- Email: [email protected]
Studies on cholesterol esterase in rat adipose tissue: Comparison of substrates and regulation of the activity
320
Biochimica et Biophystca Acta, 963 (1988) 320-328 Elsevier
BBA 52990
Studies on cholesterol esterase in rat adipose tissue: comparison of substrates and regulation of the activity Kazuki Nakamura,
Yasuhide
University
Inoue, Nobue Watanabe
of ShrruokoSchool
of Pharmuceutical
(Received
Key words:
Cholesterol
esterase;
and Takako
Sciences, Shiruoka
Tomita
*
(Japan)
12 July 1988)
cyclic AMP dependent protein kinase; Liposomal Epididymal adipose tissue; (Rat)
substrate;
Micellar
substrate;
Efficiency of substrates for cholesterol esterase (EC 3.1.1.13) assay, and regulation of the activity were investigated in rat epididymal adipose tissue. The activity in the supematant was activated by cyclic AMP-dependent protein kinase, cyclic AMP, ATP and Mg *+ , both with micellar and liposomal substrates. However, the mice&r substrate was more suitable for the assay than the liposomaf with respect to V,, and K,. Thus, the micellar substrate was employed. Pretreatment of the supematant with exogenous cyclic AMP-dependent protein kinase enhanced the activity dose dependently, whereas that with cyclic AMP decreased the activity slightly. The cyclic AMP-dependent protein kinase activity in the assay mixture was within the range which can cause changes in cholesterol esterase activity. These results suggest that the amount of cyclic AMP-dependent protein kinase, rather than the cyclic AMP level, plays an important role in the regulation of cholesterol esterase in tissues with a high cholesterol esterase activity relative to the kinase activity, such as in adipose tissue.
Introduction The presence of cholesterol esterase in adipose tissue was first demonstrated by Arnaud and Boyer [l]. This enzyme activity with a neutral pH optimum in rat adipose tissue was shown to be activated either by preincubation with cyclic AMP, ATP and Mg2+, or by pretreatment of the tissue with agents increasing the cyclic AMP level [2]. Activable cholesterol esterase has been also demonstrated in the adrenal cortex [3], corpus luteum [4], testes [5], aorta [6] and cardiac muscle [7]. Cholesterol esterase is involved in lipoprotein metabolism and syntheses of steroid hormones and bile acids in these tissues, including the liver.
Correspondence: Pharmaceutical
T. Tomita, University of Shizuoka School of Sciences. Oshika, Shizuoka 422, Japan.
0005.2760/88/$03.50
0 1988 Elsevier Science Publishers
For assay of cholesterol esterase, various substrate preparations have been used by different investigators. Pittman et al. [2] assayed cholesterol esterase in the adipose tissue using a cholesterol oleate emulsion stabilized with gum arabic or ethanol. Severson and Fletcher [8] prepared cholesterol oleate substrate dispersed with phosphatidylcholine in glycerol for aortic neutral cholesterol esterase assay. We employed a liposomal substrate for study on rat adipose tissue cholesterol esterase in a previous paper [9]. Later, Hajjar et al. [6] reported that a micellar substrate (cholesterol oleate/ phosphatidylcholine/ sodium taurocholate) was a better substrate for neutral cholesterol esterase in rabbit aortas. The present study was undertaken to compare the efficiency of the micellar substrate with that of the liposomal substrate for cholesterol esterase assay and to investigate the regulation of the
B.V. (Biomedical
Division)
321
activity in rat adipose tissue. The micellar substrate revealed a lower K, and a larger V,, than the liposomal substrate. Using a sensitive method with our micellar substrate, it was found that cyclic AMP-dependent protein kinase activity was a more important regulatory factor than the cyclic AMP level for cholesterol esterase activity in rat adipose tissue. Furthermore, the cyclic AMP-dependent protein kinase activity in the supematant was comparable to the activity required for a regulatory role in cholesterol esterase activity. Materials and Methods Materials. Cholesteryl [l-‘4C]oleate (56.6 mCi/mmol) and [y- 32P]ATP (6.45 nCi/nmol) were obtained from New England Nuclear (Boston, MA, U.S.A.); unlabelled cholesteryl oleate was from Tokyo Chemical Industries (Tokyo, Japan). ATP .2NA, adenosine 3’,5’-cyclic monophosphate (free acid), cyclic AMP-dependent protein kinase (crude lyophilized powder from bovine heart), Hl-histone from calf thymus and 2’,3’-dideoxyadenosine were from Sigma (St. Louis, MO, U.S.A.) and bovine serum albumin (fatty acid-free) was from Miles (Elkhart, IN, U.S.A.). Preparation for enzyme solution. The epididymal adipose tissue was excised from male rats of the Wistar strain at 9-15 weeks of age after a 15 h fast and diced. The tissue was homogenized with Polytron (PT 10, Kinematica, Luzem, Switzerland) in 5 vol. 50 mM Tris-HCl buffer (pH 7) containing 250 mM sucrose and 5 PM EDTA .2Na. The homogenate was consecutively centrifuged at 1500 x g for 10 min and at 43000 x g for 15 min at 4OC. The clear supematant (43000 X g) was submitted to a cholesterol esterase assay. Preparation of substrate solutions. Both labelled and unlabelled cholesteryl oleate were purified by thin-layer chromatography (developing solvent: petroleum ether/ ether/ acetic acid, 70 : 30 : 1) prior to use. Micellar substrate solution was prepared principally according to Hajjar et al. [6], but by a slightly modified method: cholesteryl oleate (0.244 pmol) containing 2.5 PCi cholesteryl [li4C]oleate (56.6 mCi/mmol), a mixture (0.95 pmol) of egg phosphatidylcholine and phosphatidylethanolamine (1 : l), and sodium taurocholate (0.5 pmol) were lyophilized for 4 h and then
sonicated (50 watts) at 46 o C for 10 min in 2 ml of 100 mM potassium phosphate buffer (pH 7.0). Liposomal substrate solution was prepared as described [9], originally according to Brecher et al. [lo]. Cholesteryl oleate (0.353 pmol) contains 2.6 PCi cholesterol [1-i4C]oleate (56.6 mCi/mmol), a mixture (20 mg) of egg phosphatidylcholine and phosphatidylethanolamine (1 : 1) were lyophilized for 4 h and then sonicated (40 watts) at 52°C for 10 min in 2 ml of 10 mM Tris-HCl buffer contains 100 mM NaCl and 0.03% NaN, (pH 7.4). Standard assay of cholesterol esterase. Cholesterol esterase activity was measured in terms of release of [l-‘4C]oleic acid from cholesteryl [l“C]oleate. The supematant (approx. 15 pg protein/10 ~1) of adipose tissue and the substrate solution (approx. 3. lo4 dpm/1.22 nmol/lO ~1) were incubated at 37’C for 10 min in 100 mM potassium phosphate buffer (pH 6) containing 0.025% bovine serum albumin (total vol. 200 ~1). The reaction was terminated by the addition of 0.3 M NaOH (0.6 ml) and a mixture (3 ml) of benzene/ CHCl,/ MeOH (1 : 0.5 : 1.2). The reaction tubes were shaken vigorously for 1 min and centrifuged (1500 x g for 20 min at 10 o C). Radioactivity of the upper layer was measured by a Liquid Scintillation Spectrometer (Aloka LSC-602, Tokyo, Japan). Assay of cyclic AMP-dependent protein kinase. Cyclic AMP-dependent protein kinase activity in the supematant (43000 x g) of the adipose tissue was assayed by incubating for 5 min the supernatant (20 ~1) 100 pg Hl-histone, 1 FM cyclic AMP, 2.5 mM magnesium acetate and 25 PM ATP containing [Y-~*P]ATP (6.45 nCi/nmol) in 25 mM Tris-HCl buffer (pH 7.5, total reaction volume, 0.2 ml) at 3O’C. The reaction was terminated with 4 ml of 10% trichloroacetic acid; the mixture was filtered through a membrane filter (size: 0.45 pm, diameter: 25 mm, Advantic Toyo, Tokyo, Japan), and a rinsed membrane filter was counted in a scintillation cocktail (toluene/Triton X-100/ EtOH, 8 : 4 : 3). Assay of phosphoprotein phosphatase. 32P-prelabelled Hl-histone was prepared by incubating for 3 h 20 mg Hl-histone from calf thymus, 0.6 mM magnesium acetate, 3.6 mg cyclic AMP-dependent protein kinase from bovine heart, 3 nM cyclic AMP and 40 PM ATP containing [y-
322
