Biochemistry Of Ketosis.pdf

Part IV: Ketosis THE BIOCHEMISTRY OF KETOSIS* Rubin Bressler Department of Medicine, Duke University Medical Center, Durham, N . C .

Although an extensive literature has accumulated on the subject of ketosis over the years, the biochemical aberrations are still not definitely Ketosis may be considered as a total catabolic state in which there are marked decreases in the syntheses of glycogen, protein, and lipid. Ketosis is promoted when carbohydrate substrate is not available, as in the fasted state or in uncontrolled diabetes mellitus. In these states carbohydrate is rapidly depleted and protein is converted in part to acetoacetate and in part to carbohydrate by gluconeogenetic mechani s m ~ . Lipid ~ is mobilized from peripheral depots to the liver,8-12and the elevations of blood lipids during these states of carbohydrate insufficiency ensure an abundant supply of substrate in the liver for the production of ketone bodies.2. Ketone-body production is a normal function of the liver, and these metabolites are continuously produced by the liver ti The liver is unique among and utilized by the extrahepatic tissues.*-3* animal tissues in accumulating ketone bodies at a time when fatty acid oxidations have increased.l3,l4 The liver is the primary source of net ketone-body production because of its ability to form these compounds in excess of its capacity to metabolize them.*-"-l6 In the liver, ketogenesis is inseparable from fatty acid oxidation, and the enzyme systems of the liver are oriented toward the production of ketone bodies.2.3, l5 The liver has been shown to take up fatty acids in proportion to their concentrations in the perfusing blood.'6 The plethora of fatty acids undergoing oxidative degradation in the liver makes for an abundance of ketone bodies. It has been shown that the ketonemia of diabetic animals correlates well with the lipid content of their 1i~ers.l~ Information concerning the metabolism of the long-chain, saturated fatty acids has become available at a rapid rate during the past ten years.18-21 This decade has seen the elucidation of the mechanism of fatty acid oxidation and the emergence of new concepts concerning the biosynthesis of long-chain, saturated fatty acids.21-26 Investigations in recent years have thrown much light on the mechanism of the &oxidation of fatty acids first proposed by Knoop in 1904.27 Fully H t y years of intensive work were required before Knoop's hypothesis of P-oxidation of fatty acids *Supported in part by grants from the U. S. Public Health Service, OG-5 and H7061. 735

736

Annals New York Academy of Sciences

could be confirmed by studies of the enzymatic steps involved. The complete definition of the individual enzymatic reactions was finally achieved in several laboratories by 1953.22* ** This accomplishment was in large part due to some important discoveries that preceded it. These included the following events: (1) the finding that the enzymes of fatty acid oxidation were located in the mitochondria and could be s o l ~ b j l i z e d30;~ ~ . (2) the discovery of coenzyme A (CoA) and the recognition of its key role in acetate metabolism by Lipmann et d 3 1 ; (3) the discovery by Lynen and Reichert that the functional group of coenzyme A was the thiol group, and the isolation of acetyl coenzyme A by these 33; (4) the recognition that the active form of not only acetate but also of higher fatty acids and substituted fatty acids were acyl thioesters of coenzyme A6v and (5) the development of methods for the large scale preparation of coenzyme A.% Shortage of coenzyme A, which had been a major problem, was no longer an impediment to work in the field of fatty acid oxidation. These developments were essential preliminaries to the reconstruction of the enzymes of the B-oxidation system. The sequence of these reactions and the enzymes involved are shown in equations (1- 6).

0

It

+ COASH + ATP ,

fatty acid thiokinaee

RCHzCHzGOH

A

0

I1

+ PPi + AMP

(1)

+ FADH2-acyldehydrogenase

(2)

RCH~CH~--C-SCOA

0

II

RCH~CHZC-SCoA

+ FAD-acyldehydrogenase 0

II

RCH =CH-G-SCoA

0

0

II

RCH =CH-G-SCoA

+ H201-,

enoyl hydrase

II

RCHOHCHzC--SCoA

(3)

0

II

+

8-hydroxyacyldehydrogenase

R C H O H C H ~ C ~ ~ C ODPN A ,-

0

0

II

I1

RG--CHz(=--SCoA

+ COASH

A

0

0

II

II

R-CCHZC-SCOA

+ DPNH + Hf

0

0

thiolase

11

RC--SCoA

II + CH3-C-SCOA

(4)

(6)

Bressler : The Biochemistry of Ketosis

737

19* 22 The Each of these reactions has been shown to be equilibrium constants of reaction 14 were ascertained to be near unity, whereas the thiolase reaction was found to lie far in the direction of cleavage of acetoacetyl C O A . ~n.. 23 The liver produces acetoacetic acid very actively from acetoacetyl CoA but is relatively incapable of reactivating the acetoacetate to acetoacetyl C O A . ~ - ~l5< The extrahepatic tissues reactivate acetoacetate by a mechanism that involves succinyl CoA and is catalyzed by the thiophorase system.6.13. l4. l9 This reaction is shown in equation 6.

CH3CCHZCOH

II

0

II

0

acetoacetic acid

+ CHzCOOH

thiophorase

I

CHzCOSCoA succinyl CoA

CHzCOOH

I

+ CH~CCHZCSCOA(6) II

CHzCOOH

0

II

0

acetoacetyl CoA

succinic acid

This enzyme system is ubiquitous except for the liver, and this enables the extrahepatic tissues to readily metabolize acetoacetate.2*3, 6 , 1 3 9 l5 The acetoacetic acid thiokinase system is another enzymatic mechanism for reactivating acetoacetate in extrahepatic tissues, and which is absent in the 1i~er.l~. 14, 36 This reaction is shown in equation 7.

CH3CCHzCOH

II

0

I1

+ ATP + COASH

acet oncetic thiokinase

0

*

+

CH~CCHZCSCOA AMP

II

II

0

+ PPi

(7)

0

However, the existence of this enzyrde system has never been adequately documented.13*14. 35 The deficiency of the acetoacetic thiophorase in hepatic tissue is manifest in vitro by an inability of liver to oxidize added acetoacetate and in vivo by a net production of a~etoacetate.l-~. 13. l4 The blood and urine ketone bodies are three in number: acetoacetic acid, its nonenzymatic decarboxylation product acetone (25-35 per cent), and D( -))P-hydroxybutyric acid (65-75 per cent).36 The D( -)@-hydroxybutyric acid that comprises over half of the liver’s output of ketone bodies can be reactivated by the liver in contrast to the liver’s inability to reactivate acetoacetate.20s36 The enzyme system is shown in equation 8.

