- Email: [email protected]
An oxidized derivative of linoleic acid affects aldosterone secretion by adrenal cells in vitro
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3),163^167 Published by Elsevier Science Ltd. doi:10.1054/plef.414, available online at http://www.idealibrary.com on
An oxidized derivative of linoleic acid affects aldosterone secretion by adrenal cells in vitro T. L. Goodfriend,1,2 D. L. Ball,1,2 H.W. Gardner3 1
Research Service,Veterans Hospital, Madison,WI, USA Departments of Medicine and Pharmacology, Medical School, University of Wisconsin, Madison,WI, USA 3 National Center for Agricultural Utilization Research, USDA, Peoria, IL, USA 2
Summary Based on the clinical observation that humans with visceral adiposity have higher plasma aldosterone levels than controls, we postulated that endogenous fatty acids can be oxidized by the liver to form stimuli of the adrenal cortex. Although we could show that hepatocytesproduced adrenalstimulifromlinoleic acidinvitro, the yieldwasvery small.To facilitate the elucidation of chemical structures, we incubated a large amount of linoleic acid with lipoxygenase, then treated the hydroperoxide with cysteine and iron.The major product of this process was12,13-epoxy-9-keto-10-trans-octadecenoic acid.This epoxy-keto compound stimulated aldosterone production at concentrations from 0.5 to15 mm. At higher concentrations, it was inhibitory.The epoxy-keto-octadecenoic acid exhibited the chromatographic characteristics of one product of the incubation of linoleic acid with hepatocytes.The results are consistent with the postulated conversion of linoleic acid to stimuli of aldosterone production.This may be a mechanistic link between visceral obesity and hypertension in humans. Published by Elsevier Science Ltd.
INTRODUCTION Aldosterone is the most potent mineralocorticoid secreted by the adrenal cortex. When administered to an animal or human ingesting a diet replete in sodium chloride, aldosterone can cause sodium retention and potassium excretion by the kidneys, expansion of the extracellular fluid volume, and increased arterial blood pressure. Because of these properties, excessive production of aldosterone has been implicated as a cause of some types of human hypertension.1 Aldosterone secretion is normally regulated by several interacting stimuli and inhibitors, including extracellular potassium, and the hormones angiotensin, adrenocorticotropic hormone
Received 22 November 2001 Accepted 3 May 2002 Correspondence to: T. L. Goodfriend, Research Service,Veterans Hospital, 2500 OverlookTerrace, Madison,WI 53705, USA.Tel.: +1-608-280-7007; Fax: +1-608-280-7244; E-mail: [email protected] Supported by grants from: The American Heart Association; Knoll Pharmaceuticals;The University of Wisconsin Graduate School, Medical School, and Department of Medicine; with additional support from the US Department of Veterans Affairs.
Published by Elsevier Science Ltd.
(ACTH), and atrial natriuretic peptide (ANP).2 Adrenal cells producing aldosterone in vitro can also be stimulated or inhibited by a wide variety of other substances, including serotonin, catecholamines, endothelin, antidiuretic hormone, melanocortin, somatostatin, adrenomedullin, calcium, and non-esterified fatty acids.3 It is unclear whether any of these known, or as-yet-unknown, substances exert important influences on aldosterone secretion in humans with normal or high blood pressure, but there are well-described clinical situations in which plasma levels of aldosterone cannot be explained by the classical regulators.4 While investigating the effects of non-esterified fatty acids on aldosterone production by rat adrenal cells, we found that addition of hepatocytes to the incubation altered the results. Some fatty acids that inhibited aldosterone production when added to adrenal cells alone stimulated aldosterone production when hepatocytes were also present. This stimulatory effect could be mimicked by the supernatant of an incubation containing only hepatocytes and fatty acid.5 The fatty acid we chose to explore most thoroughly was linoleic acid. The hepatocyte/linoleic acid supernatant contained a complex mixture of compounds, all in small amounts, which presented a daunting challenge to identification of the products.
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3), 163^167
164
Goodfriend et al.
