Lipoxygenases form a family of non-heme iron containing enzymes, which dioxygenate polyunsaturated fatty acids at specific positions to produce a series of hydroperoxy fatty acids. A variety of substrates such as free fatty acids, membrane phospholipids and lipoproteins can be used as substrates for peroxidation by the lipoxygenases. These enzymes specifically recognise the 1-cis, 4-cis pentadiene structures to introduce molecular dioxygen and create a cis-trans conjugated diene. The most accepted hypothesis for the lipoxygenase reaction is the radical hypothesis (Kühn et al., 1986b; Kühn and Thiele, 1999) according to which the reaction consists of three major steps (figure 1):
|Figure 1 Radical mechanism of the lipoxygenase reaction|
This family of enzymes is widely distributed in the animal and plant kingdom (Kühn and Thiele, 1999; Brash, 1999, Mack et al., 1987; Grechkin, 1998). Lipoxygenases have also been discovered in lower organisms, such as corals (Brash et al., 1996), fungi (Bisakowski et al., 1997; Su and Oliw, 1998) and even bacteria (Porta and Rocha-Sosa, 2001). The nomenclature of the enzyme is based upon the number of the carbon atom which is preferentially oxygenated in arachidonic acid (AA). The positional nomenclature is further specified by mentioning the tissue source of the lipoxygenase e.g. leukocyte, platelet, and the stereospecificty by adding S and R to the name. This system is, however, fraught with difficulties. The 12-lipoxygenases of leukocyte type share only 40% homology with the 12-lipoxygenase of platelet type. On the other hand they are 75% similar to 15-lipoxygenase of the reticulocyte type (Yamamoto, 1992). Phylogenetic classification (Kühn and Thiele, 1999) of the various mammalian lipoxygenases leads to the subdivision of these enzymes into 4 main subtypes (figure 2): 12/15 lipoxygenases, 5 lipoxygenases, 12 lipoxygenase of the platelet type and the epidermis type lipoxygenases. This has been proven to be a more rational system since enzymes with similar sequence, structure and activity are grouped together like 15-LOX and 12-LOX leukocyte type.
|Figure 2 Phylogenetic tree of mammalian lipoxygenases.|
So far the 12/15-lipoxygenase (12/15-LOX) family consists of murine, rat, bovine and porcine leukocyte 12-lipoxygenases, rabbit reticulocyte 15-lipoxygenase (15-LOX) and human 15-lipoxygenase-1 (15-LOX-1). These enzymes share a remarkable degree of homology with each other both at the protein, and DNA level. They also exhibit many common enzymatic properties such as broad range of substrate specificity and reaction kinetics (Yamamoto, 1992). To date, the simultaneous expression of these enzymes in the same species has been observed only in rabbits (Berger et al., 1998) leading to the hypothesis that leukocyte type 12-LOX in other species may be functionally complementary to the 15-LOXs in rabbits and humans.
A 15-LOX was first observed in rabbit reticulocytes where it was shown to play an important role in the maturation of reticulocytes to erythrocytes (Schewe et al., 1975; Rapoport et al., 1979). Later, this enzyme was also described in human eosinophils (Sigal et al., 1988) and bronchial epithelial cells (Henke et al., 1988; Nadel et al., 1991). The enzyme consists of 663 amino acids, 75 kDa, and the degree of similarity between the two species is more than 99%. In pigs, it was first purified from leukocytes (Yoshimoto et al., 1982), in mice from spleen (Chen et al., 1994) and in rat brain (Watanabe et al., 1993). The murine and porcine enzymes share a high degree (about 75%) of sequence homology with rabbit and human 15-LOXs. 12/15-LOX enzymes exhibit a broad substrate specificity. In contrast to other lipoxygenases, they are capable of oxygenating free polyunsaturated fatty acids like arachidonic acid (AA), linoleic acid (LA) and linolenic acids, esterified lipids like mono, di and tri-acylglycerols, cholesterol esters and phospholipids as well as complex substrates like biomembranes and lipoproteins (Rapoport et al., 1979; Schewe et al., 1986). These enzymes exhibit dual positional specificity with arachidonic acid. The ratio of 12- to 15-HETE varies between the lipoxygenases from different species (Yoshimoto and Yamamoto, 1995). The reaction proceeds by the abstraction of a hydrogen from the bisallylic methylene at the C-10 position (12-LOX) or at the C-13 (15-LOX) followed by insertion of oxygen. Both the processes occur antarafacially i.e. from different sides of the plane of the double bonds.
The tissue distribution of 12/15-LOXs varies in different species. In rat, the 12-leukocyte type enzyme is maximally expressed in the pineal gland followed by leukocytes, macrophages, lung, spleen, pituitary, pancreatic beta cells and aorta, while in mouse, the maximum [page 4↓]expression was seen in the peritoneal macrophages, kidney and pineal gland with weaker signal observed in intestine, spleen and aorta (Yoshimoto and Takahashi, 2002). Simultaneous expression of both 12(S) leukocyte type and 15-LOX was observed in rabbits (Berger et al., 1998). The 12-leukocyte type enzyme was detected in the tracheal epithelium and leukocytes of cows and in the pitutary gland and leukocytes of pigs. 15-LOX-1 is the only 12/15-LOX isoform which has been detected in humans. It is expressed constitutively in bronchial epithelium and eosinophils (Sigal et al., 1988) though expression is also observed in monocytes and colonic epithelium upon cytokine (IL-4) stimulation.
Another interesting feature of this enzyme is the property of suicide inactivation. Reaction with fatty acids rapidly slows down and stops within several minutes of incubation (Rapoport et al., 1984; Hada et al., 1991). This is due to a complex set of mechanisms not yet clearly understood. Kishimoto et al., 1996, have shown that 15(S)-HpETE produced as product of AA metabolism covalently binds to the 12-LOX leukocyte protein and inactivates the enzyme. This feature has also been observed from the rabbit 15-LOX. Epoxide products of 15-HpETE, such as 14,15-LTA4 involve the formation of radical intermediates, which were hypothesised to covalently link to active site residues causing enzyme inactivation. Inhibition was also observed with LA independent of covalent linkages suggesting the proposed LTA4 mechanism may not be exclusive (Wiesner et al., 2003).
