Activation of Acetyl-coA Carboxylase

THE J O U R N A L OF BIOLOGICAL CHEMISTRY IC’ 1985 by The American Society of Biological Chemists, Inc

Val. 260, No. 10, Issue of May 25, pp. 6318-6323. 1985 Printed in U. S.A.

Activation of Acetyl-coA Carboxylase PURIFICATION AND PROPERTIES OF A Mn2+-DEPENDENT PHOSPHATASE*

(Received for publication, December 13, 1984)

K. George Thampy and Salih J. Wakil

Acetyl-coA carboxylase was isolated from rat liver by polyethylene glycol precipitation and avidin affinity chromatography. Sodium dodecyl sulfate electrophoresis of the enzyme gives one protein band (Mr 250,000). Phosphate analysis of the carboxylase showed the presence of 8.3 mol of phosphate/mol of subunit (Mr 250,000). The purified carboxylase has low activity in the absence of citrate (specific activity = 0.3 unitslmg). However, addition of 10 mM citrate activates the carboxylase 10-fold, with half-maximal activation observed at 2 mM citrate, well above the physiological citrate level. Using this carboxylase as a substrate, we have isolated from rat liver a protein that activates the enzyme about 10-fold. This protein has been purified to near homogeneity (M, 90,000). Incubation of this protein with “P-labeled acetyl-coA carboxylase results in a time-dependent activation of carboxylase with concomitant release of 32Pi, indicating that this protein is a phosphoprotein phosphatase. Both activation and dephosphorylation are dependent on Mn2+, but not citrate. This phosphatase does not hydrolyze p-nitrophenyl phosphate but does show high affinity for acetyl-coA carboxylase (K , = 0.2 PM) as compared to its action on phosphorylase a (K , = 5.5 PM) and phosphohistone (K , = 20 PM). Activated acetyl-coA carboxylase was isolated after dephosphorylation by the phosphatase. Such preparations contain about 5 mol of phosphate/mol of subunit and have specific activities of 2.6-3.0 unitslmg in the absence of citrate. These activities are comparable to those of the phosphorylated carboxylase in the presence of 10 mM citrate. Thus, dephosphorylation by the Mn2+-dependent phosphatase renders the carboxylase citrate-independent, as compared to the phosphorylated form, which is citrate-dependent. To our knowledge this is the first report of a preparation of animal acetyl-coA carboxylase that has substantial catalytic activity independent of citrate.

Acetyl-coA carboxylase, a biotin-containing enzyme, was first recognized by Wakil (1) and Wakil et al. (2) as an enzymatic component of the fatty acid biosynthetic pathway. The enzyme catalyzes the carboxylation of acetyl-coA to form malonyl-CoA, the committed step in the synthesis of long chain fatty acids. The reaction involves carboxylation of the biotinyl prosthetic group (biotin carboxylase, Reaction 1) and

* This work was supported in part by Grant GM19091 from the National Institutes of Health and Grant PCM 8206562 from the National Science Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “adoertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

subsequent transfer of carboxyl group from biotin to acetyl- CoA (transcarboxylase, Reaction 2). E-biotin + ATP

MnZ+ or Mg2+ (1) + HCO; E-biotin-CO; + ADP + Pi

E-biotin-COB + acetyl-CoA + E-biotin + malonyl-CoA (2)

There was little agreement in earlier studies on the subunit structure of animal acetyl-coA carboxylase (3, 4). The disa- greement stems from the enzyme’s susceptibility to proteolysis during isolation, which resulted in cleaved fragments that retained catalytic activity. Rapid purification methods involv- ing polyethylene glycol precipitation and avidin affinity chromatography have overcome the problem of proteolysis en- countered during carboxylase isolation. The enzyme isolated from chicken liver (5, 6), rat liver (7), and rabbit mammary gland (8) is composed of identical subunits of molecular weight 225,000-250,000, each subunit containing one biotin prosthetic group, two catalytic activities, namely biotin carboxylase and transcarboxylase, and the allosteric site.

Since the acetyl-coA carboxylase reaction is the committed step in fatty acid synthesis, its regulation has been actively investigated in many laboratories. It has long been known that citrate is an allosteric activator of the enzyme. Although the exact mechanism of citrate activation is not known, recent evidence shows that in the presence of citrate the enzyme undergoes conformational changes resulting in its activation (9), followed by polymerization of the protein (10). Another way of modifying enzyme activity may be achieved by phosphorylation/dephosphorylation reactions. In 1972, Inoue and Lowenstein (4) suggested the idea of phosphorylation as a mechanism of regulation of acetyl-coA carboxylase based on their finding that the carboxylase is a phosphoprotein. Later, Carlson and Kim (11) reported a Mg2+-dependent activation of the carboxylase in crude preparations from rat liver. They also observed an ATP-dependent inhibition of acetyl-coA carboxylase in crude preparations and the incorporation of 32Pi from [y3*P]ATP into immunoprecipitated carboxylase. These experiments gave support to the idea of phosphorylation/dephosphorylation as a mechanism of regulating the formation of malonyl-CoA. Phosphorylation and simultaneous inactivation of rat liver acetyl-coA carboxylase by a CAMP-independent protein kinase (In), as well as by the catalytic subunit of CAMP-dependent protein kinase (13), have been reported. However, incorporation of 2 mol of phosphate/mol of carboxylase subunit catalyzed by two CAMP- independent kinases of rat liver (casein kinases I and 11) did not alter carboxylase activity (14). Moreover, carboxylase labeled with 9-10 phosphate groups/subunit prepared from cultured chick liver cells (15) was found to be fully active.

