- Email: [email protected]
Starvation Increases the Amount of Pyruvate Dehydrogenase Kinase in Several Mammalian Tissues
Archives of Biochemistry and Biophysics Vol. 381, No. 1, September 1, pp. 1–7, 2000 doi:10.1006/abbi.2000.1946, available online at http://www.idealibrary.com on
Starvation Increases the Amount of Pyruvate Dehydrogenase Kinase in Several Mammalian Tissues Pengfei Wu, Paul V. Blair, Juichi Sato, 1 Jerzy Jaskiewicz, Kirill M. Popov, 2 and Robert A. Harris 3 Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, IN 46202-5122
Received October 29, 1999, and in revised form March 29, 2000
Covalent modification of the pyruvate dehydrogenase complex provides an important regulatory mechanism for controlling the disposal of glucose and other compounds metabolized to pyruvate. Regulation of the complex by this mechanism is achieved in part by tissuespecific expression of the genes encoding isoenzymes of pyruvate dehydrogenase kinase (PDK). Starvation is known from our previous work to increase PDK activity of heart and skeletal muscle by increasing the amount of PDK isoenzyme 4 (PDK4) present in these tissues. This study demonstrates that increased expression of both PDK4 and PDK2 occurs in rat liver, kidney, and lactating mammary gland in response to starvation. PDK4 and PDK2 message levels were also increased by starvation in the two tissues examined (liver and kidney), suggesting enhancement of gene transcription. Changes in PDK2 message and protein were of similar magnitude, but changes in PDK4 message were greater than those in PDK4 protein, suggesting regulation at the level of translation. In contrast to these tissues, starvation had little or no effect on PDK2 and PDK4 protein in brain, white adipose tissue, and brown adipose tissue. Nevertheless, PDK4 message levels were significantly increased in brain and white adipose tissue by starvation. The findings of this study indicate that increased expression of PDK isoenzymes is an important mechanism for bringing about inactivation of the pyruvate dehydrogenase complex during starvation in many but not all tissues of the body. The absence of this mechanism preserves the capacity of neuronal tissue to utilize glucose for energy during starvation. © 2000 Academic Press 1 Current address: Department of Public Health, Nagoya City University Medical School, Nagoya, Japan. 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: pyruvate dehydrogenase complex; pyruvate dehydrogenase kinase; rat; starvation; liver; kidney; mammary gland; brown adipose tissue; adipose tissue; brain.
The pyruvate dehydrogenase (PDH) 4 complex catalyzes the oxidation of pyruvate to acetyl-coenzyme A (CoA) in all mammalian cells containing mitochondria. Since the irreversibility of the reaction catalyzed by this complex prevents conversion of acetyl-CoA back to glucose, the activity of this complex is an important determinant of the overall rate of glucose disposal in animals. The activity of PDH complex is regulated by a reversible covalent modification mechanism in which phosphorylation causes inactivation of the PDH or E1 component of the complex (reviewed in Refs. 1 and 2). PDH kinase (PDK) inactivates the PDH complex by phosphorylating specific serine residues of the ␣ subunit of E1. Pyruvate dehydrogenase phosphatase (PDP) activates the complex by dephosphorylating the phosphoserine residues of E1␣. The activity of PDH complex therefore depends on the extent of E1␣ phosphorylation, which in turn is determined by the relative PDK and PDP activities. The recent discovery that PDK activity of mammalian tissues corresponds to four PDK isoenzymes (3–7) has stimulated new interest in regulation of the PDH complex. The PDK isoenzymes exhibit unique tissue expression patterns and are differentially responsive to regulation by pyruvate, acetyl-CoA, and NADH (7). The activity of the PDH complex can now be envisaged to be a function of the relative amounts of the PDK isoenzymes present in a given tissue as well as the sum 4
Abbreviations used: PDH, pyruvate dehydrogenase complex; PDK, pyruvate dehydrogenase kinase; PDP, pyruvate dehydrogenase phosphatase; PPAR␣, peroxisomal proliferator-activated receptor ␣. 1
2
WU ET AL.
of their relative sensitivities to pyruvate inhibition and acetyl-CoA and NADH activation. Marked increases in PDK activity occur in the heart (8), skeletal muscle (9), liver (10), kidney (11), and lactating mammary gland (12) during starvation of the rat. In heart and skeletal muscle this is due to a large increase in the amount of one of the PDK isoenzymes, PDK4 (13, 14). Other tissues are examined in this study to determine whether an increase in the amount of PDK isoenzymes is a general response to starvation. Since previous studies have shown that PDK1 expression is confined primarily to the heart and PDK3 expression to the testes (7), the effects of starvation on PDK2 and PDK4 expression were determined in these experiments. MATERIALS AND METHODS Materials. 125I-labeled Protein A was obtained from ICN Biochemicals. Radioactive nucleotides were from England Nuclear Research Products. The random primed DNA labeling kit for labeling cDNA probes was from Boehringer Mannheim GmbH (Germany). Arylamine acetyltransferase was prepared from pigeon liver by the procedure of Rougraff and Paxton (15). Chemicals were from Sigma Chemical Company. Animals. Male and female Wistar rats (Harlan Industries, Indianapolis, IN) were maintained for several days on Purina rodent laboratory chow to acclimate to our light and temperature-controlled facility. Rats were either continued on the chow diet or starved for 48 h prior to sacrifice. Tissues were rapidly removed and freezeclamped for the extraction of RNA. Tissues removed for the isolation of mitochondria were placed in ice-cold isolation medium without freezing and processed immediately. Isolation of mitochondria. Mitochondria were isolated from the liver and kidney of male rats (weighing about 350 g) as described previously by Blair (16). Fresh tissues were washed and minced in an ice-cold sucrose buffer (250 mM sucrose, 0.5 mM EDTA, 0.5 mM EGTA and 10 mM Tris–HCl, pH 7.4) and then homogenized by four down– up strokes with a standard Potter–Elvehjem homogenizer with a Teflon pestle in a chuck rotating at about 300 rpm. The mitochondria were isolated by differential centrifugation at 4°C with a Sorvall RC-2B centrifuge using an SS-34 rotor (700g for 5 min and 8500g for 10 min) to sediment the mitochondria. Purification was achieved by suspending the crude mitochondria in sucrose buffer and centrifuging two more times. The procedures were identical for preparation of mitochondria from the two tissues. Succinate oxidation rates stimulated by ADP (16) were greater than 150 natoms oxygen/ min/mg protein at 30°C; respiratory control ratios exceeded 5.0. The procedure described by Titheradge and Coore (17) was modified for isolation of mitochondria from rat mammary glands. Female rats (weighing about 300 g) with litters of at least six 12-day-old pups were used. The excised tissue was cut into chunks in ice-cold sucrose buffer defined above and strained through a plastic sieve. The washed tissue was minced in the ice-cold sucrose buffer and homogenized by 12 down– up strokes in a standard Potter–Elvehjem homogenizer with a Teflon pestle in a chuck rotating at about 300 rpm. The mitochondria were purified by differential centrifugation at 2000g for 3 min and 8500g for 10 min at 4°C. The fluffy brownishwhite layer of the pellet was suspended and centrifuged once more (8500g for 10 min at 4°C) to yield reasonably purified mitochondria. This procedure was also adapted for preparing mitochondria from rat white adipose tissue (epididymal fat pads), except that male rats (weighing about 350 g) were used and the centrifuging conditions were modified (700g for 5 min and 11,000g for 10 min at 4°C).