Among the mammalian lipoxygenases, the X-ray crystal structure is available only for the rabbit 15-LOX (Sloane et al., 1990; Gillmor et al., 1997). Much of the information about structure and function of this protein has been elucidated utilising the crystal structure data along with site-directed mutagenesis and modified substrates. The enzyme is composed of a large C- terminal catalytic domain containing the non-heme iron and a small N-terminal beta barrel domain. The substrate binding site is a large boot shaped hydrophobic cavity, which contains the non-heme iron catalytic centre (Gillmor et al., 1997). The active site is lined mainly by the hydrophobic side chains of amino acids. Phe 353, Ile 418, Met 419, Ile 593 form the bottom of substrate binding cavity and a positively charged Arg 403 is positioned at the entrance of the active site. It has been proposed that the substrate fatty acid slide into the active site with its methyl end ahead and the positively charged Arg 403 interacts with the carboxy group forming a salt bridge (Gan et al., 1996). Replacement of this residue with a [page 5↓]neutral amino acid residue severely impaired the dioxygenation process. Inside the active site the methyl end interacts with various side chains to position the bisallylic methylene C-13 of the substrate close to the iron atom for hydrogen abstraction. Site-directed mutagenesis studies have revealed that Phe 353 (Borngräber et al., 1996), Ile 418, Met 419 and Ile 593 (Sloane et al., 1991; Borngräber et al., 1999) are the positional determinants of 15 lipoxygenation (C-13 abstraction) single or combined. Mutation to these amino acids to residues with less bulkier side chains (alanine) converted the enzyme to a 12-LOX. The overall cavity size of 12-LOX leukocyte type was predicted to be moderately bigger, allowing the substrate to slide in deeper and aligning the C-10 of AA atom for hydrogen abstraction at the non-heme iron.
|Figure 3 Position determinants of lipoxygenase reaction.|
This suggests that the space inside the active site cavity plays an important role in the positional specificity (Borngräber et al., 1999). The reverse process on 12-LOX works equally well (Suzuki et al., 1994; Watanabe and Haeggstrom, 1993). However, conversion to 5-LOX by mutagenesis has not been successful. The positional determinant residues on 15-LOX were mutated to those of 5-LOX but the enzyme was inactive (Sloane et al., 1990). 15-LOX possess the ability to oxygenate 15-HpETE to form 5, 15-diHpETE. Methylation of carboxy end of the substrate increased the activity significantly. This phenomenon was hypothesised to be due to an inverse orientation of the substrate at the active site. In this case the caroboxy end may slide into the cavity as suggested by experiments with modified [page 6↓]substrates and site directed mutagenesis (Schwarz et al., 1998; Walther et al., 2001). Thus, the determinant of positional specificity is not only the volume but also the orientation of the substrate in the active site.
The N-terminal domain of the enzyme does not play a major role in the dioxygenation reaction of 12/15 lipoxygenase. N-terminal domain truncations did not impair the lipoxygenase activity. The ability of the enzyme to bind to membranes, however, is impaired in the mutants (point and truncations) of the N-ternimal domain without significant alterations to the catalytic activity (Walther et al., 2002). Mutation to Trp 181, which is localised in the catalytic domain, also impaired membrane binding function. This suggests that the C-terminal domain is responsible for the catalytic activity and a concerted action of N-terminal and C-terminal domain was necessary for effective membrane binding.
The expression of 15-LOX can be controlled at the transcriptional, translational, post-translational levels (Kühn et al., 1999, Kühn et al., 2002).
Constitutive expression of rabbit 15-LOX is observed only in reticulocytes under the condition of experimental anemia, where it is strongly expressed in peripheral monocytes, lung, liver, spleen and kidneys (Schewe et al., 1975; Rapoport et al., 1979). The mechanism of induction of 15-LOX during experimental anemia is still unclear. O’Prey and Harrison, 1995, studied the promoter region of the gene and found a number of negative and positive regulatory elements, which are differentially regulated in erythroid (15-LOX expressing) and non-erythroid cell lineages (non-expressing). In 1 Kb of 5’ region flanking the 15-LOX gene, they observed multiple copies of a putative transcriptional silencer element, which function only in non-erythroid cell lineages. DNAse I mapping studies revealed protein binding to these elements only in non-erythroid cells and not in erythroid cells. This element was therefore suggested to function like a silencer element in non-erythroid cells, inhibiting the transcription of the gene. Several other positive regulatory elements were observed in erythroid cells, which could putatively bind to transcription factors such as GATA proteins.
In humans, the Th2 cytokines, IL-4 and IL-13 have been shown to upregulate the expression of this gene in peripheral monocytes (Conrad et al., 1992), A549 lung epithelial carcinoma cells (Brinckmann et al., 1996), human tracheobroncheal cells (Jayawickreme et al., 1999), endothelial cells (Lee et al., 2001) and in Caco-2 colon [page 7↓]carcinoma cells (Kamitani et al., 2000). Human alveolar macrophages appear to constitutively express 15-LOX-1 in low levels as indicated by activity assays (Levy et al., 1993). Other cytokines did not induce the enzyme. In fact, the Th1 cytokine IFNγ was observed to inhibit gene expression induced by IL-13 in monocytes (Nassar et al., 1994). The IL-4 and IL-13 pathway share JAK1/2 kinases and transcription factor STAT6 (Heim, 1999). IL-4 and IL-13 failed to elicit any response in monocytes prepared from STAT6 deficient mice (Heydeck et al., 1998) and in human monocytes antisense treatment showed the involvement of JAK2 and TYK2 kinases in the upregulation (Roy and Cathcart, 1998). Pretreatment of the cells with IFNγ inhibited the phosphorylation of these kinases and stopped the signal transduction cascade. Involvement of factors such as SOCS (suppressors of cytokine signalling) have implicated in this process (Dickensheets et al., 1999).
|Figure 4 Promoter region of human 15-LOX-1 showing putative binding sites for transcription factors|
The promoter of 15-LOX-1 revealed a putative STAT6 binding site and mutation of this site (-952) abolished the IL-4 induced 15-LOX-1 expression (Conrad and Lu, 2000; Kritzik et al., [page 8↓]1997). Several other prominent transcription factor binding sites have been identified in this promoter region. A single copy of the “silencer element” identified in rabbit 15-LOX is present in the human promoter, however its function has not been elucidated. Kevalkar et al., 1998, have identified a 29 bp region (-352 to -304) which is important for the cytokine induced expression. Deletion of this region resulted in the absence of expression. Ku 70/80 has been identified as the protein binding to this region and controlling the gene expression (Kelavkar et al., 2000b). In Caco-2 cells, histone acetylation has been shown to play an important role. Cells treated with sodium butyrate, a hisone deacetylase inhibitor, expressed 15-LOX-1 without cytokine stimulation (Kamitani et al., 2001). Histone remodelling is a well known regulator of gene transcription and adds an interesting dimension to the study of transcription regulation of 15-LOX-1.