Dephosphorylation of rat liver carboxylase by an “acetyl-

63 18

Regulation of Acetyl-coA Carboxylase 6319

CoA carboxylase phosphatase" isolated from the same tissue resulted in its activation (16). Carboxylases of rabbit mammary gland (17) and chicken liver (6) can be activated by phosphatases derived from other tissues. In contrast, alkaline phosphatase hydrolyzes 2 and 3 mol of phosphate/mol of subunit, respectively, from chicken liver (6) and rat liver (18) carboxylases without any change in carboxylase activity. Thus, the role of phosphate groups in the regulation of acetyl- CoA carboxylase remains ambiguous with several reports supporting such a mechanism (6, 8, 13, 16, 17), while others could not demonstrate direct relationship between phosphorylation and catalytic activity (9, 15, 19).

In this study, we report the isolation from rat liver of a highly purified Mn2+-dependent phosphoprotein phosphatase which activates the carboxylase with concomitant release of phosphate. The dephosphorylated carboxylase requires little or no citrate for its activity, whereas the phosphorylated carboxylase is completely dependent on citrate. Hence, we propose the names acetyl-coA carboxylase (D) and (I) to denote their dependence or independence of citrate, respectively.

MATERIALS AND METHODS

Avidin, polyethylene glylcol 8000, phenylmethylsulfonyl fluoride, lysine-rich histone, phosphorylase b, phosphorylase kinase, biotin, and benzamidine were purchased from Sigma. [14C]Bicarbonate, car- rier-free 32Pi, and [Y-~'P]ATP were purchased from ICN Biochemi- cals.

if'ication of the "COz fixation assay of Waite and Wakil (20). The Carboxylase Assay-Acetyl-coA carboxylase was assayed by a mod-

assay mixture contained 50 mM HEPES,' pH 7.5, 2.5 mM MnCI', 2.0 mM DTT, 0.125 mM acetyl-coA, 4.0 mM ATP, 12.5 mM ["C]KHCO3 ( 2 X lo6 cpm), 0.75 mg/ml BSA, 0.1-0.2 pg of affinity purified carboxylase, and 0.5 mM citrate, in a total volume of 150 pl. The reaction was initiated by the addition of acetyl-coA carboxylase. The assay was carried a t 37 "C for 2-5 min. The reaction was stopped by the addition of 50 pl of 6 N HC1. Subsequently, 150 p1 was transferred into a scintillation vial and evaporated to dryness by placing in an oven a t 100 "C for 30 min. The dried vials were cooled, 0.5 ml of water and 5 ml of Aqueous Counting Scintillant (Amersham Corp.) were added, and the radioactivity was determined in a liquid scintillation counter. The tubes, where HC1 was added before the addition of' acetyl-CoA carboxylase, served as blanks. The radioactivity in the hlank measured consistently between 30 and 50 cpm. One unit of activity is defined as 1 pmol of malonyl-CoA formed per min a t 37 "C. The specific activity is defined as units/milligram of protein.

Acetyl-coA Carboxylase (D) Phosphatase Ass~y-The carboxylase- activating enzyme was assayed by measuring the increase in the activity of affinity purified acetyl-coA carboxylase of rat liver. A typical assay mixture contained 50 mM HEPES, pH 7.5, 1.5 mM MnCI2, 0.75 mg/ml BSA, 2 mM DTT, and activating protein in a final volume of 100 pl. The reaction was initiated by the addition of 2 pg of affinity purified acetyl-coA carboxylase, and the mixture was incubated at 37 "C for variable times (0-45 min). At the end of the indicated time, a sample (10 pl) was withdrawn and added to 140 pl of the acetyl-coA carboxylase assay mixture, and the carboxylase was assayed as described above. The 15-foId dilution of the activated acetyl-CoA carboxylase essentially stopped the phosphatase action. One unit of activating phosphatase is defined as that amount of protein which causes an increase in the activity of 2 pg of acetyl-coA carboxylase measured in the presence of 0.5 mM citrate, by 1 mil- liunit/minute a t 37°C.

