Role of platelet activating factor acetylhydrolase in megakaryocytes and yeast

Platelet activating factor (PAF) is a synthesized inflammatory phospholipid that signals through a seven-transmembrane G-protein receptor. Oxidation of phospholipids with sn-2 unsaturated fatty acyl residues can fragment the esterified chain forming numerous phospholipids; some are toxic and others are PAF-like lipids which can signal through the PAF receptor. Group VII phospholipases (PAF acetylhydrolases, PAF-AH) are Ca++-independent esterases that specifically hydrolyze short sn-2 residues, including the acetyl residue of PAF. We have observed that megakaryocytes derived from human umbilical cord blood CD34+ stem cells have increased PAF-AH activity over differentiation. Inhibition of PAF-AH activity in megakaryocytes with Pefabloc SC led to increased formation of PAF-like lipids that signaled through endogenous PAF receptors. Transient morphology changes were observed in megakaryocytes treated with Pefabloc or carbamyl-PAF (cPAF), a stable form of PAR These data suggest that PAF-AHs are important for controlling aberrant PAF or PAF-like lipid formation in megakaryocytes, and that dysregulation could lead to altered pro-platelet formation or function. The fission yeast Schizosaccharomyces pombe genome contains a putative gene, SPBC106.11c, homologous to group VII hydrolases that retains a lipase motif and residues that form the esterase catalytic triad. An active PAF-AH enzyme in an organism that cannot generate PAF may instead protect against membrane oxidation and death. We cloned this SPBC106.11c locus and expressed it in distantly related Saccharomyces cerevisiae, which lacks genes related by sequence, to find it encodes a functional phospholipase we named Plg7p. Ala substitution of the conserved Ser257 in the GXSXG lipase motif and two serine-directed inhibitors of plasma PAF acetylhydrolase, Pefabloc and methyl arachidonyl fluorophosphonate, suppressed Plg7p activity. S. pombe ?plg7 had no overt phenotype, but revealed a second Ca++-independent phospholipase activity not inhibited by the Ser-directed reagents. S. cerevisiae supplemented with polyunsaturated fatty acids and exposed to environmental Cu+ resulted in oxidized membrane lipids, and reduced viability. Expression of human type II PAF acetylhydrolase or Plg7p enhanced viability in this model of oxidative death. We conclude an organism distantly related to mammals expresses a phospholipase that is able to reduce oxidative death, suggesting this is the original role of the group VII phospholipase family.

THE ROLE OF PLATELET ACTIV A TINO FACTOR ACETYLHYDROLASE IN MEOAKARYOCYTES AND YEAST by Jason Marc Foulks A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy In Experimental Pathology Department of Pathology The University of Utah May 2007 Copyright © Jason Marc Foulks 2007 All Rights Reserved THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a dissertation submitted by Jason Marc Foulks This dissertation has been read by each member of the following supervisory comnlittee and by majority vote has been found to be satisfactory. ~(;' ~ tJ~ Andrew Weyrich 25 El" abeth Leibold David Stillrilan THE UNIVERSITY OF UTAH GRADUATE SCHOOL FIN AL READING APPROVAL To the Graduate Council of the University of Utah: I have read the dissertation of _____J_ as_o_n_M_a_rc_F_o_u_l_k_s _____ in its final fonn and have found that (1) its format, citations, and bibliographic style are consistent and acceptable~ (2) its illustrative materials including figures, tables, and charts are in place; and (3) the final nlanuscript is satisfactory to the supervisory comnlittee and is ready for subnlission to The Graduate SchooL Approved for the Major Departnlent I?~--- Peter E. Jensen Chair/Dean Approved for the Graduate Council ~,~ <;: CS::>~- _ . David S. Chapman Dean of The Graduate School ABSTRACT Platelet activating factor (P AF) is a synthesized inflammatory phospholipid that signals through a seven-transmembrane G-protein receptor. Oxidation of phospholipids with sn-2 unsaturated fatty acyl residues can fragment the esterified chain forming numerous phospholipids; some are toxic and others are PAF-like lipids which can signal through the P AF receptor. Group VII phospholipases (P AF acetylhydrolases, P AF -AH) are Ca++-independent esterases that specifically hydrolyze short sn-2 residues, including the acetyl residue of P AF. We have observed that megakaryocytes derived from human umbilical cord blood CD34+ stem cells have increased PAF-AH activity over differentiation. Inhibition ofPAF-AH activity in megakaryocytes with Pefabloc SC led to increased formation of PAF-like lipids that signaled through endogenous PAF receptors. Transient morphology changes were observed in megakaryocytes treated with Pefabloc or carbamyl-PAF (cPAF), a stable form of PAF. These data suggest that PAF-AHs are important for controlling aberrant PAF or PAF-like lipid fomlation in megakaryocytes, and that dysregulation could lead to altered proplate1et formation or function. The fission yeast Schizosaccharomyces pombe genome contains a putative gene, SPBCI06.l1c, homologous to group VII hydrolases that retains a lipase motif and residues that form the esterase catalytic triad. An active PAF-AH enzyme in an organism that cannot generate P AF may instead protect against membrane oxidation and death. We cloned this SPBCI06.11c locus and expressed it in distantly related Saccharomyces cerevisiae, which lacks genes related by sequence, to find it encodes a functional phospholipase we named Plg7p. Ala substitution of the conserved Ser257 in the GXSXG lipase motif and two serine-directed inhibitors of plasma P AF acetylhydrolase, Pefabloc and methyl arachidonyl fluorophosphonate, suppressed Plg7p activity. S. pombe IIp/g7 had no overt phenotype, but revealed a second Ca++-independent phospholipase activity not inhibited by the Ser-directed reagents. S. cerevisiae supplemented with polyunsaturated fatty acids and exposed to environmental Cu + resulted in oxidized membrane lipids, and reduced viability. Expression of human type II PAF acetylhydrolase or Plg7p enhanced viability in this model of oxidative death. We conclude an organism distantly related to mammals expresses a phospholipase that is able to reduce oxidative death, suggesting this is the original role of the group VII phospholipase family. v To my parents Kathryn and Marc, for their encouragement and unconditional love, and to my wife Lacy for her patience, love and unlimited belief in me. TABLE OF CONTENTS ABSTRACT ............................................................................................... iv ACKNOWLEDGEMENTS ....................................................................... ix 1. INTRODUCTION .............................................................................. 1 Phospholipid Structure and Function .................................................. 1 Platelet Activating Factor Acetylhydrolase ...................................... 12 References ......................................................................................... 22 2. SCHIZOSACCHAROMYCES POMBE ENCODES A FUNCTIONAL PAF ACETYLHYDROLASE THAT REDUCES OXIDATIVE CELL DEATH ........................................ 28 Summary ........................................................................................... 28 Introduction ....................................................................................... 29 Materials and Methods ...................................................................... 31 Results ............................................................................................... 42 Discussion ......................................................................................... 85 References ......................................................................................... 89 3. MEGAKARYOCYTES DERIVED FROM CD34+ CELLS EXPRESS PAF ACETYLHYDROLASE TO CONTROL PAF-LIKE LIPID ACCUMULATION AND SIGNALING ............ 93 Summary ........................................................................................... 93 Introduction ....................................................................................... 94 Materials and Methods ...................................................................... 96 Results ............................................................................................. 102 Discussion ....................................................................................... 13 7 References ....................................................................................... 141 4. DISCUSSION AND SIGNIFICANCE ........................................... 143 The Role ofPlg7p in S. pombe ....................................................... 143 PAF-AH in Megakaryopoeisis ........................................................ 148 PAF-AH as a Scavenger of Oxidized Phospholipids ...................... 152 Future Directions ............................................................................ 154 References ....................................................................................... 15 7 Vll1 ACKNOWLEDGEMENTS I would like to thank David Stillman and Dennis Winge for valuable suggestions and materials. Thanks also to Gopal Marathe for his assistance with the calcium release assays. I want to thank Thomas McIntyre, Andrew Weyrich and Guy Zimmerman for being graciously accommodating and allowing me to do research in their labs, in addition to nurturing my scientific career. 1. INTRODUCTION Phospholipid Structure and Function Phospholipids are the major components of cell membranes. Membrane phospholipids typically consist of a glycerol backbone, two hydrophobic fatty acid tails, and a polar hydrophilic head group (1). The head group is linked to the glycerol backbone through a single phosphate group. Phospholipids have one of four head groups: choline, ethanolamine, serine, and inositol (Figure 1.1) (1). A final group, named sphingomyelin, has a serine derived backbone instead of glycerol and the fatty acid tails are attached by carbon-carbon or carbon-nitrogen bonds (1). Fatty acids in the other phospholipids can be attached to the glycerol backbone by ether bonds, and are termed alkinyl, or by ester bonds and are named acyl. The hydrophobic tails typically consist of long carbon chains consisting of 16-24 carbons arranged linearly. Fatty acid tails without double bonds are considered saturated, a single double bond is referred as monounsaturated, and multiple double bonds are polyunsaturated fatty acids. Membrane phospholipids are arranged in a bilayer, with the hydrophilic heads dissolved in the aqueous phase, and the hydrophilic tails of both layers facing each other (1). This arrangement allows the cell to modify its internal environment since a majority of molecules are either charged or too large to cross the hydrophobic layer. The composition of the lipid bilayer is not symmetrical; phosphatidylcholine and sphingomyelin, which both have choline based head groups, are predominantly located 0) o c: L... 0 0).0 u..:::i! >.U - m (!).o x I ~ -O-p=O { I I o 0 ===0 ===0 oI I x = CH2CH2NH3 = CH2CH2NH3 too CH3 I = CH2CH2(N)CH3 I CH3 OH I = HOCH CH HCOH I I HOCH CH HCOH I OH (Ethanolamine) (Serine) (Choline I Sphingomyelin) ( Inositol) Figure 1.1. The basic structure of common phospholipids in plasma membranes. The three main components consist of a head group (X), a glycerol backbone, and fatty acid tails. Both fatty acid tails illustrated are attached by ester bonds and are termed acyl fatty acids. Adapted from Alberts B., et al. Molecular Biology of the Cell, 3rd edition (1). 2 3 on the outer leaflet of the menlbrane (1). On the other hand, phosphatidylethanolamine and phosphatidylserine are more commonly found on the inner leaflet of the lipid bilayer (l). The relevance of having an asymmetrical membrane is important considering the types of enzymes that cleave or associate with only certain species of phospholipids (1). Also, one marker for apoptosis is phosphatidylserine ±lipped to the outer leaflet, which signals other cells to engulf the dying cell (2). Platelet Activating Factor Platelet activating factor (l-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine) is a cell synthesized phospholipid that induces potent cellular responses through a G-protein linked, seven transmembrane receptor (3). P AF was initially discovered and named in 1970 as a soluble molecule released from IgE stimulated basophils (4). The soluble factor was found to induce aggregation in rabbit platelets. The chemical structure of P AF (Figure 1.2) was solved in 1979 by three separate groups (5, 6, 7), including one group that found P AF isolated from the renal medulla induces low blood pressure in a rat hypertension model, which they named antihypertensive polar renal lipid (APRL) (8). The name PAF or PAF-acether, however, became the common names to reference the bioactive lipid. There is general agreement in the literature that the name platelet activating factor is a misnomer, since it has been reported to signal and alter the biology of a multitude of cells beyond platelets (3). Also, other lipid species exist that are similar to PAF in structure and can signal through the PAF receptor. These bioactive lipids (in ternlS of P AF receptor signaling) are typically formed through oxidation and are called P AF -like lipids due to structure and function similarities. The various pathways of P AF I o sn-1 Choline I o -o-pI =o I o I I o 1===0 sn-2 Figure 1.2. The general structure of platelet activating factor (P AF). Key characteristics of the lipid are the choline head group, and the acetate group at the sn-2 position. The sn-1 position can be either alkyl as above or acyl (like the sn-2 residue); however, the alkyl fonn is considerably more potent in signaling the P AF receptor. 