Because production of stimuli from fatty acids by hepatocytes was dependent on the presence of oxygen, we postulated that the stimulatory substances were the products of fatty acid oxidation. To accumulate sufficient quantities of stimulatory products to enable chemical characterization, we subjected bulk quantities of linoleic acid to oxidation in a cell-free system. Purification of the reaction mixture yielded individual oxidation products with the same biological and chromatographic properties as the products of hepatocyte action. We report on the structure and adrenal effects of one of these oxidation products. The possible relevance of our observations to hypertension, obesity, and nutrition is discussed. MATERIALS AND METHODS Linoleic acid and soybean lipoxidase Type V (lipoxygenase) were purchased from Sigma, St. Louis, MO. Other chemicals were reagent grade. Antibody to aldosterone was purchased from ICN, Costa Mesa, CA, radioiodinated aldosterone from Diagnostic Products Corp, Los Angeles, CA, and 14C-carboxy-labeled linoleic acid from New England Nuclear, Boston, MA. Fatty acid was oxidized in a two-step process based on a method described earlier.6 In the first step, 800 mg of linoleic acid were dissolved in 70 ml of water with the aid of 0.4 g of Tween 20 (Sigma) and potassium hydroxide. The volume was made up to 500 ml with 5 mM potassium borate (pH 10) to which was added 106 units of lipoxygenase. The mixture was chilled in ice and aerated with oxygen for 45 min. Assuming a molar extinction coefficient of 26 770 at 234 nm, we obtained approximately 700 mg of hydroperoxy-octadecadienoic acid (HPODE). Based on the specificity of the lipoxygenase, we assumed that the major part of the product was the 13S-HPODE stereoisomer. The first reaction was terminated by acidifying the reaction mixture to pH 4.0 with citric acid (1 N), and extracting the lipid into chloroform/methanol (2:1). The organic layer was washed twice with water and dried. The HPODE was partially purified by column chromatography through silicic acid, pH 4 (Mallinckrodt SilicAR-CC4). Acetone/hexane mixtures starting at an acetone concentration of 5% were used to elute the HPODE. Samples of eluate were spotted on a thin layer of fluorescent silica (F254). HPODE quenched ultraviolet fluorescence. It eluted with the second (7.5%) strength of acetone in hexane. The second step of the reaction sequence was accomplished by dissolving approximately 250 mg of 13SHPODE from the first reaction in 125 ml acetone, adding an equal volume of water, 390 mg of cysteine and 8 mmol of FeCl3, then gassing vigorously with oxygen at room temperature for 45 minutes. The organic compounds
from the iron–cysteine–HPODE reaction were extracted into chloroform and the residual aqueous layer was reextracted with chloroform/methanol (2:1). The two chloroform extracts were pooled and fractionated by an open column containing 50 g of Mallinkrodt SilicAR-CC4 eluted with eight successive mixtures of acetone in hexane ranging from 7.5 to 50% acetone and a final stripping with pure methanol. As each solvent eluted, the effluent was monitored by thin layer chromatography (TLC), and by radioactivity if labeled fatty acid had been added to the first reaction mixture. The pattern of eluted radioactivity suggested that ten separate compounds or groups of similar compounds were identifiable, and these were labeled A through J. Fractions A through J from the second silicic acid column were tested for biological activity in rat adrenal cells (see below), and those with activity were subjected to further purification by reversed-phase high-pressure liquid chromatography (HPLC). The column was a Novapak C18 radial compression cartridge (Waters Corp, Milford, MA). The eluting solvent was a linear gradient starting with 65% methanol in water and ending with 75% methanol in water, both containing 0.1% acetic acid, pumped at a rate of 2 ml/min. Fractions from the columns were tested for biological activity in rat adrenal glomerulosa cells. Capsules of adrenal glands from adult Sprague–Dawley rats (Harlan, Madison, WI) were incubated with collagenase as described,7 and the suspended cells counted in a hemocytometer. By inspection, there were no more than 4% fasciculata cells among the glomerulosa cells. Cells were diluted so that each assay tube contained 40 000 cells in 0.2 ml of medium. The medium was a mixture of 10% Medium 199 (Sigma 5017) in a HEPES-buffered balanced salt solution. The incubation medium contained the following ingredients in addition to the amino acids and vitamins in Medium 199: NaCl 116 mM; KCl 3.6 mM; MgSO4 1 mM; NaHCO3 2.6 mM; NaHPO4 0.69 mM; HEPES 6.5 mM; and glucose 7.4 mM. The pH was 7.4 during gassing with a mixture of 95% O2 and 5% CO2. Incubations were for 120 min at 371C in 95% O2 and 5% CO2. Aldosterone production was measured by radioimmunoassay of supernatants. Fractions from HPLC that altered aldosterone production by rat adrenal cells were purified by a second HPLC fractionation, then analyzed by gas chromatography/ mass spectrometry (GC/MS) and by nuclear magnetic resonance spectrometry (NMR). RESULTS The two-step oxidation of linoleic acid by lipoxygenase followed by treatment with iron–cysteine generated several substances that affected aldosterone production
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3), 163^167
Published by Elsevier Science Ltd.
Effect of linoleic acid on aldosterone secretion 165
Fig. 1 Fractionation of oxidized products of linoleic acid. Linoleic acid was oxidized by lipoxygenase, followed by treatment with cysteine and ferric chloride.The organic products were fractionated first by silicic acid chromatography.The figure shows the results of a second fractionation, by HPLC, of a part of the eluate from silicic acid. A C18 column in the reversed phase was eluted with a linear gradient of methanol in water containing acetic acid.The fine line shows absorbance of the eluate at 234 nm.The open diamonds connected by the dark line show the results of bioassays of each fraction in rat adrenal zona glomerulosa cells producing aldosterone.The two closed diamonds represent the bioassay results of fractions 21and 22 diluted ten-fold.