When "P-labeled proteins were used, the liberated 32Pi was determined by extracting the phosphomolybdate complex with 2-butanol/ benzene, and radioactivity was measured as described by Ullman and Perlman (21).

p-Nitrophenyl phosphate hydrolyzing activity was determined as described (22). 32P-labeled phosphorylase and histone were prepared

The abbreviations used are: HEPES, 4-(2-hydroxyethyl)-l-piper- azineethanesulfonic acid; PEG, polyethylene glycol; DTT, dithio- threitol; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; BSA, bovine serum albumin.

according to the method of Antoniw et d . (23). Rabbit muscle phosphorylase b was phosphorylated with phosphorylase kinase and [y- 32P]ATP. The product, [32P]phosphorylase a, was recrystallized once. It had a radiospecific activity of 800 cpmlpg of protein, corresponding to an incorporation of 0.83 mol of Pi/subunit. The catalytic activity of phosphorylase Q in the absence of AMP was over 80% of that observed in the presence of AMP. Lysine-rich histone was phosphorylated with [y-32P]ATP and CAMP-dependent protein kinase. The purified preparation had 1.7 mol of Pi/mol and a radiospecific activity of 5000 cpm/pg of protein.

Phosphate Determination-Phosphate was determined by the method of Hess and Derr (24). Alkali-labile phosphate was determined after a 3-h hydrolysis of the protein in the presence of 1.0 N NaOH a t 100 "C. At the end of hydrolysis, the NaOH was neutralized with 10 N H2S04 and the phosphate was estimated (24).

Protein Determination-Protein was determined by the method of Lowry et al. (25), using bovine serum albumin as standard. For purified carboxylase, the protein was determined from its absorbance a t 280 nm. A solution of pure carboxylase with a protein concentration of 1.0 mg/ml (based on Lowry) gave an Ai&m of 0.86.

Preparation of Auidin-Sepharose Column-Avidin was cross-linked to Sepharose according to the method described by Beaty and Lane (5), except that the ratio of volume of Sepharose to weight of avidin was doubled. The exchangeable biotin binding capacity of the column was 7 nmol/ml of packed column.

Purification of Rat Liver Acetyl-coA Carboxylase-The carboxylase was purified by a combination of PEG precipitation and avidin affinity chromatography. Rats were fasted for two days and refed with a high carbohydrate diet for 2 days before asphyxiation with carbon dioxide. The livers were quickly removed, rinsed and then homogenized in 4 volumes of 50 mM potassium phosphate, pH 7.4, 10 mM EDTA, 10 mM 2-mercaptoethanol, 2 mM benzamidine, and 0.2 mM phenylmethylsulfonyl fluoride, using a Waring blender for 1 min at 4 "C. All subsequent operations were carried out a t 4 "C. The homogenate was made 3% in polyethylene glycol by the addition of a 50% PEG solution. The solution was stirred for 1-2 min and centrifuged immediately a t 50,000 X g for 10 min. The supernatant was collected, and the PEG concentration was adjusted to 5% by the addition of 50% PEG solution. After stirring for 1-2 min, the solution was centrifuged a t 50,000 X g for 10 min. The carboxylase was precipitated in this step, while the activating phosphatase remained

buffer (100 mM Tris-HCI, pH 7.5, 1 mM EDTA, 0.1 mM DTT, 0.5 M in solution. The pellet was suspended in minimum volume of column

NaC1, and 5% glycerol) and applied to an avidin-Sepharose column at a flow rate of 10-15 ml/h (5). The column was washed with 10-15 column volumes of column buffer. When the A@' of the washings reached 0.005 or less, the carboxylase was eluted with column buffer containing 0.2 mM biotin at a flow rate of 3-5 ml/h. A single peak of carboxylase activity overlapping the protein peak was observed. The active fractions were pooled and dialyzed against 50 mM HEPES, pH 7.5, 0.1 mM DTT, 1.0 mM EDTA, and 10% glycerol for 2 h a t 4 "C. This preparation was homogeneous as judged by SDS-PAGE. The carboxylase activity of the dialyzed preparation was stable for several months a t -70 "C.

RESULTS

Preparation of 32P-labeled Carboxylase-Incorporation of "Pi into acetyl-coA carboxylase was accomplished by perfusion of rat liver with 32Pi. Rats were fasted for 2 days and refed a high carbohydrate diet for 2 days before starting the perfusion. The animal was anesthetized with an intraperiton- ial injection of Nembutal. The liver was perfused with 125 ml of Krebs-Ringer bicarbonate buffer containing 0.15 mM KH:'PO4 (250 Ci/mol) and 10 mM glucose (26). The perfusion was done at 37 "C for 90 min. At the end of this time period, the liver was removed and homogenized along with an unperfused liver. The acetyl-coA carboxylase was subsequently isolated by avidin-Sepharose chromatography as described under "Materials and Methods." As shown in Fig. l, the protein, carboxylase activity, and 32P radioactivity peaks overlapped each other. The catalytic activity of the pooled material was 2.8 units/mg in the presence of 10 mM citrate. The radiospecific activity was 500 cpm/pg, which corre- sponded to the incorporation of 1.1 mol of Pi/mol of acetyl-

6320 Regulation of Acetyl-coA Carboxylase

h -1 I

0 1 5 9 13 17

Fracuon No.

