* To whom correspondence should be addressed.
Received July 25, 1997
Most biological functions of unsaturated fatty acids are due to their ability to act as second messengers or modulators of activities of functionally important proteins; these functions are not related to oxidative metabolism of unsaturated fatty acids. These acids regulate the activity of phospholipases, ion channels, ATPases, G-proteins, and protein kinases; they also modulate the phosphoinositide and sphingomyelin cycles, the transfer of hormonal information, and gene transcription. The great diversity of effects of unsaturated fatty acids and their presence at the earliest stages of evolution suggest for these bioregulators a system-forming role in the living body.
KEY WORDS: phospholipases, sphingomyelin cycle, ion channels, second messengers, hormonal signaling, regulation of transcription
Abbreviations: AA) arachidonic acid; DAG) diacylglycerol; DHA) docosahexaenoic acid; FA) fatty acids; CBG) corticosteroid-binding globulin; OA) oleic acid; PUFA) polyunsaturated fatty acids; EPA) eicosapentaenoic acid; FABP) fatty acid-binding protein; GLUT4) glucose transporter 4; PAF) platelet-activating factor; PKC) protein kinase C; PLA2, PLC, and PLD) phospholipases A2, C, and D, respectively; PPAR) peroxisome proliferation activator receptor; TNF-alpha) tumor necrosis factor alpha.
G. Barr and M. Barr  were the first to suggest in 1930 that unsaturated fatty acids play a bioeffector role. The authors described symptoms caused by dietary insufficiency of the so-called essential FA in laboratory animals. It has been long believed that most functions of essential FA, including linoleic and linolenic acids, as well as their long-chain polyenic homologs, are mediated at the molecular level by their oxidized metabolites, whereas the acids themselves are only responsible for the integrity of cell membranes and the optimal levels of unsaturation of tissue lipids. Indeed, changes in FA profiles in membrane lipids can affect the incorporation, aggregation, and diffusive mobility of membrane components, the activity of membrane-associated enzymes, expression of receptors, the membrane permeability, and its transport properties . Many key functions of cells, such as secretion, chemotaxis, and sensitivity to microorganisms also depend on the membrane fluidity [3, 4]. After the discovery of prostaglandins and other biologically active oxidized metabolites of FA, the bioeffector functions of PUFA were explained by their conversion to these compounds ascribed to the class of oxylipins. Oxylipins are regulators of various cell functions and are involved in virtually all physiological and pathological processes. They act locally in tissues where they are produced and are rapidly converted to inactive forms.
However, recent years (the late 1980s and early 1990s) have seen an avalanche of data suggesting that free (nonesterified) FA are directly involved in many cellular processes; they regulate the activities of various enzymes, are involved (as second messengers) in intracellular signal transduction, and modulate the binding of steroid hormones to receptors, thereby modulating the transcription of certain genes (reviewed in [5-7]). This review will concern exactly these bioeffector functions of FA.
PUFA AS SECOND MESSENGERS AND MODULATORS OF INTRACELLULAR SIGNALING
PUFA belong to a large family of bioregulators that includes also oxylipins, PAF, DAG, lysophospholipids, ceramides, and FA amides. These bioregulators of the lipid nature are produced from other lipids by certain enzymes, primarily by phospholipases.
Phospholipase A2. The activity of mouse macrophage membrane-associated PLA2 was shown to be inhibited by AA, OA, linoleic acid, and linolenic acid, whereas saturated FA were not active. AA was the most potent inhibitor and behaved as a competitive inhibitor (Ki = 5 µM) . Calcium-dependent PLA2 from snake venom was strongly inhibited in vitro by micromolar concentrations of cis-unsaturated FA (OA and AA), whereas trans-unsaturated and saturated FA produced much weaker effects. The authors showed that cis-unsaturated FA bind directly to the isolated enzyme but do not affect the enzyme-substrate interactions. The in vitro inhibition of PLA2 is probably mediated by the formation of an enzymatically inactive complex of enzyme--substrate--fatty acid .
Thus, FA-mediated inhibition can be a negative feedback mechanism regulating PLA2 and blocking further production of free PUFA, thereby limiting the formation of oxylipins.
Phospholipase C. Exogenous OA and AA strongly activate the hydrolysis of membrane PtdIns in the microsomal fraction of rat liver by soluble PLC of the rat brain . The effect of AA on the activity of PLC can be mediated by certain metabolites of AA, such as endoperoxides or thromboxane A2 . However, experiments performed with acids that do not produce eicosanoids (OA and palmitoleic acid) showed that they block the signal transduction directly by affecting the receptor--PLC or PLC--substrate interaction . The activation of PLC-gamma from bovine brain by proteins belonging to the tau-family increases in the presence of unsaturated FA. The strongest activating effect (15-20-fold) was observed in the presence of 0.15 µM tau-protein and 25 µM AA, whereas saturated FA produced no such effect .
Phospholipase D. A number of works showed that PLD is stimulated by fatty acids, especially by OA and AA [14-16]. Certain PLD species, e.g., those isolated from rat liver plasma membranes, were most effectively stimulated by monounsaturated FA, whereas saturated FA caused no effect on this enzyme . However, it should be noted that this stimulatory effect of FA was only observed at rather high concentrations. For example, OA and AA caused maximal stimulatory effects on PLD from preparations of purified rat heart sarcolemmal membranes at concentrations of 5 and 4 mM, respectively .
Metabolism of PtdIns. FA modulate the phosphoinositide cycle. For example, AA produces a dose-dependent inhibition of phosphoinositides synthesis in acinous cells of rat submandibular gland [17, 18] and pancreas . This effect is PUFA-specific because arachidic acid did not affect the incorporation of Ins in PtdIns in cells of the pancreas , whereas in submandibular gland AA regulated the phosphoinositide cycle as a noncompetitive inhibitor of PtdIns4-kinase and PtdIns(4)P-5-kinase . In rat platelets, OA caused a dose-dependent inhibition of synthesis of PtdInsP and PtdIns(4,5)P2 from PtdIns. Maximal decreases in the synthesis of these phosphates by 50 and 25%, respectively, was observed within seconds after adding 20 µM OA. This effect was antagonized by PAF, which explains the inhibitory effect of OA on PAF-induced aggregation of rat platelets .