by rat adrenal cells. Of the ten distinct fractions eluted from silicic acid, fractions B–D showed the greatest stimulatory activity when diluted to 1:1 000 in the bioassay buffer. To illustrate the number of active compounds produced by oxidation of linoleic acid, these three fractions were pooled and applied to an HPLC column in the reversed phase. Figure 1 shows a profile of absorbance at 234 nm, and biological activity in the fractions from this pool. The largest amount of ultraviolet-absorbing material eluted in fractions 21 and 22, when the eluting solvent contained 68% methanol in water, a polarity index of 7.36. At the initial dilution, these two fractions showed inhibitory activity in bioassays, but when diluted further, they were highly stimulatory. This biphasic property is shown more clearly in Figure 2, depicting an experiment in which the purified material from HPLC fraction 22 was tested in serial dilutions. Mass spectrometry was used to elucidate the structure of the purified material from fraction 22. It showed a mass spectrum consistent with a linoleic derivative containing an epoxide across the 12,13 bond, a ketone at position 9, and a double bond between carbons 10 and 11. The epoxide and double bond were assigned trans configurations.8 This structure is depicted in Figure 3. Nuclear magnetic resonance spectrometry confirmed the structure and stereoisomerism deduced from mass spectrometry. The chromatographic characteristics of the epoxy-keto derivative were similar to those of one adrenal stimulus extracted into chloroform/methanol from the medium Published by Elsevier Science Ltd.
Fig. 2 Bioassays of serial dilutions of one oxidized derivative of linoleic acid. Fraction 22 from the HPLC depicted in Figure1was re-purified and tested for effects on aldosterone production by rat adrenalglomerulosa cells as describedin the text.The chemicalname was assigned to this material after GC/MS and NMR analysis.The quantity of lipid was determined by optical density at 234 nm, and by the specific activity of the compound when labeled linoleic acid was present during the oxidation reactions.
after incubating rat hepatocytes with linoleic acid (data not shown). DISCUSSION Oxygen-containing derivatives of arachidonic acid, the eicosanoids, display a wide range of biological activities, and have been the objects of intensive research for
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3), 163^167
166
Goodfriend et al.
Fig. 3 Chemical structure assigned to the major oxidation product of linoleic acid. Fraction 22 from the purification depicted in Figure1 was re-purified by HPLC, then subjected to the GC/MS and to NMR spectrometry.The structure shown was deduced from those analyses.The epoxide and double bond are both in the trans configuration.
decades. Less is known about analogous compounds derived from other unsaturated fatty acids. We believe that the compound we describe in this report has never before been studied in mammalian systems. Among oxidized linoleic acid derivatives that have been studied in animals, the two that are most similar to our active product are epoxides, but lack other oxygens aside from the carboxyl group. The first reported in mammals is 9,10-epoxy-12-octadecenoic acid, called ‘leukotoxin’ or ‘leukotoxin A’ to reflect its production by suspensions of peripheral blood leukocytes and its toxic effects on blood vessels.9 A similar compound, with the epoxide in the 12,13 position, is called isoleukotoxin or leukotoxin B. Leukotoxins relax vascular smooth muscle from pulmonary arteries, injure the pulmonary vascular bed, and cause pulmonary edema in animals.10 Leukotoxins constrict carotid arteries and reduce the strength of contraction of myocardium from cats.11 Some of these effects may be mediated by nitric oxide released in response to the lipids.12 It is probable that the reported effects of leukotoxins are not mediated by the epoxides themselves, but by their hydrolysis products, the corresponding diols.13 There are no reports of the effects of these compounds on aldosterone production, sodium retention, or blood pressure. We generated a mixture of linoleic acid oxidation products and screened them for effects on aldosterone production by adrenal cells. We found biological activity in the compound that was produced in the greatest amount by this reaction sequence, 12,13-epoxy-9-keto10-octadecenoic acid. This compound stimulated aldosterone production at concentrations between 0.5 and 15 mm. Above 15 mm, the epoxy-ketone exerted a progressively larger inhibitory effect until it reduced aldosterone production to levels below basal at 50 mm. Our study of linoleic acid oxidation products was stimulated by a clinical observation we made while studying obesity and hypertension. We found that plasma aldosterone levels were higher in subjects with visceral obesity (also called ‘abdominal’, ‘central’, or ‘upper body’ obesity) than in lean subjects or those with ‘lower body’ obesity.5 We postulated that the increased aldosterone results from the release of fatty acids from visceral adipocytes into the portal venous system and the liver.