4 5 synthesis have been reviewed and studied extensively (8, 9), and will be briefly discussed. To date, there are two commonly accepted models for PAF synthesis (10). The first de novo pathway is characterized by transfer of phospho choline to alkyl acetyl glycerol, and is generally considered to be less common. The second remodeling pathway involves replacement of a long chain fatty acyl residue with an acetyl group at the sn-2 position. The Ca2+ -dependent remodeling pathway requires a phospholipase A2 enzyme to first cleave the long sn-2 residue of the phospholipid precursor. cPLA2 has been proposed as the responsible PLA2 enzyme, since P AF synthesis in peritoneal macrophages is absent in cPLA2 knockout mice (11, 12). It is interesting to note that cPLA2 is specific for arachidonoyl residues at the sn-2 position, and that cleavage releases arachidonic acid, a precursor to eicosanoids that also are signaling molecules. Leukocytes and monocytes cultured in arachidonate-poor media display decreased P AF synthesis (3), further supporting the hypothesis that cPLA2 is required for P AF synthesis. An important note with P AF signaling is that cPLA2 does not distinguish between alkyl and acyl fatty acids at the sn-l residue (13). Alkyl-PAF is much more potent than acyl-P AF in affecting neutrophil biology, including chemotaxis, lysosomal enzyme secretion, and superoxide production (14). Thus the cell could favor the more potent alkyl-PAF over acyl-PAF by controlling the types of phospholipids that make up the membrane or by specifically synthesizing alkyl lysophospholipids. The majority of cells have only trace amounts of alkyl phospholipids; however, cells that synthesize P AF, such as endothelial cells (15) and neutrophils (16), have 10 to 40 percent alkyl phospholipids in their membranes (3). Other cells also involved in P AF synthesis, like leukocytes and 6 monocytes, are enriched for arachidonate in approximately 60 percent of their alkyl phospholipids (17, 18). Once the alkyl phospholipid is cleaved by cPLA2, another activity, acetyl-CoA­lysoP AF acetyl transferase, transfers an acetate group from acetyl-CoA to the lysophospholipid to make PAF (3). Intensive efforts to purify or clone the acetyltransferase have not been successful; however some characteristics of the enzyme have been described. The acetyltransferase appears to be activated through p38 mitogen­activated protein kinase in neutrophils (19) and through VEGF signaling in endothelial cells (20). The enzyme has little specificity for alkyl over acyl P AF, based on an observation when PMSF was added to A23187 stimulated leukocytes (21). Leukocytes stimulated with A23187 but not treated with PMSF show a 10-fold enrichment for the alkyl species of P AF (21), suggesting that degradation of acyl P AF is more efficient than alkyl PAP. PAF is synthesized in the cell, but released or displayed differentially depending on the cell involved. For endothelial cells, P AF is displayed on the cell surface along with P­selectin (an adhesion molecule involved in endothelial-leukocyte interaction) to help tether neutrophils to active sites of inflammation (3, 22). Monocytes have been shown to release most of their synthesized PAF (23), and neutrophils can release some PAF under certain conditions (24, 25). Platelet Activating Factor Receptor Platelet activating factor receptor (P AFR) is a seven-transmembrane G-coupled receptor found on all leukocytes in humans (26-28), but was originally cloned by function in the guinea pig lung (29). The P AF receptor gene has only one coding exon and is 7 located on chromosome 1 p35-p34.4 (30, 31). The two RNA isofonns of the PAFR, transcript 1 and transcript 2, differ in transcription start sites and consequently in 5' noncoding exons (32). Both isofonns have the same coding region, with the translation start site present in the third exon (exon 1 is found in transcript 1, exon 2 is found in transcript 2). The noncoding exons of the PAFR do not overlap, and exon 1 is more proximal to exon 3. Initial characterization of the promoters for each isofonn revealed different transcription factor binding sites (32). Transcript 1 contains binding sites for NF-kappa Band Sp-1, while transcript 2 has binding sites for AP-1, AP-2, and Sp-l. Further investigation revealed that leukocytes, EoL-1 cells (human eosinophilic cell line) and brain cells express transcript 1 only, while liver, heart, lung and spleen express both transcript 1 and transcript 2 (32). There seem to be no differences in function between the two P AFR isofonns, but rather differences in response to stimuli. The P AFR is constitutively expressed, but can also be induced in certain situations (3, 33, 34). PAF stimulation has been found to increase expression of transcript 1 in an NF-KB dependent manner (35). Transcript 2, conversely, is not induced by P AF but by phorbol esters, retinoic acid and thyroid hormone (35, 36). Thus there appear to be two distinct mechanisms to increase the levels of P AFR message, but the functional significance has yet to be explored. P AFR protein, on the other hand, has been more extensively studied. Numerous studies have found that stimulation of the P AFR with P AF leads to phosphatidylinositol turnover and a transient increase of calcium inside the cell, even when it was initially characterized in PAFR-transfected RBL-2H3 cells (a rat basophilic cell line) (37). Two different groups found that P AFR stimulation is only partially blocked by pertussis toxin 8 (37, 38), an inhibitor of one type of Ga subunit (39). These results suggest that there is more than one signaling pathway downstream of the P AFR. The P AF receptor localizes to the plasma membrane, although there are some reports of nuclear membrane localization and signaling (40, 41). Similar to other G-protein coupled receptors, experiments showed that the P AFR is desensitized after stimulation (3, 28). One pathway of desensitization is through phosphorylation of the C-terminus of the protein (42) by internalization by clathrin coated pits through beta-arrestill-1 (43), and degradation through ubiquitination which targets the protein to the proteasome or lysosome (44). Another pathway is through inactivation of downstream effectors, as in the case of phospholipase C (PLC) phosphorylation by protein kinase C (PKC) in RBL-2H3 cells (45). The examples above are representative of homologous desensitization where stimulation of the receptor directly results in an inability to further stimulate the receptor (3). Conversely, heterologous desensitization of receptors occurs through stimulation of other receptors with similar downstream signaling pathways. Heterologous desensitization has also been observed with the P AFR (45, 46). P AFR knock-out mice have been generated and studied for their phenotype in nonstressed or inflammatory situations. Based on numerous studies of P AF and the P AFR, it was believed that P AF plays an important role in reproduction, brain development, anaphylaxis and endotoxic shock (47). Only anaphylaxis, a fatal type of allergic reaction characterized by airway constriction, vascular leakage and hypotension, was observed with lower frequency in P AFR deficient mice (47). Reproduction, brain development and endotoxic shock were unaltered compared to wild type mice. Although 9 P AFR signaling may playa role in the conditions listed above, it appears to playa minor role at least in mice. Conversely, transgenic mice overexpressing the guinea pig PAFR displayed decreased viability in response to endotoxin, developed melanocytic tumors, and had hypersensitivity to methacholine induced bronchial reactivity (48). Combined with the results from P AFR deficient mice, it appears that P AF signaling can enhance responses to endotoxin, but that the most important role may be in airway responsiveness to various allergin challenges. The presence of melanocytic tumors in P AFR overexpressing mice was an unexpected phenotype, but coincides with reports of P AF induced neoplastic effects and skin inflammation (3). There is also evidence that PAF or PAF-like lipids may have significant roles in human biology. A P AF receptor polymorphism in a portion of the Japanese population decreased P AF signaling responses in a transfection model (49), and the mutations has been correlated with an increased susceptibility to multiple sclerosis (50). The mutation in the P AF receptor is thought to alter the receptors association with certain G-proteins which affect downstream signaling. Oxidation and Platelet Activating Factor-like Lipids Chemical oxidation is defined as the removal of electrons from elements or molecules. Oxidation is used as a mechanism for the cell to capture energy in the highly controlled process of oxidative phosphorylation, where electrons are transferred to lower energy states via the electron transport chain (l). Cells capture the energy released from the electron transport chain in the form of A TP through an electrochemical proton gradient (1). Oxidation is also used for the benefit of the cell when polymorphonuclear and mononuclear innate immune cells produce reactive oxygen species (ROS) as a 10 defense against infectious microbes. However, cells must deal with constant unregulated oxidation that can alter the function of DNA, proteins, and lipids. Oxidation of phospholipids typically occurs at double bonds in fatty acid tails, due to lower energy requirements to abstract hydrogen from a carbon hydrogen bond adjacent to double bonds. A carbon hydrogen bond next to conjugated double bonds is even less energetic and thus is more susceptible to oxidation. Monounsaturated and saturated fatty acids are not as sensitive to oxidation, but can still be fragmented with high energy species like the potent hydroxyl radical. Hydroxyl radicals are thought to form through the Fenton reaction, where iron (Fe2+) or copper (Cul+) reduce hydrogen peroxide. Fatty acid oxidation typically occurs at the sn-2 residue, since that is where unsaturated phospholipids are most commonly found (3). Free radicals attack the double bonds of polyunsaturated fatty acid tails, which can then react with oxygen to create a peroxy radical. This moiety can react in a variety of ways, including ~-scission at various sites in the fatty acid tail. The consequences of such a reaction in free fatty acids results in release of the lipid end (co) into an aqueous phase, and formation of a complex array of lipids with the acid end (3). The lipid species resulting from phospholipid oxidation can be relatively water soluble, disruptive to cellular membranes (51, 52), reactive through formation of an aldehyde residue (53, 54), or stimulate receptors as in PAF-like lipids (55). Since the PAFR is exquisitely sensitive to the structure ofPAF, PAF-like lipids must have similar characteristics in order to effectively signal through the receptor (3). These characteristics include a phosphocholine head group, a short acetyl group at sn-2, and an ether bond at the sn-l position (Figure 1.3). Changes at anyone of these sites results in I 0 Choline Choline I I 0 0 -o-pI =o -o-pI =o I I 0 0 I I I I I 0 0 0 1=0 1=0 PAF PAF-like Lipid Figure 1.3. Basic structure of PAF and one example of a PAF-like lipid. Longer fatty acid tails at the sn-2 position are typical in PAF-like lipids, and are fonned through the oxidation of double bonds. Both lipids can signal through the PAF receptor. Adapted from Prescott S.M., et al. Annu Rev Biochem (3). 11 12 decreased P AFR signaling (56, 57). Addition of more than two methylene groups at the sn-2 residue prevents P AFR signaling, but can be reversed with addition of an oxy function (3). Despite the stringency of the PAFR, there are species that are very similar to P AF and activate the receptor, yet are not synthesized by the addition of an acetyl group from acetyl-CoA. It should also be noted that although precursor lipids to PAF-like species can be rare (requiring ether bond at the sn-l position), the sensitivity of the receptor makes up for low concentrations of ligand. The P AF receptor has been documented to be sensitive to picomolar, and in some cases subpicomolar, concentrations ofPAF (57), and explains why PAF synthesis tightly regulated (3). The majority of P AF-like lipids found in an oxidized phospholipid pool, such as oxidized low-density lipoprotein (LDL), are primarily C4 PAF analogs (butanoyl sn-2 residues with or without a double bond) (57). These analogs are 10-fold less potent than P AF, and most likely constitute the majority of P AF -like lipids due to oxidation since they are not known to be synthesized in the cell (3). It should be noted that PAF-like lipids only make up a small proportion of lipids in an oxidized phospholipid pool. Oxidized lipids that are not PAF-like can also be relevant to conditions like atherosclerosis, a condition where oxidized LDL is an early and deleterious factor (58, 59). It is apparent that cells must find a way to eliminate oxidized phospholipids, which is the responsibility ofPAF acetylhydrolases (PAF-AHs). Platelet Activating Factor Acetylhydrolase Platelet activating factor acetylhydrolases (PAF-AHs) are unique esterases belonging to the group phospholipase A2, and because they do not require calcium for activation are therefore constitutively active (3). PAF-AH specificity is determined by substrate 13 composition, similar to other phospholipases. PAF-AHs are able to cleave short acyl sn-2 residues from phospholipids with either alkyl or acyl fatty acids at the sn-1 position to form lyso-PAF or lysophospholipids (Figure 1.4). PAF-AH activity is limited to short sn- 2 fatty acids, thus sparing intact phospholipids with long fatty acid chains where hydrolysis would be detrimental to the cell since PAF-AH activity is constitutive. To date there are four cloned forms of PAF-AH in two distinct families in humans: type 1b2 and 1b3 (Group VIII), or type 2 and plasma (Group VII) (Table 1.1). The type lbl isoform (~subunit, 4S-kDa) ofPAF-AH does not have activity. PAF-AHlbl is also referred to as Lis1 due to mutations in the gene that lead to Miller-Dieker lissencephaly (60), a disease characterized by severe brain malformation and seizures. PAF-AH1b1 is thought to be more of a regulatory protein that interacts in a G-protein like heterotrinler with Ib2 (al subunit, 29-kDa) or 1b3 (a2 subunit, 30-kDa) homodimers or Ib211b3 heterodimers (61). From purification studies, it appears that PAF-AH1b1 suppresses 1b2 homodimer or 1 b2-1 b3 heterodimer activity, but enhances 1 b3 homodimer activity (62). It is unclear if PAF plays a role in Miller-Dieker lissencephaly especially since PAF­AH1 b 1 can interact with other proteins through its WD-40 repeats, and P AFR knock-out mice have no defects in brain development. The human type 1 b2 and 1 b3, also known as a1 and a2 respectively, have about 60 percent amino acid homology to each other (61), and both contain catalytic serine residues in a slightly modified consensus lipase sequence Gly-X-Ser-X-Val (typical lipase motif is Gly-X-Ser-X-Gly). Both are highly specific for the sn-2 acetate residue of P AF, and are unable to cleave propionyl PAF or PAF derivatives with longer sn-2 carbon I o Choline I o -o-pI =o oI I I o 1==0 PAF PAF-AH o Choline I o -o-pI =o oI OH Lyso-PAF + -0 1==0 Acetate Figure 1.4. The canonical PAF-AH reaction, in which the acetate residue ofPAF is cleaved, releasing free acetate and forming the lysophospholipid Lyso-PAF. Notice the alcohol residue in place of acetate in Lyso-PAF (box). 14 15 Table 1.1 Isofonns ofPAF-AH that are Enzymatically Active in Humans Name Size Hydrolysis of Molecular Source oxidized lipids? Location Secreted, Macrophages, Plasma 45kDa Yes Associated with LDLIHDL platelets Liver, Kidney, Type 2 40kDa Yes Intracellular majority of tissues Type lb2a 29kDa No Intracellular Brain, majority of tissues Type Ib3a 30kDa No Intracellular Brain, majority of tissues aBoth 1 b2 and 1 b3 isoforms are inactive unless they interact with each other as homodimers or heterodimers 16 chains (62). Although all combinations of 1 b2 and 1 b3 dimers can efficiently cleave P AF, the rates of hydrolysis vary depending on the dimer composition. The 1 b3 homodimer was found to be 3-fold more efficient at cleaving P AF than a 1 b2 homodimer or 1 b2/1 b3 heterodimer (62). Alternatively, 1 b2 homo or heterodimers are more efficient at cleaving 1-0-alkyl-2-acetyl-sn-glycero-3-phosphorylethanolamine (essentially PAF with ethanolamine replacing choline) than the 1 b3 homodimer (62). Finally, the 1 b3 homodimer was the only combination that could cleave 1-0-alkyl-2-acetyl-sn-glycero-3- phosphoric acid (PAF without a head group). The 1b2 and 1b3 isoforms of PAF-AH both require a catalytic triad consisting of a serine, aspartic acid and histidine residue to be functional esterases (63), and it was found that dimerization is required for protein stabilization and activity (64). The dimerization requirement for activity is counterintuitive since the other isoforms of PAF-AH function well as monomers, and both also have the same active catalytic triad. Further studies revealed that two arginine residues (Arg22 and Arg29) of one monomer come into contact with the active site of the other monomer, and that mutations of Arg22 and Arg29 result in lower activity (66). It is unknown whether these residues affect binding or catalysis. Another group found that Leu48, Leu 1 94, and Thrl03 of PAF-AHlb2 all contact the methyl group on the sn-2 acetate tail of P AF, and could be responsible for the strict specificity of the enzyme for short fatty acids (65). Mice with targeted disruption of either PAF-AH1b2 or Ib3 showed no obvious phenotype compared to wild type mice (67). Males null for both Ib2 and Ib3 were observed to be sterile, and closer inspection of 1 b3 null mice revealed reduced testes size compared to wild type. No brain defects were observed in PAF-AHlb2 and Ib3 deficient 17 mice, arguing against the role of PAF in