Voet D., Voet Ju.G. Biochemistry

Подождите немного. Документ загружается.

B. The Mechanism of Dihydrolipoyl Dehydrogenase

The reaction catalyzed by dihydrolipoyl dehydrogenase

(E3) is more complex than Reactions 4 and 5 in Fig. 21-6

suggest. Vincent Massey demonstrated that oxidized dihy-

drolipoyl dehydrogenase contains a “redox-active” disulfide

bond, which in the enzyme’s reduced form has accepted an

electron pair through bond cleavage to form a dithiol:

He established this through the following observations in-

volving arsenite (which, as we saw in Section 21-2Ai, reacts

with vicinal sulfhydryl groups but not with disulfides).

1. The spectrum of oxidized dihydrolipoyl dehydroge-

nase (E) is unaffected by arsenite.

2. When NADH reacts with the oxidized enzyme in

the presence of arsenite, the resulting reduced enzyme

(EH

) binds arsenite to form an enzymatically inactive

species.

3. The oxidation state of the flavin in a flavoprotein

(flavin-containing protein) is readily established from its

characteristic UV–visible spectrum: FAD is an intense yel-

low, whereas FADH

is pale yellow. The spectrum of the

arsenite-inactivated EH

indicates that its FAD prosthetic

group is fully oxidized.

Oxidized dihydrolipoyl dehydrogenase must therefore

have a second electron acceptor in addition to FAD; ar-

senite’s known specificity suggests that the acceptor is a

disulfide. Dihydrolipoyl dehydrogenase’s amino acid se-

quence indicates that its redox-active disulfide bond

forms between its Cys 43 and Cys 48, which occur on a

highly conserved segment of the enzyme’s polypeptide

chain.

a. The X-Ray Structures of Dihydrolipoyl

Dehydrogenase and Glutathione Reductase

The X-ray structures of dihydrolipoyl dehydrogenase

from several microorganisms, mostly determined by Hol,

reveal that each ⬃470-residue subunit of these homo-

dimeric enzymes is folded into four domains, all of which

participate in forming the subunit’s catalytic center (Fig.

21-13a). An E2 PSBD (not present in Fig. 21-13) binds to

the bottom of the enzyme along its 2-fold axis in much

the same way that it binds to E1 (Fig. 21-12). The flavin is

almost completely buried in the protein, which prevents

the surrounding solution from interfering with the electron-

transfer reaction catalyzed by the enzyme (FADH

, but not

NADH or thiol, is rapidly oxidized by O

).The redox-active

disulfide, which is located on the opposite side of the

flavin ring from the nicotinamide ring (Fig. 21-13b), links

successive turns in a distorted segment of an ␣ helix (in

an undistorted helix, the C

␣

atoms of Cys 43 and Cys 48

would be too far apart to permit the disulfide bond

to form).

S S

Enzyme

SH HS

Enzyme

S S

Enzyme

_ _

E3’s catalytic mechanism has been largely determined

in analogy with that of the homologous (⬃33% identical)

but structurally more extensively characterized glutathione

reductase (GR). This nearly ubiquitous enzyme catalyzes

Section 21-2. Metabolic Sources of Acetyl-Coenzyme A 801

Figure 21-13 X-ray structure of dihydrolipoyl dehydrogenase

(E3) from the gram-negative bacterium Pseudomonas putida in

complex with FAD and NAD

ⴙ

. (a) The homodimeric enzyme is

viewed with its 2-fold axis of symmetry vertical and the E2

dodecahedron below. One subunit is gray and the other is colored

according to domain, with its FAD-binding domain (residues

1–142) light orange, its NAD

⫹

-binding domain (residues 143–268)

purple, its central domain (residues 269–337) yellow-green, and its

interface domain (residues 338–458) pink.The NAD

⫹

and FAD

in both subunits are shown in stick form with FAD C green, NAD

⫹

C cyan, N blue, O red, and P orange. (b) The active site region of

the left subunit in Part a in which the central and interface

domains have been deleted for clarity. The view is very roughly in

the same direction as in Part a.The redox-active portions of the

bound NAD

⫹

and FAD cofactors, the side chains of Cys 43 and

Cys 48 forming the redox-active disulfide bond, and the side chain

of Tyr 181 are shown in ball-and-stick form with FAD C green,

NAD

⫹

C cyan, side chain C magenta, N blue, O red, P magenta,

and S yellow. Note that the side chain of Tyr 181 is interposed

between the flavin and the nicotinamide rings. [Based on an X-ray

structure by Wim Hol, University of Washington. PDBid 1LVL.]