We guessed that this pathway favors oxidative metabolism of adipocyte fatty acids by the liver, yielding stimuli of aldosterone production. In support of our hypothesis, we found that rat hepatocytes produced stimuli of aldosterone production when incubated with polyunsaturated fatty acids in the presence of oxygen. We were unable to characterize the small amounts of stimuli produced by rat hepatocytes, so we explored the products of large-scale oxidation of linoleic acid catalyzed by lipoxygenase and iron/cysteine. Although we were able to identify adrenal stimuli in this oxidation mixture, we do not know if these compounds are formed in the human liver. Similar compounds have been generated from linoleic acid by addition of heme compounds in vitro, but they contained hydroxyl groups where our compound contained a ketone.14,15 Although the observation that an oxidized derivative of linoleic acid can stimulate aldosterone production by suspensions of adrenal cells is consistent with our initial hypothesis, it is far from a proof. We remain uncertain about the relevance of the compound we characterized to the association between visceral obesity and elevated aldosterone and hypertension in humans. Our previous work on fatty acids and the adrenal zona glomerulosa emphasized the inhibitory properties of unsaturated fatty acids in their non-esterified form.16 Together with the results presented in this report, our experiments show that native and oxidized polyunsaturated fatty acids exert reciprocal effects on aldosterone production. This will make it difficult to derive simple rules for the effects of polyunsaturated fatty acids on aldosterone and blood pressure. Another possible implication of our work is the effect that oxidized unsaturated fatty acids in the diet might have on aldosterone production and blood pressure. For example, linoleic acid in foods might be oxidized during preparation or storage and enter the systemic circulation from the gut. Alternatively, unmodified fatty acids might be absorbed and reach the liver via the portal vein, where they could be oxidized, just as we propose for fatty acids from visceral fat depots. Whatever their origin, oxidized products of linoleic acid present potential lipid regulators of steroid secretion and arterial blood pressure in humans.
ACKNOWLEDGEMENTS The authors are indebted to Prof. Gerhard Spiteller, Bayreuth, Germany, for many helpful discussions and analyses. We also benefited from experimental suggestions by Dr Kuni Takayama, University of Wisconsin. Nuclear magnetic resonance spectroscopy was performed by Dr Mark Anderson at the National NMR Facility at Madison.
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3), 163^167
Published by Elsevier Science Ltd.
Effect of linoleic acid on aldosterone secretion 167
REFERENCES 1. Burstyn P. G., Horrobin D. F. Possible mechanism of action for aldosterone-induced hypertension. Lancet 1970; 1: 973–976. 2. Quinn S. J., Williams G. H. Regulation of aldosterone secretion. Ann Rev Physiol 1988; 50: 409–426. 3. Nussdorfer G. G. Paracrine control of adrenal cortical function by medullary chromaffin cells. Pharmacol Rev 1996; 48: 495–530. 4. Brown R. D., Wisgerhof M., Carpenter P. C. et al. Adrenal sensitivity to angiotensin II and undiscovered aldosterone stimulating factors in hypertension. J Steroid Biochem 1979; 11: 1043–1050. 5. Goodfriend T. L., Egan B. M., Kelley D. E. Aldosterone in obesity. Endocr Res 1998; 24: 789–796. 6. Gardner H. W., Jursinic P. A. Degradation of linoleic acid hydroperoxides by a cysteine FeCl3 catalyst as a model for similar biochemical reactions. Biochim Biophys Acta 1981; 665: 100–133. 7. Goodfriend T. L., Ball D. L., Elliott M. E., Shackleton C. Lead increases aldosterone production by rat adrenal cells. Hypertension 1995; 25: 785–789. 8. Gardner H. W. Analysis of plant lipoxygenase metabolites. In: Christie W. W., ed. Advances in Lipid MethodologyFfour. Dundee: The Oily Press, 1997; 1–43. 9. Ozawa T., Hayakawa M., Takamura T. et al. Biosynthesis of leukotoxin, 9,10-epoxy-12 octadecenoate, by leukocytes in lung
Published by Elsevier Science Ltd.
10.
11.
12.
13.
14.
15.
16.
lavages of rat after exposure to hyperoxia. Biochem Biophys Res Commun 1986; 134: 1071–1078. Jia-ning H., Taki F., Sugiyama S. et al. Neutrophil-derived epoxide, 9,10-epoxy-12-octadecenoate, induces pulmonary edema. Lung 1988; 166: 327–337. Siegfried M. R., Aoki N., Lefer A. M., Elisseou E. M., Zipkin R. E. Direct cardiovascular actions of two metabolites of linoleic acid. Life Sci 1990; 46: 427–433. Ishizaki T., Shigemori K., Tamamura Y. et al. Increased nitric oxide biosynthesis in leukotoxin, 9,10-epoxy-12-octadecenoate injured lung. Biochem Biophys Res Commun 1995; 210: 133–137. Moghaddam M. F., Grant D. F., Cheek J. M., Greene J. F., Williamson K. C., Hammock B. D. Bioactivation of leukotoxins to their toxic diols by epoxide hydrolase. Nat Med 1997; 3: 562–566. Hamberg M. Decomposition of unsaturated fatty acid hydroperoxides by hemoglobin: structures of major products of 13L-hydroperoxy-9,11-octadecadienoic acid. Lipids 1975; 10: 87–92. Dix T. A., Marnett L. J. Conversion of linoleic acid hydroperoxide to hydroxy, keto, epoxyhydroxy, and trihydroxy fatty acids by hematin. J Biol Chem 1985; 260: 5351–5357. Goodfriend T. L., Ball D. L., Elliott M. E. et al. Fatty acids may regulate aldosterone secretion and mediate some of insulin’s effects on blood pressure. Prostaglandins Leukot Essent Fatty Acids 1993; 48: 43–50.