32P]ATP (6.45 nCi/nmol) in 20 ml Tris-HCl buffer (pH 7.5, total volume, 30 ml) at 30°C. Phosphoprotein phosphatase activity was measured in terms of release of 32P from 32P-labelled Hl-histone. Phosphoprotein phosphatase activity in the supernatant (43000 x g) of adipose tissue was assayed by incubating for 10 min the supematant (20 ~1) and 90 pg 32P-labelled Hl-histone in 50 mM Tris-HCl buffer containing 0.2 M NaCl and 0.5 mM dithiothreitol (pH 7.2, total reaction volume, 0.1 ml) at 30°C. The reaction was terminated by the addition of 1 ml of 2.5 mM H,SO, containing 5 mM SiO,. 12WO,, 0.25 ml of 2 M H,SO, containing 5% (NH,),MoO, and a mixture (1.5 ml) of isobutanol/benzene (1: 1). The reaction tubes were shaken vigorously for 20 s and centrifuged (1500 X g for 1 min at 4” C). Radioactivity of the upper layer was measured by a Liquid Scintillation Spectrometer. Results pH dependency of cholesterol esterase and effects of bovine serum albumin pH dependency curves of cholesterol esterase activity in the supematant of rat epididymal adipose tissue are shown in Fig. 1B (micellar substrate) and Fig. 2B (liposomal substrate). With the micellar substrate, cholesterol esterase activity was optimal at pH 7 in the absence of bovine serum albumin. An addition of bovine serum albumin (0.025%) to the reaction mixture increased the activity 2-3-fold, and shifted the pH optimum to 6. The enhancement of cholesterol esterase activity with bovine serum albumin observed in the experiment with micellar substrate will result from the removal of the reaction product. Fatty acids reportedly inhibit cholesterol esterase [ll]. In the case of the liposomal substrate, the pH optimum of cholesterol esterase was 6, both in the presence and absence of bovine serum albumin in the mixture. Addition of bovine serum albumin neither enhanced the activity nor shifted the pH optimum. Activation of cholesterol esterase activity by cyclic AMP-dependent protein kinase Activation of cholesterol esterase by cyclic AMP-dependent protein kinase was examined using micellar and liposomal substrates (Fig. 1A
PH
PH
Fig. 1. pH dependency and activation of cholesterol esterase (CEase) activity in rat epididymal adipose tissue (micellar substrate). (A) The supematant (10 ~‘1) of rat epididymal adipose tissue homogenate was preincubated at 3O’C for 10 min with cyclic AMP-dependent protein kinase from bovine heart (50 as), cyclic AMP (10 PM), ATP.2Na and magnesium acetate (1.25 mM): A, with cyclic AMP, ATP.2Na and magnesium acetate: o, without any addition; 0, in 100 mM sodium acetate buffer (pH 3.7-5.6) or 100 mM potassium phosphate buffer (pH 6.1-8.1) containing 0.025% bovine serum albumin. Assay of cholesterol esterase activity was started by addition of micellar substrate (10 JLI) to the mixture (200 pl). (B) The supematant (10 ~1) of rat epididymal adipose tissue homogenate was incubated for 10 min at 37 o C with micellar substrate in buffers with (open symbols) or without (solid symbols) an addition of 0.025% bovine serum albumin. Each point represents the mean of duplicate determination.
and Fig. 2A). The supernatant (10 ~1) was preincubated for 10 min with cyclic AMP-dependent protein kinase (50 pg), cyclic AMP (10 PM), ATP .2Na (0.5 mM) and magnesium acetate (1.25 mM) at 30 o C, and then cholesterol esterase activity was determined using the two substrates. The activity measured by the micellar substrate was enhanced at neutral pH ranges when the supematant was incubated with cyclic AMP-dependent protein kinase, cyclic AMP, ATP and magnesium acetate. However, deletion of cyclic AMP-dependent protein kinase from the preincubation medium either blocked the enhancement or even attenuated the activity slightly at pH 5.5 and 6.0 compared with the activity preincubated without any addition. This decrease may result from effects of phosphoprotein phosphatase contaminating the supernatant. No effects of the preincubation were observed at other pH ranges unless cyclic AMP-dependent protein kinase was incubated in the preincubation mixture. The activity assayed with
7
-i PH
PH
Fig. 2. pH dependency and activation of cholesterol esterase (CEase) activity in rat epididymal adipose tissue (liposomal substrate). (A) The supernatant (10 ~1) of rat epididymal adipose tissue homogenate was preincubated at 30°C for 10 min with cyclic AMP-dependent protein kinase from bovine heart (50 gg), cyclic AMP (10 PM), ATP.2Na (0.5 mM) and magnesium acetate (1.25 mM): A, with cyclic AMP, ATP’2NA and magnesium acetate; 0, without any addition; 0, in 100 mM sodium acetate buffer (pH 3.7-5.6) or 100 mM potassium phosphate buffer (pH 6.1-8.1) containing 0.025% bovine serum albumin. Assay of cholesterol esterase activity was started by addition of liposomal substrate (10 ~1) to the mixture (200 ~1). (B) The supematant (10 ~1) of rat epididymal adipose tissue homogenate was incubated for 10 min at 37 o C with liposomal substrate in buffers with (open symbols) or without (solid symbols) an addition of 0.025% bovine serum albumin. Each point represents the mean of duplicate determination. (A) and (B) are separate experiments with epididymal adipose tissue from different donors.
liposome was also enhanced by preincubation with cyclic AMP-dependent protein kinase and other cofactors at pH ranges above 6. Unlike the case of micellar substrate, the removal of cyclic AMP-dependent protein kinase from the mixture did not cause an attenuation of the activity, but it slightly increased the activity at pH 6.5 and 7. These results indicate that addition of cyclic AMP-dependent protein kinase was essential for the activation of cholesterol esterase in the adipose tissue supematant by this assay system. Substrate concentration curves of cholesterol esterase activity with micellar and liposomal substrates Fig. 3 shows substrate concentration curves of cholesterol esterase activity in rat epididymal adipose tissue with micellar and liposomal substrates. The inset shows a double-reciprocal plot
of the data. When micellar substrate was used, the apparent K, 7.9 (PM), and the apparent I$,, 45.0 (nmol fatty acid released/mg protein per h). In contrast, the liposomal substrate gave a higher K, (14.7) and lower V,,, (39.4) values than the micellar substrate. Protein dependency and time-course of cholesterol esterase activity with the micellar substrate As the micellar substrate appeared to be more suitable for measuring neutral cholesterol esterase than the liposomal substrate, a protein-dependency curve and time-course of cholesterol esterase were depicted (Fig. 4). The activity was linear up to 20 pg enzyme protein per tube when the supernatant was incubated for 10 min with 10 ~1 (2.78 . lo4 dpm/1.22 nmol) substrate in 100 mM potassium phosphate buffer (pH 6) at 37°C. The timecourse of cholesterol esterase was linear up to 10 min when 10 ~1 (14.2 pg protein) of the supernatant was incubated with 10 ~1 substrate. This assay with our micellar substrate is much more sensitive than other assays for cholesterol esterase.
Fig. 3. Substrate concentration curves of cholesterol esterase (CEase) activity from the epididymal adipose tissue. The supernatant (20 ~1) of the adipose tissue homogenate was incubated at 37 .aC either with liposomal substrate (0) or with micellar substrate (0) (l-50 pl) in 100 mM potassium phosphate buffer (pH 6.0) containing 0.025% bovine serum albumin for 10 mm (total 200 ~1). Inset shows double-reciprocal plot of the data. Micellar substrate: K, = 7.9 (PM); V,, = 45.0 (nmol fatty acid released/mg protein per h). Liposomal substrate: K, = 14.7 (PM), V,, = 39.4 (nmol fatty acid released/mg protein per h). Each point represents the mean of duplicate determination.
324
the high activity adipose tissue.
of cholesterol
esterase
in
the
Effects of adrenalin, theophylline, potassium fluoride and 2’,3’-dideoxyadenosine on cholesterol esterase activity To confirm the effect of cyclic AMP on cholesterol esterase activity, reagents known to
Fig. 4. Protein dependency and time-course of cholesterol esterase (CEase) activity in rat epididymal adipose tissue. The supernatant of rat epididymal adipose tissue homogenate was incubated at 37 o C with the micellar substrate (10 ~1) containing cholesterol [‘4C]oleate (2.78.104 dpm/1.22 nmol) in 100 mM potassium phosphate buffer (pH 6.0) containing 0.025% bovine serum albumin. (A) Incubated for 10 min: (B) 10 ~1 supernatant (14.2 pg protein) used. Each point represents the mean of duplicate determination.