The synthesis of PAF is also affected by FA. In this case, unsaturated FA (AA, OA, and linoleic acid) produced stronger inhibitory effects on PAF synthesis in microsomal preparations of glandular cells of gastric mucosa in comparison to saturated acids .
The sphingomyelin cycle. This cycle, like the phosphoinositide and phosphatidylcholine cycles, is a key link in biosynthesis of biologically active lipid mediators resulting in the production of ceramides (reviewed in ). In myeloid leukemic HL-60 cells, the sphingomyelin cycle is activated by gamma-interferon  and TNF-alpha . These effects were shown to be mediated by AA. Both gamma-interferon  and TNF-alpha  stimulate a rapid release of AA in HL-60 cells. Adding free AA to the cells stimulates their sphingomyelin hydrolysis and the accompanying formation of ceramide. Other FA, for example OA, caused the same effect, whereas methyl esters or alcohol analogs were not effective . The stimulatory effect of AA on sphingomyelin hydrolysis is not related to its conversion to oxylipins .
Effects on the Concentration of Free Ions
The concentration of free ions is a fundamental parameter ensuring the homeostasis of cells and the whole body. Mechanisms of action of many bioregulators include changes in ion conductances of membranes and/or mobilization of ions from their stores. Many examples show that FA are active modulators of the concentration of free ions.
Calcium channels and calcium pools. In digitonin-permeabilized pancreatic islets, AA (1.25-10 µM) caused the release of considerable amounts of Ca2+ from endoplasmic reticulum. This effect was observed within 2 min and was not mediated by AA metabolites . A similar effect of AA on the release of intracellular calcium was also found in human T and B lymphocytes . AA also caused the release of Ca2+ from mitochondria and microsomes isolated from rat cerebellum . One can propose that this effect of AA is mediated by its action on the activity of PtdIns(4,5)P2-specific PLC (see above) and the related formation of InsP3, which mobilizes intracellular Ca2+. However, experiments with various cell types showed that AA and other acids of the omega6 and omega3 series caused rapid increases in intracellular Ca2+ concentrations without affecting the phosphatidylinositol cycle [28, 29].
On the other side, AA (3 µM) fully blocked the increase in the intracellular Ca2+ concentration caused by mitogens, such as concanavalin A (10 µM) and phytohemagglutinin (10 µM)  or isoprenaline, an antagonist of beta-adrenergic receptors (IC50 was achieved at an AA concentration of 1.52 µM) .
The omega6 and omega3 acids that mobilize intracellular Ca2+ inhibit the anti-CD3-induced calcium entry through receptor-mediated channels in leukemic JURKAT T-cells . The authors showed that the inhibitory effects of these unsaturated FA was not due to activation of PKC and suggested that it could be mediated by binding the acids to a Ca2+ channel or its separate components.
Experiments with guinea pig heart ventricular myocytes showed that long-chain FA (OA, linoleic, linolenic, and AA) increase the potential-dependent Ca2+ current when applied at a concentration of 3-30 µM. Short-chain FA (12 carbon atoms or shorter) caused no effect on Ca2+ current . However, AA and other cis-unsaturated FA (linoleic, palmitoleic, and OA), in contrast to trans-unsaturated FA, inhibited the potential-dependent Ca2+ current in intestinal smooth muscle cells . In both cases, neither PKC and protein kinase A inhibitors nor inhibitors of the AA cascade blocked the changes in Ca2+ current. These findings suggested that FA directly affected either lipid parts of the membrane adjacent to channels or the channels themselves .
The finding that omega3 PUFA (DHA and EPA) prevent cardiac arrhythmia by decreasing the myocardial conductance, stimulated further studies that showed that these FA modulate the inactivation of the voltage-dependent currents of Na+ and Ca2+ in neurons isolated from rat hippocampal CA1 area. These acids shifted the voltage-dependent inactivation curve toward more hyperpolarized potentials, indicating that they accelerated the inactivation and retained the recovery from inactivation. The IC50 for DHA was 2.1 µM; EPA displayed IC50 values of 4 µM and more than 15 µM for Na+ and Ca2+ currents, respectively . In addition, these acids decreased the amplitudes of Na+ and Ca2+ currents without affecting the activation of potential-dependent channels. Monounsaturated and saturated acids displayed no such activity. A combination of mentioned effects of EPA and DHA may reduce the neuronal excitability and can cause an anticonvulsant effect in vivo .
Thapsigargin, a Ca2+ pump blocker, depletes Ca2+ pools in smooth muscle cells and causes the transition of cells from the steady state to proliferative growth. Although thapsigargin irreversibly blocks Ca2+ pumps, it was shown that high concentrations of serum (20%) cause the appearance of new pump protein molecules, restore the functional pools, and returns the cells to performing their cell cycle . Further studies showed that the serum-derived factors responsible for this effect are essential fatty acids: AA, linoleic, and linolenic acids. The IC50 values for AA and linoleic acid were similar (5 µM). The restoration of bradykinin-dependent pools caused by these acids occurred within 6 h and was fully blocked by protein synthesis inhibitors (cycloheximide and puromycin) . Thus, essential omega6 and omega3 fatty acids neutralize the effect of the Ca2+ pump blocker, probably through stimulation of synthesis of appropriate proteins. This effect is specific to the acids listed above because myristic, palmitic, stearic, OA, and arachidic acids did not cause such an effect .
Depending on the concentration of AA, it can cause opposite effects on the concentration of intracellular Ca2+. Thus, AA at concentrations lower then 10 µM attenuated the increase in Ca2+ level caused by 2,5-di-(tert-butyl)-1,4-hydroquinone, whereas AA produces an additional Ca2+ level increase at concentrations higher than 10 µM . When added to intact cells at a concentration of 3-20 µM, AA increased the intracellular Ca2+ level. These data suggest that the effects of AA on intracellular Ca2+ concentration are mediated via multiple mechanisms .