FIG. 1. Elution of s2P-labeled acetyl-coA carboxylase from avidin-Sepharose column. The 5% PEG precipitate was dissolved in minimum volume of column buffer and applied to 1.5 ml (packed volume) of avidin-Sepharose, as described under "Materials and Methods." The column was washed with 15 column volumes of column buffer, and the carboxylase was then eluted with column buffer containing 0.2 mM biotin. Fractions of 1.2 ml were collected. 0, Am nm; A, 32Pi, counts/minute/fraction; W, carboxylase activity/ fraction in the presence of 10 mM citrate.

CoA carboxylase subunit. The preparation of acetyl-coA carboxylase was homogenous, as judged by SDS-PAGE (Fig. 2), with an estimated molecular weight of 250,000. When '*P- labeled carboxylase was subjected to SDS-PAGE, the 250,000- dalton band was the only one containing the radiolabel "Pi (cf. Fig. 2).

Purification of Carboxylase (0) Phosphatase-The purification of carboxylase (D) phosphatase was followed by the functional assay, i.e. activation of acetyl-coA carboxylase, observed a t low citrate concentrations. A dephosphorylation assay using "lP-labeled acetyl-coA carboxylase or "2P-labeled phosphoprotamine (16) is, however, much less desirable be- cause the carboxylase protein contains 5-15 phosphate (7,17, 27) groups/subunit, and some of these phosphate groups are not directly related to activity (6, 14).

The carboxylase (D) phosphatase was purified from rat liver by fractionation of the 5% PEG supernatant solution obtained after precipitation of the carboxylase activity (see "Materials and Methods"). All purification steps were carried out at 4 "C. An equal volume of saturated ammonium sulfate solution, pH 7.4, was added to the 5% PEG supernatant fluid. The precipitated protein was collected by centrifugation and dissolved in 20 mM Tris-HC1, pH 7.5, containing 125 mM NaCl. This solution was dialyzed against the same buffer for 12 h and then applied to a DEAE-Bio-Gel column which was previously equilibrated with the same buffer. The column was then washed with 2 column volumes of the above buffer and eluted with a linear gradient of the same buffer containing NaCl from 125 to 200 mM. The carboxylase-activating phosphatase was eluted as one peak when the NaCl concentration was between 125 and 150 mM. The fractions enriched in carboxylase activating activity were pooled, and the protein was precipitated by adding an equal volume of saturated ammonium sulfate solution. The precipitate was collected and dissolved in minimum volume of 20 mM Tris-HC1, pH 7.5, 1.0 mM EDTA, 10 mM 2-mercaptoethanol, 100 mM NaC1, and 5% glycerol. This solution was applied to a Sephadex G-200 column, and the proteins were eluted with the same buffer. Fractions of 1 ml were collected. The enzyme was eluted as a single, somewhat sharp peak well beyond the main protein peaks (Fig. 3). The fractions enriched in carboxylase (D) phosphatase were pooled and concentrated by Amicon filtra- tion. The yield in this step was 50%, and the overall yield was 14% (cf. Table I). This material was nearly homogenous as judged by SDS-PAGE with an estimated molecular weight of 90,000.

A B 1 .. .: .

FIG. 2. Polyacrylamide gel electrophoresis of rat liver acetyl-coA carboxylase in the presence of SDS. Purified rat liver acetyl-coA carboxylase (5 pg) labeled with "Pi was subjected to SDS- PAGE on 5% slab gels according to the method of Laemmli (32). A , Coomassie Blue-stained gel; B, autoradiogram of the gel in A .

Dephosphorylation and Simultaneous Activation of Acetyl- CoA Carboxylase-When affinity purified and "P-labeled acetyl-coA carboxylase was incubated with purified phosphatase in the presence of Mn2+, a 3-4-fold activation of acetyl-coA carboxylase was observed (Fig. 4). This activation was pro- portional to time of incubation for up to 40 min, beyond which the rate of activation rapidly declined. This time-dependent activation was accompanied by release of 32Pi, as shown by the linear relationship between the two parameters depicted in Fig. 4 (inset). Neither activation nor "Pi release required citrate. But the activation was most apparent when the carboxylase assay contained physiological levels of citrate (0.5 mM or less). Incubation of the carboxylase in the absence of phosphatase did not alter carboxylase activity nor did it release 32Pi (Fig. 4), indicating that carboxylase preparation is free of phosphatase and that activation is due to dephosphorylation. It is of interest to note that activation reached limiting values after about 50% of the "P label had been


0.5

p 0.3

- 0.4

-

-

0.2 -

0.1 -

0 -

2

120 160 Z W Elutlon Volume (mll

FIG. 3. Elution profile of carboxylase (D) phosphatase from a Sephadex G-200 column. The pooled fractions from DEAE chromatography were concentrated as described under "Materials and Methods," applied on a G-200 column (90 X 5 cm), and eluted with 20 mM Tris-HC1, pH 7.5, 100 mM NaC1, 10 mM 2-mercaptoethanol, and 5% glycerol. Fractions of 1.0 ml were collected and assayed for acetyl-coA carboxylase activating activity. 0, Am; A, carboxylase (D) phosphatase activity in units/fraction.