(a)

(b)

JWCL281_c21_789-822.qxd 6/5/10 9:53 AM Page 801

802 Chapter 21. Citric Acid Cycle

•

NAD

Stage II:

NAD

reduction

•

E–S–S–L

•

–

NADH

–

...

NAD

–

...

–

...

–

...

–

...

–

...

His 451'

Glu 456'

Tyr 181

Cys 43

Cys 48

FAD

NAD

Stage I:

Dihydrolipoamide

oxidation

•

–

Lipoamide

(L)

Dihydro-

lipoamide

(LH

)

Figure 21-14 Catalytic reaction cycle of dihydrolipoyl

dehydrogenase. The catalytic center is surrounded by protein so

that the NAD

⫹

and dihydrolipoamide binding sites are in deep

pockets.The catalytic cycle consists of six steps: (1) The oxidized

enzyme, E, which contains a redox-active disulfide bond between

Cys 43 and Cys 48, binds dihydrolipoamide, LH

, the enzyme’s

first substrate, to form an enzyme–substrate complex, E ⴢ LH

(2) A substrate S atom nucleophilically attacks S

to yield a

disulfide bond and release S

as a thiolate ion.The proton on the

substrate’s second thiol group is abstracted by His 451¿ to yield a

second thiolate ion, E–S–S–L ⴢ S

⫺

. (3) The substrate thiolate ion

nucleophilically displaces S

aided by general acid catalysis by

His 451¿ to yield the enzyme’s lipoamide product in its complex

with the stable reduced enzyme, EH

ⴢ L, in which S

forms a

charge-transfer complex with the flavin ring (dotted red line to

the red flavin ring). (4) The lipoamide product is released,

yielding EH

.The phenol side chain of Tyr 181 continues to block

access to the flavin ring of FAD, so as to prevent the enzyme’s

oxidation by O

. (5) The enzyme’s second substrate, NAD

⫹

binds to EH

to form EH

ⴢ NAD

⫹

.The phenol side chain of Tyr

181 is pushed aside by the nicotinamide ring of NAD

⫹

. (6) The

catalytic cycle is then closed by the reduction of the NAD

⫹

to reform the oxidized enzyme E and yield the enzyme’s

second product, NADH.

JWCL281_c21_789-822.qxd 3/17/10 11:33 AM Page 802

the NADPH-dependent reduction of glutathione disulfide

(GSSG) to the intracellular reducing agent glutathione

(GSH; its physiological function is discussed in Sections

23-4E and 26-4C):

Note that this reaction is analogous to that catalyzed by di-

hydrolipoyl dehydrogenase, but that these two reactions

normally occur in opposite directions; that is, dihydrolipoyl

dehydrogenase normally uses NAD

⫹

to oxidize two thiol

groups to a disulfide (Fig. 21-6), whereas GR uses NADPH

to reduce a disulfide to two thiol groups. Nevertheless, the

active sites of these two enzymes are closely superim-

posable.

The arrangement of the groups in dihydrolipoyl dehy-

drogenase’s catalytic center and its reaction sequence are

diagrammed in Fig. 21-14. The substrate-binding site is lo-

cated in the interface between the enzyme’s two subunits

in the vicinity of the redox-active disulfide. In the absence

of NAD

⫹

, the phenol side chain of its Tyr 181 covers the

nicotinamide-binding pocket so as to shield the flavin from

contact with the solution. Indeed, Fig. 21-13b shows an

enzyme–product complex of oxidized dihydrolipoyl dehy-

drogenase with NAD

⫹

in which the side chain of Tyr 181 is

interposed between the nicotinamide ring and its binding

site at the enzyme’s catalytic center. However, in the X-ray

structure of GR in complex with NADPH, determined by

Georg Schulz and Heiner Schirmer, this side chain has

moved aside such that the reduced nicotinamide ring binds

parallel to and in van der Waals contact with the flavin ring

of the fully oxidized enzyme, E (Fig. 21-15). The H

sub-

stituent to this reduced nicotinamide’s prochiral C4 atom

(that facing the flavin), the H atom that is known to be lost

in the GR reaction, lies near flavin atom N5, the position

through which electrons often enter a flavin ring on its re-

duction. This positioning is particularly significant in view

of the catalytic mechanism described below.