Potassium channels. Several cis-PUFA and trans-PUFA, as well as cis-monounsaturated, and saturated FA, cause direct stimulatory effects on potassium channels in smooth muscle cells of the stomach , aorta, and pulmonary artery . However, unsaturated and saturated FA with a chain length of 10-14 carbon atoms, when added at a concentration of 5 µM, accelerated the inactivation of potential-dependent potassium channels in neuroblastoma cells . Moreover, in these cells unsaturated FA increased the sensitivity of potassium channels to the blocker 4-aminopyridine . AA inhibited ATP-dependent potassium channels in cells of the external medullary layer of rat kidneys  and modulated potassium conductance of oligodendrocytes, thereby causing membrane depolarization and protein phosphorylation .
Studies of the structure--activity relationships showed that negatively charged analogs of long-chain FA (but not of short-chain FA) contained in the medium, as well as certain other negatively charged lipids activate potassium channels of smooth muscle cells; positively charged lipids with medium- or long-chain hydrophobic chains suppressed this activity, whereas neutral lipids have no effect [44, 45]. Thus, these lipid effectors, in order to produce an effect, should have a hydrophobic fragment of a sufficient length and a charged group. Because negatively charged analogs of FA are incapable of flip-flop motion were only effective when introduced from the cytosolic side of the plasma membrane, the authors suggested that the site of action of FA is also located at the cytosolic side of the channel .
ATP-sensitive potassium channels of dog coronary artery smooth muscle cells were shown to respond to AA by increasing potassium current in the presence of ATP and display an opposite effect in the absence of ATP; these effects were not related to oxidative metabolism of AA .
Activation of K+ channels of gastric smooth muscle cells from the toad Bufo marinus induced by mechanical stimulation (stretch of the cell membrane) is probably also mediated by free PUFA because they potentiate this stimulatory effect. Free PUFA can be produced as a result of action of phospholipases activated by mechanical stimulation, whereas their removal (e.g., by defatted albumin) caused a rapid and irreversible decrease in the response to stimulation . This finding suggests that FA liberated in response to mechanical stimuli can act as messengers in signal transduction to mechanically sensitive K+ and other ion channels.
Sodium channels. Free FA modulate potential-dependent sodium channels in human skeletal muscle cells. Inside these cells, AA (1-20 µM), as well as OA and stearic acid, activated sodium but not potassium current, whereas AA administered extracellularly (5-10 µM) reversibly inactivated sodium current . Further studies showed that intracellular exposure of FA to sodium channels, depending on the channel isoform, could either increase or decrease the sodium current, and this effect was PKC-independent . Experiments with myocardial cells showed that FA causing anti-arrhythmic effect (EPA, DHA, eicosatetraenoic, linoleic, and linolenic) can bind to receptors on potential-dependent sodium channels thus modulating their activity. Saturated (stearic), monounsaturated (OA) acids, and EPA methyl ester did not display this activity . AA also decreased acetylcholine-induced ion currents by interacting with acetylcholine receptors . Removal of fragments of these receptors which are the targets for PKC phosphorylation abolished the regulatory effect of AA . Thus, one can propose that AA modulates acetylcholine-dependent currents by binding directly to these regions of acetylcholine receptors.
Chloride ion channels. In human fetal trachea epithelial cells, AA and other cis-unsaturated FA (including OA and linolenic) blocked Cl- channels in a dose-dependent manner. The mean time of the channel opening decreased tenfold in the presence of 25 µM AA, whereas trans-unsaturated and saturated FA produced no inhibitory effect on these channels . AA reversible inhibited Cl- channels of apical membranes of airway epithelium in normal and cystic fibrosis cells; this effect was only produced by AA added to the cytosolic surface of the membrane and was not related to oxylipin formation . Similar results were also obtained from experiments with cultured intestinal epithelial cells (Intestine 407), where IC50 for Cl- transport block was 8 µM .
Modulation of Regulatory Protein Activities
FA are active modulators of regulatory proteins, such as ATPases, G-proteins, and protein kinases. This issue was described in the literature in much detail; here, we shall consider only several examples.
ATPase. It was found that Na+,K+-dependent ATPase from rat brain synaptosomes is inhibited by micromolar concentrations of OA . The purified enzyme from rabbit kidneys was inhibited by unsaturated FA (palmitoleic, OA, linoleic, and AA) at an IC50 of less than 30 µM but not by saturated acids . However, in the presence of 50 µM ATP, PUFA activated highly purified Na+,K+-dependent kidney ATPase . The magnitude of this effect depended on the chain length and the degree of unsaturation of the fatty acid, as well as on the modification of the carboxyl group: methyl esters inhibited ATPase, whereas CoA esters and monoacyl esters of glycerol activated this enzyme .
Unsaturated FA inhibited H+,K+-dependent ATPase in the stomach; cis-monounsaturated FA were more active than trans-unsaturated acids in this respect. As in the case with Na+,K+-dependent ATPase, the FA effect depended on the length of its carbon chain, the number and positions of double bonds, and modification of the carboxyl group: esterification decreased the activity, whereas conversion of the carboxyl group to an alcohol or amide group did not decrease the activity .
G-Proteins, adenylate cyclase, and guanylate cyclase. It is well known that stimulation of G-proteins induces mobilization of AA; the liberation of AA is stimulated even by nonspecific activation of G-proteins in the presence of GTP or its nonhydrolyzable analog . However, AA and other unsaturated FA were shown to cause direct stimulatory effects on G-proteins in polymorphonuclear leukocytes. These effects lead further to cell aggregation, generation of superoxide anion radical, and release of lysosomal contents .
Changes in the activities of adenylate cyclase and guanylate cyclase as membrane-associated enzymes correlate with changes in fatty acid composition of their lipid environment and respectively with changes in physical properties of the membrane . No direct effect of nonesterified FA on adenylate cyclase was found; however, certain acids were shown to modulate the level of cAMP: exogenous gamma-linolenic acid produced a 30% decrease in hormone-induced synthesis of cAMP .
Modulation of guanylate cyclase activity by FA is not only due to changes in physical properties of the membrane. Micromolar concentrations of exogenous FA activate both the membrane-associated  and soluble isoforms of the enzyme . For example, AA at a concentration of 1-500 µM produced a direct stimulatory effect on purified soluble guanylate cyclase from rat brain, whereas higher concentrations caused a strong inhibitory effect. Other unsaturated FA, such as OA and linoleic acid, were far less potent modulators of this enzyme .
