Vol. 122, No. 2, 1984 July 31, 1984
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 694-699
MOLECULAR WEIGHTSOF SUBUNITS OF ACETYLCoA CARBOXYLASE IN RAT LIVER CYTOPLASM
Jonathan Goodson, Terri S. Pope and John B. Allred Department of Food Science and Nutrition The Ohio State University, Columbus, Ohio 43210 Received June 20, 1984
SUI~ARY: Monomeric [14C] methyl avidin was shown to bind to sodium dodecyl sulfate-denatured biotinyl proteins and remain bound through polyacrylamide gel electrophoresis which allowed their detection by fluorography. This method was used to show that purified rat liver acetyl CoA carboxylase contained two high molecular weight forms of the enzyme (MR=241,000 and 252,000) while rapidly prepared, crude rat liver cytoplasm contained two larger molecular weight (MR=257,000 and 270,000) forms. Thus, the enzyme had undergone substantial proteolysis during purification. The crude enzyme preparation also contained a smaller biotinyl protein (MR=141,000) which is likely a proteolytic product of the larger forms of acetyl CoA carboxylase.
Acetyl CoA carboxylase is a cytoplasmic, biotinyl enzyme required for catalysis of the regulatory step in fatty acid biosynthesis (1).
Early stu-
dies of purified rat liver enzyme indicated subunit MR of 215,000 (2) to 230,000 (3) which could be further converted by proteolysis into two nonidentical subunits of MR 118,000 and 125,000 (3).
In more recent reports, two
major protein bands (MR=240,O00 and 260,000) were found after SDS-PAGE when enzyme was purified from rat liver using polyethylene glycol precipitation (4) but only one major band (MR=260,O00) when an avidin a f f i n i t y column was used to purify the enzyme (5).
Regardless of the method of purification, determination
of the subunit molecular weights of the purified enzyme may not reflect those of the native enzyme because the process may either alter its structure or favor the purification of one form preferentially. We have developed a method to detect and determine the subunit molecular weights of the forms of acetyl CoA carboxylase in crude as well as purified enzyme preparations.
Two other methods have been developed to detect biotinyl
Abbreviations used: SDS, sodium dodecyl sulfate; electrophoresis. 0006-291X/84 $1.50 Copyrzght © 1984byAcademic Press, inc. Allrigh~ofreproductionin anyform reserved
694
PAGE,polyacrylamide gel
Vol. 122, No. 2, 1 9 8 4
BIOCHEMICAL AND BIOPHYSICALRESEARCHCOMMUNICATIONS
proteins in plant (6) or l i v e r (7) preparations, both of which depend upon detection of the biotinyl proteins after they have been separated by electrophoresis and transferred to nitrocellulose.
Our method depends upon the
reaction of [14C] methyl avidin with subunits of biotinyl enzymes before SDSPAGE. Sufficient radioactive avidin remains bound after electrophoresis for detection of the biotinyl proteins by fluorography.
The results indicate that
there are two subunit molecular weight forms of the enzyme in both purified and rapidly prepared mitochondrial-free supernatant but that the subunits are considerably larger in the crude compared to the purified preparation. MATERIALS AND METHODS SAMPLE PREPARATION: Purified acetyl CoA carboxylase was prepared using polyethylene glycol precipitation in the presence of protease inhibitors as previously described (4). The phosphorylated enzyme was prepared by incubating the purified enzyme with [y-32p] ATP and purified cyclic-AMP dependent protein kinase catalytic subunit (4). Crude enzyme was prepared from l i v e r of fed rats killed by decapitation. The liver was quickly removed and cooled on ice before i t was homogenized in 1.5 volumes of 0.3 M mannitol using a Potter-Elvehjem homogenizer (8). The homogenate was centrifuged in the cold for 2 minutes at 16,000 X g using an Eppendorf table top centrifuge. An aliquot of supernatant was quickly added to two volumes of hot (95 ° ) SDS-denaturing reaction mixture. Purified as well as crude enzyme was dissociated into subunits by heating (95 ° ) for four minutes in the presence of SDS (3%), 2-mercaptoethanol (2%), bromthymol blue indicator and a small amount of sucrose. After heating, the samples were cooled to ambient temperature and a small aliquot was incubated with avidin (phosphorylated enzyme) or [14C] methyl avidin in 