeric structure of mPGES-1 was recently determined at low resolution by electron crystallography. The monomers consist of four transmembrane helices (TM1-4) and enclose an inner core with a funnel-shaped opening towards the cytoplasm. The essential cofactor glutathione is bound in a U-shaped conformation at the interface between the subunits. Glutathione was proposed to attack the peroxide of PGH2 in the active site via its thiol-group during the catalytic cycle. Mutation studies suggest Arg126, which is located near the thiol of glutathione, as catalytic residue and Thr-131, Leu-135 and Ala-138 (all are located within the transmembrane-helix 4) as gate keepers for the active site. The smaller size of these gate keepers in the human enzyme compared to those in the rat enzyme may explain the failure of several mPGES-1 inhibitors to inhibit rat mPGES-1. For the related MAPEG enzyme LTC4 synthase, conformational changes from a closed to an open conformation are required for substrate entry into the active site, and a comparable mechanism might also account for mPGES-1. Molecular dynamic simulations strengthened by site-directed mutagenesis and subsequent hybridization suggest only one PGH2 substrate pocket of the trimer being occupied at the same time during the reaction cycle. While the low-resolution crystal structure of mPGES-1 seems to represent the closed confirmation, Hamza et al. proposed an mPGES-1 model in the open confirmation based on the crystal structures of MGST1 and ba3-cytochrome c oxidase.

1.4 Regulation of mPGES-1 Expression

Pro-inflammatory stimuli such as interleukin-1β, tumour necrosis factor-[apha] or lipopolysaccharide strongly induce expression of both mPGES-1 and COX-2 in a variety of tissues and cell-types (including human lung carcinoma A549 cells, macrophages, endothelial cells and others), whereas glucocorticoids reverse their up-regulation. However, differences occur in the kinetics of mPGES-1 and COX-2 expression.

Thus, mPGES-1 is constitutively expressed in diverse tissues at a low level, for example in seminal vesicles, ovary, kidney, male reproductive organs, placenta, lung, spleen and gastric mucosa of mice and rats, but not in other tissues such as heart, liver and brain. The expression of mPGES-1 is markedly induced in inflamed rodent tissue and brain under diverse pathological conditions including inflammation, fever, pain and seizure. In humans, mPGES-1 was found to be up-regulated among others in arthritic synovial tissue, in the cartilage and chondrocytes of osteoarthritic patients, in inflamed intestinal mucosa from patients with inflammatory bowel disease, in atherosclerotic carotid plaques, in Alzheimer's disease tissues, in heart tissue after acute myocardial infarction and in liver tissue from patients with hepatitis. Thus, mPGES-1 seems to play a putative role in various diseases related to inflammation.

Increasing experimental evidence implies that PGE2 also regulates critical steps in tumourigenesis by stimulating cell proliferation and angiogenesis, preventing apoptosis and inducing cancer cell migration. In fact, elevated expression of mPGES-1 and increased PGE2 levels are characteristic for many human tumours and cancer cell lines derived from colon, intestine, stomach, oesophagus, larynx, lung, liver, pancreas, breast, ovary, squamous epithelium and brain.

1.5 Redirection of the mPGES-1 Substrate PGH2 Due to the Action of mPGES-1 Inhibitors

Selective inhibition of PGE2 formation by pharmacological interference with mPGES-1 as therapeutic strategy has been questioned because of a redirection of the substrate PGH2 to other prostanoid synthases leading to an increased formation of PGI2, PGF2α, TxB2 or PGD2. The redirection pattern strongly depends on the cell type and assay conditions. In mPGES-1-deficient mice for example, the above-mentioned prostanoids (PGI2, PGF2α, TxB2 or PGD2) are increased in the stomach but not in other tissues. An increased biosynthesis of thromboxanes is associated with cardiovascular diseases such as myocardial ischemia or heart failure. However, the redirection of prostanoids might also be beneficial in some cases because of their inflammation-resolving and cytoprotective effects. Taken together, the clinical safety of mPGES-1 inhibitors is not readily known so far.

1.6 Determination of mPGES-1 Activity

In general, two different types of assays for analysis of mPGES-1 activity can be distinguished: cell-free assays and cell-based test systems. Cell-free mPGES1 activity can be assessed in microsomal preparations of cells highly expressing mPGES-1 such as interleukin-1β-stimulated human lung carcinoma A549 or human cervix carcinoma HeLa cells, but also cell lines transfected with recombinant rodent or human mPGES-1 were used as source of mPGES-1. PGH2 (1–20 µM) is added to a reaction mix containing glutathione (2.5 mM) and microsomal preparations of mPGES-1-expressing cells. After 0.5–5 min on ice or at room temperature, the reaction is stopped by converting the remaining PGH2 to PGF2α using mild reducing agents such as SnCl2 or FeCl2. Formed PGE2 is quantified by either EIA or RP-HPLC combined with UV or radiometric detection (when radioactively labelled PGH2 is supplied).

