| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Review Article |
Department of Molecular and Cellular Biochemistry, The Ohio State University, Columbus, Ohio (A.A.F., L.A.H.); and Department of Anatomy, National University of Singapore, Singapore, Singapore (W.-Y.O.)
Abstract
Abstract I. Introduction II. Multiplicity of Phospholipases A2 in Brain A. Secretory Phospholipase A2 B. Cytosolic Phospholipases A2 C. Plasmalogen-Selective Phospholipase A2 D. Calcium-Independent Phospholipases A2 III. Involvement of Phospholipase A2 Activity in Brain Injury IV. Physiological and Pharmacological Effects of Phospholipase A2 Inhibitors A. Arachidonyl Trifluoromethyl Ketone B. Methyl Arachidonyl Fluorophosphonate C. Bromoenol Lactone D. Benzenesulfonamides and Alkoxybenzamidines E. 3-(Pyrrol)-2-propionic Acid F. 2-Oxoamide and 1,3-Disubstituted Propan-2-ones G. Choline Derivatives with a Long Aliphatic Chain H. Pyrrolidine-Based Inhibitors of Phospholipase A2 I. Antimalarial Drugs J. Lithium Ion and Carbamazepine K. Vitamin E and Gangliosides L. Cytidine 5-Diphosphoamines M. Long-Chain Polyunsaturated Fatty Acids N. Phospholipase A2 Antisense Oligonucleotides and Interfering RNA V. Phospholipase A2 Activity in Kainic Acid-Induced Neural Cell Injury VI. Protection of Kainic Acid-Induced Neural Cell Injury by Phospholipase A2 Inhibitors VII. Phospholipase A2 Activity in Neurological Disorders A. Ischemia B. Alzheimer's Disease C. Experimental Model of Parkinson's Disease D. Multiple Sclerosis and Experimental Autoimmune Encephalomyelitis E. Prion Diseases F. Spinal Cord Injury G. Head Injury H. Epilepsy I. Schizophrenia and Depressive Disorders VIII. Use of Phospholipase A2 Inhibitors for the Treatment of Neurological Disorders IX. Prevention of Pain by Phospholipase A2 Inhibitors X. Perspective and Direction for Future Studies
The phospholipase A2 family includes secretory phospholipase A2, cytosolic phospholipase A2, plasmalogen-selective phospholipase A2, and calcium-independent phospholipase A2. It is generally thought that the release of arachidonic acid by cytosolic phospholipase A2 is the rate-limiting step in the generation of eicosanoids and platelet activating factor. These lipid mediators play critical roles in the initiation and modulation of inflammation and oxidative stress. Neurological disorders, such as ischemia, spinal cord injury, Alzheimer's disease, multiple sclerosis, prion diseases, and epilepsy are characterized by inflammatory reactions, oxidative stress, altered phospholipid metabolism, accumulation of lipid peroxides, and increased phospholipase A2 activity. Increased activities of phospholipases A2 and generation of lipid mediators may be involved in oxidative stress and neuroinflammation associated with the above neurological disorders. Several phospholipase A2 inhibitors have been recently discovered and used for the treatment of ischemia and other neurological diseases in cell culture and animal models. At this time very little is known about in vivo neurochemical effects, mechanism of action, or toxicity of phospholipase A2 inhibitors in human or animal models of neurological disorders. In kainic acid-mediated neurotoxicity, the activities of phospholipase A2 isoforms and their immunoreactivities are markedly increased and phospholipase A2 inhibitors, quinacrine and chloroquine, arachidonyl trifluoromethyl ketone, bromoenol lactone, cytidine 5-diphosphoamines, and vitamin E, not only inhibit phospholipase A2 activity and immunoreactivity but also prevent neurodegeneration, suggesting that phospholipase A2 is involved in the neurodegenerative process. This also suggests that phospholipase A2 inhibitors can be used as neuroprotectants and anti-inflammatory agents against neurodegenerative processes in neurodegenerative diseases.
Phospholipases A2 (PLA21; EC 3.1.1.4) form an expanding superfamily of esterases that specifically cleave the acyl ester bond at the sn-2 position of membrane phospholipids to produce a free fatty acid and lysophospholipid (Farooqui et al., 2000b
). Because a large proportion of cellular arachidonic acid is found esterified at the sn-2 position of membrane phospholipids, arachidonic acid and lysophospholipid are the major products of the PLA2-catalyzed reaction. Under normal conditions, some arachidonic acid is converted to inflammatory mediators, prostaglandins, leukotrienes, and thromboxanes, whereas a majority of arachidonic acid is reincorporated into the brain phospholipids (Rapoport, 1999; Leslie, 2004). Arachidonic acid not only acts via conversion to inflammatory metabolites, but can also directly modulate neuronal function by various mechanisms, such as altering membrane fluidity and polarization state, activating protein kinase C, and regulating gene transcription (Katsuki and Okuda, 1995; Farooqui et al., 1997b). Another product of PLA2 catalyzed reactions, 1-alkyl-2-lysophospholipid, is the immediate precursor of platelet-activating factor (PAF), another potent inflammatory mediator (Farooqui and Horrocks, 2004b).
Lysophospholipids may also change membrane fluidity and permeability. These metabolites are also involved in phospholipid remodeling and membrane perturbation. Accumulation of lysophospholipids is controlled by either reacylation to native phospholipids (Farooqui et al., 2000b) or by metabolism to water-soluble glycerophosphodiesters such as glycerophosphocholine by lysophospholipases (Farooqui et al., 1985). Thus, tight regulation of PLA2 activity is necessary for maintaining basal levels of arachidonic acid, lysophospholipid, and PAF for performing normal brain function.
Increased PLA2 activity and excessive production of proinflammatory mediators, eicosanoids, and platelet-activating factor, may potentially lead to disease states and neuronal injury. Collective evidence from many recent studies suggests that increased PLA2 activity and PLA2-generated mediators play a central role not only in acute inflammatory responses in brain but also in oxidative stress associated with neurological disorders such as ischemia, Alzheimer's disease (AD), Parkinson's disease (PD), and multiple sclerosis (MS) (Kalyvas and David, 2004; Phillis and O'Regan, 2004; Sun et al., 2004). PLA2 contributes to the pathogenesis of the above disorders by attacking neural membrane phospholipids and releasing proinflammatory lipid mediators such as prostaglandins, leukotrienes, and thromboxanes, and PAF, and also by generating 4-hydroxynonenal (4-HNE). Thus, inhibition of cPLA2 activity provides an attractive approach for designing novel drugs for the treatment of inflammation and oxidative stress associated with acute neural trauma such as ischemia, spinal cord injury, and head injury and some neurodegenerative disorders such as AD, PD, and MS. A definitive proof of whether enhanced PLA2 activity represents a part of the cellular defense mechanism or whether increased PLA2 activities contribute to the pathology of the neurological disorder awaits the development of a potent, specific, and clinically useful PLA2 inhibitor for the treatment of inflammation and oxidative stress associated with neurological disorders. The purpose of this review is to describe the current knowledge of PLA2 involvement in neurological disorders and the usefulness of PLA2 inhibitors in preventing neurotoxin-mediated damage to neurons, glial cells, and the myelin sheath in brain tissue. This discussion should initiate more studies on not only the involvement of PLA2-mediated inflammation and oxidative stress in neurological disorders but also on the development of potent, specific, and nontoxic inhibitors of PLA2 activity that can cross the blood-brain barrier without harm and can be used for the treatment of neurological disorders.