+

CHICHOHCH~COOH ATP 8-Hydroxybutyric acid

+ COASH-

fatty acid activating enzyme

+ Maff

+

A

CH~CHOHCH~COSCOA AMP p-Hydroxybutyryl CoA

+ PPi

(8)

738

Annals New York Academy of Sciences

This enzyme system has been found in heart and kidney, as well as in liver.35 A variety of explanations has been offered to account for the hyperketonemia of diabetes mellitus and starvation. A major issue of discussion is whether the hyperketonemia is a result of an overproduction of ketone bodies by the liver or an underutilization of these metabolites by the liver and peripheral tissues. Siperstein has postulated that ketosis is the result of an overabundance of acetyl CoA in the liver of the fasted or diabetic animal. The quantities of acetyl CoA that are produced from fatty acid oxidations are in excess of that which can be disposed of by a maximally functioning tricarboxylic acid cycle in the liver.3.37* as Siperstein has assumed that the peripheral metabolism of the ketone bodies is unimpaired and that the plethora of acetyl CoA results from the inability of the ketotic animal or human to synthesize fatty acids from acetyl C O A . ~The ~ ~piling ~ up of the acetyl CoA would then result in ketosis. He has localized the block in fatty acid synthesis to a deficiency of reduced triphosphopyridine nucleotide (TPNH), which would-in turn-result from a decreased oxidation of glucose via the hexose monophosphate shunt (HMP shunt).3s In the diabetic state the HMP shunt could be depressed as a result of a failure to activate glucose or changes in the HMP shunt enzyme^.^^^ 42 I n the fasted state there would be a substrate (glucose) deficit. This theory attributes the hyperketonemia to an excess of acetyl CoA that results in part from an accelerated breakdown of fatty acids to acetyl CoA and in part from a marked decrease in the utilization of acetyl CoA for fatty acid ~ y n t h e s i s . ~ ~ - ~ ~ * Siperstein's thesis has been questioned because of a failure of a number of workers to restore fatty acid synthesis in the fasted normal or diabetic liver by the addition of exogenous TPNH.41n44 Although t.here is substantial evidence that the tricarboxylic acid cycle functions normally in the extrahepatic tissues in diabetes, this view is not universally held. Some investigators have implicated a deficiency of tricarboxylic acid cycle intermediates in the ketosis of diabete~,l~-~' whereas others have found normal levels of these m e t a b o l i t i e ~ . ~An ~-~~ impaired capacity of the diabetic animal to metabolize ketone bodies has been demonstrated.''. 51 There seems t,o be agreement in that the mechanism(s) of ketosis is complex and may involve both an increased rate of ketone body production by the liver and a decreased rate of ketone body utilization by the peripheral tissues (in diabetic ketosis). The discussion of ketosis will be concerned primarily with the biochemical aspects of the process. Three topics will be considered: 1. How is acetoacetyl coenzyme A formed? 2. How is acetoacetic acid formed? 3. How is D( -)@-hydroxybutyric acid formed? 41v

Bressler : The Biochemistry of Ketosis

739

Acetoacetyl CoA Synthesis Up to 1953 the origin of acetoacetyl CoA was a subject of controversy. It was uncertain whether AcAcCoA represented the product of condensation of two acetyl CoA units or was the “stump” of a saturated long-chain fatty acid that did not undergo further thiolytic cleavage to acetyl CoA. The isotope data obtained were consistent with the view that all of the two-carbon units removed from the fatty acid chain starting from the carboxyl end were metabolically identical and formed the carboxyl end.of acetoacetate rather than the methyl end. The terminal two-carbon unit was far more reactive in the formation of the methyl end of the acetoacetate.62-54The explanation of this data was found to be a simple solution based on the mechanism of action of the thiolase In the process of fatty acid oxidation the terminal two carbons of the fatty acid represent the only acetyl unit that combines with the thiolase enzyme before passing into the acet,yl CoA “pool.” This occurs in the last step of the ,&oxidation cycle, when acetoacetyl CoA is converted by the thiolase enzyme into acetyl CoA and acety1-Senzyme. Equation 9 shows this sequence. CH3CCHsC-S-enzyme

II

0

II

+ CoASH .

thiolase

.

-

0 CH3CSCoA

II

+ CH,C-S--enzyme II

0

(9)

0

The acetyl-S-enzyme that results from the reaction may then react in either of two ways. It can transfer its acetyl group to coenzyme A (equation 10) or it can condense with a molecule of acetyl CoA from the “pool” to reform acetoacetyl CoA (equation 11). CH3G-S--enzyme

I1

+ CoASH

t.hiolnse

0 CH~C--S-enzyme

II

0

CH3CSCoA

-

II

+ enzyme-SH

(10)

0

+ CH,CSCoA I/

thiolase

0 CH,CCH,CSCoA

II

0

I1

+ enzyme-SH

(11)

0

This explanation for the formation of acetoacetyl CoA fits the experimental data well.62-64When acetoacetate was produced from carboxyl

Annals New York Academy of Sciences

740

labeled octanoic acid, it was found that the labeling in the carbox91 group of t,he acetoacetate was much greater than in the carbonyl group. If the acetoacetyl CoA was produced by a random condensation of acetyl CoA units, then the labeling in tJhe carboxyl and carbonyl groups of the resulting acetoacetate would have been equal. If the acetoacetyl CoA represented the “stump” of t,he fatty acid, then there would have been no radioactivity at all in the resulting acetoacetate. If, however, the synthesis of acetoacetyl CoA occurs by the postulated thiolase mechanism, then the terminal two-carbon group still attached to the enzyme would be unlabeled. This g o u p would take up labeled acetyl CoA from the “pool” to form radiolabeled acetoacetyl CoA, which woiild be labeled in the carboxyl group.2o.54* s6 Equations 12 and 13 show this sequence.