Determination of mPGES-1 activity in cell-based assays is less convenient. Test systems are based on isolated cells or whole blood stimulated with lipopolysaccharide (LPS) for 5 to 24 h. PGE2 is quantified either directly by EIA or after separation of PGE2 by RP-HPLC prior to immunological detection. Incubation times exceeding 24 h increase the contribution of mPGES-1 to total PGE2 synthesis, which is efficiently inhibited by selective mPGES-1 inhibitors (e.g. MF-63, 70–80% inhibition). Incubation times <5 h may minimize effects of the test compounds on gene expression, which results, however, in a higher portion of mPGES-1-independent PGE2 synthesis. Diverse mPGES-1 inhibitors such as MK-886 (30 µM) and many others maximally inhibited PGE2 formation by 40–60% under these experimental conditions. The remaining PGE2 formation was suggested to derive from constitutively expressed PGE2 synthases such as cPGES and mPGES-2.

1.7 mPGES-1 in Disease and Homeostasis – Results from KO Studies

The knockout of mPGES-1 in mice does not result in phenotypic, behavioural or histological differences. Several excellent reviews have summarized the role of mPGES-1 in inflammation, neurologic diseases, cardiovascular disease and tumourigenesis.

1.7.1 Inflammation, Fever and Pain

Knockout studies in mice have revealed a role of mPGES-1 for the induction and progression of arthritis and other inflammatory diseases, hyperalgesia and the febrile response. The role of mPGES-1 in nociception is still under discussion due to conflicting observations. Thus, mPGES-1-deficient mice have reduced pain sensation after intra-peritoneal injection of acetic acid, during collagen-induced arthritis and in a neuropathic pain model. However, knockout of mPGES-1 was without effect on the nociceptive behaviour in the formalin test and during zymosan-evoked mechanical hyperalgesia. Since COX-inhibitors were effective under the same experimental conditions and the mPGES-1 knockout reduced PGE2 levels in the spinal cord, the authors explained the differences by a redirection of PGH2 to other prostanoids (PGD2, PGF2α and 6-keto PGF1α).

1.7.2 Neurological Diseases

mPGES-1-deficient mice show a reduced severity of infarction, edema and apoptotic cell death in the cortex after transient focal ischemia, of symptoms during experimental autoimmune encephalomyelitis (a mouse model for multiple sclerosis) and of neuronal loss in response to kainaic acid. However, priming of mPGES-1-deficient mice by intra-thecal injection of LPS suggests a role of mPGES-1 also for resolving neuroinflammation.

1.7.3 Cancer

mPGES-1 is overexpressed in various tumours. Knockout of mPGES-1 reduced cell proliferation, migration and invasion ex vivo and in mouse tumour xenocraft models. Deletion of mPGES-1 counteracts azoxymethane-induced preneoplastic aberrant crypt foci formation and colorectal carcinogenesis as well as intestinal cancer growth in APC-mutant mice. However, another study using APC-mutant mice found an increased tumour formation in mPGES-1-deficent mice which is not readily understood. Moreover, the knockdown of mPGES-1 prevented apoptosis and thus induced tumourigenesis in nude mice injected with human glioblastoma cells. The growth and angiogenesis of endometrial implants is also increased in mPGES-1-deficient mice. Taken together, mPGES-1 seems to have both anti- and pro-tumoural effects depending on the experimental model.

1.7.4 Renal and Cardiovascular System

The knockout of mPGES-1 in mice increases neither thrombogenesis nor blood pressure at standard diet in contrast to a knockout of COX-2. However, the constitutive expression of mPGES-1 along the nephron and collecting duct suggests a physiological role for water and salt resorption. In fact, mPGES-1-deficient mice were a?icted with hypertension, impaired natriuretic responses and worsened cardiac parameters when challenged with a high salt-diet, angiotensin II, aldosterone or DOCA-salt. In contrast, the blood pressure was not increased at high salt diet in a study published by Cheng et al. Moreover, knockdown of mPGES-1 resulted in a detrimental left ventricular remodelling after myocardial infarction but was not associated with an increased pulmonary edema or mortality. Deletion of mPGES-1 reduced plaque formation and increased plaque stability in a mouse atherosclerosis model, reduced cisplatin-induced nephrotoxicity and retarded disease progression in a mouse model of renal mass reduction. Taken together, adverse effects of mPGES-1 inhibition might occur when risk factors are present such as chronic salt loading or cardial ischemic injury.

---

Excerpted from Anti-Inflammatory Drug Discovery by Jeremy I. Levin, Stefan Laufer. Copyright © 2012 The Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---