Experimental Hyperthyroidism Causes Inactivation of the Branched-Chain α-Ketoacid Dehydrogenase Complex in Rat Liver

Archives of Biochemistry and Biophysics Vol. 375, No. 1, March 1, pp. 55– 61, 2000 doi:10.1006/abbi.1999.1635, available online at http://www.idealibrary.com on

Experimental Hyperthyroidism Causes Inactivation of the Branched-Chain ␣-Ketoacid Dehydrogenase Complex in Rat Liver 1 Rumi Kobayashi, Yoshiharu Shimomura,* Megumi Otsuka,† Kirill M. Popov, 2 and Robert A. Harris 3 Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, Indiana 46202-5122; *Department of Bioscience, Nagoya Institute of Technology, Nagoya 466-8555, Japan; and †Department of Nutrition and Food Science, Ochanomizu University, Tokyo 112-8610, Japan

Received June 16, 1999, and in revised form November 27, 1999

Hyperthyroidism induced by 3-day treatment of rats with thyroid hormone (T 3; 3,5,3ⴕ-triiodothyronine) at 0.1 or 1 mg/kg body wt/day resulted in a reduced activity state (% of enzyme in its active, dephosphorylated state) of the hepatic branched-chain ␣-ketoacid dehydrogenase (BCKDH) complex. One treatment with 0.1 mg T 3/kg body wt caused a significant effect on the activity state of BCKDH complex after 24 h, indicating that the reduction of the activity state was triggered by the first administration of T 3. Hyperthyroidism also caused a stable increase in BCKDH kinase activity, the enzyme responsible for phosphorylation and inactivation of the BCKDH complex, suggesting that T 3 caused inactivation of the BCKDH complex by induction of its kinase. Western blot analysis also revealed increased amounts of BCKDH kinase protein in response to hyperthyroidism. No change in the plasma levels of branched-chain ␣-keto acids was observed in T 3-treated rats, arguing against an involvement of these known regulators of BCKDH kinase activity. Inactivation of the hepatic BCKDH complex as a consequence of overexpression of its kinase may save the essential branched-chain amino acids for protein synthesis during hyperthyroidism. © 2000 Academic Press 1

This work was supported by grants from the U.S. Public Health Services (NIH DK 19259), the Diabetes Research and Training Center of Indiana University School of Medicine (AM 20542), and the Grace M. Showalter Residuary Trust. 2 Current address: Division of Molecular Biology and Biochemistry, University of Missouri–Kansas City, 5100 Rockville Road, Kansas City, MO 64110. 3 To whom correspondence should be addressed at Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, 635 Barnhill Drive, Indianapolis, IN 46202-5122. Fax: (317) 274-4686. E-mail: [email protected]. 0003-9861/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

Key Words: thyroid; branched-chain amino acids; branched-chain ␣-ketoacid dehydrogenase; branchedchain ␣-ketoacid dehydrogenase kinase; rat; liver; muscle.

The branched-chain ␣-ketoacid dehydrogenase (BCKDH) 4 complex catalyzes the most important regulatory step in the catabolic pathways of the branchedchain amino acids. The relative level of expression of this complex varies among the major mammalian tissues, with a high level expressed in rat liver and a low level expressed in rat skeletal muscle (1). The BCKDH complex is in its active, dephosphorylated state when it is necessary to dispose of excess branched-chain ␣-keto acids, which are known to be toxic (1, 2). In contrast, the BCKDH complex is inactivated by phosphorylation of its E1 component by a specific kinase (BCKDH kinase) when it is important to conserve branched-chain amino acids, all of which are essential for protein synthesis. Liver and muscle BCKDH complexes are differentially regulated by a number of environmental stimuli, such as dietary protein amount, starvation, and exercise (1– 4). These factors regulate the activity state of BCKDH complex (% active as determined by phosphorylation state) by changing the BCKDH kinase activity in response to variation either in branched-chain ␣-keto acid level or in the amount of the kinase protein (5– 8). Long-term regulatory mechanisms have been shown to control the amount of protein and message for 4 Abbreviations used: BCKDH, branched-chain ␣-ketoacid dehydrogenase; T 3, Triiodothyronine.

55

56

KOBAYASHI ET AL.

