- Email: [email protected]
Regulation of branched-chain α-ketoacid dehydrogenase kinase
ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 231, No. 1, May 15, pp. 48-57, 1984
Regulation
of Branched-Chain RALPH
Department
of Biochemistry, Received
PAXTON Indiana
September
a-Ketoacid ROBERT
AND
University
School
7, 1983, and in revised
Dehydrogenase
Kinase’
A. HARRIS’
of Medicine, form
Indianapolis,
January
Indiana
~#223
23, 1984
Isolated rabbit liver branched-chain a-ketoacid dehydrogenase was inhibited in a mixed manner relative to ATP by a-ketoisocaproate, cY-keto+methylvalerate, cu-ketoisovalerate, a-ketocaproate, a-ketovalerate, and a-chloroisocaproate with Ido values (mM), respectively, of 0.065, 0.49, 2.5, 0.2, 0.5, and 0.08. The concentration (mM) of (Yketoisocaproate, cu-keto-fi-methylvalerate, and a-ketoisovalerate needed to activate branched-chain a-ketoacid dehydrogenase in the perfused rat heart to 50% of total activity was 0.07, 0.10, and 0.25, respectively. Isolated branched-chain cY-ketoacid dehydrogenase kinase was inhibited (Ido values, mM) by octanoate (0.5), acetoacetyl-CoA (O.Ol), methylmalonyl-CoA (0.2), NADP+ (1.5), and heparin (12 pg/ml). The kinase activity, in the presence or absence of ADP, was inhibited approximately 30% by 0.1 mM isobutyryl-CoA, isovaleryl-CoA, and malonyl-CoA, while not affected by NAD+ and NADH (1 mM), CoA, acetyl-CoA, methylcrotonyl-Cob, crotonyl-CoA, @hydroxy-/Imethyl-glutaryl-CoA, octanoyl-Cod, succinyl-CoA, and propionyl-CoA (0.1 mM). The following compounds at 2 mM also did not inhibit branched-chain a-ketoacid dehydrogenase kinase; acetate, propionate, P-hydroxybutyrate, lactate, acetoacetate, malonate, cu-ketomalonate, succinate, citrate, oxaloacetate, FAD, and NADPH. These findings help explain the unique effects of Leu compared with Val and Ile on branched-chain amino acid metabolism and the differences between control of the kinases associated with pyruvate dehydrogenase and branched-chain cu-ketoacid dehydrogenase.
Branched-chain cy-ketoacid dehydrogenase (EC 1.2.4.4), an intramitochondrial multienzyme complex, catalyzes the oxidative decarboxylation of the transamination products (i.e., KIC,3 KMV, and KIV) of the branched-chain amino acids, Leu, Ile, and Val. There is a linear correlation between blood branched-chain amino acids levels and blood branched-chain cY-ketoac-
ids levels (l), presumably due to a nearequilibrium transaminase in most tissues (2). The first nonreversible, rate-limiting enzyme in branched-chain amino acid catabolism is branched-chain cy-ketoacid dehydrogenase (3). Thus, the activity of this complex will determine the overall catabolism of the branched-chain amino acids. The activity of branched-chain cy-ketoacid dehydrogenase can be controlled (i.e., inhibited) by phosphorylation. This has been shown with the isolated complex (4-6) which has a copurifying kinase with several charateristics similar to the pyruvate dehydrogenase (EC 1.2.4.1) kinase (4). The specific intramitochondrial phosphatase that dephosphorylates and activates branched-chain a-ketoacid dehydrogenase has not been isolated or characterized, although a broad-specificity
i Supported in part by NIH Research Grants AM19259 and 5 SO7 RR5371 and the Grace M. ShoWalter Residuary Trust. ’ To whom correspondence should be addressed. a Abbreviations used: KIC, a-ketoisocaproate; KMV, a-keto-P-methylvalerate; KIV, cu-ketoisovalerate; EGTA, ethylene glycol bis(@-aminoethyl ether)N,N,N’,N’-tetracetic acid; Hepes, I-(2-hydroxyethyl)l-piperazineethanesulfonic acid; and TLCK, Nor-ptosyl-L-lysinechloromethyl ketone. 0003-9861/84 Copyright All rights
$3.00
0 1984 by Academic Press, Inc. of reproduction in any form reserved.
48
REGULATION
OF
BRANCHED-CHAIN
a-KETOACID
protein phosphatase has been identified (7) and utilized (8, 9) to show the alteration in phosphorylation state of the complex from several tissues of rats under various physiological treatments. The different branched-chain amino acids, although catabolized similarly through branched-chain Lu-ketoacid dehydrogenase, have unique effects on several metabolic processes. For example, Leu feeding decreases blood levels of Val and Ile, while feeding either of the latter have little effect on blood levels of Leu (10-12). Leu and KIC, but not Val, KIV, Ile, or KMV, are also insulinotropic agents (13-15) and regulators of protein synthesis/degradation (16, 17). Similarly, Leu or KIC have been shown to activate the complex in various tissue extracts (6, 18, 19). Phosphorylation of highly purified liver complex was inhibited to different degrees by the various branched-chain a-ketoacids (0.1 mM) with KIC > KMV > KIV (8). The partially purified complex from ox kidney, which contained a protein phosphatase and an ATPase (20), showed a similar effect of the branched-chain a-ketoacids on inhibiting the ATP-mediated inactivation of the complex (21). This study reports the relative effects of the different branched-chain a-ketoacids on inhibiting isolated branched-chain (Yketoacid dehydrogenase kinase and on activating branched-chain cu-ketoacid dehydrogenase in perfused rat heart. The influence of several other factors on isolated branched-chain a-ketoacid dehydrogenase kinase, particularly intramitochondrial metabolites associated with branchedchain cu-ketoacid catabolism, were also measured. Since this complex is very similar to pyruvate dehydrogenase, the results are also discussed in regards to similarities of these two complexes. EXPERIMENTAL
PROCEDURES
Materials. Reagents were obtained as previously given (4,8) or from Sigma Chemical Company. Bovine insulin was from Calbiochem. a-Chloroisocaproate was a gift from Dr. Robert J. Strohscheim and Dr. Ronald Simpson of Sandoz, Inc. Branched-chain a-ketoacid dehydrogenme isolation and assay. Rabbit liver branched-chain a-ketoacid
DEHYDROGENASE
KINASE
49
dehydrogenase with intrinsic kinase activity was isolated as previously described (4), with a final specific activity of 4 to 4.5 pm01 min-’ mg protein-’ at 30°C with 1 mM KIV. The spectrophotometric assay of the complex was done as previously described (4). Branched-chain c&&acid dehydrcgenme kimse assay. Branched-chain a-ketoacid dehydrogenase kinase was assayed by protein-bound %P in a final volume of 50 pl containing 30 mM Hepes-K, 1.5 mM MgCh 2 mM dithiothreitol, 0.2 mM EGTA (pH 7.34 at 2O”C), 2.72 pg branched-chain a-ketoacid dehydrogenase kinase complex, [y-32P]ATP (sp act 600-800 cpm/pmol) at the indicated concentration, and other additions as given in the text. As shown previously (4), =P incorporation was limited to the a-subunit of the aketoacid decarboxylase component of branched-chain a-ketoacid dehydrogenase. The reaction was conducted at 37°C for 20 min unless indicated. Determination of protein-bound %P was done as previously described (4). Isolatim of broad-spec$city protein phosphatase. Broad-specificity protein phosphatase was isolated as previously given (‘7), except from rabbit liver and the peak elution from a DEAE-cellulose column was at approximately 0.2 M NaCI. Phosphatase activity was measured by “Pi release (see below) in 165 ~1 total volume that combined 16.4 pg 3zP-labeled branchedchain a-ketoacid dehydrogenase (sp act approx 25,000 cpm/pg), 0.3 mg bovine serum albumin, 1.5 mM MnClz, 16 mM imidazole-Cl, 9 mM Hepes-K, 93 pM EDTA, 0.44 mM dithiothreitol, and 26 fig broad-specificity protein phosphatase (all at pH 7.5 at 20°C). A 30-~1 aliquot of the above mixture was mixed with 20 pl of 50% (w/v) trichloroacetic acid in an Eppendorf conical centrifuge tube, placed on ice for 15 min, and centrifuged in an Eppendorf centrifuge for 3 min. The supernatant, 40 ~1, was placed on filter paper and the radioactivity was determined (4). Phosphatase activity was hyperbolic with time and gave >90% release by 35 min at 37’C. Heart perfusions. Hearts from 300- to 350-g male Wistar rats were perfused by the flow-through Langendorff procedure as previously described (22). Hearts were perfused at 37°C for 30 min with KrebsHenseleit buffer containing 10 mM glucose, 20 mu/ ml insulin, and additions as given. Determination of percentage active fm branchedchain a-ketoacid dehydrogenase in Mused rat heart. Hearts were rapidly freeze-clamped at liquid Nz temperature and stored at -70°C. The frozen rat hearts were weighed and homogenized (Polytron at setting 4 and for 45 s) in approximately 8 vol of ice-cold 2% Triton X-100,0.2 mrd thiamine pyrophosphate, 10 mM EGTA, 10 mM EDTA, 5 mM dithiothreitol, 0.5 mM achloroisocaproate, 50 m&i Kpi, 0.5 pM leupeptin, 0.5 mM TLCK, 0.5 PM pepstatin A, 2.5 mM benzamidine, and 0.1 mg/ml trypsin inhibitor from egg white (Sigma Chemical Co., Type II) at a pH of 7.5 at 20°C.
PAXTON
50
AND
Homogenates were centrifuged at 29,OOOg (average) for 15. min at 0-2°C in a Sorvall SS-34 rotor. Supernatants were centrifuged at 165,OOOg (average) for 90 min at 0-2°C in a Beckman Ti50 rotor. Pellets were resuspended in 1.5 ml of ice-cold 50 mM Hepes-K, 2 mM dithiothreitol, 0.2 rnEb EGTA, and 0.1% Triton X100, pH 7.5, at 20°C. This extract was then assayed spectrophotometrically for branched-chain cu-ketoacid dehydrogenase activity with 1 mM KIV (4). Another portion of this extract was treated with broad-specificity protein phosphatase by incubating 75 ~1 of extract with 0.1 ml of phosphatase (0.26 mg protein) and 10 pl of 50 mM MnCl,. This mixture was incubated at 37°C for 30 and 45 min and then assayed for total branched-chain cy-ketoacid dehydrogenase activity. The percentage active form was determined by dividing activity before treatment by activity after treatment with broad-specificity protein phosphatase and multiplying by 100. Kinetic analyses and all statistical test and associated tables were as previously given (4).