Miller-Dieker lissencephaly. The role of PAF-AH 1 b2 and 1 b3 in reproduction is opaque since P AFR null mice show no obvious reproductive deficiencies and both type 1 enzymes are somewhat restricted in their ability to hydrolyze lipids other than P AF. Thus it appears that P AF signaling is not essential for virility in male mice, and it is possible that the phenotype observed in the type 1 PAF-AH knock-out mice is due to unknown alternative activities of the enzymes. The other isoforms ofPAF-AH are the type 2 (40-kDa) and the plasma form (Pla2G7, 45-kDa). PAF-AH2 and plasma PAF-AH share approximately 40 percent amino acid identity (64), but differ in their locations in the cell. Like the 1b group, the type 2 isoform is located inside the cell (3), while the plasma isoform is responsible for all the secreted PAF-AH activity in humans (3). Both type 2 and plasma PAF-AH are similar in terms of substrate specificity. P AF-AH2 allows for longer carbon chains than 1 b at the sn-2 residue (68) of PAF-like lipids. The plasma isoform is most permissive of all, but requires carbonyl functions at the (O-end of the sn-2 residue to cleave long chains (up to 9 methylene groups) (3, 69). The plasma PAF-AH isoform was isolated from human plasma and found to associate with low-density and high-density lipoproteins (LDL and HDL respectively) (70). Human plasma PAF-AH was later cloned and predicted to have 441 amino acids, the first 20 being hydrophobic and are thought to be responsible for secretion of the enzyme (71). Plasma isolated P AF -AH are missing the first 50 amino acids of the N -terminus, suggesting posttranslational processing (68, 71). Removal of a signal peptide on the N­terminus of a protein is a common pathway for secretion of some proteins (1). Since P AF is known to induce a potent inflammatory response it rodents, subsequent experiments 18 tested to see if plasma PAF-AH is antiinflammatory. Interestingly, rats injected with PAF to induce paw oedema or pleurisy were protected by preinjection of recombinant plasma PAF-AH (71). Numerous reports have shown a portion of Asian populations are deficient in plasma PAF-AH activity (72), especially in Japan where the incidence has been reported to be 4 percent of the population (3). Individuals with plasma PAF-AH deficiency are healthy, but have been found to be at higher risk for arthritis, sepsis, asthma and atherosclerosis (73). P AF -AH2 was cloned from bovine liver and kidney cDNA libraries, and then a homolog in a human cDNA library was found (68). Sequence analysis revealed similarities to the plasma PAF-AH isoform, with the exception of about 50 amino acids at the N-terminus, which are thought to be involved in secretion of the plasma isoform. In some tissues, PAF-AH2 is responsible for the majority of intracellular hydrolase activity (61). PAF-AH2 also has an N-myristoylation sequence site that has been shown to be myristoylated and allows PAF-AH2 to shuttle from the membrane to the cytosol depending on its lipidation state (74). The same study showed that overexpression of PAF-AH2 in CHO-Kl cells offered protection against oxidation induced apoptosis (74), suggesting that the enzyme acts as a scavenger molecule for oxidixed lipids that have been shown to be toxic to cells (75). Interestingly, the location of PAF-AH2 in the cell was determined by cell redox state, although the mechanism for sensing oxidation and targeting PAF-AH2 to the cell membrane is unknown. P AF -AH and Megakaryocytes It is interesting to note that to date almost every cell in humans, including red blood cells, have some level of PAF-AH activity (3, 76, unpublished observations). The 19 primary expressors and secretors of plasma PAF-AH appear to be macrophages (71). Monocyte progenitors do not express message for plasma PAF-AH, and it appears to be upregulated over time. Our lab has also shown that neutrophils express plasma PAF-AH, and that message is induced by P AF signaling (unpublished observations). Another effector of innate immulogy, the platelet, has recently been fairly well characterized to express the plasma and type 2 PAF-AH isoforms (77). Thus three of the main responders to P AF - platelets, neutrophils, and monocytes/macrophages - also express regulatory proteins to control P AF accumulation. This suggests that a link exists between P AF and PAF-AH that could simply be a negative feedback loop mechanism. Inflammatory cells could respond to the inflammatory stimuli ofPAF as well as produce PAF-AH to degrade any further PAF or PAF-like lipid generation. Platelets were observed to have message for all isoforms of PAF-AH, although low levels were detected for all but the plasma isoform (77). Protein was detected for the plasma and type 2 isoforms, and it was discovered that platelets release plasma PAF-AH in microparticles (77). Our interest in platelets and PAF-AH was generated through a megakaryocyte cell culture model. Similar to data from other groups, we are currently able to isolate CD34+ stem cells from human cord blood and culture to differentiated megakaryocytes with the addition of thrombopoietin (TPO) and stem cell factor (SCF) (78). After two weeks of culture in serum free X-Vivo 20 media (Cambrex) and growth factors, mature megakaryocytes begin to form, giving off proplatelet extensions (79). These proplatelet extensions are thought bud off from the megakaryocyte and form mature platelets in the bloodstream. 20 Since it is known that platelets contain PAF-AH activity that is secretable (77, 80), and that monocytes upregulate plasma PAF-AH over differentiation (81), we were interested in examining megakaryocytes over differentiation for PAF-AH activity and expression of the PAF-AH isoforms. We were also interested to find a potential function for PAF-AH in megakaryocytes and platelets that may exist other than to degrade PAF. PAF-AH and Schizosaccharomyces pombe BLAST analysis revealed a PAF-AH homolog in S. pombe that has striking similarity to the plasma and type 2 PAF-AH human isoforms. Other homologs to PAF-AH are present in sequences from bacteria, fungi, metazoans and even one case in archae bacteria. Plants and viruses are notable exceptions where no instances of PAF-AH sequence homologs are found. Analysis of PAF-AH in species other than mammals has been limited to Caenorhabditis elegans and Drosophila melanogaster. In the case of C. elegans, two isoforms of P AF -AH exist (paf-l and paf-2), and both share significant homology to type 2 PAF-AH. Worms deficient in paf-l showed no abnormal phenotype, but paf-2 deficiency led to gross epithelial morphology changes and resulted in embryonic lethality (82). An antibody recognizing both isoforms revealed expression in epithelial cells, consistent with the deletion experiments. It is unknown why paf-2 was lethal as opposed to paf-l, especially since both proteins were shown to have similar activity and specificities. It is also unknown why PAF-AH activity is essential for epithelial morphology in worms. Conversely, D. melanogaster was found to have a homolog to PAF-AHI b2 and 1 b3 that was expressed in the brain. Surprisingly, despite the proteins overall similarity to the 21 PAF-AHlb isoforms, the enzyme is catalytically inactive presumably due to loss of two of the three critical residues for hydrolysis. Thus D. melanogaster has a paralog to P AF­AH with an unknown function. Due to these two diverging examples ofPAF-AH in nonmammalian species, we were interested in understanding whether the S. pombe PAF-AH homolog is enzymatically active and try to determine the function of the protein. It is interesting to note the lack of PAF receptor homo logs outside of metazoans, including S. pombe. This suggests that PAF-AHs are more ancient that the PAF receptor and that their role is more substantial than P AF signaling. 22 References 1. Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., Watson, J.D., Molecular Biology of the Cell, 3rd edition (New York & London: Garland Publishing, Inc., 1994), xliii+ 1294pp. 2. Wu, Y., Tibrewal, N., Birge, R.B. (2006) Trends Cell Bioi 16(4), 189-97. 3. Prescott, S. M., Zimmennan, G. A., Stafforini, D. M., and McIntyre, T. M. (2000) Annu Rev Biochem 69, 419-45. 4. Benveniste, J., Henson, P. M., and Cochrane, C. G. (1972) J Exp Med 136, 1356- 1377. 5. Benveniste, J., Tence, M., Varenne, P., Bidault, J., Boullet, C., and Polonsky, J. (1979) C R Acad Sci Paris 289,1037-1040. 6. Blank, M. L., Snyder, F., Byers, L. W., Brooks, B., and Muirhead, E. (1979) Biochem Biophys Res Commun 90, 1194-1200. 7. Demopoulos, C. A., Pinckard, R. N., and Hanahan, D. J. (1979) J Bioi Chem 254, 9355-9358. 8. Snyder, F. (1990) Am J Physiol259, C697-C708. 9. Snyder, F. (1995) Biochem J305 (Pt 3),698-705. 10. Prescott, S. M., McIntyre, T. M., and Zimmennan, G. A. (1999) Inflammation: Basic Principles and Clinical Correlates (Philadelphia: Lippincott/Williams & Wilkins), pp. 387-96. 11. Bonventre, J. V., Huang, Z., Taheri, M. R., O'Leary, E., Li, E., Moskowitz, M. A., and Sapirstein, A. (1997) Nature 390 (6660), 622-625. 12. Uozumi, N., Kume, K., Nagase, T., Nakatani, N., Ishii, S., Tashiro, F., Komagata, Y., Maki, K., Ikuta, K., Ouchi, Y., Miyazaki, J., and Shimizu, T. (1997) Nature 390 (6660), 618-622. 13. Leslie, C. C., Voelker, D. R., Channon, J. Y., Wall, M. M., and Zelamey, P. T. (1988) Biochim Biophys Acta 963, 476-492. 14. Pinckard, R. N., Showell, H. J., Castillo, R., Lear, C., Breslow, R., McManus, L. M., Woodard, D. S., and Ludwig, J. C. (1992) J ImmunoI148(11), 3528-3535. 15. Murphy, E. J., Joseph, L., Stephens, R., and Horrocks, L. A. (1992) Lipids 27(2), 150-153. 23 16. Ramesha, C. S., and Pickett, W. C. (1987)J Lipid Res 28(3),326-31. 17. Nakagawa, Y., Kurihara, K., Sugiura, T., and Waku, K. (1986) Biochim Biophys Acta 876 (3),601-610. 18. Chilton, F. H., and Connell, T. R. (1988) J BioI Chem 263(11),5260-5265. 19. Nixon, A. B., O'Flaherty, 1. T., Salyer, 1. K., and Wykle, R. L. (1999) J BioI Chem 274(9), 5469-5473. 20. Bematchez, P. N., Winstead, M. V., Dennis, E. A., and Sirois, M. G. (2001) Br J PharmacoI134(1), 197-205. 21. Sturk, A., Schaap, M. C., Prins, A., ten Cate, 1. W., and van den Bosch, H. (1989) Biochim Biophys Acta 993(2-3), 148-156. 22. Prescott, S. M., Zimmerman, G. A., and McIntyre, T. M. (1984) Proc Natl Acad Sci USA 81(11), 3534-3538. 23. Elstad, M. R., Prescott, S. M., and McIntyre, T. M. (1988) J Immunol 140(5), 1618-1624. 24. Cluzel, M., Undem, B. 1., and Chilton, F. H. (1989) J ImmunoI143(11), 3659- 3665. 25. Miwa, M., Sugatani, J., Ikemura, T., Okamoto, Y., Ino, M., Saito, K., Suzuki, Y., and Matsumoto, M. (1992) J ImmunoI148(3), 872-880. 26. Ye, R. D., Prossnitz, E. R., Zou, A., Cochrane, C. G. (1991) Biochem Biophys Res Commun 180( 1), 105-111. 27. Nakamura, M., Honda, Z., Izumi, T., Sakanaka, C., Mutoh, H., Minami, M., Bito, H., Seyama, Y., Matsumoto, T., Noma, M., and Shimizu, T. (1991) J Bioi Chem 266(30), 20400-20405. 28. Kunz, D., Gerard, N. P., Gerard, C. (1992) J Bioi Chem 267(13), 9101-9106. 29. Honda, Z., Nakamura, M., Miki, 1., Minami, M., Watanabe, T., Seyama, Y., Okado, H., Toh, H., Ito, K., Miyamoto, T., and Shimizu, T. (1991) Nature 349( 6307), 342-346. 30. Seyfried, C. E., Schweickart, V. L., Godiska, R., and Gray, P. W. (1992) Genomics. 13(3), 832-834. 31. Chase, P. B., Yang, J. M., Thompson, F. H., Halonen, M., Regan, 1. W. (1996) Cytogenet. Cell Genet 72(2-3):205-207. 24 32. Mutoh, H., Bito, H., Minami, M., Nakamura, M., Honda, Z., Izumi, T., Nakata, R., Kurachi, Y., Terano, A., and Shimizu, T. (1993) FEBS Lett 322(2), 129-134. 33. Rola-Pleszczynski, M., Thivierge, M., Ouellet, S., Dagenais, P., Parent, J. L., and Stankova, J. (1995) Adv Prostaglandin Thromboxane Leukot Res 23,287-292. 34. Dagenais, P., Thivierge, M., Parent, J. L., Stankova, J., and Rola-Pleszczynski, M. (1997) J Leukoc BioI 61( 1), 106-112. 35. Mutoh, H., Ishii, S., Izumi, T., Kato, S., and Shimizu, T. (1994) Biochem Biophys Res Commun 205(2), 1137-1142. 36. Mutoh, H., Fukuda, T., Kitamaoto, T., Masushige, S., Sasaki, H., Shimizu, T., and Kato, S. (1996) Proc Natl Acad Sci USA 93(2), 774-779. 37. Ali, H., Richardson, R. M., Tomhave, E. D., DuBose, R. A., Haribabu, B., and Snyderman, R. (1994) J BioI Chem 269(40), 24557-24563. 38. Honda, Z., Takano, T., Gotoh, Y., Nishida, E., Ito, K., and Shimizu, T. (1994) J BioI Chem 269(3), 2307-2315. 39. Bokoch, G. M., Katada, T., Northup, J. K., Hewlett, E. L., and Gilman, A. G. (1983) J BioI Chem 258(4), 2072-2075. 40. Marrache, A. M., Gobeil, F. Jr., Bernier, S. G., Stankova, J., Rola-Pleszczynski, M., Choufani, S., Bkaily, G., Bourdeau, A., Sirois, M. G., Vazquez-Tello, A., Fan, L., Joyal, J. S., Filep, J. G., Varma, D. R., Ribeiro-Da-Silva, A., Chemtob, S. (2002) J ImmunoI169(ll), 6474-6481. 41. Zhu, T., Gobeil, F., Vazquez-Tello, A., Leduc, M., Rihakova, L., Bossolasco, M., Bkaily, G., Peri, K., Varma, D. R., Orvoine, R., Chemtob, S. (2006) Can J Physiol Pharmacol84(3-4), 377-391. 42. Takano, T., Honda, Z., Sakanaka, C., Izumi, T., Kameyama, K., Haga, K., Haga, T., Kurokawa, K., and Shimizu, T. (1994) J BioI Chem 269(35), 22453-22458. 43. McLaughlin, N. J., Banerjee, A., Kelher, M. R., Gamboni-Robertson, F., Hamiel, C., Sheppard, F. R., Moore, E. E., and Silliman, C. C. (2006) J ImmunoI176(ll), 7039-7050. 44. Dupre, D. J., Chen, Z., Le GouiU, C., Theriault, C., Parent, J. L., Rola­Pleszczynski, M., and Stanko va, J. (2003) J BioI Chem 278(48), 48228-48235. 45. Ali, H., Fisher, I., Haribabu, B., Richardson, R. M., and Snyderman, R. (1997) J BioI Chem 272(18), 11706-11709. 25 46. Richardson, R. M., Haribabu, B., Ali, H., and Snydennan, R. (1996) J Bioi Chern 271(45), 28717-28724. 47. Ishii, S., Kuwaki, T., Nagase, T., Maki, K., Tashiro, F., Sunaga, S., Cao, W. H., Kume, K., Fukuchi, Y., Ikuta, K., Miyazaki, J., Kumada, M., and Shimizu T. (1998) J Exp Med 187(11), 1779-1788. 48. Ishii, S., Nagase, T., Tashiro, F., Ikuta, K., Sato, S., Waga, 1., Kume, K., Miyazaki, J., and Shimizu, T. (1997) EMBO J 16(1), 133-142. 49. Fukunaga, K., Ishii, S., Asano, K., Yokomizo, T., Shiomi, T., Shimizu, T., Yamaguchi, K. (2001) J Bioi Chern 276(46), 43025-43030. 50. Osoegawa, M., Miyagishi, R., Ochi, H., Nakamura, 1., Niino, M., Kikuchi, S., Murai, H., Fukazawa, T., Minohara, M., Tashiro, K., Kira, J. (2005) J NeuroirnrnunoI161(1-2), 195-198. 51. Dousset, N., Dousset, 1. C., and Douste-Blazy, L. (1981) Biochirn Biophys Acta 641(1), 1-10. 52. Itabe, H., Kushi, Y., Handa, S., and Inoue, K. (1988) Biochirn Biophys Acta. 962(1), 8-15. 53. Itabe, H., Yamamoto, H., Suzuki, M., Kawai, Y., Nakagawa, Y., Suzuki, A., Imanaka, T., and Takano, T. (1996) J Bioi Chern 271(52),33208-33217. 54. Horkko, S., Miller, E., Branch, D. W., Palinski, W., and Witztum, J. L. (1997) Proc Natl Acad Sci USA 94(19), 10356-10361. 55. McIntyre, T. M., Zimmennan, G. A., and Prescott, S. M. (1999) J Bioi Chern 274(36):25189-25192. 56. Tanaka, T., limori, M., Tsukatani, H., and Tokumura, A. (1994) Biochirn Biophys Acta 1210(2), 202-8. 57. Marathe, G. K., Davies, S. S., Harrison, K. A., Silva, A. R., Murphy, R. C., Castro-Faria-Neto, H., Prescott, S. M., Zimmennan, G. A., and McIntyre, T. M. (1999) J Bioi Chern 274(40),28395-28404. 58. Steinberg, D., Parthasarathy, S., Carew, T. E., Khoo, J. C., and Witztum, J. L. (1989) N Engl J Med 320(14),915-924. 59. Homma, Y. (2004) J Atheroscler Thrornb 11(5), 265-270. 60. Hattori, M., Adachi, H., Tsujimoto, M., Arai, H., and Inoue, K. (1994) Nature 370(6486), 216-218. 26 61. Arai, H., Koizumi, H., Aoki, J., and Inoue, K. (2002) J Biochem 131(5), 635-640. 62. Manya, H., Aoki, J., Kato, H., Ishii, J., Hino, S., Arai, H., and Inoue, K. (1999) J BioI Chem 274(45),31827-31832. 63. Adachi, H., Tsujimoto, M., Hattori, M., Arai, H., and Inoue, K. (1997) Biochem Biophys Res Commun 233( 1), 10-13. 64. Hattori, K., Hattori, M., Adachi, H., Tsujimoto, M., Arai, H., and Inoue, K. (1995) J BioI Chem 270(38), 22308-22313. 