Expression of 12/15-LOX was induced in porcine smooth muscle cells (Natarajan et al., 1996) and murine system by IL-4 and IL-13 (Conrad et al., 1992), however, the involvement of STAT6 is not clear. No induction of 12/15-LOX expression by IL-4 was observed in peritoneal macrophages from STAT6 deficient mice (Heydeck et al., 1998) but the constitutive expression of the enzyme was not affected (Sendorby et al., 1998). Another cytokine IL-1β has been observed to upregulate rat 12/15-LOX expression in rat insulinoma cells both at the mRNA and protein levels (Bleich et al., 1995). Other authors have suggested that IL-1β induced expression of 12/15-LOX depends on an increased availability of substrate and on the cellular nitric oxide levels (Ma et al., 1996). Melatonin, a hormone involved in seasonal control of metabolism, significantly downregulated 12/15-LOX activity in pineal gland. It has been suggested that melatonin acts as an endogenous modulator of the 12 Lipoxygenase protein (Zhang et al., 1999). Though, a number of molecules appear to modulate the activity and expression of 12/15-LOX, the mechanisms or the signal transduction pathways responsible have not been elucidated in detail.
Translational regulation of 12/15 lipoxygenase expression has been investigated in detail in rabbit reticulocytes (Thiele et al., 1982), where it plays a major role in the maturation of the cell to erythrocyte. Abundant quantities of 15-LOX mRNA is present in immature rabbit reticulocyte but no active LOX enzyme could be detected. During maturation, the protein is expressed which by acting specifically on the mitochondrial membranes and promotes organelle degradation (Schewe et al., 1975). This process is controlled by translational [page 9↓]silencing by hnRNP k and E1 (Ostareck et al., 1997). These two RNA binding proteins bind to specific elements known as DICE (differentiation control elements) present in the 3’untranslated region of the 15-LOX mRNA . These elements consists of poly UAA stretches repeated several times. In cell free system, the hnRNP proteins specifically inhibit the assembly of 80S ribosome on the 15-LOX mRNA, thus preventing translation (Ostareck et al., 2001). As the process of maturation proceeds, the two proteins detach from the mRNA, thus permitting the translation to proceed.
A major factor in the regulation of the 12/15 LOX activity is the hydroperoxide tone, which quantifies the sum of all hydroperxides present in the cell (Weitzel and Wendel, 1993). Hydroperoxides activate LOXs by converting the non-heme iron from the ferrous to a ferric form. Inhibition of this conversion, by specific inhibitor like OPP, prevents the catalytic action of these enzymes (Richards et al., 1999). In vivo the hydroperoxide tone in the cell is regulated by glutathione dependent peroxidases. These enzymes convert hydroperoxy fatty acid to the hydroxide form, and this is paralleled by oxidation of GSH to GSSG (Hurst et al., 2001).
The Glutathione peroxidase (GPx) family consists of 4 major subfamilies: i) classical intracellular glutathione peroxidase (GPx-1), ii) plasma GPx, iii) gastro-intestinal Gpx (GI-GPx) and iv) phospholipid hydroperoxide GPx (PHGPx) (Flóhe et al., 1973; Takahashi et al., 1987; Chu et al., 1993). The common structural feature of these enzymes is presence of a selenocysteine residue. Of these, GPx-1 and PHGPx are the best characterised isoforms and they are found in almost all cell types. GPx-1 is a 26 kDa cytosolic protein, which exists as a homotetramer and utilises intracellular free hydroperoxy fatty acids as substrates (Flóhe, 1989). PHGPx, on the other hand, is a 23 kDa protein, which exists as a monomer and has the ability to react not only with free hydroperoxy fatty acids but also with esterified hydroperoxy fatty acids. It is preferentially associated with the membranes (Ursini et al., 1985; Thomas et al., 1990; Roveri et al., 1994).
These enzymes constitute an important part of the cellular anti-oxidant defence system. Overexpression of PHGPx has been observed to reduce apoptosis caused by oxidative stress in rat basophilic leukemia cells (Imai et al., 1996). The mitochondrial form of the enzyme has been found to be particularly effective as indicated by inhibition of apoptosis and increase in the level of Bcl-2, a mitochondrial anti-apoptotic protein (Nomura et al., 1999). 12/15-LOX can utilise membrane phospholipids as substrates for oxygenation. Pre-incubation of [page 10↓]membranes with PHGPx inhibited the oxygenation reaction, which was reversed by the addition of hydroperoxides (Schnurr et al., 1996). IL-4 and IL-13 treatment of A549 cells causes induction of 15-LOX-1 but also a concomitant decrease in the levels of PHGPx (Schnurr et al., 1999). Analysis of various tissues of mice overexpressing IL-4 revealed a negative correlation between the PHGPx levels and increased arachidonic acid oxygenase activity in certain tissues. This observation suggests that lipid peroxide generating and reducing enzymes are inversely regulated in various mammalian cells. It has been reported, that PHGPx is primarily responsible for the regulation of 5-lipoxygenase activity in rat leukocyte (Weitzel and Wendel, 1993).
A major reduction in the glutathione peroxidase activity was observed in the platelets from selenium deficient rats (Bryant et al., 1980). These platelets produced at least seven fold higher amounts of 12-HpETE as compared to control rats ((Bryant et al., 1983). Similar results were obtained in leukocytes from selenium deficient rats (Ho et al., 1997). Moreover, the selenium deficient rat platelets showed a three fold increase in the synthesis of isomeric trihydroxy fatty acids, TrXA3 and TrXB3 (Bryant et al., 1983). The isomerization pathway produces HXA3 and HXB3 from 12 HpETE which are rapidly hydrolysed by epoxide hydrolases to TrXA3 (Pace-Asciak and Lee, 1989). As the reductase activity of glutathione peroxidase is limited in selenium deficient rat platelets, the metabolism of 12-HpETE is diverted towards the isomerization route. This suggests the importance of glutathione peroxidases in the reduction of 12-HpETE. These authors, however, assume that the classical glutathione peroxidase, GPx-1 is the sole selenium dependent glutathione peroxidase enzyme present in platelets. However, platelets from GPx-1 knockout mice incubated with 25 µM arachidonic acid still synthesised 12-HETE (Ho et al., 1997). Furthermore, Gpx-1 deficient mice exhibited no phenotypic changes and the rate of lipid peroxidation and consumption of exogenous H2O2 was not altered in the tissues when compared with normal mice. This implies the ability of other enzyme(s) to take over the function of GPx-1, namely PHGPx. Selenium depletion studies also revealed a different response pattern between GPx-1 and PHGPx (Bermano et al., 1995). Severe selenium deficiency in rats caused complete loss of GPx-1 activity and GPx-1 mRNA levels in liver and heart, while the PHGPx activity was reduced by only 75%, and the mRNA levels were unaffected. In selenium-deficient rat basophilic leukemia cells, <1% GPx activity and 35% PHGPx activity were observed as compared to control cells and upon activation an enhanced 12-lipoxygenase activity was observed. Replenishment of selenium to these cells restored the PHGPx activity within 8 h, while the GPx-1 activity needed 7 days to return to normal levels (Weitzel and Wendel, [page 11↓]1993). The role of PHGPx and GPx-1 in the regulation of hydroperoxide tone seems to be different. An important factor for consideration is the localisation of the enzymes. PHGPx is preferentially associated with the membranes, which is also the site of eicosanoid synthesis. Similar observation has been reported in human platelets where PHGPx may have a special function in the regulation of 12-LOX activity (Sutherland et al., 2001).