Preparation of mitochondria from rat brown adipose tissue was adapted from the procedure described by Gianotti et al. (18) and the procedure used by Hatefi and Lester (19) for the isolation of heavy beef heart mitochondria. The fresh tissue from male rats (weighing 200 g and not cold adapted) was cut into chunks and washed through a plastic sieve with the ice-cold sucrose buffer. The washed tissue was minced in ice-cold sucrose buffer and homogenized by five down– up strokes with a standard Potter–Elvehjem homogenizer with a Teflon pestle in a chuck rotating at about 300 rpm. The mitochondria were purified by differential centrifugation at 700g for 5 min and 11,000g for 10 min at 4°C. The crude mitochondrial pellet was suspended and centrifuged once more (8500g for 10 min at 4°C). Preparation of mitochondria from the rat brain followed the method described by Lai and Clark (20) with some modification. The fresh brain was washed twice and minced in the ice-cold sucrose buffer and homogenized by 12 down– up strokes in a glass homogenizer with a Teflon pestle in a chuck rotating at about 300 rpm. The mitochondria were purified by differential centrifugation at 200g for 3 min and 38,000g for 8 min at 4°C. The crude mitochondria pellets were suspended in 5 ml of Ficoll medium (120 mM mannitol, 60 mM sucrose, 50 M EDTA, 5 mM Tris, and 3% Ficoll 400, pH 7.4), layered onto 35 ml of 6% Ficoll 400 medium and centrifuged at 30,000g for 10 min at 4°C. The brown bottom layer of the pellet was washed twice with the sucrose buffer by suspension and centrifugation (30,000g for 10 min at 4°C) to further purify the mitochondria and remove Ficoll. The purified brain mitochondria had a respiratory control greater than 2.5 when oxidizing succinate and an ADP-stimulated respiratory rate greater than 50 natoms oxygen/min/mg protein at 30°C. Assay of PDK activity. PDK activity was determined as described previously (13). Activities are expressed as first-order rate constants for ATP-dependent inactivation of PDH complex activity calculated with the single exponential decay option of GraFit version 3.03, Erithacus Software. Western blot and immunoquantification of PDK2 and PDK4. Ten micrograms of mitochondria protein of different tissues was separated by SDS/PAGE as described previously (13). Western blot analysis was conducted by the protocol described previously (21). Polyclonal antisera of PDK isozymes were generated in rabbit using recombinant enzyme proteins by a standard protocol (22). The E1␣ subunit of pyruvate dehydrogenase complex was quantified with rabbit antiserum raised against purified rat pyruvate dehydrogenase complex. Northern blot analysis of message levels for PDK2 and PDK4. Total cellular RNA was extracted from freeze-clamped liver and kidney with Tri Reagent (Sigma Bioscience) following the instructions of the manufacturer. Northern blotting was conducted by the protocol previously described (4). 32P-labeled cDNAs of rat PDK2, PDK4, and 28S ribosome RNA were used as probes for the hybridization. The PDH kinase isozyme cDNAs used in this study were all cloned in this laboratory (4, 7). The oligonucleotide used as the probe to detect rat 28S rRNA was based on the sequence published by Chan et al. (23). Image quantification and statistical analysis. Quantitative results of Western and Northern blot analyses were obtained by densitometry in an Image Analysis Systems (version 2.1, Bio-Rad Laboratories). Statistical analysis of the data was carried out with SigmaPlot 3.0 (Jandel Scientific). All values are presented as means ⫾ SE.
RESULTS
Effect of starvation on expression of PDK2 and PDK4 in rat liver. As expected from the work of others (10, 24), starvation of rats caused a significant increase in PDK activity in liver mitochondria (1.48 ⫾ 0.12 min ⫺1 for starved rats versus 0.63 ⫾ 0.02 min ⫺1 for fed control rats, means ⫾ SE for four rats/group, P ⬍ 0.01). Also as
EXPRESSION OF PYRUVATE DEHYDROGENASE KINASE ISOENZYME
3
Thus, both PDK2 and PDK4 contribute to the starvation-induced increase in PDH kinase activity in liver. Effect of starvation on expression of PDK2 and PDK4 in rat kidney. Starvation caused a large increase in PDK4 protein (Fig. 2A, relative amounts by densitometry of 3.8 ⫾ 0.2 for starved rats versus 1.0 ⫾ 0.2 for fed control rats, P ⬍ 0.001) and PDK4 message (Fig. 2B, relative abundance by densitometry of 8.9 ⫾ 0.7 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.001) expressed in the kidney of rats. Starvation caused smaller but nevertheless significant increases in PDK2 protein (Fig. 2A; relative amounts by densitometry of 1.5 ⫾ 0.1 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.001) and in PDK2 message levels in the kidney (Fig. 2B; relative abundance by densitometry of 2.5 ⫾ 0.2 for starved rats versus 1.0 ⫾ 0.2 for fed control rats, P ⬍ 0.001). Thus, both PDK2 and PDK4 increase, but PDK4 probably contributes more than PDK2 to the increase in kidney PDK activity induced by starvation. Effect of starvation on expression of PDK2 and PDK4 in rat white adipose tissue. In contrast to the findings with liver and kidney, starvation had no effect on the FIG. 1. Effect of starvation on PDK2 and PDK4 protein amounts and mRNA abundance in rat liver mitochondria. (A) PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of liver mitochondrial protein from three rats in each indicated group. Antiserum raised against the PDH complex was used to measure the amount of PDH E1␣ as loading control. Similar results were obtained in separate independent experiments. (B) PDK2 mRNA (2.4 kb) and PDK4 mRNA (3.9 kb) were analyzed by Northern blotting in 20 g of total RNA from the liver of three rats in each indicated group. A 32P-labeled oligonucleotide was used to detect the corresponding 28S rRNA as loading control. Similar results were obtained in separate independent experiments.
expected from previous reports (25, 26), this treatment increased the amount of PDK2 protein in liver mitochondria (Fig. 1A). In this experiment starvation caused a doubling of PDK2 protein in the mitochondria (relative amounts by densitometry of 2.1 ⫾ 0.2 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.001). The present study further shows that the mRNA encoding this kinase was likewise nearly doubled in abundance by starvation (Fig. 1B; relative abundance by densitometry of 1.9 ⫾ 0.2 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.01), suggesting that increased transcription of the PDK2 gene may be induced by starvation. Expression of the PDK4 gene, as measured at both the protein and message levels, was likewise increased by starvation (Fig. 1A, relative amounts of PDK4 protein by densitometry: 1.9 ⫾ 0.3 for starved rats versus 1.0 ⫾ 0.2 for fed control rats, P ⬍ 0.01; Fig. 1B, relative abundance by densitometry of PDK4 mRNA: 3.0 ⫾ 0.4 for starved rats versus 1.0 ⫾ 0.2 for fed control rats, P ⬍ 0.01).
FIG. 2. Effect of starvation on PDK2 and PDK4 protein amounts and mRNA abundance in rat kidney mitochondria. (A) PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of kidney mitochondrial protein from three rats in each indicated group. Other details are the same as described in the legend to Fig. 1. (B) PDK2 mRNA (2.4 kb) and PDK4 mRNA (3.9 kb) were analyzed by Northern blotting in 20 g of total RNA from the kidney of three rats in each indicated group. Other details are the same as described in the legend to Fig. 1.
4
WU ET AL.
g (mainly with milk depletion) and induced a large increase in PDK4 protein (relative amounts by densitometry of 5.0 ⫾ 0.5 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.001) and a smaller but significant increase in PDK2 protein (relative amounts by densitometry of 1.8 ⫾ 0.1 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.001) in mitochondria isolated from lactating mammary gland (Fig. 5). Thus, increased amounts of both isoenzymes probably contribute to the starvation-induced inactivation of the PDH complex in the lactating mammary gland, although PDK4 again dominates. Effect of starvation on expression of PDK2 and PDK4 in rat brown adipose tissue. Only a modest or questionable increase in PDK4 protein in brown adipose tissue occurred in response to starvation (Fig. 6, relative amounts by densitometry of 1.4 ⫾ 0.1 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.05). These values are corrected for some increase in the amount of the E1␣ subunit of the PDH complex (Fig. 6). No significant difference in the amount of PDK2 protein between starved and control rats was observed for this tissue (Fig. 6). FIG. 3. Effect of starvation on PDK2 and PDK4 protein amounts and mRNA abundance in white adipose tissue. (A) PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of white adipose tissue mitochondrial protein from three rats in each indicated group. Other details are the same as described in the legend to Fig. 1. (B) PDK2 mRNA (2.4 kb) and PDK4 mRNA (3.9 kb) were analyzed by Northern blotting in 20 g of total RNA from the adipose tissue of three rats in each indicated group. Other details are the same as described in the legend to Fig. 1.
amounts of either PDK2 or PDK4 protein in white adipose tissue after these values were corrected for some decrease in the amount of the E1␣ subunit of the PDH complex (Fig. 3A). Nevertheless, a large increase in PDK4 message (Fig. 3B, relative abundance by densitometry of 4.1 ⫾ 0.4 for starved rats versus 1.0 ⫾ 0.2 for fed control rats, P ⬍ 0.01) and a marginal increase in PDK2 message (Fig. 3B, relative abundance by densitometry of 1.3 ⫾ 0.1 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⫽ 0.04) were observed. Effect of starvation on expression of PDK2 and PDK4 in brain. Both PDK2 and PDK4 were found expressed in the brain but starvation had no effect on their protein amounts relative to fed control rats (Fig. 4A). The abundance of PDK2 mRNA was also not altered (Fig. 4B). Nevertheless, starvation caused a significant increase in PDK4 mRNA level (Fig. 4B; relative abundance by densitometry of 2.3 ⫾ 0.3 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.05). Effect of starvation on expression of PDK2 and PDK4 in rat lactating mammary gland. Starvation of lactating female rats for 48 h reduced the combined weight of the mammary glands from 20 –25 g to 12–14
FIG. 4. Effect of starvation on PDK2 and PDK4 protein amounts and mRNA abundance in brain. (A) PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of brain mitochondrial protein from three rats in each indicated group. Other details are the same as described in the legend to Fig. 1. (B) PDK2 mRNA (2.4 kb) and PDK4 mRNA (3.9 kb) were analyzed by Northern blotting in 20 g of total RNA from the brain of three rats in each indicated group. Other details are the same as described in the legend to Fig. 1.