HARRIS
FIG. 2. Lineweaver-Burk plot showing branched-chain cy-ketoacid dehydrogenase KIV. Same conditions as in Fig. 1.
inhibition kinase
of by
RESULTS
Branched-chain a-ketoacid dehydrogenase kinase was inhibited by KIC (Fig. l), KIV (Fig. Z), KMV, a-ketocaproate, a-ketovalerate, and a-chloroisocaproate relative to ATP. The data for the latter four compounds are not shown but were ob-
FIG. 1. Lineweaver-Burk branched-chain a-ketoacid tivity (nmol a? 20 min-’ various concentrations of of slopes and y intercepts versus KIC concentration. scribed under Experimental
plot showing inhibition of dehydrogenase kinase acmg protein-‘; y axis) by KIC. Insert shows replot of Lineweaver-Burk plot Assay conditions as deProcedures.
tained in identical fashion and gave the same inhibition pattern as shown for KIC and KIV. The reciprocal plots of velocity versus ATP concentration were linear with various concentrations of each inhibitor. However, the lines with various inhibitor concentrations did not appear to have a common intersection. Replots of slopes and y intercepts with inhibitor concentrations (see inserts of Fig. 1 and 2) were nonlinear with an apparent hyperbolic shape. This type of inhibition would appear to be slopehyperbolic, y intercept-hyperbolic noncompetitive inhibition. This type of inhibition is also seen with isolated pyruvate dehydrogenase kinase with respect to pyruvate and dichloroacetate (23). The complexity of the initial plots, however, limit definitive determination of the mechanism of inhibition. Thus, defining the inhibition with a more general term such as mixed is probably more appropriate. Further characterization of the mechanism of inhibition by these compounds is beyond the scope of this paper. Since the mechanism of inhibition by these compounds toward the kinase is not known, the relative inhibitory abilities of the compounds can not be definitely determined. However, the relative potency of these compounds to inhibit the kinase was
REGULATION
OF
BRANCHED-CHAIN
wKETOACID
estimated by selecting only a single ATP concentration (75 PM). While these determinations do not provide a substrate (ATP) concentration-free estimate of relative inhibitory abilities, they do allow a comparison of the relative potency of the various compounds to inhibit the kinase. It should be pointed out that these estimates of the relative inhibitory abilities of several compounds tested were consistent with the relative potency of these compounds to inhibit the kinase in the perfused rat heart (see below). I40 values (amount of compound needed to inhibit 40% of the total kinase activity; Fig. 3) were selected since the inhibition was nonlinear and it appeared that the inhibition pattern at 40% inhibition was fairly uniform among the various compounds. Ido values for several other compounds (Table I) not given above were also determined, although the type of in-
FIG. 3. Inhibition (I,, values) of branched-chain (Yketoacid dehydrogenase (BCKDH) kinase by KIV (A), KMV (B), and KIC (C). BCKDH kinase was assayed as given in Fig. 1, except assay time was 30 min and ATP concentration was 75 FM, with other additions as given. Under these conditions 100% BCKDH kinase activity (nmol 32p bound 30 min-’ mg protein-‘) was 5.13 rt 0.13 (jt f SE) for 33 determinations.
DEHYDROGENASE TABLE
KINASE
51
I
Ido VALUES FOR VARIOUS COMPOUNDS WITH ISOLATED BRANCHED-CHAIN (Y-KETOACID DEHYDROGENASE KINASE“ Compounds
Lo (rnM) 0.065 0.08 0.20 0.49 0.50 2.0 2.5 0.01 0.20 0.50 1.5
a-Ketoisocaproate a-Chloroisocaproate a-Ketocaproate a-Keto-fl-methylvalerate a-Ketovalerate a-Ketoadipate or-Ketoisovalerate Acetoacetyl-CoA Methylmalonyl-CoA Oetanoate NADP+ Heparin (1 Conditions
12 pg/ml as given
in Fig.
3.
hibition relative to ATP was not determined. KIC was about 7.5- and 38-fold more potent at inhibiting branched-chain a-ketoacid dehydrogenase kinase than KMV and KIV, respectively. a-Chloroisocaproate, a known inhibitor of isolated branchedchain Lu-ketoacid dehydrogenase kinase and activator of the complex in perfused rat heart (8), was almost as potent as KIC at inhibiting isolated branched-chain a-ketoacid dehydrogenase kinase. Sensitivity of the kinase to inhibition decreased as the carbon length (e.g. KIV) and structure (e.g. KMV) of the inhibitor varied from KIC. In a comparison of various straight-chain (Yketoacids (i.e., pyruvate, cY-ketobutyrate, LYketovalerate, a-ketocaproate, a-ketooctanoate, and cY-ketononanoate), all at 2 InM, maximum inhibition was seen with carbon length of 6, with diminishing potency occurring as the carbon length changed from 6 (data not shown). Besides the a-ketoacid, several other compounds were inhibitors of the branchedchain cY-ketoacid dehydrogenase kinase in a concentration-dependent manner (Table I). Although probably not of physiological importance, heparin was a very effective inhibitor of the kinase. Inhibition by heparin links branched-chain cr-ketoacid dehydrogenase kinase with a group of protein
52
PAXTON
AND
kinases such as casein kinase II (24) and glycogen synthase kinase P&T (25). However, numerous other protein kinases (e.g. casein kinase I, type I and type II CAMPdependent protein kinases, protease-activated kinase I, and the hemin-controlled repressor kinase) are not inhibited by heparin (24). What physiological role the inhibition of branched-chain a-ketoacid dehydrogenase kinase by cY-ketoadipate, an intermediate of lysine catabolism, and NADP+ (Id,, values of 2.0 and 1.5 mM, respectively; Table I) may have is not clear. Inhibition of branched-chain a-ketoacid dehydrogenase kinase by acetoacetyl-CoA and methylmalonyl-CoA, with Ido values of 0.01 and 0.2 mM, respectively, may have physiological significance (see Discussion) since both are distal metabolites of branched-chain amino acid oxidation and acetoacetyl-CoA is formed from ketone bodies in peripheral tissues. Octanoate inhibition of branched-chain cu-ketoacid dehydrogenase kinase (Ido value, 0.5 mM) may have little physiological significance since octanoate is not a major energy source nor tissue metabolite. However, octanoate is frequently used to determine the effects of fatty acid oxidation on branched-chain amino acid and branched-chain cy-ketoacid metabolism (see Discussion). Octanoyl-CoA had no effect on isolated branched-chain a-ketoacid dehydrogenase kinase activity (Table II). The following compounds at 2 mM had no effect (~10%) on branched-chain cy-ketoacid dehydrogenase kinase activity measured with 0.2 mM ATP as substrate: acetate, propionate, P-hydroxybutyrate, y-hydroxybutyrate, isobutyrate, lactate, acetoacetate, malonate, cu-ketomalonate, succinate, citrate, oxaloacetate, NADPH, NAD+, and NADH (data not shown). Of particular interest is the lack of direct effect of acetoacetate and /?-hydroxybutyrate since both have been suggested to have an influence on branched-chain amino acid metabolism (26; see Discussion). Isolated pyruvate dehydrogenase kinase is inhibited by substrates (CoA and NAD+) and activated by products (acetyl-CoA and NADH) of the pyruvate dehydrogenase-
HARRIS TABLE EFFEGT CHAIN
II
OF VARIOUS COMPOUNDS ON BRANCHEDWKETOACID DEHYDROGENASE KINASE ASSAYED WITHORWITHOUT ADP”
Compound
Concentration (m@
None NAD+
1.0
NADH NAD+ + NADH NAD+ + NADH CoA Acetyl-CoA Acetoacetyl-CoA Methylmalonyl-CoA Succinyl-CoA Octanoyl-CoA Isovaleryl-CoA Isobutyryl-CoA Malonyl-CoA
1.0
0.75 + 0.25 0.25 + 0.75 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
Percentage of BCKDH kinase activity -ADP
+ADP
100 96
100
99 91 95
93 112 102 100
98 102
106 102
56 64
44 70
110 100 70
60 74
109 110 89
84 71
a Conditions as given in Fig. 3, except ATP and ADP concentrations were 0.05 and 0.5 mM. ADP inhibited branched-chain wketoacid dehydrogenase (BCKDH) kinase activity by 50% under these conditions.
catalyzed reaction, but only in the presence of ADP (27). The influence of various intramitochondrial metabolites on branchedchain a-ketoacid dehydrogenase kinase activity were measured in the presence and absence of ADP (Table II). Several compounds not shown in Table II had ~10% effect on branched-chain cY-ketoacid dehydrogenase activity with or without ADP. These included NADPH (1 mM); and methylcrotonyl-CoA, valeryl-CoA, propionylCoA, oleoyl-CoA, P-hydroxy+methylglutaryl-CoA, and crotonyl-CoA (all at 0.1 mM). NAD+, NADH, combinations of both NAD+ and NADH, CoA, and acteyl-CoA were also without effect on branched-chain a-ketoacid dehydrogenase kinase activity with or without ADP. ADP did not seem to change the direction or degree of inhibition promoted by the various compounds. Of particular note in this group of metabolites (Table II) is the inhibition of the kinase by the immediate CoA-ester
REGULATION
OF
BRANCHED-CHAIN
LU-KETOACID
products of branched-chain cy-ketoacid dehydrogenase-catalyzed reaction, i.e., isovaleryl-CoA and isobutyryl-CoA, and by a more distal product, methylmalonyl-CoA. However, the final products of branchedchain amino acid oxidation, succinyl-CoA and acetyl-CoA, both intermediates of the citric acid cycle, had no effect on the isolated branched-chain cr-ketoacid dehydrogenase kinase activity. KIC and acetoacetyl-CoA appear to be better inhibitors of the kinase than methylmalonyl-CoA, malonyl-CoA, isobutyryl-CoA, or isovaleryl-CoA. The perfused rat heart was used to verify relative potency of different branchedchain a-ketoacids to inhibit branchedchain a-ketoacid dehydrogenase kinase as revealed by the percentage active form of the complex (Fig. 4). Perfused rat heart was used because conditions had already been determined that showed maximum inhibition of extracted branched-chain (Yketoacid dehydrogenase activity (22, 28). Using these conditions (perfusion with 10
FIG. 4. Activation (A, values) of branched-chain cY-ketoacid dehydrogenase (BCKDH) in perfused rat hearts by KIC (A), KMV (B), and KIV (C). Perfusions of hearts, extraction of BCKDH, and determination of BCKDH activity were as given under Experimental Procedures. Values given as it + SE with 3 hearts for each point except there were 12 hearts with 0.1 mM KIV. All points are significantly different from no additions at P < 0.05, based on a two-tailed Student’s t test, except with 0.1 mM KIV (C).