CHCH NHNH CC

COOCCCH CHNH NH

COO

NADPH

NADP

2 CH

COOC

CCH CHNH NH

COO

Glutathione disulfide (GSSG)

Glutathione (GSH)

(␥-

L-Glutamyl-L-cysteinylglycine)

glutathione

reductase

b. Catalytic Mechanism

Protein X-ray structures neither reveal H atom posi-

tions (except in the most highly resolved structures) nor in-

dicate pathways of electron transfer. The electron-transfer

pathway in the glutathione reductase reaction has never-

theless been inferred from the X-ray structures of a series

of stable enzymatic reaction intermediates as augmented

with a variety of enzymological data. Here we present this

mechanism in terms of the reaction catalyzed by the closely

similar dihydrolipoyl dehydrogenase (Fig. 21-14).

Dihydrolipoyl dehydrogenase has a Ping Pong mecha-

nism (Section 14-5Aa); each of its two substrates, dihy-

drolipoamide and NAD

⫹

, react in the absence of the other

(Fig. 21-6, Reactions 4 and 5). Stage I of the catalytic reac-

tion (Fig. 21-14, Steps 1–4) involves a disulfide interchange

reaction between the first substrate, dihydrolipoamide

(LH

), and the redox-active disulfide on the enzyme, a

process in which His 451¿ functions as a general acid–base

(primed residues refer to the opposite subunit). In fact, Glu

456¿, His 451¿, and Cys 43 are arranged much like the cat-

alytic triad of serine proteases (Section 15-3B), with the

Cys SH replacing the Ser OH. The importance of His 451¿

as an acid–base catalyst was established by Charles

Williams through his observation that mutationally chang-

ing it to Gln yields an enzyme that retains only ⬃0.4% of

the wild-type catalytic activity.

The thiolate anion formed from Cys 48 in Step 2 forms a

charge-transfer complex with the flavin in which S

–

(the S

atom of Cys 48) contacts the flavin ring near its 4a position

(a charge-transfer complex is a noncovalent interaction in

Section 21-2. Metabolic Sources of Acetyl-Coenzyme A 803

Figure 21-15 The complex of the flavin ring, the nicotinamide

ring, and the side chain of Cys 63 observed in the X-ray structure

of human erythrocyte glutathione reductase. The view is

perpendicular to the flavin ring.Atoms are colored according to

type with nicotinamide C cyan, flavin C green, Cys 63 C purple,

N blue, O red, and S yellow. The two planar heterocycles are

parallel and in van der Waals contact. The Cys 63 S atom

(equivalent to S

of dihydrolipoyl dehydrogenase), a member of

the redox-active disulfide, is also in van der Waals contact with

the flavin ring on the opposite side of it from the nicotinamide

ring. [Based on an X-ray structure by Andrew Karplus and

Georg Schulz, Institut für Organische Chemie und Biochemie,

Freiburg, Germany. PDBid 1GRB.]

JWCL281_c21_789-822.qxd 3/17/10 11:33 AM Page 803

which an electron pair is partially transferred from a donor,

in this case S

–

, to an acceptor, in this case the oxidized

flavin ring; the red color of this complex is indicative of the

formation of the charge-transfer complex).

Stage II of this Ping Pong reaction (Fig. 21-14, Steps 5

and 6) involves the binding and reduction of NAD

⫹

fol-

lowed by the release of NADH to regenerate the oxidized

enzyme. The path of the electrons from the reactive disul-

fide in its reduced form through the FAD to NAD

⫹

has

been elucidated by spectroscopic studies and the chemistry

of model compounds.These indicate that an electron pair is

rapidly transferred from S

–

to the flavin ring through the

transient formation of a covalent bond from S

to flavin

atom 4a (Fig. 21-16, Step 1). His 451¿ then abstracts the pro-

ton from the S

thiol to form a thiolate ion, which nucle-

ophilically attacks S

to reform the redox-active disulfide

group (Fig. 21-16, Step 2). The resulting reduced flavin an-

ion (FADH

⫺

) has but transient existence.The H atom sub-

stituent to its N5 is immediately transferred (formally as a

hydride ion) to the juxtaposed C4 atom of the nicoti-

namide ring (Fig. 21-15), yielding FAD and the reaction’s

second product, NADH, thereby completing the catalytic

cycle.Thus, the FAD appears to function more like an elec-

tron conduit between the reduced form of the redox-active

disulfide and NAD

⫹

than as a source or sink of electrons.