BCKDH kinase (8), suggesting regulation of the expression of the gene encoding this enzyme. The occurrence of circadian rhythms in BCKDH complex activity state, BCKDH kinase activity, and the amount of BCKDH kinase protein in the liver of female rats (9) have stimulated interest in hormonal factors that may regulate expression of the kinase gene. Although evidence has been presented that glucocorticoids (10, 11) and sex hormones (9) are involved in regulation of the activity state of the BCKDH complex, the possible involvement of other hormones has been largely ignored. Thyroid hormones are known to affect not only protein degradation but also protein synthesis and therefore to accelerate protein turnover of the body. Indeed, numerous effects of thyroid hormones on enzymes involved in protein metabolism in muscle, liver, and kidney have been reported (12–14). Although hyperthyroidism has been reported to cause a modest increase in leucine oxidation in humans (15), to our knowledge there have been no studies on the effects of hyperthyroidism on the enzymatic capacities of tissues for branched-chain amino acid catabolism. In the present study the effect of thyroid hormone administration on the activity state of the BCKDH complex in rat liver and skeletal muscle was investigated. Triiodothyronine (T 3), the active form of the thyroid hormones, was injected to induce hyperthyroidism. Marked changes were found in the activity state of the liver BCKDH complex activity, activity of BCKDH kinase, and the amount of BCKDH kinase protein. EXPERIMENTAL PROCEDURES Materials. 125I-protein A was obtained from ICN Biochemicals Inc. (Irvine, CA). Broad-specificity phosphoprotein phosphatase was isolated from bovine heart as described previously (16). Lambda protein phosphatase was obtained from New England BioLabs (Beverly, MA). All other chemicals were from Sigma Chemical Co. (St. Louis, MO). Female Wistar rats (9 weeks old) were purchased from Harlan Industries (Indianapolis, IN). Assay of enzyme activities. Extraction of the BCKDH complex from tissues was performed essentially by the procedure of Shimomura et al. (17), with the exception that 2000 U/ml lambda protein phosphatase and 2 mM Mn 2⫹ were used to achieve complete dephosphorylation of the BCKDH complex prior to the assay of total BCKDH activity. The assay of BCKDH complex activity of liver was performed spectrophotometrically as described previously (18) with the exception that the pH of the assay cocktail was optimized at 7.0. One unit of BCKDH complex activity is defined as the amount of enzyme that catalyzed the formation of 1 ␮mol of NADH per minute. Reactions were initiated by the addition of 1 mM ␣-ketoisovalerate. Muscle (gastrocnemius) BCKDH complex activity was measured by the rate of ␣-keto-[1- 14C]isocaproate decarboxylation, as described previously (17). One unit of BCKDH complex activity catalyzed the decarboxylation of 1 ␮mol of ␣-ketoisocaproate per minute. For the assay of BCKDH kinase, tissue extracts were prepared in the same manner as for the assay of the BCKDH complex except that ␣-chloroisocaproate and thiamin pyrophosphate, which are known kinase inhibitors, were omitted from extraction and suspending buffers to prevent carryover into the final preparation. The broad-specificity phosphoprotein phosphatase purified from bovine heart was

used for complete activation of the BCKDH complex prior to assay of kinase activity. The lambda protein phosphatase could not be used for this purpose because Mn 2⫹, which is required for lambda protein phosphatase activity, inhibits BCKDH kinase activity. The extract was incubated with 15 mM MgSO 4 and an appropriate amount of the broad-specificity phosphoprotein phosphatase for 30 min at 37°C. The BCKDH kinase complex was precipitated again with 9% polyethylene glycol and the pellet was resuspended in a kinase assay buffer consisting of 20 mM Hepes, 1.5 mM MgCl 2, 2 mM dithiothreitol, 25 mM potassium phosphate, 50 mM KF, 2.5 ␮g/ml oligomycin, and 20% glycerol, pH 7.5. The reaction was initiated by addition of 0.5 mM ATP to the mixture prewarmed to 30°C. Samples corresponding to ⬃8 munits of the complex were taken at specified time points (30, 60, 90, and 120 s) for assay of remaining BCKDH complex activity. The activity of BCKDH kinase is expressed as the first-order rate constant for inactivation of BCKDH complex, determined from the slope of a semilogarithmic plot of BCKDH complex activity remaining versus incubation time with ATP. Western blot analysis. The BCKDH kinase complex was isolated by immunoprecipitation with goat anti-BCKDH complex IgG coupled to BrCN-activated Sepharose 4B (Sigma Chemical Co.) (8). Western blot analysis of immunoaffinity-purified BCKDH kinase complexes was used to quantify the amounts of kinase and E2 proteins as described previously (8). Determination of branched-chain ␣-keto acids in serum. Blood was withdrawn from the vena cava of pentobarbital-anesthetized rats. Serum was collected by centrifugation at 1000g for 5 min. Total branched-chain ␣-keto acids were measured spectrophotometrically by the procedure described by Goodwin et al. (19). Animal procedures. Wistar rats (female for most experiments, two per cage) were housed in a temperature- and light-controlled room (12-h light/12-h dark cycle; light from 07:00 h) and were fed Purina Rodent Laboratory Chow 7001 ad libitum throughout the experiments. After 4 days of acclimation, rats were divided into control and T 3-treated groups. Rats in the T 3-treated groups were injected subcutaneously with triiodothyronine at 0.1 or 1 mg/kg body wt in 10 mM NaOH/0.03% bovine serum albumin (pH 10.5) at 10:00 h either once or daily for 3 consecutive days. Rats in the control group were injected subcutaneously with equivalent amounts of the carrier solution. Twenty-four hours after the last injection rats were anesthetized with pentobarbital, blood was collected, and liver and gastrocnemius muscle were removed, freeze clamped, and stored at ⫺80°C until analysis. Statistics. Data are presented as means ⫾ SEM. Statistical analysis was performed by a one-factorial ANOVA and LSD test (20). Differences with a P value less than 0.05 were considered significant.