DEHYDROGENASE
KINASE
53
mM glucose + 20 mu/ml insulin), the percentage active form of the complex was very low, 3.1 f 0.6 (g f SE for three hearts). The total activity (nmol min-’ g wet wt-‘) of the complex, determined by adding exogenous phosphatase for 84 hearts done in a similar fashion (the combined number of hearts done in this study and (29)) was 248 f 40 (ii + S.D.). All concentrations of additions resulted in a statistically significant increase in the percentage active form of the complex except 0.1 mM KIV (4.3% + 1.0). The lowest concentration of KIC (15 PM) and KMV (75 PM) used in the perfused heart experiments increased the percentage active form of the complex to approximately 20%. Concentrations of KIC, KMV, and KIV that gave 50% activation (A& of total branched-chain a-ketoacid dehydrogenase activity were 0.07,0.11, and 0.25 mM, respectively. Several other compounds (i.e., clofibric acid, dichloroacetate, and phenylpyruvate) that inhibited isolated branched-chain (Yketoacid dehydrogenase kinase were also used to measure activation of the complex in perfused rat heart (29). There was a linear correlation (Fig. 5) between the Idovalues obtained with isolated branched-chain a-ketoacid dehydrogenase kinase and the AbO values obtained with perfused rat hearts for KIC, KIV, KMV, clofibric acid, dichloroacetate, and phenylpyruvate. For reasons not presently understood, the isolated branched-chain a-ketoacid dehydrogenase kinase was much less sensitive to KIV inhibition (Ido = 2.5 mM) than was the KIV activation in perfused rat heart (As0 = 0.25 mM). One possible explanation may involve the production from KIV oxidation of other inhibitors of branched-chain Lu-ketoacid dehydrogenase kinase (e.g. methylmalonyl-CoA) that potentiate KIV activation of the complex. DISCUSSION
The type of inhibition of branched-chain a-ketoacid dehydrogenase kinase (mixed relative to ATP) caused by several compounds is also seen with isolated pyruvate dehydrogenase kinase with respect to pyruvate and dichloroacetate (23). All of the
54
FIG. 5. Correlation between Aw values various compounds (DCA is dichloroacetate) as given in Fig. 4 and Ido values (y axis) given in Fig. 3. Significance of correlation with a two-tailed test of the correlation
PAXTON
AND
(z axis) for obtained obtained as determined coefficient.
branched-chain a-ketoacid dehydrogenase kinase inhibitors we have studied in detail are either substrates (KIC, KMV, KIC, (Yketocaproate, and a-ketovalerate) or inhibitors of branched-chain cy-ketoacid dehydrogenase (cu-chloroisocaproate (8), dichloroacetate (4, 8), clofibric acid and phenylpyruvate (29,30)). This relationship suggests that these compounds inhibit branched-chain cy-ketoacid dehydrogenase kinase by interaction at a site which binds a-ketoacid substrate. This site could be at either the a-subunit of the branched-chain a-ketoacid decarboxylase component (subunit that binds substrate and becomes phosphorylated) of the complex or through direct binding on the kinase. Octanoate, but not octanoyl-CoA, inhibited isolated branched-chain cY-ketoacid dehydrogenase kinase. Octanoate increases decarboxylation of branched-chain amino acids in rat skeletal muscle homogenates (31), perfused rat hindquarters (32, 33), isolated rat skeletal muscle mitochondria (33), rat diaphragms (34,35), and perfused rat hearts (34). However, octanoate inhibits branched-chain amino acid decarbox-
HARRIS
ylation in perfused rat liver (33) as well as in rat sciatic nerve, spinal cord, and aorta (35). Contrary to (34), octanoate inhibited KIC oxidation in perfused rat heart (36), although the effect was on flux through the complex and not on the extractable activity of the complex (28). Since octanoate can inhibit isolated branchedchain cr-ketoacid dehydrogenase kinase, which should lead to activation of the complex, then those tissues where the percentage active form of the complex is low (e.g., skeletal muscle (37)) may show a stimulation by octanoate. Those tissues with a high percentage active form of the complex (e.g., liver (9)) or where the activity state has been increased due to activation by a branched-chain a-ketoacid (e.g., heart; (28, 36) and this report) will probably show decreased branched-chain amino and a-ketoacid oxidation since octanoate oxidation will reduce substrate (NAD+ and CoA) availability and lead to product inhibition (38). Intracellular levels of CoA and various CoA esters are altered with branchedchain cY-ketoacid metabolism in perfused rat heart (39) and isolated rat hepatocytes (39,40). CoA and succinyl-CoA decrease in both isolated rat hepatocytes and perfused rat heart with KIC and KIV oxidation. Acetyl-CoA levels decrease with KIV oxidation in isolated rat hepatocytes but do not change with KIC oxidation in either isolated rat hepatocytes or perfused rat heart. P-Hydroxy+methylglutaryl-CoA levels also do not change with KIC oxidation in isolated rat hepatocytes. None of these compounds which either do not change or are lowered by KIC or KIV oxidation (i.e., CoA, acetyl-CoA, P-hydroxylP-methylglutaryl-CoA, and succinyl-CoA) had any effect on isolated branched-chain cz-ketoacid dehydrogenase kinase activity assayed with or without ADP. In isolated rat hepatocytes, KIV oxidation increases methylmalonyl-CoA, propionyl-CoA, and isobutyryl-CoA levels, whereas KIC oxidation increases methylcrotonyl-CoA and isovaleryl-CoA levels. Of those CoA esters that increased with branched-chain ketoacid oxidation, propionyl-CoA and methylcrotonyl-CoA, at 0.1 mM, had little effect
REGULATION
OF
BRANCHED-CHAIN
a-KETOACID
on isolated branched-chain cu-ketoacid dehydrogenase kinase activity, while the others (i.e., methylmalonyl-CoA, isobutyryl-CoA, and isovaleryl-CoA) inhibited the kinase by 30 to 40%. Thus, several CoA esters that are elevated with KIC and KIV oxidation also inhibited isolated branchedchain cu-ketoacid dehydrogenase kinase. This suggests that oxidation of branchedchain cu-ketoacid may lead to product activation of the complex through inhibition of its associated kinase (see below). The most potent CoA ester inhibitor of isolated branched-chain a-ketoacid dehydrogenase kinase was acetoacetyl-CoA (Ido value, 0.01 mM). Heart mitochondrial concentration of acetoacetyl-CoA increases lofold (2 to 25 pM) in ketotic diabetic rat (41). The estimated mitochondrial acetoacetylCoA concentration in heart, under ketotic diabetic conditions, is greater than the Id,-, value for the isolated branched-chain (Yketoacid dehydrogenase kinase. Thus, the complex may be activated under physiological conditions where ketone bodies are elevated, e.g. starvation and diabetic. Starvation leads to increased branchedchain amino acid oxidation in rats (42), humans (43), rat skeletal muscle (31), intact perfused rat hindquarters (44), rat diaphragms (45), rat kidney slices (45), and rat atria (46). Diabetes also leads to increased branched-chain amino acid oxidation in rat diaphragms (47) and increased uptake of branched-chain amino acid in human skeletal muscle (48). There is also a linear correlation between Leu oxidation in rat skeletal muscle and altered tissue acetoacetate content due to various physiological conditions (26). Acetoacetate and P-hydroxybutyrate (2 mM) had no effect on isolated branched-chain cY-ketoacid dehydrogenase kinase, suggesting that conversion to acetoacetyl-CoA is necessary for activation of the complex. Thus, inhibition of isolated branched-chain a-ketoacid dehydrogenase kinase by physiological concentrations of acetoacetyl-CoA suggests that this may be part of the mechanism whereby elevated branchedchain amino acid oxidation can occur under conditions of ketosis. The lowest concentrations of KIC and
DEHYDROGENASE
KINASE
55
KMV used in the perfused rat heart experiments were, respectively, 15 and ‘75 pM, and these concentrations produced an increased percentage active form of the complex (3.1 versus approximately 20%). The lowest KIV concentration used, 0.1 mM, did not change the percentage active form of the complex (4.3%). Human plasma concentrations of KIC, KMV, and KIV are 30, 22, and 8 pM, respectively, and increases lo-fold with Maple Syrup Urine disease (49). Rat plasma concentration of branched-chain a-ketoacids are similar in relative distribution as that seen in humans with KIC concentration ranging from 16 PM for untreated rats to 32 pM for highprotein fed, and diabetic rats (49). Plasma KIC, but not KMV or KIV, is, therefore, within the range observed to activate heart branched-chain a-ketoacid dehydrogenase through inhibition of its associated kinase. Dietary studies have shown that Leu feeding reduces plasma levels of Ile and Val, while Ile or Val feeding have little effect on plasma levels of the other two in humans (10, 11) and rats (12). Leu/KIC have also been shown to have an activating effect on branched-chain a-ketoacid dehydrogenase activity not seen with Ile/ KMV and Val/KIV in rat adipose tissue (19) and perfused rat heart (28). Similarly, KIC is much more potent than KIV in preventing the pyruvate-mediated decrease in activity of and flux through branched-chain a-ketoacid dehydrogenase in perfused rat heart (28,36). These observations are possibly explained by the relative sensitivity of the branched-chain cu-ketoacid dehydrogenase kinase (as seen with both the isolated complex and perfused rat heart) to the different branched-chain a-ketoacids with relative potency of KIC > KMV > KIV. Leu and KIC alter several other biological processes. Leu or KIC, but not Ile, Val, or KIV, have been shown to stimulate protein synthesis and inhibit protein degradation in heart (16), although the effect of KIC may only be on proteolysis (17). Similarly, Leu and KIC, but not KIV, can promote insulin release, although KIC is much more potent (13-15). Since both of these effects on protein synthesis/degradation
56
PAXTON
and insulin release appear to be related to catabolism of Leu and KIC, which is limited by branched-chain cu-ketoacid dehydrogenase, then an increase in the percentage active form of the complex, due to KIC inhibition of its associated kinase, may be part of the mechanism for these effects and may also explain why Leu/KIC, but not Ile/KMV and Val/KIV, have these properties. The differences between control of the kinases for pyruvate dehydrogenase and branched-chain a-ketoacid dehydrogenase may be important because of the different metabolic role that each complex plays in controlling their respective pathways. For example, with starvation and diabetes, where fatty acid oxidation, branched-chain amino acid oxidation, and gluconeogenesis are elevated, pyruvate dehydrogenase is inactivated (see (50)) in order to conserve pyruvate for gluconeogenesis. On the other hand, activation of branched-chain a-ketoacid dehydrogenase would lead to the production of precursors for gluconeogenesis and ketogenesis as well as maintenance of an energy supply independent of glucose. The inactivation of pyruvate dehydrogenase is due, presumably, to the increased NADH/NAD+ and acetyl-CoA/CoA ratios (27,50) from elevated fatty acid oxidation. It is important to note that these metabolites have little effect on the phosphorylation state of branched-chain a-ketoacid dehydrogenase, as shown in this paper for the highly purified rabbit liver complex and for the partially purified ox kidney complex (21). However, several other factors associated with starvation and diabetes were shown to inhibit the isolated branchedchain cY-ketoacid dehydrogenase kinase which may lead to activation of the complex. Thus, elevated levels of branchedchain amino acids from increased peripheral proteolysis would promote increased formation of branched-chain a-ketoacids (1, 49). The increased levels of branchedchain a-ketoacids, besides directly inhibiting the branched-chain a-ketoacid dehydrogenase kinase (particularly KIC), would also promote increased flux through the complex with subsequent production of other branched-chain a-ketoacid de-
AND
HARRIS
hydrogenase kinase inhibitors (i.e., isovaleryl-CoA, isobutyryl-CoA, methylmalonyl-CoA, and acetoacetyl-CoA). Ketone body utilization by peripheral tissues will also lead to elevated acetoacetyl-CoA levels (41), which may activate the complex though inhibition of its associated kinase. The signal to inactivate branched-chain QIketoacid dehydrogenase would be a decrease in branched-chain amino acids levels (and ketone bodies), possibly as a result of increased protein synthesis and decreased protein degradation in response to insulin. ACKNOWLEDGMENTS We wish to thank Richard Landry for technical assistance and Dr. Arthur Schulz for helpful diseussion. REFERENCES 1. MATTHEWS, MOTIL,
(1982) 2. KREBS,
D. E., SCHWARZ, H. P., YANG, J., YOUNG, V. R., AND BIER,
K.
