Biochimica et Biophysica Acta, 1171 (1992) 27-34 © 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4781/92/$05.00

27

BBAEXP 92426

Arachidonic acid regulates unsaturated fatty acid synthesis in lymphocytes by inhibiting stearoyl-CoA desaturase gene expression Paul W. Tebbey and Thomas M. Buttke Department of Microbiology and Immunology, East Carolina University School of Medicine, Greem;ille, NC (USA) (Received 30 January 1992) (Revised manuscript received 11 June 1992)

Key words: Arachidonic acid; Stearoyl-CoA desaturase; Lymphocyte; Gene expression

This work was based upon the observation that a reduction in the level of serum, provided to murine lymphocytes in culture, augmented endogenous unsaturated fatty acid (UFA) synthesis. Since the phospholipids of BW5147 cells grown in 1% serum were especially deficient in arachidonic acid (20 : 4), and given the findings of previous workers, we suspected that the availability of exogenous 20:4 in serum might correlate with the squelching of UFA synthesis. Indeed, after a 5 h exposure to 4-28 /zM 20:4, the 20:4 content of BW5147 cell phospholipids increased from 1% to 15% of the total fatty acids with a coincident reduction in 18 : 1 synthesis to approx. 30% of starting values. Subsequent studies were done to define the mechanism by which 20:4 down-regulates 18:1 synthesis. The results indicated that 20:4 inhibited endogenous 18:1 synthesis by reducing stearoyl-CoA desaturase (SCD) enzyme activity. Moreover, as determined by Northern blot analyses, the inhibitory effect of 20 : 4 on stearoyl-CoA desaturase activity coincided with decreased stearoyl-CoA desaturase mRNA levels. Exposure of BW5147 cells to either 20 : 4, actinomycin D, or both, resulted in a temporal decay of stearoyl-CoA desaturase mRNAs with half-lives ranging from 4.0 h to 4.4 h. Such a similarity in decay times implied that 20 : 4 regulates stearoyl-CoA desaturase expression by inhibiting transcription. This was confirmed by nuclear run-on studies in which 20:4 was found to inhibit transcription of nascent stearoyl-CoA desaturase mRNA. Collectively, these findings implicate 20:4 as an important regulator of stearoyl-CoA desaturase gene expression, and hence UFA synthesis, in lymphoid cells.

Introduction The relationship between plasma m e m b r a n e fluidity and cellular growth or function has been well documented in numerous cell systems [1-3]. Mammalian cell m e m b r a n e s consist of approximately equal proportions of saturated and unsaturated fatty acids, the balance of which regulates bilayer fluidity and thus, m e m b r a n e function [1-3]. The predominant saturated fatty acids (SFAs) of mammalian cells are palmitic (16:0) and stearic (18:0), both of which are normally

Correspondence to: T.M. Buttke, Department of Microbiology and Immunology, East Carolina University School of Medicine, Greenville, NC 27858-4354, USA. Abbreviations: 5-HETE, (5S)-5-hydroxy-6,8,11,14-eicosatetraenoic acid; 5-HPETE, (5S)-5-hydroperoxy-6,8,11,14-eicosatetraenoic acid; FCS, fetal calf serum; FBS, fetal bovine serum; 20:4, arachidonic acid; 18:0, stearic acid; 16:0, palmitic acid; 18: 1, oleic acid; 18:2, linoleic acid; 16: 1, palmitoleic acid; SCD, stearoyl-CoA desaturase; SFA, saturated fatty acid; UFA, unsaturated fatty acid; FABPs, fatty acid-binding proteins; GLUT-l, glucose transporter type 1.

synthesized endogenously via the activities of acetylCoA carboxylase and fatty acid synthase enzyme complexes [4]. Unsaturated fatty acids (UFAs) can be synthesized endogenously by the actions of stearoyI-CoA desaturase (SCD) or may be derived from exogenous sources [2,5]. Stearoyl-CoA desaturase is a A9-fatty acyl-CoA desaturase which requires the involvement of cytochrome bs, NADH-cytochrome bs-reductase and molecular oxygen in order to catalyze the insertion of a double bond between the C-9 and C-10 atoms of either palmitic or stearic acids in the formation of palmitoleic acid (16 : 1) and oleic acid (18 : 1), respectively [5,6]. One of the major UFAs present in mammalian cells is arachidonic acid (20:4) which may account for as much as 25% of all phospholipid fatty acids [7]. Arachidonic acid (20:4) is principally synthesized in the liver from dietary linoleic acid (18:2) and then transported to other cell types via serum albumin or lipoproteins [8]. Serum levels of 20 : 4 are low relative to other fatty acids but many cells possess a high affinity arachidonyl-CoA synthetase which facilitates selective accumulation of 20:4 even when other fatty acid species