RESULTS

Effect of T 3 treatment on body weights, liver weights, and serum T 3 levels. As expected, body weights of rats were significantly decreased after the 3-day treatment with T 3: ⫺3.8 ⫾ 1.7 and ⫺13.5 ⫾ 0.3 g for the 0.1 and 1 mg T 3/kg body wt treated groups, respectively (mean values ⫾ SEM: four rats per group). The control group of four rats showed a ⫹21.3 ⫾ 3.3 g gain in body weight. Liver weights were likewise decreased in T 3treated groups, but there was no difference in the ratio of liver to body weight of each group (data not shown). Treatment for 1 day with 0.1 mg T 3/kg body wt had no effect upon body and liver weights (data not shown). Serum T 3 levels measured on samples collected 24 h after the last injection were significantly increased in the rats treated with 1.0 but not 0.1 mg T 3/kg body wt

HYPERTHYROIDISM AND RAT LIVER ␣-KETOACID DEHYDROGENASE TABLE I

Effect of T 3 Treatment on the Activity of Liver BCKDH Complex and Its Kinase in Female Rats BCKDH complex activity

Treatment Control T 3 (0.1 mg/kg) T 3 (1.0 mg/kg)

Before After activation activation (mU/g wet wt)

Complex in active state (%)

BCKDH kinase activity (min ⫺1)

567 ⫾ 46 51 ⫾ 12* 127 ⫾ 57*

64 ⫾ 5 7 ⫾ 2* 17 ⫾ 5*

0.39 ⫾ 0.05 1.26 ⫾ 0.08* 1.34 ⫾ 0.22*

892 ⫾ 48 743 ⫾ 142 728 ⫾ 122

Note. T 3 was injected subcutaneously once daily for 3 consecutive days in the amounts indicated. Rats were killed 24 h after the third injection. Values are means ⫾ SEM for four rats in each group. * Significantly different from control group (P ⬍ 0.05).

for 3 days. A tendency to be greater was apparent in the 0.1 mg T 3-treated rats, but the increase was not significantly greater than that of carrier-treated rats (data not shown). Lack of weight gain indicates that the animals treated in this manner were surely rendered hyperthyroid, but that blood levels of T 3 had returned to near normal at the time blood was drawn for analysis. Serum T 3 levels were significantly increased in the experiment in which rats were treated once with 0.1 mg T 3/kg body wt and then sacrificed 24 h later (data not shown). Effect of hyperthyroidism on the activity of the liver BCKDH complex and its kinase. Treatment of female rats for 3 days with T 3 caused a marked reduction in basal BCKDH complex activity (before activation by dephosphorylation) but had no effect on total BCKDH complex activity (activity after complete activation with phosphatase) (Table I). As a consequence the activity state of the BCKDH complex was significantly decreased after 3 days of treatment with T 3 (Table I). Both amounts of T 3 used to treat these rats caused very large changes in the activity state of the complex (Table I). Thus, hyperthyroidism produced by just 0.1 mg T 3/kg body wt was sufficient to induce a dramatic effect on the enzymatic capacity of the liver for branchedchain amino acid catabolism. The effect of treatment for 1 day with the lower dose of T 3 (0.1 mg/kg body wt) was investigated in order to determine whether this was sufficient to induce a change in the phosphorylation state of the BCKDH complex. Although the effect was clearly less pronounced than that observed after the 3-day protocol, the activity state of the BCKDH complex was significantly decreased by the low-dose 1-day treatment (from 73 to 33%) (Table II). The total activity (activity obtained after activation) of the BCKDH complex in 1-day treated rats was also decreased significantly (Table II), whereas no difference