Using a sensitive HPLC assay, glutathione transferases (GST) from α, μ and θ groups were observed to reduce phospholipid hydroperoxides to corresponding hydroxides using GSH as the reducing agent (Hurst et al., 1997). The α group exhibited the maximum activity, though it was significantly lesser than that of PHGPx. Also, in contrast to the glutathione peroxidases, these enzymes were inactive in the presence of detergent, Triton X100 (Hurst et al., 1998). The μ enzymes were, however, inacapable of reducing phospholipid hydroperoxides. Although, the activity of the GSTs is significantly lower than that of the glutathione peroxidases, their relative abundance (nearly 5% of total cellular protein) could be important. The activity of this class of enzymes with free fatty acid hydroperoxide, however, has not been reported so far.
Recently, a novel non-selenium glutathione peroxidase has been identified. The enzyme, 1-Cys peroxiredoxin, was first isolated from the bovine ciliary body in the olifactory mucosa (Shichi and Demar, 1990) and has been shown to catalyze the reduction of both free fatty acid hydroperoxides and phospholipid hydroperoxides using glutathione as electron donor (Fisher et al., 1999). This enzyme had no glutathione transferase activity when assayed with a spectrum of potential SH donors (Shichi and Demar, 1990). Furthermore, 1-Cys peroxiredoxin also exhibited PLA2 like activity and both the activities were reported to reside in distinctive active sites (Chen et al., 2000).
Nitric oxide has been shown to regulate the activity of 12/15-LOXs in vitro. Short incubations of the enzyme with NO considerably increases the kinetic lag phase which was shortened upon long term incubations (Holzhütter et al., 1997). Together with EPR and X-ray absorption studies, these data suggest that during NO-LOX interaction oxidation of ferrous non-heme iron may have occurred (Wiesner et al., 1996).
12/15-LOXs exhibit the capability to oxidise biomembranes. In vitro reticulocyte-type 15-LOX was observed to bind reversibly to biomembranes such as, submitochondrial particles and erythrocyte ghosts (Brinckmann et al., 1998). This membrane [page 12↓]binding activity was enhanced by the addition of calcium. Interestingly, the addition of calcium also enhanced the membrane oxygenase activity of the enzyme. This feature was confirmed in vivo. In rabbit reticulocytes and IL-4 stimulate peripheral monocytes, electron microscopic studies showed the localisation of the enzyme to the plasma and intracellular vesicles. Calcium ionophore augmented the membrane binding share of 15-LOX in human eosinophils, where it is expressed constitutively. Only under these conditions were specific lipoxygenase products of membrane lipids observed membrane oxygenase activity observed (Brinckmann et al., 1998). Thus, calcium not only enhances the membrane translocation but also the catalytic activity of 15-LOX in hematopoetic cells. Similar observations were made in colon carcinoma, Caco-2, cells. The presence of calcium was essential for the translocation and the catalytic activity of the 15-LOX-1, which was overexpressed in these cells (Hsi et al., 2001).
The predominant product of AA and LA (the major polyunsaturated fatty acids in mammalian cells) metabolism is 12-HpETE, 15-HpETE and 13-HpODE. These metabolites can be further processed into 3 primary types of products.
12-HpETE, 15-HpETE and 13-HpODE can be converted to the corresponding hydroxides by the action of glutathione peroxidases.
Lipoxin B4 (5S,14R,15S-trihydroxy-6E,8Z,10E,12E-eicosatetraenoic acid) is a product of the 15 lipoxygenase (Kühn et al., 1987). Lipoxin B4 is a biologically important compound. It is involved in the activation of leukocytes and the inhibition of neutrophil recruitment, chemotaxis and adhesion (Fierro et al., 2002). Lipoxins, especially aspirin triggered 15-epi-lipoxin A4, are potent inhibitors of acute inflammation (Takano et al., 1997).
A concerted action of the oxygenase and hydroperoxidase activty leads to the formation epoxy-leukotrienes such as 14,15 epoxyleukotriene A4 (Bryant et al., 1985).
The hydroperoxy group is homolytically cleaved to alkoxy radical initiating the formation of epoxyhydroxy compounds (involving the hydroperoxidase activity), keto-dienes, aldehydes and alkanes (Veldink et al., 1997). Of these, the epoxyhdroxy compounds of 12-LOX , hepoxilins, have been best studied (Pace-Asciak and Asotra, 1989).
Hepoxilins are monohydroxy-epoxy derivatives of arachidonic acid which are products of 12-HpETE. Two forms of hepoxilin are known, HxA3 (8(R) and 8(S)-hydroxy-11(S),12(S)-epoxyeicosa-5Z,9E,14Z-trienoic acid and HxB3 (10(R) and 10(S)-hydroxy-11(S),12(S)-epoxyeicosa-5Z,8Z,14Z-trienoic acid) (Pace-Asciak et al., 1983). The intra molecular re-arrangement of 12-HpETE can proceed by two main pathways.
|Figure 5 Bifurcation of the 12-Lipoxygenase pathway.|
12(S)-HpETE undergoes heme-dependent isomerization to produce HxA3 and HxB3 in equimolar amounts (Pace-Asciak, 1984a; Pace-Asciak and Asotra, 1984; Pace-Asciak, 1984b). Incubation of a racemic mixture of 12(R/S)-HpETE with hemin produced an equimolar mixture of HxA3 and HxB3 and the R derivatives produced the corresponding hepoxilins which do not occur in vivo (Pace-Asciak et al., 1995). This reaction was insensitive to heat. Both the insensitivity to heat and lack of substrate selectivity suggested that hemin catalysed reactions were non-enzymatic. Bryant et al., 1980; 1983, observed that platelets from selenium deficient rats produced more 12-HpETE and tri-hydroxy derivatives upon incubation with AA. This product was identified as a mixture of isomers of Trioxilin A3 (TrXA3), hydrolysis product of HxA3.