28 are in excess [9,10]. Such a characteristic implies a unique role for 20:4 in cellular growth or function. Indeed, accumulating evidence indicates a role for 20:4 metabolites in the cellular signalling pathways of various cell types [11]. For example, in the adipogenic TA1 ceil line, tumor necrosis factor mediates its functions by activating two secondary messenger pathways, one of which relies upon the release of 20:4 from membrane phospholipids and its subsequent metabolism to 5 - H P E T E [12]. Metabolism of 20:4 by the cyclooxygenase pathway concludes with the production of prostaglandins and thromboxane whereas the actions of lipoxygenases result in the formation of leukotrienes. Interestingly, some cells, including lymphocytes, do not seem to produce prostaglandins or leukotrienes suggesting a possible non-eicosanoid role for 20:4 in these cell types [13]. Alterations in dietary lipid intake can have profound influences on the fatty acid profiles of mammalian cell membranes [7,14-17]. For example, diets rich in polyunsaturated fatty acids lead to increased proportions of membrane polyenoics with a corresponding decrease in monounsaturated fatty acids [2,7]. The change in membrane fatty acid composition results, in part, from effects on endogenous 18:1 synthesis. Polyunsaturated fatty acids such as 20:4, have been suggested to inhibit endogenous 18:1 synthesis in both rat lung and hamster kidney cells, thereby allowing preferential incorporation of the exogenously-derived fatty acids into cellular phospholipids [18,19]. However, the precise mechanism by which polyenoic fatty acids influence the activity of stearoyl-CoA desaturase remains obscure. The present studies initiate the characterization of the mechanism of action of 20:4 upon stearoyl-~oA desaturase gene expression in lymphocytes. The data presented herein define a role for 20:4 as a potent regulator of 18 : 1 synthesis in lymphocytes. These studies also demonstrate that 20:4 acts at the level of transcription to repress stearoyl-CoA desaturase gene expression rather than by modulating either translation or the activity of pre-formed enzyme. Thus, the regulation of stearoyl-CoA desaturase gene expression by 20:4 is analogous to the regulation of other genes involved in fatty acid synthesis [20,21]. Materials and Methods

Cell culture BW5147 cells (mouse lymphoma, ATCC TIB 48) were cultured in RPMI-1640 supplemented with 1% or 10% fetal calf serum (FCS) (Hyclone) or 1% iron-supplemented fetal bovine serum (FBS) (Hyclone) in order to induce stearoyl-CoA desaturase gene expression. Cultures were maintained at 37°C in a humidified atmosphere of 6% CO~-94% air.

Lipid extraction and analyses Cellular lipids were extracted with chloroformmethanol and separated by thin layer chromatography (TLC) on silica gel plates as previously described [7]. Briefly, total phospholipids were isolated by TLC using a solvent system of petroleum ether/diethyl e t h e r / acetic acid; 70 : 30 : 1, v/v. The region corresponding to phospholipids was scraped and fatty acids were liberated and converted to methyl esters by heating in a solution of 5% (w/v) HC1 in MeOH at 75°C for 1 h. Methyl esters were extracted with pentane and analysed with a Varian 3700 gas-liquid chromatograph (GLC) interfaced to a Varian 4290 integrator. The methyl esters were separated on a 6-ft column of 1(1% SP-2330 on 100/120 Chromosorb WAW (Supelco), with a temperature program of 5 min at 170°C followed by a temperature increase of 5°C/min to 200°C. Fatty acids were identified by comparing their retention times to authentic fatty acid standards obtained from Supelco (Bellefonte, PA), Sigma (St. Louis, MO), or Nu Chek Prep (Elsian, MN).

Assay of endogenous 18:1 synthesis In order to assay endogenous fatty acid synthesis, BW5147 cells (10 7 cells per sample) were pulsed with 10/zCi of [1-14C]acetate (55 mCi/mmol; New England Nuclear) for 1 h. Cells were harvested, washed twice in phosphate-buffered saline and [ipids extracted and methylated as described above. The radiolabeled esters were separated on the basis of their degree of unsaturation using argentation TLC [22]. In this TLC system, the monoenoics 18:1 and 16:1 are resolved and well separated from 16:0 and 18:0 which migrate with the solvent front. BW5147 cells cultured in the absence of 20:4 synthesized primarily 16:0 and 18:1 with trace amounts of 18:0 and eicosenoic acid. No 16:1 or polyenoic fatty acids were detected (data not shown). Radiolabeled lipids were quantified using a RITA linear radioactivity detector (Raytest, Muenster. Germany).

Isolation and assay of stearoyl-CoA desaturase actit,ity BW5147 cells cultured in 1% FBS medium (_+ 20:4) were disrupted by sonication using a Biosonik Ill sonicator (Bronwill) at maximal output [23]. Following a slow-speed centrifugation (10 rain, 2 0 0 0 x g , 4°C) to remove nuclei and unbroken cells, membranes were pelleted by centrifugation at 105000 x g for 45 rain. The membranes were resuspended in 10 mM Tris buffer (pH 7.6) containing 0.5 mM MgCI 2 and 150 /xg/ml phenyimethylsulfonyl fluoride. Protein was subsequently quantitated by a modified Lowry procedure [24]. Stearoyl-CoA desaturase activity of cell-free extracts was measured by monitoring the conversion of [114C]18 : 0-CoA to [1-~4C]18 : 1 [23,25]. Equivalent