II. Multiplicity of Phospholipases A2 in Brain
Recent advances in molecular and cellular biology of PLA2 have led to the identification of more than 20 isoforms with PLA2 activity. PLA2 enzymes are subdivided into several groups depending upon their structure, enzymic properties, subcellular localization, and cellular function (Table 1) (Farooqui et al., 1997c; Chakraborti, 2003; Phillis and O'Regan, 2004; Sun et al., 2004). These groups include secretory phospholipase A2 (sPLA2), cytosolic phospholipase A2 (cPLA2), plasmalogen-selective phospholipase A2 (PlsEtn-PLA2), and calcium-independent phospholipase A2 (iPLA2). Each class of PLA2 is further subdivided into isozymes of which there are 14 for sPLA2, at least 4 for cPLA2, and 2 for iPLA2. Genes coding for sPLA2, cPLA2, and iPLA2 have been shown to occur in different regions of brain, in neurons, microglia, and astrocytes (Molloy et al., 1998; Zanassi et al., 1998; Balboa et al., 2002). PLA2 isoforms have been partially purified and characterized from brain tissue (Hirashima et al., 1992; Ross et al., 1995; Yang et al., 1999), but none have been cloned and fully characterized. The following section briefly describes properties of brain PLA2 isoforms, their roles, and importance in signal transduction processes in brain metabolism (Fig. 1).
|
|
sPLA2 is synthesized intracellularly; then it is secreted and acts extracellularly. sPLA2 has a molecular mass of 14 kDa and is mainly associated with synaptosomes and synaptic vesicle fractions (Kim et al., 1995; Matsuzawa et al., 1996). PLA2 binds to two types of cell surface receptors, namely the N type, identified in neurons, and the M type, identified in skeletal muscles, of sPLA2 receptors (Hanasaki and Arita, 2002). Brain sPLA2 contains a secretion peptide and requires millimolar concentrations of Ca2+ for enzymic activity. It shows no selectivity for particular fatty acyl chains in the phospholipids. This enzyme is present in all regions of mammalian brain. The highest activities of sPLA2 are found in medulla oblongata, pons, and hippocampus, moderate activities in the hypothalamus, thalamus, and cerebral cortex, and low activities in the cerebellum and olfactory bulb (Thwin et al., 2003). At the cellular level, the sPLA2 transcript is found in astrocytes (Mosior et al., 1998b; Zanassi et al., 1998). sPLA2 is present in differentiated PC12 cells and in rat brain synaptic vesicles, indicating that neurons also express sPLA2 activity (Matsuzawa et al., 1996).
Rat brain synaptosomes or differentiated PC12 cells release sPLA2 upon stimulation via acetylcholine and glutamate receptors or via voltage-dependent calcium channels through depolarization. Thus, sPLA2 may play an important role in neuronal metabolism (Kim et al., 1995; Matsuzawa et al., 1996). Based on pharmacological studies, the sPLA2 released from neuronal cells may modulate the degranulation process leading to the release of neurotransmitters. Inhibitors of sPLA2 activity block this release. For the expression of neurotoxicity, the released sPLA2 binds to the presynaptic membrane, enters the lumen of the synaptic vesicle during retrieval of the vesicle from the plasma membrane, and hydrolyzes phospholipids of the inner leaflet of synaptic vesicles, changing the phospholipid composition and thus impairing its endocytosis. The stimulation of sPLA2 in synaptic vesicles correlates with the induction of vesicle-vesicle aggregation. This process plays a central role in presynaptic neurotransmission (Moskowitz et al., 1983; Matsuzawa et al., 1996; Wei et al., 2003). In brain, astrocytes express sPLA2, which can be induced in response to proinflammatory cytokines such as tumor necrosis factor-
and interleukin-1
(Lin et al., 2004).
Mitochondrial fractions from rat brain, PC12, and U251 astrocytoma cell cultures contain significant sPLA2 and iPLA2 activities (Macchioni et al., 2004). The mechanism for a secretory protein (like sPLA2) targeting an intracellular organelle (like mitochondria) remains unknown. However, it is proposed that at the molecular level, heparan sulfate, a glycosaminoglycan, may play an important role in internalization and attachment of PLA2 isoforms to intracellular organelles (Farooqui et al., 1994b; Boilard et al., 2003). A reduction in the mitochondrial membrane potential causes the release of sPLA2 and this sPLA2 along with other PLA2 isoforms may be involved in neural cell injury (Farooqui et al., 1997d; Macchioni et al., 2004).
Glutamate and its analogs stimulate sPLA2 activity in a dose- and time-dependent manner (Kim et al., 1995; Xu et al., 2003). The neurotoxicity of glutamate is synergistically increased with the addition of sPLA2 to cortical cultures. This observation suggests that glutamatergic synaptic activity may be modulated by sPLA2 and its receptors on the neuronal surface (DeCoster et al., 2002; Kolko et al., 2002). In PC12 cells, sPLA2 induces neurite outgrowth. Mutants with reduced sPLA2 activity exhibit a comparable reduction in neurite-inducing activity (Nakashima et al., 2003), indicating that sPLA2 performs a neurotrophin-like role in the central nervous system.