65. Ho, Y. S., Swenson, L., Derewenda, U., Serre, L., Wei, Y., Dauter, Z., Hattori, M., Adachi, T., Aoki, 1., Arai, H., Inoue, K., and Derewenda, Z. S. (1997) Nature 385( 6611), 89-93. 66. McMullen, T. W., Li, 1., Sheffield, P. J., Aoki, J., Martin, T. W., Arai, H., Inoue, K., and Derewenda, Z. S. (2000) Protein Eng 13(12), 865-871. 67. Koizumi, H., Yamaguchi, N., Hattori, M., Ishikawa, T. 0., Aoki, J., Taketo, M. M., Inoue, K., and Arai, H. (2003) J BioI Chem 278(14), 12489-12494. 68. Hattori, K., Adachi, H., Matsuzawa, A., Yamamoto, K., Tsujimoto, M., Aoki, 1., Hattori, M., Arai, H., and Inoue K. (1996) J Bioi Chem 271(51), 33032-33038. 69. Stremler, K. E., Stafforini, D. M., Prescott, S. M., and McIntyre, T. M. (1991) J Bioi Chem 266(17), 11095-11103. 70. Stafforini, D. M., Prescott, S. M., and McIntyre, T. M. (1987) J Bioi Chem 262(9), 4223-4230. 71. Tjoelker, L. W., Wilder, C., Eberhardt, C., Stafforini, D. M., Dietsch, G., Schimpf, B., Hooper, S., Le Trong, H., Cousens, L. S., Zimmerman, G. A., Yamada, Y., McIntyre, T. M., Prescott, S. M., and Gray, P. W. (1995) Nature 374(6522), 549-553. 72. Miwa, M., Miyake, T., Yamanaka, T., Sugatani, J., Suzuki, Y., Sakata, S., Araki, Y., and Matsunl0to, M. (1988) J Clin Invest 82(6), 1983-1991. 73. Imaizumi, T. A., Stafforini, D. M., Yamada, Y., McIntyre, T. M., Prescott, S. M., and Zimmerman, G. A. (1995) J Intern Med 238(1):5-20. 74. Matsuzawa, A., Hattori, K., Aoki, J., Arai, H., and Inoue, K. (1997) J Bioi Chem 272(51),32315-32320. 75. Chisolm, G. M. 3rd, and Chai, Y. (2000) Free Radic Bioi Med 28(12), 1697-1707. 27 76. Karasawa, K., Shirakura, M., Harada, A., Satoh, N., Yokoyama, K., Setaka, M., and Inoue, K. (2005) J Biochem (Tokyo) 138(4), 509-517. 77. Mitsios, 1. V., Vini, M. P., Stengel, D., Ninio, E., and Tselepis A. D. (2006) Arterioscler Thromb Vasc BioI 26(8), 1907-1913. 78. Patel, S. R., Hartwig, 1. H., and Italiano, J. E. Jr. (2005) J Clin Invest 115(12), 3348-3354. 79. Denis, M. M., Tolley, N. D., Bunting, M., Schwertz, H., Jiang, H., Lindemann, S., Yost, C. C., Rubner, F. 1., Albertine, K. H., Swoboda, K. J., Fratto, C. M., Tolley, E., Kraiss, L. W., McIntyre, T. M., Zimmerman, G. A., Weyrich, A. S. (2005) Cell 122(3), 379-391. 80. Suzuki, Y., Miwa, M., Harada, M., and Matsumoto, M. (1988) Eur J Biochem 172(1), 117-120. 81. Elstad, M. R., Stafforini, D. M., McIntyre, T. M., Prescott, S. M., and Zimmerman, G. A. (1989)J BioI Chem 264(15),8467-8470. 82. Inoue, T., Sugimoto, A., Suzuki, Y., Yamamoto, M., Tsujimoto, M., Inoue, K., Aoki, J., and Arai, H. (2004) Proc Natl Acad Sci USA 101(36), 13233-13238. 2. SCHIZOSACCHAROMYCES POMBE ENCODES A FUNCTIONAL PAF ACETYLHYDROLASE THAT REDUCES OXIDATIVE CELL DEATH Summary Oxidation of phospholipids with sn-2 unsaturated fatty acyl residues fragments the esterified chain forming numerous phospholipids, some of which are toxic. Group VII phospholipases (PAF acetylhydrolases, PAF-AH) are Ca++-independent esterases that specifically hydrolyze short sn-2 residues, including the acetyl residue of PAF. The fission yeast S. pombe genome contains an open reading frame and that potential encodes a gene homologous to group VII hydrolases and, although it is only 24 percent identical, it retains a lipase motif and residues that form the esterase catalytic triad. An active enzyme in an organism that cannot generate P AF may instead protect against membrane oxidation and death. We cloned this SPBCI06.11c locus and expressed it in distantly related Saccharomyces cerevisiae, which lacks genes related by sequence, to find it encodes a functional phospholipase that we named Plg7p. Ala substitution of the conserved Ser257 in the GXSXG lipase motif abolished activity and two serine-directed inhibitors of plasma P AF acetylhydrolase, Pefabloc SC and methyl arachidonyl fluorophosphonate, suppressed Plg7p activity. S. pombe ~plg7 had no overt phenotype, but also revealed a second Ca2+ -independent phospholipase activity not inhibited by the Ser-directed reagents. S. cerevisiae supplemented with polyunsaturated fatty acids and exposed to environmental Cu + resulted in oxidized membrane lipids, which reduced viability. Expression of human type II P AF acetylhydrolase or Plg7p enhanced viability 29 in this model of oxidative death. We conclude an organism distantly related to mammals expresses a phospholipase with the capacity to reduce oxidative death, suggesting this is the original role of the group VII phospholipase family. Introduction Platelet activating factor (P AF) acetylhydrolases, classified as group VII and group VIII phospholipases A2, are Ca ++ -independent esterases that display an unusual specificity for the target sn-2 residue. The mammalian genome encodes two members of the Group VII that are 42 percent identical (1), contain the esterase GXSXG motif with the acceptor Ser residue, and have identically positioned Asp and His residues that complete a catalytic esterase triad (2). The two members of group VIII phospholipases (1 b2 and 1 b3) are related to one another, but not to the group VII PAF acetylhydrolase family (1, 3). PAF is a phospholipid autocoid, a locally acting hormone, intimately involved in the acute inflammatory response (4) that signals to a single receptor expressed by all cells of the innate immune system (5, 6). Structurally, PAF is a phosphatidylcholine with an sn-l ether bond and a short chain sn-2 acetyl residue, both of which are required for the very high affinity interaction with the PAF receptor (7,8). PAF is rapidly, and transiently, synthesized by inflammatory cells after stimulation by classic inflammatory agonists. Plasma P AF acetylhydrolase, a group VII enzyme (PLA2g7), circulates in a fully active state (9, 10) and acts on PAF displayed on the surface of activated endothelial cells (11). This reduces endothelial cell-dependent leukocyte activation and trafficking, and so the deduced role of plasma P AF acetylhydrolase is to resolve inflammatory signaling. 30 The role of type II PAF acetylhydrolase (PAF-AH2) is opaque. This intracellular enzyme is found in noninflammatory cells-initially it was purified and cloned from liver (3, 12)-and is seemingly not positioned to participate in resolution of PAF signaling. However, both group VII enzymes will hydrolyze substrates other than PAF (12, 13), in contrast to the two group VIII phospholipases A2 that display a stringent requirement for the sn-2 acetyl residue ofPAF (12, 14). The function of at least PAF-AH2 may therefore relate to hydrolysis of substrates other than P AF. The two group VII enzymes are inactive against natural, long chain phospholipids of cellular membranes and circulating lipoproteins, but will hydrolyze phospholipids where the sn-2 residue is short or contains introduced oxy functions. Phospholipids with these characteristics are formed by oxidative fragmentation of esterified polyunsaturated phospholipids (15-17). Some phospholipid oxidation products are toxic (18)(submitted, McIntyre TM et a1.), and overexpression of either PLA2g7 (19, 20) or PAF-AH2 (21, 22) suppresses oxidant-induced toxicity. PAF-AH2, and perhaps as an ancillary function of circulating PLA2g7, may function to detoxify oxidatively damaged phospholipids. Murine knockouts for neither group VII gene is reported, but C. elegans has two group VII homologs with one of these predominantly expressed in epithelial cells. Genetic ablation of one locus disrupts normal epithelial morphogenesis in an undefined way (23), while disruption of the other locus has no effect on the animal. In contrast to group VII phospholipase A2, the group VIII phospholipases A2 selectively hydrolyze P AF. Yet, targeted disruption of one of the group VIII genes has no effect on normal murine physiology, while disruption of the second of the two family members impaired 31 spermatogenesis (24). Currently, the role of either group of phospholipases A2 cannot easily be ascribed to either P AF metabolism or resolution of oxidative stress. The genome of Schizosaccharomyces pombe contains a putative gene that would encode a protein with 25 percent identity to PLA2g7, and 24 percent identity to P AF­AH2 (25). Additionally, this SPBC106.11c locus retains residues that form the catalytic triad, suggesting that the translated product could be expected to have esterolytic activity. A key question is why such a distantly related organism, and one not engaged in phospholipid intercellular signaling, would possess such an activity. We postulated that understanding the role of a group VII homolog in an ancient organism would offer insight into the role of the homologous genes retained by mammals. S. pombe are unusually enriched in unsaturated fatty acids (26), and we show here that the S. pombe locus encodes a functional phospholipase that protects against oxidative cell death. Materials and Methods Strains, Media and Plasmids All mammalian cell lines were purchased from ATCC (CV-1, CCL-70; COS-7, CRL- 1651; CHO-K1, CCL-61; 293T, CRL-11268). Rosetta-gami B competent Escherichia coli [F- ompT hsdSB(rB- mB-) gal dcm lacY1 aphC (DE3) gor522::Tn10 trxB pLysSRARE2 (CamR , KanR , TetR )] cells were purchased from Novagen. INVSc1 (MATa his3D1 leu2 trpl-289 ura3-52 MATa his3D1 leu2 trpl-289 ura3-52) and TCP1 (h- leu1- 32) yeast strains were purchased from Invitrogen. Strains DY 1838 (MATa pep4-3 prb1- 1122 HISIJ.:p GAL 1 O:GAL4 his3IJ. leu2 trp1 ura3-52), DY 4165 (MA Ta ade2 ade3 can1 his5 leu2 lys2 trp1 ura3), and DY 851 (MATa can1 his7 hom3 leu2 trp1 ura3) were generous gifts from D.l. Stillman. BY4741jet3IJ. (MATa, ura3-52, leu2-3, 2-112, trp1-1, 32 his3-11, 3-15, ade2-1, can1-100, fet3::HIS3) was kindly provided by D.R. Winge. S. pombe strain CHP428 (h+ ura4-D18 leul-32 ade6-M210 his7-366) was purchased from ATCC (201399). All mammalian cell lines were grown in 90 percent Dulbecco's modified Eagle's medium (DMEM) with 4 mM L-glutamine, 1.5 giL sodium bicarbonate, 4.5 giL glucose, and 10 percent fetal bovine serum (Sigma). S. cerevisiae were grown in synthetic complete (SC) minimal media without uracil (2 percent glucose or 2 percent galactose, and 1 percent raffinose for expression studies), SC-Iow ade (2 percent galactose, 1 percent raffinose, and 0.01 percent adenine), or YEPD (1 percent yeast extract, 2 percent peptone, and 2 percent glucose). S. pombe were grown in SC minimal media without histidine, YES media (0.5 percent yeast extract, 3 percent glucose, and 0.2 percent ade, his, leu, ura, lys) or Edinburgh Minimal Media (EMM, Gibco) with or without 10 JlM Thiamine (Invitrogen). Bacteria were grown in Miller's Luria Broth (Fisher) with 50 Jlg/mL Kanamycin, Ampicillin, Tetracycline, and Chloramphenicol (Sigma). Plasmids pENTR™/D-TOPO®, pET-DEST42, pDEST24, pNMTl and pYES­DEST52 were purchased from Invitrogen. Plasmid pCI-Neo was purchased from Promega. Plasmid pEZZ 18 was purchased from Amersham. Plasmid M3153 was kindly provided by D. J. Stillman. All vectors were sequenced after construction to confirm correct inserts and sequences. Cosmid 1 06 (Welcome Trust Sanger Institute) was used to clone SPBCI06.11c into pYES-DEST52 using the pENTRTM/D-TOPO® cloning kit. PCR ofSPBC106.11c (plg7+) with Platinum® Taq (Invitrogen) to incorporate into pYES­DEST52 with a V5/6xHis C-terminus tag was done using forward primer 5'-CAC CGA AAT GGG ATT GGG ATT TTC TTC G and reverse primer 5'-GTA CAT AAT TCT 33 TIC CCA CCC AGG. The QuikChange® II site-directed mutagenesis kit (Stratagene) was used to mutate Ser257 of pig 7+ to Ala with the primers: 5'- AA T TGA TTG TTG CTG GTC ATG CAT TTG GTG CCG CTA CTT GC and 5'-GCA AGT AGC GGC ACC AAA TGC ATG ACC AGC AAC AAT CAA TT. The V5/6xHis tag ofplg7+ and plg7-S257A was removed using the primers 5'-TCC CTC GCA TGG TAC TAA GG and 5'-GGT CGG CGC GCC CAC CCT TTC ACA TAA TTC TIT CCC AC. The PCR product was cleaved with Hind III and BssH II (New England Biolabs) and inserted into pDEST52-plg7+ or plg7-S257 A cut with the same enzymes. Empty vector pDEST52 was made by cleaving pDEST52-plg7+ with BssH II and Sac II, blunted using Klenow fragment polymerase (Promega), and ligated with T4 Ligase. A human PAF-AH type 2 cDNA clone was purchased from Origene (Clone AB3241_B06) and shuttled into pYES-DEST52 using the pENTRTM/D-TOPO® cloning kit with the primers 5'-CAC CCT GGG TCG TTT CTC ATT TCC and 5'-GGA AAT GGC CAG TTG TGC GTA C, and the resulting plasmid did not have a V5/6xHis tag. The V5/6xHis tag was added onto pDEST52-HPAF-AH2 using the primers: 5'-GGA ATG GAT CCC TTT CCG TC and 5'-GGT CGG CGC GCC CAC CCT TGG AAA TGG CCA GTT GTG CCC GCA GGC TGG AC. The PCR product was then cleaved with BamH I and BssH II and inserted into pDEST52-HPAF-AH2 cut with the same enzymes. The I1plg7 construct was made using the his7+ gene from pEA2 (ATCC) cut with Xba I and EcoR I and inserted into pCI-Neo. The PCR product of the plg7+ N-terminus using the primers 5'-CCT AGC TAG CGG GAT TGG GAT TTT CTT CG and 5'-CGG AAT TCC GAA AAC CTI TCG CAA CTI C was cut with Nhe I and EcoR I and inserted into pCI-neo-his7. The PCR product of the plg7+ C-tenninus using the primers 34 5'-TGC TCT AGA ITC CCA CGT GTT TGT TTA TGA and 5'-ATA AGA ATG CGG CCG CAT TCT TTC CCA CCC AGG AAT was cut with Xba I and Not I and inserted into pCI-neo-his7-N-tenninus plg7+ to make the IJ.plg7 construct. S. cerevisiae strains were transfonned using the s.c. EasyCompTM Transfonnation Kit (Invitrogen); S. pombe strains were transfonned with the YEASTMAKER ™ yeast transfonnation system 2 (Clontech). Amplfication of plg7+ from pDEST52-plg7+ N5/6xHis was done using the primers YES-pCI-5' (5' GGA AAT CAA CCC CGG ATC GGA CTA CTA) and YES-pCI-3' (5' GCT CTA GAT ACA TGA TGC GGC CCT CTA) for insertion into the pCI-Neo vector using EcoR I and Xba I sites internal to the primers. The resulting pCI-Neo­pig 7+ N5/6xHis vector was transfected into CV -1 cells using Lipofectamine (Invitrogen), along with a positive control of human PAF-AH2 in a CMV expression construct. To make the pCI-Neo-plgr vector without a V5/6xHis tag, the primers spCl1 (5' CTA GGC TAG CGA AAT GGG ATT GGG ATT TTC) and spCI2 (5' ATA AGA ATG CGG CCG CTC ACA TAA TTC TIT CCC ACC) were used to amplify plg7+ from pDEST52-plg7+ using Platinum Taq polymerase. PCR products and pCI-neo were cleaved with Not I and Nco I, and ligated using T4 Ligase (Promega). For bacteria expression vectors, PCR-amplification of pig 7+ was done using the same primers to insert the gene into pDEST52, but this time the product was inserted into the pENTRISDID-TOPO vector which contains a Shine-Delgarno sequence that has a ribosomal binding site to improve translation efficiency. The resulting vector, pENTRISD/D-TOPO-plg7+ was used in an LR recombination reaction with pDEST24 or pET -DEST42. The pDEST24 plasmid has a C-tenninal Glutathione S-transferase (GST) 35 tag, while pET-DEST42 has a C-terminal V5/6xHis tag. Primer design and insertion of plg7+ permitted both tags to be in frame with the protein in the respective plasmids. A T7 promoter drives expression in both pDEST24 and pET-DEST42 plasmids, which was utilized in a system that allowed for isopropyl ~-D-l-thiogalactopyranoside (IPTG)­inducible expression ofT7 RNA polymerase. The vectors pDEST52-plg7-G2A-V5/6xHis and pDEST52-plg7- G2A1S257 A/V5/6xHis were made by amplifying plg7+ with the primers g2a-mutl (5' CAT TTC TCG AGC CAA CCA AT) and g2a-mut2 (5' CTC CGC GGC CGC CCC CTT CAC CGA AAT GGC CTT GGG ATT TTC). The PCR product, pDEST52-plg7+ and pDEST52-plg7-S257 A were cleaved with Not I and Xho I and ligated with T4 DNA Ligase. The plasmids pNMTl-plg7+ N5/6xHis and pNMTl-plg7-G2A1V5/6xHis were made by amplifying pDEST52-plg7+ N5/6xHis or pDEST52-plg7-G2A1V5/6xHis with the primers NMT-toprol (5' GAA ATG GGA TTG GGA TTT TC) and NMT-topro2 (5' GAT GCG GCC CTC TAG GAT CAG) and inserted into pNMTI using the TOPO Cloning reaction (Invitrogen). The pDEST52-plgr-protein A/V5/6xHis vector was made by using the PCR primers 5'GCT GCC GGC CGT TCC TTG GAA GTA TAA GTT TTC CCC GGG TAC CGA GCT CGA AT and 5' GCG CAC GGC CGA AAA AAA TGG CTG CAA ATG CTG CGC AAC AC to amplify the protein A tag from pEZZ 18 and include the Tobacco Etch Virus protease target sequence: Glu-Asn-Leu-Tyr-Phe-Gln-Gly. Primers also include cut sites for Not I, which was used to insert and tag the N-terminus of pDEST52-plg7+. Clones were screened by PCR analysis to determine correct orientation of the insert, and confirmed by sequencing. 