29 amounts of protein were added to assay mixtures containing 0.1 M Tris (pH 7.2), 1 mM NADH, and 70 ~M [1-14C]18:0-COA (NEN; 5.5 mCi/mmol), and the mixture was incubated for 10-30 min at 37°C. Reactions were stopped by addition of 10% (w/v) K O H / M e O H followed by heating at 80°C for 30 min. After acidification, fatty acids were recovered and converted to methyl esters for analysis by argentation TLC. Northern blot analyses Expression of stearoyl-CoA desaturase mRNA was analysed using a cDNA probe for rat liver stearoyl-CoA desaturase obtained from P. Strittmatter [26]. GLUT-1 gene expression was measured using a murine cDNA probe obtained from P. Pekala [27] and /3-actin was assayed using a probe for chicken /3-actin purchased from Oncor (Gaithersburg, MD). Total cellular RNA was isolated from 5" 107 cells cultured with 12 ~zM 20:4, 5 /xg/ml actinomycin D, or both, for various time periods. RNA was extracted with guanidium isothiocyanate and centrifuged (30 000 x g, 24 h) through 5.7 M CsC1 containing 100 mM EDTA [28]. For Northern analyses, 20 /~g of total RNA was separated by electrophoresis through 1.2% formaldehyde-agarose gels and transferred to Zeta-probe membranes (Bio-Rad) as described [23,29]. Prehybridization was performed for 4 h at 42°C in a solution containing 50% deionized formamide, 4 x SSC (1 x SSC = 150 mM sodium chloride/15 mM sodium citrate), 5 x Denhardt's solution (1 × Denhardt's = 0.02% each of polyvinylpyrrolidone, bovine serum albumin, and ficoll), denatured salmon sperm DNA at 25/zg/ml, 50 mM sodium phosphate buffer (pH 7), 0.5 mg/ml sodium pyrophosphate and 1% SDS. Hybridizations were performed with 2" 107 cpm/ml of cDNA probes labeled with [a-SZp]dCTP (NEN, 3000 Ci/mmol) by random priming [30] in a similar solution except that the Denhardt's solution was 1 x . After 16 to 24 h, filters were washed twice (each for 30 min) in 3 x SSC/0.1% SDS, twice in 0.3 X SDS/0.1% SDS, once in 0.1 x SSC/0.1% SDS at room temperature before a final wash in 0.1 x SSC/0.1% SDS at 60-65°C. Washed filters were subjected to autoradiography using Kodak XAR-5 film in cassettes containing DuPont lightening plus intensifying screens for 1-3 days at -80°C. The relative intensities of the bands on the autoradiographs were quantified by scanning on a LKB Ultrascan XL laser densitometer with Gelscan XL software. Nuclei isolation and nuclear run-on transcription assays BW5147 cells (~ 5 • 1 0 7 per time point) cultured in 1% FBS/RPMI were exposed to 12 /xM 20:4 for 0 and 2 h. Nuclei were isolated as previously described by Cornelius et al. [31]. Subsequently, nuclei were subjected to nuclear run-on transcription assay in order to measure the rate of nascent mRNA synthesis. Run-

on assays were performed exactly as described previously [31]. Results

Effect of serum deprivation on UFA synthesis in BW5147 cells As a prelude to studies regarding the regulation of fatty acid synthesis in lymphoid ceils, the BW5147 cell line was adapted to growth in the presence of low serum by gradually reducing the level of FCS provided from 10% to 1%. One effect of the reduction in FCS was a substantial increase in endogenous 18:1 synthesis. Whereas BW5147 cells grown in 10% FCS medium synthesized relatively low levels of 18:1 (< 15% of total fatty acids), in cells grown in 1% FCS, 18:1 accounted for approx. 50% of the endogenously synthesized fatty acids (data not shown). These findings were in good agreement with a previous study in which delipidation of serum was shown to induce stearoylCoA desaturase activity in Chinese hamster lung fibroblasts [32]. We subsequently compared the phospholipid fatty acid contents of 10% versus 1% FCS-supplemented BW5147 cells and found that the latter were especially deficient in 20:4 (4.8% vs. 0.7% respectively, data not shown). Given the ability of lymphoid cells to selectively incorporate 20:4 [10] and in view of similar studies conducted by other researchers [19,21,33], our results suggested that the availability of exogenous 20:4 might regulate the expression of stearoyl-CoA desaturase in lymphoid cells. Inhibition of UFA synthesis by 20:4 in BW5147 cells To more clearly establish 20:4 as a component of serum capable of suppressing endogenous 18 : 1 synthesis, we determined the effects of 20 : 4 supplementation upon cellular UFA synthesis and phospholipid fatty acid content. BW5147 cells cultured in 1% FBS were supplemented with 0-28 /zM 20:4. The 20:4 was complexed to bovine serum albumin (BSA) (FA/BSA = 1.5:1) prior to addition, since we have previously shown that addition of fatty acids as BSA-complexes facilitates their efficient esterification into phospholipids and triacylglycerides [23]. After a 5 h exposure to 20:4, culture aliquots were removed a n d pulsed with [1-14C]acetate to label newly-synthesized fatty acids. The remainder of the cultures were harvested for subsequent analyses of phospholipid fatty acids by GLC. By this methodology, we determined that the membrane phospholipids, although originally containing only small amounts (< 1%) of 20:4, became enriched with the polyenoic fatty acid in a dose-dependent fashion (Fig. 1). The level of phospholipid-associated 20 : 4 plateaued at approx. 15% of total fatty acids, a level similar to that seen previously in freshly isolated murine lymphocytes [7]. The incorporation of 20 : 4 was accom-