B. Cytosolic Phospholipases A2
Although brain tissue contains cPLA2 activity, it has never been purified to homogeneity and characterized from brain. The cytosolic fraction from rat brain contains two forms of PLA2 activity, PLA2-H and PLA2-L. PLA2-H has an apparent molecular mass of 200 to 500 kDa. Its activity is partially inhibited by Ca2+. In contrast, PLA2-L has a molecular mass of 100 kDa and requires Ca2+ (Yoshihara and Watanabe, 1990). Based on several enzymic properties such as Ca2+ sensitivity, molecular mass, and Ca2+-mediated translocation, PLA2-L seems to be identical to cPLA2 (Yoshihara et al., 1992). cPLA2 prefers arachidonic acid over other fatty acids and does not use Ca2+ for catalysis, although submicromolar Ca2+ concentrations are needed for membrane binding (Clark et al., 1987; Farooqui et al., 2000b). Owing to the presence of a Ca2+-dependent phospholipid-binding domain at the N-terminal region, cPLA2 is translocated in a Ca2+-dependent manner from cytosol to the nuclear or other cellular membranes (Clark et al., 1987; Hirabayashi et al., 2004), in which other downstream enzymes, including the cyclooxygenases and lipoxygenases responsible for the metabolism of arachidonic acid to eicosanoids, are located. This gives cPLA2 access to its membrane-associated phospholipid substrate. However, we do not know the manner in which cPLA2 is activated by extracellular stimuli and whether this activation occurs at specific sites. The C-terminal region of cPLA2 contains the phosphorylation site and catalytic site. These sites may be involved in regulation of the enzymic activity. The activation of cPLA2 can be through serine residues, notably Ser-505 and Ser-727, by mitogen-activated protein kinase and protein kinase C (PKC) (Hirabayashi and Shimizu, 2000; Hirabayashi et al., 2004). In neural membranes, cPLA2 activity and arachidonic acid release are linked to dopamine, glutamate, serotonin, P2-purinergic, muscarinic, cytokine, and growth factor receptors through different coupling mechanisms (Table 2). Some receptors involve G-proteins and others do not. The ligand-mediated stimulation of the above receptors modulates the release of arachidonic acid and levels of other second messengers in brain tissue (Farooqui et al., 2000b).
|
cPLA2 activity can also be modulated through a cooperative binding mechanism with glycerophospholipids containing arachidonic acid (Burke et al., 1995) or through binding of anionic phospholipids, such as phosphatidylinositol 4,5-bisphosphate, phosphatidylinositol 3,4,5-trisphosphate, and ceramide 1-phosphate (Hirabayashi et al., 2004; Pettus et al., 2004), to a pleckstrin homology domain (Mosior et al., 1998a). Three paralogs of cPLA2 occur in brain and other non-neural tissues (Diaz-Arrastia and Scott, 1999; Farooqui et al., 2000b; Hirabayashi et al., 2004). They are cPLA2- (molecular mass 85 kDa), cPLA2- (molecular mass 114 kDa), and cPLA2-
(molecular mass 61 kDa). cPLA2- is found mainly in the cerebellum and shares more similarities with cPLA2- than with cPLA2-. cPLA2- lacks the C2 domain, but contains a prenyl group-binding motif that behaves as a lipid anchor and allows binding of the enzyme to the membrane.
Recombinantly expressed cPLA2- liberates arachidonic acid from phosphatidylcholine. Unlike cPLA2-, cPLA2- also acts on other fatty acid residues at the sn-2 and sn-1 positions of glycerophospholipids. cPLA2- hydrolyzes fatty acids at the sn-2 position, cPLA2- prefers to cleave fatty acids at the sn-1 position, and cPLA2- efficiently hydrolyzes fatty acid at the sn-1 as well as sn-2 positions of the glycerol moiety (Song et al., 1999). The overexpression of cPLA2- increases the proportions of polyunsaturated fatty acids in phosphatidylethanolamine, indicating that this paralog can modulate the phospholipid composition (Asai et al., 2003). cPLA2- is constitutively expressed in the endoplasmic reticulum where it is involved in remodeling and maintaining membrane phospholipid composition under oxidative stress.
cPLA2- displays much lower activity with [2-arachidonyl]PtdCho than do the other two paralogs. The genes for human cPLA2-, -, and - map to chromosomes 1, 15, and 19, respectively. Mitogen-activated protein kinase phosphorylation sites are only present in cPLA2- and are not conserved in cPLA2- and cPLA2-. cPLA2- activity is uniformly distributed in various regions of rat brain (Farooqui et al., 2000b). Recent studies have indicated the presence of a new paralog of cPLA2. This paralog is mainly found in skin and has been named cPLA2-
(molecular mass 109 kDa) (Chiba et al., 2004). In contrast with other cPLA2 paralogs, cPLA2- has a preference for linoleic acid release instead of arachidonic acid release.
Considerable information is available on cPLA2-. From immunolabeling and in situ hybridization studies, cPLA2- is localized in somata and dendrites of Purkinje cells, whereas cPLA2- is present in the granule cells of rat brain (Shirai and Ito, 2004). In addition, cPLA2- is predominantly found in astrocytes of gray matter (Farooqui et al., 2000b; Pardue et al., 2003), as well as in hippocampal neurons (Sandhya et al., 1998; Kishimoto et al., 1999; Strokin et al., 2003), in which under physiological conditions cPLA2- may be involved in second-messenger generation and long-term potentiation (LTP), a mechanism involved in memory storage. More recently, the mRNAs for cPLA2- and cPLA2- have been identified by reverse transcription-polymerase chain reaction analysis in human brain tissue (Pickard et al., 1999; Song et al., 1999; Hirabayashi et al., 2004), but the role of these paralogs of cPLA2 in brain tissue remains speculative. Because cPLA2- mainly occurs in skin, it is proposed that this paralog plays a critical role in inflammation in psoriatic lesions (Chiba et al., 2004).
Even though cPLA2 isozymes are often considered to be the enzymes responsible for stimulus-mediated arachidonic acid release and eicosanoid formation, several studies also implicate sPLA2 activation. cPLA2 and sPLA2 may modulate arachidonic acid metabolism in the astrocytoma cell line 1321N1 via the mitogen-activated protein kinase pathway (Hernandez et al., 2000). At the cellular level, interactions and interplay among calcium mobilization, cPLA2 phosphorylation, and extracellular receptors of sPLA2 may be responsible for increased eicosanoid production (Fig. 1). cPLA2 may be a good candidate for triggering down-regulation of nitricoxide synthase activity and may thus be an important component of the cross-talk between calcium and nitric oxide-regulated signal transduction pathways in neuronal cells (Palomba et al., 2004). Under pathological conditions, the interactions among calcium, cPLA2, and nitric-oxide synthase may play an important role in the pathophysiology of neurological disorders associated with oxidative stress and inflammation (see below).