36 The pDEST52-HPAF-AH2-ADE3+ vector was constructed in a two step process. First, a restriction enzyme site was introduced into M3153 by hybridizing the 5' phosporylated oligos 3153-nde 1 (5' GAT CCC ATA TGA CGC GTG ATA TCG CTA GCC CGC GGG TCG ACC TCG AGG) and 3153-nde 2 (5' GAT CCC TCG AGG TCG ACC CGC GGG CTA GCG ATA TCA CGC GTC ATA TGG). The linker was inserted into BamH I cut M3153, which was pre-treated with Calf Intestine Phosphatase (CIP, New England Biolabs) prior to ligation. The resulting vector 3153-ndel, was cleaved with Nde I to remove theADE3 gene and inserted into Nde I cleaved pDEST52-HPAF-AH2. BLAST analysis comparing SPBCI06.11c to human plasma and type 2 PAF-AH was done using the NCBI website http://www.ncbi.nlm.nih.govlblastlbI2seq/wblast2.cgi. ClustalW alignment was done using the default settings at website http://align.genome.jp/ (CLUSTAL output, BLOSUM weight matrix). Expression studies A single colony from transformed S. cerevisiae was grown in 5 mL SC-ura (glucose) at 30°C overnight, centrifuged at 400xg for 5 minutes, washed once with sterile water and resuspended in 25 mL SC-ura (galactose). Cells were grown overnight at 30°C, centrifuged and resuspended in SDS buffer (l percent SDS, 5 percent glycerol, 125 mM Tris pH 6.8) + protease inhibitor cocktail (Pierce) at an optical density (OD) 550 nm of 100 (GENESYSTM 5, Spectronic). Cells were lysed by vortexing with glass beads, boiling for 5 minutes, then clarified in a centrifuge for 2 nlinutes at 20000xg in a microcentrifuge. 20 J.!L of clarified lysates were loaded onto a 12 percent SDS-PAGE gel, transferred to a PVDF membrane (lmmobilon™, Millipore) and blotted for the V5 epitope (R960, Invitrogen) at 1:2500 and actin (691001, MP Biomedicals) at 1:1000. For 37 activity assays, cells were lysed in 50 mM Tris buffer pH 7.8 with 20 uM E-64, pepstatin A, and 2 mM benzamidine HCI. PAF-AH activity assays were done as previously described (2), using 25 JlL celllysates and incubating at 37°C for 15 minutes with 40 JlL 1 nmol 3H-PAF, 5 JlL 400 mM DTT and 0.5 JlL 0.5 M EDTA. The PAF-AH inhibitors Pefabloc SC (Boehringer Mannheim 1429 868) and methyl arachidonyl fluorophosphonate (MAFP, BIOMOL ST-360) were added to lysates at a final concentration of 1 mM and incubated for 30 minutes at 37°C prior to incubation with 3H_ PAF. The reactions were quenched with 50 JlL of 10 M glacial acetic acid and excess 0.1 M sodium acetate. Cleaved 3H-acetate was isolated using BAKERBOND™ spe Octadecyl (CIS) extraction columns (J.T. Baker). Lysates were normalized using the Coomassie Protein Assay Kit from Pierce. For immunofluorescence, paraformaldehyde fixed cells were treated with 1 mg/mL Lyticase (Sigma), then adhered to poly-L-Iysine coated 8-well borosilicate chamber slides. Cells were permeabilized with ice-cold methanol for 5 minutes, then ice-cold acetone for 30 seconds. Cells were blocked with PBS/l percentBSA, then incubated overnight at 4°C with primary antibodies at a final concentration of 2 Jlg/mL. The cells were washed 3x with PBS, and incubated with secondary antibodies for 1 hour at room temperature at a final concentration of 2 Jlg/mL. Cells were then stained with TO-PRO-3 (Invitrogen) at 1:400 dilution for 5 minutes, washed 3 times and visualized by confocal microscopy. Confocal microscopy was done using a 488 nm Argon and 543 or 633 HeNe excitation laser, and a 520/10 nm, 605/25 nm or 700/20 nm bandwidth emission filter (60X 1.42NA oil objective on an FV300 Olympus IX81 microscope). 38 Primary IgG 1 antibodies against dolichol phosphate mannose synthase, porin, 3- phosphoglycerate kinase, and carboxypeptidase Y and IgG2a anti-V5 were purchased from Invitrogen (A-6429, A-6449, A-6457, A-6428, and R960 respectively). Secondary antibodies (goat anti-mouse IgG 1-546 and IgG2a-488) were also purchased from Invitrogen. Expression and purification of Plg7p-protein Al6xHisN5 was done by inducing expression in DY 1838 cells grown to an OD550 of 0.7 in 2 liters of SC-ura (galactose, raffinose) media. Cells were lysed by glass beads in a bead beater, and clarified lysates were loaded onto IgG Sepharose 6 Fast Flow columns from Amersham. Lysates were washed with 50 mM Tris buffer, pH 7.6, 150 mM NaCI and 0.05 percent Tween 20. Columns were incubated with AcTEV protease (Invitrogen) for 1 hour at room temperature, and eluted with lysis buffer (50 mM Tris buffer pH 7.8 with 20 uM E-64, pepstatin A, and 2 mM benzamidine HCI). Eluted fractions were concentrated in 30000 Dalton molecular weight spin columns from Millipore. For bacteria expression studies, competent Rosetta-gami B cells from Novagen were transformed with pDEST24-plg7+ or pET -DEST42-plg7+. These cells are optimized for foreign gene expression through expression of rare tRNAs that are not commonly found in bacteria, a lack of the thioredoxin reductase (trxB) and glutathione reductase (gor) genes allowing for disulfide bond formation in the cytosol, and are deficient in Ion and ompT proteases that gives better stability of the expressed proteins. Cultures were grown in small volume (5 mL) in Luria-Broth media (Fisher) supplemented with 50 ug/mL Kanamycin, Tetracycline, Chloramphenicol, and Ampicillin overnight. Cultures were diluted and grown for 3-4 hours until they reached 39 an OD 600nm of 0.4. Expression was induced with 1 mM IPTG for 2 hours, then cells were centrifuged and lysed with 250 J.tL SDS-PAGE loading buffer (1 percent sodium dodecyl sulfate (SDS), 5 percent glycerol, 125 mM Tris-HCI pH 6.8). Lysates for soluble and insoluble fractions were generated by incubation with 10 ug/mL lysozyme in 50 mM Tris-HCI buffer for 30 minutes at 37°C, then centrifuged at 20000xg for 5 minutes in an Eppendorf 5417R centrifuge. Twenty five J.tL of the samples were loaded onto a 9 or 12 percent SDS-PAGE gel, electrophoresed, transferred to an Immobilon P membrane, and probed with anti-V5 or anti-GST antibodies (Santa Cruz, sc-459, 1 :100). Secondary antibodies were purchased from Biosource (goat anti-mouse Ig's HRP conjugate, AMI0404, and goat anti-rabbit Ig's HRP conjugate, ALI3404) and diluted 1 :300 in HBSS+ 1 0 percent goat serum prior to incubation with membranes. Membranes were developed using ECL from Amersham-Biosciences and a Kodak RP X-OMAT processor. For mammalian expression studies, cells were plated at 70-80 percent confluency the day prior to transfection on a 6-well tissue culture treated plate (Falcon). Two J.tg of plasmid was used to transfect an individual well with serum free media and Lipofectamine for 4 hours at 37°C in a humid incubator. Cells were then washed, DMEM media added, and incubated overnight at 37°C in a humid incubator. Cells were lysed in 20 mM Tris, 15 mM CHAPS buffer after incubation and assayed for P AF-AH activity and normalized to protein content by Bradford assay. l:!p/g7 generation S. pombe strain CHP428 was transformed with the fip/g7 construct and grown on SC­his media (3 percent agar) at 30°C. Colonies were replica plated 3 times on SC complete media, then transferred back to SC-his plates to select for stable transformants. Colonies 40 were screened for correct insertion of construct by colony PCR as described previously (27, 28). Briefly, a colony was resuspended in 10 )lL zymolyase mixture (2.5 mglmL zymolyase (Sigma), 1.2 M sorbitol, 0.1 M sodium phosphate, pH 7.4) and incubated for 15 minutes at 37°C. 2 )lL of the spheroplasts were used in a 25 )ll PCR reaction with Taq DNA polymerase (Promega) using primers: dl 5' CCT TAA TCA TCG CGG TCC TA with d2 5' AGG CTT TTT CCA TCT CCT GA, or d3 5' TG CAA ACG AAA GAT TCA CA with d4 5' AAA ACG AAC CGG CTA AAA GG to detect successful integration. Primers d4 with Y5 5' TCT CGC GAT ACT GAA CAA CG were used to detect the presence of wild type pig 7+. Supplementation and oxidation assays S. cerevisiae were grown for 2 days shaking in 25 mL SC-ura (galactose) + 1 percent 1gepal CA-630 at 25°C. Cells were centrifuged at 400xg, washed once with sterile water and a portion of the cells were resuspended in 19 mL of SC-ura (galactose) + 1 mL filter sterilized 5 percent 1gepal or 20 mM linolenic acid (Sigma) in 5 percent 1gepal to an OD550 of approximately 0.2. The cells were grown overnight at 25°C shaking, and 1 mL was removed for oxidation detection by fluorescence. Five )lL of 2 mM Bodipy® 5811591 C11 (Invitrogen) were added to 1 mL of cells, and incubated by rocking at room temperature for 30 minutes. Cells were collected by centrifuge at 400xg for 5 minutes and resuspended in MES to obtain an OD550 of 1. Then, 250 )lL was removed for untreated, and the remaining amount treated with a final concentration of 50 )lM CUS04. 125 )lL of cells were ali quoted in duplicate into a flat-bottomed black 96-well plate and fluorescence was measured in 5 minutes increments at 485/20 nm bandwidth excitation, 528/20 nm bandwidth emission at a sensitivity of 50 in a Synergy HT fluorimeter (B10- 41 TEK®). Alternatively, cells were adhered to 8-well chamber coverslips coated with 1 mglmL poly-L-Iysine. Thirty III of cells were allowed to adhere for 1-2 minutes, and washed once with MES buffer. Cells were left untreated or treated with a final concentration of 50 IlM CUS04 and visualized after 60 minutes by confocal microscopy using a 488 nm Argon laser, 520/10 nm emission filter. Viability assays S. cerevisiae were grown in mock or linolenic acid supplemented media as with the oxidation detection assay and collected the next day at 400xg for 5 minutes. Pellets were washed with sterile water twice, and final cell pellets were resuspended in 1 mL of water. Suspensions were added to 20 nIL of sterile 10 mM 2-( 4-morpholino )-ethane sulfonic acid (MES) buffer pH 5.5, 1 percent glucose to an OD 550 nm of 0.6-0.8 (OD of cultures were matched to within 0.01). After 10 minutes of equilibration, 1 mL was removed for untreated control, and the rest of the cells were treated with 190 III of 50 IlM CUS04. Aliquots were removed after 15 and 60 minutes, diluted 10-fold serially and 7 III of each dilution were spot plated on YEPD (3 percent agar). Cells were grown for 4 days at room temperature. For cell counting, 100 III of 10-3 and 10-4 dilutions were plated on YEPD media and grown for 4 days at room temperature. Viability was determined as the fraction of cells that formed colonies after treatment with CUS04 compared to untreated. Functional Cloning Assay An overnight culture of pDEST52-HPAF-AH2-ADE3+ transformed DY 4165 cells was grown in 5 mL of SC-ura (glucose) media at 30°C. The culture was diluted 1 :200, and 200 ilL was spread onto SC low ade (galactose) plates. A total of 150 plates were 42 used in the experiment, and each was exposed to 30 mJ of UV light in a UV Stratalinker 2400 (Stratagene). A control, nonexposed plate was included to check viability and was prepared by spreading 100 p.L of a 1: 1 0000 dilution of the overnight culture. All plates were incubated at 35°C for 5 days. The resulting red colonies were streaked again on SC low ade (galactose) plates to confirm positive colonies. The colonies that remained red on the second screen were then streaked onto YEPD plates and grown at 30°C to rule out false positives. Results BLAST analysis against human plasma P AF acetylhydrolase revealed a putative PAF-AH open reading frame in the S. pombe genome. This ORF, SPBC106.11c, has 25 percent identity and 44 percent similarity to the human plasma PAF-AH isoform and 24 percent identity and 41 percent similarity to human type 2 PAF-AH (Figure 2.1). SPBC1 06.11 c, which we later named pig 7+ due to homology to phospholipase group VII, does not share homology with the N-terminus of plasma PAF-AH, including the 17- amino acid sequence thought to be important for secretion (29). ClustalW analysis reveals the consensus lipase sequence GXSXG is conserved in the yeast and human isoforms (Figure 2.1). Also conserved are the anlino acids Ser257, Asp291, and His368, previously shown to be critical for enzyme activity (Ser273/236, Asp296/259, His351/315 for plasma and type 2 PAF-AH respectively) (l, 30). These residues form a catalytic triad characteristic of esterases and lipases, and retention of all the essential residues suggest SPBC1 06.11 c might encode a functional enzyme. 43 Plasma PAF-AH: SPBCl06.1le PAF-AH type 2: 1 MVPPKLHVLFCLCGCLAVVYPFDWQYINPVAHMKSSAWVNKIQVLMAAASFGQTK1PRGII 1 --------------------------------MGLGFSSKKQLPAYCGPLPVGSL)LELI 1 -----------------------------------------------MGVNQSVGjppvr Plasma PAF-AH: SPBCl06.11e PAF-AH type 2: Plasma PAF-AH: SPBCI06.1le PAF-AH type 2: Plasma PAF-AH: SPBC106.l1e PAF-AH type 2 : Plasma PAF-AH: SPBC106.lle PAF-AH type 2 : Plasma PAF-AH: SPBCl06.lle PAF-AH type 2: n!H~~IE 117 86 WLLRA 72 174 146 131 227 206 190 282 TLI<I >i---------- 266 IsiSTKSLYNDYM 245 AI.---------- EIGDIS-wlYLRTLKQ VVQyRjISDiYADATVV NE S LQI'E-wlp FRRVEE ~DLK!D-­CSlIDSSIYN .' lfiPGGiD---. * Plasma PAF-AH: 328 ___________________ * II~R·~t~v~ln SPBC106.1le 326 ALESWLVNKDSENQNAGESADEuMilltuullllT PAF-AH type 2: 291 -------------------VN~·H'_~.YnI~1 * FQ Plasma PAF-AH: 367 SPBCl06.lle 385- PAF-AH type 2: 332 -CDLIP ..-...'. ••.' R..:•.. ' .•_.'..,. DLDQET •;.•'.. •: . · .'.R ..: •.l . ·. r,'-.IENE RINITMT-N-LI. IGP~T· PHHLSS Plasma PAF-AH: 427 QHIMLQNSSGIEKYN SPBCl06.l1e 438 --------------­PAF- AH type 2: 392 L-------------- Figure 2.1. S. pombe has a PAF-AH honl010g to the human plasma, type 2 PAF-AH isofonns. Sequence alignment of the S. pombe putative PAF-AH (SPBC106.11c), human plasma and type 2 PAF-AH using the ClustalW, BLOSUM program. Characters highlighted in black are exact matches, characters highlighted in gray are similar in identity. Amino acids marked (*) are essential for enzyme activity. 44 Expression of Plg7p Various expression systems were analyzed for ability to express SPBC 106.11 c. Initial attempts were in a mammalian system, potentially recapitulating the system used previously to show that PAF-AH2 protects against oxidative stress (21). CV -1 (green monkey kidney) cells transfected with pCI-Neo-plg7+ N5/6xHis had a very minimal increase of activity over mock transfected samples, while pCMV-HPAF-AH2 produced a 4-fold increase of activity (Figure 2.2). Similar results were obtained using other cell types such as COS-7, 293T and CHO­K1 cells, and other transfection reagents like Fugene (Roche) (data not shown).Westerns using anti-6xHis or anti-V5 antibodies were unable to detect appropriate size bands of Plg7pN5/6xHis (55 kDa) in soluble or insoluble lysates from transfected cultures that were distinct from mock transfected (data not shown). Thus there was no detectable activity or protein for Plg7p expressed in a mammalian system. Next, expression of plg7+ in bacteria expression plasmids was examined. The goal was to overexpress plg7+ and purify it for activity studies. Plg7p was tagged with C­terminal V5/6xHis or GST and expressed in competent E. coli bacteria. Western expression studies revealed the presence of correct sized bands, but also an abundant amount of intermediate lower sized bands in IPTG-induced samples for either type of tag (Figure 2.3, data not shown). Follow up experiments revealed that the majority of protein expressed was insoluble (Figure 2.4). Foreign gene expression in bacteria frequently results in insoluble protein formation and was an expected but unfortunate result. Although various methods were tried to increase the solubility, such as different media, induction conditions, or bacteria strains, no change significantly increased the amount of -C) ::l 1:: .c ~ a. -c- "C Q) II) as Q) "i ~ .! S Q) u « :I: ('II') 2000 "I 1800 1600 1400 1200 1000 800 600 400 ~ 200 0 Mock pCI-Neo- pCMV-HPAF-AH2 pig 7+N5/6xHi s Figure 2.2. PIg7pN5/6xHis expressed in a mammalian cell line does not increase soluble PAF-AH activity. CMV-promoter based plasmids expressingplg7+N5/6xHis or human PAF-AH2 were transfected into CV- 1 cells. Cells were lysed in a Tris-CHAPS buffer and assayed for P AF -AH activity. Released acetate from 3H-PAF was measured in a scintillation counter (LS 6500, Beckman Coulter), and normalized by protein concentration (Coommassie Protein Assay, Pierce). 