57

in total activity of the complex occurred after 3 days of treatment (Table I). These results suggest that the first administration of T 3 caused a decrease in liver total BCKDH complex activity that was reversed by longerterm treatment with the hormone. Treatment of female rats with T 3 at 0.1 and 1.0 mg/kg body wt for 3 days caused 3.2- and 3.4-fold increases, respectively, in BCKDH kinase activity (Table I). One-day treatment with the lower amount of T 3 caused a 2.6-fold increase in BCKDH kinase activity (Table II). These findings establish an inverse relationship between the activity state of the BCKDH complex and the activity of its kinase, suggesting that the observed inactivation of the BCKDH complex in the liver of hyperthyroid rats is a consequence of induction of greater BCKDH kinase activity by T 3. Because gender differences have been found with respect to factors that regulate the activity state of the BCKDH complex (9), the key observation of the above studies was repeated with male rats (Table III). Also examined was the question of whether starvation would reverse or prevent the effects of T 3 treatment on the BCKDH complex and its kinase. Relative to the corresponding values for the female rats (Tables I and II), the total activity of the hepatic BCKDH complex was higher in the male rats, and more of the complex was in its dephosphorylated, active form in these rats, as reported previously (9). Nevertheless, the effects of T 3 treatment were the same. The percentage of the complex in its active state was reduced from 89 to 16% by treatment of male rats with T 3 at 1 mg/kg body wt for 3 days (Table III). A 2.4-fold increase in BCKDH kinase activity was likewise induced by this treatment. Starvation of both control and T 3-treated animals caused some reduction of BCKDH kinase activity without affecting the activity state of the complex (Table III).

TABLE II

Effect of 1-Day Treatment with T 3 on the Activity of Liver BCKDH Complex and Its Kinase in Female Rats BCKDH complex activity Before After activation activation (mU/g wet wt) Control T 3 (0.1 mg/kg)

588 ⫾ 110 187 ⫾ 39*

806 ⫾ 76 574 ⫾ 79*

Complex in active state (%)

BCKDH kinase activity (min ⫺1)

73 ⫾ 9 33 ⫾ 7*

0.45 ⫾ 0.05 1.16 ⫾ 0.14*

Note. T 3 was injected subcutaneously one time in the amount indicated. Rats were killed 24 h after the injection. Values are means ⫾ SEM for three rats in each group. * Significantly different from control group (P ⬍ 0.05).

58

KOBAYASHI ET AL. TABLE III

Effects of T 3 Treatment on Activity of Liver BCKDH Complex and Its Kinase in Male Rats BCKDH complex activity Before activation

After activation

Complex in active state (%)

BCKDH kinase (min ⫺1)

1298 ⫾ 61 1095 ⫾ 173 974 ⫾ 186 918 ⫾ 294

89 ⫾ 2 87 ⫾ 2 16 ⫾ 2* 23 ⫾ 3**

0.44 ⫾ 0.07 0.26 ⫾ 0.02 1.06 ⫾ 0.12* 0.73 ⫾ 0.09** ,***

(mU/g wet wt) Control Starved T 3 treated T 3 ⫹ starved

1159 ⫾ 34 950 ⫾ 164 159 ⫾ 34* 213 ⫾ 34**

Note. T 3 was injected subcutaneously once daily for 3 consecutive days at 1.0 mg/kg body wt. Rats were killed 24 h after the last injection. Values are means ⫾ SEM for four rats in each group. * Significantly different from control group (P ⬍ 0.05). ** Significantly different from starved group (P ⬍ 0.05). *** Significantly different from T 3-treated group (P ⬍ 0.05).