C. Control of Pyruvate Dehydrogenase

The PDC regulates the entrance of acetyl units derived from

carbohydrate sources into the citric acid cycle. The decar-

boxylation of pyruvate by E1 is irreversible and, since there

are no other pathways in mammals for the synthesis of

acetyl-CoA from pyruvate, it is crucial that the reaction be

carefully controlled.Two regulatory systems are employed:

1. Product inhibition by NADH and acetyl-CoA

(Fig. 21-17a).

2. Covalent modification by phosphorylation/dephos-

phorylation of the pyruvate dehydrogenase (E1) subunit

(Fig.21-17b; enzymatic regulation by covalent modification

is discussed in Section 18-3Ba).

a. Control by Product Inhibition

NADH and acetyl-CoA compete with NAD

⫹

and CoA

for binding sites on their respective enzymes.They also drive

the reversible transacetylase (E2) and dihydrolipoyl dehy-

drogenase (E3) reactions backward (Fig. 21-17a). High ra-

tios of [NADH]/[NAD

⫹

] and [acetyl-CoA]/[CoA] there-

fore maintain E2 in the acetylated form, incapable of

accepting the hydroxyethyl group from the TPP on E1.

This, in turn, ties up the TPP on the E1 subunit in its hy-

droxyethyl form, decreasing the rate of pyruvate decar-

boxylation.

b. Control by Phosphorylation/Dephosphorylation

Control by phosphorylation/dephosphorylation occurs

only in eukaryotic enzyme complexes. These complexes

contain pyruvate dehydrogenase kinase and pyruvate

804 Chapter 21. Citric Acid Cycle

FAD

–

Cys 48Cys 43

10a

Charge-transfer complex

Cys 48Cys 43

Covalent adduct

FADH

anion

–

Cys 48Cys 43

Redox-active disulfide

Figure 21-16 The reaction transferring an electron pair from

dihydrolipoyl dehydrogenase’s redox-active disulfide in its

reduced form to the enzyme’s bound flavin ring. (1) The collapse

of the charge-transfer complex between the Cys 48 thiolate ion

and the flavin ring (dashed red line) to form a covalent bond

between S

and flavin atom C4a. S

is located out of the plane

of the flavin, as Fig. 21-15 indicates. Flavin atom N5 acquires a

proton, possibly from S

, which becomes a thiolate ion. (2) The

thiolate nucleophilically attacks S

to form the redox-active

disulfide bond, thereby releasing the reduced flavin anion

FADH

⫺

JWCL281_c21_789-822.qxd 3/17/10 11:33 AM Page 804

NADH

Acetyl –CoA

NAD

pyruvate

dehydrogenase

phosphatase

pyruvate

dehydrogenase

kinase

E1–OPO

(inactive)

2–

E1– OH (active)

ADP

ATP

FAD

(b)

Activators

Acetyl-CoA

NADH

ctivators

Inhibitors

Pyruvate

ADP

(high Mg

)

Covalent modification

(a)

Product inhibition

FAD

Dihydrolipoamide

Acetyl-

dihydrolipoamide

CoA

Lipoamide

Hydroxyethyl-

TPP

Pyruvate

dehydrogenase phosphatase bound to the dihydrolipoyl

transacetylase (E2) core. The kinase inactivates the pyru-

vate dehydrogenase (E1) subunit by catalyzing the ATP-

dependent phosphorylation of a Ser residue on the more

C-terminal loop at the entrance of the active site channel

(Fig. 21-17b). Moreover, phosphorylation at only one of

E1’s two active sites inactivates the entire enzyme, thereby

further demonstrating the obligatory nature of the out of

phase coupling between these two active sites (Section 21-

2Al). Hydrolysis of this phosphoSer residue by the phos-

phatase reactivates the complex.

Pyruvate dehydrogenase kinase is activated through its

interaction with the acetylated form of E2. Consequently,

the products of the reaction, NADH and acetyl-CoA, in

addition to their direct effects on the PDC, indirectly acti-

vate pyruvate dehydrogenase kinase. The resultant phos-

phorylation inactivates the complex just as the products

themselves inhibit it.Acetyl-CoA and NADH are products

of fatty acid oxidation (Section 25-2) so that this inhibition

of PDC serves to preserve carbohydrate stores when lipid

fuels are available.