C. Plasmalogen-Selective Phospholipase A2
This enzyme hydrolyzes arachidonic acid and docosahexaenoic acid from the sn-2 position of plasmalogens, a special type of glycerophospholipid with a vinyl ether linkage at the sn-1 position of the glycerol backbone (Farooqui and Horrocks, 2001). This enzyme has been purified and characterized from bovine brain cytosol (Hirashima et al., 1992), rabbit kidney (Portilla and Dai, 1996), and rabbit heart (Hazen and Gross, 1993). Bovine brain PlsEtn-PLA2 has an apparent molecular mass of 39 kDa. Nonionic detergents, Triton X-100 and Tween 20, stimulate the enzymic activity. It is not inhibited by bromoenol lactone, an inhibitor that markedly inhibits iPLA2. Low micromolar concentrations of ATP have no effect on PlsEtn-PLA2 activity, but 2 mM ATP markedly inhibits its activity. Bovine brain PlsEtn-PLA2 is inhibited by 5,5'-dithiobis(2-nitrobenzoic acid, iodoacetate, and N-ethylmaleimide in a dose-dependent manner (Farooqui et al., 1995). Various polyvalent anions, citrate > sulfate > phosphate, and metal ions, Ag+, Hg2+, and Fe3+, also inhibit this enzyme in a dose-dependent manner. Glycosaminoglycans markedly inhibit bovine brain PlsEtn-PLA2 with an inhibition pattern of heparan sulfate > hyaluronic acid > chondroitin sulfate > heparin. This PLA2 is also inhibited by N-acetylneuraminic acid, gangliosides, and sialoglycoproteins (Yang et al., 1994b). Other glycosphingolipids, such as cerebrosides and sulfatides, have no effect. However, ceramide markedly stimulates PlsEtn-PLA2 activity in a time-and dose-dependent manner (Latorre et al., 2003). Treatment of rat brain slices with Staphylococcus aureus sphingomyelinase or C2-ceramide produces a marked decrease in PlsEtn levels, suggesting stimulation of PlsEtn-PLA2 activity. Bromoenol lactone, a potent inhibitor of iPLA2, does not affect this stimulation, but quinacrine and gangliosides, nonspecific inhibitors of PlsEtn-PLA2, completely block it (Latorre et al., 2003; Yang et al., 1994a,b). These studies have led to the suggestion that the degradation of plasmalogen by PlsEtn-PLA2 is a receptor-mediated process (Farooqui and Horrocks, 2001; Farooqui et al., 2003a; Latorre et al., 2003) and may involve an interaction between plasmalogen metabolism and sphingolipid metabolism.
PlsEtn-PLA2 has been localized immunochemically in neurons and astrocytes (Farooqui and Horrocks, 2001). The colocalization of PlsEtn-PLA2 with glial fibrillary acidic protein suggests that this PLA2 is predominantly associated with astrocytes. This is in contrast with cPLA2-, which is present in neurons as well as astrocytes (Sandhya et al., 1998; Kishimoto et al., 1999). cPLA2 (molecular mass 85 kDa) and PlsEtn-PLA2 (molecular mass 39-42 kDa) release arachidonic acid and docosahexaenoic acid in rat brain astrocytes and cyclic AMP and Ca2+ regulate these enzymes differentially (Strokin et al., 2003). Because plasmalogens are major phospholipids of neural membranes, PlsEtn-PLA2 may be mainly involved in generating docosahexaenoic acid (DHA), a 22-carbon essential fatty acid with 6 double bonds. This fatty acid is highly enriched in synaptosomal membranes, synaptic vesicles, and growth cones and accounts for >17% by weight of the total fatty acids in the brain of adult rats (Hamano et al., 1996).
D. Calcium-Independent Phospholipases A2
The brain cytosolic fraction contains an 80-kDa Ca2+-independent PLA2 activity. This enzyme has been purified from rat brain to homogeneity using multiple column chromatographic procedures with a very low yield. The purified enzyme has a specific activity of 4.3 µmol/min/mg. The peptide sequence of this enzyme has considerable homology to sequences of the iPLA2 from P388D1 macrophages, Chinese hamster ovary cells, and human B lymphocytes (Yang et al., 1999). This iPLA2 hydrolyzes the sn-2 fatty acid from PtdCho with preferences linoleoyl > palmitoyl > oleoyl > arachidonyl group. iPLA2 has an unique amino acid sequence containing a lipase consensus sequence and eight ankyrin repeats. This enzyme is strongly inhibited by bromoenol lactone, and ATP augments its activity. iPLA2 is present in all brain regions with the highest activity in striatum, hypothalamus, and hippocampus. The gene encoding iPLA2 has been identified (Molloy et al., 1998). Alternative splicing can generate multiple iPLA2 isoforms with distinct tissue distribution and localization (Larsson et al., 1998). Truncated splice variant proteins that prevent the formation of active iPLA2 tetramers may negatively regulate iPLA2 (Larsson et al., 1998; Seashols et al., 2004). Native iPLA2 is a homotetramer that is potentially formed through interactions between N-terminal ankyrin repeats (Ackermann and Dennis, 1995). Five splice variants of iPLA2 occur in various tissues (Larsson et al., 1998; Shirai and Ito, 2004). One splice variant (iPLA2-1) lacks exon 9 (165 base pairs), whereas the other four variants (iPLA2-2, iPLA2-3, iPLA2-ankyrin-1, and iPLA2-ankyrin-2) contain exon 9. The presence of this exon makes these splice variants membrane-bound because exon 9 encodes hydrophobic amino acids (Larsson et al., 1998; Shirai and Ito, 2004). Rat cerebellum contains iPLA2-1 and iPLA2-2 or iPLA2-3, but not iPLA2-ankyrin-1 or iPLA2-ankyrin-2. From immunolabeling studies in rat brain, granule cells, stellate cells, and the nucleus of Purkinje cells contain iPLA2 (Shirai and Ito, 2004). Strong signals of iPLA2 immunoreactivity are observed in olfactory bulb, hippocampus CA1-3, dentate gyrus, and brain stem. In non-neural cells, the cleavage of iPLA2 by caspase-3 is associated with the execution of apoptosis (Atsumi et al., 1998, 2000). The proposed role of iPLA2 in phospholipid remodeling and apoptosis is based on the use of bromoenol lactone, thought to be a specific inhibitor, but this compound actually inhibits other enzymes such as diacylglycerol lipase and phosphatidate phosphohydrolase (see below). This makes it difficult to define the role of iPLA2 in phospholipid metabolism (Farooqui et al., 2000a; Pérez et al., 2004). iPLA2 may play an important role not only in long-term potentiation and long-term depression and activity-dependent changes in synaptic strength believed to underlie certain forms of learning and memory in the hippocampus (Fitzpatrick and Baudry, 1994; Wolf et al., 1995; Fujita et al., 2001) but also in neural cell proliferation, apoptosis, and differentiation (Farooqui et al., 2004a).