Metabolism
31, 1105-1112.
H. A., ANDLUND,
P. (1977)
Adwan
R. O., D. M.
Enzyme
15.375-394.
Reg. 3. SHINNICK,
F. L., AND HARPER,
Biophys. 4. PAXTON,
Aeta
A. E.
(1976) Biochim
437, 477-486.
R., AND HARRIS,
J. Bid
R. A. (1982)
C~GWL
257, 14433-14439. R. (1982) Bio~hem J. 204, 353-356. H. R., LA& K. S., AND RANDLE, P. J.
5. ODESSEY, 6. FATANIA, (1981)
FEBS L&t. 132,285-289. A., PAXTON, R., AND PARKER, B&hem, Biophys. Res. Cbmmun
7. HARRIS, (1982)
R.
R. A.
107,
1497-1503. R. A., PAXTON, A. A. (1982) J. BioL
8. HARRIS, 9. GILLIM,
S. E., PAXTON,
R. A. (1982)
R., ANDDEPAOLI-ROACH,
Chem,
257,13915-13918.
R., COOK,
Biochem
G. A., AND HARRIS,
Biuphys.
Res. Cwrnmun
111, 74-81. 10.
SWENDSEID,
M.
E.,
VILLALOBOS,
W. S., AND DRENICK, Nutr. 17, 317-321. 11.
HAMBRAEUS, SHAW,
J.,
L., BILMAZES, N., AND YOUNG,
FIGUEROA,
Amer.
E. J. (1965)
C., DIPPLE, V. R.
J. Clin.
C., SCRIM-
(1976) J. Nutr.
106,
230-240. 12. SHINNICK,
F. L., AND HARPER,
.I Nutr.
A. E. (1977)
107.887-895. 13.
PANTEN, U., SCHGNBORN,
FEBS 14. LENZEN,
L&t.
VON KRIEGSTEIN, E., J., AND HASSELBLAIT,
20. 225-228. Biochem
S. (1978)
Pha
POSER, W., A. (1972)
?-mad
27,1321-
1324. 15. SENER, A., MALAISSE-LAGUE, W. J. (1983) J. Biol Chem.
F.,
AND
MALAISSE,
258,6693-6694.
REGULATION 16.
CHUA,
OF
B., SIEHL,
J. BioL
D. L., AND
Chem.
1’7. TISCHLER,
BRANCHED-CHAIN MORGAN,
H. E. (1979)
J. Biol
M., AND
Chem
19.
FRICK,
20.
MAN, H. M. (1981) .I Bid FATANIA, H. R., PATSTON, P. J. (1983) FEBS L&t. LAU,
G. P., TAI,
K.
FEBS
22.
SANS,
R. M.,
23.
(1980) Arch. PRATT, M. L.,
Chem.
25.
P.
P.
L., AND
(1981)
A.,
AND
AND
W.,
B&hem.
AND
PAUL,
H.
HARRIS,
28.
WAYMACK,
29.
PAXTON,
30.
DANNER,
Biochem (1982) 31. 32.
R.,
ADIBI,
200,336-345.
Metabolism
(1978)
J. W., ANDREED,
J. Biol.
M.
Chem
HARRIS,
Biophys.
CORKEY,
65, 575-582. S., AND
OLSON,
255,9773-9781. R.
A.
Arch
(1984)
Biophys. 34.
O., AND
Chem
WILLIAMSON,
257, 9668-9676.
42.
MEIKLE,
E. T., AND
ELSAS,
L. J.
A. W.,
PhysioL 43.
ADIBI,
S. A.,
(1982) 44.
HOKLAND,
B. (1981)
AND STANKO,
Biochim.
BUSE, M. G., BIGGERS, F., FRIDERICI, K. H., AND BUSE, J. F. (1972) J. Biol. Chem. 247,8085-8096.
Amer.
G. J. (1972)
R.
Metabolism
J.
T., AND
MORSE,
E.
L.
31, 578-588.
HUTSON, S. M., ZAPALOWSKI, C., CREE, HARPER, A. E. (1980) J. Biol. Chem.
T. C., AND
255,2418-
2426. 45. GOLDBERG, A. L., AND ODESSEY, J. Physiol. 223, 1384-1391. 46.
TISCHLER,
M. E., AND GOLDBERG,
J. Physiol. 47.
BUSE,
M.
48.
G.,
WAHREN,
(1972) 49.
HUTSON,
50.
HERLONG,
J., FELIG, S. M.,
Nutr.
Amer.
A. L. (1980)
Amer.
AND
AND
WEIGAND,
E., AND
LUFT,
R.
51, 1870-1878.
HARPER,
A. E. (1981)
Amer.
34, 173-183.
D. L., STANSBIE,
Mol.
F.,
88,1166-1175.
P., CERASI,
DENTON, R. M., RANDLE, COOPER, R. H., KERBEY, ERSON,
H.
Endocrinology
J. CZin. Invest.
S. (1975)
R. (1972)
238, E487-E493.
D. A. (1973)
676, 279-288.
KLAIN,
222, 1246-1250.
J. Clin Acta
A., WALAJTYS-
R. J., AND
MENAHAN, L. A., HRON, W. T., HINKELMAN, D. G., AND MIZIORKO, H. M. (1981) Eur. J. Biochem.
257,659-662.
PAUL, H. S., AND ADIBI, S. A. (1976) J. Nutr. 106, 1079-1088. SPYDEVOLD, 0. (1979) Eur. J. B&hem. 97, 389SPYDEVOLD,
B. E., MARTIN-REQUERO,
41.
394. 33.
RANDLE,
RODE, E., WILLIAMS, J. R. (1982) J. Biol.
231,56-64.
Chem.
Biochem
OLSON,
40.
L. J. (1975)
Res. Commun
D. J., SEWELL,
R. (1980) P. J., AND
M. S., AND
254, 10453-10458. J. 192, 155-163. P. J. (1978) Biochem
J. 171,751-757.
J. Biol
217,305-311. S. A.
J. Biol. AND
C. K., DEBUYSERE, J. Biol. Ch.em.
119,287-294.
P. P., DEBUYSERE,
M. S. (1980)
S. S. (1975)
CORKEY, B. E., BRANDT, M., WILLIAMS, R. J., AND WILLIAMSON, J. R. (1981) AnaL Biochem. 118,
R. A.
27,185-200. 27. PEITIT, F. H., PELLEY, B&hem Biophys.
REID,
30-41. P. J.
T. E. (1979)
Biophys.
S., AND
S., AND
J. 148, 363-374.
39.
254, 7191-7196.
Biochem.
G., JURSINIC,
ODESSEY, PARKER,
RANDLE,
RANDLE,
Biophys.
ROCHE,
57
KINASE
37. 38.
GOOD-
256,2618-2620.
HATHAWAY, G. M., LUBBEN, T. H., AND TRAUGH, J. A. (1980) J. BioL Chem. 255, 8038-8041. DEPAOLI-ROACH, A. A., AND ROACH, P. J. (1982)
Arch. 26.
W.
M.
M. S. (1979)
158,234-238. R.,
BUSE,
36. BUFFINGTON,
144,57-62.
JOLLY, AND
A.
Chem
H.
Lett
GOLDBERG,
257, 1613-1621.
BLINDER,
S., FATANIA,
(1982)
24.
L.-R.,
DEHYDROGENASE
B&hem.
18. HUGHES, W. A., AND HALESTRAP, B&hem. J. 196.459-469.
21.
35.
254, 8358-8362.
M. E., DESAUTELS,
A. L. (1982)
a-KETOACID
P. J., BRIDGES, B. J., A. L., PASK, H. T., SEVD.,
Cell. B&hem
AND
WHITEHOUSE,
9,27-53.