+CH3(CH&i‘OSCoA

@-oxidationsequence

+ + CH360SCo,4+CH,CO-S-enzpe + CH360SCoA --+ CH3COCJH2kOSCoA + enzyme-SH ___)

2 CH&OSCoA

+CH3CO-S-enzyme

(12)

thiulaw

(13)

The omega acetyl group still att.ached to the enzyme takes up labeled acetyl CoA from the “pool,” resulting in carboxyl labeled acetoacetyl CoA. If the omega carbon is labeled, then the label appears in the methyl group of acetoacetate (equat.ions12 and 13). These data confirm the view that the synthesis of acetoacetyl CoA occurs via a condensation of ncetyl CoA units catalyzed by the thiolase enzyme. Amtoacetic Acid Synthesis

Two enzymatic pathways have beeu demonstrated in the production of acetoacetate from acetoacetyl CoA. Lynen and his co-workers57~513 have proposed a two-stage sequence for the production of acetoacetate. This sequence is shown by equations 14 and 16 and is known as the “HMG

+

CH3CCH2CSCoA CH&SCoA

II

0

II

0

ncetonretyl CoA

II

curidensirla enzyme ~

--,

0 ncetyl CoA

HO

I I

CH&CH&SCoA

II

+ GOASH

1 0 CH~OOH 8-h ydroxy-8-methyl glutaryl CoA (HMGCoA)

(14)

Bressler : The Biochemistry of Ketosis

741

HO

I

A *A

cleavage enzyme

CH~CCH~CSCOA

* CH3CCH260H

II

0

I1

0

+

CH3CSCoA

I

(16)

0

HZCOOH

HMGCoA

acetoacetate

aceiyl CoA

shunt." These workers used a31 acetone powder of ox liver as a source of the cleavage enzyme and purified this enzyme by heat denaturation and ammonium sulfate fractionation procedure^.^^ A yeast extract was used as a source of the condensing enzyme.67, 68 Acetoacetyl CoA was generated from acetyl CoA in the presence of thiolase, which waa contained in the yeast preparation.67*59 The HMGCoA condensing enzyme yeast, the thioIase from yeast, and the HMGCoA cleavage enzyme from liver were all shown to be sulfhydryl dependent enzymes.67.69. The condensing enzyme and the cleavage enzyme were both found to be located primarily in the mitochondria.68, 6 1 Condensing enzyme activity was also found to be present in the microsomes that were free of the cleavage enzyme.68.62 The microsomes are loci of cholesterol synthesis, and HMGCoA is a cholesterol precursor.62-64 Hence HMGCoA can be converted to either cholesterol or acetoacetic acid and acetyl CoA, depending on whether the HMGCoA cleavage enzyme is present." The Lynen group demonstrated that acetyl CoA augmented the disappearance of AcAcCoA and increased the formation of acetoacetic acid. They showed that in the presence of unlabeled AcAcCoA and labeled acetyl CoA, the acetoacetate produced was labeled in the carboxyl group, which would have been expected if HMGCOA was an obligatory intermediate in the formation of acetoacetate from AcAcCoA. This is shown by equations 14 and 16. Stern et ~ 1 . 5 9 .Bo* and Segal and Menon61.67. G8 have presented equally convincing evidence that supports the formation of acetoacetic acid by a direct deacylation of acetoacetyl CoA. This reaction is shown by equation 16.

+

AcAcCoA deacylase

CH~COCH~COSCOAH20 --

-.+

CH3COCH2COOH

+ COASH

(16)

Stern's group employed rat and beef liver mitochondria as a source of enzyme and found the following evidence in favor of a deacylation mechanism for the product.ion of acetoacetate from acetoacetyl CoA: (1) acetoacetyl CoA deacylaae activity was present in the mitochondria in sufficient B'; (2) quantity to account for almost all of the acetoacetate formedsS~ acetoacetate production from acetoacetyl CoA is unimpaired in the face of complete inhibition of the HMGCoA condensing and cleavage enzymes

742

Annals New York Academy of Sciences

by iodoacetamide 60, 66; and (3) the procedure used by the Lynen group to purify the HMGCoA cleavage enzyme involves a heating step that causes a loss of both the HMGCoA condensing enzyme (which Lynen added back to his system from a yeast preparation) and the acetoacetyl CoA deacylase enzyme.57+ Stmn ascertained that subjecting the mitochondrial preparation to 50" C. for twenty minutes caused a loss of 90 per cent of its acetoacetyl CoA deacylase activity.69 Stern's work demonstrated that both the HMGCoA shunt mechanism enzymes and the acetoacetyl CoA deacylase enzyme were present in mitochondria, but that the deacylase was the quantitatively more important system. He also found that the Lynen preparation of the HMGCoA cleavage enzyme involved a procedure that destroyed the acetoacetyl CoA deacylase enzyme and thus explained the failure of Lynen's group to find the deacylase activity in their enzyme preparations. Segal and Menon investigated the production of acetoacetate from acetoacetyl CoA in mitochondria1 preparations from normal and alloxan diabetic rat livers.61*67. 68 Their work showed that the formation of acetoacetate proceeds almost, exclusively by the direct, deacylation of acetoacetyl CoA. Radioactive acetyl CoA was incubated with unlabeled acetoacetyl CoA in the presence of mitochondria1 enzymes. Acetoacetate production was ascertained, and the radioactivity in it was determined. 67 If the acetoacetate had been formed via the HMGCoA shunt, it would have had the same high specific activity as that of the starting radioactive acetyl CoA (see equations 14 and 16). The remaining acetoacetyl CoA would have remained unlabeled except for the small amount of radioactive acetoacetyl CoA, which would have formed as a result of the thiolase enzyme acting on radioactive acetyl CoA. If, however, acetoacetate was formed by a direct deacylation of acetoacetyl CoA, it would only have that small amount of radioactivity that was introduced into acetoacetyl CoA by the action of the thiolase enzyme on the radioactive acetyl CoA (see equation 16). The acetoacetate that was produced in these experiments contained only traces of radioactivity, which was consistent with its formation by a direct deacylation of acetoacetyl COA.~' The elimination of acetyl CoA in these experiments did not diminish the amount of acetoacetate formed. This was further evidence for the lack of an HMGCoA shunt operating to produce any appreciable quantity of acet~acetate.~~. 59 These workers studied the formation of acetoacetate from acetoacetyl CoA in alloxan diabetic rat liver preparations.61, 67 A two-fold increase in acetoacetate production was found in the mitochondrial preparations from the diabetic animals.67 The experimental data of Segal and Menon indicate that the bulk of the acetoacetate forming capacity of the liver residues in the mitochondria and that the major, s99