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3), 163^167
An oxidized derivative of linoleic acid affects aldosterone secretion by adrenal cells in vitro T. L. Goodfriend,1,2 D. L. Ball,1,2 H.W. Gardner3 1
Research Service,Veterans Hospital, Madison,WI, USA Departments of Medicine and Pharmacology, Medical School, University of Wisconsin, Madison,WI, USA 3 National Center for Agricultural Utilization Research, USDA, Peoria, IL, USA 2
Summary Based on the clinical observation that humans with visceral adiposity have higher plasma aldosterone levels than controls, we postulated that endogenous fatty acids can be oxidized by the liver to form stimuli of the adrenal cortex. Although we could show that hepatocytesproduced adrenalstimulifromlinoleic acidinvitro, the yieldwasvery small.To facilitate the elucidation of chemical structures, we incubated a large amount of linoleic acid with lipoxygenase, then treated the hydroperoxide with cysteine and iron.The major product of this process was12,13-epoxy-9-keto-10-trans-octadecenoic acid.This epoxy-keto compound stimulated aldosterone production at concentrations from 0.5 to15 mm. At higher concentrations, it was inhibitory.The epoxy-keto-octadecenoic acid exhibited the chromatographic characteristics of one product of the incubation of linoleic acid with hepatocytes.The results are consistent with the postulated conversion of linoleic acid to stimuli of aldosterone production.This may be a mechanistic link between visceral obesity and hypertension in humans. Published by Elsevier Science Ltd.
INTRODUCTION Aldosterone is the most potent mineralocorticoid secreted by the adrenal cortex. When administered to an animal or human ingesting a diet replete in sodium chloride, aldosterone can cause sodium retention and potassium excretion by the kidneys, expansion of the extracellular fluid volume, and increased arterial blood pressure. Because of these properties, excessive production of aldosterone has been implicated as a cause of some types of human hypertension.1 Aldosterone secretion is normally regulated by several interacting stimuli and inhibitors, including extracellular potassium, and the hormones angiotensin, adrenocorticotropic hormone
Received 22 November 2001 Accepted 3 May 2002 Correspondence to: T. L. Goodfriend, Research Service,Veterans Hospital, 2500 OverlookTerrace, Madison,WI 53705, USA.Tel.: +1-608-280-7007; Fax: +1-608-280-7244; E-mail: [email protected] Supported by grants from: The American Heart Association; Knoll Pharmaceuticals;The University of Wisconsin Graduate School, Medical School, and Department of Medicine; with additional support from the US Department of Veterans Affairs.
Published by Elsevier Science Ltd.
(ACTH), and atrial natriuretic peptide (ANP).2 Adrenal cells producing aldosterone in vitro can also be stimulated or inhibited by a wide variety of other substances, including serotonin, catecholamines, endothelin, antidiuretic hormone, melanocortin, somatostatin, adrenomedullin, calcium, and non-esterified fatty acids.3 It is unclear whether any of these known, or as-yet-unknown, substances exert important influences on aldosterone secretion in humans with normal or high blood pressure, but there are well-described clinical situations in which plasma levels of aldosterone cannot be explained by the classical regulators.4 While investigating the effects of non-esterified fatty acids on aldosterone production by rat adrenal cells, we found that addition of hepatocytes to the incubation altered the results. Some fatty acids that inhibited aldosterone production when added to adrenal cells alone stimulated aldosterone production when hepatocytes were also present. This stimulatory effect could be mimicked by the supernatant of an incubation containing only hepatocytes and fatty acid.5 The fatty acid we chose to explore most thoroughly was linoleic acid. The hepatocyte/linoleic acid supernatant contained a complex mixture of compounds, all in small amounts, which presented a daunting challenge to identification of the products.
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3), 163^167
164
Goodfriend et al.