Effect of hyperthyroidism on liver BCKDH kinase protein. Western blot analysis was conducted to determine whether the increase of BCKDH kinase activity in T 3-treated rats involved a change in the amount of BCKDH kinase protein. Western blot analysis of the amount of E2 protein was used to establish that similar amounts of the BCKDH complex from each animal were loaded on the gel (Fig. 1). Rats treated for 3 days with 0.1 and 1 mg T 3/kg body wt showed significant increases of 1.8 ⫾ 0.4- and 1.6 ⫾ 0.3-fold, respectively, in the amounts of BCKDH kinase protein relative to that of the control animals (Fig. 1). Calculating the relative changes on the basis of the ratios of BCKDH kinase protein bands to BCKDH complex E2 protein bands gave comparable results. One-day treatment with T 3 at 0.1 mg/kg body weight as described in Table II caused a 1.6-fold increase in the amount of BCKDH kinase (data not shown). Effect of hyperthyroidism on serum levels of branched-chain ␣-keto acids. The branched-chain ␣-keto acids, particularly ␣-ketoisocaproate, are known

to inhibit BCKDH kinase activity (5, 6), and a direct correlation between activity state of the BCKDH complex and serum levels of the branched-chain ␣-keto acids has been observed in rats fed varying amounts of dietary protein (21). However, no effect of hyperthyroidism on serum concentrations of the branched-chain ␣-keto acids was observed: 44 ⫾ 4 ␮M for the control group, 41 ⫾ 4 ␮M for the 0.1 mg T 3/kg body wt treated group, and 47 ⫾ 2 ␮M for the 1 mg T 3/kg body wt treated group. Thus, it seems unlikely that changes in branched-chain ␣-keto acids levels have a role in the mechanism responsible for inactivation of the liver BCKDH complex in hyperthyroidism. Effect of hyperthyroidism on BCKDH complex activity in skeletal muscle (gastrocnemius). No effect of hyperthyroidism was observed on the activity state or total activity of the BCKDH complex in skeletal muscle (Table IV). As expected, the complex was almost completely phosphorylated and inactive in the tissue of control animals. Treatment of rats with either the low or the high amount of T 3 was without effect. TABLE IV

Effect of T 3 Treatment on BCKDH Activity in Skeletal Muscle of Female Rats BCKDH complex activity

FIG. 1. Effect of treatment of rats with T 3 for 3 days on the amount of BCKDH kinase protein in liver. Top: Western blot for BCKDH kinase protein. Bottom: Western blot for BCKDH complex subunits to establish that the loading of the BCKDH complex was the similar for each extract. Lanes 1– 4, Western blots for control rats; lanes 5– 8, Western blots for rats injected with T 3 at 1 mg/kg once daily for 3 consecutive days; lanes 9 –12, Western blots for rats injected with T 3 at 0.1 mg/kg once daily for 3 consecutive days. Similar results were obtained in a second independent experiment.

Before After activation activation (mU/g wet wt) Control T 3 (0.1 mg/kg) T 3 (1.0 mg/kg)

2.0 ⫾ 0.2 2.2 ⫾ 0.3 2.3 ⫾ 0.2

71 ⫾ 4 78 ⫾ 4 75 ⫾ 6

Complex in active state (%) 2.9 ⫾ 0.4 2.8 ⫾ 0.3 3.0 ⫾ 0.2

Note. T 3 was injected subcutaneously daily for 3 consecutive days in the amounts indicated. Rats were killed 24 h after the last injection. Values are means ⫾ SEM for four rats in each group.