Bressler : The Biochemistry of Ketosis if not the sole, pathway of acetoacetate formation in normal and diabetic 67, livers is through a direct deacylation of acetoacetyl COA.~’. Freezing and thawing the mitochondria caused an augmentation of acetoacetate production in these experiments.61 This treatment causes a disruption of the mitochondrial structure and breaks down the permeability barrier of the intact mitochondria.70 It permits a freer access of added substrate to the mitochondria1 enzymes. This suggests that the intracellular precursors of acetoacetate are formed within the intact mitochondria, with good access to the acetoacetate producing enzymes. The long-chain saturated fatty acids that are undergoing oxidative degradation within the mitochondria are the precursors of acetoacetate.s. Is, l9 Although these fatty acids are converted to their coenzyme A derivatives or “activated” by microsomal enzymes, the oxidative degradation of these “activated” long-chain fatty acyl CoA derivatives occurs n, 29 This is consonant with the observation that in the there is an accumulation of lipid in the liver in starvation and It has become clear that there are at least two enzyme mechanisms concerned with the production of acetoacetate from acetoacetyl CoA that can be demonstrated in liver mitochondrial preparations.”. Bo* 71 The quantitative significance and physiological variations of these systems have not yet been well established. Wieland el al., using rats as experimental animals, found differences in the patterns of enzymatic activity in the ketosis of fasting and the ketosis of alloxan diabetes.?’ They found that in fasting there was .an increase of HMGCoA in the liver which was associated with an eight-foId increase in HMGCoA cleavage enzyme activity and a marked decrease in HMGCoA reductase activity. The cleavage enzyme produces acetoacetate from HMGCoA, whereas the reductase enzyme catalyzes the conversion of HMGCoA to mevalonic acid, a cholesterol precursor*, 6Q This decrease of HMGCoA reductase activity would explain the decreased formation of cholesterol in the fasted state.“, 68. 71 The ketosis of fasting could be explained on the basis of an increased production of acetoacetate via the HMGCoA shunt. The decrease in HMGCoA reductase activity would result in the accumulation of HMGCoA and enhance the formation of acetoacetate by the HMGCoA cleavage enzyme. The acetoacetyl CoA deacylase activity was not elevated in the livers of the fasted animals. The livers of the diabetic animals had a normal HMGCoA reductase activity, a moderate increase in the activity of the HMGCoA cleavage enzyme, and a marked elevation of acetoacetyl CoA deacylase activity. This marked elevation of the deacylase activity and the moderate. elevation of the HMGCoA cleavage enzyme activity both could contribute to the production of acetoacetate by the diabetic rat liver. How-

744

Annals New York Academy of Sciences

ever, since the HMGCoA reductase activity of the diabetic livers was not reduced, HMGCoA could serve as a substrate for both the production of acetoacetate and cholesterol synthesis. The synthesis of cholesterol by the diabetic liver is not 72 These data would attribute the increased production of acetoacetate in fasting to increased activity in the HMGCoA shunt and the increased production of acetoacetate in diabetes to increased activity of the acetoacetyl CoA deacylase enzyme. It has been shown in the fasted guinea pig that the increased production of acetoacetate is in a large part, but not entirely, through the HMGCoA shunt pat,hway." D( -)@-Hydroxybutyric

Acid Metabolism.

It has been known since 1901 that the fasted normal or uncontrolled diabetic animal or human excretes primarily, or solely, the D ( - ) isomer of @-hydroxybutyric acid.73 Since this compound represents from 50 to 75 per cent of the blood ketone a discussion of its metabolism is important in a consideration of the biochemistry of ketosis. There are five enzyme systems that relate to the metabolism of the D ( - ) & hydroxybutyric acid. These will be briefly discussed. I . D( -)j3-Hydroxybutyric acid dehydrogenase. This enzyme system has been found in many tissues, including the heart and 75 The oxidation of D( -)@-hydroxybutyric acid by DPN+ was first described in 1937 by Green et al.74 This reaction is shown by equation 17. The CH3CHOHCHpCOOH D(

-)@-hydroxybutyric acid

+ DPN-t-

CH&OCH&OOH acetoacetic acid

+ DPNH + H+

(17)

enzyme that catalyzed this reaction was shown to be tightly bound to mitochondria and to be closely associated with the electron transport and oxidative phosphorylation systems.''. 75 The enzyme was recently solubilized and was found to have an absolute requirement for l e ~ i t h i n . ~ ~ - ~ ~ This enzyme system is capable of producing D( - )@-hydroxybutyric acid (which is found in the blood and excreted in the urine in ketotic states) from acetoacetic acid, which is excessively produced in the mitochondria in ketotic states. Although the liver is incapable of converting acetoacetate to acetoacetyl CoA, it can reactivate the D( -)p-hydroxybutyric acid to a thioester of coenzyme A (equation 8).a5 Once the D( -) acid has been converted to the D(-)p-hydroxybutyryl CoA, it can be oxidized in the tricarboxylic acid cycle after passing through acetoacetyl CoA and being acted upon by the thiolase enzyme (equations 9 and 10)-which produces acetyl CoA. Scow and Chernick have reported that the utilization of infused D( -)@-hydroxybutric acid by the diabet.ic rat is impaired

745

Bressler : The Biochemistry of Ketosis

and that this defect is corrected by insulin therapy." Beatty et al. have shown that the utilization of acetoacetate by diabetic muscle is depresseds1 These findings mitigate strongly against the long-held belief that the utilization of ketone bodies by the peripheral tissues of the diabetic is normal.78 The evidence now favors a combination of overproduction of acetoacetate by the liver and an underutilization of both acetoacetate and D( -)P-hydroxybutyrate by the peripheral tissues as being responsible for the more severe ketosis of diabetes. In the milder ketosis encountered in the fasted state, overproduction of acetoacetate by the liver is probably the cause of the ketosis because utilization of ketone bodies has been found, to be normal.2,13, l9 The impaired utilization of ketone bodies by the muscle in diabetes has not been found in the liver.', 15, 79 It i s important to note that the impaired utilization of acetoacetate and D( -)P-hydroxybutyrate does not mean that there is necessarily a defect in the tricarboxylic acid cycle of the diabetic muscle. The utilization of these compounds entails overcoming a permeability barrier and an activation process, either of which may be at fault (equations 6 and 8). There have been speculations that the ketosis of diabetes is a result of a deficiency of tricarboxylic acid cycle intermediates in the liver,m and decreased levels of oxalacetate have been found by some.% However, this view is not shared by other investigators who have found that the respiration of the diabetic rat liver is norma1,*1as are the hepatic levels of oxa1acetate.m 2. L( +)P-Hydroxybutyryl CoA dehydrogenase. This enzyme catalyzes the reversible oxidation of L( +)P-hydroxybutyryl CoA by DPN+. This is shown by equation 18. This system was described almost simultaneously 177