EXPRESSION OF PYRUVATE DEHYDROGENASE KINASE ISOENZYME
FIG. 5. Effect of starvation on PDK2 and PDK4 protein amounts in lactating mammary gland mitochondria. PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of mammary gland mitochondrial protein from three rats in each indicated group. Other details are the same as described in the legend to Fig. 1.
DISCUSSION
This study extends previous work by others (10, 24 – 27) on the mechanism by which starvation increases PDK activity in liver. The mechanism is important because increased PDK activity is likely responsible in large part for phosphorylation and inactivation of the liver PDH complex during starvation. Without inactivation of the PDH complex, compounds like lactate, pyruvate, and alanine that should be converted to glucose in the liver would be irreversibly diverted into ketone bodies via acetyl-CoA produced by the complex. Starvation is known from previous work to increase the amount of PDK2 protein present in liver mitochondria (25–27). We confirmed this finding and extended it by showing that starvation increases the relative abundance of PDK2 mRNA in the liver. Moreover, we found that starvation also increases PDK4 expression in rat liver as measured by the amount of PDK4 protein present in mitochondria and the relative abundance of its message. Thus, an increase in the amount of PDK4 as well as PDK2 contributes to the stable increase in liver PDK activity induced by starvation. This finding stands in contrast to heart and skeletal muscle in which only PDK4 is increased by starvation (13, 14). PDK4 probably contributes more than PDK2 to the stable increase in PDK activity since the specific enzyme activity of PDK4 is significantly greater than that of PDK2, at least when assayed with recombinant proteins of these isoenzymes generated in a heterologous expression system (7). Although small relative to the size of the liver, the kidney has significant capacity for gluconeogenesis and becomes increasingly important in providing glucose in long-term starvation. Consistent with the need to also conserve three carbon compounds in this tissue, starvation is known to cause inactivation of the PDH com-
5
plex in the kidney (28) and to increase PDK activity and the amount of PDK2 protein present in the kidney (11). We extended this work in the present study with the finding that starvation causes a remarkable increase in the amount of PDK4, similar in magnitude to that induced by starvation in the heart (13) and skeletal muscle (14). Starvation was also found in this study to cause a significant increase in PDK2 protein in the kidney as reported previously by Sugden et al. (11). Thus, the increase in PDK activity induced by starvation in rat kidney is due to increase expression of both isoenzymes, but increased expression of PDK4 is dominant over PDK2. The relative increases in level of liver and kidney PDK2 protein (about 2-fold) induced by starvation were about the same as the increases in PDK2 mRNA (also about 2-fold) in both liver and kidney, suggesting that regulation of PDK2 expression may occur primarily at the level of transcription. In contrast, the relative increase in PDK4 protein (1.9-fold) was less than the increase in its message (3.0-fold) in liver. This was also true for the kidney (3.8-fold for protein and 8.9-fold for message). Thus, both transcriptional and translational mechanisms may regulate PDK4 expression in these tissues. High activity of the PDH complex is required for fatty acid synthesis during milk production by the lactating mammary gland (29). Starvation is known to bring about inhibition of lipogenesis in this tissue (30). The inactivation of the PDH complex by phosphorylation is believed to be responsible in part for this inhibition (31, 32). The present work provides the first evidence that starvation increases the amounts of PDK2 and PDK4 expressed in the lactating mammary gland. We propose that increased amounts of these enzymes contribute to the inactivation of the PDH
FIG. 6. Effect of starvation on PDK2 and PDK4 protein amounts in brown adipose tissue mitochondria. PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of brown adipose tissue mitochondrial protein from three rats in each indicated group. Other details are the same as described in the legend to Fig. 1.
6
WU ET AL.
complex and therefore inhibition of lipogenesis in the mammary gland during starvation. Starvation is known to decrease the activity of the PDH complex and to inhibit lipid synthesis in brown (33) and white (34) adipose tissue. Based on our findings with other tissues, we anticipated that starvation would induce large changes in PDK2 and PDK4 proteins in both types of adipose tissue. However, starvation induced only a very slight increase in PDK4 and no change in PDK2 in brown adipose tissue, and it caused no change in the amount of either isoenzyme in white adipose tissue. Why the response of these tissues is different from that of heart, skeletal muscle, liver, kidney, and lactating mammary gland is not known. The result may have been different for brown fat tissue if cold-adapted rats had been used. An increase in the abundance of the mRNA encoding the PDKs, particularly the PDK4 message, was induced in white adipose tissue by starvation. Why a corresponding increase in PDK4 protein was not also induced is unknown, but, as suggested above for liver and kidney, control of translation of the PDK4 message is a possibility. Since the PDK4 message was dramatically increased, perhaps a longer period of starvation than used in this study would have invoked an increase in PDK4 protein in this tissue. On the other hand, the white adipocyte is unique in that it is the only cell in which insulin has been firmly established to affect short-term regulation of the PDH complex activity (35), apparently by controlling PDH phosphatase rather than PDK activity (36). Perhaps short-term regulation of the PDH phosphatase by insulin provides sufficient control of PDH complex activity in this tissue. Perhaps a long-term control mechanism would interfere with operation of a short-term, rapidly responding mechanism that either is needed or is more appropriate for control of the PDH complex activity in adipose tissue. Starvation has no effect on the phosphorylation and activity state of the PDH complex in the brain of the rat (37). This makes brain unique among all tissues of the rat that have been examined. It is consistent with glucose remaining the most important fuel for the brain energy requirement during starvation of the rat (38, 39). However, this stands in contrast to the human brain in which ketone bodies replace a significant part of the need for glucose for the brain energy requirement during starvation for several weeks (40, 41). PDK and PDH phosphatase activities are expressed in rat brain (42) and altered activities of the complex as a consequence of phosphorylation and dephosphorylation have been demonstrated in vitro in response to various incubation conditions (43) and in vivo in response to the PDK inhibitor dichloroacetate (44, 45). Stable changes in PDK activity that would impose a longer-term regulation have not been reported for rat brain, and our finding that starvation is without effect
upon PDK isoenzyme protein in this tissue is consistent with these findings. It is presumably unnecessary and perhaps undesirable to establish a long-term control mechanism that would interfere with rapid and complete activation of the brain PDH complex of the rat. Although no increase in brain PDK4 protein occurred in response to starvation, a large increase in the abundance of brain PDK4 mRNA was induced. As discussed before, this may reflect control at the level of translation. No attempt was made in this study to determine the regional and cell specificity of PDK4 expression in the brain. Nevertheless, since the message for PDK4 is increased, the potential is there for increased expression of PDK4 protein in at least some cells of the brain, which might occur with starvation of longer duration. An increase in the amount of specific PDK isoenzymes in response to starvation has now been demonstrated in five metabolically important tissues of the body, i.e., heart (13), skeletal muscle (14), and, in this study, liver, kidney, and lactating mammary gland. This change in the PDK protein amount and therefore PDK activity causes inactivation of the PDH complex by phosphorylation. Alteration of PDK gene expression may therefore provide a long-term mechanism for the regulation of the PDH complex. This is supplemented by short-term control exerted by acetyl-CoA, NADH, and pyruvate upon the specific activity of the PDH complex. Increased amounts of PDK2 and PDK4 during starvation may function as a safeguard against fluctuation in concentrations of these compounds that affect kinase activity in the short term. Furthermore, the time required for down regulation of the PDK isoenzymes imposes a lag upon the rate at which the PDH complex is reactivated in response to food intake by a starved animal (46). Maintaining the inactive state of pyruvate dehydrogenase in the early stages of these treatments assures continued availability of gluconeogenic substrates for the resynthesis of glycogen stores in tissues. We have proposed that free fatty acids are physiologically important inducing agents for PDK4 gene expression in skeletal muscle (14). Treatment of animals with WY-14,643, a synthetic agonist for the peroxisomal proliferator-activated receptor ␣ (PPAR␣), induces PDK4 in skeletal muscle (14) and several other tissues. (P. Wu and R. A. Harris, unpublished studies). The message level for PDK4 in the heart of the ground squirrel also increases dramatically during hibernation (47), a metabolic state like starvation and diabetes in which stored body fat is mobilized and serum free fatty acid levels are increased. Since some of the naturally occurring fatty acids are potent PPAR␣ agonists (48), activation of this receptor by free fatty acids may be a major part of the mechanism responsible for increased PDK4 gene expression in various tissues dur-
EXPRESSION OF PYRUVATE DEHYDROGENASE KINASE ISOENZYME
ing starvation, diabetes, and hibernation. However, it remains to be established whether free fatty acids are truly responsible for increased PDK4 gene expression. Factors involved in regulation of PDK2 expression likewise remain to be established. Nonetheless, the present study makes it apparent that starvation increases the relative amounts of PDK2 and PDK4 in several major tissues of the body as part of the mechanism designed to inactivate the PDH complex and thereby conserve compounds that can be converted to glucose. ACKNOWLEDGMENTS This work was supported by grants from U.S. Public Health Services (PHS DK47844), the Diabetes Research and Training Center of Indiana University School of Medicine, the Grace M. Showalter Residuary Trust, and the American Heart Association, Indiana Affiliate, Inc. (Postdoctoral Fellowship, P.W.).