Regulation
of Branched-Chain RALPH
Department
of Biochemistry, Received
PAXTON Indiana
September
a-Ketoacid ROBERT
AND
University
School
7, 1983, and in revised
Dehydrogenase
Kinase’
A. HARRIS’
of Medicine, form
Indianapolis,
January
Indiana
~#223
23, 1984
Isolated rabbit liver branched-chain a-ketoacid dehydrogenase was inhibited in a mixed manner relative to ATP by a-ketoisocaproate, cY-keto+methylvalerate, cu-ketoisovalerate, a-ketocaproate, a-ketovalerate, and a-chloroisocaproate with Ido values (mM), respectively, of 0.065, 0.49, 2.5, 0.2, 0.5, and 0.08. The concentration (mM) of (Yketoisocaproate, cu-keto-fi-methylvalerate, and a-ketoisovalerate needed to activate branched-chain a-ketoacid dehydrogenase in the perfused rat heart to 50% of total activity was 0.07, 0.10, and 0.25, respectively. Isolated branched-chain cY-ketoacid dehydrogenase kinase was inhibited (Ido values, mM) by octanoate (0.5), acetoacetyl-CoA (O.Ol), methylmalonyl-CoA (0.2), NADP+ (1.5), and heparin (12 pg/ml). The kinase activity, in the presence or absence of ADP, was inhibited approximately 30% by 0.1 mM isobutyryl-CoA, isovaleryl-CoA, and malonyl-CoA, while not affected by NAD+ and NADH (1 mM), CoA, acetyl-CoA, methylcrotonyl-Cob, crotonyl-CoA, @hydroxy-/Imethyl-glutaryl-CoA, octanoyl-Cod, succinyl-CoA, and propionyl-CoA (0.1 mM). The following compounds at 2 mM also did not inhibit branched-chain a-ketoacid dehydrogenase kinase; acetate, propionate, P-hydroxybutyrate, lactate, acetoacetate, malonate, cu-ketomalonate, succinate, citrate, oxaloacetate, FAD, and NADPH. These findings help explain the unique effects of Leu compared with Val and Ile on branched-chain amino acid metabolism and the differences between control of the kinases associated with pyruvate dehydrogenase and branched-chain cu-ketoacid dehydrogenase.
Branched-chain cy-ketoacid dehydrogenase (EC 1.2.4.4), an intramitochondrial multienzyme complex, catalyzes the oxidative decarboxylation of the transamination products (i.e., KIC,3 KMV, and KIV) of the branched-chain amino acids, Leu, Ile, and Val. There is a linear correlation between blood branched-chain amino acids levels and blood branched-chain cY-ketoac-
ids levels (l), presumably due to a nearequilibrium transaminase in most tissues (2). The first nonreversible, rate-limiting enzyme in branched-chain amino acid catabolism is branched-chain cy-ketoacid dehydrogenase (3). Thus, the activity of this complex will determine the overall catabolism of the branched-chain amino acids. The activity of branched-chain cy-ketoacid dehydrogenase can be controlled (i.e., inhibited) by phosphorylation. This has been shown with the isolated complex (4-6) which has a copurifying kinase with several charateristics similar to the pyruvate dehydrogenase (EC 1.2.4.1) kinase (4). The specific intramitochondrial phosphatase that dephosphorylates and activates branched-chain a-ketoacid dehydrogenase has not been isolated or characterized, although a broad-specificity
i Supported in part by NIH Research Grants AM19259 and 5 SO7 RR5371 and the Grace M. ShoWalter Residuary Trust. ’ To whom correspondence should be addressed. a Abbreviations used: KIC, a-ketoisocaproate; KMV, a-keto-P-methylvalerate; KIV, cu-ketoisovalerate; EGTA, ethylene glycol bis(@-aminoethyl ether)N,N,N’,N’-tetracetic acid; Hepes, I-(2-hydroxyethyl)l-piperazineethanesulfonic acid; and TLCK, Nor-ptosyl-L-lysinechloromethyl ketone. 0003-9861/84 Copyright All rights
$3.00
0 1984 by Academic Press, Inc. of reproduction in any form reserved.
48
REGULATION
OF
BRANCHED-CHAIN
a-KETOACID
protein phosphatase has been identified (7) and utilized (8, 9) to show the alteration in phosphorylation state of the complex from several tissues of rats under various physiological treatments. The different branched-chain amino acids, although catabolized similarly through branched-chain Lu-ketoacid dehydrogenase, have unique effects on several metabolic processes. For example, Leu feeding decreases blood levels of Val and Ile, while feeding either of the latter have little effect on blood levels of Leu (10-12). Leu and KIC, but not Val, KIV, Ile, or KMV, are also insulinotropic agents (13-15) and regulators of protein synthesis/degradation (16, 17). Similarly, Leu or KIC have been shown to activate the complex in various tissue extracts (6, 18, 19). Phosphorylation of highly purified liver complex was inhibited to different degrees by the various branched-chain a-ketoacids (0.1 mM) with KIC > KMV > KIV (8). The partially purified complex from ox kidney, which contained a protein phosphatase and an ATPase (20), showed a similar effect of the branched-chain a-ketoacids on inhibiting the ATP-mediated inactivation of the complex (21). This study reports the relative effects of the different branched-chain a-ketoacids on inhibiting isolated branched-chain (Yketoacid dehydrogenase kinase and on activating branched-chain cu-ketoacid dehydrogenase in perfused rat heart. The influence of several other factors on isolated branched-chain a-ketoacid dehydrogenase kinase, particularly intramitochondrial metabolites associated with branchedchain cu-ketoacid catabolism, were also measured. Since this complex is very similar to pyruvate dehydrogenase, the results are also discussed in regards to similarities of these two complexes. EXPERIMENTAL
PROCEDURES
Materials. Reagents were obtained as previously given (4,8) or from Sigma Chemical Company. Bovine insulin was from Calbiochem. a-Chloroisocaproate was a gift from Dr. Robert J. Strohscheim and Dr. Ronald Simpson of Sandoz, Inc. Branched-chain a-ketoacid dehydrogenme isolation and assay. Rabbit liver branched-chain a-ketoacid
DEHYDROGENASE
KINASE
49
dehydrogenase with intrinsic kinase activity was isolated as previously described (4), with a final specific activity of 4 to 4.5 pm01 min-’ mg protein-’ at 30°C with 1 mM KIV. The spectrophotometric assay of the complex was done as previously described (4). Branched-chain c&&acid dehydrcgenme kimse assay. Branched-chain a-ketoacid dehydrogenase kinase was assayed by protein-bound %P in a final volume of 50 pl containing 30 mM Hepes-K, 1.5 mM MgCh 2 mM dithiothreitol, 0.2 mM EGTA (pH 7.34 at 2O”C), 2.72 pg branched-chain a-ketoacid dehydrogenase kinase complex, [y-32P]ATP (sp act 600-800 cpm/pmol) at the indicated concentration, and other additions as given in the text. As shown previously (4), =P incorporation was limited to the a-subunit of the aketoacid decarboxylase component of branched-chain a-ketoacid dehydrogenase. The reaction was conducted at 37°C for 20 min unless indicated. Determination of protein-bound %P was done as previously described (4). Isolatim of broad-spec$city protein phosphatase. Broad-specificity protein phosphatase was isolated as previously given (‘7), except from rabbit liver and the peak elution from a DEAE-cellulose column was at approximately 0.2 M NaCI. Phosphatase activity was measured by “Pi release (see below) in 165 ~1 total volume that combined 16.4 pg 3zP-labeled branchedchain a-ketoacid dehydrogenase (sp act approx 25,000 cpm/pg), 0.3 mg bovine serum albumin, 1.5 mM MnClz, 16 mM imidazole-Cl, 9 mM Hepes-K, 93 pM EDTA, 0.44 mM dithiothreitol, and 26 fig broad-specificity protein phosphatase (all at pH 7.5 at 20°C). A 30-~1 aliquot of the above mixture was mixed with 20 pl of 50% (w/v) trichloroacetic acid in an Eppendorf conical centrifuge tube, placed on ice for 15 min, and centrifuged in an Eppendorf centrifuge for 3 min. The supernatant, 40 ~1, was placed on filter paper and the radioactivity was determined (4). Phosphatase activity was hyperbolic with time and gave >90% release by 35 min at 37’C. Heart perfusions. Hearts from 300- to 350-g male Wistar rats were perfused by the flow-through Langendorff procedure as previously described (22). Hearts were perfused at 37°C for 30 min with KrebsHenseleit buffer containing 10 mM glucose, 20 mu/ ml insulin, and additions as given. Determination of percentage active fm branchedchain a-ketoacid dehydrogenase in Mused rat heart. Hearts were rapidly freeze-clamped at liquid Nz temperature and stored at -70°C. The frozen rat hearts were weighed and homogenized (Polytron at setting 4 and for 45 s) in approximately 8 vol of ice-cold 2% Triton X-100,0.2 mrd thiamine pyrophosphate, 10 mM EGTA, 10 mM EDTA, 5 mM dithiothreitol, 0.5 mM achloroisocaproate, 50 m&i Kpi, 0.5 pM leupeptin, 0.5 mM TLCK, 0.5 PM pepstatin A, 2.5 mM benzamidine, and 0.1 mg/ml trypsin inhibitor from egg white (Sigma Chemical Co., Type II) at a pH of 7.5 at 20°C.
PAXTON
50
AND
Homogenates were centrifuged at 29,OOOg (average) for 15. min at 0-2°C in a Sorvall SS-34 rotor. Supernatants were centrifuged at 165,OOOg (average) for 90 min at 0-2°C in a Beckman Ti50 rotor. Pellets were resuspended in 1.5 ml of ice-cold 50 mM Hepes-K, 2 mM dithiothreitol, 0.2 rnEb EGTA, and 0.1% Triton X100, pH 7.5, at 20°C. This extract was then assayed spectrophotometrically for branched-chain cu-ketoacid dehydrogenase activity with 1 mM KIV (4). Another portion of this extract was treated with broad-specificity protein phosphatase by incubating 75 ~1 of extract with 0.1 ml of phosphatase (0.26 mg protein) and 10 pl of 50 mM MnCl,. This mixture was incubated at 37°C for 30 and 45 min and then assayed for total branched-chain cy-ketoacid dehydrogenase activity. The percentage active form was determined by dividing activity before treatment by activity after treatment with broad-specificity protein phosphatase and multiplying by 100. Kinetic analyses and all statistical test and associated tables were as previously given (4).