Because production of stimuli from fatty acids by hepatocytes was dependent on the presence of oxygen, we postulated that the stimulatory substances were the products of fatty acid oxidation. To accumulate sufficient quantities of stimulatory products to enable chemical characterization, we subjected bulk quantities of linoleic acid to oxidation in a cell-free system. Purification of the reaction mixture yielded individual oxidation products with the same biological and chromatographic properties as the products of hepatocyte action. We report on the structure and adrenal effects of one of these oxidation products. The possible relevance of our observations to hypertension, obesity, and nutrition is discussed. MATERIALS AND METHODS Linoleic acid and soybean lipoxidase Type V (lipoxygenase) were purchased from Sigma, St. Louis, MO. Other chemicals were reagent grade. Antibody to aldosterone was purchased from ICN, Costa Mesa, CA, radioiodinated aldosterone from Diagnostic Products Corp, Los Angeles, CA, and 14C-carboxy-labeled linoleic acid from New England Nuclear, Boston, MA. Fatty acid was oxidized in a two-step process based on a method described earlier.6 In the first step, 800 mg of linoleic acid were dissolved in 70 ml of water with the aid of 0.4 g of Tween 20 (Sigma) and potassium hydroxide. The volume was made up to 500 ml with 5 mM potassium borate (pH 10) to which was added 106 units of lipoxygenase. The mixture was chilled in ice and aerated with oxygen for 45 min. Assuming a molar extinction coefficient of 26 770 at 234 nm, we obtained approximately 700 mg of hydroperoxy-octadecadienoic acid (HPODE). Based on the specificity of the lipoxygenase, we assumed that the major part of the product was the 13S-HPODE stereoisomer. The first reaction was terminated by acidifying the reaction mixture to pH 4.0 with citric acid (1 N), and extracting the lipid into chloroform/methanol (2:1). The organic layer was washed twice with water and dried. The HPODE was partially purified by column chromatography through silicic acid, pH 4 (Mallinckrodt SilicAR-CC4). Acetone/hexane mixtures starting at an acetone concentration of 5% were used to elute the HPODE. Samples of eluate were spotted on a thin layer of fluorescent silica (F254). HPODE quenched ultraviolet fluorescence. It eluted with the second (7.5%) strength of acetone in hexane. The second step of the reaction sequence was accomplished by dissolving approximately 250 mg of 13SHPODE from the first reaction in 125 ml acetone, adding an equal volume of water, 390 mg of cysteine and 8 mmol of FeCl3, then gassing vigorously with oxygen at room temperature for 45 minutes. The organic compounds
from the iron–cysteine–HPODE reaction were extracted into chloroform and the residual aqueous layer was reextracted with chloroform/methanol (2:1). The two chloroform extracts were pooled and fractionated by an open column containing 50 g of Mallinkrodt SilicAR-CC4 eluted with eight successive mixtures of acetone in hexane ranging from 7.5 to 50% acetone and a final stripping with pure methanol. As each solvent eluted, the effluent was monitored by thin layer chromatography (TLC), and by radioactivity if labeled fatty acid had been added to the first reaction mixture. The pattern of eluted radioactivity suggested that ten separate compounds or groups of similar compounds were identifiable, and these were labeled A through J. Fractions A through J from the second silicic acid column were tested for biological activity in rat adrenal cells (see below), and those with activity were subjected to further purification by reversed-phase high-pressure liquid chromatography (HPLC). The column was a Novapak C18 radial compression cartridge (Waters Corp, Milford, MA). The eluting solvent was a linear gradient starting with 65% methanol in water and ending with 75% methanol in water, both containing 0.1% acetic acid, pumped at a rate of 2 ml/min. Fractions from the columns were tested for biological activity in rat adrenal glomerulosa cells. Capsules of adrenal glands from adult Sprague–Dawley rats (Harlan, Madison, WI) were incubated with collagenase as described,7 and the suspended cells counted in a hemocytometer. By inspection, there were no more than 4% fasciculata cells among the glomerulosa cells. Cells were diluted so that each assay tube contained 40 000 cells in 0.2 ml of medium. The medium was a mixture of 10% Medium 199 (Sigma 5017) in a HEPES-buffered balanced salt solution. The incubation medium contained the following ingredients in addition to the amino acids and vitamins in Medium 199: NaCl 116 mM; KCl 3.6 mM; MgSO4 1 mM; NaHCO3 2.6 mM; NaHPO4 0.69 mM; HEPES 6.5 mM; and glucose 7.4 mM. The pH was 7.4 during gassing with a mixture of 95% O2 and 5% CO2. Incubations were for 120 min at 371C in 95% O2 and 5% CO2. Aldosterone production was measured by radioimmunoassay of supernatants. Fractions from HPLC that altered aldosterone production by rat adrenal cells were purified by a second HPLC fractionation, then analyzed by gas chromatography/ mass spectrometry (GC/MS) and by nuclear magnetic resonance spectrometry (NMR). RESULTS The two-step oxidation of linoleic acid by lipoxygenase followed by treatment with iron–cysteine generated several substances that affected aldosterone production
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3), 163^167
Published by Elsevier Science Ltd.
Effect of linoleic acid on aldosterone secretion 165
Fig. 1 Fractionation of oxidized products of linoleic acid. Linoleic acid was oxidized by lipoxygenase, followed by treatment with cysteine and ferric chloride.The organic products were fractionated first by silicic acid chromatography.The figure shows the results of a second fractionation, by HPLC, of a part of the eluate from silicic acid. A C18 column in the reversed phase was eluted with a linear gradient of methanol in water containing acetic acid.The fine line shows absorbance of the eluate at 234 nm.The open diamonds connected by the dark line show the results of bioassays of each fraction in rat adrenal zona glomerulosa cells producing aldosterone.The two closed diamonds represent the bioassay results of fractions 21and 22 diluted ten-fold.