45 1 mM IPTG Plg7pN5/ -+ 6xHis * * * 1 + 2 3 4 + + + 92 kDa Figure 2.3. The majority of PIg7pN5/6xHis expressed in bacteria is not full length protein. Expression of pET -DEST42-plg7+ N5/6xHis in Rosetta-gami B competent bacteria reveals the presence of full length PIg7p/V5/6xHis, as well as degradation products (stars). Four different clones were used for expression, and induced for 2 hours at 37° C with 1 mM IPTG. Totallysates were loaded on a 12 percent SDS-PAGE gel for western blot. Membranes were probed with anti-V5 and exposed to film for 15 minutes. The predicted protein size ofPlg7p-V5/6xHis is 55 kDa. 46 1 mM IPTG Plg7pN5! ----+ 6xHis Insoluble + Soluble + Figure 2.4. PIg7pN5/6xHis expressed in bacteria is mostly insoluble. A single clone of pET-DEST42-plg7+N5/6xHis in Rosetta-gami B competent bacteria was induced for 2 hours at 37° C with 1 mM IPTG. Soluble and insoluble lysates were loaded on a 12 percent SDS-PAGE gel for western blot. Membranes were probed with anti-V5 and exposed to film for 20 minutes. The predicted protein size of PIg7p-V5/6xHis is 55 kDa, full length protein is marked with an arrow. 47 48 soluble protein. None of the soluble protein resulted in an increase of PAF-AH activity in bacteria (data not shown). We chose S. cerevisiae to express SPBC106.11c because this distantly related yeast had low levels of endogenous PAF-AH activity compared to S. pombe (Table 2.1) and a search of its genome reveals that it lacks any homolog to SPBC106.11c or other PAF acetylhydrolases. SPBC106.11c and a mutated form that replaces what is predicted to be the active site serine (Ser257 Ala) were cloned into the p YES-DEST52 vector with introduced V5 and 6xHis tags at the C-terminus. S. cerevisiae strain DY 1838 was transformed with these plasmids (Figure 2.5) and expressed encoded proteins at the predicted size of 55 kDa. SPBC106.11c therefore codes for a protein, which we named Plg7p due to homology to the phospholipase group VII hydro lases. We determined whether Plg7p is a catalytically active enzyme, but found that Plg7pN5/6xHis in crude lysates or after purification on a chelation affinity column (31) failed to hydrolyze PAF (data not shown). Expressing Plg7p-V5/6xHis and Plg7p­S257A- V5/6xHis in the INVSc1 expression strain did not result in a large amount of increased soluble activity above the mutant protein (data not shown), and it was found that the majority of protein expressed was insoluble (Figure 2.6). The insolubility of Plg7p-V5/6xHis in S. cerevisiae could be due to accumulation in inclusion bodies or localization at the plasma membrane (or in other membranes in the cell). This possibility was examined by immunofluorescence using antibodies against the V5 epitope. Immunofluorescence showed that induced protein was detected, and that it localized in a ring type arrangement around the nucleus (Figure 2.7). Subsequent immunofluorescence using colocalization markers to the vacuolar lumen (carboxypep- Table 2.1 P AF -AH activity in clarified lysates from S. cerevisiae and S. pombe S. cerevisiae (INVSc 1) S. pombe (CHP 428) 16.3±6.4a 19S.9±S4.3a aValues are presented as 3H-acetate released in DPMlhourlJlg. 49 Empty p/g7+ p/g7-S2S7A 54 38 Figure 2.S. pig 7+ NS/6xHis and plg7-S2S7 ANS/6xHis are expressed as full length proteins in S. cerevisiae. Western blot of lysates from emtpy vector, plg7+ NS/6xHis, or plg7-S2S7 ANS/6xHis-pDESTS2 transformed DY 1838 cells with antibodies against actin (42 kDa) and the VS epitope tag. so Soluble Insoluble WT p/g7-S257 A p/g7+ p/g7-S257 A p/g7+ 94 kDa 54 37 29 Figure 2.6. The majority of Plg7p/V5/6xHis (and Plg7p-S257 A) are insoluble in S. cerevisae. Western blot of pDEST52-plg7+/V5/6xHis or pDEST52-plg7-S257 AlV5/6xHis INVScl transformed cells or nontransformed cells (wild type = WT). Two separate clones were incubated in SC-ura (Galactose) media overnight at 30°C. The antibodies used were against actin (42 kDa) and the V5 epitope. The predicted protein size ofPlg7p-V5/6xHis is 55 kDa. 51 Anti-V5 Nuclei Noninduced Induced Figure 2.7. Plg7p localizes to a ring outside the nucleus in S. cerevisiae. Immunofluorescence of INVScl cells transformed with pDESTS2-plg7+ NS/6xHis. Expression was either induced in SC­ura (Galactose) media, or cells were left under noninducing conditions in SC-ura (Glucose). Cells were stained with anti-VS (green), and TO-PRO-3® (blue) which stains nuclei and visualized by confocal microscopy. Scale bar = 10 )lm. 52 53 tidase Y), endoplasmic reticulum (dolichol phosphate mannose synthase), cytosol (3- phosphoglycerate kinase), and mitochondria (porin) showed no substantial colocalization with Plg7p-V5/6xHis (Figure 2.8, A-D). Also interesting was the fact that not every cell expressed high levels ofPlg7p-V5/6xHis, which could be due to variability in the number of plasmids in each cell. The further decrease in PAF-AH activity in protein A-tagged Plg7p/V5/6xHis raised the issue that perhaps modification of Plg7p was interfering with activity. In an effort to improve the solubility of Plg7p-V 5/6xHis, the glycine at position 2 of the protein was mutated to an alanine. A report on the human type 2 PAF-AH discovered a myristoylation sequence at the N-terminus of the protein (MGXXXS) that was found to influence protein presence on the plasma membrane through lipidation of the glycine residue (21). A similar sequence is observed in Plg7p, which could influence localization of the protein in the same manner. Expression of Plg7p-G2A1V5/6xHis (or Plg7p­G2A1S257A/ V5/6xHis) in strain INVScl revealed even less soluble protein than previously found without the mutation (Figure 2.9). It was next determined whether Plg7p/V5/6xHis localization to the membrane is also observed when expressed in S. pombe. Plg7pN5/6xHis and Plg7p-G2A/V5/6xHis were expressed under NMT promoters (no message in thiamine) in transformed TCP 1 cells using EMM media without thiamine and examined by immunofluorescence. Localization of Plg7p/V5/6xHis was similar to that seen in S. cerevisiae, showing a ring-like expression on the perimeter of the cell and was unaffected with the G2A substitution (Figure 2.10). Also similar to S. cerevisiae expression of Plg7p/V5/6xHis was the varia- «>.1cJ0:. x A B c D Figure 2.8. Plg7p does not colocalize with markers for vacuoles, endoplasmic reticulum, mitochondria, or the cytosol. Immunofluorescence of INVScl cells transformed with pDEST52-plg7+ N5/6xHis. Cells were stained with anti-V5 (green), TO-PRO-3® (nuclei, blue) and colocalization markers (X, red). Antibodies against colocalization markers were as follows: X = carboxypeptidase Y (A, vacuolar lumen), dolichol phosphate mannose synthase (B, endoplasmic reticulum membrane), porin (C, mitochondria), 3-phosphoglycerate kinase (D, cytosol). Cells were visualized by confocal microscopy. Scale bar = 10 J.llTI. 54 Soluble E G2A G2A1 S257A E Insoluble G2A1 G2A S257A 52 kDa 35 29 Figure 2.9. Plg7p-G2A is insoluble when expressed in S. cerevlszae. Western blot of pDEST52-p/g7-G2A-V5/6xHis (G2A), pDEST52-p/g7- G2A/S257A1V5/6xHis (G2NS257A) and pDEST52 empty vector (E) INVScl transformed cells. Antibodies against actin (42 kDa) and the V5 epitope were used for blotting. 55 L() > • ..L +"' C « CD 13 :::J Z A B c Figure 2.10. Plg7pNS/6xHis overexpressed in S. pombe localizes in an outer ring in the cells, and is unaffected by mutation of Glycine 2 to Alanine. Immunofluorescence of TCP 1 cells transformed with pNMTl-plg7+NS/6xHis (A, B) or pNMTI-plg7- G2AN5/6xHis (C). Expression was induced by incubating the cells in EMM media (B, C) or in noninducing media EMM+ 10 uM thiamine (A). Cells were stained with anti-V5 (green) and TO­PRO- 3 (nuclei, blue) and visualized by confocal microscopy. Scale bar = 10 Jlm. 56 57 bility of protein expression within a culture. TCP 1 cells expressing Plg7pN5/6xHis did not have substantial increases in activity compared to noninduced cells (with 10 J.lM thiamine). Thus in four different expression systems Plg7p/V5/6xHis did not have significant amounts ofPAF-AH activity. In another effort to improve the solubility and potential for purification of Plg7p, a construct was generated with a protein A tag on the N-terminus of the protein that could then be purified on an IgG column. Release of the protein from the IgG column involves cleavage with TEV (Tobacco Etch Virus) protease that recognizes a peptide sequence placed between the protein A tag and Plg7p. Western analysis revealed that Plg7p/V5/6xHis is more soluble with a protein A tag, with approximately equal amount of protein in the soluble and insoluble fractions (Figure 2.11). Protein eluted from the TEV treated column resulted in no visible bands, but concentration of the eluted fraction with a 30 kDa molecular weight cut-off spin filter showed the presence of protein at a lower than expected size (less than 50 kDa). The presence of a smear in the concentrated sample suggested proteolysis or breakdown of Plg7pN5/6xHis. It seems unlikely that TEV with its high specificity would be responsible, especially without any similar consensus sequence sites in Plg7pN5/6xHis. Thus either the protein is inherently unstable, or susceptible to other serine proteases that may be present in S. cerevisiae. This last point is especially relevant, since soluble Plg7pN5/6xHis was very difficult to detect by western blot unless a significant amount of protease inhibitors were present at the time oflysis. Since increased solubility was observed with Plg7p/V5/6xHis/Protein A, we were interested if increased solubility led to increased activity. PAF-AH assay of soluble lys- Lysate Insol Sol Eluted Cone 92 kDa 52 35 28 Figure 2.11. Protein A-Plg7pN5/6xHis is approximately 50 percent soluble, but susceptible to proteolysis. Western blot of expressed Plg7p­protein AlV5/6xHis in insoluble (Insol) or soluble (Sol) lysates from transformed DY 1838 (s. cerevisiae). Soluble lysates were purified on an IgG Fast Flow purification column, and eluted by cleavage with TEV protease (Eluted). Eluted samples were concentrated in a 30 kDa molecular weight cut-off filter (Conc). Membranes were probed with an anti-V5 antibody and exposed for 5 minutes after blotting with secondary antibodies. The predicted size of the intact protein is 72 kDa, and after cleavage is 56 kDa. 58 59 ates showed no increase of activity, but rather a small decrease in total activity (Figure 2.12). This was confirmed using intact S. cerevisiae that were concentrated down to a high OD 550 nm (around 30), incubated directly in a PAF-AH reaction with 3H-PAF, and lysed prior to analysis of free 3H-acetate (Figure 2.12, intact). The fact that intact S. cerevisiae can take up P AF suggests a transporter is involved (32). It has been consistantly observed that intact PAF-AH assays correlates well to the activity found in lysates. The decrease in PAF-AH activity in protein A-tagged Plg7pN5/6xHis raised the issue that perhaps modification of Plg7p by the incorporated tags was interfering with activity. This was especially intriguing since the human type 2 PAF-AH that was expressed in the same experiment showed significant activity, and did not have an epitope tag. We expressed endogenous Plg7p in S. cerevisiae, and added a C-terminus V5/6xHis tag to HPAF-AH2 to assess for activity changes. The results showed that indeed the V5/6xHis tag interfered with PAF-AH activity (Figure 2.13). Removal of the tag on Plg7p resulted in a 6-fold increase in activity compared to Plg7p/V5/6xHis. Conversely, addition of a V5/6xHis tag on HPAF-AH2 resulted in a 2.5-fold decrease activity compared to nontagged HPAF-AH2.These data suggest that both HPAF-AH2 and Plg7p require an unaltered C-terminus to function properly. Decreased function with protein modifications may be due to alterations in folding that could affect binding or hydrolysis and should be studied further. Further studies of Plg7p without a tag showed the activity levels of Plg7p were similar to the human intracellular type 2 PAF-AH expressed under the same conditions (Figure 2.14). The mutant Plg7p-S257 A had little or no activity suggesting this conserved ... 0 CD ~ i:: -.c ::::!!l 0..- cC -0 "i:: CII.c UI_ m::::!!l ~a.. &!C -CII -m CII u « ::z:: (II) II Soluble • Intact 120 100 80 60 40 ! 20 0 V5/6xHis S257A1V5/6xHis V5/6xHis/Prot A HPAF-AH2 Figure 2.12. Intact cells have comparable activity to soluble lysates from S. cerevisiae and both show low activity with Plg7p/V5/6xHis. PAF-AH activity assay on strain DY 851 transformed with pDEST52-plg7+ with the tags listed above (except for HPAF-AH2, which was expressed without a tag in pDEST52). Cells were lysed and centrifuged to separate the soluble fraction. Cells were also concentrated down to an OD of 30, and incubated directly in the reaction without lysis. Activity numbers were normalized by protein concentration for lysates and OD550 for intact cells. 60 Q o 1.c: 2000 i ~ 1500 'tJ CI) lQ 1000 .! ~ -CI) co Ci) u <C J: M 500 o ~ ---"1 --_. Empty plg?+ pig? +N5/6xHis HPAF-AH2 HPAF­AH2N5/ 6xHis Figure 2.13. The V5/6xHis C-tenninus tag interferes with Plg7p and HPAF-AH2 activity. PAF-AH assay of intact INVScl cells transfonned with pDEST52 vectors with the listed expression constructs. Cells were concentrated down to an OD of 30, and incubated directly in the reaction without lysis. 61 -en ::l 1:: -J: :::!: c.. -c ~ 'S: ~ 600 * * 500 400 I 300 200 100 a Empty PIg7p Plg7p-S257 A PAF-AH2 Figure 2.14. Plg7p has Ser257 dependent phospholipase A2 activity. P AF­AH activity assay of celllysates from empty vector, plg7\ plg7-S257 A, or human PAF-AH2-pDEST52 transformed INVScl cells normalized by total protein. n=4, * = P<O.05 vs empty, Plg7p-s257a (One Way ANOVA, Tukey Test). 62 63 serine IS essential for catalysis by Plg7p. We tested whether targeted covalent modification of the active site serine by Pefabloc, a modified sulfonyl fluoride, that inhibits human plasma and type 2 PAF-AH (33) or methyl arachidonyl fluorophosphonate (MAFP), that also targets plasma and type 2 PAF-AH (34), would have an effect on Plg7p activity. Each mechanism based reagent greatly reduced, but did not abolish, Plg7p and HPAF -AH2 activity in crude lysates (Figure 2.15). Similar results were observed using recombinant plasma PAF-AH in the absence of yeast lysate, although activity was completely abolished in this case. Since Plg7p has significant PAF-AH activity, we examined the stability of the protein when expressed without a tag in S. cerevisiae. Plg7p expression was inducedovemight in SC-ura (galactose) med.ia, then diluted 1: lOin SC-ura (glucose) and incubated at room temperature for 2 constxutive days. An intact PAF-AH activity assay was done on each day, normalizing the ct:lls to the same OD 550 nm. Interestingly, the protein was fairly stable after being switched into glucose for one day (Figure 2.16). Two days of incubation resulted in a 5-fold decrease in activity. It is possible that activity decreased on the second day due to cell division, and thus dilution of the protein, however the OD 550 nM only increased 2-fold from day 0 to day 2, thus accounting for only about a 2- fold decrease in activity. Plg7p is also fairly stable in celllysates, generally confirmed by observations that several freeze thaw cycles have no effect on activity. Since it was discovered that the V5/6xHis tag was interfering with activity ofPlg7p, a follow up experiment was done using nontagged Plg7p expressed in pCI-Neo in mammalian cells. COS-7 cells transfected with pCI-Neo-plg7+ actually had lower activity compared to mock transfected, while pCMV-HPAF-AH2 produced a 3-fold in- 600 * - 500 CI :::l ".si::. 