HARRIS
FIG. 2. Lineweaver-Burk plot showing branched-chain cy-ketoacid dehydrogenase KIV. Same conditions as in Fig. 1.
inhibition kinase
of by
RESULTS
Branched-chain a-ketoacid dehydrogenase kinase was inhibited by KIC (Fig. l), KIV (Fig. Z), KMV, a-ketocaproate, a-ketovalerate, and a-chloroisocaproate relative to ATP. The data for the latter four compounds are not shown but were ob-
FIG. 1. Lineweaver-Burk branched-chain a-ketoacid tivity (nmol a? 20 min-’ various concentrations of of slopes and y intercepts versus KIC concentration. scribed under Experimental
plot showing inhibition of dehydrogenase kinase acmg protein-‘; y axis) by KIC. Insert shows replot of Lineweaver-Burk plot Assay conditions as deProcedures.
tained in identical fashion and gave the same inhibition pattern as shown for KIC and KIV. The reciprocal plots of velocity versus ATP concentration were linear with various concentrations of each inhibitor. However, the lines with various inhibitor concentrations did not appear to have a common intersection. Replots of slopes and y intercepts with inhibitor concentrations (see inserts of Fig. 1 and 2) were nonlinear with an apparent hyperbolic shape. This type of inhibition would appear to be slopehyperbolic, y intercept-hyperbolic noncompetitive inhibition. This type of inhibition is also seen with isolated pyruvate dehydrogenase kinase with respect to pyruvate and dichloroacetate (23). The complexity of the initial plots, however, limit definitive determination of the mechanism of inhibition. Thus, defining the inhibition with a more general term such as mixed is probably more appropriate. Further characterization of the mechanism of inhibition by these compounds is beyond the scope of this paper. Since the mechanism of inhibition by these compounds toward the kinase is not known, the relative inhibitory abilities of the compounds can not be definitely determined. However, the relative potency of these compounds to inhibit the kinase was
REGULATION
OF
BRANCHED-CHAIN
wKETOACID
estimated by selecting only a single ATP concentration (75 PM). While these determinations do not provide a substrate (ATP) concentration-free estimate of relative inhibitory abilities, they do allow a comparison of the relative potency of the various compounds to inhibit the kinase. It should be pointed out that these estimates of the relative inhibitory abilities of several compounds tested were consistent with the relative potency of these compounds to inhibit the kinase in the perfused rat heart (see below). I40 values (amount of compound needed to inhibit 40% of the total kinase activity; Fig. 3) were selected since the inhibition was nonlinear and it appeared that the inhibition pattern at 40% inhibition was fairly uniform among the various compounds. Ido values for several other compounds (Table I) not given above were also determined, although the type of in-
FIG. 3. Inhibition (I,, values) of branched-chain (Yketoacid dehydrogenase (BCKDH) kinase by KIV (A), KMV (B), and KIC (C). BCKDH kinase was assayed as given in Fig. 1, except assay time was 30 min and ATP concentration was 75 FM, with other additions as given. Under these conditions 100% BCKDH kinase activity (nmol 32p bound 30 min-’ mg protein-‘) was 5.13 rt 0.13 (jt f SE) for 33 determinations.
DEHYDROGENASE TABLE
KINASE
51
I
Ido VALUES FOR VARIOUS COMPOUNDS WITH ISOLATED BRANCHED-CHAIN (Y-KETOACID DEHYDROGENASE KINASE“ Compounds
Lo (rnM) 0.065 0.08 0.20 0.49 0.50 2.0 2.5 0.01 0.20 0.50 1.5
a-Ketoisocaproate a-Chloroisocaproate a-Ketocaproate a-Keto-fl-methylvalerate a-Ketovalerate a-Ketoadipate or-Ketoisovalerate Acetoacetyl-CoA Methylmalonyl-CoA Oetanoate NADP+ Heparin (1 Conditions
12 pg/ml as given
in Fig.
3.
hibition relative to ATP was not determined. KIC was about 7.5- and 38-fold more potent at inhibiting branched-chain a-ketoacid dehydrogenase kinase than KMV and KIV, respectively. a-Chloroisocaproate, a known inhibitor of isolated branchedchain Lu-ketoacid dehydrogenase kinase and activator of the complex in perfused rat heart (8), was almost as potent as KIC at inhibiting isolated branched-chain a-ketoacid dehydrogenase kinase. Sensitivity of the kinase to inhibition decreased as the carbon length (e.g. KIV) and structure (e.g. KMV) of the inhibitor varied from KIC. In a comparison of various straight-chain (Yketoacids (i.e., pyruvate, cY-ketobutyrate, LYketovalerate, a-ketocaproate, a-ketooctanoate, and cY-ketononanoate), all at 2 InM, maximum inhibition was seen with carbon length of 6, with diminishing potency occurring as the carbon length changed from 6 (data not shown). Besides the a-ketoacid, several other compounds were inhibitors of the branchedchain cY-ketoacid dehydrogenase kinase in a concentration-dependent manner (Table I). Although probably not of physiological importance, heparin was a very effective inhibitor of the kinase. Inhibition by heparin links branched-chain cr-ketoacid dehydrogenase kinase with a group of protein
52
PAXTON
AND
kinases such as casein kinase II (24) and glycogen synthase kinase P&T (25). However, numerous other protein kinases (e.g. casein kinase I, type I and type II CAMPdependent protein kinases, protease-activated kinase I, and the hemin-controlled repressor kinase) are not inhibited by heparin (24). What physiological role the inhibition of branched-chain a-ketoacid dehydrogenase kinase by cY-ketoadipate, an intermediate of lysine catabolism, and NADP+ (Id,, values of 2.0 and 1.5 mM, respectively; Table I) may have is not clear. Inhibition of branched-chain a-ketoacid dehydrogenase kinase by acetoacetyl-CoA and methylmalonyl-CoA, with Ido values of 0.01 and 0.2 mM, respectively, may have physiological significance (see Discussion) since both are distal metabolites of branched-chain amino acid oxidation and acetoacetyl-CoA is formed from ketone bodies in peripheral tissues. Octanoate inhibition of branched-chain cu-ketoacid dehydrogenase kinase (Ido value, 0.5 mM) may have little physiological significance since octanoate is not a major energy source nor tissue metabolite. However, octanoate is frequently used to determine the effects of fatty acid oxidation on branched-chain amino acid and branched-chain cy-ketoacid metabolism (see Discussion). Octanoyl-CoA had no effect on isolated branched-chain a-ketoacid dehydrogenase kinase activity (Table II). The following compounds at 2 mM had no effect (~10%) on branched-chain cy-ketoacid dehydrogenase kinase activity measured with 0.2 mM ATP as substrate: acetate, propionate, P-hydroxybutyrate, y-hydroxybutyrate, isobutyrate, lactate, acetoacetate, malonate, cu-ketomalonate, succinate, citrate, oxaloacetate, NADPH, NAD+, and NADH (data not shown). Of particular interest is the lack of direct effect of acetoacetate and /?-hydroxybutyrate since both have been suggested to have an influence on branched-chain amino acid metabolism (26; see Discussion). Isolated pyruvate dehydrogenase kinase is inhibited by substrates (CoA and NAD+) and activated by products (acetyl-CoA and NADH) of the pyruvate dehydrogenase-
HARRIS TABLE EFFEGT CHAIN
II
OF VARIOUS COMPOUNDS ON BRANCHEDWKETOACID DEHYDROGENASE KINASE ASSAYED WITHORWITHOUT ADP”
Compound
Concentration (m@
None NAD+
1.0
NADH NAD+ + NADH NAD+ + NADH CoA Acetyl-CoA Acetoacetyl-CoA Methylmalonyl-CoA Succinyl-CoA Octanoyl-CoA Isovaleryl-CoA Isobutyryl-CoA Malonyl-CoA
1.0
0.75 + 0.25 0.25 + 0.75 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
Percentage of BCKDH kinase activity -ADP
+ADP
100 96
100
99 91 95
93 112 102 100
98 102
106 102
56 64
44 70
110 100 70
60 74
109 110 89
84 71
a Conditions as given in Fig. 3, except ATP and ADP concentrations were 0.05 and 0.5 mM. ADP inhibited branched-chain wketoacid dehydrogenase (BCKDH) kinase activity by 50% under these conditions.
catalyzed reaction, but only in the presence of ADP (27). The influence of various intramitochondrial metabolites on branchedchain a-ketoacid dehydrogenase kinase activity were measured in the presence and absence of ADP (Table II). Several compounds not shown in Table II had ~10% effect on branched-chain cY-ketoacid dehydrogenase activity with or without ADP. These included NADPH (1 mM); and methylcrotonyl-CoA, valeryl-CoA, propionylCoA, oleoyl-CoA, P-hydroxy+methylglutaryl-CoA, and crotonyl-CoA (all at 0.1 mM). NAD+, NADH, combinations of both NAD+ and NADH, CoA, and acteyl-CoA were also without effect on branched-chain a-ketoacid dehydrogenase kinase activity with or without ADP. ADP did not seem to change the direction or degree of inhibition promoted by the various compounds. Of particular note in this group of metabolites (Table II) is the inhibition of the kinase by the immediate CoA-ester
REGULATION
OF
BRANCHED-CHAIN
LU-KETOACID
products of branched-chain cy-ketoacid dehydrogenase-catalyzed reaction, i.e., isovaleryl-CoA and isobutyryl-CoA, and by a more distal product, methylmalonyl-CoA. However, the final products of branchedchain amino acid oxidation, succinyl-CoA and acetyl-CoA, both intermediates of the citric acid cycle, had no effect on the isolated branched-chain cr-ketoacid dehydrogenase kinase activity. KIC and acetoacetyl-CoA appear to be better inhibitors of the kinase than methylmalonyl-CoA, malonyl-CoA, isobutyryl-CoA, or isovaleryl-CoA. The perfused rat heart was used to verify relative potency of different branchedchain a-ketoacids to inhibit branchedchain a-ketoacid dehydrogenase kinase as revealed by the percentage active form of the complex (Fig. 4). Perfused rat heart was used because conditions had already been determined that showed maximum inhibition of extracted branched-chain (Yketoacid dehydrogenase activity (22, 28). Using these conditions (perfusion with 10
FIG. 4. Activation (A, values) of branched-chain cY-ketoacid dehydrogenase (BCKDH) in perfused rat hearts by KIC (A), KMV (B), and KIV (C). Perfusions of hearts, extraction of BCKDH, and determination of BCKDH activity were as given under Experimental Procedures. Values given as it + SE with 3 hearts for each point except there were 12 hearts with 0.1 mM KIV. All points are significantly different from no additions at P < 0.05, based on a two-tailed Student’s t test, except with 0.1 mM KIV (C).