30

---- 6 0 -15 ~ 5o ~

O"""".0 /

f

~ 0

gs

2O

5

0 0 5 Dose of

10 20:4

effect on stearoyl-CoA desaturase per se. Indeed, in contrast to a previous report with rat liver stearoyl-CoA desaturase [34], the addition of up to 1 mM 20:4 to cell free extracts from BW5147 cells had no effect upon enzyme activity (data not shown). Collectively, these data suggest that 20 : 4 inhibits endogenous 18 : 1 synthesis via a pre-translational rather than a posttranslational mechanism.

~

o~

.F

15 (UM)

Fig. 1. Effect of exogenous 20:4 on phospholipid fatty acid composition and 18:1 synthesis in BW147 cells. BW5147 cells cultured in 1% F B S / R P M I were supplemented with varying doses (0-28 /xM) of 20:4. After 5 h, [1-14C]acetate was added and the incubation continued for 1 h. Cells were harvested and total phospholipid fatty acids were isolated and methylated as described in Materials and Methods. Fatty acid methyl esters were analyzed by either G L C for total phospholipid 20:4 content or by argentation TLC to assay endogenous 18 : 1 synthesis.

panied by a decrease in endogenous 18:1 synthesis. BW5147 cells cultured in the absence of 20:4 synthesized roughly equal amounts of SFA and UFA. However, the addition of 20:4 resulted in a dose-dependent decrease in endogenous 18 : 1 synthesis, with maximal (70%) inhibition occuring at 12/xM 20:4 (Fig. 1). These observations implicate 20:4 as an important physiological regulator of endogenous U F A synthesis in lymphoid ceils.

Effect of exogenous 20:4 on stearoyl-CoA desaturase actiuity in BW5147 cells The reduction in endogenous 18:1 synthesis in 20:4-supplemented cells could result from a reduction in stearoyl-CoA desaturase activity. To test this possibility, BW5147 cells cultured in 1% FBS medium were supplemented with 0, 6 or 12 /xM 20:4, and after 1 h and 8 h time periods, microsomal membranes were isolated and stearoyl-CoA desaturase activity was measured [23]. As shown in Fig. 2 (left panel), culturing BW5147 cells in the presence of 20:4 for 1 h did not significantly alter their level of stearoyl-CoA desaturase activity. By contrast, when microsomal extracts from cells exposed to 6 or 12 /xM 20:4 for 8 h were tested, their level of stearoyl-CoA desaturase activity was reduced by 90% (Fig. 2, right panel). Since BW5147 cells cultured for 1 h in the presence of 20:4 incorporated substantial amounts of the polyenoic fatty acid (Clarke, A. and Buttke, T.M., unpublished observations) the absence of a detectable change in stearoylCoA desaturase activity in such cells (Fig. 2, left panel) suggests that 20:4 does not have a direct inhibitory

Northern blot analyses of stearoyl-CoA desaturase mRNA in BW5147 cells To determine whether the observed decrease in UFA synthesis could be correlated with a reduction in stearoyl-CoA desaturase m R N A levels, Northern blot analyses were performed on BW5147 cells exposed to various doses of 20:4. Cells were cultured for 5 h in the presence of 0-28 /xM 20:4, after which time aliquots were removed and labeled with [1-t4C]acetate to assay endogenous fatty acid synthesis. Total cellular RNAs were isolated from the remaining cells and subjected to Northern blot hybridization using a - 3 2 p labeled cDNA probes corresponding to either rat liver stearoyl-CoA desaturase [26] or chicken /3-actin (Fig. 3). While the levels of stearoyl-CoA desaturase mRNA visibly decreased with increasing doses of 20:4, no similar effect was apparent with regard to /3-actin. Furthermore, as shown in Fig. 4, the inhibitory effect of 20:4 on endogenous 18:1 synthesis coincided with decreased levels of stearoyl-CoA desaturase mRNA. At 20:4 doses > 14 /xM, the level of stearoyl-CoA desaturase m R N A declined by approx. 90% compared to untreated controls (Fig. 4). One possible mechanism for the inhibitory effect of 20:4 on 18:1 synthesis would involve enhanced stea-

1 Hr Exposure

400

~ 2" 200 t

&

8 Hr Exposure 500

15o1

300

too

200 •

Q ,~

"

50

0

:

t

0 UN 20:4 6 UM 2 0 4 • 12 ~N 2 0 4

100 0

10

2'0

3'0

Assay Time (rain)

0 UM 2 0 4 6 UN 204 12 UM 204

113 2'0 3'0 Assay Time (rain)

Fig. 2. Effect of 20:4 supplementation on stearoyl-CoA desaturase enzyme activity. BW5147 cells cultured in 1% F B S / R P M I were additionally supplemented with 0, 6 or 12 /xM 20:4. At 1 h and 8 h after 20:4 addition cells were harvested, washed and disrupted by sonication. Cell-free extracts were assayed for stearoyl-CoA desaturase activity by measuring the conversion of [1-I4C]18:0-CoA to [1-14Cl18 : 1-CoA.