TABLE I Purification of rat liver acetyl-coA carboxylase-activating

phosphatase The carboxylase (D) phosphatase was purified from the 5% PEG

supernatant (derived from 200 g of liver from fasted and refed rats) as described under "Materials and Methods." The purification was followed by the activation of affinity purified rat liver acetyl-coA carboxylase.

Total Total Fraction Pro- Vol- Pro- Specific Activ- Yield

tein ume tein activity ity

Ammonium sulfate, 1.0 800 800 1.0 800 100'

DEAE 4.0 6 24 10.0 240 30 Sephadex G-200 2.4 0.25 0.6 181 109 14

0-50%

" Arbitrarily take as 100.

hydrolyzed. The remaining 50% of radioactive 32P remained bound to the enzyme and was presumed to be not available to this phosphatase.

Both activation and dephosphorylation required the presence of Mn2+. When "P-labeled carboxylase was incubated with either the phosphatase or Mn2+, neither activation nor significant %'Pi release could be observed. But when both the carboxylase (D) phosphatase and MnZ+ were present, a 3-fold activation as we11 as significant release of 32Pi could be dem- onstrated (Table 11).

The activation of acetyl-coA carboxylase was pH-dependent. It was optimal around pH 7.5 despite some loss of carboxylase activity on incubation at this pH.

Substrate Specificity of Carboxylase (D) Phosphatase-The rate of activation of acetyl-coA carboxylase by the activating phosphatase was dependent on carboxylase concentration. At low carboxylase concentrations (10-100 nM, subunit M, - 250,000), the activation by phosphatase followed Michaelis- Menten kinetics. A double reciprocal plot of the data gave a straight line with a K, value of 0.2 ~ L M for acetyl-coA carboxylase. The phosphatase was also active toward other phospho- proteins. The enzyme hydrolyzed [32P]phosphorylase a and [:"P]histone with varying affinities: K, = 5.5 p~ for phosphorylase a and 20 p~ for histone. Thus, the phosphatase exhibits significant affinity toward carboxylase when compared to other substrates. The crude preparations of the phosphatase hydrolyzed p-nitrophenyl phosphate, but the purified preparations did not show detectable activity toward this substrate at neutral or alkaline pH.

E . .- n

A ACC lUimgl

I I ' 0 0 20 40 60 90 120

Time (mid FIG. 4. Activation and simultaneous dephosphorylation of

S2P-labeled acetyl-coA carboxylase by acetyl-coA carboxylase-activating phosphatase. The reaction mixture contained 50 mM HEPES, pH 7.5, 32P-labeled acetyl-coA carboxylase (40 pg), 1.0 mM MnCIZ, 0.75 mg of BSA, 2 mM DTT in a final volume of 1.0 ml. The mixture was incubated at 37 "C in the presence (solid symbols) or absence (open symbols) of acetyl-coA carboxylase-activating phosphatase (6 pg). At the times indicated, 5 pl was withdrawn, diluted 30-fold into acetyl-coA carboxylase assay buffer, and assayed for acetyl-coA carboxylase (squares). For the determination of 32P,, 150 pl was withdrawn, added to an equal volume of 25% trichloroacetic acid, cooled over ice, and centrifuged. A sample (280 pl) was withdrawn from the supernatant fluid and counted in a liquid scintillation spectrophotometer (circles). Alternatively, the 32Pi released was determined after extraction of the phosphomolybdate complex with 2- butanol/henzene (27). Both methods gave the same results. Inset, plot of increase in 3zPi counts/minute Llersus increase in acetyl-coA carboxylase activity.

TABLE I1 Mn2+ dependence of the carboxylase activation and

dephosphorylation 32P-labeled acetyl-coA carboxylase (5.8 pg) was incubated at 37 "C

for 1 h in a total volume of 150 pl in the presence of Mn2+ (1.0 mM) or carboxylase (D) phosphatase (0.6 pg) or both, as indicated. The carboxylase activity and 32Pi released were assayed as described in the legend t,o Fig. 4.

