Oxidized derivatives of linoleic acid have the potential to alter steroidogenesis. One such derivative is 12,13-epoxy-9- keto-10-(trans)-octadecenoic acid (EKODE). To evaluate the effect of EKODE on corticosterone production, dispersed rat zona fasciculata/reticularis (subcapsular) cells were incubated for 2 h with EKODE alone or together with rat ACTH (0, 0.2, or 2.0 ng/ml). In the absence of ACTH, EKODE (26 μM) increased corticosterone production from 5.3 ± 2.3 to 14.7 ± 5.0 ng · 106 cells · h−1. The stimulatory effect of ACTH was increased threefold in the presence of EKODE (26.0 μM). Cholesterol transport/P-450scc activity was assessed by measuring basal and cAMP-stimulated pregnenolone production in the presence of cyanoketone (1.1 μM). EKODE (13.1 and 26.0 μM) significantly increased basal and cAMP-stimulated (0.1 mM) pregnenolone production. In contrast, EKODE decreased the effect of 1.0 mM cAMP. EKODE had no effect on early or late-pathway activity in isolated mitochondria. We conclude that EKODE stimulates corticosterone biosynthesis and amplifies the effect of ACTH. Increased levels of fatty acid metabolites may be involved in the increased glucocorticoid production observed in obese humans.
- fatty acid
- zona fasciculata
linoleic acid(9,12-octadecadienoic acid) and its metabolites exert a variety of effects on physiological and pathophysiological processes (10,19, 20). Metabolites of linoleic acid have been shown to activate vascular endothelial cells (14), disrupt ionic pump currents (Na+/K+) in oligodendrocytes (12), and alter the characteristics of mitochondria (27). Free fatty acids such as linoleic acid play a role in cellular regulation in many endocrine tissues, including adrenocortical cells (25, 30).
Recent studies have identified and characterized a linoleic acid metabolite, formed by rat hepatocytes, that modulates aldosterone production in rat zona glomerulosa (ZG) cells in vitro (8). 12,13-Epoxy-9-keto-10-(trans)-octadecenoic acid (EKODE) is an oxidized metabolite of linoleic acid with a structure similar to products of leukocytes, such as leukotoxin and isoleukotoxin (13). However, leukotoxins display toxicity only upon hydrolysis of the epoxide to a dihydroxyl conformation. EKODE, on the other hand, requires the epoxide ring for its steroidogenic activity (8). The purpose of the current study was to evaluate the interaction of EKODE with ACTH-stimulated corticosterone production from rat zona fasciculata/reticularis (ZFR) cells in vitro. Furthermore, the sites of action of EKODE were analyzed by studying the early and late steroidogenic pathways in intact cells and isolated mitochondria. These experiments with fatty acids and their derivatives may bear upon the observed coincidence of obesity and increased glucocorticoid production in humans.
MATERIALS AND METHODS
Linoleic acid, progesterone, 11-DOC, dibutyryl-cAMP, BSA, isocitrate, halothane, and 25-hydroxycholesterol (25OH-cholesterol) were purchased from Sigma Chemical (St. Louis, MO). Collagenase was purchased from Worthington Biochemical (Freehold, NJ). Cyanoketone was kindly donated by Sanofi-Synthelabo (Malvern, PA). Rat ACTH-(1–39) was purchased from Peninsula Laboratories (Belmont, CA). All other chemicals were of reagent grade and were purchased from Sigma or Fisher Scientific (Fair Lawn, NJ).
All experimentation was approved by the Institutional Animal Care and Use Committees of the Medical College of Wisconsin and St. Luke's/Sinai Medical Center. Male Sprague-Dawley rats (n = 80; Harlan Sprague Dawley, Indianapolis, IN) were obtained at ∼5 wk of age (100–125 g). Animals were maintained on a standard sodium diet (Richmond Standard 5001, Brentwood, MO) and water ad libitum in a controlled environment (lights on, 0600–1800). On the morning of an experimental day, rats were lightly anesthetized with halothane and then decapitated. Adrenal glands were removed quickly, and subcapsules (ZFR) were manually separated from capsules. Subcapsules were immediately immersed in ice-cold Krebs-HEPES-Ca2+ buffer (pH 7.4) containing BSA (1 mg/ml) and subsequently incubated with collagenase (4 mg/ml) to digest the tissue. Each batch of cells was counted and assessed for viability by trypan blue exclusion and then diluted in fresh buffer to a final concentration of 100,000 cells/ml. All incubations were performed in a shaker bath at 37°C.