Calcium ion is an important second messenger signaling

the need for increased energy (e.g., for muscle contrac-

tion). Increasing [Ca

2⫹

] enhances pyruvate dehydrogenase

phosphatase activity, thus activating PDC.

Insulin is involved in the control of this system through

its indirect activation of pyruvate dehydrogenase phos-

phatase. Recall that insulin activates glycogen synthesis as

well by activating phosphoprotein phosphatase-1 (Section

18-3Cg). Insulin, in response to increases in blood glucose,

is now seen as promoting the synthesis of acetyl-CoA as

well as glycogen. As we shall see in Section 25-4, acetyl-

CoA is the precursor to fatty acids in addition to being the

fuel for the citric acid cycle. Various other activators and

inhibitors regulate the pyruvate dehydrogenase system

(Fig. 21-17b); in contrast to the glycogen metabolism con-

trol system (Section 18-3), however, it is unaffected by

cAMP.

Section 21-2. Metabolic Sources of Acetyl-Coenzyme A 805

Figure 21-17 Factors controlling the activity of the PDC.

(a) Product inhibition. NADH and acetyl-CoA, respectively,

compete with NAD

⫹

and CoA in Reactions 3 and 5 of the

pyruvate dehydrogenase reaction sequence. When the relative

concentrations of NADH and acetyl-CoA are high, the

reversible reactions catalyzed by E2 and E3 are driven

backward (red arrows), thereby inhibiting further formation of

acetyl-CoA. (b) Covalent modification in the eukaryotic

complex. Pyruvate dehydrogenase (E1) is inactivated by the

specific phosphorylation of one of its Ser residues in a reaction

catalyzed by pyruvate dehydrogenase kinase (right).This

phosphoryl group is hydrolyzed through the action of pyruvate

dehydrogenase phosphatase (left), thereby reactivating E1. The

activators and inhibitors of the kinase are listed on the right and

the activators of the phosphatase are listed on the left.

JWCL281_c21_789-822.qxd 6/5/10 9:53 AM Page 805

3 ENZYMES OF THE CITRIC

ACID CYCLE

In this section we discuss the reaction mechanisms of the

eight citric acid cycle enzymes. Our knowledge of these

mechanisms rests on an enormous amount of experimental

work; as we progress, we shall pause to examine some of

these experimental details. Consideration of how this cycle

is regulated and its relationship to cellular metabolism are

the subjects of the following sections.

A. Citrate Synthase

Citrate synthase (originally named citrate condensing en-

zyme) catalyzes the condensation of acetyl-CoA and ox-

aloacetate (Reaction 1 of Fig. 21-1). This initial reaction of

the citric acid cycle is the point at which carbon atoms are

“fed into the furnace” as acetyl-CoA. The citrate synthase

reaction proceeds via an Ordered sequential kinetic mech-

anism (Section 14-5Aa), with oxaloacetate adding to the

enzyme before acetyl-CoA.

The X-ray structure of the free dimeric enzyme, deter-

mined by James Remington and Robert Huber, shows it to

adopt an “open form” in which the two domains of each sub-

unit form a deep cleft that contains the oxaloacetate binding

site (Fig. 21-18a). On binding oxaloacetate, however, the

smaller domain undergoes a remarkable 18º rotation rela-

tive to the larger domain, which closes the cleft (Fig. 21-18b).

The X-ray structures of two inhibitors of citrate synthase

in ternary complex with the enzyme and oxaloacetate have

also been determined. Acetonyl-CoA, an inhibitory analog

of acetyl-CoA in the ground state, and carboxymethyl-

CoA, a proposed transition state analog (see below),

CoAS C

Acetyl-CoA

CoAS C

Acetonyl-CoA

(ground-state analog)

CoAS C

Carboxymethyl-CoA

(transition state analog)

CoAS C

Proposed enol

intermediate

806 Chapter 21. Citric Acid Cycle

(a) (b)

Figure 21-18 Conformational changes in citrate synthase.