by rat adrenal cells. Of the ten distinct fractions eluted from silicic acid, fractions B–D showed the greatest stimulatory activity when diluted to 1:1 000 in the bioassay buffer. To illustrate the number of active compounds produced by oxidation of linoleic acid, these three fractions were pooled and applied to an HPLC column in the reversed phase. Figure 1 shows a profile of absorbance at 234 nm, and biological activity in the fractions from this pool. The largest amount of ultraviolet-absorbing material eluted in fractions 21 and 22, when the eluting solvent contained 68% methanol in water, a polarity index of 7.36. At the initial dilution, these two fractions showed inhibitory activity in bioassays, but when diluted further, they were highly stimulatory. This biphasic property is shown more clearly in Figure 2, depicting an experiment in which the purified material from HPLC fraction 22 was tested in serial dilutions. Mass spectrometry was used to elucidate the structure of the purified material from fraction 22. It showed a mass spectrum consistent with a linoleic derivative containing an epoxide across the 12,13 bond, a ketone at position 9, and a double bond between carbons 10 and 11. The epoxide and double bond were assigned trans configurations.8 This structure is depicted in Figure 3. Nuclear magnetic resonance spectrometry confirmed the structure and stereoisomerism deduced from mass spectrometry. The chromatographic characteristics of the epoxy-keto derivative were similar to those of one adrenal stimulus extracted into chloroform/methanol from the medium Published by Elsevier Science Ltd.
Fig. 2 Bioassays of serial dilutions of one oxidized derivative of linoleic acid. Fraction 22 from the HPLC depicted in Figure1was re-purified and tested for effects on aldosterone production by rat adrenalglomerulosa cells as describedin the text.The chemicalname was assigned to this material after GC/MS and NMR analysis.The quantity of lipid was determined by optical density at 234 nm, and by the specific activity of the compound when labeled linoleic acid was present during the oxidation reactions.
after incubating rat hepatocytes with linoleic acid (data not shown). DISCUSSION Oxygen-containing derivatives of arachidonic acid, the eicosanoids, display a wide range of biological activities, and have been the objects of intensive research for
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3), 163^167
166
Goodfriend et al.
Fig. 3 Chemical structure assigned to the major oxidation product of linoleic acid. Fraction 22 from the purification depicted in Figure1 was re-purified by HPLC, then subjected to the GC/MS and to NMR spectrometry.The structure shown was deduced from those analyses.The epoxide and double bond are both in the trans configuration.
decades. Less is known about analogous compounds derived from other unsaturated fatty acids. We believe that the compound we describe in this report has never before been studied in mammalian systems. Among oxidized linoleic acid derivatives that have been studied in animals, the two that are most similar to our active product are epoxides, but lack other oxygens aside from the carboxyl group. The first reported in mammals is 9,10-epoxy-12-octadecenoic acid, called ‘leukotoxin’ or ‘leukotoxin A’ to reflect its production by suspensions of peripheral blood leukocytes and its toxic effects on blood vessels.9 A similar compound, with the epoxide in the 12,13 position, is called isoleukotoxin or leukotoxin B. Leukotoxins relax vascular smooth muscle from pulmonary arteries, injure the pulmonary vascular bed, and cause pulmonary edema in animals.10 Leukotoxins constrict carotid arteries and reduce the strength of contraction of myocardium from cats.11 Some of these effects may be mediated by nitric oxide released in response to the lipids.12 It is probable that the reported effects of leukotoxins are not mediated by the epoxides themselves, but by their hydrolysis products, the corresponding diols.13 There are no reports of the effects of these compounds on aldosterone production, sodium retention, or blood pressure. We generated a mixture of linoleic acid oxidation products and screened them for effects on aldosterone production by adrenal cells. We found biological activity in the compound that was produced in the greatest amount by this reaction sequence, 12,13-epoxy-9-keto10-octadecenoic acid. This compound stimulated aldosterone production at concentrations between 0.5 and 15 mm. Above 15 mm, the epoxy-ketone exerted a progressively larger inhibitory effect until it reduced aldosterone production to levels below basal at 50 mm. Our study of linoleic acid oxidation products was stimulated by a clinical observation we made while studying obesity and hypertension. We found that plasma aldosterone levels were higher in subjects with visceral obesity (also called ‘abdominal’, ‘central’, or ‘upper body’ obesity) than in lean subjects or those with ‘lower body’ obesity.5 We postulated that the increased aldosterone results from the release of fatty acids from visceral adipocytes into the portal venous system and the liver.