50 mM sodium phosphate buffer (pH 7.4) for four hours at 37° . The samples were then d i l u t ed with I/2 volume of stacking gel buffer (0.15M Tris Chloride, pH 6.8) and refrigerated overnight. ELECTROPHORESIS AND FLUOROGRAPHY/RADIOAUTOGRAPHY: Samples were subjected to slab SDS-PAGE using a 3% polyacrylamide stacking gel and a 5% separating gel (9). After electrophoresis overnight, the gels were stained with Coomassie Blue R-250 and destained with methanol-acetic acid. For detection of [14C] by fluorography, the gels were treated with a s c i n t i l l a t i o n solution (Enlighten) before drying. For detection of [32p], gels were dried after destaining. The dried gels were then exposed to X-ray film (XAR-2) overnight which was subsequently developed according to manufacturer's directions. MATERIALS: Scintillation solution (Enlighten), y-[32p]-ATP and [14C] methyl avidin (New England Nuclear, Boston), SDS and electrophoresis chemicals (BioRad, Richmond, California) and X-ray film (Eastman-Kodak, Rochester) were obtained from the indicated sources. Avidin and protein molecular weight standards, with values of subunit molecular weights, were obtained from Sigma Chemical Co., St. Louis, except that purified rabbit mammary gland fatty acid synthetase was a generous g i f t of A. D. McCarthy and D. G. Hardie. RESULTS
Rat liver acetyl CoA carboxylase, purified by polyethylene glycol precipitation, was found to contain two major protein bands after SDS-PAGE (4). 695
Vol. 122, No. 2, 1 9 8 4
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
The obvious question was whether both proteins were acetyl CoA carboxylase. When an SDS-denatured preparation of the purified enzyme was incubated with [14C] methyl avidin before SDS-PAGE, i t was shown that both protein bands contained radioactivity (Figure i ) .
Non-specific binding of the radioactive avi-
din to proteins was shown not to account for the results because free biotin added to the avidin before, during and after the incubation with the denatured protein resulted in the lack of any radioactivity above the dye front (results not shown). The result (Figure 1) not only indicated that both protein subunits contained biotin and therefore likely represented different forms of acetyl CoA carboxylase, i t also demonstrated that avidin would bind to the biotinyl moiety of proteins even in the presence of SDS with sufficient radioactivity staying bound through SDS-PAGE to allow detection of the biotinyl proteins. Further evidence that avidin would remain bound to biotinyl protein subunits through SDS-PAGE was obtained using [32p]-labelled acetyl CoA carboxylase (Figure 2).
Conditions were chosen so that both of the major protein
bands were phosphorylated (4) and therefore could be detected by radioautography. After phosphorylation, the SDS-denatured proteins were incubated without (Lane 1) or with (Lane 2) non-radioactive avidin.
Estimation of subunit
molecular weights, using protein standards (Figure 3), gave values of 241,000 and 252,000 for the proteins without avidin (Figure 2, Lane 1) and 257,000 and 268,000 for the biotinyl protein:avidin complexes (Figure 2, Lane 2).
Thus,
reaction with avidin increased the molecular weight by about 16,000 which corresponds to the molecular weight of monomeric avidin (10). When mitochondrial-free supernatant was incubated with [14C] methyl avidin followed by SDS-PAGE, a large number of protein bands, stained with Coomassie Blue, are observed but only three radioactive bands were detected (Figure 2, Lane 3).
The two larger biotinyl protein subunits (MR=257,000 and 270,000
after subtraction of the molecular weight of monomeric avidin) are likely forms of acetyl CoA carboxylase because:
a) of the several biotinyl proteins
occurring in liver, only acetyl CoA carboxylase is present in the cytoplasm 696
Vol. 122, No. 2, 1984
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
",-Origin
I i
~
268 k
~
257 k
2
3
~?~
' '-~
286 k 273 k
;-~'~i :~C
%>
Origin
i ;,?;:
,'
157 k
~
it<
' "'
LL:~
/%i:, :~':;~
Front
@
~
Front .-,