III. Involvement of Phospholipase A2 Activity in Brain Injury
A growing body of evidence suggests the involvement of isoforms of PLA2 in neurotransmitter release, LTP, long-term depression, membrane repair, neurodegeneration, neural cell proliferation, differentiation, and apoptosis (Fig. 2) (Farooqui et al., 1997c). It is not known which PLA2 isoform performs which central nervous system function. The multiplicity of PLA2 and the interplay among lipid mediators generated by PLA2 in brain tissue provide diversity in function and specificity of various isoforms of this family of enzymes in the regulation of enzymic activity in response to a wide range of extracellular signals. However, this diversity complicates the analysis of their function (Fig. 1). The complexity of this problem becomes obvious when one considers the coupling of various isoforms of PLA2 with different receptors in a single neural cell and tries to associate PLA2 activity with specific neuronal functions and disease processes.
|
). Various isoforms of PLA2 probably act on different cellular pools of phospholipids located in different types of neural cells and these isoforms may be regulated by different coupling mechanisms involving common second messengers (Sun et al., 2004). The synthesis of eicosanoids depends not only on PLA2 activity for the generation of arachidonic acid, but also on cyclooxygenase or lipoxygenase activities (Sun et al., 2004). Tight regulation of PLA2 activity is necessary for normal brain function (Farooqui et al., 2000b; Phillis and O'Regan, 2004).
Activities of PLA2 isoforms are also regulated through modulation of gene expression. Interleukin (IL)-1, (Xu et al., 2003), tumor necrosis factor (TNF)- (Pirianov et al., 1999; Tong et al., 1999; Jupp et al., 2003), interferon- (Xu et al., 2003), and several growth factors induce these enzymes (Jupp et al., 2003; Akiyama et al., 2004; Wang et al., 2004b). Regulatory interactions and coupling between PLA2 activity and the expression of genes encoding cyclooxygenase and lipoxygenases have been suggested (Doug et al., 2003). In brain tissue, the coupling between cPLA2 depends not only on the type of neural cell but also on the stimulatory status of neural cell involved.
Stimulation of PLA2 isoforms may contribute to brain damage in several ways:
). Arachidonic acid produces mitochondrial swelling in glial cells and induces changes in membrane permeability by regulating ion channels (Farooqui et al., 1997b,c). It also inhibits glutamate uptake. In the nucleus, arachidonic acid may also interact with elements of gene structure, such as promoters, enhancers, suppressors, and others, in a specific manner that is not shared by eicosanoids or other fatty acids. These interactions modulate gene expression (Farooqui et al., 1997c). 
B activity and gene expression (Mazière et al., 1999). Another consequence of increased polyunsaturated free fatty acid-induced oxidative stress is the activation and inactivation of redox-sensitive proteins (Wang, 2003). , 1994). The most compelling evidence for the involvement of cPLA2 in neurodegeneration comes from reports indicating that mice with a targeted deletion of the gene that encodes cPLA2 show reduced infarct size after cerebral ischemia (Bonventre et al., 1997) and resistance to MPTP-induced neurotoxicity (Klivenyi et al., 1998). Studies on brain lipid metabolism in cPLA2 knockout mice indicate there is no net change in unesterified arachidonic acid. However, there was a 50% reduction in esterified arachidonic acid in phosphatidylcholine, indicating involvement of cPLA2 in wild-type mice (Rosenberger et al., 2003). The knockout mice also have reduced rates of arachidonic acid incorporation into ethanolamine and choline glycerophospholipids but elevated rates into phosphatidylinositol. cPLA2-deficient mice also show a 62% reduction in the rate of formation of prostaglandin (PG) E2, suggesting a coupling between cPLA2 and cyclooxygenase activities (Murakami et al., 1997; Bosetti and Weerasinghe, 2003). The deletion of cPLA2 also causes dysregulation ofinsulin-like growth factor-1 signaling and stimulates striated muscle growth (Obata et al., 2003).
4-HNE has three functional groups that confer to its molecule a very high reactivity toward thiol and amino groups. 4-HNE has been reported to cause a number of deleterious effects in cells including inhibition of DNA synthesis, disturbance in calcium homeostasis, and inhibition of mitochondrial respiration. All of these processes may result in neuronal injury in ischemia and glutamate-mediated neurotoxicity (Farooqui et al., 1997c). In brain tissue, 4-HNE also produces alterations in the function of key membrane proteins including glucose transporter, glutamate transporter, and sodium potassium ATPase (Friguet et al., 1994; Jamme et al., 1995). Thus, high levels of this metabolite are toxic for brain tissue (Farooqui et al., 1997c).
ROS inactivate membrane proteins and DNA (Berlett and Stadtman, 1997). The reaction between ROS and proteins or unsaturated lipids in the plasma membrane leads to a chemical cross-linking of membrane proteins and lipids and a reduction in membrane unsaturation. This depletion of unsaturation in membrane lipids is associated with decreased membrane fluidity and decreases in the activity of membrane-bound enzymes, ion channels, and receptors (Ray et al., 1994).
IV. Physiological and Pharmacological Effects of Phospholipase A2 Inhibitors
The stimulation of PLA2 isoforms, release of arachidonic acid, and generation of platelet-activating factor are important events in the inflammation and oxidative stress associated with acute neural trauma and chronic neurological disorders (Farooqui and Horrocks, 1994; Phillis and O'Regan, 2004). Treatment of these disorders requires potent and selective inhibitors of PLA2 activity that can be used as drugs. The problem with available inhibitors of PLA2 isoforms has been their specificity. Many PLA2 inhibitors originally thought to be selective for a specific PLA2 isoform are now known to not only inhibit other PLA2 isoforms but also block activities of different enzymes (Farooqui et al., 1999; Cummings et al., 2000; Fuentes et al., 2003). For example, in non-neural cells, arachidonyl trifluoromethyl ketone inhibits not only cPLA2 activity but also cyclooxygenase and acyltransferase activities (Cummings et al., 2000; Fuentes et al., 2003). Methyl arachidonyl fluorophosphonate, another inhibitor of brain cPLA2, also inhibits bovine brain iPLA2.
The discovery of potent and specific inhibitors of PLA2 isoforms is an important approach, not only for establishing functional roles of a given PLA2 isoform in a specific type of neural cell in brain tissue but also for treating oxidative stress and inflammation caused by neurodegenerative process. Studies on this important topic are beginning to emerge (Farooqui et al., 1999; Cummings et al., 2000; Miele, 2003a; Scott et al., 2003; Clark and Tam, 2004). These studies are complicated not only by the lack of information on the availability of specific inhibitors but also by the occurrence of isoforms of PLA2 activity and in vivo effects of PLA2 inhibitors on enzymic activity. Furthermore, the effect of inhibitors on the physical state of substrate aggregates in neural membranes remains unknown. In searching for good cPLA2 inhibitors, kinetic analysis is not enough to evaluate whether an inhibitor can block PLA2 activity by affecting the interfacial quality of phospholipid in lipid bilayer or by directly inhibiting the interaction between the phospholipids and the active site of the enzyme. An ideal inhibitor should have regional specificity and should be able to reach the site where cells are under oxidative stress and inflammatory and neurodegenerative processes are taking place.