Sutherland et al., 2001, have demonstrated that platelets produce an equimolar mixture of HxA3 and HxB3 from 12-HpETE. The inability of the earlier authors to observe both the hepoxilins in platelets has been attributed to the instabilty of hepoxilins in acidic extraction conditions (Pace-Asciak, 1994a). Similar observations were made in trout gills (German and [page 14↓]Kinsella, 1986). Rat pineal gland slices, on the other hand incubated with 12(R/S)-HpETE produced exclusively HxA3. Here 12(S)-HpETE was selectively consumed (Reynaud et al., 1994). Tissue, boiled prior to use, did not produce any hepoxilins. Taken together, these data indicate the presence of a heat labile enzyme which stereoselectively converts 12(S)-HpETE to HxA3.Similar reaction was also observed in rat brain and skin (Pace-Asciak et al., 1993). This enzyme, designated as “hepoxilin synthase”, did not seem to require any cofactors.
Another probable pathway for the production of hepoxilins is the cytochrome P450 pathway. Incubation of 15(S)-HpETE with rat liver microsomes produced the mono-hydroxyepoxy derivative, 11S-hydroxy-14S,15S-trans-epoxyeicosa-5Z,8Z,12E-trienoic acid (Weiss et al., 1987). 15(R)-HpETE was converted into the corresponding 15R derivative. The P450 2B1 enzyme gave products similar to those obtained from the purified microsomes. Similar derivatives were also observed upon the incubation of 15-HpETE with garlic root tips (Reynaud et al., 1999). Reticulocyte 15LOX, under anaerobic conditions, produces significant amounts of a 13-hydroxy-14,15-epoxy eicosatrienoic acid. This compound was also observed under aerobic conditions but to a significantly lesser extent (Kühn et al., 1986a). In algae, another enzyme has been isolated which produces epoxy hydroxy derivatives from free fatty acids (Moghaddam et al., 1990).
HxA3 is chemically and biologically unstable due to the intrinsic instability of the epoxide ring (Pace-Asciak and Lee, 1989; Pace-Asciak and Asotra, 1989). The epoxide ring is rapidly hydrolysed in vivo by epoxide hydrolases to form the tri-hydroxy derivative, TrXA3 (8,11,12-trihydroxy-eicosatrienoic acid) (Pace-Asciak and Lee, 1989; Pace-Asciak et al., 1986). Similar reaction also occurs in acidic media, particularly during the acidic extraction of lipids. The inhibition of epoxide hydrolase by trichloropropene oxide (TCPO) (Pace-Asciak et al., 1986) diverts the conversion of HxA3 towards conjugation. Purified Yb2 subunits of glutathione transferases produce a glutathione conjugate upon reaction with HxA3 (Laneuville et al., 1991).
15-LOX was discovered in rabbit reticulocytes as the agent responsible for the destruction of mitochondria during its maturation to erythrocytes. It was hypothesised that peroxidation of lipids leads to membrane destruction (Rapoport et al., 1982; Grullich et al., 2001). Recent evidence has shown that the enzyme may actually oligomerise and integrate into the lipid bilayer leading to the formation of hydrophilic pores (van Leyen et al., 1998). 12/15-LOX [page 15↓]deficient mice, however, did not exhibit any significant defects in erythropoiesis (Sun and Funk, 1996). These mice did not exhibit any major malfunctions even under conditions of experimental anemia. Eye lens cells undergo a similar process of enucleation and destruction of organelles during maturation (van Leyen et al., 1998). 15-LOX has been shown to play an important role in this process by preferentially targetting the organelle membranes. These reports also suggest that similar mechanism might be operative in keratinocytes. A role for 15-LOX in differentiation of human tracheobronchial cells has also been suggested (Hill et al., 1998b). Retinoic acid induced differentiation and the expression of 15-LOX was observed in the later stages of the process, while undifferentiated cells do not produce any 15-LOX. 12/15-LOX translocated to the plasma membrane in monocytes incubated with apoptotic cells to the sites where actual contact was observed and altered the actin polymerization suggesting a role for this enzyme in the process of phagocytosis (Miller et al., 2001). However, 12/15-LOX deficient mice did not exhibit any physical abnormalities such as defective eye lens or any immune abnormalities. The underlying mechanisms need to be elucidated.
Metabolites of arachidonic acid may play a fundamental role as mediators in the development of airway inflammation in asthmatics (Chavis et al., 1998). Since the identification of “slow reacting substance of anaphylaxis” as leukotrienes C4/D4 most of the research on asthma was concentrated around the key enzyme 5-lipoxygenase, which led to development of numerous drugs. However, a real breakthrough has not yet been achieved. It is now known that the bronchial hyperreactivity is not solely caused by increased release of mediators of bronchoconstriction and inflammation, but the increased sensitivity of bronchial receptors towards these mediators plays a pivotal role in the pathogenesis of asthma. The lung epithelial cells express 5- and 15- lipoxygenases, which produce a variety of metabolites, such as 5-HETE, 15-HETE and leukotrienes, but also cyclooxygenases (Salari and Chan-Yeung, 1989; Hill et al., 1998a). The expression of these enzymes is tightly regulated in different types of cells (Moore et al., 2001; Conrad et al., 1992). Expression of 15-LOX enzyme is significantly increased in bronchial epithelial cells, and eosinophils (Conrad et al., 1992; Levy et al., 1993; Conrad and Lu, 2000). Consequently, large amounts of 15-HETE were observed in bronchial epithelial cells of asthmatic and emphysema patients as compared with those of normal subjects (Campbell et al., 1993). Since 15-LOX dioxygenates not only free arachidonic acid, but also biomembrane phospholipids, a considerable alteration in the affinity of various receptors with respect to 15-HETE and other lipid mediators may be expected. IL-4, which is also upregulated during asthma promotes incorporation of 15-HETE into the phospholipids [page 16↓](Profita et al., 1999). Similarly exposure of human tracheal epithelial cells to ozone, a major oxidant in environmental pollution, increases 15-HETE production and its esterification into phospholipids (Alpert and Walenga, 1995). Early allergic response was enhanced upon pre-inhaling 15-HETE with no effect on the late response (Lai et al., 1990b). Inhaled 15-HETE as such did not have differential effects on the airway of either normal or asthmatic individuals (Lai et al., 1990a).
A number of reports have suggested a role for 15-LOX in other bronchial diseases like bronchitis and chronic obstructive pulmonary disease (COPD). In COPD patients, alveolar apoptosis was observed along with decrease in 15-LOX protein and vascular endothelial growth factor RNA levels (Kasahara et al., 2000). Recently, increased 15-LOX-1 protein and mRNA expression was observed in the bronchial biopsies from patients suffering from chronic bronchitis (Zhu et al., 2002). Further, increased IL-4 protein and mRNA levels were also associated with this condition. The association of IL-4 and 15-LOX-1 in bronchial diseases is interesting as it has been previously observed that IL-4 strongly upregulated 15-LOX-1 levels in bronchial cells (Conrad et al., 1992).