DEHYDROGENASE
KINASE
53
mM glucose + 20 mu/ml insulin), the percentage active form of the complex was very low, 3.1 f 0.6 (g f SE for three hearts). The total activity (nmol min-’ g wet wt-‘) of the complex, determined by adding exogenous phosphatase for 84 hearts done in a similar fashion (the combined number of hearts done in this study and (29)) was 248 f 40 (ii + S.D.). All concentrations of additions resulted in a statistically significant increase in the percentage active form of the complex except 0.1 mM KIV (4.3% + 1.0). The lowest concentration of KIC (15 PM) and KMV (75 PM) used in the perfused heart experiments increased the percentage active form of the complex to approximately 20%. Concentrations of KIC, KMV, and KIV that gave 50% activation (A& of total branched-chain a-ketoacid dehydrogenase activity were 0.07,0.11, and 0.25 mM, respectively. Several other compounds (i.e., clofibric acid, dichloroacetate, and phenylpyruvate) that inhibited isolated branched-chain (Yketoacid dehydrogenase kinase were also used to measure activation of the complex in perfused rat heart (29). There was a linear correlation (Fig. 5) between the Idovalues obtained with isolated branched-chain a-ketoacid dehydrogenase kinase and the AbO values obtained with perfused rat hearts for KIC, KIV, KMV, clofibric acid, dichloroacetate, and phenylpyruvate. For reasons not presently understood, the isolated branched-chain a-ketoacid dehydrogenase kinase was much less sensitive to KIV inhibition (Ido = 2.5 mM) than was the KIV activation in perfused rat heart (As0 = 0.25 mM). One possible explanation may involve the production from KIV oxidation of other inhibitors of branched-chain Lu-ketoacid dehydrogenase kinase (e.g. methylmalonyl-CoA) that potentiate KIV activation of the complex. DISCUSSION
The type of inhibition of branched-chain a-ketoacid dehydrogenase kinase (mixed relative to ATP) caused by several compounds is also seen with isolated pyruvate dehydrogenase kinase with respect to pyruvate and dichloroacetate (23). All of the
54
FIG. 5. Correlation between Aw values various compounds (DCA is dichloroacetate) as given in Fig. 4 and Ido values (y axis) given in Fig. 3. Significance of correlation with a two-tailed test of the correlation
PAXTON
AND
(z axis) for obtained obtained as determined coefficient.
branched-chain a-ketoacid dehydrogenase kinase inhibitors we have studied in detail are either substrates (KIC, KMV, KIC, (Yketocaproate, and a-ketovalerate) or inhibitors of branched-chain cy-ketoacid dehydrogenase (cu-chloroisocaproate (8), dichloroacetate (4, 8), clofibric acid and phenylpyruvate (29,30)). This relationship suggests that these compounds inhibit branched-chain cy-ketoacid dehydrogenase kinase by interaction at a site which binds a-ketoacid substrate. This site could be at either the a-subunit of the branched-chain a-ketoacid decarboxylase component (subunit that binds substrate and becomes phosphorylated) of the complex or through direct binding on the kinase. Octanoate, but not octanoyl-CoA, inhibited isolated branched-chain cY-ketoacid dehydrogenase kinase. Octanoate increases decarboxylation of branched-chain amino acids in rat skeletal muscle homogenates (31), perfused rat hindquarters (32, 33), isolated rat skeletal muscle mitochondria (33), rat diaphragms (34,35), and perfused rat hearts (34). However, octanoate inhibits branched-chain amino acid decarbox-
HARRIS
ylation in perfused rat liver (33) as well as in rat sciatic nerve, spinal cord, and aorta (35). Contrary to (34), octanoate inhibited KIC oxidation in perfused rat heart (36), although the effect was on flux through the complex and not on the extractable activity of the complex (28). Since octanoate can inhibit isolated branchedchain cr-ketoacid dehydrogenase kinase, which should lead to activation of the complex, then those tissues where the percentage active form of the complex is low (e.g., skeletal muscle (37)) may show a stimulation by octanoate. Those tissues with a high percentage active form of the complex (e.g., liver (9)) or where the activity state has been increased due to activation by a branched-chain a-ketoacid (e.g., heart; (28, 36) and this report) will probably show decreased branched-chain amino and a-ketoacid oxidation since octanoate oxidation will reduce substrate (NAD+ and CoA) availability and lead to product inhibition (38). Intracellular levels of CoA and various CoA esters are altered with branchedchain cY-ketoacid metabolism in perfused rat heart (39) and isolated rat hepatocytes (39,40). CoA and succinyl-CoA decrease in both isolated rat hepatocytes and perfused rat heart with KIC and KIV oxidation. Acetyl-CoA levels decrease with KIV oxidation in isolated rat hepatocytes but do not change with KIC oxidation in either isolated rat hepatocytes or perfused rat heart. P-Hydroxy+methylglutaryl-CoA levels also do not change with KIC oxidation in isolated rat hepatocytes. None of these compounds which either do not change or are lowered by KIC or KIV oxidation (i.e., CoA, acetyl-CoA, P-hydroxylP-methylglutaryl-CoA, and succinyl-CoA) had any effect on isolated branched-chain cz-ketoacid dehydrogenase kinase activity assayed with or without ADP. In isolated rat hepatocytes, KIV oxidation increases methylmalonyl-CoA, propionyl-CoA, and isobutyryl-CoA levels, whereas KIC oxidation increases methylcrotonyl-CoA and isovaleryl-CoA levels. Of those CoA esters that increased with branched-chain ketoacid oxidation, propionyl-CoA and methylcrotonyl-CoA, at 0.1 mM, had little effect
REGULATION
OF
BRANCHED-CHAIN
a-KETOACID
on isolated branched-chain cu-ketoacid dehydrogenase kinase activity, while the others (i.e., methylmalonyl-CoA, isobutyryl-CoA, and isovaleryl-CoA) inhibited the kinase by 30 to 40%. Thus, several CoA esters that are elevated with KIC and KIV oxidation also inhibited isolated branchedchain cu-ketoacid dehydrogenase kinase. This suggests that oxidation of branchedchain cu-ketoacid may lead to product activation of the complex through inhibition of its associated kinase (see below). The most potent CoA ester inhibitor of isolated branched-chain a-ketoacid dehydrogenase kinase was acetoacetyl-CoA (Ido value, 0.01 mM). Heart mitochondrial concentration of acetoacetyl-CoA increases lofold (2 to 25 pM) in ketotic diabetic rat (41). The estimated mitochondrial acetoacetylCoA concentration in heart, under ketotic diabetic conditions, is greater than the Id,-, value for the isolated branched-chain (Yketoacid dehydrogenase kinase. Thus, the complex may be activated under physiological conditions where ketone bodies are elevated, e.g. starvation and diabetic. Starvation leads to increased branchedchain amino acid oxidation in rats (42), humans (43), rat skeletal muscle (31), intact perfused rat hindquarters (44), rat diaphragms (45), rat kidney slices (45), and rat atria (46). Diabetes also leads to increased branched-chain amino acid oxidation in rat diaphragms (47) and increased uptake of branched-chain amino acid in human skeletal muscle (48). There is also a linear correlation between Leu oxidation in rat skeletal muscle and altered tissue acetoacetate content due to various physiological conditions (26). Acetoacetate and P-hydroxybutyrate (2 mM) had no effect on isolated branched-chain cY-ketoacid dehydrogenase kinase, suggesting that conversion to acetoacetyl-CoA is necessary for activation of the complex. Thus, inhibition of isolated branched-chain a-ketoacid dehydrogenase kinase by physiological concentrations of acetoacetyl-CoA suggests that this may be part of the mechanism whereby elevated branchedchain amino acid oxidation can occur under conditions of ketosis. The lowest concentrations of KIC and
DEHYDROGENASE
KINASE
55
KMV used in the perfused rat heart experiments were, respectively, 15 and ‘75 pM, and these concentrations produced an increased percentage active form of the complex (3.1 versus approximately 20%). The lowest KIV concentration used, 0.1 mM, did not change the percentage active form of the complex (4.3%). Human plasma concentrations of KIC, KMV, and KIV are 30, 22, and 8 pM, respectively, and increases lo-fold with Maple Syrup Urine disease (49). Rat plasma concentration of branched-chain a-ketoacids are similar in relative distribution as that seen in humans with KIC concentration ranging from 16 PM for untreated rats to 32 pM for highprotein fed, and diabetic rats (49). Plasma KIC, but not KMV or KIV, is, therefore, within the range observed to activate heart branched-chain a-ketoacid dehydrogenase through inhibition of its associated kinase. Dietary studies have shown that Leu feeding reduces plasma levels of Ile and Val, while Ile or Val feeding have little effect on plasma levels of the other two in humans (10, 11) and rats (12). Leu/KIC have also been shown to have an activating effect on branched-chain a-ketoacid dehydrogenase activity not seen with Ile/ KMV and Val/KIV in rat adipose tissue (19) and perfused rat heart (28). Similarly, KIC is much more potent than KIV in preventing the pyruvate-mediated decrease in activity of and flux through branched-chain a-ketoacid dehydrogenase in perfused rat heart (28,36). These observations are possibly explained by the relative sensitivity of the branched-chain cu-ketoacid dehydrogenase kinase (as seen with both the isolated complex and perfused rat heart) to the different branched-chain a-ketoacids with relative potency of KIC > KMV > KIV. Leu and KIC alter several other biological processes. Leu or KIC, but not Ile, Val, or KIV, have been shown to stimulate protein synthesis and inhibit protein degradation in heart (16), although the effect of KIC may only be on proteolysis (17). Similarly, Leu and KIC, but not KIV, can promote insulin release, although KIC is much more potent (13-15). Since both of these effects on protein synthesis/degradation
56
PAXTON
and insulin release appear to be related to catabolism of Leu and KIC, which is limited by branched-chain cu-ketoacid dehydrogenase, then an increase in the percentage active form of the complex, due to KIC inhibition of its associated kinase, may be part of the mechanism for these effects and may also explain why Leu/KIC, but not Ile/KMV and Val/KIV, have these properties. The differences between control of the kinases for pyruvate dehydrogenase and branched-chain a-ketoacid dehydrogenase may be important because of the different metabolic role that each complex plays in controlling their respective pathways. For example, with starvation and diabetes, where fatty acid oxidation, branched-chain amino acid oxidation, and gluconeogenesis are elevated, pyruvate dehydrogenase is inactivated (see (50)) in order to conserve pyruvate for gluconeogenesis. On the other hand, activation of branched-chain a-ketoacid dehydrogenase would lead to the production of precursors for gluconeogenesis and ketogenesis as well as maintenance of an energy supply independent of glucose. The inactivation of pyruvate dehydrogenase is due, presumably, to the increased NADH/NAD+ and acetyl-CoA/CoA ratios (27,50) from elevated fatty acid oxidation. It is important to note that these metabolites have little effect on the phosphorylation state of branched-chain a-ketoacid dehydrogenase, as shown in this paper for the highly purified rabbit liver complex and for the partially purified ox kidney complex (21). However, several other factors associated with starvation and diabetes were shown to inhibit the isolated branchedchain cY-ketoacid dehydrogenase kinase which may lead to activation of the complex. Thus, elevated levels of branchedchain amino acids from increased peripheral proteolysis would promote increased formation of branched-chain a-ketoacids (1, 49). The increased levels of branchedchain a-ketoacids, besides directly inhibiting the branched-chain a-ketoacid dehydrogenase kinase (particularly KIC), would also promote increased flux through the complex with subsequent production of other branched-chain a-ketoacid de-
AND
HARRIS
hydrogenase kinase inhibitors (i.e., isovaleryl-CoA, isobutyryl-CoA, methylmalonyl-CoA, and acetoacetyl-CoA). Ketone body utilization by peripheral tissues will also lead to elevated acetoacetyl-CoA levels (41), which may activate the complex though inhibition of its associated kinase. The signal to inactivate branched-chain QIketoacid dehydrogenase would be a decrease in branched-chain amino acids levels (and ketone bodies), possibly as a result of increased protein synthesis and decreased protein degradation in response to insulin. ACKNOWLEDGMENTS We wish to thank Richard Landry for technical assistance and Dr. Arthur Schulz for helpful diseussion. REFERENCES 1. MATTHEWS, MOTIL,
(1982) 2. KREBS,
D. E., SCHWARZ, H. P., YANG, J., YOUNG, V. R., AND BIER,
K.