Neurons are more susceptible to free radical-mediated neuroinflammation and oxidative stress than glial cells (Adibhatla et al., 2003; Ajmone-Cat et al., 2003). In fact, activated glial cells, including astroglia and microglia, sustain inflammatory processes initiated by arachidonic acid-generated metabolites. This suggests that signals modulating the induction, expression, and stimulation of PLA2 isoforms may play an important role in neurodegenerative diseases associated with neuroinflammation and oxidative stress (Farooqui and Horrocks, 1994; Farooqui et al., 2003b, 2004b). For the successful treatment of inflammatory and oxidative stress in neurological disorders, timely delivery of a well-tolerated, chronically active, and specific inhibitor of PLA2 that can bypass or cross the blood-brain barrier without harm is required. Some nonspecific PLA2 inhibitors (see below) have been used for the treatment of ischemia, spinal cord injury, and AD (Sano et al., 1997), but no compound with real clinical potential has emerged.
A. Arachidonyl Trifluoromethyl Ketone
Arachidonyl trifluoromethyl ketone (Fig. 3) is a potent inhibitor of cPLA2. NMR studies show that the carbon chain of AACOCF3 binds in a hydrophobic pocket and the carbonyl group of AACOCF3 forms a covalent bond with the serine 228 in the active site, generating a charged hemiketal oxoanion that interacts with a positively charged group of the enzyme (Street et al., 1993; Trimble et al., 1993). AACOCF3 is a 500-fold more potent inhibitor of cPLA2 than sPLA2 (Trimble et al., 1993), indicating that it is a more selective inhibitor of cPLA2 than sPLA2. This inhibitor also blocks cyclooxygenase activity (Riendeau et al., 1994).
|
). AACOCF3 inhibits bovine brain cPLA2 and iPLA2 in a dose-dependent manner with IC50 values of 1.5 and 6.0 µM, respectively.
The treatment of NG 108-15 cells with AACOCF3 decreases initial neurite formation in a concentration-dependent manner (Smalheiser et al., 1996). The pharmacological blockade of cPLA2 by a low concentration, 10 µM, of AACOCF3 significantly inhibits neuronal death in the CA1 region of rat hippocampus. In primary neuronal cultures, this PLA2 inhibitor prevents caspase-3 activation and neurodegeneration induced by -amyloid peptide (A) and human prion protein peptide (Bate et al., 2004), suggesting the role of PLA2 isoforms in neurodegenerative processes (Farooqui and Horrocks, 1994; Farooqui et al., 1997c). In primary neuronal cultures, AACOCF3 also abolishes methylmercury (MeHg2+)-mediated stimulation of cPLA2 and arachidonic acid release (Shanker et al., 2004), suggesting that cPLA2 plays an important role in MeHg2+-induced neurotoxicity. Similarly, in astrocytes MeHg2+-induced ROS generation is strongly inhibited by AACOCF3 (Shanker and Aschner, 2003). In neural cell cultures AACOCF3 also shortens the association of PKC- with plasma membrane indicating that this isoform of PKC may be involved in neuronal plasticity (Yagi et al., 2004). AACOCF3 also blocks L-buthionine sulfoximine toxicity in glutathione-depleted mesencephalic cultures (Kramer et al., 2004).
AACOCF3 induces the dispersal of Golgi stack and trans-Golgi network resident proteins. This suggests that cPLA2 isozymes play a crucial role in membrane trafficking and in maintenance of Golgi architecture. In non-neural cultured cells, AACOCF3 inhibits the expression of IL-2 at both the mRNA and protein levels, indicating that cPLA2 may have marked effects on T-cell function (Amandi-Burgermeister et al., 1997; Ouyang and Kaminski, 1999). AACOCF3 inhibited DNA fragmentation during apoptosis in U937 cells, but failed to affect the morphological changes that occur during apoptosis, suggesting that the use of AACOCF3 may distinguish between cytoplasmic and nuclear events that occur during apoptotic cell death (Vanags et al., 1997). AACOCF3 is described as a specific inhibitor of cPLA2, but recent studies in non-neural cells indicate it may also inhibit cyclooxygenases and 5-lipoxygenase (Cummings et al., 2000; Fuentes et al., 2003). These observations strongly suggest that AACOCF3 is not a specific inhibitor of cPLA2.
B. Methyl Arachidonyl Fluorophosphonate
MAFP (Fig. 3) is an irreversible inhibitor of bovine brain cPLA2 (IC50 0.5 µM) and has no effect on sPLA2.It inhibits enzymic activity by reacting with a serine residue at the active site. MAFP also inhibits bovine brain iPLA2 in a dose-dependent manner with an IC50 value of 0.75 µM. At 5.0 µM, MAFP completely inhibits bovine brain iPLA2 activity. In addition, MAFP inhibits A-mediated stimulation of cPLA2 activity in cortical neuronal cultures (Kriem et al., 2005).
MAFP induces irreversible inhibition of the enzymic hydrolysis of arachidonyl ethanolamide (anandamide) by fatty acid amide hydrolase. Based on various pharmacological studies with cannabinoid CB1 receptors and MAFP, it has been suggested that MAFP is an irreversible cannabinoid CB1 receptor antagonist (Fernando and Pertwee, 1997).
Because MAFP interacts with several PLA2 isoforms and with fatty amide hydrolase, it cannot be considered as a specific inhibitor of cPLA2. MAFP also blocks L-buthionine sulfoximine toxicity in glutathione-depleted mesencephalic cultures (Kramer et al., 2004). Intrathecal injections of MAFP in rats produce antinociceptive effects (Ates et al., 2003), suggesting that PLA2 isoforms may play some role during pain states (see below).
BEL (Fig. 3) is a potent inhibitor of bovine brain iPLA2 and PlsEtn-PLA2 with IC50 values of 60 and 40 nM, respectively. BEL has a structural resemblance to plasmalogen. It inhibits brain cPLA2 and sPLA2 at a very high concentration (500 µM). The injection of BEL (10 µM) into postsynaptic CA1 pyramidal neurons produces a robust increase in the amplitude of -amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) receptor-mediated excitatory postsynaptic currents, suggesting that iPLA2 plays an important role in AMPA-mediated synaptic plasticity (St-Gelais et al., 2004). AACOCF3 and palmityl trifluoromethyl ketone, which mainly interact with cPLA2, have no effect on AMPA-mediated synaptic transmission. The inhibition of iPLA2 by BEL and the enhancement of AMPA subunit immunoreactivity in brain homogenates and slices support the above electrophysiological studies. Taken together, these results support the hypothesis that BEL-mediated antagonism of AMPA receptors is involved in long-term potentiation and long-term depression during the regulation of synaptic plasticity. Based on the blockage of induction of hippocampal long-term potentiation, brain iPLA2 may be involved in learning and memory (Wolf et al., 1995; Fujita et al., 2001).