HYPERTHYROIDISM AND RAT LIVER ␣-KETOACID DEHYDROGENASE

DISCUSSION

The initial working hypothesis for this study was that hyperthyroidism would have little or no effect on the activity state of the BCKDH complex in the liver but that it would completely activate the complex in the skeletal muscle. Thus, it was expected that T 3 treatment to induce hyperthyroidism would promote conversion of the BCKDH complex into its most dephosphorylated state in tissues of the animals, thereby maximizing the capacity for oxidative disposal of branched-chain amino acids. Lack of much effect on the liver was anticipated because in fed rats most of the liver complex is already in its dephosphorylated form. A large effect on skeletal muscle was anticipated because most of the complex in this tissue is present in its phosphorylated form. This scenario seemed likely because hyperthyroidism is known to increase the basal metabolic rate, stimulate the respiration rate, accelerate most if not all metabolic processes, and promote heat production by animals (reviewed in Ref. 22). It is also known that thyroid hormone can promote gluconeogenesis (23), which could be supported by increased catabolism of valine and isoleucine. Hyperthyroidism can also cause negative nitrogen balance, increase proteolysis in some tissues, promote muscle wasting, and increase blood levels of branched-chain amino acids (24). The latter would seem particularly important because the branched-chain ␣-keto acids, which might also be expected to rise in the blood and tissues of hyperthyroid animals, are known inhibitors of the BCKDH kinase, which would be expected to promote activation of the BCKDH complex. However, the activity state of the BCKDH complex responded to hyperthyroidism in exactly the opposite manner. Under conditions where the activity state of the hepatic complex of euthyroid male and female rats is well established to be primarily in its active dephosphorylated state, T 3 caused a marked reduction in the activity state of the hepatic BCKDH complex in both sexes. A significant effect was observed just 24 h after the single administration of a relatively low dose of T 3. Both male and female rats responded in a similar way to T 3 injection, ruling out possible gender effects that are known to exist for the regulation of the hepatic BCKDH complex (9). Likewise, in contrast to our prediction, no effect of T 3 treatment was found on the activity state of the BCKDH complex in skeletal muscle, which normally exists predominately in this tissue in its phosphorylated and therefore inactive form. T 3 treatment of rats changed neither the basal activity nor the total activity of the skeletal muscle complex. Thus, in the hyperthyroid state the BCKDH complex is phosphorylated and inactive in liver and skeletal muscle, the two tissues of the body that are perhaps of greatest importance for the catabolism of branched-chain amino acids in the

59

body. In retrospect, these findings may be consistent with thyroid hormone’s ability to stimulate protein synthesis in some tissues and to cause overall positive nitrogen balance (22), as discussed below. The marked decrease in activity state of liver BCKDH complex after T 3 administration suggests branched-chain amino acid catabolism should be suppressed under conditions that cause hyperthyroidism. Protein synthesis is stimulated in liver, kidney, and small intestine in rats treated with T 3 (14). However, increased levels of many free amino acids occur in the liver of hyperthyroid rats, indicating hyperthyroidism also stimulates proteolysis (14). Urinary excretion of urea decreases in hyperthyroid rats and increases in hypothyroid rats (14, 25). These findings suggest that thyroid hormone stimulates both protein synthesis and proteolysis by thyroid hormone, but that protein synthesis is stimulated more than degradation in some tissues. Thus, a decreased activity state of the BCKDH complex in hyperthyroidism may be needed to conserve branched-chain amino acids for the synthesis of specific proteins in a tissue-specific manner. This also follows from what is known about regulation of the BCKDH complex in high- and low-protein-fed rats (1). The activity state of the hepatic BCKDH complex is increased in rats fed a high-protein diet in order to rid the body of excess branched-chain amino acids whereas it is decreased in rats fed a low-protein diet in order to conserve branched-chain amino acids for the synthesis of critically important proteins. Some initial insight into the molecular basis responsible for decreased BCKDH complex activity state in the liver of hyperthyroid rats was gained in this study. The substantial increase in liver BCKDH kinase activity occurred in response to T 3 likely explains much if not all of the decrease in hepatic BCKDH activity state induced by treatment of rats with this hormone. The activity state of the BCKDH complex is determined by reversible covalent modification in which phosphorylation causes inactivation of the BCKDH or E1 component of the complex (reviewed in 1, 2). BCKDH kinase inactivates the BCKDH complex by phosphorylating specific serine residues of the ␣ subunit of E1. BCKDH phosphatase activates by dephosphorylating the phosphoserine residues of E1␣. Therefore, the activity of the BCKDH complex depends on the extent of BCKDH E1␣ phosphorylation, which in turn is determined by the relative BCKDH kinase and BCKDH phosphatase activities. Thus, an increase in the ratio of kinase activity to phosphatase activity promotes phosphorylation of the complex and causes inactivation of its enzymatic activity. The mechanism responsible for increasing BCKDH kinase activity in the hyperthyroid state then becomes of interest. A change in concentration of the branched-chain ␣-keto acids, known inhibitors of the BCKDH kinase, is one possibility. A reduction in

60

KOBAYASHI ET AL.