Figure 1. Fluorographic analysis of purified rat l i v e r acetyl CoA carboxylase incubated with [14C] methyl avidin. Purified enzyme (3ug) was dissociated into subunits, incubated with 0.Bug [14¢] methyl avidin (65,000 DPM) and subjected to SDS-PAGE as described in Methods. The resulting gel was stained for protein, destained, soaked in scintillation solution, dried and exposed to X-ray film. Figure 2. Comparison of relative mobility of purified acetyl CoA carboxylase, purified enzyme:avidin complex and crude cytoplasmic biotinyl protein: avidin complexes. Phosphorylated acetyl CoA carboxylase (5.5ug, with 2,200 CPM protein bound) was incubated without (.Lane 1) or with 2.8~g avidin (Lane 2) and subjected to SDS-PAGE as described in Methods. Incubation for Lane 3 was the same as for Figure 1 except that SDS-denatured crude rat liver cytoplasm (equivalent to 4.4mg liver) served as the source of biotinyl proteins. Radioactivity was detected on the dried gel by radioautography for Lanes 1 and 2 and fluorography for Lane 3. The dye front is marked by unreacted [y-32p] ATP in Lanes 1 and 2 and by unreacted [14C] methyl avidin in Lane 3,
(11);
b) the other biotinyl enzymes have substantially smaller subunit mole-
cular weights than acetyl CoA carboxylase (12);
and c) the two high molecular
weight biotinyl proteins in the purified (Figure 2, Lane 2) and crude (Figure 2, Lane 3) preparations are remarkably similar with respect to each other. The subunits were larger in the crude compared to the purified preparation by 697
Vol. 122, No. 2, 1 9 8 4
BIOCHEMICAL AND BIOPHYSICAL RESEARCHCOMMUNICATIONS
7,o0
1
250 200
o_ 15o
7
o
I~
E
I00 9O 8O 7o 6O I
0
I
I
I
I
I
I
I
I
I
0.1 0.2 0.3 04 0.5 0.6 0.7 0.8 0.9 1.0 RELATIVE MOBILITY
Figure 3. Subunit molecular weights of standards and biotinyl proteins. Standard proteins (with subunit molecular weights X 10-3 ) were: A, crosslinked bovine serum albumin tetramer (266); B, fatty acid synthetase (250); C, cross-linked bovine serum albumin trimer (198); D, cross-linked bovine serum albumin dimer (132); E, urease (120); F, phosphorylase "b" (94); and G, bovine serum albumin (66). Biotinyl proteins (with subunit molecular weights X 10-3 ) were: i and 2, upper bands protein:avidin complexes in crude cytoplasm (286 and 273); 3 and 4, purified acetyl CoA carboxylase:avidin complexes (268 and 257); 5 and 6, purified acetyl CoA carboxylase (252 and 241); and 7, smaller biotinyl protein:avidin complex in crude cytoplasm (157).
about 16,000 which indicates that the enzyme underwent substantial proteolysis during p u r i f i c a t i o n
in spite of the use of protease i n h i b i t o r s .
Neither the o r i g i n nor i d e n t i t y of the smaller b i o t i n y l
protein
(MR=141,O00 a f t e r correction f o r bound avidin) has been established, although f o r the f i r s t
reason given above, i t is l i k e l y to be the b i o t i n - c o n t a i n i n g
fragment of acetyl CoA carboxylase previously shown to be produced by proteol y s i s of the larger forms of the enzyme (3). not be detected.
This smaller b i o t i n y l
The fragment without b i o t i n would
protein may be produced in l i v e r during
the normal turnover of acetyl CoA carboxylase but i t is possible that i t was formed by proteolysis during the time between decapitation of the rat and three minutes l a t e r when the homogenate was added to the hot SDS.
ACKNOWLEDGEHENT
Supported in part by Grant AM 19555, USPHS and by a grant from the Central Ohio Chapter, American Heart Association. We are also grateful to P.K. Stumpf and coworkers f o r sharing t h e i r manuscript p r i o r to publication. 698
Vol. 122, No. 2, 1 9 8 4
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
REFERENCES
1. 2. 3.
Volpe, J.J. and Vagelos, P.R. (1976) Physiol. Rev. 56, 339-417. Inoue, H. and Lowenstein, J.M. (1975) Methods Enzymo]. 35, 3-11. Tanabe, T., Wada, K., Okazaki, T. and Numa, S. (1975) Eur. J. Biochem. 51, 15-24. 4. Allred, J.B., Harris, G.J. and Goodson, J. (1983) J. Lipid Res. 24, 449-455. 5. Song, C.S. and Kim, K.-H. (1981) J. Biol. Chem. 256, 7786-7788. 6. Nikolar, B.J., Wurtele, E.S. and Stumpf, P.K. (1984) Plant Physiol. (in press). 7. Mapes, J. (1983) Fed. Proc. 42, 1854. 8. Allred, J.B. and Roehrig, K.L. (1978) J. Biol. Chem. 253, 4826-4829. 9. Hardie, D.G. and Guy, P.S. (1980) Eur. J. Biochem. 110, 167-177. 10. Moss, J. and Lane, M.D. (1971) Adv. Enzymol. 35, 321-442. 11. Wolf, B. and Feldman, G.L. (1982) Am. J. Hum. Genet. 34, 699-716. 12. Wood, H.G. and Barden, R.E. (1977) Ann. Rev. Biochem. 46, 385-413.
699