400 :::::: :::E • None ~ 300 • 1 m M Pefabloc SC 01 mMMAFP -:>::­-,~ 200 u < 100 o PAf.·AH2 PIg7p Figure 2.15. Plg7p shows similar sensitivity to serine-directed inhibitors as human PAF-AH2. PAF-AH activity assay of cell lysates from p/g7+ or human PAF-AH2··pDEST52 DY 1838 transformed cells, pretreated with 1 mM Pefabloc SC or MAFP. n=3, * = P<O.05 vs Pefabloc SC, MAFP (One Way ANOY A, Tukey Test), 64 -Q 0 't: -J: ::::IE Q. -Q 'tl Q) 1/1 CIS Q) ~ -Q) ! .'i J: ~ 600 500 400 300 200 100 0 Rec: plasma PAF-AH/10 pDEST52-plg7+ o Day 0 Il:I Day 1 • Day 2 Figure 2.16. Plg7p is fairly stable after induced expression in S. cerevisiae. PAF-AH assay of intact INVSc1 transformed with pDEST52-plg7+, switched from SC-ura Galactose media to SC-ura Glucose media for the indicated number of days. Cells were concentrated to an OD of 40, and incubated directly in the reaction without lysis. Recombinant human plasma PAF-AH was used as a positive control for the assay. 65 66 crease of activity over mock (Figure 2.17). Thus there is some unknown inherent property of pig 7+ that prevents it from being expressed or active in mammalian cells, but could reflect codon usage differences. Gene Replacement of Plg7p S. pombe has a single homolog ofPAF-AH by sequence analysis that was targeted by gene replacement to produce Ilplg7+::his7+. A plasmid construct was generated with approximately 200 bp of the 5' and 3' ends ofplgT flanking the his7+ gene from pEA2 (Figure 2.18). Transfonmation of strain CHP428 with the gene replacement construct yielded colonies that were selected for stable incorporation of the plasmid. Successful integration was analyzed in 85 colonies using PCR primers internal to the his7+ gene and external to plgT for detection of Ilplg7 (Figure 2.18). Generation of Ilplg7 was successful in 5 out of 85 colonies, illustrated in the PCR results from one of the colonies (Figure 2.19). Appropriate sized bands for replacement were present in Ilplg7 but not in wild type cells; the opposite was true for detection of intact plg7+ (Figure 2.19) and confirmed by sequence a:nalysis. We found that Ilplg7 displayed reduced PAF hydrolytic activity compared to wild type S. pombe, but that half of this activity remained after p/gT deletion (Figure 2.20). The selective inhibitor Pe:fabloc reduced wild type P AF hydrolysis by 50 percent, and the residual activity in IIp/g7 was not sensitive to Pefabloc SC. Similar results were obtained in experiments using MAFP (data not shown). A significant change in phenotype of Ilplg7 versus wild-type was not observed using various challenges (heavy metal, oxidative stress, or temperature). S. pombe therefore constitutively express at least two 7000 -I I -C) :::l 6000 I:: -.s::: :D!E.. 5000 -C " 4000 G) ecna .G!!) 3000 0:: S -ca 2000 G) to) <C 1000 -; ::I: (") 0 Mock pC I-Neo-plg7 + pCMV-HPAF-AH2 Figure 2.17. Plg7p without an altered C-terminus does not express detectable soluble PAF-AH activity above background when expressed in a mammalian system. CMV-promoter based plasmids expressing pig 7+ or human PAF-AH2 were transfected into COS-7 cells. Cells were lysed in a Tris-CHAPS buffer and assayed for PAF-AH activity. 67 d4 ~ Wild type ~ Y5 d4 ~ I1plg7 ~ ~ d3 d1 Figure 2.18. Gene deletion scheme to replace plg7+ with his 7+ in S. pombe strain CHP428. Approximately 200 base pairs of both ends of pig 7+ were added to a construct with the his7+ gene. Translation start sites are indicated with black arrows. 68 A B Wt l!plg? Wt !!pIg? Wt !!pIg? primers: d1+d2 d3+d4 d4+Y5 predicted ~ ~ '----y----J product size: 646bp 770 bp 935 bp Figure 2.19. Successful replacement of pIg 7+ with his7+ in S. pombe. PCR analysis of one transformant and wild type CHP428 on both sides of the his7+ insertion using genome and his7+ specific primers (A). PCR detecting intact genomic pIg 7+ in wild type and /1plg7 cells (B). 69 .~:; 13 ca .i. .. .0. s:: CD ~ CD Q. 70 100 90 80 I ~j I 70 60 * * 50 • Mock 40 • Pefabloc 30 20 10 0 p/g7+ IIp/g7 Figure 2.20. /1p/g7 cells have reduced, but residual PAF-AH activity that is insensitive to the serine-directed inhibitor Pefabloc SC. PAF-AH activity assay of OD550 normalized wild type and /1p/g7 cells, lysed and pretreated with mock or 100 j..lM Pefabloc SC. Values are expressed relative to mock treated wild type (wt) lysates, n=2, * = P<O.OOI vs p/gT (One Way Anova, Student t-test). 71 enzymes with phospholipase activity against short chain phospholipids, with apparently only one, Plg7p, containing a catalytically essential serine residue. Plg7p Ortholog Search Our experiments with Plg7p expression in S. cerevisiae revealed a baseline PAF-AH activity exists in S. cerevisiae. It was also observed in our S. pombe t::.p/g7 cells that a secondary source of P AF-AH activity exists. Finally, it was observed in both S. pombe and S. cerevisiae that the covalent PAF-AH inhibitors Pefabloc SC and MAFP had no affect on the unknown sources of activity. Since the residual PAF-AH activity in S. cerevisiae and S. pombe showed similar characteristics, presumably the genes responsible could contain significant homology. A system developed in S. cerevisiae to clone genes based on function was used to discover the possible source (35). The cloning by function assay is based on a color change in S. cerevisiae involving the adenine synthesis pathway. Cells with an ade2 mutation accumulate a red color when grown on media with low adenine. Cells with an ade3 mutation (a gene epistatic to ade2) and double mutant ade2 ade3 cells produce a white color on low adenine solid media. The ade2 ade3 double mutant is transformed with a plasmid expressing both ADE3 and a gene of interest. Since a certain amount of plasmid loss occurs in S. cerevisiae, a single colony can show red and white sectoring in the assay. Sectoring is the result of a portion of the cells losing the plasmid expressing ADE3 within a colony, thus reverting to a white phenotype which will also be seen in its progeny. Thus random mutations are introduced into ade2 ade3 cells transformed with ADE3 and the gene of interest, and the viable cells are screened for those that lack red/white sectoring and are completely red. The red cells 72 potentially have a mutation in an essential orthologous gene to the one expressed on the plasmid, and thus loss of that plasmid results in death. The cloning by function assay was recapitulated using HPAF-AH2 as the gene of interest. Strain DY 4165 was used (ade2 ade3) for the assay, and was transformed with pDEST52-HPAF-AH2/ADE3+. The cells were treated with UV light at 30 mJ (a fairly toxic level) and plated on SC low adenine media with galactose and grown for 5 days at 35°C to select for potential temperature sensitive mutants. As shown in Table 2.2, approximately 0.4-4 percent of the cells were viable and resulted in a total number of 26000 viable colonies. 112 of the colonies appeared to be completely red on an initial screen, of which 8 were validated after restreaking on SC low adenine media with galactose and grown for 5 days at 35°C. Of the 8 positives, all were also completely red on YEPD media containing glucose instead of galactose, and thus are false positives since HPAF-AH2 cannot be expressed in the presence of glucose. Most likely, these cells have undergone homologous recombination in the ADE3 gene and thus have reverted to a single ade2 phenotype. Oxidation Experiments Overexpression of type II P AF -AH protects mammalian cells against oxidative stress (21, 22), presumably because toxic short chain phospholipids are substrates of the type II and plasma PAF-AH isoforms (13, 36, 37). We used the metal dependent oxidation model developed by Avery et al. (38) to test whether Plg7p expressed in S. cerevisiae conferred protection from oxidant injury. We found (Figure 2.21) that supplementation of INVScl with linolenic acid (18:3) caused a time dependent increase in fluorescence from Bodipy® 5811591 Cl1, an indicator lipid that shifts its fluorescence in response to oxida- Table 2.2 Cloning by Function Experiment to Find the PAF-AH Ortholog in S. cerevisiae Colonies screened Number of colonies 30000a Positive Positive after second screen aDY 4165 strain was transfonned with pDEST52-HPAF-AH2/ADE3+, spread onto 150 SC-Iow ade (galactose) plates, treated with 30 mJ UV light, and grown at 35°C for 5 days to develop red color indicating positive colonies. Viabilities of each plate with UV treated cells were about 4-0.4 percent compared to untreated control. bColonies that were completely red after growing for 5 days at 35°C. cColonies positive from the initial screen were restreaked onto SC-Iow ade (galactose) plates and grown at 35°C for confinnation. 73 G) U C G) u II) ! 0 ~ u::: c CIS G) :E 100 --0- None 90 -0- +18:3 80 - +50 uM CuS04 70 -- +18:3,50 uM CuS04 60 50 40 30 5 15 25 35 45 55 65 75 Time (minutes) Figure 2.21. S. cerevisiae supplemented with 18:3 and treated with CuS04 show an increased oxidized environment over nonsupplemented and nontreated cells. INVSc1 cells loaded with Bodipy® 5811591 C11 were examined for oxidation products by fluorimetry using 488 nm excitation, 528 emission in a representative experiment done in duplicate. 74 75 tion. The addition ofthe transition metal Cu+ alone to INVScl did not indicate membrane oxidation, but a combination of linolenic acid supplementation followed by exposure to Cu+ increased the fluorescence over that produced by 18:3 supplementation alone. We used confocal microscopy to image the oxidative environment detected by the Bodipy® 581/591 Cll dye, and found punctate intracellular staining with a diffuse halo closely associated with the plasma membrane (Figure 2.22). However, there was marked variation in the level of membrane oxidative stress individual cells experienced, with some cells apparently unaffected by Cu+ exposure. Previous works shows there is individual cell variation in this system (39) for undefined reasons. We considered that intracellular Cu + might be limiting because Cu + limits its own uptake (40) and so used fet3!1 cells that displays extreme sensitivity to Cu + due to abnormalities in Fe/Cu homeostasis (41). These cells displayed increased levels of Bodipy® 5811591 C 11 fluorescence in an oxidized lipid environment when supplemented with C18:3 and treated with Cu+, and this was suppressed by y-tocopherol (Figure 2.23). These results were confirmed by measuring oxidation in a fluorimeter with intactfet3~ cells (Figure 2.24). Thus, an antioxidant effectively inhibited formation of oxidized lipids in our model. However, even these cells were heterogeneous in their response to oxidative stress. Although phenotypic heterogeneity was ameliorated in fet3~ cells, it was still observed that some cells within the culture were more prone to producing oxidized lipid­dependent fluorescence. Indeed this can be seen in the transmission images from an oxidation experiment (Figure 2.23). The cells which have dark invaginations were responsible for all the fluorescence in the 18:3 supplemented culture, while opaque cells Figure 2.22. An oxidation environment is detected in 18:3 supplemented, and copper treated S. cerevisiae cells. INVSc1 strain was supplemented with 18:3, loaded with Bodipy® 581/591 C11, and treated with 50 ~M CUS04 for 60 minutes. Cells were visualized by confocal microscopy using a 488 nm excitation laser (left), or by transmitted light (right). Scale bar = 10 ~m. 76 Non-supplemented +18:3 +18:3+ Vitamin E Figure 2.23. fet313. S. cerevlszae show increased oxidation over a population of cells when supplemented with 18:3 and treated with copper. Confocal microscopy of fet313. S. cerevisiae incubated with Bodipy CII 581/591 fluorescent oxidation marker, then treated with 50 ).lM CUS04 for I hour. For the vitamin E sample, 200 ).lM 1-Tocopherol was added to the cells while incubating with Bodipy C11 and washed prior to treatment with copper. Positive oxidation signal is indicated in green. Phenotypic heterogeneity in response to copper is shown with arrows. 77 25 20 CD u c::: u(I) 15j II) .(.I.) • None 0 ::3 u:: 10 -, III +200 uM Vitamin E c::: CG (I) :liE 5 0 i ~J --~------............ 18:3 Figure 2.24. Vitamin E inhibits copper induced oxidation of 18:3 supplementedfet3~ S. cerevisiae. Fluorimetric assay offet3~ S. cerevisiae incubated with Bodipy CII 5811591 fluorescent oxidation marker, then treated with 50 JlM CUS04 for I hour. Prior to incubation with Bodipy Cll, cells were grown overnight in media supplemented with I mM 18:3. For the vitamin E sample, 200 JlM "{-Tocopherol was added to the cells while incubating with Bodipy CII and washed prior to treatment with copper. Fluorescence values were subtracted by values from 18:3 supplemented cells alone. 78 79 did not produce any oxidation induced fluorescence (Figure 2.23, arrows). The observed opaque and dark phenotypes do not seem to be due to an oxidation induced effect since the same populations of cells are observed in nonsupplemented and vitamin E treated cultures. Further experiments may shed light on the nature of phenotype heterogeneity based on these initial observations. We reduced the effect of individual cell variation to oxidative stress (39) by plating INVScl to assess colony number, an approach where the background of cells not subject to oxidative stress is quantifiable. We treated INVSc 1 with Cu + for 0, 1 ° or 60 minutes in culture and then plated serial dilutions to immobilize the cells. Exposure to Cu + reduced the number of viable cells in a time-dependent way (Figure 2.25), and cells supplemented with C 18:3 prior to this exposure displayed enhanced sensitivity to the transition metal. Introduction of a p/g7+ expression plasmid into copper challenged, 18:3- supplemented INVScl suppressed Cu+ toxicity (Figure 2.26), and resulted in an average 6-fold increase in viability compared to no plasmid (Figure 2.27). Copper challenged, 18:3-supplementedfet3~ cells, expressing Plg7p also had increased viability compared to cells expressing the inactive Plg7p-S257 A (Figure 2.28). Similar levels of oxidation were detected by Bodipy® 5811591 Cll fluorimetry in both Plg7p and Plg7p-S257A expressing groups (not shown), so each strain encountered the same level of oxidative stress. Finally, similar results were observed infot311 expressing human type 2 PAF-AH compared to empty vector (Figure 2.29). These results show that S. cerevisiae is sensitive to oxidation when supplemented with polyunsaturated fatty acid and that viability can be protected by overexpression ofPlg7p. Time (min) o 15 60 Mock 18:3 Supplemented Figure 2.25. S. cerevisiae cells supplemented with 18:3 are more sensitive to copper induced cell death. Viability assay of INVSc1 cells -/+ 18:3 supplementation, treated with 50 ~M CUS04 for 10 or 60 minutes, diluted 10-fold serially and plated on YEPD plates. 80 o 15 60 Empty Vector p/g7+ Figure 2.26. Plg7p expressing cells are more resistant to copper induced oxidative death. 18:3 supplemented INVScl cells were transformed with empty vector or pDEST52-plgT and challenged with 50 ~M CUS04 for indicated times, then serially diluted and plated to test viability. 81 c 0 :0:; (,) -d) .0.. C. 'tJ '0 LL 10 9 8 7 6 4 1 3 j 2 ~ 1 0 ~ - -,-~------ no plasmid plg7+ Figure 2.27. Plg7p expressing cells are more resistant to copper induced oxidative death in a cell counting assay. 18:3 supplemented INVScl cells were transformed with or without pDEST52-plg7+ and challenged with 25-50 ~M CUS04 for 60 minutes, then serially diluted and counted. Viability was determined by comparing number of colonies before and after treatment. Values are represented as fold protection compared to no plasmid (n=2). 82 -, p/g7-S2S7A p/g?