12/15-lipoxygenase and its metabolites seem to exhibit both pro-inflammatory and anti-inflammatory activities. Elevated levels of 15-HETE and 13-HODE are associated with inflammation in various disorders such as human proctocolitis (Donoiwitz, 1985; Zijlstra et al., 1991), psoriasis (Duell et al., 1988). Psoriasis is an interesting example of the complexity of the lipoxygenase family. Both 12/15-LOX and 12R-LOX have been identified in psoriatic scales from patients (Boeglin et al., 1998). Increased levels of 15(S)-HETE, 12(S)-HETE, 13(S)-HODE and 12(R)-HETE have been observed in psoriatic cells when compared to normal cells (Baer et al., 1991). The relevance of these metabolites in the progression of the disease is unclear though pro-inflammatory cytokines such as IL-4 may play an important role in this disease (Asadullah et al., 2002). Human rheumatoid arthritis B synoviocytes express 15-LOX-1 and produce 15-HETE (Liagre et al., 1999). The role of this metabolite has not been studied in disease progression even though IL-4 levels were shown to be elevated under these conditions. 13-HODE, 12-HETE induced chemotaxis in neutrophils (Henricks et al., 1991; Cunninghham et al., 1986), however, 15-HETE inhibited the migration of these cells across cytokine activated endothelium (Takata et al., 1994a).
The 12/15-LOX products are capable of exhibiting anti-inflammatory properties in a protective role or reverse the inflammatory symptoms. Leukotriene B4, a major 5-LOX product is one of the most important mediators of acute inflammation. Remodelling of [page 17↓]polymorphonuclear leukocyte membranes with 15-HETE reduced the leukotriene B4 cell surface receptor affinity (Takata et al., 1994b), similarly increased levels of 15-HETE were correlated with inhibition of leukotriene B4 formation and synovial cell proliferation (Herlin et al., 1990) in experimental arthritis. Leukotriene B4 synthesis and its chemotactic activity were specifically antagonised by 15-HETE in rat experimental glomulonephritis (Fischer et al., 1992). Experimental glomerulonephritis was prevented in rat kidneys transfected with human 15-LOX-1 gene indicating a major anti-inflammatory role for 12/15-LOX (Munger et al., 1999). 15-HETE inhibited superoxide formation and exocytosis in neutrophils stimulated with phorbol ester or platelet activating factor (Smith et al., 1993).
Many contradictory reports have appeared with respect to the role of 12/15 lipoxygenases in carcinogenesis. Transfection of PC3 prostate cancer cell line with 15-LOX-1 showed a pro-carcinogenic effect as measured by several parameters (Kelavkar et al., 2001). These cells exhibited decreased anchorage dependent growth, increased proliferation and an increase in levels of angiogenic factors such as vascular endothelial growth factor. The cells also produced larger and more tumors when transplanted into athymic nude mice as compared to control cells. Prostatectomy specimens showed a significant co-expression of 12/15-LOX and mutant p53 coupled with the fact that elevated levels of 13-HODE were found in prostate carcinoma samples (Kelavkar et al., 2000a; Spindler et al., 1997). These data also correlated with elevated Gleason rating and the level of 12/15-LOX expression. A report which appeared six months later, however, showed that 15-LOX-2 metabolite 15-HETE caused dose dependent inhibition of PC3 proliferation by activating PPARγ transcription factors (Shappell et al., 2001). Recently, Hsi et al., 2002, observed that 13-HODE (15-LOX-1) and 15-HETE (15-LOX-2) have opposing effects on the regulation of PPARγ in epidermal growth factor signalling in prostate cancer cells and offer this as an explanation to the contradictory results obtained. Thrombin produced rapid pseudopod formation and detachment involving the formation of 12 and 15-HETE in rat prostatic carcinoma cells (Ross et al., 2000).
Similar contrasting results were obtained in studies performed on colon carcinoma. Increased expression of 12/15-LOX was observed in colon carcinoma tissue as compared to matched normal tissue as measured by PCR, protein expression and immunohistochemistry by Ikawa et al., 1999. On the other hand, Shureiqi et al., 1999, found significantly reduced amounts of immunohistochemical staining and 13-HODE levels in colon carcinoma tissue as compared to normal ones. The same group also reported that non-steroidal anti inflammatory drugs (NSAIDs) could upregulate 15-LOX-1 and 13-HODE and induce apoptosis in these cells [page 18↓](Shureiqi et al., 2000). Caco-2 colon carcinoma cells exhibited apoptosis upon sodium butyrate treatment which corresponded to increased 12/15-LOX expression (Kamitani et al., 1998, 2001).
Honn and group have shown in a number of reports that 12-HETE and platelet type 12-LOX can function as inducers of carcinogenesis and metastasis (Nie et al., 2001; Honn et al., 1994a). Furthermore, lipoxygenase inhibitor NDGA induces apoptosis in a number of carcinoma cell lines. The role of 12/15 lipoxygenase and its metabolites is complex and the mechanism of action of these contradictory effects is yet unclear.
Atherosclerotic lesions develop in a characteristic fashion,first appearing as fatty streaks. They consist of lipid-richmacrophages, foam cells localised beneath an intactendothelial cell layer in the arteries (Ross, 1999). Oxidation of LDL plays a key role in theearly pathogenic events, leading to foam-cell formation andfatty streaks (Witzum, 1994; Steinberg and Witzum, 1999). Products of oxidized LDL (oxLDL) are chemotacticfor monocytes, promote endothelial cell binding of monocytes,and once the monocytes have been recruited, inhibit the motility of macrophages. Macrophagescan both initiate the oxidation of LDL and take up oxLDL inan unregulated manner, leading to foam-cell formation (Steinberg and Witzum, 1999; Cushing et al., 1990; Watson et al., 1997). There isnow considerable evidence to support the presence of oxLDL in vivo, at least in animal models (Steinberg and Witzum, 1999).Many enzyme systems or nonenzymatic oxidative mechanisms havebeen demonstrated to induce oxidation in vitro (Heinecke, 1997). In vitro 15-lipoxygenase has been demonstrated to oxidise LDL (Kühn et al., 1994a). 15-LOX transfected fibroblasts and monocytic cells exhibited increased conversion to the atherogenic form of LDL (Benz et al., 1995) and impaired oxidation ability was observed in zymosan treated macrophages from 12/15-LOX deficient mice (Sun and Funk, 1996). Several other lines of evidence support the pro-atherogenic role of 12/15-LOX. The mechanism(s)responsible for oxidation of LDL in vivo, however, remain tobe defined.Human atherosclerotic lesions showed 12/15-LOX activity (Henriksson et al., 1985). Furthermore, 12/15-LOX mRNA and protein have also been detected in many different animal models (Hiltunen et al., 1995; Yla-Herttuala et al., 1990; Hugou et al., 1995), however, due to non-specificity of the experimental procedures these results need to be confirmed. The oxLDL from cholesterol fed rabbits showed the presence of 12/15-LOX specific products and the presence correlated temporally with the onset of lipid deposition on the arterial wall (Kühn et al., 1994b). Though similar results were obtained in human atherosclerotic lesions, the share of 12/15-LOX product is much reduced [page 19↓](Folcik et al., 1995; Kühn et al., 1997). Added to this is the clarification of the exact enzymatic nature of the oxLDL formation as several other enzymatic systems such as cytochrome P450, myeloperoxidase and cyclooxyneases and non-enzymatic process can confuse the results.