+ DPN+

L( +)CH~CHOHCH~COSCOA L( +)@-hydroxybutyrylCoA

L -

+

CH~COC&COSCOA DPNH Acetoacetyl CoA

+ H+

(18)

by investigators at the Enzyme Institute of the University of Wisconsina2 and by Lynen and Ochoa.22 The enzyme is mitochondrial. It is found in association with the enzymes of Bdxidation.6sn*66 The L( +) isomer produced may be converted into the D( -) isomer. 3. P-Hydroxybutyryl CoA racemase and D( -)P-hydroxybutyryl CoA dehydrogenase. Stern et al.% have reported on the interconversion of D(-) P-hydroxybutyryl CoA into L( +)@-hydroxybutyryl CoA. The racemase enzyme that catalyzes this interconversion was found in mitochondria1 extracts of liver. The reaction shown in equation 19 is reversible and D( -)CH~CHOHCH&OSCOA

-racewe

L( +)CH~CHOHCH~COSCOA

(19)

746

Annals New York Academy of Sciences

was not affected by the addition of DPN+, which suggested that the interconversion mechanism was a direct Wakil isolat,ed a mitochondria1 enzyme fraction from liver, which catalyzed the conversion of D( -)&hydroxybutyrvl CoA to L (+) phythoxybutyryl CoA in the presence of DPN+. Catalytic amounts of DPN+ (10-5 M) were an absolute requirement.& This led Wakil to postulate a D( -)ghydroxybutyryl CoA dehydrogenase in his mitochondrial preparations, and he proposed the following reaction sequence for the interconversion of the L ( + ) CoA and D ( - ) CoA isomers. This sequence is shown in equations 20-22. @

D(

+ llPN+.

-)CH3CHOHCH&OSCoA D(

D(-)

dehydrogenase

a

-)&hydroxybutyryl CoA

CHaCOCHZCOSCoA acetoacetyl CoA

+

CH~COCH~COSCOA DPNH

+ H+

(20)

L( +) dehydrogentwe -

acctoacetyl CoA

D

+ DPNH + H+ +

L(+)CH~CHOHCH~COSCOADPNf

(21)

(-) CH~CHOHCH~COSCOA L (+) CH~CHOHCH&OSCOA SUM

(22)

-

a

Although preparations of the L(+) CoA dehydrogenase have been obtained that are free of the D(-) CoA dehydrogenase, the reverse separation has not yet been achieved. The interrelationships of these enzyme systems are shown in equation 23. All of these metabolites can enter the raceme

L(+)COA-D(-)COA

(23)

tricarboxylic acid cycle. In addition, they are capable of becoming ketone bodies, if acted upon by deacylase enzymes. Although the acetoacetyl CoA deacylase is known, a D( -)&hydroxybutyryl CoA deacylase has not yet been described. The two isomers of 8-hydroxybutyryl CoA, however, are also potential ketone bodies because of their capacities to be converted to acetoacetyl CoA. 4. Aceioacetyl CoA reductase. Wakil and BressleF and Lynens8 demonstrated the presence of an enzyme in liver and yeast that catalyzes the reduction of acetoacetyl CoA to D( -)p-hydroxybutyryl CoA in the presence of TPNH. This reaction is shown in equation 24. This enzyme can

747

Bressler : The Biochemistry of Ketosis

+

CH~COCH~COSCOATPNH acetoacetyl CoA

+ H+

---f

D( -)CH3CHOHCH2COSCoA D( -)@-hydroxybutyryl CoA

+ TPN+

(24)

be separated from the L(+))P-hydroxybutyryl CoA, as well as from the D( - )P-hydroxybutyryl CoA and racemase. DPNH could substitute for TPNH to the extent of about 10-20 per cent.44 The physiological role of the reductase is not known ns yet. Its role, if any, in the synthesis of long-chain saturated fatty acids is still not ~ e t t l e d . ~A~ *possible ~ role for this enzyme in ketone body formation has been suggested.44 It is known that the excess of acetoacetate that occurs in diabetes and fasting may be reduced by DPNH to ~(-))P-hydroxybutyric acid by the mitochondrial D( - )@-hydroxybutyric acid dehydr~genase.~~, 76 Though this reaction is reversible, its link to the electron transport chain and oxidative phosphorylation systems suggests that its physiological role may be in an oxidative pathway, rather than in the formation of D( -)P-hydroxybu tyrate. Wakil and Bressler have proposed an alternative pathway for the formation of D( -)/3-hydroxybutyrate.44 This pathway involves the reduction of acetoacetyl CoA by TPNH to D( -))P-hydroxybutyric acid and coenzyme A. This reaction occurs in the extramitochondrial portion of the cell. The D(-))P-hydroxybutyric acid could then be oxidized by mitochondria in the presence of DPN+ to acetoacetate and DPNH. The DPNH could then be oxidized by the specific electron transport chain (by oxygen) with production of ATP. This scheme would be a means of introducing “reducing equivalents” into themito~hondria.8~ Normally, TPNH functions as the electron donor in fatty acid m The fatty acids then undergo oxidative degradation to COz, water, and energy. In starvation and diabetes, fatty acid synthesis is markedly whereas TPNH levels are not.40.41 This pathway reduced,3@. 89, would provide an alternative route for the oxidation of TPNH by mitochondria. This acetoacetyl CoA reductase activity has been shown to rise in the fasted state, at a time when fatty acid synthesis was depressed .90 per cent.” Hence the TPNH “reducing equivalents” would be transferred to the mitochondria for energy production, even though the intermediary phase of fatt,y acid synthesis, ie., TPNH storage, did not occur. This scheme is shown in equation 26.