Protein kinases. In 1984, it was found that AA and other unsaturated FA activate PKC, and this effect is not mediated by their oxidized metabolites because inhibitors of cyclooxygenase and lipoxygenase did not affect this activation . After this publication, many studies showed an important role of fatty acids in the mechanism of activation of PKC; it was suggested that AA liberated from phospholipids can act as a second messenger in signal transduction mechanisms involving PKC activity [67-70]. Linolenic acid and AA activate PKC II and PKC III in epithelial cells of the mammary gland and mediate the effect of a growth factor . Other FA can also act as second messengers to activate various isoforms of PKC independently from DAG, phosphatidylserine , and Ca2+ . FA strongly increase the DAG-induced activation of PKC, that leads to the nearly maximal enzyme activity at low, near-basal levels of Ca2+ . In addition, unsaturated FA caused time- and dose-dependent translocation of PKC; specifically, AA and OA (in the presence of Ca2+) caused the translocation of PKC from the cytosol to the plasma membrane (an effect similar to that caused by stimulation of neutrophils with phorbol ester or formyl-Met-Leu-Phe ). This is not a detergent-like effect because it was not reproduced by administration of saturated and trans-unsaturated FA at the same concentrations .
There are specific differences in the activation of PKC by various classes of FA. A purified PKC from rat colon was activated by several C18 and C20 acids of both omega3 and omega6 series in the absence of other lipid activators; however, the same acids inhibited phosphatidylserine-DAG-stimulated PKC activity . DHA displayed a special type of activity, which did not stimulate PKC in the absence of phosphatidylserine or DAG but was a stronger inhibitor than other FA of phosphatidylserine-DAG-stimulated activity .
MODULATION OF HORMONAL SIGNALS
Many studies showed that nonesterified FA can modulate certain stages of transduction of hormonal information. FA caused specific changes in physicochemical properties of certain plasma protein that bind steroids, including corticosteroid-binding globulin , mouse alpha-fetoprotein , and vitamin D-binding protein . FA modulate the binding of estrogens, glucocorticoids , and vitamin D  to their receptors.
Regulation of binding activity of CBG in blood plasma. Glucocorticoid binding by CBG in blood plasma of immature rats is considerably decreased after increasing the concentration of nonesterified FA by stimulation of lipoprotein lipase by heparin. Studies performed in vitro showed that unsaturated FA are the most effective inhibitors. The concentrations of all types of FA increased by 150-200% during 60 min after an injection of heparin. After this treatment, the plasma levels of CBG and corticosterone did not change; however, there was a 60% decrease in the binding activity of CBG. This inhibition was explained by a 53% decrease in the apparent number of binding sites; the CBG-corticosterone affinity constant remained unchanged . In contrast, in adult animals an increase in plasma levels of free FA increased the binding of corticosterone by CBG in vivo and in vitro . This is explained by the dependence of the binding on the FA/CBG ratio: it increases at a FA/CBG ratio of 500 and considerably decreases at a FA/CBG ratio of 2000 . In immature and adult rats, this ratio is 6500 and 350, respectively. In addition, the CBG binding activity depends on its carbohydrate environment , which undergoes age-related changes: a highly glycosylated CBG species accounts for 9% of total CBG in immature rats and 26% in adult rats . The authors suggested that FA can cause various conformational changes in proteins depending on their different glycan coats, which result in different effects of FA on the binding activity depending on the carbohydrate environment.
Similar results were obtained from studies of the effects of changes in profiles of free FA in blood serum after food intake in adult women. An increase in the concentration of monounsaturated FA (especially OA) increased the binding activity of CBG because of conformational changes .
Regulation of binding of steroid hormones to their tissue receptors. An increase in the level of free FA in liver cytosol is accompanied by a decrease in the binding capacity of liver glucocorticoid receptors because of decreases in their binding constants and the total number of binding sites . It is proposed that steroid hormone receptors may have a binding site for endogenous modulators, such as FA, in the receptor fragment containing the hormone-binding domain and certain C-terminal sequences of the DNA-binding domain . In addition, recent studies showed that FA can bind to the receptor of peroxisomal proliferation factor, a member of the superfamily of nuclear steroid receptors ; therefore, FA not only modulate the binding of steroid hormones to receptors but also act as endogenous ligands of other members of this receptor family.
The profile of free FA in blood serum depends on sex, age, feeding regimen, and hormonal status; for example, adrenalectomy and castration also change the FA profile . In turn, changes in free FA concentrations causes considerable changes in two stages of transduction of hormonal signals: binding of glucocorticoids to their specific carrier proteins in blood plasma and tissue receptors.
TRANSCRIPTIONAL REGULATION OF CERTAIN GENES
Dietary PUFA of omega6 and omega3 families suppress lipogenesis in the liver, whereas monounsaturated and saturated FA produce no such effect [87-90]. It has been long known that inhibition of enzyme activity is due to a decrease in the number of the enzyme molecules rather than to changes in the catalytic efficiency . Later studies showed that dietary fatty acids inhibit the expression of fatty acid synthase and S14 protein in liver by suppressing the synthesis of mRNA species encoding these proteins . In addition, dietary FA block a 20-30-fold increase in mRNA levels in rat pups receiving a carbohydrate-rich diet . The mRNA level is decreased because of inhibition of transcription of genes encoding FA synthase and S14 by products of Delta6-desaturase of omega6 and omega3 FA . The concentration of FA synthase limits the capacity of tissues to synthesize FA de novo. In certain tissues (liver and adipose tissue) the level of FA synthase mRNA is sensitive to hormonal and alimentary factors, whereas in other tissues this sensitivity was not found . On the basis of these data a conception that the tissue-specific expression of the gene of FA synthase is mediated via a tissue-specific mechanism of inhibition of this expression  was formulated.
