Characterization of human glutathione-dependent microsomal prostaglandin E synthase-1

University dissertation from Stockholm : Karolinska Institutet, Department of Medical Biochemistry and Biophysics

Abstract: Prostaglandins (PGs) are lipid mediators, which act as local hormones. PGs are formed in most calls and are synthesized de novo from membrane-released arachidonic acid (AA) upon cell activation. Prostaglandin H synthase (PGHS) -1 or 2, also referred to as COX-1 and COX-2, metabolize AA to PGH2, which is subsequently converted in a cell-specific manner by downstream enzymes to biologically active prostanoids, i.e. PGE2, PGD2, PGF2alpha, PGI2 or TXA2. PGHS-1 is constitutively expressed in many calls and is mainly involved in housekeeping functions, such as vascular homeostasis, whereas PGHS-2 can be induced by proinflammatory cytokines at sites of inflammation. Prostaglandin E synthase (PGES) specifically catalyzes the conversion of PGH2 to PGE2, which is a biologically potent prostaglandin involved in several pathological conditions; including pain, favor, inflammation and possibly some forms of cancers and neurodegenerative diseases. mPGES-1 was initially identified as a homologue to microsomal glutathione transferase-1 (MGST1) with 37% identity on the amino acid sequence level and referred to as MGST1-like 1 (MGST1-L1). Based on the properties of MGST1-L1, regarding size, amino acid sequence, hydropathy and membrane localization, the protein was identified as a member of the MAPEG-superfamily (membrane-associated proteins in eicosanoid and glutathione metabolism). The superfamily consists of 16- 18 kDa, integral membrane proteins with typical hydropathy profiles and diverse functions. The MAPEG family comprises six human members, which in addition to mPGES-1 are; 5-lipoxygenase activating protein (FLAP), leukotriene C4 synthase (LTC4S), MGST1, MGST2 and MGST3. MGST1 -2 and -3 are glutathione transferases as well as glutathione-dependent peroxidases, while FLAP and LTC4S are crucial for leukotriene biosynthesis. Human mPGES-1 was cloned and characterized as a 16 kDa, inducible GSH-dependent microsomal PGE synthase. Northern dot blot analysis of mPGES-1 mRNA demonstrated a low expression in most tissues, medium expression in reproductive organs and a high expression in two cancer cell lines (A549 and HeLa). A549 cells had been used earlier as a model system to study PGHS-2 induction by the proinflammatory cytokine IL-1beta and mPGES-1 was also induced by IL-1beta in these calls. A protein of similar size was detected in microsomes from sheep vesicular glands, which are known to contain a highly efficient microsomal PGES, indicating that mPGES-1 was the long-sought membrane bound PGES. Furthermore, a time study of PGHS-2 and mPGES-1 expression revealed a coordinate induction of these enzymes, which was correlated with increased PGES activity in the microsomal fraction. Tumor necrosis factor-alpha (TNF-alpha) also induced mPGES-1 in these cells and dexamethasone was found to counteract the effect of these cytokines on mPGES-1 induction. A method based on RP-HPLC and UV-detection was developed to efficiently quantify PGES activity. A small set of potential mPGES-1 inhibitors were tested and NS-398, Sulindac sulfide and LTC4 were found to inhibit PGES activity with IC50-values of 20 µm, 80 µm and 5 µm, respectively. The human mPGES-1 gene structure was investigated. The mPGES-1 gene span a region of approximately 15 kb, is divided into three exons, and is localized on chromosome 9q34.3. A 682 bp fragment directly upstream of the translation start site exhibited promoter activity when transfected in A549 calls. The putative promoter is GC-rich, lacks a TATA box at a functional site and contains numerous potential transcription factor binding-sites. Two GC-boxes, two tandem Barble-boxes and an aryl hydrocarbon response element were identified. The putative promoter region of mPGES1 was transcriptionally active and reporter constructs were regulated by IL-1beta and phenobarbital. The expression of mPGES-1 was investigated in synovial tissues from patients suffering from rheumatoid arthritis (RA). Primary synovial cells obtained from patients with RA were treated with IL-1beta or TNF-alpha. Both cytokines were found to induce mPGES-1 mRNA from low basal levels to maximum levels after 24 hours and the induction by IL-1beta was inhibited by dexamothasone in a dose-dependent manner. The protein expression of mPGES-1 was also induced by IL-1beta with a linear increase up to 72 h. In contrast, the PGHS-2 induction demonstrated an earlier peak expression (4-8 h). Furthermore, the protein expression of mPGES-1 was correlated with increased microsomal PGES activity. In these biochemical experiments any significant contribution of cytosolic PGES or other cytosolic or nonn-inducible membrane bound PGE syntheses was ruled out. A purification protocol for mPGES-1 was developed. Human mPGES-1 was expressed with a histidine tag in Eschericha coli, solubilized by Triton X-100 and purified by a combination of hydroxyapatite and immobilized metal affinity chromatography. mPGES-1 catalyzed a rapid GSH-dependent conversion of PGH2 to PGE2 (170 µmol/min mg). The enzyme, also displayed a high GSH-dependent activity against PGG2, forming 15hydroperoxy PGE2 (250 µmol/min mg). In addition, mPGES-1 possessed several other activities; glutathionedependent peroxidase activity towards cumene hydroperoxide, 5-HpETE and 15-hydroperoxy-PGE2, as well as conjugation of 1-chloro-2,4-dinitrobenzene (CDNB) to GSH. These activities likely reflect the relationship with other MAPEG enzymes. Two-dimensional crystals of purified mPGES-1 were obtained and a 10 A projection map was determined by electron crystallography. Hydrodynamic studies were also performed on the mPGES-1-Triton X-100 complex to investigate the oligomeric state of the protein. Electron crystallography and hydrodynamic studies independently demonstrated a trimeric organization of mPGES-1. Together with other studies published to date, mPGES-1 has been verified biologically as a drug target and the next stop in this validation process requires specific inhibitors to be tested in animal disease models.

  This dissertation MIGHT be available in PDF-format. Check this page to see if it is available for download.