REFERENCES 1. Yeaman, S. J. (1989) Biochem. J. 257, 625– 632. 2. Behal, R. H., Buxton, D. B., Robertson, J. G., and Olson, M. S. (1993) Annu. Rev. Nutr. 13, 497–520. 3. Popov, K. M., Kedishvili, N. Y., Zhao, Y., Shimomura, Y., Crabb, D. W., and Harris, R. A. (1993) J. Biol. Chem. 268, 26,602– 26,606. 4. Popov, K. M., Kedishvili, N. Y., Zhao, Y., Gudi, R., and Harris, R. A. (1994) J. Biol. Chem. 269, 29,720 –29,724. 5. Gudi, R., Bowker-Kinley, M. M., Kedishvili, N. Y., Zhao, Y., and Popov, K. M. (1995) J. Biol. Chem. 270, 28,989 –28,994. 6. Rowles, J., Scherer, S. W., Xi, T., Majer, M., Nickle, D. C., Rommens, J. M., Popov, K. M., Harris, R. A., Riebow, N. L., Xia, J., Tsui, L. C., Bogardus, C., and Prochazka, M. (1996) J. Biol. Chem. 271, 22,376 –22,382. 7. Bowker-Kinley, M. M., Davis, W. I., Wu, P., Harris, R. A., and Popov, K. M. (1998) Biochem. J. 329, 191–196. 8. Hutson, N. J., and Randle, P. J. (1978) FEBS Lett. 92, 73–76. 9. Fuller, S. J., and Randle, P. J. (1984) Biochem. J. 219, 635– 646. 10. Denyer, G. S., Kerbey, A. L., and Randle, P. J. (1986) Biochem. J. 239, 347–354. 11. Sugden, M. C., Holness, M. J., Donald, E., and Lall, H. (1999) Metabolism 48, 707–715. 12. Baxter, M. A., and Coore, H. G. (1978) Biochem. J. 174, 553–561. 13. Wu, P., Sato, J., Zhao, Y., Jaskiewicz, J., Popov, K. M., and Harris, R. A. (1998) Biochem. J. 329, 197–201. 14. Wu, P., Inskeep, K., Bowker-Kinley, M. M., Popov, K. M., and Harris, R. A. (1999) Diabetes 48, 1593–1599. 15. Rougraff, P. M., and Paxton, R. (1987) Comp. Biochem. Physiol. 86B, 601– 606. 16. Blair, P. V. (1977) Arch. Biochem. Biophys. 181, 550 –568. 17. Titheradge, M. A., and Coore, H. G. (1997) Int. J. Biochem. 8, 433– 436. 18. Gianotti, M., Clapes, J., Llado, I., and Palou, A. (1998) Life Sci. 62, 1889 –1899. 19. Hatefi, Y., and Lester, R. L. (1958) Biochim. Biophys. Acta 27, 83– 88. 20. Lai, J. C., and Clark, J. B. (1979) Methods Enzymol. 55, 51– 60.
7
21. Zhao, Y., Popov, K. M., Shimomura, Y., Kedishvili, N. Y., Jaskiewicz, J., Kuntz, M. J., Zhang, B., and Harris, R. A. (1994) Arch. Biochem. Biophys. 308, 446 – 453. 22. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2 ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 23. Chan, Y. L., Olvera, J., and Wool, I. G. (1983) Nucleic Acids Res. 11, 7819 –7831. 24. Jones, B. S., Yeaman, S. J., Sugden, M. C., and Holness, M. J. (1992) Biochim. Biophys. Acta 1134, 164 –168. 25. Priestman, D. A., Donald, E., Holness, M. J., and Sugden, M. C. (1997) FEBS Lett. 419, 55–57. 26. Sugden, M. C., Fryer, L. G., Orfali, K. A., Priestman, D. A., Donald, E., and Holness, M. J. (1998) Biochem. J. 329, 89 –94. 27. Priestman, D. A., Mistry, S. C., Halsall, A., and Randle, P. J. (1994) Biochem. J. 300, 659 – 664. 28. Wieland, O., Siess, E., Schulze-Wethmar, F. H., Funcke, H. G., and Winton, B. (1971) Arch. Biochem. Biophys. 143, 593– 601. 29. Munday, M. R., and Williamson, D. H. (1981) Biochem. J. 196, 831– 837. 30. Mercer, S. W., and Williamson, D. H. (1987) Biochem. J. 242, 235–243. 31. Hagopian, K., Butt, J., and Munday, M. R. (1991) Comp. Biochem. Physiol. 100, 527–534. 32. Hagopian, K., and Munday, M. R. (1997) Biochim. Biophys. Acta 1336, 474 – 484. 33. Holness, M. J., and Sugden, M. C. (1990) Biochem. Soc. Trans. 18, 591–592. 34. Stansbie, D., Denton, R. M., Bridges, B. J., Pask, H. T., and Randle, P. J. (1976) Biochem. J. 154, 225–236. 35. Denton, R. M., McCormack, J. G., Rutter, G. A., Burnett, P., Edgell, N. J., Moule, S. K., and Diggle, T. A. (1996) Adv. Enzyme Regul. 36, 183–198. 36. Rutter, G. A., Diggle, T. A., and Denton, R. M. (1992) Biochem. J. 285, 435– 439. 37. Siess, E. A., Wittmann, J., and Wieland, O. H. (1971) HoppeSeyler’s Z. Physiol. Chem. 352, 447– 452. 38. Hawkins, R. A., Mans, A. M., and Davis, D. W. (1986) Am. J. Physiol. 250, E169 –E178. 39. Cherel, Y., Burnol, A. F., Leturque, A., and Le Maho, Y. (1988) Metabolism 37, 1033–1039. 40. Owen, O. E., Morgan, A. P., Kemp, H. G., Sullivan, J. M., Herrera, M. G., and Cahill, G. F., Jr. (1967) J. Clin. Invest. 46, 1589 –1595. 41. Hasselbalch, S. G., Knudsen, G. M., Jakobsen, J., Hageman, L. P., Holm, S., Paulson, and OB (1994) J. Cereb. Blood Flow Metab. 14, 125–131. 42. Sheu, K. F., Lai, J. C., and Blass, J. P. (1984) J. Neurochem. 42, 230 –236. 43. Browning, M., Baudry, M., Bennett, W. F., and Lynch, G. (1981) J. Neurochem. 36, 1932–1940. 44. Morgan, D. G., and Routtenberg, A. (1982) Brain Res. 251, 391– 394. 45. Abemayor, E., Kovachich, G. B., and Haugaard, N. (1984) J. Neurochem. 42, 38 – 42. 46. Holness, M. J., French, T. J., and Sugden, M. C. (1986) Biochem. J. 235, 441– 445. 47. Andrews, M. T., Squire, T. L., Bowen, C. M., and Rollins, M. B. (1998) Proc. Natl. Acad. Sci. USA 95, 8392– 8397. 48. Forman, B. M., Chen, J., and Evans, R. M. (1997) Proc. Natl. Acad. Sci. USA 94, 4312– 4317.