Linoleic acid was oxidized by a cell-free mechanism, as described previously (7). Briefly, linoleic acid was incubated under O2 in the presence of lipoxygenase type V to yield 13-hydroperoxyoctadecadienoic acid, which was subsequently incubated under O2 in the presence of cysteine and FeCl3. Chromatographic fractions (HPLC) were tested for biological activity in rat adrenal cells, and the active fractions were purified further. The chemical structure of EKODE was confirmed by GC-MS and NMR. Purified EKODE was reconstituted in ethanol and diluted in cell buffer, with the final experimental ethanol concentration never exceeding 0.2%.
To study the entire steroidogenic pathway, ZFR cells were incubated with EKODE (0, 7.8, or 26.0 μM) and/or ACTH (0, 0.2, or 2.0 ng/ml). EKODE was added to the cell suspension, followed immediately by other stimuli or substrates. The reaction was started by moving the tubes to a 37°C shaker bath. The corticosterone response to linoleic acid (1.7–49.9 μM) was used as a control. Cell suspensions were incubated for 2 h, and each treatment was run in triplicate. Tubes not treated with EKODE received vehicle, which exposed all cells to the same concentration of ethanol. At the end of incubation, cells were centrifuged quickly at 4°C, and supernatants were transferred to new tubes and frozen at −20°C until further analysis.
To examine early vs. late steroidogenic pathways, ZFR cells were incubated in the presence of cyanoketone (1.1 μM), an inhibitor of 3β-hydroxysteroid dehydrogenase, which was added to cells before EKODE. The early steroidogenic pathway (cholesterol transport in the mitochondria and side-chain cleavage of cholesterol) was assessed by measurement of pregnenolone. Cells were stimulated with dibutyryl-cAMP (0, 0.1, or 1.0 mM) and/or EKODE. The late pathway was assessed by measuring the conversion of exogenous progesterone (0, 0.3, or 3.2 μM) to corticosterone, in the presence or absence of EKODE.
Mitochondria were isolated as previously described to study the direct effects of EKODE on steroidogenic enzymes (23). The effect of EKODE (0, 7.8, or 26.0 μM) on P-450scc activity was assessed by the addition of isocitrate (10 mM), cyanoketone (1.1 μM), and 25OH-cholesterol (2.5 μM) with samples drawn at 0 and 30 min, rapidly cooled, centrifuged, and frozen for future measurement of pregnenolone. P-450c11β activity in the presence of EKODE was assessed by the addition of isocitrate, cyanoketone, and 11-DOC (7.2 μM), with measurement of corticosterone. Linoleic acid (23.8 μM) was also tested for steroidogenic activity in both early and late-pathway mitochondria studies. The integrity of the mitochondrial preparation was confirmed by a brisk conversion of 25OH-cholesterol to pregnenolone and 11-DOC to corticosterone in the absence of EKODE.
The concentrations of corticosterone and pregnenolone were measured by previously described RIA protocols (24). All data were analyzed by two-way ANOVA for repeated measures and Duncan's Multiple Range test for multiple comparisons (SigmaStat 2.03). P< 0.05 was considered significant. Data are presented as means ± SE.
EKODE by itself significantly increased corticosterone production from ZFR cells, with an ED50 of 22 μM and a sensitivity of 10.0–13.0 μM (Table 1). Linoleic acid (1.7–49.9 μM) had no effect on corticosterone production in ZFR cells.
There was a significant interaction between ACTH and EKODE in adrenal cells (Fig. 1). Low-dose ACTH (0.2 ng/ml) did not stimulate corticosterone production from ZFR cells in the absence of EKODE, but, in the presence of EKODE (7.8 μM), there was a significant increase. Corticosterone production in the presence of a higher concentration of ACTH (2.0 ng/ml) was increased significantly compared with basal, and addition of EKODE increased steroid production still more. The increment of corticosterone produced at lower doses of ACTH and EKODE was greater than the effect of either stimulus alone.
Figure 2 shows assessment of both the early and late steroidogenic pathways in ZFR cells. Pregnenolone production from endogenous cholesterol was measured without and with EKODE and cAMP (Fig. 2 A). In the absence of cAMP, EKODE significantly increased pregnenolone production in a concentration-related manner. Under low cAMP stimulation (0.1 mM), the magnitude of the EKODE effect was similar to its effect under basal conditions. At a higher concentration, cAMP (1.0 mM) alone increased pregnenolone production nearly five times. EKODE (26.0 μM) significantly decreased pregnenolone production at this level of cAMP stimulation. Figure 2 B, summarizes results from the analysis of the late pathway (exogenous progesterone to corticosterone). With no exogenous progesterone, EKODE (26.0 μM) significantly increased corticosterone production, although values were very low. In the presence of a low concentration of progesterone (0.3 μM), EKODE had a tendency to increase corticosterone production, but the differences were not statistically significant. When higher concentrations of progesterone (3.2 μM) were added, EKODE significantly decreased corticosterone production.