FA also inhibit the transcription of genes of other lipogenic proteins in the liver: Delta9-desaturase (thereby preventing the accumulation of omega7 and omega9 acids [95-97]), acetyl-CoA carboxylase and malic enzyme [98, 99], and glucose-6-phosphate dehydrogenase . An increase in the level of omega3 acids in the diet of rats with hereditary hypertriglyceridemia decreases the level of triglycerides through inhibition of expression of genes of lipogenic enzymes by omega3 acids . In addition, long-chain FA (palmitic, OA, linoleic, linolenic, and AA) stimulate the expression of FABP .
Recent data suggest that FA regulate the transcription of genes encoding not only lipogenic enzymes. For example, various long-chain FA including AA decrease the concentration of mRNA encoding GLUT4 (a glucose-transporting protein) in adipocytes [103, 104], and regulate the expression of the lipoprotein lipase gene in preadipose and adipose cells ; omega3-FA decrease the amount of mRNA of platelet growth factor  and interleukin-1beta [107, 108]. Platelet growth factor is an agent involved in pathogenesis of atherosclerosis and other inflammatory and proliferative diseases. Suppression of its synthesis is one of the mechanisms responsible for the antisclerotic effects of omega3-FA. DHA and dihomo-gamma-linolenic acid increase the number of transcripts of plasminogen activator inhibitor in endothelial cells . Long-chain FA suppress the glucocorticoid-induced expression of genes responsible for enzymes of the ornithine cycle . DHA is a potent inhibitor of expression of NO synthase in murine macrophages ; this can explain a higher susceptibility to infection in neonates because their serum DHA level is 10-50 times higher than in adult specimens. In cytokine-stimulated endothelial cells, mRNA of certain proteins is expressed and AA is liberated. The liberated AA in turn suppresses the synthesis of some of these proteins at the transcriptional level; therefore, the authors suggested that in this case there is regulation by a negative feedback mechanism .
A model explaining the mechanism of effects of FA on the rate of gene transcription was proposed in 1993. PUFA are transported through the plasma membrane and then interact with cytosolic FABP, which carries them to Delta6-desaturase and further transports the products generated by the enzyme to the nucleus. In the nucleus, polyenic FA ligands of the cytosolic FABP are transferred to a specific nuclear FABP. Having interacted with a FA, the nuclear FABP is activated and binds either to DNA directly or to a DNA-binding protein, thereby modulating the rate of transcription .
By that time, proteins belonging to the superfamily of steroid receptors capable of binding FA had been already discovered [113-115]. One such protein is PPAR. Agents that bind this receptor (e.g., clofibrate) can induce peroxisomal enzymes that oxidize FA; they also suppress the expression of FA synthase and inhibit the synthesis of FA in the liver [6, 114]. FA, like these agents, can induce the formation of peroxisomes  and inhibit the synthesis of FA in the liver (as described above). Further studies showed that FA bind to PPAR and activate the transcription of genes encoding peroxisomal enzymes . A PPAR-responsive element capable of binding to PPAR was found in the promoter of the gene of peroxisomal oxidase of acyl-CoA. Incubation of adipocytes with AA decreased this binding and caused a parallel inhibition of the expression of the gene encoding CoA-FA oxidase . These findings suggest that FA regulate the binding of PPAR to specific elements on gene promoters. In Saccharomyces cerevisiae, activation sequences of Delta9-desaturase gene was found to be responsible for activation of transcription and regulation by FA . Studies of genes encoding acyl-CoA-oxidase , cytochrome P-450 , and lipoprotein lipase  showed that PPAR binds to the PPAR-responsive element in a heterodimeric form associated with another member of the nuclear receptor family--retinoic acid X-receptor. PPAR isoforms were found in many tissues: bones, heart, skin, lungs, and kidneys [115, 119] and are subdivided into two major types: PPAR-alpha whose regulation is modulated by long-chain FA, and PPAR-gamma whose regulation is modulated by antidiabetic agents . PPAR proteins are now ascribed to a special family of nuclear receptors that regulate the synthesis of proteins responsible for the synthesis, storage, and catabolism of FA .
PPAR proteins bind polyunsaturated, monounsaturated, and saturated FA. However, it is well known that it is exactly omega3 and omega6 PUFA species that specifically inhibit the transcription of genes of lipogenic and glycolytic proteins. It is possible that PPAR and regulation by omega3/omega6-FA may represent two distinct and independent mechanisms . Certain authors also suggest that multiple mechanisms mediate the regulation of gene transcription by FA. For example, inhibition of synthesis of GLUT4 mRNA is mediated by AA (through a PPAR-dependent mechanism) and PGE2, a product of oxidative metabolism of AA . In addition, recent studies showed that FA can regulate the level of mRNA by affecting the stability of already synthesized transcripts rather than their synthesis [122, 123]. Thus, in the presence of PUFA there was a fourfold decrease in half-life of mRNA encoding Delta9-desaturase in S. cerevisiae  and mRNA encoding stearoyl-CoA-desaturase in adipocytes . It was shown that omega3-FA suppress the expression of interleukin-1beta; the initial rate of transcription of this gene is not affected but the transcription is rapidly terminated in cells stimulated by lipopolysaccharide and phorbol ester .
DNA polymerase is also sensitive to FA. Thus, experiments in vitro showed that linoleic and certain other FA inhibit alpha- and beta-DNA polymerases; the former was noncompetitively inhibited by linoleic acid, and the latter was competitively inhibited in relation to DNA and the substrate (dTTP) .
Reviewed examples show that FA display all properties of endogenous bioregulators with a broad spectrum of activity; however, these data do not represent all details of this spectrum. The difficulties in elucidation of specific bioeffector properties of FA are explained by rapid binding of free FA immediately upon their release by specific proteins, or incorporation into lipids, as well as by the fact that FA may undergo oxidative metabolism resulting in the formation of oxylipins that display similar activity spectra. Nevertheless, the data accumulated suggest that FA act as second messengers mediating the effects of many other biologically active molecules. They also modulate the activities of many functionally significant proteins. FA influence many fundamental regulatory processes of the living body, such as ion homeostasis, gene transcription, hormonal signal transduction, synthesis of various highly active lipid bioregulators, and functioning of many regulatory proteins. In the majority of cases the fine mechanisms of effects produced by FA are to be discovered. A great diversity of effects of unsaturated FA and the fact that they are found at the earliest stages of evolution suggest that these bioregulators play a system-forming role in the organism.