Starvation Increases the Amount of Pyruvate Dehydrogenase Kinase in Several Mammalian Tissues Pengfei Wu, Paul V. Blair, Juichi Sato, 1 Jerzy Jaskiewicz, Kirill M. Popov, 2 and Robert A. Harris 3 Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, IN 46202-5122
Received October 29, 1999, and in revised form March 29, 2000
Covalent modification of the pyruvate dehydrogenase complex provides an important regulatory mechanism for controlling the disposal of glucose and other compounds metabolized to pyruvate. Regulation of the complex by this mechanism is achieved in part by tissuespecific expression of the genes encoding isoenzymes of pyruvate dehydrogenase kinase (PDK). Starvation is known from our previous work to increase PDK activity of heart and skeletal muscle by increasing the amount of PDK isoenzyme 4 (PDK4) present in these tissues. This study demonstrates that increased expression of both PDK4 and PDK2 occurs in rat liver, kidney, and lactating mammary gland in response to starvation. PDK4 and PDK2 message levels were also increased by starvation in the two tissues examined (liver and kidney), suggesting enhancement of gene transcription. Changes in PDK2 message and protein were of similar magnitude, but changes in PDK4 message were greater than those in PDK4 protein, suggesting regulation at the level of translation. In contrast to these tissues, starvation had little or no effect on PDK2 and PDK4 protein in brain, white adipose tissue, and brown adipose tissue. Nevertheless, PDK4 message levels were significantly increased in brain and white adipose tissue by starvation. The findings of this study indicate that increased expression of PDK isoenzymes is an important mechanism for bringing about inactivation of the pyruvate dehydrogenase complex during starvation in many but not all tissues of the body. The absence of this mechanism preserves the capacity of neuronal tissue to utilize glucose for energy during starvation. © 2000 Academic Press 1 Current address: Department of Public Health, Nagoya City University Medical School, Nagoya, Japan. 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: pyruvate dehydrogenase complex; pyruvate dehydrogenase kinase; rat; starvation; liver; kidney; mammary gland; brown adipose tissue; adipose tissue; brain.
The pyruvate dehydrogenase (PDH) 4 complex catalyzes the oxidation of pyruvate to acetyl-coenzyme A (CoA) in all mammalian cells containing mitochondria. Since the irreversibility of the reaction catalyzed by this complex prevents conversion of acetyl-CoA back to glucose, the activity of this complex is an important determinant of the overall rate of glucose disposal in animals. The activity of PDH complex is regulated by a reversible covalent modification mechanism in which phosphorylation causes inactivation of the PDH or E1 component of the complex (reviewed in Refs. 1 and 2). PDH kinase (PDK) inactivates the PDH complex by phosphorylating specific serine residues of the ␣ subunit of E1. Pyruvate dehydrogenase phosphatase (PDP) activates the complex by dephosphorylating the phosphoserine residues of E1␣. The activity of PDH complex therefore depends on the extent of E1␣ phosphorylation, which in turn is determined by the relative PDK and PDP activities. The recent discovery that PDK activity of mammalian tissues corresponds to four PDK isoenzymes (3–7) has stimulated new interest in regulation of the PDH complex. The PDK isoenzymes exhibit unique tissue expression patterns and are differentially responsive to regulation by pyruvate, acetyl-CoA, and NADH (7). The activity of the PDH complex can now be envisaged to be a function of the relative amounts of the PDK isoenzymes present in a given tissue as well as the sum 4
Abbreviations used: PDH, pyruvate dehydrogenase complex; PDK, pyruvate dehydrogenase kinase; PDP, pyruvate dehydrogenase phosphatase; PPAR␣, peroxisomal proliferator-activated receptor ␣. 1
2
WU ET AL.
of their relative sensitivities to pyruvate inhibition and acetyl-CoA and NADH activation. Marked increases in PDK activity occur in the heart (8), skeletal muscle (9), liver (10), kidney (11), and lactating mammary gland (12) during starvation of the rat. In heart and skeletal muscle this is due to a large increase in the amount of one of the PDK isoenzymes, PDK4 (13, 14). Other tissues are examined in this study to determine whether an increase in the amount of PDK isoenzymes is a general response to starvation. Since previous studies have shown that PDK1 expression is confined primarily to the heart and PDK3 expression to the testes (7), the effects of starvation on PDK2 and PDK4 expression were determined in these experiments. MATERIALS AND METHODS Materials. 125I-labeled Protein A was obtained from ICN Biochemicals. Radioactive nucleotides were from England Nuclear Research Products. The random primed DNA labeling kit for labeling cDNA probes was from Boehringer Mannheim GmbH (Germany). Arylamine acetyltransferase was prepared from pigeon liver by the procedure of Rougraff and Paxton (15). Chemicals were from Sigma Chemical Company. Animals. Male and female Wistar rats (Harlan Industries, Indianapolis, IN) were maintained for several days on Purina rodent laboratory chow to acclimate to our light and temperature-controlled facility. Rats were either continued on the chow diet or starved for 48 h prior to sacrifice. Tissues were rapidly removed and freezeclamped for the extraction of RNA. Tissues removed for the isolation of mitochondria were placed in ice-cold isolation medium without freezing and processed immediately. Isolation of mitochondria. Mitochondria were isolated from the liver and kidney of male rats (weighing about 350 g) as described previously by Blair (16). Fresh tissues were washed and minced in an ice-cold sucrose buffer (250 mM sucrose, 0.5 mM EDTA, 0.5 mM EGTA and 10 mM Tris–HCl, pH 7.4) and then homogenized by four down– up strokes with a standard Potter–Elvehjem homogenizer with a Teflon pestle in a chuck rotating at about 300 rpm. The mitochondria were isolated by differential centrifugation at 4°C with a Sorvall RC-2B centrifuge using an SS-34 rotor (700g for 5 min and 8500g for 10 min) to sediment the mitochondria. Purification was achieved by suspending the crude mitochondria in sucrose buffer and centrifuging two more times. The procedures were identical for preparation of mitochondria from the two tissues. Succinate oxidation rates stimulated by ADP (16) were greater than 150 natoms oxygen/ min/mg protein at 30°C; respiratory control ratios exceeded 5.0. The procedure described by Titheradge and Coore (17) was modified for isolation of mitochondria from rat mammary glands. Female rats (weighing about 300 g) with litters of at least six 12-day-old pups were used. The excised tissue was cut into chunks in ice-cold sucrose buffer defined above and strained through a plastic sieve. The washed tissue was minced in the ice-cold sucrose buffer and homogenized by 12 down– up strokes in a standard Potter–Elvehjem homogenizer with a Teflon pestle in a chuck rotating at about 300 rpm. The mitochondria were purified by differential centrifugation at 2000g for 3 min and 8500g for 10 min at 4°C. The fluffy brownishwhite layer of the pellet was suspended and centrifuged once more (8500g for 10 min at 4°C) to yield reasonably purified mitochondria. This procedure was also adapted for preparing mitochondria from rat white adipose tissue (epididymal fat pads), except that male rats (weighing about 350 g) were used and the centrifuging conditions were modified (700g for 5 min and 11,000g for 10 min at 4°C).