+ o 15 60 Figure 2.28. Plg7p expressing fet3!1 cells are more resistant to copper induced oxidative death than cells expressing the S257 A active site mutant. Viability assay using fet3!1 cells transformed with pDEST52- plg7+ or plg7-S257A, supplemented with 1 mM 18:3, treated with 50 J.tM CuS04, serially diluted and plated onto YEPD. Results are representative of 3 separate experiments. 83 Time (min) o 15 60 Empty HPAF-AH2 Figure 2.29. Human PAF-AH2 expressingfet38 cells are more resistant to copper induced oxidative death than cells with empty vector. Viability assay usingfet38 cells transformed with HPAF-AH2-pDEST52 or empty vector, supplemented with 1 mM 18:3, treated with 50 JlM CuS04, serially diluted and plated onto YEPD. Results are representative of 2 separate experiments. 84 85 Discussion The genome of S. pombe, a fission yeast distantly related to both animals and the budding yeast S. cerevisiae, has been fully sequenced revealing a genome with introns, genes with homology to human disease genes, and clusters of genes regulating cell function related to those found in higher eukaryotes (25). The Gene Ontology Project (http:www.geneontology.org) infers by sequence homology that SPBC106.11c should be a member of the phospholipase family because sequence codes for the lipase motif GXSXG, and the positions of the aspartyl and histidyl residues are the same in Plg7 as in the two mammalian phospholipases. Blast analysis shows the open reading frame SPB 1 06.11 c is the only related sequence in S. pombe, and that S. cerevisiae lacks related sequences. We cloned the S. pombe open reading frame SPBC 106.11 c and expressed it in S. cerevisiae, which lacks confounding sequence homologs, to find that this gene encodes a functional member of the group VII phospholipase family. We found that like the mammalian group VII enzymes Plg7p was Ca ++ -independent and required the Ser residue in the midst of the GXSXG lipase motif for catalysis. Additional evidence that Plg7 contains essential features of the mammalian ortholog is that Plg7 was inhibited by the selective serine-directed inhibitors Pefabloc and MAFP that irreversibly inhibit human plasma P AF acetylhydrolase. The group VII family of phospholipases A2 was originally described in studies of P AF catabolism (9), but their role in homeostasis has not yet been genetically defined. The only reported genetic ablation of a group VII family member was accomplished in C. eiegans, where disruption of the locus interfered with normal epithelial morphogenesis 86 (23). The two mammalian members of the family (3, 36) and this C. elegans ortholog (23) all hydrolyze phospholipids other than PAF when the sn-2 residue is short or oxygenated. PAF is the only described phospholipid with a radically short sn-2 residue produced by normal metabolism, but such lipids are formed by the nonenzymatic oxidation of sn-2 polyunsaturated fatty acyl residues. The ectopically expressed S. pombe enzyme hydrolyzed the short chain phospholipid P AF and it hydrolyzed oxidized dioleoyl phosphatidy1choline after oxidative fragmentation. Like other members of this phospholipase family, Plg7p does not hydrolyze phospholipids when the sn-2 residue is a long chain acyl residue. s. pombe as a free living organism encounters environmental stresses, and several hundred genes are induced in response to events such as the oxidative load presented by H20 2 and Cd++ exposure (42). The open reading frame SPBCI06.l1c was among the genes rapidly responding to these two oxidative stresses with increased message accumulation (42). S. pombe phospholipids have a remarkably homogenous fatty acyl chain composition, where 85 percent of their fatty acids are esterified oleoyl residues (26). This monounsaturated fatty acids will oxidize, albeit less readily than polyunsaturated fatty acids (15). S. pombe, like S. cerevisiae, lack the desaturase that would introduce a second double bond and so make no polyunsaturated fatty acids. However, these yeasts will (26, 38), and do (43), accumulate and incorporate polyunsaturated fatty acids from their environment. We found that deletion of the pig 7+ locus caused no overt change in the phenotype of cells whether or not they had been supplemented with polyunsaturated fatty acid and challenged with Cu+. We also found that S. pombe expressed an unknown activity that 87 was as effective as Plg7p in hydrolyzing short chain phosphatidylcholines. Lacking a way to test our hypothesis in S. pombe that Plg7p protects the organism from the consequences of membrane lipid oxidation, we ectopically expressed Plg7p in S. cerevisiae to test its function in a system where lipid peroxidation is lethal. The level of polyunsaturated fatty acids can be increased in S. cerevisiae by supplementing their growth media with linoleate, rendering them susceptible to Cu+-induced lipid peroxidation and cell death (38). We found by serial plating that S. cerevisiae expressing Plg7p were protected against this oxidative stress, and that this protection required the deduced active site serine be present. We found a similar level of protection when we expressed the mammalian group VII enzyme PAF-AH2. The results from this reduced system indicate that membrane lipid peroxidation and death can be suppressed after oxidative attack, and that both the human and S. pombe enzyme act in the same way to hydrolyze structurally damaged phospholipids and maintain viability. Cloning the entire genome of S. pombe and expression tagging the resulting proteins reveals that SPBC 106.11 c is a minor protein expressed equally in nuclear and cytoplasmic compartments (45). The motif dictating this location is not clear from the primary sequence, but we noted that introduction of any of several tags at the C-tem1inus of Plg7p or human PAF-AH2 reduced activity. The intracellular group VII family member PAF AH2 has a sequence directing its N-terminal myristoylation, and incorporates radio labeled palmitate when expressed in cultured mammalian cells (21). Immunofluorescence microscopy shows PAF-AH2 protein to be distributed between the cytosol and membranes close to the nucleus and is translocated to cellular membranes when the cells are challenged by exposure to a lipid soluble hydroperoxide (21). Plg7p 88 retains the N tenninal sequence MGXXXS myristoylation sequence (46) found in PAF­AH2 (21), and the S. pombe ortholog of neuronal calcium sensors requires N-tenninal myristolation to translocate from cytosol to membranes (47). Recent data from our lab shows that certain oxidized phospholipids are toxic because they specifically damage mitochondrial integrity and start the intrinsic apoptotic pathway (McIntyre et al. submitted), and targeting oxidized phospholipid phospholipase activity to the membrane may be an effective way to maintain membrane structure. S. pombe contains neither the ether phospholipids required to synthesize high affinity PAF receptor ligands, enzymes of ether lipid metabolism, nor a relevant G protein receptor for PAF. Instead, our data are consistent with Plg7p primarily protecting against the consequences of phospholipid peroxidation. S. pombe (25) has a spare genome and expresses just 4900 genes, suggesting the protection conferred by Plg7p is crucial to this organism. This function is retained by the orthologous mammalian enzymes, and a primary purpose of the group VII phospholipases may be catabolism of the unusual phospholipid structures fonned by unregulated nonenzymatic phospholipid oxidation that starts the process of regulated cell death. 89 References 1. Arai, H., Koizumi, H., Aoki, J., and Inoue, K. (2002) J Biochem (Tokyo) 131(5), 635-640. 2. Tjoelker, L. W., Eberhardt, c., Unger, J., Trong, H. L., Zimmerman, G. A, McIntyre, T. M., Stafforini, D. M., Prescott, S. M., and Gray, P. W. (1995) Journal of Biological Chemistry 270, 25481-25487. 3. Hattori, K., Adachi, H., Matsuzawa, A, Yamamoto, K., Tsujimoto, M., Aoki, J., Hattori, M., Arai, H., and Inoue, K. (1996)J BioI Chem 271(51), 33032-33038. 4. Gomes, R. N., Bozza, F. A, Amancio, R. T., Japiassu, AM., Vi anna, R. C., Larangeira, A. P., Gouvea, J. M., Bastos, M. S., Zimmerman, G. A, Stafforini, D. M., Prescott, S. M., Bozza, P. T., and Castro-Faria-Neto, H. C. (2006) Shock 26(1),41-49. 5. Prescott, S. M., Zimmerman, G. A, Stafforini, D. M., and McIntyre, T. M. (2000) Ann. Rev. Biochem. 69,419-445. 6. Ishii, S., Nagase, T., and Shimizu, T. (2002) Prostaglandins Other Lipid Mediat 68-69,599-609. 7. O'Flaherty, J. T., Salzer, W. L., Cousart, S., McCall, C. E., Piantadosi, C., Surles, J. R., Hammett, M. J., and Wykle, R. L. (1983) Research Communications in Chemical Pathology and Pharmacology 39,291-309. 8. Shen, T. Y., Hwang, S.-B., Doebber, T. W., and Robbins, J. C. (1987) The chemical and biological properties of P AF agonists, antagonists, and biosynthetic inhibitors. In: Snyder, F. (ed). Platelet-Activating Factor and Related Lipid Mediators, Plenum Press, New York. 9. FaIT, R. S., Cox, C. P., Wardlow, M. L., and Jorgensen, R. (1980) Clinical Immunology and Immunopathology 15, 318-330. 10. Tjoelker, L. W., and Stafforini, D. M. (2000) Biochim Biophys Acta 1488(1-2), 102-123. 11. Zimmerman, G. A, McIntyre, T. M., Mehra, M., and Prescott, S. M. (1990) Journal of Cellular Biology 110,529-540. 12. Hattori, K., Hattori, M., Adachi, H., Tsujimoto, M., Arai, H., and Inoue, K. (1995) J BioI Chem 270(38), 22308-22313. 13. Stremler, K. E., Stafforini, D. M., Prescott, S. M., and McIntyre, T. M. (1991) J BioI. Chem. 266, 11095-11103. 90 14. Ho, Y. S., Sheffield, P. 1., Masuyama, 1., Arai, H., Li, J., Aoki, J., Inoue, K., Derewenda, u., and Derewenda, Z. S. (1999) Protein Eng 12(8), 693-700. 15. Frankel, E. N. (1984) Progress in Lipid Research 23, 197-221. 16. Podrez, E. A., Poliakov, E., Shen, Z., Zhang, R., Deng, Y., Sun, M., Finton, P. J., Shan, L., Gugiu, R, Fox, P. L., Hoff, H. F., Salomon, R. G., and Hazen, S. L. (2002) J BioI Chem 277(41),38503-38516. 17. Tokumura, A., Sumida, T., Toujima, M., Kogure, K., Fukuzawa, K., Takahashi, Y., and Yamamoto, S. (2000) J Lipid Res 41(6),953-962. 18. Itabe, H., Kushi, Y., Handa, S., and Inoue, K. (1988) Biochim. Biophys. Acta 962, 8-15. 19. Ogden, F., DeCoster, M. A., and Bazan, N. G. (1998) J Neurosci Res 53(6),677- 684. 20. Hirashima, Y., Veno, H., Karasawa, K., Yokoyama, K., Setaka, M., and Takaku, A. (2000) Brain Res 885(1), 128-132. 21. Matsuzawa, A., Hattori, K., Aoki, J., Arai, H., and Inoue, K. (1997) J BioI Chem 272(51),32315-32320. 22. Marques, M., Pei, Y., Southall, M. D., Johnston, J. M., Arai, H., Aoki, J., Inoue, T., Seltmann, H., Zouboulis, C. c., and Travers, J. R (2002) J Invest Dermatol 119(4),913-919. 23. Inoue, T., Sugimoto, A., Suzuki, Y., Yamamoto, M., Tsujimoto, M., Inoue, K., Aoki, J., and Arai, H. (2004) Proc Nat! Acad Sci USA 101(36), 13233-13238. 24. Koizurni, H., Yamaguchi, N., Hattori, M., Ishikawa, T. 0., Aoki, J., Taketo, M. M., Inoue, K., and Arai, H. (2003) J BioI Chem 278(14), 12489-12494. 25. Wood, V., Gwilliam, R., Rajandream, M. A., Lyne, M., Lyne, R., Stewart, A., Sgouros, J., Peat, N., Hayles, J., Baker, S., Basham, D., Bowman, S., Brooks, K., Brown, D., Brown, S., Chillingworth, T., Churcher, C., Collins, M., Connor, R., Cronin, A., Davis, P., Feltwell, T., Fraser, A., Gentles, S., Goble, A., Hamlin, N., Harris, D., Hidalgo, 1., Hodgson, G., Holroyd, S., Hornsby, T., Howarth, S., Huckle, E. J., Hunt, S., Jagels, K., James, K., Jones, L., Jones, M., Leather, S., McDonald, S., McLean, 1., Mooney, P., Moule, S., Mungall, K., Murphy, L., Niblett, D., Odell, C., Oliver, K., O'Neil, S., Pearson, D., Quail, M. A., Rabbinowitsch, E., Rutherford, K., Rutter, S., Saunders, D., Seeger, K., Sharp, S., Skelton, J., Simmonds, M., Squares, R., Squares, S., Stevens, K., Taylor, K., Taylor, R. G., Tivey, A., Walsh, S., Warren, T., Whitehead, S., Woodward, J., Volckaert, G., Aert, R., Robben, J., Grymonprez, B., Weltjens, 1., Vanstreels, E., 91 Rieger, M., Schafer, M., Muller-Auer, S., Gabel, C., Fuchs, M., Dusterhoft, A., Fritze, c., Holzer, E., Moestl, D., Hilbert, H., Borzym, K., Langer, 1., Beck, A., Lehrach, H., Reinhardt, R., Pohl, T. M., Eger, P., Zimmermann, W., Wedler, H., Wambutt, R., Pumelle, B., Goffeau, A., Cadieu, E., Dreano, S., Gloux, S., Lelaure, V., Mottier, S., Galibert, F., Aves, S. 1., Xiang, Z., Hunt, C., Moore, K., Hurst, S. M., Lucas, M., Rochet, M., Gaillardin, c., Tallada, V. A., Garzon, A., Thode, G., Daga, R. R., Cruzado, L., Jimenez, J., Sanchez, M., del Rey, F., Benito, 1., Dominguez, A., Revuelta, J. L., Moreno, S., Armstrong, J., Forsburg, S. L., Cerutti, L., Lowe, T., McCombie, W. R., Paulsen, 1., Potashkin, 1., Shpakovski, G. V., Ussery, D., Barrell, B. G., Nurse, P., and Cerrutti, L. (2002) Nature 415(6874),871-880. 26. McDonough, V. M., and Roth, T. M. (2004) Antonie Van Leeuwenhoek 86(4), 349-354. 27. Ling, M., Merante, F., and Robinson, B. H. (1995) Nucleic Acids Res 23(23), 4924-4925. 28. Chen, H. R., Hsu, M. T., and Cheng, S. C. (1995) Biotechniques 19(5), 744-746, 748. 29. Tjoelker, L. W., Wilder, C., Eberhardt, c., Stafforini, D. M., Dietsch, G., Schimpf, B., Hooper, S., Le Trong, H., Cousens, L. S., Zimmerman, G. A., Yamada, Y., McIntyre, T. M., Prescott, S. M., and Gray, P. W. (1995) Nature 374(6522),549-553. 30. Wei, Y., Swenson, L., Castro, C., Derewenda, U., Minor, W., Arai, H., Aoki, J., Inoue, K., Servin-Gonzalez, L., and Derewenda, Z. S. (1998) Structure 6(4), 511- 519. 31. Albert, D. H., and Snyder, F. (1983) J Bioi Chern 258( 1), 97-102. 32. Hanson, P. K. 33. Dentan, c., Tse1epis, A. D., Chapman, M. J., and Ninio, E. (1996) Biochim Biophys Acta 1299(3),353-357. 34. Kell, P. J., Creer, M. H., Crown, K. N., Wirsig, K., and McHowat, J. (2003) J Pharmacol Exp Ther 307(3), 1163-1170. 35. Kranz, J. E., Holm, C. (1990) Proc Natl Acad Sci USA 87(17), 6629-6633. 36. Stremler, K. E., Stafforini, D. M., Prescott, S. M., Zimmerman, G. A., and McIntyre, T. M. (1989) Journal of Biological Chemistry 264, 5331-5334. 92 37. Hattori, M., Arai, H., and Inoue, K. (1993) Journal of Biological Chemistry 268, 18748-18753. 38. Avery, S. V., Howlett, N. G., and Radice, S. (1996) Appl Environ Microbiol 62(11),3960-3966. 39. Howlett, N. G., and Avery, S. V. (1999) FEMS Microbiol Lett 176(2), 379-386. 40. Eide, D. J. (1998) Annu Rev Nutr 18, 441-469. 41. Shi, x., Stoj, C., Romeo, A., Kosman, D. J., and Zhu, Z. (2003) J BioI Chem 278(50),50309-50315. 42. Chen, D., Toone, W. M., Mata, J., Lyne, R., Bums, G., Kivinen, K., Brazma, A., Jones, N., and Bahler, 1. (2003) Mol Bioi Cell 14(1), 214-229. 43. Howlett, N. G., and Avery, S. V. (1997) Appl Environ Microbiol 63(8), 2971- 2976. 44. Blagovic, B., Rupcic, J., Mesaric, M., and Marie, V. (2005) Folia Microbiol (Praha) 50(1), 24-30. 45. Matsuyama, A., Arai, R., Yashiroda, Y., Shirai, A., Kamata, A., Sekido, S., Kobayashi, Y., Hashimoto, A., Hamamoto, M., Hiraoka, Y., Horinouchi, S., and Yoshida, M. (2006) Nat BiotechnoI24(7), 841-847. 46. Resh, M. D. (1999) Biochim Biophys Acta 1451(1), 1-16. 47. Hamasaki-Katagiri, N., Molchanova, T., Takeda, K., and Ames, J. B. (2004) J Bioi Chem 279(13), 12744-12754. 3. MEGAKARYOCYTES DERIVED FROM CD34+ CELLS EXPRESS PAF ACETYLHYDROLASE TO CONTROL PAF-LIKE LIPID ACCUMULATION AND SIGNALING Summary Platelet activating factor (P AF) is an inflammatory phospholipid that signals through a seven-transmembrane G-protein receptor present on many cells, including platelets. Oxidation of unsaturated fatty acyl residues of phospholipids can lead to the formation of PAF-like lipids which can also signal through the PAF receptor. Both PAF and PAF-like lipids are hydrolyzed by PAF acetylhydrolases (PAF-AHs) to the relatively inactive Lyso-PAF. We have observed that megakaryocytes, which are platelet precursors, have increased PAF-AH activity over differentiation from CD34+ stem cells from human umbilical cord blood. Inhibition ofPAF-AH activity with Pefabloc SC in megakaryocytes led to increased formation of PAF-like lipids that were found to signal through endogenous P AF receptors. Transient changes in morpholog