the level of these keto acids would cause less inhibition of BCKDH kinase activity and could therefore indirectly cause an increase in the degree to which the BCKDH complex is phosphorylated. However, serum branched-chain ␣-keto acids did not decrease in T 3treated rats, making it unlikely that a change in their mitochondrial concentration could be a significant factor in determining the activity state of the BCKDH complex. Another possibility would be an increase in the amount of BCKDH kinase protein expressed in the liver of T 3-treated rats, a mechanism demonstrated previously to be responsible for the increase in hepatic BCKDH kinase activity induced by feeding rats a lowprotein diet (8). Indeed an increase in the amount of BCKDH kinase protein that can explain a significant part of the increase in hepatic BCKDH kinase activity was found in the present study. This finding raises the possibility that BCKDH kinase gene expression may be regulated by T 3. In this regard, glucocorticoids have been shown to downregulate BCKDH kinase gene expression (11), but no evidence implicating thyroid hormone in the regulation of BCKDH kinase gene expression has been reported previously. Although the observed increase in BCKDH kinase protein can explain part of the increase in BCKDH kinase activity that occurred in response to T 3 treatment, the magnitude of the increase in protein was significantly less than the increase in enzyme activity. This raises the question of whether the pull-down immunoprecipitation technique used to measure BCKDH kinase protein underestimated the magnitude of the increase. The shortcoming of the technique is that it measures the amount of BCKDH kinase bound to the BCKDH complex rather than total enzyme protein present in the mitochondrial matrix space. However, evidence has been presented previously (8, 9) that this technique is appropriate for measuring changes in the amount of BCKDH kinase expressed in response to various conditions (8, 9). This is because the BCKDH complex binds the kinase very tightly, there appear to be many unoccupied binding sites for the kinase present on the complex, and immunoprecipitation of the complex does not release the kinase from its binding site (8, 9). Thus, we believe the technique provides a good estimate of the total amount of the kinase present in the mitochondrial matrix space. The finding that the increase in BCKDH kinase protein does not completely explain the increase in kinase activity suggests that another mechanism contributes to the increase in BCKDH kinase activity induced by T 3. Although no evidence exists for any other mechanism, activation of the kinase by some posttranslational modification of the enzyme is an obvious possibility. Hypothyroidism has been shown previously to cause an increase in activity of the BCKDH complex in adipose tissue and liver (26, 27). The demonstration in

this study that the opposite occurs in hyperthyroidism is consistent with these previous findings. Increased leucine oxidation in adipose tissue of thyroidectomized rats is corrected by administration of T 3 (26), but it was not clear from that study whether altered phosphorylation state of the BCKDH complex was involved. The increase in the BCKDH complex activity was maintained in the presence of saturating amounts of leucine and insulin (26), both of which promote activation of the complex by dephosphorylation, suggesting hypothyroidism may increase expression of the BCKDH complex in adipose tissue. Hyperthyroidism is known to decrease skeletal muscle mass (12, 13) by increasing proteolysis and decreasing protein synthesis in hyperthyroid rats (12). The activity state of muscle BCKDH complex of sedentary rats fed regular laboratory diet is low (⬃10% of the enzyme is active) (1), whereas that of the liver BCKDH complex of rats under the same condition is close to 100% (1–3). With the expectation of a relatively large increase in its activity state to dispose of branchedchain amino acids and their corresponding ␣-keto acids induced by stimulated protein degradation, skeletal muscle BCKDH complex activity was measured in the present study. However, no effect of hyperthyroidism on skeletal muscle BCKDH complex activity was observed. Although muscle protein content was not measured, it is likely that muscle mass was decreased after T 3 treatment because this has been the finding in previous studies (12, 13, 28) and the rats lost body weight. It has been reported that alanine release from muscle is stimulated as soon as 24 h after a single T 3 administration (12). Since no change in the level of branchedchain ␣-keto acids in rat serum was observed in the present study, branched-chain ␣-keto acids produced in muscle may not be transported to liver for further metabolism in hyperthyroid animals. Although muscle BCKDH kinase activity was not measured in this study, it is possible that the BCKDH complex was not activated in T 3-treated rats because of increased BCKDH kinase activity. This theoretically could cause the BCKDH complex to remain in its inactive state even when challenged by conditions (e.g., increased branched-chain ␣-keto acid levels) that might otherwise cause activation. Further investigation is clearly needed to evaluate the physiological mechanisms by which branched-chain amino acid metabolism is controlled by thyroid status in the major tissues of the body. In conclusion, the present study demonstrates that T 3 administration inactivates the hepatic BCKDH complex by increasing BCKDH kinase protein and its activity. It is proposed that saving branched-chain amino acids for protein synthesis during hyperthyroidism is the purpose of this mechanism.