1.Barr, G. O., and Barr, M. M. (1930) J. Biol.
Chem., 86, 587-621.
2.Mead, J. F. (1984) J. Lipid Res., 25, 1517-1521.
3.Gill, R., and Clark, W. (1980) J. Immunol., 125, 689-695.
4.Heron, D. S., Shinitzky, M., Hershkowitz, M., and Samuel, D. (1980) Proc. Natl. Acad. Sci. USA, 77, 7463-7467.
5.Graber, R., Sumida, C., and Nunez, E. A. (1994) J. Lipid Med. Cell Sign., 9, 91-116.
6.Clarke, S. D., and Jump, D. B. (1993) Prog. Lipid Res., 32, 139-149.
7.Haourigui, M., Vallett, G., Martin, M. E., Sumida, C., Benassayag, C., and Nunez, E. A. (1994) Steroids, 59, 46-54.
8.Lister, M. D., Deems, R. A., Watanabe, Y., Ulevitch, R. J., and Dennis, E. A. (1988) J. Biol. Chem., 263, 7506-7513.
9.Raghupathi, R., and Franson, R. C. (1992) Biochim. Biophys. Acta, 1126, 206-221.
10.Irvin, R. F., Letcher, A. J., and Dawson, R. M. C. (1979) Biochem. J., 178, 497-500.
11.Siess, W., Siegel, F. L., and Lapetina, E. G. (1983) J. Biol. Chem., 258, 11236-11242.
12.Casabiell, X., Pandiella, V. M., and Casanueva, F. F. (1991) Biochem. J., 278, 679-687.
13.Hwang, S. C., Jhon, D. Y., Bae, Y. S., Kim, J. H., and Rhee, S. G. (1996) J. Biol. Chem., 271, 18342-18349.
14.Siddiqui, R. A., and Exton, J. H. (1992) Eur. J. Biochem., 210, 601-607.
15.Dai, J., Williams, S. A., Ziegelhoffer, A., and Panagia, V. (1995) Prostaglandins, Leukotrienes and Essential Fatty Acids, 52, 167-171.
16.Kanfer, J. N., McCartney, D., Singh, I. N., and Freysz, L. (1996) J. Neurochem., 67, 760-766.
17.Chung, H. C., and Flemin, N. (1992) J. Dent. Res., 71, 1462-1467.
18.Chung, H. C., and Fleming, N. (1995) Pflugers Arch. - Eur. J. Physiol., 429, 789-796.
19.Chaudhry, A., Laychock, S. G., and Rubin, R. P. (1987) J. Biol. Chem., 262, 17426-17431.
20.Nunez, D., Randon, J., Gandhi, C., Siafaka-Kapadai, A., Olson, M. S., and Hanahan, D. J. (1990) J. Biol. Chem., 265, 18330-18338.
21.Fernandez-Gallardo, S., Gijon, M. A., Garcia, M. C., Cano, E., and Sanchez Crespo, M. (1988) Biochem. J., 254, 707-714.
22.Hannun, Y. A., Obeid, L. M., and Dbaibo, G. S. (1996) in Handbook of Lipid Research (Bell, R. M., ed.) Vol. 8, Plenum Press, New York, pp. 177-204.
23.Visnjic, D., Batinic, D., and Banfic, H. (1997) Blood, 89, 81-91.
24.Jayadev, S., Linardic, C. M., and Hannun, Y. A. (1994) J. Biol. Chem., 269, 5757-5763.
25.Wolf, B. A., Turk, J., Sherman, W. R., and McDaniel, M. L. (1986) J. Biol. Chem., 261, 3501-3511.
26.Corado, J., Le Deist, F., Griscelli, C., and Fisher, A. (1990) Cell Immunol., 126, 245-254.
27.Huang, W. C., and Chueh, S. H. (1996) Brain Res., 718, 151-158.
28.Volpi, M., Yassin, R., Tao, W., Molski, T. F. P., Naccache, P. H., and Shaafi, R. I. (1984) Proc. Natl. Acad. Sci. USA, 81, 5966-5969.
29.Chow, S. C., and Jondal, M. (1990) J. Biol. Chem., 265, 902-907.
30.Astashkin, E. I., Khodorova, A. B., and Surin, A. M. (1993) FEBS Lett., 329, 72-74.
31.Petit-Jacques, J., and Hartzell, H. C. (1996) J. Physiol., 493, 67-81.
32.Chow, S. C., Ansotegui, I. J., and Jondal, M. (1990) Biochem. J., 267, 727-732.
33.Huang, J. M. C., Xian, H., and Bacaner, M. (1992) Proc. Natl. Acad. Sci. USA, 89, 6452-6456.
34.Shimada, T., and Somlyo, A. P. (1992) J. Gen. Physiol., 100, 27-44.
35.Vreugdenhil, M., Bruehl, C., Voskuyl, R. A., Kang, J. X., Leaf, A., and Wadman, W. J. (1996) Proc. Natl. Acad. Sci. USA, 93, 12559-12563.
36.Waldron, R. T., Short, A. D., Meadows, J. J., Ghosh, T. K., and Gill, D. L. (1994) J. Biol. Chem., 269, 11927-11933.
37.Graber, M. N., Alfonso, A., and Gill, D. L. (1996) J. Biol. Chem., 271, 883-888.
38.Khodorova, A. B., and Astashkin, E. I. (1994) FEBS Lett., 353, 167-170.
39.Ordway, R. W., Walsh, J. V., Jr., and Singer, J. J. (1989) Sience, 244, 1176-1179.
40.Kirber, M. T., Ordway, R. W., Clapp, L. H., Walsh, J. V., Jr., and Singer, J. J. (1992) FEBS Lett., 297, 24-28.
41.Rousaire-Dubois, B., Gerard, V., and Dubois, J. M. (1991) Pflugers Arch., 419, 467-471.
42.Macica, C. M., Yang, Y., Hebert, S. C., and Wang, W. H. (1996) Am. J. Physiol., 271, F588-F594.
43.Soliven, B., and Wang, N. (1995) Am. J. Physiol., 269, C341-C348.