SC
_I 0
/
40
/e
30
Effects of exogenous cyclic AMP-dependent protein kinase and cyclic AMP on cholesterol esterase activity The results in Fig. 1 indicate that in this cholesterol esterase assay system, an addition of cyclic AMP, ATP and Mg2+ appeared not to be essential for the activation of cholesterol esterase, but an addition of exogenous cyclic AMP-dependent protein kinase from bovine heart was essential for the enhancement of activity. Thus, the effect of exogenous cyclic AMP-dependent protein kinase was examined in detail. Fig. 5 clearly shows that preincubation of the adipose tissue supernatant with varying amounts of cyclic AMP-dependent protein kinase increased cholesterol esterase activity in a dose-dependent manner. A 64% increase in the activity was observed by preincubation with 50 pg cyclic AMP-dependent protein kinase. In contrast, preincubation with over 4 PM cyclic AMP alone slightly decreased the activity. These results suggest that cyclic AMP-dependent protein kinase rather than cyclic AMP is one of the important regulatory factors for cholesterol esterase in this assay with the rat adipose tissue supematant. This effect of cyclic AMP-dependent protein kinase might have become evident because our reaction mixture did not contain enough endogenous cyclic AMP-dependent protein kinase due to the very sensitive assay method, and also
20
ia
a
-10
-20
a
cyclic L 0
20
la AMP
30
dependent
Proteln
4
2
cyclic
6 AMP
40
50
kinarehg) 3
10
h8M)
Fig. 5. Effects of exogenous cyclic AMP-dependent protein kinase or cyclic AMP on cholesterol esterase (CEase) activity in the supematant. The supematant (5 ~1) of rat epididymal adipose tissue was preincubated at 30 Oc in 100 mM potassium phosphate buffer (pH 6.0) containing 0.025% bovine serum albumin in the presence and absence of varying amounts of cyclic AMP-dependent protein kinase from bovine heart (solid symbols) or cyclic AMP (open symbols) for 10 min. Assay of cholesterol esterase activity was started by an addition of mice&u substrate (10 ~1) to the above mixture (200 ~1). Each point represents the mean of duplicate determination. Cholesterol esterase activity in the absence of cyclic AMP-dependent protein kinase and cyclic AMP was 27.68 (nmol fatty acid released/mg protein per h).
325 TABLE
I
EFFECTS OF ACTIVATORS OR INHIBITORS OF ADENYLATE CYCLASE ON CHOLESTEROL ESTERASE ACTIVITY A piece of the epididymal adipose tissue (0.2 g) was preincubated at 37’C in 1 ml Krebs Henseleit buffer (pH 7.4) in the presence or absence of effecters for 30 min. The washed fat pad was then homogenized and the supematant was assayed for cholesterol esterase in 100 mM potassium phosphate buffer (pH 7.0) without bovine serum albumin. Adipose tissues in each experiment are from different donors. Values are mean of duplicate experiments. Preincubated
with
Expt. 1 None Adrenalin (10 PM) Theophylline (1 mM) 2’,3’-Dideoxyadenosine (0.5 mM) Expt. 2 None Adrenalin (10 PM) + theophylline (1 mM) Expt. 3 None GTP (10 FM) GTP (10 pM)+potassium fluoride (10 mM) GTP (10 PM) + isoproterenol (10 PM)
Cholesterol esterase activity (nmol fatty acid released/mg protein per h)
3.27 3.90 2.83 4.59
9.17 10.09
10.62 11.26 11.32 9.31
affect adenylate cyclase were applied to change the cyclic AMP level in the preincubation (Table I). Fat pads (approx. 0.2 g) were incubated with adrenalin (10 PM) in the presence or absence of theophylline (1 mM) at 37°C for 30 min and then the supernatant of the homogenate was prepared as well. There was no enhancement in cholesterol esterase activity by adrenalin although lipolysis was almost maximally increased. The preincubation of the supernatant of adipose tissue with potassium fluoride (10 mM), which activated adenylate cyclase, did not alter the cholesterol esterase activity. 2’,3’-Dideoxyadenosine (0.5 mM), an inhibitor of adenylate cyclase, increased the activity slightly. This result suggests that cyclic
AMP is not a regulatory factor for cholesterol esterase activity in the adipose tissue. Cyclic AMP may play such a role in tissues containing relatively a high activity of the kinase to cholesterol esterase activity. Cyclic AMP-dependent protein kinase and phosphoprotein phosphatase activities in the supernatant Since cyclic AMP-dependent protein kinase appeared to be one of the important regulatory factors for cholesterol esterase activity in the supernatant of adipose tissue, cyclic AMP-dependent protein kinase and phosphoprotein phosphatase activities were assayed in the supernatant. Cyclic AMP-dependent protein kinase activity assayed by the incorporation of [Y-‘~P]ATP into HI-histone was 0.4-3.0 (nmol 32P incorporated/mg protein per min). Since one assay tube contains approx. 15 pg protein, cyclic AMPdependent protein kinase activity in the reaction mixture will be 6-45 pmol/min. This activity is comparable to the amounts of protein kinase added in the experiment shown in Fig. 5. The activity of protein kinase used is 1.2 pmol/pg protein per min. Phosphoprotein phosphatase activity (pmol “P released/mg protein per min) expressed in the term of 32P released from 32P-labeled Hl-histone was 37 5 3.6 [lo] (mean f S.D. (n)). Thus, the level of cyclic AMP-dependent protein kinase activity was one order of magnitude higher than that of the phosphoprotein phosphatase activity in the supernatant. Effects of preincubation with Mg’+ and EDTA Both activation by cyclic AMP-dependent protein kinase and deactivation by phosphoprotein phosphatase of cholesterol esterase need the presrequires ence of Mg ‘+ in the mixture. Deactivation higher concentrations of Mg2+ than activation. Preincubation of the supernatant for 10 min at 32” C with varying concentrations of MgZt (1.25-40 mM) decreased the cholesterol esterase activity. The decrease reached a maximum at the concentration of 20 mM. Fig. 6 shows the timecourse of deactivation of cholesterol esterase activity by preincubation with 5 mM Mg’+. Deactivation of cholesterol esterase proceeded rapidly; the activity (nmol fatty acid released/mg protein per
326
h) decreased from 44.0 to 21.7 in 5 min, after which no significant change was observed. The supernatant was preincubated for 10 min with 1 mM EDTA or 20 mM Mg2+, and then an aliquot of the supernatant was incubated in the presence or absence of cyclic AMP-dependent protein kinase, cyclic AMP and ATP (Table II). Cholesterol esterase activity was enhanced 74% by preincubation with EDTA and reduced 52% by Mg2+ preincubation compared to control. The activity which was raised by EDTA was further enhanced over 2-fold after an incubation in the presence of cofactors. The concentration of EDTA in the second incubation medium of EDTA-preincubated samples was 50 PM. It was assumed that EDTA inhibited phosphoprotein phosphatase, so that deactivation of cholesterol esterase was prevented, while Mg2+ activated the phosphoprotein phosphatase, which in turn deactivated cholesterol esterase. Thus, the endogenous (partially phosphorylated) activity of cholesterol esterase would be 11.5 (nmol fatty acid released/mg protein per h), and the basal (dephosphorylated) activity, ap-
TABLE II CHANGES IN CHOLESTEROL ESTERASE ACTIVITY OF THE SUPERNATANT BY PREINCUBATION WITH EDTA AND Mg’+ The supematant of adipose tissue homogenate was preincubated at 32’C for 10 min with EDTA’2Na (1 mM) and Mg2+ (20 mM), and then 10 ~1 of the supernatant was incubated in the presence or absence of cyclic AMP (10 pM), magnesium acetate (1.25 mM), ATP.2Na (0.5 mM) and cyclic AMP-dependent protein kinase from bovine heart (50 pg) in 200 pl of 100 mM potassium phosphate buffer (pH 6.0) containing 0.025% bovine serum albumin for another 10 min. Assay of cholesterol esterase was started by addition of the micellar substrate (10 al) to the mixture (200 ~1). Values are means of triplicate determination. Preincubation
None EDTA (1 mM) Mg*+ (20 mM)
Cholesterol esterase activity (nmol fatty acid released/ mg protein per h) Additions:
cyclic AMP, MgAc, ATP. ZNA, protein kinase
no
additions
+ additions
6.6 15.1 (+ 128%) b 11.5 (+ 74%) a 26.3 (+ 129%) b 3.2(-52%)” 5.7 (+79%) b
a % change in the presence of EDTA or Mg2+, relative to their absence. ’ % change in the presence of cyclic AMP-dependent protein kinase relative to respective value (left column) in its absence.