+

+

+

+

(l)AcAcCoA TPNH H+ + D( -)&OH TPN+ CoASH ( 2 ) ~ ( -)&OH DPN+ + AcAc DPNH H+

+

SUM = TPNH+

+

+ H+-+ DPNH+ + H+

+

(26)

748

Anrials New York Academy of Sciences

The same result would obtain whether D( -)p-hydroxybutyric acid or resultled in reaction 25.' There are DPN+linked dehydrogenase enzymes for both these substrates in the mitochondria, and DPNH would result in either case. These are shown in equations 17 and 20. D( - ))P-hydroxybutyryl CoA

Sicmrrary The ketosis of diabetes appears to be a result of an overproduction of acetoacetate by the liver and an underutilization of both acetoacetate and D( -))P-hydroxybutyrate by the muscle. The ketosis of fasting results from hepatic overproduction of acetoacetate. The biochemistry of acetoacetyl CoA, acetoacetic acid, and D( -)@-hydroxybutyric acid was discussed. Acetoacet.yl CoA is formed by the condensation of acet,yl CoA moities catalyzed by the thiolase enzyme. There are two known mechanisms for the production of acetoacetic acid, and these may vary differently in the fasted and diabetic states. The metabolism of p-hydroxybutyric acid and its acyl CoA derivatives has been discussed. It has been proposed that acetoacetyl CoA reductase plays a role in energy transfer in the ketotic state. References 1. LANGDON,R. G.

1960. Hormonal regulat,ion of fatty acid metabolism. I n Lipide Metabolism. B.Bloch, Ed. John Wiley and Sons. New York, N. Y. Pp. 238-285. 2. VAN ITALLIE, T. B. & S. S. BERGEN.1961. Iietogenesis and hyperketonemia. Am. J. Med. 31: 909. 3. KREBS,H. -4. 1961. The biochemical leison in ketosis. ,4.M.A. Arch. Int. Med. 107: 51. 4. ENGEL, F. L. 1957. The influence of the endocrine glands on fatty acid and ketone body metabolism. A.M.A. Arch. Internal Med. 100: 18. 5. STADIE, W-. 1958. Ketogenesis. Diabetes. 7: 175. 6. GREEN,D. E. 1954. Fatty acid oxidation in soluble systems of animal tissues. Biol. Rev. 29: 330. 7. GREENBERG,D. M. 1961. Carbon catabolism of amino acids. In Metabolic Pathways. D. M. Greenberg, Ed. Academic Press, New York, N. Y. Pp. 79-162. R . S., JR. & A. CHERKES. 1956. Unesterified fat,t,yacid in human blood 8. GORDON, plasma. J. Clin. Invest. 36: 206. 9. RIERMAN, E. L., V. P. DOLEBE T. N. ROBERTS. 1957. An abnormality of nonesterified fatty acid metabolism in diabetes mellitus. Diabetes. 6: 475. 10. DOLE,V. P. 1958. Pat metabolism in diabetes. Bull. 11'. Y. Acad. Med. 34: 21. 11. I,AURELL, S. 1956. Plasma free fatty acids in diabetic acidosis and starvation. Scand. J. Clin. & Lab. Invest. 8: 81. 12. GORDON,R. S., JR. 1957. Unesterified fatty acid in human blood plasma. 11. Transport function of unesterified fatty acid. J. Clin. Invest. 36: 810. 13. MCCANN,W. P. 1957. Oxidation of ketone bodies by miiochondria from liver and peripheral tissues. J. Biol. Chem. 226: 15.

Bressler : The Biochemistry of Ketosis

749

14. STERN,J. R., M. J. COON& A. DELCAMPILLO.1953. Enzymatic breakdown and synthesis of acetoacetate. Nature. 171: 28. 15. FRITZ, I. 1961. Factors influencing the rates of long chain fatty acid oxidation and synthesis in mammalian systems. Physiol. Rev. 41: 52. 16. GORDON, R. S., JR. 1957. Unesterified fatty acid in h u m m blood plasma. 11. Transport function of unesterified fatty acid. J. Clin. Invest. 36: 810. 17. Scow, R. 0. & S. S. CHERNICK.1960. Hormonal control of protein and fat metabolism in the pancreatectomized rat. Recent Prop. in Hormone Research. 16: 497. 18. LYNEN, F. 1952-1953. Acetyl Coenzyme A and the “fatty acid cycle.” Harvey Lectures. 48: 210. 19. GREIN, D. E. & S. J. WAKIL. 1960. Fatty acid oxidation and synthesis. I n Lipide Metabolism. John Wiley and Sons, Inc. New York, N. Y. Pp. 1 4 0 . 20. GREEN,D. E. 1956. Fatty acid oxidation and synthesis in a system of soluble enzymes. In Biochemical Problems of Lipids. Pp. 233-245. Popjak and LeBreton, Eds. Rutterworth. London, England. 21. WAKIL,S. J. 1961. The mechanism of fatty acid synthesis. J . Lipid Research. 2: 1. 22. LYNEN,F. & S. OCHOA. 1953. Enzymes of fatty acid metabolism. Biochim. et Biophys. Acta. 12: 299. 23. LYNEN,F. 1959. Participation of acyl-CoA in carbon chain biosynthesis. J. Cellular Comp. Physiol. 54: 33. -. 24. BRADY,R. 0. & S. GURIN. 1952. Biosynthesis of fatty acids by cell free or water soluble enzyme systems. J. Biol. Chem. 199: 421. 25. DITURI, F. & S. GURIN. 1953. Lipogenesis by homogenates or particle free extracts of rat liver. Arch. Biochim. et Biophys. 43: 231. 1959. The mechanism of fatty acid synthesis. 26. WAKIL, S. J. & J. GANGULY. J. Amer. Chem. SOC.81: 2597. 27. KNOOP,F. 1904. Der abbau aromatischer fettsiiuren im Tierkorper. Beitr. chem. physiol. Path. 6: 150. 28. BEINERT,H., R. M. BOCK,D. S. GOLDMAN, D. E. GREEN,H. R. MAHLER,S. MII, P. G. STANBLY & S. J. WAKIL. 1953. The reconstruction of the fatty acid oxidase system of animal tissues. J. Amer. Chem. SOC.78: 4111. 29. LEHINGER,A. L. & E. P. KENNEDY.1949. Oxidation of fatty acids and tricarboxylic acid cycle intermediates by isolated rat liver mitochondria. J. Biol. Chem. 179: 957. 30. DRYBDALE, G. R. & H. A. LARDY. 1953. Fatty acid oxidation by a soluble enzyme system from mitochondria. J. Biol. Chem. 202: 119. 31. LIPMANN, F. 1948-1949. Biosynthesis mechanisms. Harvey Lectures. 44: 99. 32. LYNEN,F. & E. REICHERT. 1951. Zur chemischen struktur der aktivierten essigsaure. Z. Angew. Chem. 63: 47. 33. LYNEN,F. & K. DECKER. 1957. Das coenzym A und seine biologischen functionen. Ergeb. Physiol. biol. chem. u. exptl. pharmakol. 49: 327. D. E. GREEN,D. A. BUYEKE, R. E. HANDBCHU34. BEINERT,H., R. W. VON KORFF, MACHER, H. HIQGINB & F. STRONG A method for the purification of coenzyme A fromyeast. J. Biol. Chem. 200: 385. 35. MAELER,H. R., S. J. WAIUL& R. M. BOCK. 1953. Studies on fatty acid oxidation. I. Enzymatic activation of fatty acids. J. Biol. Chem. 204: 453. 36. G-LTOFT, A. 1951. Ratio of 8-hydroxybutyric acid to acetoacetic acid in the blood under various experimental conditions. Acta Physiol. Scand. 24: 35. 37. SIPERSTEIN,M. D. & V. M. FAGAN.1958. Studies on the relationship between glucose oxidation and intermediary metabolism. 11. The role of glucose oxidation in lipogenesis in diabetic rat liver. J. Clin. Invest. 37: 1196. 38. SIPERBTEIN,M. D. 1959. Inter-relationships of glucose and lipid metabolism. Am. J. Med. 26: 685.