The conversion of 25OH-cholesterol to pregnenolone in isolated mitochondria was unaffected by EKODE or linoleic acid (data not shown). EKODE also had no effect on the conversion of 11-DOC to corticosterone in isolated mitochondria.
The present study assessed the effects of an oxidized metabolite of linoleic acid (EKODE) on rat adrenal glucocorticoid production. Micromolar amounts of EKODE stimulated corticosterone production by rat ZFR cells. Moreover, EKODE augmented the effect of ACTH on steroid production.
The effect of EKODE on corticosterone was different from its effect on aldosterone production by rat ZG cells (7). EKODE (at concentrations as high as 26.0 μM) stimulated corticosterone production from ZFR cells. By contrast, the peak aldosterone response to EKODE occurred at 5.0–15.0 μM, whereas higher concentrations were inhibitory (8). Furthermore, EKODE clearly amplified ACTH stimulation, whereas EKODE did not potentiate ANG II-stimulated aldosterone production (7).
Because pregnenolone production increased in whole cells, EKODE must have stimulated the movement of cholesterol from the cytosol into the mitochondria, the rate-determining step in steroidogenesis. Under low stimulation by cAMP (0.1 mM), EKODE potentiated pregnenolone production in whole cells; the magnitude of the increase was not different from the effect in the absence of cAMP. EKODE did not have a direct effect on P-450scc activity in isolated mitochondria using a permeable form of cholesterol as substrate. Steroidogenic acute regulatory protein (StAR) and peripheral-type benzodiazepine receptors (PBR) are known to be involved in cholesterol transport (4, 22,26, 29), and our mitochondria experiments utilized a permeable cholesterol analog. This analog enters the mitochondria without the mediation of StAR and PBR. Therefore, we have not excluded the possibility that EKODE altered the activity of these two important proteins in intact cells (27).
Under high levels of stimulation with cAMP (1.0 mM) or in the presence of excess substrate for the late pathway (3.2 μM progesterone), EKODE decreased pregnenolone and corticosterone production, respectively. This effect may be toxic and would agree with previous studies that found linoleic acid metabolites to be toxic in a number of cell types (10, 21). ACTH signaling is complex and does not act solely through the classical cAMP-protein kinase A signaling pathways (2, 6, 16). It is possible that EKODE modulates the diacylglycerol/inositol trisphosphate/Ca2+/protein kinase C cascade. Fatty acids impair insulin action in muscle via activation of protein kinase C, occurring through changes in intracellular diacylglycerol concentration (11, 15). EKODE may have a similar action on protein kinase C in steroidogenic cells. There are a number of other possible intracellular mechanisms through which EKODE could increase steroid production, but their discussion is beyond the scope of this report.
The isolation and characterization of EKODE was stimulated by the observation of elevated plasma aldosterone in adults with visceral obesity (9). The rate of production of glucocorticoids has also been shown to be elevated in humans with increased abdominal obesity (5, 17). Visceral fat is unique in that its venous blood flows directly into the portal vein. An excess of portal long-chain fatty acids, such as linoleic acid, can participate in the cluster of abnormalities referred to as the metabolic syndrome or syndrome X (1). EKODE has been detected in rat and human plasma with levels ranging from 10−9 to 5 × 10−7 M (Ref. 8 and unpublished results). Because a linoleic acid epoxygenase (CYP2C9) has been detected in hepatic microsomes (3), a testable hypothesis can be formulated: as fatty acids pass through the liver, metabolites such as EKODE are produced that stimulate glucocorticoid production. The elevated glucocorticoids initiate a vicious cycle by promoting further accumulation of visceral adipose tissue (18). Evidence for an effect of high-fat feeding on adrenocortical activity in the rat provides some support for this theory (28).
The results of the present study confirm that oxidized metabolites of linoleic acid modulate steroidogenesis in the adrenal cortex. EKODE increased both basal and ACTH-stimulated corticosterone production from rat ZFR cells in vitro. This effect was observed when the early steroidogenic pathway was analyzed separately in whole cells and is probably not caused by a direct effect on mitochondrial steroidogenicP-450 enzymes. To better understand the cellular mechanism of action of EKODE, it will be relevant to evaluate cAMP and other second messenger systems. Further research on the effects of this linoleic acid metabolite on glucocorticoid production may provide insight into adrenal regulation and metabolic diseases.
We thank Sanofi-Synthelabo, Inc., for the generous donation of cyanoketone and Barbara M. Jankowski for expert technical assistance.
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-54685.
Address for reprint requests and other correspondence: H. Raff, Endocrinology and Metabolism, St. Luke's Physician's Office Bldg., 2801 W. KK River Pkwy., Suite 245, Milwaukee, WI 53215 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2003 the American Physiological Society