44.Petrou, S., Ordway, R. W., Hamilton, J. A., Walsh, J. V., Jr., and Singer, J. J. (1994) J. Gen. Physiol., 103, 471-486.
45.Petrou, S., Ordway, R. W., Kirber, M. T., Dopico, A. M., Hamilton, J. A., Walsh, J. V., Jr., and Singer, J. J. (1995) Prostaglandins, Leukotrienes and Essential Fatty Acids, 52, 173-178.
46.Xu, X., and Lee, K. S. (1996) Am. J. Physiol., 270, H1957-H1962.
47.Ordway, R. W., Petrou, S., Kirber, M. T., Walsh, J. V., Jr., and Singer, J. J. (1995) J. Physiol., 484, 331-337.
48.Wieland, S. J., Fletcher, J. E., and Gong, Q. H. (1992) Am. J. Physiol., 263, C308-C312.
49.Wieland, S. J., Gong, Q., Poblete, H., Fletcher, J. E., Chen, L. Q., and Kallen, R. G. (1996) J. Biol. Chem., 271, 19037-19041.
50.Kang, J. X., and Leaf, A. (1996) Proc. Natl. Acad. Sci. USA, 93, 3542-3546.
51.Ikeuchi, Y., Nishizaki, T., Matsuoka, T., and Sumikawa, K. (1996) Biochem. Biophys. Res. Commun., 221, 716-721.
52.Hwang, T. C., Guggino, S. E., and Guggino, W. B. (1990) Proc. Natl. Acad. Sci. USA, 87, 5706-5709.
53.Anderson, M. P., and Welsh, M. J. (1990) Proc. Natl. Acad. Sci. USA, 87, 7334-7338.
54.Kubo, M., and Okada, Y. (1992) J. Physiol., 456, 351-371.
55.Oishi, K., Zheng, B., and Kuo, J. F. (1990) J. Biol. Chem., 265, 70-75.
56.Swarts, H. G. P., Schuurmans Stekhoven, F. M. A. H., and De Pont, J. J. H. H. M. (1990) Biochim. Biophys. Acta, 1024, 32-40.
57.Jack-Hays, M. G., Xie, Z., Wang, Y., Huang, W. H., and Askari, A. (1996) Biochim. Biophys. Acta, 1279, 43-48.
58.Murakami, S., Muramatsu, M., Araki, H., and Otomo, S. (1994) Res. Commun. Mol. Pathol. Pharmacol., 85, 57-66.
59.DuBourdieu, D. J., and Morgan, D. W. (1990) Biochim. Biophys. Acta, 1054, 326-332.
60.Abramson, S. B., Leszczynska-Piziak, J., and Weissman, G. (1991) J. Immunol., 147, 231-236.
61.Delton-Vandenbroucke, I., Sarda, N., Moliere, P., Lagarde, M., and Gharib, A. (1996) Eur. J. Pharmacol., 312, 379-384.
62.Cantrill, R. C., Patterson, P. P., Ells, G. W., and Horrobin, D. F. (1996) Cancer Lett., 100, 17-21.
63.Wallach, D., and Pastan, I. (1976) J. Biol. Chem., 251, 5802-5809.
64.Waldman, S. A., and Murad, F. (1987) Pharmacol. Rev., 39, 163-196.
65.Louis, J. C., Basset, P., Revel, M. O., Vincendon, G., and Zwiller, J. (1991) Neurochem. Int., 18, 131-135.
66.McPhail, L. S., Claiton, C. C., and Snyderman, R. (1984) Science, 224, 622-626.
67.Nakamura, S., and Nishizuka, Y. (1994) J. Biochem., 115, 1029-1034.
68.Naor, Z. (1991) Mol. Cell. Endocr., 80, C181-C186.
69.Nishizuka, Y. (1992) Science, 258, 607-614.
70.Huang, K. P., and Huang, F. L. (1993) Neurochem. Int., 22, 417-433.
71.Bandyopadhyay, G. K., Hwang, S., Imagawa, W., and Nandi, S. (1993) Prostaglandins, Leukotrienes and Essential Fatty Acids, 48, 71-78.
72.Murakami, K., Chan, S. Y., and Ruttenberg, A. (1986) J. Biol. Chem., 261, 15424-15429.
73.Nakanishi, H., and Exton, J. H. (1992) J. Biol. Chem., 267, 16347-16354.
74.Kent, J. D., Sergeant, S., Burns, D. J., and McPhail, L. C. (1996) J. Immunol., 157, 4641-4647.
75.Holian, O., and Nelson, R. (1992) Anticancer Res., 12, 975-980.
76.Martin, M. E., Benassayag, C., and Nunez, E. A. (1988) Endocrinology, 123, 1178-1186.
77.Savu, L., Benassayag, C., Vallett, G., Christeff, N., and Nunez, E. A. (1981) J. Biol. Chem., 256, 9414-9418.
78.Bouillon, R., Xiang, D. Z., Convents, R., and van Baelen, H. (1992) J. Steroid Biochem., 42, 855-861.
79.Kato, J. (1989) J. Steroid Biochem., 34, 219-227.
80.Chen, T. C., Mullen, J. P., and Meglin, N. J. (1984) J. Lipid Res., 25, 1306-1312.
81.Haourigui, M., Martin, M. E., Thobie, N., Benassayag, C., and Nunez, E. A. (1993) Endocrinology, 133, 183-191.
82.Avvakumov, G. V., Warmels-Rodenhiser, S., and Hammond, G. L. (1993) J. Biol. Chem., 268, 862-866.
83.Haourigui, M., Sakr, S., Martin, M. E., Thobie, N., Girard-Globa, A., Benassayag, C., and Nunez, E. A. (1995) Am. J. Physiol., 269, E1067-1075.
84.Sumida, C., Vallette, G., and Nunez, E. A. (1993) Acta Endocrinol., 129, 348-355.
85.Gottlicher, M., Widmark, E., Li, Q., and Gustafsson, J. A. (1992) Proc. Natl. Acad. Sci. USA, 89, 4653-4657.
86.Christeff, N., Homo-Delarche, F., Thobie, N., Durant, S., Dardenne, M., and Nunez, E. A. (1994) Prostaglandins, Leukotrienes and Essential Fatty Acids, 51, 125-131.