We guessed that this pathway favors oxidative metabolism of adipocyte fatty acids by the liver, yielding stimuli of aldosterone production. In support of our hypothesis, we found that rat hepatocytes produced stimuli of aldosterone production when incubated with polyunsaturated fatty acids in the presence of oxygen. We were unable to characterize the small amounts of stimuli produced by rat hepatocytes, so we explored the products of large-scale oxidation of linoleic acid catalyzed by lipoxygenase and iron/cysteine. Although we were able to identify adrenal stimuli in this oxidation mixture, we do not know if these compounds are formed in the human liver. Similar compounds have been generated from linoleic acid by addition of heme compounds in vitro, but they contained hydroxyl groups where our compound contained a ketone.14,15 Although the observation that an oxidized derivative of linoleic acid can stimulate aldosterone production by suspensions of adrenal cells is consistent with our initial hypothesis, it is far from a proof. We remain uncertain about the relevance of the compound we characterized to the association between visceral obesity and elevated aldosterone and hypertension in humans. Our previous work on fatty acids and the adrenal zona glomerulosa emphasized the inhibitory properties of unsaturated fatty acids in their non-esterified form.16 Together with the results presented in this report, our experiments show that native and oxidized polyunsaturated fatty acids exert reciprocal effects on aldosterone production. This will make it difficult to derive simple rules for the effects of polyunsaturated fatty acids on aldosterone and blood pressure. Another possible implication of our work is the effect that oxidized unsaturated fatty acids in the diet might have on aldosterone production and blood pressure. For example, linoleic acid in foods might be oxidized during preparation or storage and enter the systemic circulation from the gut. Alternatively, unmodified fatty acids might be absorbed and reach the liver via the portal vein, where they could be oxidized, just as we propose for fatty acids from visceral fat depots. Whatever their origin, oxidized products of linoleic acid present potential lipid regulators of steroid secretion and arterial blood pressure in humans.
ACKNOWLEDGEMENTS The authors are indebted to Prof. Gerhard Spiteller, Bayreuth, Germany, for many helpful discussions and analyses. We also benefited from experimental suggestions by Dr Kuni Takayama, University of Wisconsin. Nuclear magnetic resonance spectroscopy was performed by Dr Mark Anderson at the National NMR Facility at Madison.
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3), 163^167
Published by Elsevier Science Ltd.
Effect of linoleic acid on aldosterone secretion 167
REFERENCES 1. Burstyn P. G., Horrobin D. F. Possible mechanism of action for aldosterone-induced hypertension. Lancet 1970; 1: 973–976. 2. Quinn S. J., Williams G. H. Regulation of aldosterone secretion. Ann Rev Physiol 1988; 50: 409–426. 3. Nussdorfer G. G. Paracrine control of adrenal cortical function by medullary chromaffin cells. Pharmacol Rev 1996; 48: 495–530. 4. Brown R. D., Wisgerhof M., Carpenter P. C. et al. Adrenal sensitivity to angiotensin II and undiscovered aldosterone stimulating factors in hypertension. J Steroid Biochem 1979; 11: 1043–1050. 5. Goodfriend T. L., Egan B. M., Kelley D. E. Aldosterone in obesity. Endocr Res 1998; 24: 789–796. 6. Gardner H. W., Jursinic P. A. Degradation of linoleic acid hydroperoxides by a cysteine FeCl3 catalyst as a model for similar biochemical reactions. Biochim Biophys Acta 1981; 665: 100–133. 7. Goodfriend T. L., Ball D. L., Elliott M. E., Shackleton C. Lead increases aldosterone production by rat adrenal cells. Hypertension 1995; 25: 785–789. 8. Gardner H. W. Analysis of plant lipoxygenase metabolites. In: Christie W. W., ed. Advances in Lipid MethodologyFfour. Dundee: The Oily Press, 1997; 1–43. 9. Ozawa T., Hayakawa M., Takamura T. et al. Biosynthesis of leukotoxin, 9,10-epoxy-12 octadecenoate, by leukocytes in lung
Published by Elsevier Science Ltd.
10.
11.
12.
13.
14.
15.
16.
lavages of rat after exposure to hyperoxia. Biochem Biophys Res Commun 1986; 134: 1071–1078. Jia-ning H., Taki F., Sugiyama S. et al. Neutrophil-derived epoxide, 9,10-epoxy-12-octadecenoate, induces pulmonary edema. Lung 1988; 166: 327–337. Siegfried M. R., Aoki N., Lefer A. M., Elisseou E. M., Zipkin R. E. Direct cardiovascular actions of two metabolites of linoleic acid. Life Sci 1990; 46: 427–433. Ishizaki T., Shigemori K., Tamamura Y. et al. Increased nitric oxide biosynthesis in leukotoxin, 9,10-epoxy-12-octadecenoate injured lung. Biochem Biophys Res Commun 1995; 210: 133–137. Moghaddam M. F., Grant D. F., Cheek J. M., Greene J. F., Williamson K. C., Hammock B. D. Bioactivation of leukotoxins to their toxic diols by epoxide hydrolase. Nat Med 1997; 3: 562–566. Hamberg M. Decomposition of unsaturated fatty acid hydroperoxides by hemoglobin: structures of major products of 13L-hydroperoxy-9,11-octadecadienoic acid. Lipids 1975; 10: 87–92. Dix T. A., Marnett L. J. Conversion of linoleic acid hydroperoxide to hydroxy, keto, epoxyhydroxy, and trihydroxy fatty acids by hematin. J Biol Chem 1985; 260: 5351–5357. Goodfriend T. L., Ball D. L., Elliott M. E. et al. Fatty acids may regulate aldosterone secretion and mediate some of insulin’s effects on blood pressure. Prostaglandins Leukot Essent Fatty Acids 1993; 48: 43–50.
Prostaglandins, Leukotrienes and Essential FattyAcids (2002) 67(2^3), 163^167