L 0
5
. 10
Praincubation
20
timetmtnl
Fig. 6. Time-course of deactivation of cholesterol esterase (CEase) from rat epididymal adipose tissue. The supematant (10 ~1) of adipose tissue homogenate was preincubated at 32OC for various periods in 100 mM potassium phosphate buffer (pH 6.0) containing 0.025% bovine serum albumin and 5 mM magnesium acetate. Assay of cholesterol esterase was started by an addition of the micellar substrate (10 ~1) to the above mixture (200 PI). Each point represents the mean of duplicate determination.
prox. 3.2, and the full (fully phosphorylated) ity, 26.3.
activ-
Discussion Two types of cholesterol esterase were demonstrated in the aorta and in the liver: acid cholesterol esterase in the lysosome and neutral cholesterol esterase in the cytosol fraction. Acid cholesterol esterase activity can be efficiently revealed by liposomal [lo] or phospholipid-digitonin-dispersed substrates [12] containing cholesteryl oleate. Several substrates for neutral cholesterol esterase have been reported: acetone substrate [13], ethanol or gum arabic-stabilized substrate [2], phosphatidylcholine-glycerol suspension [ 81, a micellar substrate [6] and a liposomal substrate [lo]. Using a micellar substrate containing phosphatidylcho-
327
line/ sodium taurocholate/ cholesteryl oleate (4: 2: l), Hajjar et al. [6] have recently demonstrated the presence of a neutral cholesterol esterase in rabbit aortas which was activated by cyclic AMP. Although our previous studies [9] on adipose tissue cholesterol esterase have been carried out with a liposomal substrate, the micellar substrate is assumed to be suited to assay of adipose tissue cholesterol esterase. In this study, therefore, the efficiency of the liposomal and the micellar substrate was compared in an assay of adipose tissue cholesterol esterase. The micellar substrate gave a larger apparent V,, and smaller apparent K, than the liposomal substrate for cholesterol esterase in the supematant of rat adipose tissue homogenate. Pittman et al. [2] had reported that the supernatant of the epididymal adipose tissue homogenate from Sprague-Dawley rats had high cholesterol esterase activity which was activated by cyclic AMP/ATP. Mg. Pretreatment of the rats with glucagon or other hormones increasing the cyclic AMP level was found to activate cholesterol esterase activity. Similar cholesterol esterase activity in macrophages [14], adrenocortical cells [3] and arterial smooth muscle cells [6] was activated by addition of cyclic AMP and ATP . Mg or agents to increase the cyclic AMP level (adrenalin, theophylline). These findings suggested that cholesterol esterase activity was regulated by the cyclic AMP level in the cells. None of these experiments regarded cyclic AMP-dependent protein kinase activity as an important regulatory factor. In experiments with adipocytes of SpragueDawley rats [15], lipolysis rates correlated with cyclic AMP-dependent protein kinase activity ratios (cyclic AMP-dependent protein kinase activity in the absence or in the presence of cyclic AMP) below 0.35 of the ratios. As the cyclic AMP-dependent protein kinase activity ratio rises above that level, the rate of dephosphorylation of the lipase by the phosphoprotein phosphatase exceeds the rate of phosphorylation, so that the lipolysis ratio remains constant. The cyclic AMP concentration seems to be a rate-limiting factor for lipolysis only within limited ranges. In our experiments, cholesterol esterase activity in the supematant of the epididymal adipose tis-
sue from the Wistar strain was assayed by a micellar substrate. The assay system employed here with our micellar substrate provided a sensitive method for the measurement of cholesterol esterase, and approx. 15 pg protein of the supernatant (43000 x g) was enough for an assay. The cholesterol esterase activity was not activated by preincubation with cyclic AMP/ATP * Mg, while it was enhanced dose-dependently by preincubation with exogenous cyclic AMP-dependent protein kinase of bovine heart (see Fig. 5), Pretreatments of adipose tissue with adrenalin/ theophylline, isoproterenol, potassium fluoride or 2’,3’-dideoxyadenosine did not influence cholesterol esterase activity although the same treatment affected lipolysis activity in the adipose tissue or the adipocytes. These results imply that cyclic AMP-dependent protein kinase activity rather than cyclic AMP is an important regulator of cholesterol esterase activity in this system. This hypothesis is further supported by the fact that cyclic AMP-dependent kinase activity in the assay mixture, measured by incorporation of [y- 32P]ATP into histone (see Results) was in the range comparable to the activity exogenously added (Fig. 5). One of the reasons that the regulatory role of cyclic AMP-dependent protein kinase on cholesterol esterase became evident in our system will be because only a small amount of the enzyme protein was used for the assay due to the high activity in the adipose tissue and due to our sensitive method; thus, there was not enough cyclic AMP-dependent protein kinase present in the mixture. If we used more enzyme protein, the effect of the kinase would be attenuated, or not apparent. In the report by Pittman et al. [2], who showed the effect of cyclic AMP on cholesterol esterase, more than lo-fold the enzyme protein (supematant of 100000 x g) that we used in the supematant of 43000 X g was employed. Cholesterol esterase activity in aorta, adrenal cortex, liver, myocardiac cells and testes is not as high as that in adipose tissue. The enzyme preparations from these tissues contain a relatively high activity of cyclic AMP-dependent protein kinase compared to that of cholesterol esterase. In these tissues, cyclic AMP may play a regulatory role for cholesterol esterase, as Hajjar et al. [6] showed in rabbit aorta.
328
It was reported that some pituitary stimulating hormones induce the change in cyclic AMP-dependent protein kinase activity in organs where steroid hormones are produced; in the adrenal cortex, adrenocorticotropic hormone activates cyclic AMP-dependent protein kinase by increasing the catalytic unit [16], and in rat ovary, the combined action of estradiol and follicle-stimulating hormone regulates cyclic AMP-dependent protein kinase activity through the increased transcription of the gene of the regulatory unit (RII,,) [17]. Gonadotropin-induced steroidgenesis in testicular interstitial cells is also intermediated by cyclic AMP-dependent protein kinase [5]. We have found recently that treatment of rats of the Wistar strain with estradiol enhanced markedly cholesterol esterase activity in the adipose tissue, with a concomitant change in cyclic AMP-dependent protein kinase activity. Ovariectomy resulted in an attenuation of cholesterol esterase and cyclic AMP-dependent protein kinase activities in the parametrial adipose tissue. There was no change in the cyclic AMP level in the tissue [18]. Enhancement in testicular cholesterol esterase by follicle-stimulating hormone and luteinizing hormone might be mediated through a similar mechanism [9]. Thus, we conclude that cholesterol esterase activity in the supematant of adipose tissue homogenate from the Wistar rat strain is regulated primarily by cyclic AMP-dependent protein kinase activity rather than by cyclic AMP levels. Acknowledgements This investigation has been supported in part by a grant-in-aid for scientific research from the Ministry of Japan (No. 60571051,1985-1986) and
by a medical research grant from the Ministry of Education of Japanese Health and Welfare (198331985). References 1 Amaud, J. and Boyer, J. (1974) B&him. Biophys. Acta 337, 165-168. 2 Pittman, R.C., Khoo, J.C. and Steinberg, D. (1975) J. Biol. Chem. 250, 4505-4511. 3 Trzeciak, W.H., Mason, J.I. and Boyd, G.S. (1979) FEBS Lett. 102, 13-17. 4 Bisgaier, CL., Treadwell, C.R. and Vahouny, G.V. (1979) Lipids 14, 1-14. 5 Bailey, M.L. and Grogan, W.M. (1986) J. Biol. Chem. 261, 7717-7722. 6 Hajjar, D.P., Minick, C.R. and Fowler, S. (1983) J. Biol. Chem. 258, 192-198. 7 Goldberg, D.I. and Khoo, J.C. (1985) J. Biol. Chem. 260, 5879-5882. 8 Severson, D.L. and Fletcher, T. (1978) Atherosclerosis 31, 21-32. 9 Tomita, T., Yonekura, I., Umegaki, K., Okada, T. and Hayashi, E. (1984) Hot-m. Metabol. Res. 16, 461-464. 10 Brecher, R., Chobanian, J., Small, D.M. and Chobanian, A.V. (1976) J. Lipid Res. 17, 239-247. A.V. 11 Brecher, P., Kessler, M., Clifford, C. and Chobanian, (1973) B&him. Biophys. Acta 316, 386-394. 12 Haley, N.J., Fowler, S. and De Duve, C. (1980) J. Lipid Res. 21, 961-969. A.V. (1977) J. 13 Brecher, P., Pyun, H.Y. and Chobanian, Lipid Res. 18, 154-163. 14 Khoo, J.C., Mahoney, E.M. and Steinberg, D. (1981) J. Biol. Chem. 256, 12659-12661. 15 Honnor, R.C., Dhillon, G.S. and Londos, C. (1985) J. Biol. Chem. 260, 15130-15138. C. (1987) Endo16 Murray, S.A., Champ, C. and Lagenauer, crinology 120, 1921-1927. G.S., Lifka, J., Durica, J.M. and 17 Hedin, L., Mcknight, Richards, J.S. (1987) Endocrinology 120, 192881935. K., Inoue, Y., Ikeda, M., Talceshita, N. and 18 Nakamura, Tomita, T. (1988) J. Jpn. Atheroscler. Sot. 16, 349-351. 19 Durham, L.A., III and Grogan, W.M. (1984) J. Biol. Chem. 259, 7433-7438.