Addition Carboxylase activity 32Pi

released

unitslmg cpm None 0.74 175 Phosphatase 0.74 210 Mn2+ 0.66 180 Phosphatase + Mn2+ 2.30 910

Preparation and Isolation of Dephosphorylated Acetyl-coA Carboxylase-The production of dephosphorylated carboxylase on a preparative scale was readily accomplished by in- creasing the size of the reaction mixture described for the activity assay. It was necessary to keep the concentration of carboxylase below 40 pg/ml to avoid its precipitation in the presence of Mn2+ ions. Acetyl-coA carboxylase (2.8 mg) was incubated with DEAE-purified carboxylase (D) phosphatase (8 mg) in the presence of 1 mM MnC12, 2 mM DTT, 0.75 mg/ ml BSA, and 0.2 mM phenylmethylsulfonyl fluoride in a total volume of 100 ml for 40 min at 37 "C. The time course of carboxylase activation was followed by assaying carboxylase activity in the presence and absence of citrate. The carboxylase activity, measured in the absence of citrate, increased from 0.3 to 2.3 units/mg during the first 20 min of incubation with the phosphatase. By the end of another 20 min of incubation, the activity of the carboxylase reached 3.0 units/ mg, which is about the limiting value obtained in the presence of citrate. At the end of incubation with phosphatase, the reaction mixture was cooled to 4 "C and the carboxylase was

6322 Regulation of Acetyl-coA Carboxylase

precipitated by addition of PEG (50%) to give a final concentration of 5%. The precipitate was suspended in minimum volume of column buffer and applied to an avidin-Sepharose column. The acetyl-coA carboxylase was eluted essentially as described under “Materials and Methods.” The isolated carboxylase had a specific activity of 2.6 units/mg assayed in the absence of citrate.

The native carboxylase and the activated enzyme were indistinguishable (Fig. 5) in their electrophoretic pattern on SDS-PAGE. Both native and activated forms had the same subunit molecular weight (-250,000), suggesting that the activation may not be due to some gross proteolysis. Alkali- labile phosphate content of these forms of the carboxylase revealed that the native enzyme contained 8.3 mol of Pi/mol of carboxylase subunit, while the activated form contained 5.1 mol of Pi/mol of subunit. These results, therefore, strongly indicate that activation is due to dephosphorylation.

Citrate Dependence of Phosphorylated and Dephosphoryl- ated Carboxylase-The acetyl-coA carboxylase isolated from livers of fasted-refed rats had very little activity in the absence of citrate. The enzyme showed a strong dependence on citrate (Fig. 6). Addition of citrate to 10 mM stimulated the activity 10-fold (0.3 to 3.0 units/mg). A double reciprocal plot (not shown) of data in Fig. 6 gave a half-maximal activation at 2 mM citrate, well above the physiological level. The dephosphorylated form was also activated by citrate, reaching a

A B

FIG. 5. SDS-PAGE of phosphorylated and dephosphorylated acetyl-coA carboxylase. Native acetyl-coA carboxylase (5 pg) and dephosphorylated acetyl-coA carboxylase (5 pg) were subjected to polyacrylamide gel electrophoresis in the presence of SDS (32) on 7.5% gels. The gel was stained with Coomassie Blue. A, native acetyl-coA carboxylase; €3, dephosphorylated acetyl-coA carboxyl-

- . E 2 0 0

3.0

2.0

L ‘ 0 1 2.5 5 10

Citrate (mMI

FIG. 6. Citrate dependence of phosphorylated and dephosphorylated acetyl-coA carboxylase. Carboxylase activity of phosphorylated and dephosphorylated acetyl-coA carboxylase was determined using 0.2 pg of carboxylase protein as described under “Materials and Methods.” The citrate concentration was varied as indicated. Inset, ratio of activities of the dephosphorylated form to phosphorylated form was plotted against citrate concentration.

limiting value a t 2.5 mM citrate, but the magnitude of activation was much less as shown in Fig. 6. I t is to be noted that a t 10 mM citrate, the concentration used in the standard carboxylase assay, the difference in activity between the native form and dephosphorylated form was less apparent (3.0 and 3.5 units/mg, respectively). On the other hand, when the carboxylase assay was carried out in the absence of citrate, a signifxant difference in activity between the native and dephosphorylated form was observed 0.3 uers’sus 2.6 units/mg, respectively (Fig. 6). The latter activity is about 87% of that of the native enzyme in the presence of 10 mM citrate. The relative sensitivities of the two forms of carboxylase to various citrate concentrations were best observed in their activity ratios as depicted in Fig. 6 (inset). It is apparent that the highest activity ratio of dephosphorylated to phosphorylated form was observed in the absence of citrate. As the citrate level was increased, the ratio dropped rapidly, and at concentrations above 2 mM citrate, the ratio was consistently less than 2, and at 10 mM citrate, the activity ratio was only 1.3.

DISCUSSION

Phosphorylation as well as dephosphorylation of animal acetyl-coA carboxylase have been reported by several inves- tigators. There is general consensus that the carboxylase protein is a phosphoprotein containing several phosphate groups/subunit. For instance, the carboxylase from chicken liver has 4.9 f 0.2 mol of Pi/mol of subunit (6); rabbit mammary gland enzyme, 5.9 mol of Pi/subunit (17); rat mammary gland, 15 mol of PJsubunit (27); and rat liver, 5.5 mol of PJsubunit (7). But there has been considerable controversy regarding the relationship between phosphorylation state and catalytic activity (13-15, 19, 28).