Preparation of mitochondria from rat brown adipose tissue was adapted from the procedure described by Gianotti et al. (18) and the procedure used by Hatefi and Lester (19) for the isolation of heavy beef heart mitochondria. The fresh tissue from male rats (weighing 200 g and not cold adapted) was cut into chunks and washed through a plastic sieve with the ice-cold sucrose buffer. The washed tissue was minced in ice-cold sucrose buffer and homogenized by five down– up strokes with a standard Potter–Elvehjem homogenizer with a Teflon pestle in a chuck rotating at about 300 rpm. The mitochondria were purified by differential centrifugation at 700g for 5 min and 11,000g for 10 min at 4°C. The crude mitochondrial pellet was suspended and centrifuged once more (8500g for 10 min at 4°C). Preparation of mitochondria from the rat brain followed the method described by Lai and Clark (20) with some modification. The fresh brain was washed twice and minced in the ice-cold sucrose buffer and homogenized by 12 down– up strokes in a glass homogenizer with a Teflon pestle in a chuck rotating at about 300 rpm. The mitochondria were purified by differential centrifugation at 200g for 3 min and 38,000g for 8 min at 4°C. The crude mitochondria pellets were suspended in 5 ml of Ficoll medium (120 mM mannitol, 60 mM sucrose, 50 M EDTA, 5 mM Tris, and 3% Ficoll 400, pH 7.4), layered onto 35 ml of 6% Ficoll 400 medium and centrifuged at 30,000g for 10 min at 4°C. The brown bottom layer of the pellet was washed twice with the sucrose buffer by suspension and centrifugation (30,000g for 10 min at 4°C) to further purify the mitochondria and remove Ficoll. The purified brain mitochondria had a respiratory control greater than 2.5 when oxidizing succinate and an ADP-stimulated respiratory rate greater than 50 natoms oxygen/min/mg protein at 30°C. Assay of PDK activity. PDK activity was determined as described previously (13). Activities are expressed as first-order rate constants for ATP-dependent inactivation of PDH complex activity calculated with the single exponential decay option of GraFit version 3.03, Erithacus Software. Western blot and immunoquantification of PDK2 and PDK4. Ten micrograms of mitochondria protein of different tissues was separated by SDS/PAGE as described previously (13). Western blot analysis was conducted by the protocol described previously (21). Polyclonal antisera of PDK isozymes were generated in rabbit using recombinant enzyme proteins by a standard protocol (22). The E1␣ subunit of pyruvate dehydrogenase complex was quantified with rabbit antiserum raised against purified rat pyruvate dehydrogenase complex. Northern blot analysis of message levels for PDK2 and PDK4. Total cellular RNA was extracted from freeze-clamped liver and kidney with Tri Reagent (Sigma Bioscience) following the instructions of the manufacturer. Northern blotting was conducted by the protocol previously described (4). 32P-labeled cDNAs of rat PDK2, PDK4, and 28S ribosome RNA were used as probes for the hybridization. The PDH kinase isozyme cDNAs used in this study were all cloned in this laboratory (4, 7). The oligonucleotide used as the probe to detect rat 28S rRNA was based on the sequence published by Chan et al. (23). Image quantification and statistical analysis. Quantitative results of Western and Northern blot analyses were obtained by densitometry in an Image Analysis Systems (version 2.1, Bio-Rad Laboratories). Statistical analysis of the data was carried out with SigmaPlot 3.0 (Jandel Scientific). All values are presented as means ⫾ SE.
RESULTS
Effect of starvation on expression of PDK2 and PDK4 in rat liver. As expected from the work of others (10, 24), starvation of rats caused a significant increase in PDK activity in liver mitochondria (1.48 ⫾ 0.12 min ⫺1 for starved rats versus 0.63 ⫾ 0.02 min ⫺1 for fed control rats, means ⫾ SE for four rats/group, P ⬍ 0.01). Also as
EXPRESSION OF PYRUVATE DEHYDROGENASE KINASE ISOENZYME
3
Thus, both PDK2 and PDK4 contribute to the starvation-induced increase in PDH kinase activity in liver. Effect of starvation on expression of PDK2 and PDK4 in rat kidney. Starvation caused a large increase in PDK4 protein (Fig. 2A, relative amounts by densitometry of 3.8 ⫾ 0.2 for starved rats versus 1.0 ⫾ 0.2 for fed control rats, P ⬍ 0.001) and PDK4 message (Fig. 2B, relative abundance by densitometry of 8.9 ⫾ 0.7 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.001) expressed in the kidney of rats. Starvation caused smaller but nevertheless significant increases in PDK2 protein (Fig. 2A; relative amounts by densitometry of 1.5 ⫾ 0.1 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.001) and in PDK2 message levels in the kidney (Fig. 2B; relative abundance by densitometry of 2.5 ⫾ 0.2 for starved rats versus 1.0 ⫾ 0.2 for fed control rats, P ⬍ 0.001). Thus, both PDK2 and PDK4 increase, but PDK4 probably contributes more than PDK2 to the increase in kidney PDK activity induced by starvation. Effect of starvation on expression of PDK2 and PDK4 in rat white adipose tissue. In contrast to the findings with liver and kidney, starvation had no effect on the FIG. 1. Effect of starvation on PDK2 and PDK4 protein amounts and mRNA abundance in rat liver mitochondria. (A) PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of liver mitochondrial protein from three rats in each indicated group. Antiserum raised against the PDH complex was used to measure the amount of PDH E1␣ as loading control. Similar results were obtained in separate independent experiments. (B) PDK2 mRNA (2.4 kb) and PDK4 mRNA (3.9 kb) were analyzed by Northern blotting in 20 g of total RNA from the liver of three rats in each indicated group. A 32P-labeled oligonucleotide was used to detect the corresponding 28S rRNA as loading control. Similar results were obtained in separate independent experiments.
expected from previous reports (25, 26), this treatment increased the amount of PDK2 protein in liver mitochondria (Fig. 1A). In this experiment starvation caused a doubling of PDK2 protein in the mitochondria (relative amounts by densitometry of 2.1 ⫾ 0.2 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.001). The present study further shows that the mRNA encoding this kinase was likewise nearly doubled in abundance by starvation (Fig. 1B; relative abundance by densitometry of 1.9 ⫾ 0.2 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.01), suggesting that increased transcription of the PDK2 gene may be induced by starvation. Expression of the PDK4 gene, as measured at both the protein and message levels, was likewise increased by starvation (Fig. 1A, relative amounts of PDK4 protein by densitometry: 1.9 ⫾ 0.3 for starved rats versus 1.0 ⫾ 0.2 for fed control rats, P ⬍ 0.01; Fig. 1B, relative abundance by densitometry of PDK4 mRNA: 3.0 ⫾ 0.4 for starved rats versus 1.0 ⫾ 0.2 for fed control rats, P ⬍ 0.01).
FIG. 2. Effect of starvation on PDK2 and PDK4 protein amounts and mRNA abundance in rat kidney mitochondria. (A) PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of kidney mitochondrial protein from three rats in each indicated group. Other details are the same as described in the legend to Fig. 1. (B) PDK2 mRNA (2.4 kb) and PDK4 mRNA (3.9 kb) were analyzed by Northern blotting in 20 g of total RNA from the kidney of three rats in each indicated group. Other details are the same as described in the legend to Fig. 1.
4
WU ET AL.
g (mainly with milk depletion) and induced a large increase in PDK4 protein (relative amounts by densitometry of 5.0 ⫾ 0.5 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.001) and a smaller but significant increase in PDK2 protein (relative amounts by densitometry of 1.8 ⫾ 0.1 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.001) in mitochondria isolated from lactating mammary gland (Fig. 5). Thus, increased amounts of both isoenzymes probably contribute to the starvation-induced inactivation of the PDH complex in the lactating mammary gland, although PDK4 again dominates. Effect of starvation on expression of PDK2 and PDK4 in rat brown adipose tissue. Only a modest or questionable increase in PDK4 protein in brown adipose tissue occurred in response to starvation (Fig. 6, relative amounts by densitometry of 1.4 ⫾ 0.1 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.05). These values are corrected for some increase in the amount of the E1␣ subunit of the PDH complex (Fig. 6). No significant difference in the amount of PDK2 protein between starved and control rats was observed for this tissue (Fig. 6). FIG. 3. Effect of starvation on PDK2 and PDK4 protein amounts and mRNA abundance in white adipose tissue. (A) PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of white adipose tissue mitochondrial protein from three rats in each indicated group. Other details are the same as described in the legend to Fig. 1. (B) PDK2 mRNA (2.4 kb) and PDK4 mRNA (3.9 kb) were analyzed by Northern blotting in 20 g of total RNA from the adipose tissue of three rats in each indicated group. Other details are the same as described in the legend to Fig. 1.
amounts of either PDK2 or PDK4 protein in white adipose tissue after these values were corrected for some decrease in the amount of the E1␣ subunit of the PDH complex (Fig. 3A). Nevertheless, a large increase in PDK4 message (Fig. 3B, relative abundance by densitometry of 4.1 ⫾ 0.4 for starved rats versus 1.0 ⫾ 0.2 for fed control rats, P ⬍ 0.01) and a marginal increase in PDK2 message (Fig. 3B, relative abundance by densitometry of 1.3 ⫾ 0.1 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⫽ 0.04) were observed. Effect of starvation on expression of PDK2 and PDK4 in brain. Both PDK2 and PDK4 were found expressed in the brain but starvation had no effect on their protein amounts relative to fed control rats (Fig. 4A). The abundance of PDK2 mRNA was also not altered (Fig. 4B). Nevertheless, starvation caused a significant increase in PDK4 mRNA level (Fig. 4B; relative abundance by densitometry of 2.3 ⫾ 0.3 for starved rats versus 1.0 ⫾ 0.1 for fed control rats, P ⬍ 0.05). Effect of starvation on expression of PDK2 and PDK4 in rat lactating mammary gland. Starvation of lactating female rats for 48 h reduced the combined weight of the mammary glands from 20 –25 g to 12–14
FIG. 4. Effect of starvation on PDK2 and PDK4 protein amounts and mRNA abundance in brain. (A) PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of brain mitochondrial protein from three rats in each indicated group. Other details are the same as described in the legend to Fig. 1. (B) PDK2 mRNA (2.4 kb) and PDK4 mRNA (3.9 kb) were analyzed by Northern blotting in 20 g of total RNA from the brain of three rats in each indicated group. Other details are the same as described in the legend to Fig. 1.