(a) The open conformation. (b) The closed, substrate-binding

conformation. In both forms, the homodimeric protein is viewed

along its 2-fold axis and is represented by its transparent

molecular surface with its polypeptide chains drawn in worm

form colored in rainbow order from blue at their N-termini to

red at their C-termini.The reaction product citrate, which is

bound to both enzymatic forms, and coenzyme A, which is also

bound to the closed form, are shown in space-filling form with

citrate C cyan, CoA C green, N blue, O red, and P orange.The

large conformational shift between the open and closed forms

entails 18° rotations of the small domains (upper left and lower

right) relative to the large domains resulting in relative

interatomic movements of up to 15 Å. [Based on X-ray

structures by James Remington and Robert Huber, Max-Planck-

Institut für Biochemie, Martinsried, Germany. PDBids 1CTS and

2CTS.]

See Interactive Exercise 15

JWCL281_c21_789-822.qxd 10/19/10 7:42 AM Page 806

bind to the enzyme in its “closed” form, thereby identifying

the acetyl-CoA binding site. The existence of the “open”

and “closed” forms explains the enzyme’s ordered sequen-

tial kinetic behavior: The conformational change induced

by oxaloacetate binding generates the acetyl-CoA binding

site while sealing off the solvent’s access to the bound ox-

aloacetate. This is a classic example of the induced-fit

model of substrate binding (Section 10-4C). Hexokinase

exhibits similar behavior (Section 17-2Aa).

The citrate synthase reaction is a mixed aldol–Claisen es-

ter condensation, subject to general acid–base catalysis and

the intermediate participation of the enol(ate) form of

acetyl-CoA. The X-ray structure of the enzyme in ternary

complex with oxaloacetate and carboxymethyl-CoA reveals

that three of its ionizable side chains are properly oriented

to play catalytic roles: His 274,Asp 375, and His 320.The N1

atoms in both of these His side chains are hydrogen bonded

to two backbone NH groups indicating that these N1 atoms

are not protonated. The participation of His 274, Asp 375,

and His 320 has been confirmed by kinetic studies of mutant

enzymes generated by site-directed mutagenesis. The fol-

lowing three-step mechanism, formulated mainly by Rem-

ington, takes into account these observations (Fig. 21-19):

1. The enolate of acetyl-CoA is generated in the rate-

limiting step of the reaction, with the catalytic participation

of Asp 375 acting as a base to remove a proton from the

methyl group and His 274, in its neutral form, H-bonding

the enolate oxygen (whose protonated carbonyl form is

normally far more acidic than a neutral His side chain).

Much as we discussed for a similar step in the triose phos-

phate isomerase reaction (Section 17-2Eb), the pK of the

protonated form of the substrate thioester carbonyl oxy-

gen increases to ⬃14 on enolization, which approximates

that of neutral His 274. Hence, it has been proposed that this

step is facilitated through the formation of a low-barrier

hydrogen bond (which, it will be recalled, is a particularly

strong form of hydrogen bonding interaction in which the

hydrogen atom is more or less equally shared between the

“donor” and “acceptor” atoms; Section 15-3Dd). However,

whether or not this low-barrier hydrogen bond actually

forms is a subject of controversy. The presence of the

thioester bond to CoA facilitates the enolization; enoliza-

tion of acetate alone would require the generation of a

much more highly charged and therefore less stable inter-

mediate.

2. Citryl-CoA is formed in a second acid–base-catalyzed

reaction step in which the acetyl-CoA enolate nucleophili-

cally attacks oxaloacetate while His 320, also in its neutral

form, donates a proton to oxaloacetate’s carbonyl group.

The citryl-CoA remains bound to the enzyme.

3. Citryl-CoA is hydrolyzed to citrate and CoA, with

His 320, now in the anionic form, abstracting a proton from

water as it attacks the carbonyl group of the citryl-CoA

and the acidic form of Asp 375 donating a proton to the

CoA as it leaves. This hydrolysis provides the reaction’s

thermodynamic driving force (⌬G°¿ ⫽ –31.5 kJ ⴢ mol

⫺1

We shall see presently why the reaction requires such a

large, seemingly wasteful, release of free energy.