Systematic investigations of the role of phosphorylation in the regulation of acetyl-coA carboxylase have been carried out in our laboratory. A rapid procedure for the isolation of highly purified preparations of acetyl-coA carboxylase was developed. Electrophoresis of such preparations on sodium dodecyl sulfate-polyacrylamide gels showed the presence of one protein band with an estimated molecular weight of 250,000. Phosphate analysis of the protein showed the presence of about 8 mol of Pi/mol of subunit which is within the ranges reported earlier (7, 27). The enzyme has little activity

ase. in the absence of citrate. However, upon addition of citrate,


it undergoes a rapid activation (10-fold). The half-maximal activation was observed at about 2.0 mM, which is within the reported values of 2-10 mM citrate, depending upon the source of enzyme (28, 29). However, intracellular citrate concentration is about 0.6 mM (33), and since most of this citrate is in the mitochondria, the citrate concentration in the cytosol where the carboxylase is located should be lower than 0.6mM. Therefore, the observed for citrate activation must be regarded as very high, and hence the physiological role of citrate in the regulation of carboxylase has been questioned (34).

Using these carboxylase preparations as substrates, we were able to isolate an activating protein from crude liver extract. The protein requires Mn2+ for its activation of the carboxylase, a process which is accompanied by the release of phosphate from the carboxylase. Hence, the protein is a phosphoprotein phosphatase. No detectable proteolysis of the carboxylase subunit was noted after incubation with the phosphatase, indicating that activation was not due to proteolysis as reported earlier for limited tryptic digestion of the carboxylase (30).

This phosphatase is different from the “carboxylase phosphatase” reported earlier by Krakower and Kim (16). The Krakower and Kim acetyl-coA carboxylase phosphatase has no metal ion requirement and has an estimated molecular weight of 71,000. An interesting difference between their phosphatase and the one described in this report is concerned with the requirement for citrate by the activated product carboxylase. These authors’ phosphatase yields acetyl-coA carboxylase which is dependent on citrate for its activity (16), even though crude preparations of their phosphatase con- verted carboxylase to a form that was very sensitive to physiological levels of citrate (11,31). Dephosphorylation of acetyl- CoA carboxylase by our purified phosphatase converts it from a form that has low activity (0.3 unit/mg) to a form that has high activity (2.6 units/mg) in the absence of citrate. The latter activity is comparable to that obtained in the presence of 10 mM citrate prior to dephosphorylation of the carboxylase. Addition of citrate at concentrations lower than 2 mM will further stimulate carboxylase activity as shown in Fig. 6. Thus, dephosphorylation of the carboxylase by our purified phosphatase results in elimination or considerable decrease in citrate requirement. To our knowledge, this is the first report of a carboxylase from an animal source that has substantial catalytic activity independent of citrate. The relationship between phosphorylation/dephosphorylation and citrate requirement may be interrelated in a manner similar to glycogen synthetase requirement for glucose 6-phosphate (35) or the phosphorylase requirement for AMP (36). Glucose 6- phosphate is an allosteric activator of glycogen synthetase (D). However, conversion of glycogen synthetase (D) to glycogen synthetase (I) by dephosphorylation eliminates or re- duces the requirement for glucose 6-phosphate. Similarly, glycogen phosphorylase b is allosterically activated by AMP, whereas, after phosphorylation to phosphorylase a, it loses its AMP requirement. With these examples in mind, it is tempt- ing to suggest that the acetyl-coA carboxylase preparation, which is citrate-dependent, be referred to as carboxylase (D) to distinguish it from the citrate-independent form, carboxylase (I) . Since carboxylase (D) is a substrate for the purified

phosphatase, we would like to attribute the name acetyl-coA carboxylase (D) phosphatase to our enzyme. In uiuo, the two forms (carboxylases D and I) may be interconvertible and, therefore, play a role in the physiological regulation of fatty acids. This hypothesis becomes more attractive if one consid- ers that the I form is more sensitive to citrate and at physiological citrate concentration (-0.6 mM) is fully activated (cf. Fig. 6). In this way, citrate becomes an important allosteric activator of acetyl-coA carboxylase and consequently a key metabolic regulator of fatty acid synthesis.

Acetyl-coA carboxylase of animal contains many phosphorylated sites (5-18 phosphates/mol of subunit) (7, 27) and, therefore, is one of the most highly phosphorylated multifunctional enzymes reported. Some of these phosphates may be important for catalytic activity, while others may be either silent (6, 18) or somehow contribute to the structure of the protein. The Mn2+-dependent phosphatase removes three phosphates/mol of carboxylase subunit; however, it is not known whether these phosphates originate from one or more subdomains of this multifunctional subunit. However, the relationship between these phosphates is not known. Clearly, further information is needed in order to identify the sites of these phosphates on the subunit and their role in regulating catalytic activity.