Biochimica et Biophystca Acta, 963 (1988) 320-328 Elsevier
BBA 52990
Studies on cholesterol esterase in rat adipose tissue: comparison of substrates and regulation of the activity Kazuki Nakamura,
Yasuhide
University
Inoue, Nobue Watanabe
of ShrruokoSchool
of Pharmuceutical
(Received
Key words:
Cholesterol
esterase;
and Takako
Sciences, Shiruoka
Tomita
*
(Japan)
12 July 1988)
cyclic AMP dependent protein kinase; Liposomal Epididymal adipose tissue; (Rat)
substrate;
Micellar
substrate;
Efficiency of substrates for cholesterol esterase (EC 3.1.1.13) assay, and regulation of the activity were investigated in rat epididymal adipose tissue. The activity in the supematant was activated by cyclic AMP-dependent protein kinase, cyclic AMP, ATP and Mg *+ , both with micellar and liposomal substrates. However, the mice&r substrate was more suitable for the assay than the liposomaf with respect to V,, and K,. Thus, the micellar substrate was employed. Pretreatment of the supematant with exogenous cyclic AMP-dependent protein kinase enhanced the activity dose dependently, whereas that with cyclic AMP decreased the activity slightly. The cyclic AMP-dependent protein kinase activity in the assay mixture was within the range which can cause changes in cholesterol esterase activity. These results suggest that the amount of cyclic AMP-dependent protein kinase, rather than the cyclic AMP level, plays an important role in the regulation of cholesterol esterase in tissues with a high cholesterol esterase activity relative to the kinase activity, such as in adipose tissue.
Introduction The presence of cholesterol esterase in adipose tissue was first demonstrated by Arnaud and Boyer [l]. This enzyme activity with a neutral pH optimum in rat adipose tissue was shown to be activated either by preincubation with cyclic AMP, ATP and Mg2+, or by pretreatment of the tissue with agents increasing the cyclic AMP level [2]. Activable cholesterol esterase has been also demonstrated in the adrenal cortex [3], corpus luteum [4], testes [5], aorta [6] and cardiac muscle [7]. Cholesterol esterase is involved in lipoprotein metabolism and syntheses of steroid hormones and bile acids in these tissues, including the liver.
Correspondence: Pharmaceutical
T. Tomita, University of Shizuoka School of Sciences. Oshika, Shizuoka 422, Japan.
0005.2760/88/$03.50
0 1988 Elsevier Science Publishers
For assay of cholesterol esterase, various substrate preparations have been used by different investigators. Pittman et al. [2] assayed cholesterol esterase in the adipose tissue using a cholesterol oleate emulsion stabilized with gum arabic or ethanol. Severson and Fletcher [8] prepared cholesterol oleate substrate dispersed with phosphatidylcholine in glycerol for aortic neutral cholesterol esterase assay. We employed a liposomal substrate for study on rat adipose tissue cholesterol esterase in a previous paper [9]. Later, Hajjar et al. [6] reported that a micellar substrate (cholesterol oleate/ phosphatidylcholine/ sodium taurocholate) was a better substrate for neutral cholesterol esterase in rabbit aortas. The present study was undertaken to compare the efficiency of the micellar substrate with that of the liposomal substrate for cholesterol esterase assay and to investigate the regulation of the
B.V. (Biomedical
Division)
321
activity in rat adipose tissue. The micellar substrate revealed a lower K, and a larger V,, than the liposomal substrate. Using a sensitive method with our micellar substrate, it was found that cyclic AMP-dependent protein kinase activity was a more important regulatory factor than the cyclic AMP level for cholesterol esterase activity in rat adipose tissue. Furthermore, the cyclic AMP-dependent protein kinase activity in the supematant was comparable to the activity required for a regulatory role in cholesterol esterase activity. Materials and Methods Materials. Cholesteryl [l-‘4C]oleate (56.6 mCi/mmol) and [y- 32P]ATP (6.45 nCi/nmol) were obtained from New England Nuclear (Boston, MA, U.S.A.); unlabelled cholesteryl oleate was from Tokyo Chemical Industries (Tokyo, Japan). ATP .2NA, adenosine 3’,5’-cyclic monophosphate (free acid), cyclic AMP-dependent protein kinase (crude lyophilized powder from bovine heart), Hl-histone from calf thymus and 2’,3’-dideoxyadenosine were from Sigma (St. Louis, MO, U.S.A.) and bovine serum albumin (fatty acid-free) was from Miles (Elkhart, IN, U.S.A.). Preparation for enzyme solution. The epididymal adipose tissue was excised from male rats of the Wistar strain at 9-15 weeks of age after a 15 h fast and diced. The tissue was homogenized with Polytron (PT 10, Kinematica, Luzem, Switzerland) in 5 vol. 50 mM Tris-HCl buffer (pH 7) containing 250 mM sucrose and 5 PM EDTA .2Na. The homogenate was consecutively centrifuged at 1500 x g for 10 min and at 43000 x g for 15 min at 4OC. The clear supematant (43000 X g) was submitted to a cholesterol esterase assay. Preparation of substrate solutions. Both labelled and unlabelled cholesteryl oleate were purified by thin-layer chromatography (developing solvent: petroleum ether/ ether/ acetic acid, 70 : 30 : 1) prior to use. Micellar substrate solution was prepared principally according to Hajjar et al. [6], but by a slightly modified method: cholesteryl oleate (0.244 pmol) containing 2.5 PCi cholesteryl [li4C]oleate (56.6 mCi/mmol), a mixture (0.95 pmol) of egg phosphatidylcholine and phosphatidylethanolamine (1 : l), and sodium taurocholate (0.5 pmol) were lyophilized for 4 h and then
sonicated (50 watts) at 46 o C for 10 min in 2 ml of 100 mM potassium phosphate buffer (pH 7.0). Liposomal substrate solution was prepared as described [9], originally according to Brecher et al. [lo]. Cholesteryl oleate (0.353 pmol) contains 2.6 PCi cholesterol [1-i4C]oleate (56.6 mCi/mmol), a mixture (20 mg) of egg phosphatidylcholine and phosphatidylethanolamine (1 : 1) were lyophilized for 4 h and then sonicated (40 watts) at 52°C for 10 min in 2 ml of 10 mM Tris-HCl buffer contains 100 mM NaCl and 0.03% NaN, (pH 7.4). Standard assay of cholesterol esterase. Cholesterol esterase activity was measured in terms of release of [l-‘4C]oleic acid from cholesteryl [l“C]oleate. The supematant (approx. 15 pg protein/10 ~1) of adipose tissue and the substrate solution (approx. 3. lo4 dpm/1.22 nmol/lO ~1) were incubated at 37’C for 10 min in 100 mM potassium phosphate buffer (pH 6) containing 0.025% bovine serum albumin (total vol. 200 ~1). The reaction was terminated by the addition of 0.3 M NaOH (0.6 ml) and a mixture (3 ml) of benzene/ CHCl,/ MeOH (1 : 0.5 : 1.2). The reaction tubes were shaken vigorously for 1 min and centrifuged (1500 x g for 20 min at 10 o C). Radioactivity of the upper layer was measured by a Liquid Scintillation Spectrometer (Aloka LSC-602, Tokyo, Japan). Assay of cyclic AMP-dependent protein kinase. Cyclic AMP-dependent protein kinase activity in the supematant (43000 x g) of the adipose tissue was assayed by incubating for 5 min the supernatant (20 ~1) 100 pg Hl-histone, 1 FM cyclic AMP, 2.5 mM magnesium acetate and 25 PM ATP containing [Y-~*P]ATP (6.45 nCi/nmol) in 25 mM Tris-HCl buffer (pH 7.5, total reaction volume, 0.2 ml) at 3O’C. The reaction was terminated with 4 ml of 10% trichloroacetic acid; the mixture was filtered through a membrane filter (size: 0.45 pm, diameter: 25 mm, Advantic Toyo, Tokyo, Japan), and a rinsed membrane filter was counted in a scintillation cocktail (toluene/Triton X-100/ EtOH, 8 : 4 : 3). Assay of phosphoprotein phosphatase. 32P-prelabelled Hl-histone was prepared by incubating for 3 h 20 mg Hl-histone from calf thymus, 0.6 mM magnesium acetate, 3.6 mg cyclic AMP-dependent protein kinase from bovine heart, 3 nM cyclic AMP and 40 PM ATP containing [y-
322