Contrasting data showing the anti-atherogenic effects also exist. 15-LOX transgenic rabbits developed significantly lesser amount of atherosclerotic lesions when fed western type diet (Shen et al., 1996). Experimental anaemia in rabbits causes the overexpression of 12/15-LOX and these animals fed on cholestrol rich diet developed fewer lesions as compared to control animals (Trebus et al., 2002). Similar effects were also seen in mice (Paul et al., 1999). Atherosclerosis is a complex disease requiring the interplay of several factors and the elucidation of the role of 12/15-LOX in this process requires intensive research.
Another area in which the role of 12/15-LOXs have been studied is diabetes melitus. In type I insulin dependent diabetes, the insulin producing β islet cells in the pancreas are destroyed by a cytokine dependent autoimmune process. The cytokines primarily involved are IL-1 β, IFNγ and TNFα and according to current hypothesis, this is achieved via the activation of nitric oxide (NO) (Eizirik and Mandrup-Poulsen, 2001). IL-1 β upregulates the expression of 12/15-LOX in rat insulinoma cell lines and in freshly isolated islet β cells (Bleich et al., 1995). Increase in the 12/15-LOX mRNA, protein and 12-HETE were observed post IL-1 β treatment. Some authors have, however, argued that this increase in 12-HETE production is due to increased substrate availability facilitated by NO mediated pathways rather than an upregulation of the enzyme activity as such (Ma et al., 1996). 12/15-LOX deficient mice have been observed to be highly resistant to streptozotocin induced diabetes (Bleich et al., 1999) and increase in 12-HETE levels were observed in patients suffering from this disease (Antonpillai et al., 1996). 12/15-LOXs are lipid peroxidising enzymes and have the ability to generate free radical mediated oxidative stress. Rat insulinoma cells, Rin m5F, transfected with cGPx earlier demonstrated as an antagonist of 12/15-LOX, show increased resistance to cytokine induced apoptosis and necrosis (Lortz et al., 2000). Cells transfected with other antioxidant enzymes such as, superoxide dismutase and catalase also show similar resistance (Tiedge et al., 1999).
A different picture emerges of the role of 12/15-LOX in type-II, non-insulin dependent, diabetes. Hepoxilin induces the production of insulin when injected in rats (Pace-Asciak et al., 1999) and 12-HETE also induces the production of glucagon from the α cells (Falck et [page 20↓] al., 1983). Thus, the exact role of 12/15-LOX in both types of diabetes needs to be elucidated, preferably by the use of specific inhibitors.
Apoptosis, or programmed cell death, is considered a normal physiological process and a major form of cell death that is used to remove damaged or infected cells throughout the life. Apoptosis is also a mechanism by which the organism deals with stress, injury and factors threatening its integrity such as infection. Apoptosis is therefore important in normal cell development, occurring during embryogenesis as well as in the maintenance of tissue homeostasis (Nagata, 1997).
|Figure 6Overview of apoptotic signalling pathways.|
Apoptosis is a regulated physiological process leading to cell death characterized by cell shrinkage, membrane blebbing and DNA fragmentation (Kerr et al., 1972). Caspases, a family of cysteine proteases, are central elements of apoptosis. Initiator caspases (including 8, 9, 10 and 12) are closely coupled to pro-apototic signals. Once activated, these caspases cleave and activate downstream effector caspases (including 3, 6 and 7) which in turn cleave cytoskeletal and nuclear proteins and induce apoptosis (Thornberry and Lazebnik, 1998). Principally two pathways initiate the signalling cascade that results in apoptosis.
This pathway is triggered in response to DNA damage, oxidative stress, chemotherapeutic agents and other types of stress (Vaux, 2002). It involves the Bcl-2family of proteins which consists of pro- and anti-apoptotic members (Gross et al., 1999). The anti-apoptotic members such as Bcl-2 and Bcl-XL exist in the mitochondria (Kluck et al., 1997; Vander Heiden et al., 1997). Upon receiving the stress signal, the pro-apoptotic memebrs (bax, bid, bak, PUMA) translocate to the mitochondria and neutralise the anti-apoptotic members by oligomerisation (Shimizu et al., 1999; Rosse et al., 1998; Luo et al., 1998). This results in the permeabilisation of the membrane resulting in the release of apoptogenic factors such as SMAC/Diablo, cytochrome c, SIMPs and AIF (Garl and Rudin, 1998; Matsuyama and Reed, 2000). Cytochrome c binds to APAF-1 to form the apoptosome resulting in the cleavage of caspase-9 and progression of the cascade via the effector caspases (Li et al., 1997).
A number of pro-apoptotic stimuli including FasL, TNF, DNA damage and ER stress can initiate this pathway (Budihardjo et al., 1999; Nagakawa, 1999). Cytosolic p53, induced by DNA damage, has been observed to activate bax, PUMA, Noxa mitochondrial translocation (Oda et al., 2000). This process can be inhibited by the overexpression of Bcl-2 and Bcl-XL (Camilleri-Broet et al., 2000). Fas ligand or TNF receptors can also induce apoptosis utilising both intrinsic and extrinsic pathway (Scaffidi et al., 1998). The caspase-8 activated by these signals cleaves Bid which then translocates to the mitochondria to bind to Bcl-XL triggering cytochrome c release (Li et al., 1998). ER stress leads to the calcium-mediated activation of caspase 12 and proceeds via the cleavage of caspase-3 (Nagakawa, 1999). Anti-apoptotic ligands including growth factors and cytokines activate AKT kinase and p90RSK, which inhibit Bad and prevent cytochrome c release (Bonni et al., 1999; Datta et al., 1999). TNFR can also stimulate an anti-apoptotic pathway by inducing IAP, which directly inhibits [page 22↓]caspases 3, 7 and 9 (Roy et al., 1997). AIF (apoptosis inducing factor) released from the mitochondria (like cytochrome c) can directly bind to the nuclear DNA and induce its damage leading to cell death (van Loo et al., 2002).