750

Annals New York Academy of Sciences

39. VANBRUWEN,J. T., T. T. HUTCHENS,C. K. CLAYCOMB, W. J. CATHEY& E. S. WEST. 1952. The effect of fasting on lipogenesis in the rat. J. Biol. Chem. 196: 389. 40. MATTEES, K . J., S. ABRAHAM & I. L. CHAIKOFF. 1960. Fatty acid synthesis from acetate by normal and diabetic rat liver homogenate fractions. 11. Effect of microsomes and oxidation of substrates. J. Biol. Chem. 236: 2560. 41. ABRAHAM, S., K. J. MA~THES & I. L. CHAIKOFF. 1960. Fatty acid synt.hesis from acetate by normal and diabetic rat liver homogenate fractions. I . A comparison of cofactor requirements. J. Biol. Chem. 236: 2551. 42. GLOCK,G. E., P. A. MCLEAN. 1955. Preliminary investigation of the hormonal control of the hexose monophosphate oxidative pathway. Biochem. J. 61:390. 43. SAUER,F. 1960. Fatty acid, cholesterol and acetoacetate biosynthesis in liver homogenates from normal and starved guinea pigs. Can. J. Biochem. Physiol. 38: 635. 44. WAKIL,S. J. & R. BRESSLER. 1962. Studies on the mechanism of fatty acid synthesis. X. Reduced triphosphopyridine nucleotide-acetoacetyl coenzyme A reductase. J. Biol. Chem. 237: 687. 45. FROHMAN, C. E., J. M . ORTEN& A. H. SMITH. 1951. Levels of acids of the citric acid cycle in tissues of normal and diabetic rats. J. Biol. Chem. 193: 803. 46. STADIE,W. C., J. A. ZAPP & F. D. LUKENS. 1940. The effect of insulin on the ketone metabolism of normal and diabetic cats. J. Biol. Chem. 132: 423. 47. HEAITY,C. H., E. S. WEST & R. M. BOCEK. 1958. Effect of succinate, fummarate and oxalacet.ate on ketone body production by liver slices from nondiabetic and diabetic rats. J. Biol. Chem. 230: 725. 48. PASELLA, P., C. BAGLIONI, C. TURANO & N. SILIPRANDI.1958. Action of citrate and oxalacetate on dietary and diabetic ketosis. Lancet. 1: 1097. 49. DEUL,H. J., JR.,S. MURRAY & L. HALLMAN. 1937. A comparison of the ketolytic effect of succinic acid with glucose. Proc. SOC.Exptl. Biol. Med. 37: 413. 50. SHAW,W. V. & D. F. TAPLEY.1958. Owloacetate in the livers of alloxan diabetic rats. Biochim. et Biophys. Acta. 30: 426. 51. BEATTY, C. H., R. D. PETERSON, R. M. BOCEK& E. S. WEST. 1959. Acetoacetate and glucose uptake by diaphragm and skeletal muscle from control and diabetic rats. J. Biol. Chem. 234: 11. 52. CRANDALL, D. I . & S. GURIN. 1949. Studies of acetoacetate formation with labeled carbon. I. Experiments with pyruvate, acetate and fatty acids in washed liver homogenates. J. Biol Chem. 181: 829. 53. GEYER,11. P., M. CUNNINGE~AM & T. PENDERGAST. 1951. Inhibition studies on radioactive fatty acid metnboliim. J. Biol. Chem. 188: 185. 54. LEHNINGER, A. L. 1952. Enzymatic oxidation of fatty acids. Biochemical Society Symposia No. 9. Lipid Metabolism. Pp. 66-79. 1953. Assymmetric labeling of acetoacetate by 55. BEINERT,II. & P. G. STANSLY. enzymatic exchange with acetyl coenzyme A. J. Biol. Chem. 204: 67. 56. LYNEN,F. 1955. Lipide metabolism. Ann. Review Biochem. 24: 653. C. BUBLITZ,B. SORBO & L. KROPLIN-RUEFF. 1958. 57. LYNEN,P.,U. HENNING, Der. chemische mechanisms der acetessigsaurebiedung in der leber. Biochem. Z. 330: 269. 58. BUCENER,N., P. OVERATE & F. LYNEN. 1960. ,¶-hydroxy-fl-methylglutaryl coenzyme A reductase, cleavage and condensing enzymes in relation to cholesterol formation in rat liver. Biochim. et Biophys. Acta. 40: 491. 59. STERN,J. R. & G. E. MILLER. 1959. On the enzymatic mechanism of acetoacetate synthesis. Biochim. et Biophys. Acta. 36: 576. 60. DRIIMMOND, G. 1. & J. R. STERN.1960. Enzymes of ketone body metabolism: 11. Properties of an acetoacetate synthesizing enzyme prepared from ox liver. J. Biol. Chem. 236: 318. 61. SEGAL, H. L. & G. K. K.MENON. 1961. Acetoacetate formation from acetoacetyl coenzyme A in rat liver mitochondria. Effects of endocrine state and nature of the system. J. Biol. Chem. 236: 2872.