32P]ATP (6.45 nCi/nmol) in 20 ml Tris-HCl buffer (pH 7.5, total volume, 30 ml) at 30°C. Phosphoprotein phosphatase activity was measured in terms of release of 32P from 32P-labelled Hl-histone. Phosphoprotein phosphatase activity in the supernatant (43000 x g) of adipose tissue was assayed by incubating for 10 min the supematant (20 ~1) and 90 pg 32P-labelled Hl-histone in 50 mM Tris-HCl buffer containing 0.2 M NaCl and 0.5 mM dithiothreitol (pH 7.2, total reaction volume, 0.1 ml) at 30°C. The reaction was terminated by the addition of 1 ml of 2.5 mM H,SO, containing 5 mM SiO,. 12WO,, 0.25 ml of 2 M H,SO, containing 5% (NH,),MoO, and a mixture (1.5 ml) of isobutanol/benzene (1: 1). The reaction tubes were shaken vigorously for 20 s and centrifuged (1500 X g for 1 min at 4” C). Radioactivity of the upper layer was measured by a Liquid Scintillation Spectrometer. Results pH dependency of cholesterol esterase and effects of bovine serum albumin pH dependency curves of cholesterol esterase activity in the supematant of rat epididymal adipose tissue are shown in Fig. 1B (micellar substrate) and Fig. 2B (liposomal substrate). With the micellar substrate, cholesterol esterase activity was optimal at pH 7 in the absence of bovine serum albumin. An addition of bovine serum albumin (0.025%) to the reaction mixture increased the activity 2-3-fold, and shifted the pH optimum to 6. The enhancement of cholesterol esterase activity with bovine serum albumin observed in the experiment with micellar substrate will result from the removal of the reaction product. Fatty acids reportedly inhibit cholesterol esterase [ll]. In the case of the liposomal substrate, the pH optimum of cholesterol esterase was 6, both in the presence and absence of bovine serum albumin in the mixture. Addition of bovine serum albumin neither enhanced the activity nor shifted the pH optimum. Activation of cholesterol esterase activity by cyclic AMP-dependent protein kinase Activation of cholesterol esterase by cyclic AMP-dependent protein kinase was examined using micellar and liposomal substrates (Fig. 1A
PH
PH
Fig. 1. pH dependency and activation of cholesterol esterase (CEase) activity in rat epididymal adipose tissue (micellar substrate). (A) The supematant (10 ~‘1) of rat epididymal adipose tissue homogenate was preincubated at 3O’C for 10 min with cyclic AMP-dependent protein kinase from bovine heart (50 as), cyclic AMP (10 PM), ATP.2Na and magnesium acetate (1.25 mM): A, with cyclic AMP, ATP.2Na and magnesium acetate: o, without any addition; 0, in 100 mM sodium acetate buffer (pH 3.7-5.6) or 100 mM potassium phosphate buffer (pH 6.1-8.1) containing 0.025% bovine serum albumin. Assay of cholesterol esterase activity was started by addition of micellar substrate (10 JLI) to the mixture (200 pl). (B) The supematant (10 ~1) of rat epididymal adipose tissue homogenate was incubated for 10 min at 37 o C with micellar substrate in buffers with (open symbols) or without (solid symbols) an addition of 0.025% bovine serum albumin. Each point represents the mean of duplicate determination.
and Fig. 2A). The supernatant (10 ~1) was preincubated for 10 min with cyclic AMP-dependent protein kinase (50 pg), cyclic AMP (10 PM), ATP .2Na (0.5 mM) and magnesium acetate (1.25 mM) at 30 o C, and then cholesterol esterase activity was determined using the two substrates. The activity measured by the micellar substrate was enhanced at neutral pH ranges when the supematant was incubated with cyclic AMP-dependent protein kinase, cyclic AMP, ATP and magnesium acetate. However, deletion of cyclic AMP-dependent protein kinase from the preincubation medium either blocked the enhancement or even attenuated the activity slightly at pH 5.5 and 6.0 compared with the activity preincubated without any addition. This decrease may result from effects of phosphoprotein phosphatase contaminating the supernatant. No effects of the preincubation were observed at other pH ranges unless cyclic AMP-dependent protein kinase was incubated in the preincubation mixture. The activity assayed with
7
-i PH
PH
Fig. 2. pH dependency and activation of cholesterol esterase (CEase) activity in rat epididymal adipose tissue (liposomal substrate). (A) The supernatant (10 ~1) of rat epididymal adipose tissue homogenate was preincubated at 30°C for 10 min with cyclic AMP-dependent protein kinase from bovine heart (50 gg), cyclic AMP (10 PM), ATP.2Na (0.5 mM) and magnesium acetate (1.25 mM): A, with cyclic AMP, ATP’2NA and magnesium acetate; 0, without any addition; 0, in 100 mM sodium acetate buffer (pH 3.7-5.6) or 100 mM potassium phosphate buffer (pH 6.1-8.1) containing 0.025% bovine serum albumin. Assay of cholesterol esterase activity was started by addition of liposomal substrate (10 ~1) to the mixture (200 ~1). (B) The supematant (10 ~1) of rat epididymal adipose tissue homogenate was incubated for 10 min at 37 o C with liposomal substrate in buffers with (open symbols) or without (solid symbols) an addition of 0.025% bovine serum albumin. Each point represents the mean of duplicate determination. (A) and (B) are separate experiments with epididymal adipose tissue from different donors.
liposome was also enhanced by preincubation with cyclic AMP-dependent protein kinase and other cofactors at pH ranges above 6. Unlike the case of micellar substrate, the removal of cyclic AMP-dependent protein kinase from the mixture did not cause an attenuation of the activity, but it slightly increased the activity at pH 6.5 and 7. These results indicate that addition of cyclic AMP-dependent protein kinase was essential for the activation of cholesterol esterase in the adipose tissue supematant by this assay system. Substrate concentration curves of cholesterol esterase activity with micellar and liposomal substrates Fig. 3 shows substrate concentration curves of cholesterol esterase activity in rat epididymal adipose tissue with micellar and liposomal substrates. The inset shows a double-reciprocal plot
of the data. When micellar substrate was used, the apparent K, 7.9 (PM), and the apparent I$,, 45.0 (nmol fatty acid released/mg protein per h). In contrast, the liposomal substrate gave a higher K, (14.7) and lower V,,, (39.4) values than the micellar substrate. Protein dependency and time-course of cholesterol esterase activity with the micellar substrate As the micellar substrate appeared to be more suitable for measuring neutral cholesterol esterase than the liposomal substrate, a protein-dependency curve and time-course of cholesterol esterase were depicted (Fig. 4). The activity was linear up to 20 pg enzyme protein per tube when the supernatant was incubated for 10 min with 10 ~1 (2.78 . lo4 dpm/1.22 nmol) substrate in 100 mM potassium phosphate buffer (pH 6) at 37°C. The timecourse of cholesterol esterase was linear up to 10 min when 10 ~1 (14.2 pg protein) of the supernatant was incubated with 10 ~1 substrate. This assay with our micellar substrate is much more sensitive than other assays for cholesterol esterase.
Fig. 3. Substrate concentration curves of cholesterol esterase (CEase) activity from the epididymal adipose tissue. The supernatant (20 ~1) of the adipose tissue homogenate was incubated at 37 .aC either with liposomal substrate (0) or with micellar substrate (0) (l-50 pl) in 100 mM potassium phosphate buffer (pH 6.0) containing 0.025% bovine serum albumin for 10 mm (total 200 ~1). Inset shows double-reciprocal plot of the data. Micellar substrate: K, = 7.9 (PM); V,, = 45.0 (nmol fatty acid released/mg protein per h). Liposomal substrate: K, = 14.7 (PM), V,, = 39.4 (nmol fatty acid released/mg protein per h). Each point represents the mean of duplicate determination.
324
the high activity adipose tissue.
of cholesterol
esterase
in
the
Effects of adrenalin, theophylline, potassium fluoride and 2’,3’-dideoxyadenosine on cholesterol esterase activity To confirm the effect of cyclic AMP on cholesterol esterase activity, reagents known to
Fig. 4. Protein dependency and time-course of cholesterol esterase (CEase) activity in rat epididymal adipose tissue. The supernatant of rat epididymal adipose tissue homogenate was incubated at 37 o C with the micellar substrate (10 ~1) containing cholesterol [‘4C]oleate (2.78.104 dpm/1.22 nmol) in 100 mM potassium phosphate buffer (pH 6.0) containing 0.025% bovine serum albumin. (A) Incubated for 10 min: (B) 10 ~1 supernatant (14.2 pg protein) used. Each point represents the mean of duplicate determination.