Intracerebroventricular injections of BEL, 3 nmol, markedly affect spatial performance in mice (Fujita et al., 2000), indicating that iPLA2 is involved in spatial memory formation. BEL also modulates intracellular membrane trafficking (Kuroiwa et al., 2001; Brown et al., 2003) by inhibiting iPLA2 activity in membrane tubule formation during reassembly of the Golgi complex. In addition, BEL treatment also interferes with membrane fusion events during endocytosis and exocytosis.
D. Benzenesulfonamides and Alkoxybenzamidines
Benzenesulfonamide (Fig. 3) and its piperidine derivative are the most potent inhibitors of membrane-bound heart PLA2 activity with IC50 values of 28.0 and 9.0 nM, respectively (Oinuma et al., 1991). Intravenous injections of these inhibitors protect rats against ischemic damage in acute myocardial infarction. These compounds are relatively metabolically stable in plasma with half-lives of 1 to 2 h. These compounds also inhibit brain cPLA2 activity in a dose-dependent manner with IC50 values of 23.0 and 10.0 nM, respectively. Nothing is known about their ability to cross the blood-brain barrier. The synthesis of 4-alkoxybenzamidines as PLA2 inhibitors has also been described previously (Aitdafoun et al., 1996). These compounds competitively inhibit bovine pancreatic and rabbit platelet PLA2 activities with IC50 values of 3.0 and 5.0 µM, respectively. It is interesting to note that 4-tetradecyloxybenzamidine has an anti-inflammatory effect in vivo on carrageenan-mediated rat paw edema.
E. 3-(Pyrrol)-2-propionic Acid
Lehr (1996, 1997a,b) has synthesized many compounds (Fig. 3) that inhibit cPLA2 activity in platelets with IC50 values varying from 0.5 to 10 µM. Effects of these compounds on other isoforms of PLA2 remain unknown. These inhibitors have not been used to block the activity of brain PLA2 isoforms and have not been injected into intact animals or animal models of neurological disorders; therefore, nothing is known about their tolerance, half-lives, and toxicity. The ability of these inhibitors to cross the blood-brain barrier also remains unknown.
F. 2-Oxoamide and 1,3-Disubstituted Propan-2-ones
Long-chain 2-oxoamides of -aminobutyric acid and -norleucine (AX006, AX007, and AX008) (Fig. 3) reversibly inhibit cPLA2 activity in a dose- and time-dependent manner (Kokotos et al., 2004). These inhibitors block the LPS-mediated release of arachidonic acid due to the stimulation of cPLA2 in murine P388 D1 macrophages with IC50 values of 8.0, 7.6, and 4.6 µM, respectively. These IC50 values are lower than the IC50 value reported for MAFP (25 µM). This strongly suggests that 2-oxoamides are more potent inhibitors of cPLA2 than MAFP. Anti-inflammatory and analgesic activities of 2-oxoamide have been tested in the rat paw carrageenan-mediated edema assay. Carrageenan-induced edema in rat paw can be prevented with 2-oxoamides (Kokotos et al., 2004). Effects of these inhibitors on brain cPLA2 remain unknown.
1,3-Disubstituted propan-2-ones (Fig. 4) were recently synthesized (Connolly et al., 2002). These compounds inhibit cPLA2 activity with an IC50 value of 0.03 µM in an in vitro assay. They are 10-fold more effective than 2-oxoamides and AACOCF3 in inhibiting cPLA2 activity. 1,3-Disubstituted propan-2-ones inhibit arachidonic acid production in HL60 cells with an IC50 value of 2.8 µM. These inhibitors have not been injected into intact animals or animal models of neurological disorders; therefore, nothing is known about their tolerance, half-lives, and toxicity. The ability of these inhibitors to cross the blood-brain barrier also remains unknown.
|
Recently, a new class of hydrophobic inhibitors that partition into the lipid bilayer and compete with monomers of the glycerophospholipid substrate was described previously (Burke et al., 1999). These compounds include 2-(2-benzyl-4-chlorophenoxy)ethyldimethyl-n-octadecyl-ammonium chloride and 2-(2-benzyl-4-chlorophenoxy)ethyldimethyl-n-octyl-ammonium bromide (Fig. 4). Both compounds inhibit cPLA2 activity in a competitive manner with IC50 values of 5 and 13 µM, respectively. The length of the N-alkyl chain plays an important role in the degree of inhibition. Shortening of the N-alkyl chain considerably decreases the percentage of inhibitor partitioned into the glycerophospholipid bilayer and increases the IC50 value. These effects may be due to diminished hydrophobic interaction between the shorter alkyl chain and the fatty acid tails of the glycerophospholipids making up the bilayer. In contrast, lengthening of the N-alkyl chain increases the percentage of inhibitor partitioned into the lipid bilayer and decreases the IC50 value. The synthesis of these inhibitors represents an important step in the development of potent in vivo cPLA2 inhibitors because these compounds inhibit enzymic activity at the interface and provide a weak interaction with the glycerophospholipid bilayer (Burke et al., 1999). Based on kinetic studies, the potential in vivo efficacy of these intracellular inhibitors can be more potent than other inhibitors such as AACOCF3. These inhibitors have not been used for in vivo studies so nothing is known about their tolerance, toxicity, and half-life. It is also not known whether these inhibitors can cross blood-brain barrier.
H. Pyrrolidine-Based Inhibitors of Phospholipase A2
Pyrrolidine-containing compounds (Seno et al., 2000; Ghomashchi et al., 2001) markedly inhibit cPLA2- in vitro and block arachidonate release in Ca2+ ionophore-stimulated non-neural cells (Seno et al., 2000). The structure of the most potent inhibitor, pyrrolidine-1, is shown in Fig. 4. In a fluorometric assay, pyrrolidine-1 inhibits cPLA2- activity in a dose-dependent manner with an IC50 value of 0.18 µM. In a mixed-micelle assay system, this compound inhibits cPLA2- activity with an IC50 value of 1.8 µM. This difference in the IC50 value may be due to the differences in interface concentrations of pyrrolidine-1 in the different assay systems. The treatment of Chinese hamster ovary cells with pyrrolidine-1 results in marked inhibition of A23187-induced arachidonic acid release with an IC50 value of 0.2 to 0.5 µM. The degree of inhibition approaches 100% with 4 to 10 µM pyrrolidine-1 (Ghomashchi et al., 2001). Similarly the treatment of Madin-Darby canine kidney cells with pyrrolidine-1 also results in marked inhibition of ATP-induced arachidonic acid release with an IC50 value of 0.8 µM. It must be stated here that pyrrolidine-1 also inhibits cPLA2- and iPLA2- at a very high concentration but it does not inhibit sPLA2.