31 royl-CoA desaturase m R N A turnover. To examine this possibility, BW5147 cells were grown in 1% FBS in RPMI to induce maximal stearoyl-CoA desaturase expression, after which time the cultures were supplemented with either 2 0 : 4 (12 /zM), actinomycin D (5 ~ g / m l ) , or both 2 0 : 4 and actinomycin D. Subsequent Northern blot analyses showed that exposure to 20:4, actinomycin D, or both (Fig. 5) resulted in temporal decreases in stearoyl-CoA desaturase mRNA. By comparison, 20 : 4 did not affect the levels of either GLUT-1 or /3-actin m R N A s (Fig. 5), thus confirming 2 0 : 4 ' s specificity of action. Quantitation of m R N A half-lives was determined by densitometric scanning of the autoradiograms. As shown in Fig. 6A, following the addition of 12 /xM 20:4, stearoyl-CoA desaturase m R N A declined with a t~/2 = 4.0 h while the levels of GLUT-1 remained relatively unchanged (Fig. 6A). In the presence of actinomycin D alone (Fig. 6B), stearoyl-CoA

(pM)

+20:4 0

4

7

~, ,'---

~T

4o~ _,'2,,,° ° / /% -

l'8°~

~ \!

o SFA • UFA

.E-b 30-1~;

v, -~

I\

* mRNA

.

-60 >

E 20- ,'%

lo

,40

2o

".....,

o

........ ._~

o 6

lb

o

2'0

3°

Dose of 20:4 (uM) Fig. 4. Effect of 20:4 on stearoyl-CoA desaturase mRNA levels. BW5147 cells grown in 1% FBS/RPMI were supplemented with the indicated doses of 20:4 and 5 h later aliquots were removed and labeled with [1-14C]acetate to assay endogenous fatty acid synthesis. RNA was isolated from the remaining cultures. Northern blots were hybridized with a cDNA probe for stearoyl-CoAdesaturase, stripped and rehybridized with a cDNA probe for /3-actin. Stearoyl-CoA desaturase mRNA levels were normalized to /3-actin following densitometric scanning of the resultant autoradiograms.

14 28

SCD

5.0kb

,B-Actin

1.9kb

288rRNA

4.7kb

desaturase m R N A decayed with t½= 4.4 h while GLUT-1 decayed with t~ = 1.6 h, the latter being in good agreement with the reported short half-life for GLUT-1 m R N A [31]. The addition of both actinomycin D and 2 0 : 4 to the culture medium resulted in stearoyl-CoA desaturase and GLUT-1 m R N A half-lives of t~= 4.1 h and 1.3 h, respectively (Fig. 6C). The similarity in the half-life of stearoyl-CoA desaturase m R N A in response to all three culture conditions suggested that 2 0 : 4 does not inhibit stearoyl-CoA desaturase gene expression by altering m R N A stability. Furthermore, the effects of 2 0 : 4 are highly specific in that the decrease in stearoyl-CoA desaturase m R N A occurred in the absence of a similar reduction in either GLUT-1 or/3-actin m R N A levels (Fig. 5).

Inhibition of nascent stearoyl-CoA desaturase mRNA synthesis by 20:4

188rRNA

1.9kb

Fig. 3. Northern analyses of the 20:4-mediated reduction of stearoyl-CoA desaturase mRNA. RNA was isolated from cells exposed to the indicated doses of 20:4 after a time period of 5 h. Northern blots were hybridized with cDNA probes encoding the full-length rat liver stearoyl-CoA desaturase coding region (SCD), stripped and subsequently rebybridized with a cDNA for chicken /3-actin. The bottom panels indicate ethidium bromide stained gels showing the 28S and 18S ribosomal RNA subunits. The size of each mRNA species is given in kilobases (kb) to the right of each panel. The autoradiogram is representative of an experiment performed three times in which identical results were obtained.

The data shown in Figs. 2 - 6 suggest that 2 0 : 4 affects stearoyl-CoA desaturase activity at the level of gene transcription. Accordingly, nuclear transcription run-on assays were performed to determine the transcription rates of stearoyl-CoA desaturase m R N A in BW5147 cells cultured in the presence of 12/~M 2 0 : 4 for 0 and 2 h. The transcription rate of the 13-actin gene in the same cells was also measured as a control for gene specificity. Following autoradiography, densitometric analyses of hybridization intensity were normalized to the signal generated by hybridization of in vitro transcribed R N A to BW5147 genomic DNA. The results obtained (Fig. 7) show that after a 2 h exposure