Bressler : The Biochemistry of Ketosis

75 1

62. DURR,I. F. & H. RUDNEY. 1960. The reduction of fl-hydroxy-8-methylglutaryl coenzyme A to mevalonic acid. J. Biol. Chem. 236: 2572. 1957. The biosynthesis of 8-hydroxy-8-methyl63. RUDNEY,H. & J. J. FERGUSON. glutaryl coenzyme A. J. h e r . Chem. Soc. 79: 5680. 64. CORNFORTH, J. W. 1959. Biosynthesis of fatty acids and cholesterol considered as chemical processes. J. Lipid Research. 1: 3. B. K., W. C. ROBINSON& M. J. COON. 1955. The enzymatic 65. BACHHAWAT, cleavage of @-hydroxy-fl-methylglutarylcoenzyme A to acetoacetate and acetyl coenzyme A. J. Biol. Chem. 216: 1955 M. J. COON& A. DEL CAMPILLO. 1960. Enzymes 66. STERN,J. R., G. I. DRUMMOND, of ketone body metabolism. I. Purification of an acetoacetate synthesizine enzyme from ox liver. J. Biol. Chem. 236: 313. 67. SEGAL,H. L. & G. K. K. MENON. 1960. Elevation in acetoacetate formation from acetoacetyl coenzyme A in diabetic rat liver preparations. Biochem. Biophys. Research Comm. 3: 28. 68. SEQAL,H. L. & G. K. K. MENON. 1960. Evidence for the formation of acetoacetate by direct deacylation of acetoacetyl CoA in liver mitochondria. Biochem. Biophys. Research Comm. 3:406. 1961. Enzymes of acetoacetate formation. 69. CALDWELL, I. C. & G. I. DRUMMOND. Biochem. Biophys. Research Comm. 4: 127. 70. GREEN, D. E. & J. JHRNEFELT. 1959. Enzymes and biological organization. 1959. Perspectives Biol. Med. 2: 163. 71. WIELAND,O., G . LOFFLER, and I. NEUFELDT. 1960. Zur Acetessigsaure und choleskrinbildung bei experimenteller ketose. Biochem. Z. 333: 10. 72. HOTFA,S., R.HILL & I. L. CHAIKOFF. 1954. Mechanism of increased cholesterogenesis in diabetes. Its relation to carbohydrate utilization. J. Biol. Chem. 206: 835. 731 MAGNUS-LEVY, A. 1901. Untersuchungen uber die acidosus im diabetes melitas und die saurintoxication im coma diabeticum. Arch. Exptl. Pathol. Pharmakol. NaunynSchmiedelberg’s. 46: 389. 74. GREEN,D. E., J. G. DEWAN& L. F. LELOIR. 1937. CXXV. The fl-hydroxybutyric dehydrogenase of animal tissues. Biochem. J. 31: 934. 75. LEHNINGER, A. L., H. C. SUDDUTH & J. B. WISE. 1960. D( -) 0-hydroxybutyric dehydrogenase of mitochondria. J. Biol. Chem. 236: 2450. 76. SEKUZU, I., P. JURTSHUK, JR.& D. E. GREEN. 1961. On the isolation and properties of the D(-)@-hydroxybutyric dehydrogenase of beef heart mitochondria. Biochem. Biophys. Research Comm. 6: 71. 77. JURTSHUK, P., JR., I. SEKUZU & D. E . GREEN. 1961. The interaction of the D( -) 8-hydroxybutyric apoenzyme with lecithin. Biochem. Biophys. Research Comm. 6: 76. 78. CHAIKOFF, I. L. & S. SOSKIN. 1928. The utilization of acetoacetic acid by normal and diabetic dogs. Am. J. Physiol. 87: 58. 79. SAUER,F. 1961. Acetoacetate and 8-hydroxy-8-methylglutaryl coenzyme A metabolism in normal and ketotic guinea pigs. Can. J . Biol. Physiol. 39: 1635. 80. WEINHOUSE,S. 1952. Factors involved in the formation and utilization of ketone bodies. Brookhaven Symposia in Biol. KO. 5. September. U. S. Atomic Energy Commission. Pp. 201-222. 81. FELTS, J . M., I. L. CHAIKOFF & M. J. OSBORN. 1951. Insulin and the fate of lactate in the diabetic liver. J. Biol. Chem. 191:683. 82. WAKIL,S. J., D. E. GREEN,5.MII & H. R.MAHLER. 1954. Studies on the fatty acid oxidizing system of animal tissues. VI. 8-hydroxyacyl coenzyme A dehydrogenase. J. Biol. Chem. 207: 631. 83. STERN, J. R., A. DEL CAMPILLO & A. L. LEHNINGER.1955. Enzymatic racemization of 8-hydroxybutyryl-&coenzyme A and the sterospecificit,y of the enzymes of the fatty acid cycle. J. Am. Chem. SOC.77: 1073. 84. STERN,J. R. 1957. PARTIAL resolution of fl-hydroxybutyryl CoA racemase and P-hydroxybutyryl dehydrogenase. Biochim. st Biophys. Acta. 26: 661.

752

Annds New York Academy of Sciences

85. WAKIL, S. J. 1955. I)(-) 8-hydroxybutyryl CoA dehydrogenase. Biochim. et Biophys. Acta. 18: 314. 86. LYNEN,F. 1961. Biosynthesis of saturated fatty acids. Federation Proc. 20: 941. 1961. Pathways of intracellular hydrogen trans87. BOXER, G. E. & T. M. DEVLIN. port. Science. 134: 1. 88. BRESSLER, R. & S. J. WAKIL. 1961. Studies on the mechanism of fatty acid synthesis. IX. The conversion of malonyl CoA to,long chain fatty acids. J. Biol. Chem. 236: 1643.