EXPRESSION OF PYRUVATE DEHYDROGENASE KINASE ISOENZYME
FIG. 5. Effect of starvation on PDK2 and PDK4 protein amounts in lactating mammary gland mitochondria. PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of mammary gland mitochondrial protein from three rats in each indicated group. Other details are the same as described in the legend to Fig. 1.
DISCUSSION
This study extends previous work by others (10, 24 – 27) on the mechanism by which starvation increases PDK activity in liver. The mechanism is important because increased PDK activity is likely responsible in large part for phosphorylation and inactivation of the liver PDH complex during starvation. Without inactivation of the PDH complex, compounds like lactate, pyruvate, and alanine that should be converted to glucose in the liver would be irreversibly diverted into ketone bodies via acetyl-CoA produced by the complex. Starvation is known from previous work to increase the amount of PDK2 protein present in liver mitochondria (25–27). We confirmed this finding and extended it by showing that starvation increases the relative abundance of PDK2 mRNA in the liver. Moreover, we found that starvation also increases PDK4 expression in rat liver as measured by the amount of PDK4 protein present in mitochondria and the relative abundance of its message. Thus, an increase in the amount of PDK4 as well as PDK2 contributes to the stable increase in liver PDK activity induced by starvation. This finding stands in contrast to heart and skeletal muscle in which only PDK4 is increased by starvation (13, 14). PDK4 probably contributes more than PDK2 to the stable increase in PDK activity since the specific enzyme activity of PDK4 is significantly greater than that of PDK2, at least when assayed with recombinant proteins of these isoenzymes generated in a heterologous expression system (7). Although small relative to the size of the liver, the kidney has significant capacity for gluconeogenesis and becomes increasingly important in providing glucose in long-term starvation. Consistent with the need to also conserve three carbon compounds in this tissue, starvation is known to cause inactivation of the PDH com-
5
plex in the kidney (28) and to increase PDK activity and the amount of PDK2 protein present in the kidney (11). We extended this work in the present study with the finding that starvation causes a remarkable increase in the amount of PDK4, similar in magnitude to that induced by starvation in the heart (13) and skeletal muscle (14). Starvation was also found in this study to cause a significant increase in PDK2 protein in the kidney as reported previously by Sugden et al. (11). Thus, the increase in PDK activity induced by starvation in rat kidney is due to increase expression of both isoenzymes, but increased expression of PDK4 is dominant over PDK2. The relative increases in level of liver and kidney PDK2 protein (about 2-fold) induced by starvation were about the same as the increases in PDK2 mRNA (also about 2-fold) in both liver and kidney, suggesting that regulation of PDK2 expression may occur primarily at the level of transcription. In contrast, the relative increase in PDK4 protein (1.9-fold) was less than the increase in its message (3.0-fold) in liver. This was also true for the kidney (3.8-fold for protein and 8.9-fold for message). Thus, both transcriptional and translational mechanisms may regulate PDK4 expression in these tissues. High activity of the PDH complex is required for fatty acid synthesis during milk production by the lactating mammary gland (29). Starvation is known to bring about inhibition of lipogenesis in this tissue (30). The inactivation of the PDH complex by phosphorylation is believed to be responsible in part for this inhibition (31, 32). The present work provides the first evidence that starvation increases the amounts of PDK2 and PDK4 expressed in the lactating mammary gland. We propose that increased amounts of these enzymes contribute to the inactivation of the PDH
FIG. 6. Effect of starvation on PDK2 and PDK4 protein amounts in brown adipose tissue mitochondria. PDK2 and PDK4 proteins were analyzed by Western blotting in 10 g of brown adipose tissue mitochondrial protein from three rats in each indicated group. Other details are the same as described in the legend to Fig. 1.
6
WU ET AL.
complex and therefore inhibition of lipogenesis in the mammary gland during starvation. Starvation is known to decrease the activity of the PDH complex and to inhibit lipid synthesis in brown (33) and white (34) adipose tissue. Based on our findings with other tissues, we anticipated that starvation would induce large changes in PDK2 and PDK4 proteins in both types of adipose tissue. However, starvation induced only a very slight increase in PDK4 and no change in PDK2 in brown adipose tissue, and it caused no change in the amount of either isoenzyme in white adipose tissue. Why the response of these tissues is different from that of heart, skeletal muscle, liver, kidney, and lactating mammary gland is not known. The result may have been different for brown fat tissue if cold-adapted rats had been used. An increase in the abundance of the mRNA encoding the PDKs, particularly the PDK4 message, was induced in white adipose tissue by starvation. Why a corresponding increase in PDK4 protein was not also induced is unknown, but, as suggested above for liver and kidney, control of translation of the PDK4 message is a possibility. Since the PDK4 message was dramatically increased, perhaps a longer period of starvation than used in this study would have invoked an increase in PDK4 protein in this tissue. On the other hand, the white adipocyte is unique in that it is the only cell in which insulin has been firmly established to affect short-term regulation of the PDH complex activity (35), apparently by controlling PDH phosphatase rather than PDK activity (36). Perhaps short-term regulation of the PDH phosphatase by insulin provides sufficient control of PDH complex activity in this tissue. Perhaps a long-term control mechanism would interfere with operation of a short-term, rapidly responding mechanism that either is needed or is more appropriate for control of the PDH complex activity in adipose tissue. Starvation has no effect on the phosphorylation and activity state of the PDH complex in the brain of the rat (37). This makes brain unique among all tissues of the rat that have been examined. It is consistent with glucose remaining the most important fuel for the brain energy requirement during starvation of the rat (38, 39). However, this stands in contrast to the human brain in which ketone bodies replace a significant part of the need for glucose for the brain energy requirement during starvation for several weeks (40, 41). PDK and PDH phosphatase activities are expressed in rat brain (42) and altered activities of the complex as a consequence of phosphorylation and dephosphorylation have been demonstrated in vitro in response to various incubation conditions (43) and in vivo in response to the PDK inhibitor dichloroacetate (44, 45). Stable changes in PDK activity that would impose a longer-term regulation have not been reported for rat brain, and our finding that starvation is without effect
upon PDK isoenzyme protein in this tissue is consistent with these findings. It is presumably unnecessary and perhaps undesirable to establish a long-term control mechanism that would interfere with rapid and complete activation of the brain PDH complex of the rat. Although no increase in brain PDK4 protein occurred in response to starvation, a large increase in the abundance of brain PDK4 mRNA was induced. As discussed before, this may reflect control at the level of translation. No attempt was made in this study to determine the regional and cell specificity of PDK4 expression in the brain. Nevertheless, since the message for PDK4 is increased, the potential is there for increased expression of PDK4 protein in at least some cells of the brain, which might occur with starvation of longer duration. An increase in the amount of specific PDK isoenzymes in response to starvation has now been demonstrated in five metabolically important tissues of the body, i.e., heart (13), skeletal muscle (14), and, in this study, liver, kidney, and lactating mammary gland. This change in the PDK protein amount and therefore PDK activity causes inactivation of the PDH complex by phosphorylation. Alteration of PDK gene expression may therefore provide a long-term mechanism for the regulation of the PDH complex. This is supplemented by short-term control exerted by acetyl-CoA, NADH, and pyruvate upon the specific activity of the PDH complex. Increased amounts of PDK2 and PDK4 during starvation may function as a safeguard against fluctuation in concentrations of these compounds that affect kinase activity in the short term. Furthermore, the time required for down regulation of the PDK isoenzymes imposes a lag upon the rate at which the PDH complex is reactivated in response to food intake by a starved animal (46). Maintaining the inactive state of pyruvate dehydrogenase in the early stages of these treatments assures continued availability of gluconeogenic substrates for the resynthesis of glycogen stores in tissues. We have proposed that free fatty acids are physiologically important inducing agents for PDK4 gene expression in skeletal muscle (14). Treatment of animals with WY-14,643, a synthetic agonist for the peroxisomal proliferator-activated receptor ␣ (PPAR␣), induces PDK4 in skeletal muscle (14) and several other tissues. (P. Wu and R. A. Harris, unpublished studies). The message level for PDK4 in the heart of the ground squirrel also increases dramatically during hibernation (47), a metabolic state like starvation and diabetes in which stored body fat is mobilized and serum free fatty acid levels are increased. Since some of the naturally occurring fatty acids are potent PPAR␣ agonists (48), activation of this receptor by free fatty acids may be a major part of the mechanism responsible for increased PDK4 gene expression in various tissues dur-
EXPRESSION OF PYRUVATE DEHYDROGENASE KINASE ISOENZYME
ing starvation, diabetes, and hibernation. However, it remains to be established whether free fatty acids are truly responsible for increased PDK4 gene expression. Factors involved in regulation of PDK2 expression likewise remain to be established. Nonetheless, the present study makes it apparent that starvation increases the relative amounts of PDK2 and PDK4 in several major tissues of the body as part of the mechanism designed to inactivate the PDH complex and thereby conserve compounds that can be converted to glucose. ACKNOWLEDGMENTS This work was supported by grants from U.S. Public Health Services (PHS DK47844), the Diabetes Research and Training Center of Indiana University School of Medicine, the Grace M. Showalter Residuary Trust, and the American Heart Association, Indiana Affiliate, Inc. (Postdoctoral Fellowship, P.W.).