Metabolism
31, 1105-1112.
H. A., ANDLUND,
P. (1977)
Adwan
R. O., D. M.
Enzyme
15.375-394.
Reg. 3. SHINNICK,
F. L., AND HARPER,
Biophys. 4. PAXTON,
Aeta
A. E.
(1976) Biochim
437, 477-486.
R., AND HARRIS,
J. Bid
R. A. (1982)
C~GWL
257, 14433-14439. R. (1982) Bio~hem J. 204, 353-356. H. R., LA& K. S., AND RANDLE, P. J.
5. ODESSEY, 6. FATANIA, (1981)
FEBS L&t. 132,285-289. A., PAXTON, R., AND PARKER, B&hem, Biophys. Res. Cbmmun
7. HARRIS, (1982)
R.
R. A.
107,
1497-1503. R. A., PAXTON, A. A. (1982) J. BioL
8. HARRIS, 9. GILLIM,
S. E., PAXTON,
R. A. (1982)
R., ANDDEPAOLI-ROACH,
Chem,
257,13915-13918.
R., COOK,
Biochem
G. A., AND HARRIS,
Biuphys.
Res. Cwrnmun
111, 74-81. 10.
SWENDSEID,
M.
E.,
VILLALOBOS,
W. S., AND DRENICK, Nutr. 17, 317-321. 11.
HAMBRAEUS, SHAW,
J.,
L., BILMAZES, N., AND YOUNG,
FIGUEROA,
Amer.
E. J. (1965)
C., DIPPLE, V. R.
J. Clin.
C., SCRIM-
(1976) J. Nutr.
106,
230-240. 12. SHINNICK,
F. L., AND HARPER,
.I Nutr.
A. E. (1977)
107.887-895. 13.
PANTEN, U., SCHGNBORN,
FEBS 14. LENZEN,
L&t.
VON KRIEGSTEIN, E., J., AND HASSELBLAIT,
20. 225-228. Biochem
S. (1978)
Pha
POSER, W., A. (1972)
?-mad
27,1321-
1324. 15. SENER, A., MALAISSE-LAGUE, W. J. (1983) J. Biol Chem.
F.,
AND
MALAISSE,
258,6693-6694.
REGULATION 16.
CHUA,
OF
B., SIEHL,
J. BioL
D. L., AND
Chem.
1’7. TISCHLER,
BRANCHED-CHAIN MORGAN,
H. E. (1979)
J. Biol
M., AND
Chem
19.
FRICK,
20.
MAN, H. M. (1981) .I Bid FATANIA, H. R., PATSTON, P. J. (1983) FEBS L&t. LAU,
G. P., TAI,
K.
FEBS
22.
SANS,
R. M.,
23.
(1980) Arch. PRATT, M. L.,
Chem.
25.
P.
P.
L., AND
(1981)
A.,
AND
AND
W.,
B&hem.
AND
PAUL,
H.
HARRIS,
28.
WAYMACK,
29.
PAXTON,
30.
DANNER,
Biochem (1982) 31. 32.
R.,
ADIBI,
200,336-345.
Metabolism
(1978)
J. W., ANDREED,
J. Biol.
M.
Chem
HARRIS,
Biophys.
CORKEY,
65, 575-582. S., AND
OLSON,
255,9773-9781. R.
A.
Arch
(1984)
Biophys. 34.
O., AND
Chem
WILLIAMSON,
257, 9668-9676.
42.
MEIKLE,
E. T., AND
ELSAS,
L. J.
A. W.,
PhysioL 43.
ADIBI,
S. A.,
(1982) 44.
HOKLAND,
B. (1981)
AND STANKO,
Biochim.
BUSE, M. G., BIGGERS, F., FRIDERICI, K. H., AND BUSE, J. F. (1972) J. Biol. Chem. 247,8085-8096.
Amer.
G. J. (1972)
R.
Metabolism
J.
T., AND
MORSE,
E.
L.
31, 578-588.
HUTSON, S. M., ZAPALOWSKI, C., CREE, HARPER, A. E. (1980) J. Biol. Chem.
T. C., AND
255,2418-
2426. 45. GOLDBERG, A. L., AND ODESSEY, J. Physiol. 223, 1384-1391. 46.
TISCHLER,
M. E., AND GOLDBERG,
J. Physiol. 47.
BUSE,
M.
48.
G.,
WAHREN,
(1972) 49.
HUTSON,
50.
HERLONG,
J., FELIG, S. M.,
Nutr.
Amer.
A. L. (1980)
Amer.
AND
AND
WEIGAND,
E., AND
LUFT,
R.
51, 1870-1878.
HARPER,
A. E. (1981)
Amer.
34, 173-183.
D. L., STANSBIE,
Mol.
F.,
88,1166-1175.
P., CERASI,
DENTON, R. M., RANDLE, COOPER, R. H., KERBEY, ERSON,
H.
Endocrinology
J. CZin. Invest.
S. (1975)
R. (1972)
238, E487-E493.
D. A. (1973)
676, 279-288.
KLAIN,
222, 1246-1250.
J. Clin Acta
A., WALAJTYS-
R. J., AND
MENAHAN, L. A., HRON, W. T., HINKELMAN, D. G., AND MIZIORKO, H. M. (1981) Eur. J. Biochem.
257,659-662.
PAUL, H. S., AND ADIBI, S. A. (1976) J. Nutr. 106, 1079-1088. SPYDEVOLD, 0. (1979) Eur. J. B&hem. 97, 389SPYDEVOLD,
B. E., MARTIN-REQUERO,
41.
394. 33.
RANDLE,
RODE, E., WILLIAMS, J. R. (1982) J. Biol.
231,56-64.
Chem.
Biochem
OLSON,
40.
L. J. (1975)
Res. Commun
D. J., SEWELL,
R. (1980) P. J., AND
M. S., AND
254, 10453-10458. J. 192, 155-163. P. J. (1978) Biochem
J. 171,751-757.
J. Biol
217,305-311. S. A.
J. Biol. AND
C. K., DEBUYSERE, J. Biol. Ch.em.
119,287-294.
P. P., DEBUYSERE,
M. S. (1980)
S. S. (1975)
CORKEY, B. E., BRANDT, M., WILLIAMS, R. J., AND WILLIAMSON, J. R. (1981) AnaL Biochem. 118,
R. A.
27,185-200. 27. PEITIT, F. H., PELLEY, B&hem Biophys.
REID,
30-41. P. J.
T. E. (1979)
Biophys.
S., AND
S., AND
J. 148, 363-374.
39.
254, 7191-7196.
Biochem.
G., JURSINIC,
ODESSEY, PARKER,
RANDLE,
RANDLE,
Biophys.
ROCHE,
57
KINASE
37. 38.
GOOD-
256,2618-2620.
HATHAWAY, G. M., LUBBEN, T. H., AND TRAUGH, J. A. (1980) J. BioL Chem. 255, 8038-8041. DEPAOLI-ROACH, A. A., AND ROACH, P. J. (1982)
Arch. 26.
W.
M.
M. S. (1979)
158,234-238. R.,
BUSE,
36. BUFFINGTON,
144,57-62.
JOLLY, AND
A.
Chem
H.
Lett
GOLDBERG,
257, 1613-1621.
BLINDER,
S., FATANIA,
(1982)
24.
L.-R.,
DEHYDROGENASE
B&hem.
18. HUGHES, W. A., AND HALESTRAP, B&hem. J. 196.459-469.
21.
35.
254, 8358-8362.
M. E., DESAUTELS,
A. L. (1982)
a-KETOACID
P. J., BRIDGES, B. J., A. L., PASK, H. T., SEVD.,
Cell. B&hem
AND
WHITEHOUSE,
9,27-53.