SC
_I 0
/
40
/e
30
Effects of exogenous cyclic AMP-dependent protein kinase and cyclic AMP on cholesterol esterase activity The results in Fig. 1 indicate that in this cholesterol esterase assay system, an addition of cyclic AMP, ATP and Mg2+ appeared not to be essential for the activation of cholesterol esterase, but an addition of exogenous cyclic AMP-dependent protein kinase from bovine heart was essential for the enhancement of activity. Thus, the effect of exogenous cyclic AMP-dependent protein kinase was examined in detail. Fig. 5 clearly shows that preincubation of the adipose tissue supernatant with varying amounts of cyclic AMP-dependent protein kinase increased cholesterol esterase activity in a dose-dependent manner. A 64% increase in the activity was observed by preincubation with 50 pg cyclic AMP-dependent protein kinase. In contrast, preincubation with over 4 PM cyclic AMP alone slightly decreased the activity. These results suggest that cyclic AMP-dependent protein kinase rather than cyclic AMP is one of the important regulatory factors for cholesterol esterase in this assay with the rat adipose tissue supematant. This effect of cyclic AMP-dependent protein kinase might have become evident because our reaction mixture did not contain enough endogenous cyclic AMP-dependent protein kinase due to the very sensitive assay method, and also
20
ia
a
-10
-20
a
cyclic L 0
20
la AMP
30
dependent
Proteln
4
2
cyclic
6 AMP
40
50
kinarehg) 3
10
h8M)
Fig. 5. Effects of exogenous cyclic AMP-dependent protein kinase or cyclic AMP on cholesterol esterase (CEase) activity in the supematant. The supematant (5 ~1) of rat epididymal adipose tissue was preincubated at 30 Oc in 100 mM potassium phosphate buffer (pH 6.0) containing 0.025% bovine serum albumin in the presence and absence of varying amounts of cyclic AMP-dependent protein kinase from bovine heart (solid symbols) or cyclic AMP (open symbols) for 10 min. Assay of cholesterol esterase activity was started by an addition of mice&u substrate (10 ~1) to the above mixture (200 ~1). Each point represents the mean of duplicate determination. Cholesterol esterase activity in the absence of cyclic AMP-dependent protein kinase and cyclic AMP was 27.68 (nmol fatty acid released/mg protein per h).
325 TABLE
I
EFFECTS OF ACTIVATORS OR INHIBITORS OF ADENYLATE CYCLASE ON CHOLESTEROL ESTERASE ACTIVITY A piece of the epididymal adipose tissue (0.2 g) was preincubated at 37’C in 1 ml Krebs Henseleit buffer (pH 7.4) in the presence or absence of effecters for 30 min. The washed fat pad was then homogenized and the supematant was assayed for cholesterol esterase in 100 mM potassium phosphate buffer (pH 7.0) without bovine serum albumin. Adipose tissues in each experiment are from different donors. Values are mean of duplicate experiments. Preincubated
with
Expt. 1 None Adrenalin (10 PM) Theophylline (1 mM) 2’,3’-Dideoxyadenosine (0.5 mM) Expt. 2 None Adrenalin (10 PM) + theophylline (1 mM) Expt. 3 None GTP (10 FM) GTP (10 pM)+potassium fluoride (10 mM) GTP (10 PM) + isoproterenol (10 PM)
Cholesterol esterase activity (nmol fatty acid released/mg protein per h)
3.27 3.90 2.83 4.59
9.17 10.09
10.62 11.26 11.32 9.31
affect adenylate cyclase were applied to change the cyclic AMP level in the preincubation (Table I). Fat pads (approx. 0.2 g) were incubated with adrenalin (10 PM) in the presence or absence of theophylline (1 mM) at 37°C for 30 min and then the supernatant of the homogenate was prepared as well. There was no enhancement in cholesterol esterase activity by adrenalin although lipolysis was almost maximally increased. The preincubation of the supernatant of adipose tissue with potassium fluoride (10 mM), which activated adenylate cyclase, did not alter the cholesterol esterase activity. 2’,3’-Dideoxyadenosine (0.5 mM), an inhibitor of adenylate cyclase, increased the activity slightly. This result suggests that cyclic
AMP is not a regulatory factor for cholesterol esterase activity in the adipose tissue. Cyclic AMP may play such a role in tissues containing relatively a high activity of the kinase to cholesterol esterase activity. Cyclic AMP-dependent protein kinase and phosphoprotein phosphatase activities in the supernatant Since cyclic AMP-dependent protein kinase appeared to be one of the important regulatory factors for cholesterol esterase activity in the supernatant of adipose tissue, cyclic AMP-dependent protein kinase and phosphoprotein phosphatase activities were assayed in the supernatant. Cyclic AMP-dependent protein kinase activity assayed by the incorporation of [Y-‘~P]ATP into HI-histone was 0.4-3.0 (nmol 32P incorporated/mg protein per min). Since one assay tube contains approx. 15 pg protein, cyclic AMPdependent protein kinase activity in the reaction mixture will be 6-45 pmol/min. This activity is comparable to the amounts of protein kinase added in the experiment shown in Fig. 5. The activity of protein kinase used is 1.2 pmol/pg protein per min. Phosphoprotein phosphatase activity (pmol “P released/mg protein per min) expressed in the term of 32P released from 32P-labeled Hl-histone was 37 5 3.6 [lo] (mean f S.D. (n)). Thus, the level of cyclic AMP-dependent protein kinase activity was one order of magnitude higher than that of the phosphoprotein phosphatase activity in the supernatant. Effects of preincubation with Mg’+ and EDTA Both activation by cyclic AMP-dependent protein kinase and deactivation by phosphoprotein phosphatase of cholesterol esterase need the presrequires ence of Mg ‘+ in the mixture. Deactivation higher concentrations of Mg2+ than activation. Preincubation of the supernatant for 10 min at 32” C with varying concentrations of MgZt (1.25-40 mM) decreased the cholesterol esterase activity. The decrease reached a maximum at the concentration of 20 mM. Fig. 6 shows the timecourse of deactivation of cholesterol esterase activity by preincubation with 5 mM Mg’+. Deactivation of cholesterol esterase proceeded rapidly; the activity (nmol fatty acid released/mg protein per
326
h) decreased from 44.0 to 21.7 in 5 min, after which no significant change was observed. The supernatant was preincubated for 10 min with 1 mM EDTA or 20 mM Mg2+, and then an aliquot of the supernatant was incubated in the presence or absence of cyclic AMP-dependent protein kinase, cyclic AMP and ATP (Table II). Cholesterol esterase activity was enhanced 74% by preincubation with EDTA and reduced 52% by Mg2+ preincubation compared to control. The activity which was raised by EDTA was further enhanced over 2-fold after an incubation in the presence of cofactors. The concentration of EDTA in the second incubation medium of EDTA-preincubated samples was 50 PM. It was assumed that EDTA inhibited phosphoprotein phosphatase, so that deactivation of cholesterol esterase was prevented, while Mg2+ activated the phosphoprotein phosphatase, which in turn deactivated cholesterol esterase. Thus, the endogenous (partially phosphorylated) activity of cholesterol esterase would be 11.5 (nmol fatty acid released/mg protein per h), and the basal (dephosphorylated) activity, ap-
TABLE II CHANGES IN CHOLESTEROL ESTERASE ACTIVITY OF THE SUPERNATANT BY PREINCUBATION WITH EDTA AND Mg’+ The supematant of adipose tissue homogenate was preincubated at 32’C for 10 min with EDTA’2Na (1 mM) and Mg2+ (20 mM), and then 10 ~1 of the supernatant was incubated in the presence or absence of cyclic AMP (10 pM), magnesium acetate (1.25 mM), ATP.2Na (0.5 mM) and cyclic AMP-dependent protein kinase from bovine heart (50 pg) in 200 pl of 100 mM potassium phosphate buffer (pH 6.0) containing 0.025% bovine serum albumin for another 10 min. Assay of cholesterol esterase was started by addition of the micellar substrate (10 al) to the mixture (200 ~1). Values are means of triplicate determination. Preincubation
None EDTA (1 mM) Mg*+ (20 mM)
Cholesterol esterase activity (nmol fatty acid released/ mg protein per h) Additions:
cyclic AMP, MgAc, ATP. ZNA, protein kinase
no
additions
+ additions
6.6 15.1 (+ 128%) b 11.5 (+ 74%) a 26.3 (+ 129%) b 3.2(-52%)” 5.7 (+79%) b
a % change in the presence of EDTA or Mg2+, relative to their absence. ’ % change in the presence of cyclic AMP-dependent protein kinase relative to respective value (left column) in its absence.