REFERENCES 1. Yeaman, S. J. (1989) Biochem. J. 257, 625– 632. 2. Behal, R. H., Buxton, D. B., Robertson, J. G., and Olson, M. S. (1993) Annu. Rev. Nutr. 13, 497–520. 3. Popov, K. M., Kedishvili, N. Y., Zhao, Y., Shimomura, Y., Crabb, D. W., and Harris, R. A. (1993) J. Biol. Chem. 268, 26,602– 26,606. 4. Popov, K. M., Kedishvili, N. Y., Zhao, Y., Gudi, R., and Harris, R. A. (1994) J. Biol. Chem. 269, 29,720 –29,724. 5. Gudi, R., Bowker-Kinley, M. M., Kedishvili, N. Y., Zhao, Y., and Popov, K. M. (1995) J. Biol. Chem. 270, 28,989 –28,994. 6. Rowles, J., Scherer, S. W., Xi, T., Majer, M., Nickle, D. C., Rommens, J. M., Popov, K. M., Harris, R. A., Riebow, N. L., Xia, J., Tsui, L. C., Bogardus, C., and Prochazka, M. (1996) J. Biol. Chem. 271, 22,376 –22,382. 7. Bowker-Kinley, M. M., Davis, W. I., Wu, P., Harris, R. A., and Popov, K. M. (1998) Biochem. J. 329, 191–196. 8. Hutson, N. J., and Randle, P. J. (1978) FEBS Lett. 92, 73–76. 9. Fuller, S. J., and Randle, P. J. (1984) Biochem. J. 219, 635– 646. 10. Denyer, G. S., Kerbey, A. L., and Randle, P. J. (1986) Biochem. J. 239, 347–354. 11. Sugden, M. C., Holness, M. J., Donald, E., and Lall, H. (1999) Metabolism 48, 707–715. 12. Baxter, M. A., and Coore, H. G. (1978) Biochem. J. 174, 553–561. 13. Wu, P., Sato, J., Zhao, Y., Jaskiewicz, J., Popov, K. M., and Harris, R. A. (1998) Biochem. J. 329, 197–201. 14. Wu, P., Inskeep, K., Bowker-Kinley, M. M., Popov, K. M., and Harris, R. A. (1999) Diabetes 48, 1593–1599. 15. Rougraff, P. M., and Paxton, R. (1987) Comp. Biochem. Physiol. 86B, 601– 606. 16. Blair, P. V. (1977) Arch. Biochem. Biophys. 181, 550 –568. 17. Titheradge, M. A., and Coore, H. G. (1997) Int. J. Biochem. 8, 433– 436. 18. Gianotti, M., Clapes, J., Llado, I., and Palou, A. (1998) Life Sci. 62, 1889 –1899. 19. Hatefi, Y., and Lester, R. L. (1958) Biochim. Biophys. Acta 27, 83– 88. 20. Lai, J. C., and Clark, J. B. (1979) Methods Enzymol. 55, 51– 60.
7
21. Zhao, Y., Popov, K. M., Shimomura, Y., Kedishvili, N. Y., Jaskiewicz, J., Kuntz, M. J., Zhang, B., and Harris, R. A. (1994) Arch. Biochem. Biophys. 308, 446 – 453. 22. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2 ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 23. Chan, Y. L., Olvera, J., and Wool, I. G. (1983) Nucleic Acids Res. 11, 7819 –7831. 24. Jones, B. S., Yeaman, S. J., Sugden, M. C., and Holness, M. J. (1992) Biochim. Biophys. Acta 1134, 164 –168. 25. Priestman, D. A., Donald, E., Holness, M. J., and Sugden, M. C. (1997) FEBS Lett. 419, 55–57. 26. Sugden, M. C., Fryer, L. G., Orfali, K. A., Priestman, D. A., Donald, E., and Holness, M. J. (1998) Biochem. J. 329, 89 –94. 27. Priestman, D. A., Mistry, S. C., Halsall, A., and Randle, P. J. (1994) Biochem. J. 300, 659 – 664. 28. Wieland, O., Siess, E., Schulze-Wethmar, F. H., Funcke, H. G., and Winton, B. (1971) Arch. Biochem. Biophys. 143, 593– 601. 29. Munday, M. R., and Williamson, D. H. (1981) Biochem. J. 196, 831– 837. 30. Mercer, S. W., and Williamson, D. H. (1987) Biochem. J. 242, 235–243. 31. Hagopian, K., Butt, J., and Munday, M. R. (1991) Comp. Biochem. Physiol. 100, 527–534. 32. Hagopian, K., and Munday, M. R. (1997) Biochim. Biophys. Acta 1336, 474 – 484. 33. Holness, M. J., and Sugden, M. C. (1990) Biochem. Soc. Trans. 18, 591–592. 34. Stansbie, D., Denton, R. M., Bridges, B. J., Pask, H. T., and Randle, P. J. (1976) Biochem. J. 154, 225–236. 35. Denton, R. M., McCormack, J. G., Rutter, G. A., Burnett, P., Edgell, N. J., Moule, S. K., and Diggle, T. A. (1996) Adv. Enzyme Regul. 36, 183–198. 36. Rutter, G. A., Diggle, T. A., and Denton, R. M. (1992) Biochem. J. 285, 435– 439. 37. Siess, E. A., Wittmann, J., and Wieland, O. H. (1971) HoppeSeyler’s Z. Physiol. Chem. 352, 447– 452. 38. Hawkins, R. A., Mans, A. M., and Davis, D. W. (1986) Am. J. Physiol. 250, E169 –E178. 39. Cherel, Y., Burnol, A. F., Leturque, A., and Le Maho, Y. (1988) Metabolism 37, 1033–1039. 40. Owen, O. E., Morgan, A. P., Kemp, H. G., Sullivan, J. M., Herrera, M. G., and Cahill, G. F., Jr. (1967) J. Clin. Invest. 46, 1589 –1595. 41. Hasselbalch, S. G., Knudsen, G. M., Jakobsen, J., Hageman, L. P., Holm, S., Paulson, and OB (1994) J. Cereb. Blood Flow Metab. 14, 125–131. 42. Sheu, K. F., Lai, J. C., and Blass, J. P. (1984) J. Neurochem. 42, 230 –236. 43. Browning, M., Baudry, M., Bennett, W. F., and Lynch, G. (1981) J. Neurochem. 36, 1932–1940. 44. Morgan, D. G., and Routtenberg, A. (1982) Brain Res. 251, 391– 394. 45. Abemayor, E., Kovachich, G. B., and Haugaard, N. (1984) J. Neurochem. 42, 38 – 42. 46. Holness, M. J., French, T. J., and Sugden, M. C. (1986) Biochem. J. 235, 441– 445. 47. Andrews, M. T., Squire, T. L., Bowen, C. M., and Rollins, M. B. (1998) Proc. Natl. Acad. Sci. USA 95, 8392– 8397. 48. Forman, B. M., Chen, J., and Evans, R. M. (1997) Proc. Natl. Acad. Sci. USA 94, 4312– 4317.