32

+ 12pM 20:4 00.5

1

2

4

6

+ 5 p g / m l Act D + 12pM 20:4

+5jJg/ml A c t D Hr

8 10

0 0.5 1 2

4

6

8

10

Hr

0 0.5 1

2

4

6 8 10

SCD -

5.0kb

GLUT- 1 -

2.8kb

~B-Actin -

1.9kb

28SrRNA

4.7kb

18SrRNA 1.9kb

Fig. 5. Northern analyses of stearoyl-CoA desaturase m R N A stability. BW5147 cells grown in 1% F B S / R P M I were exposed to either 12 # M 20:4 (left), 5 / ~ g / m l actinomycin D (middle), or both (right), for the indicated times. At each time point, culture aliquots were removed and total R N A isolated. Northern blots were sequentially hybridized with c D N A probes corresponding to stearoyl-CoA desaturase (SCD), G L U T - l , and /3-actin.

to 12 /~M 20:4, the rate of stearoyl-CoA desaturase mRNA synthesis declined by 61% while the transcription of/3-actin increased 2.7-fold. The results obtained for stearoyl-CoA desaturase are consistent with the premise that the decreases in m R N A levels shown in Figs. 4 and 6 are primarily due to a reduction in nascent stearoyl-CoA desaturase mRNA transcription. Interestingly, the apparent increase observed in/3-actin transcription in response to 20:4-treatment mimics

c"

+20:4 0

100"

2

Hr

SCD

E SCD

• o

I0-

GLUT-1

,B-Actin

i i i

z E

that seen after treatment of 3T3-L1 adipocytes with tumor necrosis factor [27], an effect which may involve the release of 20:4 from cellular phospholipids [12]. Overall, the data shown in Fig. 7 clearly identifies 20 : 4 as a potent gene regulator which acts at the level of transcription initiation.

A i

0

•

|

•

i

•

5

i

•

I

o,

PGEM

IC, ......

•

I0

0

5

10

0

5

I0

Time ( H r ) Fig. 6. Effect of 20:4 on stearoyl-CoA desaturase m R N A turnover. The Northern blot analyses shown in Fig. 5 were analyzed by densitometric scanning and the data plotted as the percent of mR N A s remaining in BW5147 cells exposed to either 12 p.M 20:4 (A), 5 ~ g / m l actinomycin D (B) or both 20:4 and actinomycin D (C), for the indicated times. In each case the levels of stearoyl-CoA desaturase and GLUT-1 m R N A s were normalized to /3-actin. Data are presented as the mean value generated from independent experiments: A, n = 4: B, n = 4: C, n = 1. The m R N A half-lives calculated from lines plotted according to linear regression values, were; A, S C D = 4 . 0 h; B, S C D = 4 . 4 h, G L U T - l = 1 . 6 h; C, S C D = 4 . 1 h, GLUT-1 = 1.3 h.

DNA Fig. 7. Regulation of stearoyI-CoA desaturase gene transcription by 20:4. Nuclei were isolated from BW5147 cells exposed to 12 ~ M 20:4 for 0 and 2 h and subsequently used in transcription run-on assays performed as described in Materials and Methods. In vitro transcribed a-32p-labeled R N A s were subsequently hybridized to stearoyl-CoA desaturase,/3-actin and P G E M c D N A s (2/~g each), or BW5147 genomic D N A (0.2 g g / f i l t e r ) immobilized on nylon membranes (Zeta-probe). Following autoradiography for 24 h at - 8 0 ° C , the relative levels of gene expression were quantified by densitometric scanning. A representative autoradiogram is presented from an experiment performed twice with independent nuclear isolations. Each experiment yielded identical results.

33

Discussion The data obtained in this study implicate 20 : 4 as an important regulator of stearoyl-CoA desaturase gene expression in murine lymphocytes. The results therefore confirm and extend previous reports by other workers showing that polyenoics are more potent than SFAs and monounsaturated fatty acids in down-regulating UFA synthesis and indeed, fatty acid synthesis in general [18-21]. Our results also provide new information regarding mechanisms by which 20:4 regulates 18:1 synthesis. While previous workers have reported the effects on 18:1 synthesis in intact cells by merely measuring the uptake and incorporation of [1~4C]acetate [18,19], we have additionally demonstrated a reduction in actual stearoyl-CoA desaturase enzyme activity. The observed reduction could occur via pre- or post-translational mechanisms. While our studies do not preclude enzyme modifications, we clearly show a regulation at the transcriptional level. Such regulation may thus involve similar processes to those occurring in the suppression of fatty acid synthase gene expression by 20 carbon polyenes, including 20:4 [20,21]. Evidence for transcriptional mechanisms of regulation of 18:1 synthesis have been described in both rats and chickens wherein, dietary UFAs were seen to reduce mRNA levels of stearoyl-CoA desaturase [26,35]. Also, m R N A levels of OLE-1, the apparent structural gene for the yeast A-9 desaturase, declined in the presence of exogenous UFAs such as linoleic acid [36]. Using OLE-1 promoter-lacZ fusions, the region responsive to UFA regulation was localized to a 111 bp fragment approx. 500 bp upstream of the transcriptional start site of the gene [37]. In mammalian cells, UFA regulation is more complex in that at least two genes encoding stearoyl-CoA desaturase have been identified [38,39]. Whereas liver expresses SCD-1 exclusively, spleen, heart and brain express only SCD-2 [39]. By contrast, both SCD-1 and SCD-2 are expressed in adipose tissue, lung and kidney to varying extents [39]. Such cell-type specific expression of the two genes may arise from positional differences of various transcriptional elements within the respective promoter regions [39]. Similarly, the 20:4responsiveness of stearoyl-CoA desaturase gene expression may also be maintained via regulatory elements within the promoter. Since we have recently identified SCD-2 as the form of stearoyl-CoA desaturase expressed in lymphocytes and further shown its responsiveness to 20:4 [50], there seems to be precedent for the existence of a putative 'arachidonic acidresponse element' within the SCD-2 promoter. Such a scenario would be akin to the regulation of HMG-CoA reductase by cholesterol [40], mediated at the transcriptional level by a sterol-response element [41]. Our results implicate 20 : 4 per se, or a metabolite of