87.Blake, W. L., and Clarke, S. D. (1992) J. Nutr., 120, 1727-1729.
88.Clarke, S. D., Romsos, D. R., and Leveille, G. A. (1977) J. Nutr., 107, 1170-1180.
89.Clarke, S. D., Romsos, D. R., and Leveille, G. A. (1976) Lipids, 11, 485-492.
90.Katsurada, A., Iritani, N., Fukuda, H., Noguchi, T., and Tanaka, T. (1987) Eur. J. Biochem., 168, 487-492.
91.Clarke, S. D., Wilson, M. D., and Ibnoughazala, T. (1984) J. Nutr., 114, 598-605.
92.Clarke, S. D., Armstrong, M. K., and Jump, D. B. (1990) J. Nutr., 120, 225-232.
93.Clarke, S. D., and Jump, D. B. (1992) in CRC Press Reviews (Berdanier, C. D., and Hargrove, J. L., eds.) CRC Press, Boca Raton, p. 226.
94.Clarke, S. D. (1993) J. Animal Sci., 71, 1957-1965.
95.Ntambi, J. M. (1991) J. Biol. Chem., 267, 10925-10930.
96.McDonough, V. M., Stukey, J. E., and Martin, C. E. (1992) J. Biol. Chem., 267, 5931-5936.
97.Choi, J. Y., Stukey, J., Hwang, S. Y., and Martin, C. E. (1996) J. Biol. Chem., 271, 3581-3589.
98.Katsurada, A., Iritani, N., Fukuda, H., Matsumura, Y., Nishimoto, N., Noguchi, T., and Tanaka, T. (1990) Eur. J. Biochem., 190, 435-441.
99.Goodridge, A. G., Klautky, S. A., Fantozzi, D. A., Baillie, R. A., Hodnett, D. W., Chen, W., Thurmond, D. C., Xu, G., and Roncero, C. (1996) Progr. Nucleic Acids Res. Mol. Biol., 52, 89-122.
100.Tomlinson, J. E., Nakayama, R., and Holten, D. (1988) J. Nutr., 118, 808-814.
101.Sebokova, E., Klimes, I., Gasperikova, D., Bohov, P., Langer, P., Lavau, M., and Clandinin, M. T. (1996) Biochim. Biophys. Acta, 1303, 56-62.
102.Meunier-Durmort, C., Poirier, H., Niot, I., Forest, C., and Besnard, P. (1996) Biochem. J., 319, 483-487.
103.Tebbey, P. W., McGowan, K. M., Stephens, J. M., Buttke, T. M., and Pekala, P. H. (1994) J. Biol. Chem., 269, 639-644.
104.Long, S. D., and Pekala, P. H. (1996) J. Biol. Chem., 271, 1138-1144.
105.Amri, E. Z., Teboul, L., Vannier, C., Grimaldi, P. A., and Ailhaud, G. (1996) Biochem. J., 314, 541-546.
106.Kaminski, W. E., Jendraschek, E., Kiefl, R., and von Schaky, C. (1993) Blood, 81, 1871-1879.
107.Urakaze, M., Sugiyama, E., Xu, L., Auron, P., Yeh, E., and Robinson, D. (1991) Clin. Res., 23, 111-115.
108.Robinson, D. R., Urakaze, M., Huang, R., Taki, H., Sugiyama, E., Knoell, C. T., Xu, L., Yeh, E. T., and Auron, P. E. (1996) Lipids, 31, Suppl., S23-S31.
109.Kariko, K., Rosenbaum, H., Kuo, A., Zurier, R. B., and Barnathan, E. S. (1995) FEBS Lett., 361, 118-122.
110.Tomomura, M., Tomomura, A., Dewan, M. A., and Saheki, T. (1996) FEBS Lett., 399, 310-312.
111.Khair-El-Din, T., Sicher, S. C., Vazquez, M. A., Chung, G. W., Stallworth, K. A., Kitamura, K., Miller, R. T., and Lu, C. Y. (1996) J. Exp. Med., 183, 1241-1246.
112.Stuhlmeier, K. M., Tarn, C., Csizmadia, V., and Bach, F. H. (1996) Eur. J. Immunol., 26, 1417-1423.
113.Issemann, I., and Green, S. (1990) Nature, 347, 645-650.
114.Ladis, J. A., and Karathanasis, S. K. (1991) Science, 251, 561-565.
115.Schmidt, A., Endo, N., Rutledge, S. J., Vogel, R., Shinar, D., and Rodan, G. A. (1992) Mol. Endocrin., 6, 1634-1641.
116.Lock, A. B., Mitchell, A. M., and Elcombe, C. R. (1989) Ann. Rev. Pharmacol. Toxicol., 29, 145-163.
117.Issemann, I., Prince, R. A., Tugwood, J. D., and Green, S. (1993) J. Mol. Endocrin., 11, 37-47.
118.Johnson, E. F., Palmer, C. N., Griffin, K. J., and Hsu, M. H. (1996) FASEB J., 10, 1241-1248.
119.Schoonjans, K., Peinado-Onsurbe, J., Lefebvre, A. M., Heyman, R. A., Briggs, M., Deeb, S., Staels, B., and Auwerx, J. (1996) EMBO J., 15, 5336-5348.
120.Forman, B. M., Chen, J., and Evans, R. M. (1996) Ann. New York Acad. Sci., 804, 266-275.
121.Clarke, S. D., and Jump, D. B. (1996) Lipids, 31, Suppl., S7-S11.
122.Gonzalez, C. I., and Martin, C. E. (1996) J. Biol. Chem., 271, 25801-25809.
123.Sessler, A. M., Kaur, N., Palta, J. P., and Ntambi, J. M. (1996) J. Biol. Chem., 271, 29854-29858.
124.Mizushina, Y., Tanaka, N., Yagi, H., Kurosawa, T., Onoue, M., Seto, H., Horie, T., Aoyagi, N., Yamaoka, M., Matsukage, A., Yoshida, S., and Sakaguchi, K. (1996) Biochim. Biophys. Acta, 1308, 256-262.