This pathway is used in a number important physiological functions such as cytotoxic T cell mediated apoptosis. External signals such as FasL (produced by cytotoxic T cells on their surface) and TNFα can activate pathway specific receptors of the TNF superfamily known as death receptors (TNF-R1, CD95, TRAIL R1 and R2) (Baker and Reddy, 1998; Ashkenazi and Dixit, 1998). The receptor oligomerisation results in the recruitment of specific adapter proteins such as FADD, TRADD, RIP and DAXX (Chinnaiyan et al., 1995; Hsu et al., 1995; Yang et al., 1997; Chaudhary et al., 1997). This complex is designated as DISC (death- inducing signalling complex). The formation of DISC leads to the recruitment and cleavage of pro-caspase-8 and 10 to their active forms. Caspase-8 and 10 function as initiator caspases leading to the activation of effector caspases. Cells can be divided into type I and type II depending upon the pathway utilised by activated caspase-8 (Scaffidi et al., 1998). In type I cells, sufficient amounts of caspase-8 exists and therefore directly cleave the effector caspases. While the type II, as described earlier, due to limited amounts of caspase-8, the mitochondrial amplification step is utilised.
Arachidonic acid and its lipoxygenase and cyclooxygenase products have been observed to modulate apoptosis signalling (Tang et al., 2002). This occurs either by various metabolites of the AA cascade acting as ligands for signalling pathways or via the generation of oxidative stress by lipid hydroperoxides. 12/15 lipoxygenases can act on both free fatty acid and on membrane phopholipids to produce lipid hydroperoxides. In rat glioma cells, it has been observed that glutathione depletion leads to increased production of 12/15-LOX and causes cell death (Higuchi and Yoshimoto, 2002). NSAIDs induce the expression of 15-LOX-1 in colorectal and gastric cancer cells and trigger apoptosis (Shureiqi et al., 2000; Wu et al., 2003). SC-236, a specific inhibitor of COX-2 induced 15-LOX-1 and 13-HODE production (not 15-HETE) and initiated apoptosis in gastric cancer cells (Wu et al., 2003). Apoptosis was abolished by the inhibition of 15-LOX-1. Sodium butyrate treatment of rat intestinal epithelial cells led to differentiation and apoptosis, induction of 12/15-LOX has been suggested to be involved in this process (Kamitani et al., 1999). Treatment of RBL2H3 cells with 2-deoxyglucose leads to the production of lipid hydroperxides and subsequent apoptosis. [page 23↓]Overexpression of the mitochondrial form of PHGPx, a scavenger of lipid hydroperoxide radical, led to the reduction in hydroperoxide levels and suppression of apoptosis (Imai et al., 1996; Nomura et al., 1999). Similar protection against apoptosis by PHGPx was observed in rabbit aortic smooth muscle cells challenged with linoleic acid hydroperoxide (Brigelius-Flóhe et al., 2000).
Increased levels of 12-HETE, produced by the reduction of 12-HpETE by glutathione peroxidase has been observed in several cancers and is correlated to their metastaic potential (Tang and Honn, 1999). Overexpression of 12-LOX platelet type enzyme in MCF-7 mammary carcinoma cells, in human prostate adenocarcinoma cells resulted in cell proliferation and larger tumors (Liu et al., 1996; Connolly and Rose, 1998). Inhibition of 12-LOX platelet type enzyme by NDGA and other inhibitors in Walker 256 carcinoma cells, gastric cancer cells and in lewis lung carcinoma cells resulted in the growth inhibition and apoptosis (Wong et al., 2001; Tang et al., 1997; Honn et al., 1994b). Similar effect has also been observed in melanoma and prostate cancer cells (Onoda et al., 1994; Pidgeon et al., 2002). Thus, the two 12 lipoxygenase isoforms have opposite effects on cell proliferation and apoptosis. It could be argued that conversion of 12-HpETE to 12-HETE reduces the oxidative stress in the cell. Another possibility is that 12-HETE functions as a ligand for proliferative pathways. AA and its products act as ligands for PPAR and other receptors. These receptors could have variable effects on cell proliferation.
Peroxisome proliferator-activated receptors (PPARs) are transcription factors belonging to the nuclear receptor gene family and are important regulators of fatty acid metabolism. PPARs act in a similar manner to other nuclear hormone receptors (Kersten et al., 2000). First, they bind a specific element in the promoter region of target genes. PPAR bind to the promoter as a heterodimer with retinoid X receptor (RXR) (Gearing et al., 1993). Second, they activate transcription in response to the binding of ligand. Eicosanoids, fatty acids and synthetic compounds known as peroxisome proliferators act as ligands to PPARs.
Three PPAR isotypes have been identified: α,β and γ. Some ligands are shared by the three isotypes, such as polyunsaturated fatty acids and probably oxidized fatty acids. Several compounds bind with high affinity to PPARα, including long-chain unsaturated fatty acids such as linoleic acid, branched, conjugated and oxidized fatty acids such as phytanic acid and conjugated linolenic acid, and eicosanoids such as 8(S)-HETE and leukotriene (LT) B4. The [page 24↓]prostaglandin 15-deoxy-Δ12,14-prostaglandin J2 is the most potent natural ligand of PPARγ. 12/15-LOX products 13-HODE and 15-HETE also act as ligands for this receptor (Kersten et al., 2000; Gearing v 1993; Hihi et al., 2002; Forman et al., 1995; Lehmann et al., 1995). Much of the function of PPARs can be extrapolated from the identity of their target genes, which so far all belong to pathways of lipid transport and metabolism.
The role of 12/15-LOX and PPARγ in atherosclerosis is quite confusing. PPARγ activation in monocytes and macropphages induce their apoptosis (Chinetti et al., 1998) and PPARγ liagnds reduce the development of atherosclerotic lesions in LDL-receptor knockout mice indicating anti-atherosclerotic effect (Li et al., 2000). While, pro-atherosclerotic effect was the induction of CD36, a scavenger receptor of oxLDL by IL-4 and 12/15-LOX induced PPARγ activation (Huang et al., 1999). This was clearly demonstrated by the lack of CD36 induction in 12/15-LOX deficient mice and in PPARγ deficient stem cells (Chawla et al., 2001).
PPARγ activation has also been implicated in lung diseases. Increased PPARγ activation was observed in asthmatic patients as compared to normal controls and was correlated to apoptosis, airway remodelling and inflammation (Benayoun et al., 2001). Further, PPARγ agonists induced apoptosis in human lung carcinoma cells (Theocharis et al., 2002; Satoh et al., 2002; Inoue et al., 2001; Tsubouchi et al., 2000). Activation of death receptors, specifically TRAIL and inhibition of anti-apoptotic subunit of NF-κB p65/RelA have been implicated in PPARγ induced apoptosis, especially in macrophages (Chinetti et al., 1998; Ji et al., 2001).
|© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.|
|DiML DTD Version 3.0||Zertifizierter Dokumentenserver|
der Humboldt-Universität zu Berlin