it, as a major influential component of stearoyl-CoA desaturase gene expression. Reports that 20:4 has the capability of binding to DNA thereby controlling oncogene expression or inducing chromosomal abberations in human lymphocytes implies a non-eicosanoid role for 20:4 [42]. Alternatively, 20:4 may need to be metabolized in order to affect gene expression. Although prostaglandins and leukotrienes may not be synthesized in lymphocytes [13], evidence for 5-1ipoxygenase activity in Epstein-Barr virus transformed human B cells [43] and two mouse T cell lines [44,45] suggests that 20 : 4 metabolites such as 5-HETE may be involved in lymphocyte activation. In other cell types, such metabolites have been implicated in a variety of actions at the gene level including induction of c-los and Egr-1 immediate early genes in rat mesangial cells [46] and production of interferon- in mouse spleen cells [47]. In the adipogenic TA1 cell line, 20:4 was seen to induce c-los transcription by a pathway that was inhibitable by nordihydroguaiaretic acid (NDGA) indicating a reliance upon 5-1ipoxygenase activity for the genetic effect [12]. In contrast, other fatty acids, such as 18:1, were unable to induce c-los transcription. In similar experiments with BW5147 cells we have also observed an induction of c-los in response to 20 : 4 but not 18:1 [51]. It is therefore conceivable that regulation of stearoyl-CoA desaturase transcription by 20:4 is dependent on both the activity of 5-1ipoxygenase and the induction of one or more immediate early genes. Alternatively, the genetic effect of 20 : 4 may be mediated by fatty acid binding proteins (FABPs), in a manner reminiscent of the action of retinoic acid [48,49]. Such a model would involve a 20:4-FABP association and subsequent DNA binding. However, the function of FABPs remains largely unknown, although it has been speculated that they may play a role in gene regulation and expression [48]. In summary, this research defines an important role for 20:4 in the regulation of endogenous UFA synthesis in lymphocytes. Furthermore, like other genes involved in fatty acid synthesis [20,21], the control of stearoyl-CoA desaturase gene expression is mediated at the transcriptional level. Future experiments are aimed at more fully characterizing the 20:4-mediated regulation of stearoyl-CoA desaturase with particular emphasis on the possible role of 20 : 4-derived metabolites and immediate early genes.

Acknowledgments These studies were supported, in part, by a grant from the East Carolina University School of Medicine Diabetes Research Council to T.M.B. The authors wish to thank Drs. J.S. Stephens and P.H. Pekala for expert experimental help and advice, and Mr. S. Van Cleave for technical assistance.

34 References 1 Quinn, P.J. and Chapman, D. (1980) CRC Crit. Rev. Biochem. 8, 1-117. 2 Stubbs, C.D. and Smith, A.D. (1984) Biochim. Biophys. Acta 779, 89-137. 3 Tebbey, P.W. and Buttke, T.M. (1990) Immunology 70, 379-384. 4 Volpe, J.J. and Vagelos, P.R. (1976) Physiol. Rev. 56, 339-417. 5 Jeffcoat, R. (1977) Biochem. Soc. Trans. 5,811-818. 6 Bloomfield, D.K. and Bloch, K. (1960) J. Biol. Chem. 235, 337344. 7 Buttke, T.M., Mallett, G.S. and Meydrech, E.F. 11985) Nutr. Rep. Int. 32, 749-756. 8 Habenicht, A.J.R., Salbach, P., Goerig, M., Zeh, W., Janssent, U., Blattner, C., King, W.C. and Glomset, J.A. (1990) Nature 345, 634-636. 9 Neufeld, E.J., Wilson, D.B., Sprecher, H. and Majerus, P.W. (1983) J. Clin. Invest. 72, 214-220. 10 Taylor, A.S., Sprecher, H. and Russell, J.H. 11985) Biochim. Biophys. Acta 833, 229-238. 11 Mason-Garcia, M. and Beckman, B.S. (1991) FASEB J. 5, 29582964. 12 Haliday, E.M., Ramesha, C.S. and Ringold, G. (1991) EMBO J. 10, 109-115. 13 Goldyne, M.E. (1988) Prog. Allergy 44, 140-152. 14 Kollmorgen, G.M., Sansing, W.A., Lehman, A.A., Fischer, G., Longley, R.E., Alexander, S.S., King, M.M. and McCay, P.B. (1979) Cancer Res. 39, 3458-3462. 15 Erickson, K.L., McNeill, C.J., Gershwin, M.E. and Ossmann, J.B. (1980) J. Nutr. 110, 1555-1572. 16 Levy, J.A., Ibrahim, A.B., Shirai, T., Ohta, K., Nagasawa, R., Yoshida, H., Estes, J. and Gardner, M. (1982) Proc. Natl. Acad. Sci. USA 79, 1974-1978. 17 Locniskar, M., Nauss, K.M. and Newberne, P.M. 11983) J. Nutr. 113, 951-961. 18 Balint, J.A., Kyriakides, E.C. and Beeler, D.A. (1980) J. Lipid Res. 21,868-873. 19 Stern, W. and Pullman, M.E. (1983) J. Biol. Chem. 258, 1112811135. 20 Clarke, S.D., Armstrong, M.K. and Jump, D.B. (1990) J. Nutr. 120, 225-231. 21 Armstrong, M.K., Blake, W.L. and Clarke, S.D. (1991) Biochem. Biophys. Res. Comm. 177, 1056-1061. 22 Buttke, T.M. and Ingram, L.O. (1978) Biochemistry 17, 52825286. 23 Buttke, T.M., Van Cleave, S., Steelman, L. and McCubrey, J.A. 11989) Proc. Natl. Acad. Sci. USA 86, 6133-6137. 24 Markwell, M.A.K., Haas, S.M., Tolbert, N.E. and Bieber, L.L. (1981) Methods Enzymol. 72, 296-303. 25 Chang, T.-Y. and Vagelos, P.R. (1976) Proc. Natl. Acad. Sci. USA 73, 24-28.