HYPERTHYROIDISM AND RAT LIVER ␣-KETOACID DEHYDROGENASE

REFERENCES 1. Harper, A. E., Miller, R. H., and Block, K. P. (1984) Annu. Rev. Nutr. 4, 409 – 454. 2. Harris, R. A., Zhang, B., Goodwin, G. W., Kuntz, M. J., Shimomura, Y., Rougraff, P., Dexter, P., Zhao, Y., Gibson, R., and Crabb, D. W. (1990) Adv. Enzyme Regul. 30, 245–263. 3. Harris, R. A., Powell, S. M., Paxton, R., Gillim, S. E., and Nagae, H. (1985) Arch. Biochem. Biophys. 243, 542–555. 4. Shimomura, Y., Fujii, H., Suzuki, M., Fujitsuka, N., Naoi, M., Sugiyama, S., and Harris, R. A. (1993) Biochim. Biophys. Acta 1157, 290 –296. 5. Lau, K. S., Fatania, H. R., and Randle, P. J. (1982) FEBS Lett. 144, 57– 62. 6. Paxton, R., and Harris, R. A. (1984) Arch. Biochem. Biophys. 231, 48 –57. 7. Harris, R. A., Popov, K. M., and Zhao, Y. (1995) J. Nutr. 125, 1758S–1761S. 8. Popov, K. M., Zhao, Y., Shimomura, Y., Jaskiewicz, J., Kedishvili, N. Y., Irwin, J., Goodwin, G. W., and Harris, R. A. (1995) Arch. Biochem. Biophys. 316, 148 –154. 9. Kobayashi, R., Shimomura, Y., Murakami, T., Nakai, N., Fujitsuka, N., Otsuka, M., Arakawa, N., Popov, K. M., and Harris, R. A. (1997) Biochem. J. 327, 449 – 453. 10. Block, K. P., Richmond, W. B., Mehard, W. B., and Buse, M. G. (1987) Am. J. Physiol. 252, E396 –E407. 11. Huang, Y-S., and Chuang, D. T. (1999) Biochem. J. 339, 503–510. 12. Carter, W. J., Benjamin-van der Weijden, W. S., and Faas, F. H. (1981) Biochem. J. 194, 685– 690. 13. Brown, J. G., and Millward, D. J. (1983) Biochim. Biophys. Acta 757, 182–190.

61

14. Hayase, K., Yonekawa, G., Yokogoshi, H., and Yoshida, A. (1991) J. Nutr. 121, 970 –978. 15. Hagg, S. A., and Adibi, S. A. (1985) Metabolism 34, 813– 816. 16. Harris, R. A., Paxton, R., and Parker, R. A. (1982) Biochem. Biophys. Res. Commun. 107, 1497–1503. 17. Shimomura, Y., Suzuki, T., Saitoh, S., Tasaki, Y., Harris, R. A., and Suzuki, M. (1990) J. Appl. Physiol. 68, 161–165. 18. Paxton, R., Scislowski, P. W. D., Davis, E. J., and Harris, R. A. (1986) Biochem. J. 234, 295–303. 19. Goodwin, G. W., Kuntz, M. J., Paxton, R., and Harris, R. A. (1987) Anal. Biochem. 162, 536 –539. 20. Ott, R. L. (1993) An Introduction to Statistical Methods and Data Analysis, pp. 767– 841, Duxbury Press, Belmont, CA. 21. Soemitro, S., Block, K. P., Crowell, P. L., and Harper, A. E. (1989) J. Nutr. 119, 1203–1212. 22. Silva, J. E. (1995) Thyroid 5, 481– 492. 23. Menahan, L. A., and Wieland, O. (1969) Eur. J. Biochem. 10, 188 –194. 24. Tauveron, I., Charrier, S., Champredon, C., Bonnet, Y., Berry, C., Bayle, G., Prugnaud, J., Obled, C., Grizard, J., and Thie´blot, P. (1998) Am. J. Physiol. 269, E499 –E507. 25. Hayase, K., Yonekawa, G., and Yoshida, A. (1992) J. Nutr. 122, 1143–1148. 26. Coiro, V., Frick, G. P., Braverman, L. E., and Goodman, H. M. (1981) Am. J. Phys. 240, E669 –E676. 27. Sullivan, S. G., Dancis, J., and Cox, R. P. (1978) Biochim. Biophys. Acta 539, 135–141. 28. Angerås, U., and Hasselgren, P-O. (1987) Endocrinology 120, 1417–1421.