REFERENCES

2. Wakil, S. J., Titchener, E. B., and Gibson, D. M. (1958) Biochim. Biophys. 1. Wakil, S. J. (1958) J. Am. Chem. Soc. 8 0 , 6 4 6 5

3. Guchhait, R. B., Zwergel, E. E., and Lane, M. D. (1974) J . Biol. Chem. Acta 2 9 , 2 2 5

249.4776-4780 4. Inoue, H . , and Lowenstein, J. M. (1972) J . B i d . Chem. 2 4 7 , 4825-4832 5. Beaty, N. B., and Lane, M. D. (1982) J. Bid. Chem. 257,924-929 6. Wada. K.. and Tanabe. T. (1983) Eur. J. Biochem. 135. 17-23 7. Witters, L. A,, and Vogt, B. (1981) J. Lipid Res. 22, 364-369 8. Hardie, D. G., and Cohen, P. (1978) FEBS Lett. 91, 1-7

10. Beaty, N. B., and Lane, M. D. (1983) J. B i d . Chem. 258, 13051-13055 9. Beaty, N. B., and Lane, M. D. (1983) J. Biol. Chem. 258, 13043-13050

11. Carlson, C. A,, and Kim, K.-H. (1973) J . Biol. Chem. 248,378-380 12. Lent, B., and Kim, K:H. (1982) J. Bzol. Chem. 257,1897-1901 13. Tipper, J . P., and Witters, L. A. (1982) Biochim. Biophys. Acta 7 1 5 , 162-

1 fi9 14.

15.

16. 17. 18. 19.

20. 21.

Tipper, J. P., Bacon, G. W., and Witters, L. A,, (1983) Arch. Biochem. ””

Pekala, P. H., Meredith, M. J., Tarlow, D. M., and Lane, M. D. (1978) J . Biophys. 227,386-396

Biol. Chem. 253.5267-5269 Krakower, G. R., and Kim, K.-H. (1981) J . Biol. Chem. 256,2408-2413 Hardie, D. G., and Cohen, P. (1979) FEES Lett. 1 0 3 , 333-338 Song, C. S., and Kim, K.-H. (1981) J . Biol. Chem. 256, 7786-7788 Lane, M. D., Moss, J., and Polakiss, E. (1974) Curr. Top. Cell. Regul. 8 ,

Waite, M., and Wakil, S. J. (1962) J. Biol. Chem. 2 3 7 , 2750-2757 Ullman, B., and Perlman, R. C. (1975) Biochim. Biophys. Acta 403, 393-

139-195

A1 1 22. Applebury, M. L., Johnson, B. P., and Coleman, d . E. (1970) J . Biol. Chem.

23. Antoniw, J. F., Nimmo, H. G., Yeaman, S. J., and Cohen, P. (1977) Biochem. 2 4 5 , 4968-4976

24. Hess, H. H., and Derr, J. E. (1975) A n d . Biochem. 63 607-613 25. Lowrv. 0. H.. Rosebroueh. N. d.. Farr. A. L.. and RaAdall. R. J. (1951) J.

J . 162,423-433

Bid: Chem: 193,2651275 26. Bartosek, I., Guaitani, A., and Miller, L. L. (eds) (1973) Isolated Liver

27. Ahmad, F., and Ahmad: P . M. (1981) Methods Enzymol. 7 1 , 16-26 Perfusion and its A p lication, pp. 219-223, Raven Press, New York

28. Kim, K.-H. (1979) Mol. Cell. Biochern. 28, 27-43 29. Numa, S., and Yamashita, S. (1974) Curr. Top. Cell. Regul. 8 , 197-246 30. Guy, P . S., and Hardie, D. G . (1981) FEBS Lett. 132,67-70 31. Carlson, C. A,, and Kim, K.-H. (1974) Arch. Biochem. Biophys. 1 6 4 , 490-

32. Laemmli, U. K. (1970) Nature (Lond.) 227,680-685 33. Siess, E. A., Brocks, D. G., and Wieland, 0. H. (1978) Hoppe-Seyler’s 2.

Physiol. Chern. 359,785-798 34. Wakil, S. J., and Barnes, Jr., E. M. (1971) in Comprehensive Biochemistry

Holland Biochemical Press. Amsterdam (Florkin, M., and Stotz, E. H., eds) Vol. 185, pp. 57-104, Elsevier/North-

. , . ,~

501

35. Larner, J., and Villar-Palasi, C. (1971) Curr. Top. Cell. Re&. 3 , 196-236 36. Fischer, E. H. Heilmeyer, L. M. G., Jr., and Haschke, R. H. (1972) Curr.

Top. Cell. Regul. 4, 211-251

Documents

Activation of Acetyl-coA Carboxylase