Pyrrophenone, a triphenylmethylthioether derivative of pyrrolidine, is a 39-fold more potent inhibitor (Ki 4.2 nM) of cPLA2 activity than pyrrolidine-1 (Ono et al., 2002). Pretreatment of rats with pyrrolidine dithiocarbamate, a powerful thiol antioxidant, protects against kainate (KA)-mediated neurotoxicity (Shin et al., 2004). In vitro studies indicate that pyrrophenone may be a potential therapeutic agent for inflammatory diseases (Ono et al., 2002). Pyrrolidine-containing PLA2 inhibitors have not been injected in animal models of neurological disorders so their half-life and side effects remain unknown.
All isoforms of bovine brain PLA2 are strongly inhibited by antimalarial drugs in a dose-dependent manner with rank order potency of chloroquine > quinacrine > hydroxychloroquine > quinine (Lu et al., 2001b). Chloroquine, quinacrine, hydroxychloroquine, and quinine inhibit bovine brain cPLA2 with IC50 values of 125, 200, 185, and 250 µM. It is suggested that among PLA2 isoforms, PlsEtn-PLA2 and cPLA2 may be associated with proximal events involved in the induction and maintenance of inflammatory processes after ischemic and traumatic brain injuries (Farooqui et al., 2004b) and sPLA2 may be involved in intensification (Han et al., 2003) of inflammation during later stages of the inflammatory reaction. At low concentrations (<50 µM), these inhibitors have no effect on the growth of neuron-enriched cultures from rat brain cortex, but at high concentrations (>1000 µM), these inhibitors are toxic.
J. Lithium Ion and Carbamazepine
Lithium ion is a mood stabilizer. It has been used for the treatment of bipolar disorders for almost a half-century (Corbella and Vieta, 2003). Lithium ion has a neuroprotective effect on brain tissue. Chronic lithium ion administration in rats result in 50% reduction in mRNA and protein levels of cPLA2 with no changes observed in iPLA2 and sPLA2 protein during these studies (Chang and Jones, 1998; Rintala et al., 1999). Lithium ion does not reduce phosphorylated cPLA2 protein. Thus, the decrease in brain cPLA2 enzyme activity induced by chronic lithium ion treatment is due to down-regulation of cPLA2 transcription (Weerasinghe et al., 2004). This down-regulation of cPLA2 transcription may be responsible for a selective reduction of arachidonic acid turnover compared with docosahexaenoic acid in rat brain phospholipids (Basselin et al., 2003). The labeling of Purkinje cell dendrites with cPLA2 and cyclooxygenase (COX-2) antibodies is inhibited by lithium ion, indicating the functional coupling at brain synapses between cPLA2 and COX-2 enzymes (Weerasinghe et al., 2004).
An anticonvulsant drug, carbamazepine (CBZ), has been used for the treatment of bipolar disorders for many years. Chronic administration of CBZ not only inhibits cPLA2 activity but also alters its protein and mRNA levels. In contrast, it did not affect iPLA2 and sPLA2 activities and protein levels (Ghelardoni et al., 2004). These effects are accompanied by a decrease in COX-2 activity and prostaglandin E2 levels in brain tissue, suggesting that CBZ blocks the cPLA2-mediated release of arachidonic acid and its conversion via COX-2 to prostaglandin E2 (Ghelardoni et al., 2004). The protein levels of other arachidonic acid-metabolizing enzymes such as 5-lipoxygenase and cytochrome P450 and their reaction product leukotriene B4 are not affected by CBZ. Thus, nonspecific PLA2 inhibitors, such as lithium ion and CBZ, may have beneficial effects, not only in neurological diseases (see below), but also in bipolar disorders and manic-depressive patients.
Vitamin E (-tocopherol) is another nonspecific inhibitor of cPLA2 activity (Douglas et al., 1986). It modulates the production of arachidonic acid and eicosanoids (Tran et al., 1996). It inhibits bovine brain cPLA2 and iPLA2 activities in a dose- and time-dependent manner with IC50 values of 500 and 750 µM, respectively. Vitamin E crosses the blood-brain barrier and with time accumulates in brain (Pentland et al., 1992). It reduces lipid peroxidation and stabilizes neuronal membranes. Several reports indicate that ischemia is accompanied by an increase in PLA2 activity and a reduction in the levels of vitamin E and glutathione (Farooqui et al., 1994a). A deficiency of vitamin E and selenium in rats leads to a biphasic increase in iPLA2 activity in non-neural cells (Burgess and Kuo, 1996), once again supporting the view that PLA2 activity is regulated by vitamin E. Vitamin E may modulate activities of PLA2 isozymes by two mechanisms: first, by direct incorporation into substrate vesicles; and second, by stimulation of the activity of PLA2 isoforms by up-regulating the rate of their synthesis at the transcriptional or translation level (Tran et al., 1996).
GM1 and GM3 gangliosides also inhibit cPLA2 and PlsEtn-PLA2 activities in a dose-dependent manner (Yang et al., 1994a,b). With PlsEtn-PLA2 the IC50 values for GM1 and GM3 gangliosides were 150.0 and 75.0 µg/ml, respectively. The IC50 values for brain cPLA2 for GM1 and GM3 were 250.0 and 100.0 µg/ml, respectively. The mechanism of inhibition by gangliosides remains unknown. However, the orientation of N-acetylneuraminic acid residues in glycoconjugates is important for inhibitory activity (Yang et al., 1994b). Gangliosides not only stabilize neural membranes but also regulate calcium influx and enzyme activities associated with signal transduction.
CDP-amines are key intermediates in the biosynthesis of phosphatidylcholine and phosphatidylethanolamine. CDP-choline (citicoline) decreases cPLA2 stimulation and hydroxyl radical generation after transient cerebral ischemia (Adibhatla and Hatcher, 2003). This process results in lowering of the concentration of free fatty acids in a dose- and time-dependent manner. CDP-choline protects neural membranes, not only by accelerating the re-synthesis of phospholipids but also by quenching free radicals generated by PLA2 isozymes (Rao et al., 2001; Adibhatla et al., 2002