L 0
5
. 10
Praincubation
20
timetmtnl
Fig. 6. Time-course of deactivation of cholesterol esterase (CEase) from rat epididymal adipose tissue. The supematant (10 ~1) of adipose tissue homogenate was preincubated at 32OC for various periods in 100 mM potassium phosphate buffer (pH 6.0) containing 0.025% bovine serum albumin and 5 mM magnesium acetate. Assay of cholesterol esterase was started by an addition of the micellar substrate (10 ~1) to the above mixture (200 PI). Each point represents the mean of duplicate determination.
prox. 3.2, and the full (fully phosphorylated) ity, 26.3.
activ-
Discussion Two types of cholesterol esterase were demonstrated in the aorta and in the liver: acid cholesterol esterase in the lysosome and neutral cholesterol esterase in the cytosol fraction. Acid cholesterol esterase activity can be efficiently revealed by liposomal [lo] or phospholipid-digitonin-dispersed substrates [12] containing cholesteryl oleate. Several substrates for neutral cholesterol esterase have been reported: acetone substrate [13], ethanol or gum arabic-stabilized substrate [2], phosphatidylcholine-glycerol suspension [ 81, a micellar substrate [6] and a liposomal substrate [lo]. Using a micellar substrate containing phosphatidylcho-
327
line/ sodium taurocholate/ cholesteryl oleate (4: 2: l), Hajjar et al. [6] have recently demonstrated the presence of a neutral cholesterol esterase in rabbit aortas which was activated by cyclic AMP. Although our previous studies [9] on adipose tissue cholesterol esterase have been carried out with a liposomal substrate, the micellar substrate is assumed to be suited to assay of adipose tissue cholesterol esterase. In this study, therefore, the efficiency of the liposomal and the micellar substrate was compared in an assay of adipose tissue cholesterol esterase. The micellar substrate gave a larger apparent V,, and smaller apparent K, than the liposomal substrate for cholesterol esterase in the supematant of rat adipose tissue homogenate. Pittman et al. [2] had reported that the supernatant of the epididymal adipose tissue homogenate from Sprague-Dawley rats had high cholesterol esterase activity which was activated by cyclic AMP/ATP. Mg. Pretreatment of the rats with glucagon or other hormones increasing the cyclic AMP level was found to activate cholesterol esterase activity. Similar cholesterol esterase activity in macrophages [14], adrenocortical cells [3] and arterial smooth muscle cells [6] was activated by addition of cyclic AMP and ATP . Mg or agents to increase the cyclic AMP level (adrenalin, theophylline). These findings suggested that cholesterol esterase activity was regulated by the cyclic AMP level in the cells. None of these experiments regarded cyclic AMP-dependent protein kinase activity as an important regulatory factor. In experiments with adipocytes of SpragueDawley rats [15], lipolysis rates correlated with cyclic AMP-dependent protein kinase activity ratios (cyclic AMP-dependent protein kinase activity in the absence or in the presence of cyclic AMP) below 0.35 of the ratios. As the cyclic AMP-dependent protein kinase activity ratio rises above that level, the rate of dephosphorylation of the lipase by the phosphoprotein phosphatase exceeds the rate of phosphorylation, so that the lipolysis ratio remains constant. The cyclic AMP concentration seems to be a rate-limiting factor for lipolysis only within limited ranges. In our experiments, cholesterol esterase activity in the supematant of the epididymal adipose tis-
sue from the Wistar strain was assayed by a micellar substrate. The assay system employed here with our micellar substrate provided a sensitive method for the measurement of cholesterol esterase, and approx. 15 pg protein of the supernatant (43000 x g) was enough for an assay. The cholesterol esterase activity was not activated by preincubation with cyclic AMP/ATP * Mg, while it was enhanced dose-dependently by preincubation with exogenous cyclic AMP-dependent protein kinase of bovine heart (see Fig. 5), Pretreatments of adipose tissue with adrenalin/ theophylline, isoproterenol, potassium fluoride or 2’,3’-dideoxyadenosine did not influence cholesterol esterase activity although the same treatment affected lipolysis activity in the adipose tissue or the adipocytes. These results imply that cyclic AMP-dependent protein kinase activity rather than cyclic AMP is an important regulator of cholesterol esterase activity in this system. This hypothesis is further supported by the fact that cyclic AMP-dependent kinase activity in the assay mixture, measured by incorporation of [y- 32P]ATP into histone (see Results) was in the range comparable to the activity exogenously added (Fig. 5). One of the reasons that the regulatory role of cyclic AMP-dependent protein kinase on cholesterol esterase became evident in our system will be because only a small amount of the enzyme protein was used for the assay due to the high activity in the adipose tissue and due to our sensitive method; thus, there was not enough cyclic AMP-dependent protein kinase present in the mixture. If we used more enzyme protein, the effect of the kinase would be attenuated, or not apparent. In the report by Pittman et al. [2], who showed the effect of cyclic AMP on cholesterol esterase, more than lo-fold the enzyme protein (supematant of 100000 x g) that we used in the supematant of 43000 X g was employed. Cholesterol esterase activity in aorta, adrenal cortex, liver, myocardiac cells and testes is not as high as that in adipose tissue. The enzyme preparations from these tissues contain a relatively high activity of cyclic AMP-dependent protein kinase compared to that of cholesterol esterase. In these tissues, cyclic AMP may play a regulatory role for cholesterol esterase, as Hajjar et al. [6] showed in rabbit aorta.
328
It was reported that some pituitary stimulating hormones induce the change in cyclic AMP-dependent protein kinase activity in organs where steroid hormones are produced; in the adrenal cortex, adrenocorticotropic hormone activates cyclic AMP-dependent protein kinase by increasing the catalytic unit [16], and in rat ovary, the combined action of estradiol and follicle-stimulating hormone regulates cyclic AMP-dependent protein kinase activity through the increased transcription of the gene of the regulatory unit (RII,,) [17]. Gonadotropin-induced steroidgenesis in testicular interstitial cells is also intermediated by cyclic AMP-dependent protein kinase [5]. We have found recently that treatment of rats of the Wistar strain with estradiol enhanced markedly cholesterol esterase activity in the adipose tissue, with a concomitant change in cyclic AMP-dependent protein kinase activity. Ovariectomy resulted in an attenuation of cholesterol esterase and cyclic AMP-dependent protein kinase activities in the parametrial adipose tissue. There was no change in the cyclic AMP level in the tissue [18]. Enhancement in testicular cholesterol esterase by follicle-stimulating hormone and luteinizing hormone might be mediated through a similar mechanism [9]. Thus, we conclude that cholesterol esterase activity in the supematant of adipose tissue homogenate from the Wistar rat strain is regulated primarily by cyclic AMP-dependent protein kinase activity rather than by cyclic AMP levels. Acknowledgements This investigation has been supported in part by a grant-in-aid for scientific research from the Ministry of Japan (No. 60571051,1985-1986) and
by a medical research grant from the Ministry of Education of Japanese Health and Welfare (198331985). References 1 Amaud, J. and Boyer, J. (1974) B&him. Biophys. Acta 337, 165-168. 2 Pittman, R.C., Khoo, J.C. and Steinberg, D. (1975) J. Biol. Chem. 250, 4505-4511. 3 Trzeciak, W.H., Mason, J.I. and Boyd, G.S. (1979) FEBS Lett. 102, 13-17. 4 Bisgaier, CL., Treadwell, C.R. and Vahouny, G.V. (1979) Lipids 14, 1-14. 5 Bailey, M.L. and Grogan, W.M. (1986) J. Biol. Chem. 261, 7717-7722. 6 Hajjar, D.P., Minick, C.R. and Fowler, S. (1983) J. Biol. Chem. 258, 192-198. 7 Goldberg, D.I. and Khoo, J.C. (1985) J. Biol. Chem. 260, 5879-5882. 8 Severson, D.L. and Fletcher, T. (1978) Atherosclerosis 31, 21-32. 9 Tomita, T., Yonekura, I., Umegaki, K., Okada, T. and Hayashi, E. (1984) Hot-m. Metabol. Res. 16, 461-464. 10 Brecher, R., Chobanian, J., Small, D.M. and Chobanian, A.V. (1976) J. Lipid Res. 17, 239-247. A.V. 11 Brecher, P., Kessler, M., Clifford, C. and Chobanian, (1973) B&him. Biophys. Acta 316, 386-394. 12 Haley, N.J., Fowler, S. and De Duve, C. (1980) J. Lipid Res. 21, 961-969. A.V. (1977) J. 13 Brecher, P., Pyun, H.Y. and Chobanian, Lipid Res. 18, 154-163. 14 Khoo, J.C., Mahoney, E.M. and Steinberg, D. (1981) J. Biol. Chem. 256, 12659-12661. 15 Honnor, R.C., Dhillon, G.S. and Londos, C. (1985) J. Biol. Chem. 260, 15130-15138. C. (1987) Endo16 Murray, S.A., Champ, C. and Lagenauer, crinology 120, 1921-1927. G.S., Lifka, J., Durica, J.M. and 17 Hedin, L., Mcknight, Richards, J.S. (1987) Endocrinology 120, 192881935. K., Inoue, Y., Ikeda, M., Talceshita, N. and 18 Nakamura, Tomita, T. (1988) J. Jpn. Atheroscler. Sot. 16, 349-351. 19 Durham, L.A., III and Grogan, W.M. (1984) J. Biol. Chem. 259, 7433-7438.