26 Strittmatter, P., Thiede, M.A., Hackett, C.S. and Ozols, J. (1988) J. Biol. Chem. 263, 2532-2535. 27 Kaestner, K.H., McLenithan, J.C., Christy, R., Braiteiman, L.T., Cornelius, P., Pekala, P.H. and Lane, M.D. (1989) Proc. Natl. Acad. Sci. USA 86, 3150-3154. 28 Chirgwin, J.M., Przbyla, A.E., MacDonald, R.J. and Rutter, W.J. 11979) Biochemistry 18, 5294-5299. 29 Maniatis, T., Fritsch, E.F. and Sambrook, J. 11982) Molecular cloning: a laboratory manual, Cold Spring Harbor Laboratory, Cold Spring Harbor. 30 Feinberg, A.P. and Vogelstein, B. (1983) Anal. Biochem. 132, 6-13. 31 Cornelius, P., Marlowe, M., Lee, M.D. and Pekala, P.H. (1990) J. Biol. Chem. 265, 20506-20516. 32 Hayashi, Y., Urade, R. and Kito, M. (1987) Biochim. Biophys. Acta 918, 267-273. 33 Lagarde, M., Sicard, B., Guihardant, Felisi, O. and Dechavanne, M. (1984) In Vitro 20, 33-37. 34 Oshino, N. and Sato, R. (1972) Arch. Biochem. Biophys. 149, 369-377. 35 Prasad, M.R. and Joshi, V.C. (1979) J. Biol. Chem. 254, 63626369. 36 Bossie, M.A. and Martin, C.E. (1989)J. Bacteriol. 171, 6409-6413. 37 Hwang, S.-Y., Stukey, J. and Martin, C.E. (1991) FASEB J. 5, All60. 38 Ntambi, J.M., Buhrow, S.A., Kaestner, K.H., Christy, R.J., Sibley, E., Kelly, T.J. and Lane, M.D. (1988) J. Biol. Chem. 263, 1729117300. 39 Kaestner, K.H., Ntambi, J.M., Kelly, T.J. and Lane, M.D. 11989) J. Biol. Chem. 264, 14755-14761. 40 Goldstein, J.L. and Brown, M.S. (1990) Nature 343, 425-430. 41 Osborne, T.F., Gil, G., Goldstein, J.L. and Browm, M.S. (1988) J. Biol. Chem. 263, 3380-3387. 42 Das, U.N. (1991) Prostaglandins Leukot. Essent. Fatty Acids 42, 241-244. 43 Schulam, P.G. and Shearer, W.T. (1990) J. Immunol. 144, 26962701. 44 Ambrus, J.L., Jurgensen, C.H., Witzel, N.L., Lewis, R.A., Butler, J.L. and Fauci, A.S. (1988) J. Immunol. 140, 2382-2388. 45 Abraham, R.T., McKinney, M.M., Forray, C., Shipley, G.D. and Handwerg, B.S. (1986) J. lmmunopharm. 138, 165-204. 46 Sellmayer, A., Uedelhoven, W.M., Weber, P.C. and Bonventre, J.V. 11991) J. Biol. Chem. 266, 3800-3807. 47 Johnson, H.M., Russell, J.K. and Tortes, B.A. (1986) J. lmmunol. 137, 3053-3056. 48 Clarke, S.D. and Armstrong, M.K. (1989) FASEB J. 3, 2480-2487. 49 Evans, R.M. (1989) Science 240, 889-895. 50 Tebbey, P.W. and Buttke, T.M. (1992) Biochem. Biophys. Res. Commun. 186, 531-536. 51 Tebbey, P.W. and Buttke, T.M. 11992) FASEB J. 6, A1935.