Section 21-3. Enzymes of the Citric Acid Cycle 807

–

OOC

COO

–

SCoA

His

274

Acetyl-CoA

Possible low-barrier

hydrogen bond

C+

C Asp 375

–

C Asp 375

His 320

Oxaloacetate

(S)-Citryl-CoA

–

OOC

SCoAH

C Asp 375

COO

–

His 320

C Asp 375C Asp 375

–

OOC

–

OOC

SCoA

CoASH

C CH

His

–

320

Citrate

pro S-arm

COO

–

OOC

–

OOC H

C CH

His

274

His

274

pro R-arm

Figure 21-19 Mechanism and stereochemistry of the citrate

synthase reaction. His 274 and His 320 in their neutral forms and

Asp 375 have been implicated as general acid–base catalysts.The

overall reaction’s rate-limiting step is the formation of the

acetyl-CoA enolate, which may be stabilized by a low-barrier

hydrogen bond to His 274.The acetyl-CoA enolate then

nucleophilically attacks the si face of oxaloacetate’s carbonyl

carbon.The resulting intermediate, (S)-citryl-CoA, is hydrolyzed

to yield citrate and CoA. [Mostly after Remington, S.J., Curr.

Opin. Struct. Biol. 2, 732 (1992)].

JWCL281_c21_789-822.qxd 6/5/10 9:53 AM Page 807

Enzyme-catalyzed reactions, as we have previously

noted, are stereospecific. The aldol–Claisen condensation

that occurs here involves attack of the acetyl-CoA enolate

exclusively at the si face of oxaloacetate’s carbonyl carbon

atom, thereby yielding (S)-citryl-CoA (chirality nomencla-

ture is presented in Section 4-2C).The acetyl group of acetyl-

CoA thereby only forms citrate’s pro-S carboxymethyl group.

B. Aconitase

Aconitase catalyzes the reversible isomerization of citrate

and isocitrate with cis-aconitate as an intermediate (Reac-

tion 2 of Fig. 21-1). Although citrate has a plane of symme-

try and is therefore not optically active, it is nevertheless

prochiral; aconitase can distinguish between citrate’s pro-R

and pro-S carboxymethyl groups.

A combination of X-ray crystallography and site-

directed mutagenesis by Helmut Beinert and David Stout

identified several amino acid residues that participate in

catalysis. Formation of the cis-aconitate intermediate in-

volves a dehydration in which Ser 642 alkoxide, acting as a

general base, abstracts the pro-R proton at C2 of citrate’s

pro-R carboxymethyl group (Fig. 21-20, top). [The O

␥

Ser 642 occupies a sort of oxyanion hole (Section 15-3Db)

that apparently stabilizes its otherwise highly basic alkox-

ide form. The carbanion formed in the transition state of

this reaction is not shown.] This is followed by the loss of

the OH group at C3 in a trans elimination of H

O to form

the cis-aconitate intermediate. The latter reaction step is

facilitated through general acid catalysis by His 101, whose

imidazolium group is polarized through ion pairing to the

side chain of Asp 100. This model is corroborated by the

observations that the mutagenesis of Asp 100, His 101, or

Ser 642 results in a 10

- to 10

-fold reduction in aconitase’s

catalytic activity without greatly affecting its substrate

binding affinity.

808 Chapter 21. Citric Acid Cycle

Asp His100 101

OSer 642 :

COO

OOC

pro-S

pro-R

Arg 580

COO

OOC

Arg

Ser 642

580

Citrate cis-Aconitate intermediate

[4Fe-4S] cluster

Asp His100 101

COO

Asp His100 101

OSer 642 :

COO

OOC

Arg 580

(2R,3S)-Isocitrate

COO

OOC

Arg

Ser 642

580

Asp

His

100

101

COO

180⬚ flip

Figure 21-20 Mechanism and stereochemistry of the aconitase

reaction. Fe

of the enzyme’s [4Fe–4S] cluster coordinates the

citrate hydroxyl and central carboxyl groups;Arg 580 forms a

salt bridge with the pro-S carboxyl group; Ser 642, in its alkoxide

form, acts as a general base; and the Asp 100-polarized His 101

acts as a general acid in the elimination of water to form

cis-aconitate. Note the unusual 180° flip that cis-aconitate

apparently undergoes, possibly while remaining bound to the

active site. Thus rehydration takes place on the opposite face of

the substrate from which dehydration occurred, yielding

(2R,3S)-isocitrate.

JWCL281_c21_789-822.qxd 6/5/10 9:53 AM Page 808

Aconitase contains a covalently bound [4Fe–4S]

iron–sulfur cluster, which is required for catalytic activity

(the properties of iron–sulfur clusters are discussed in Sec-

tion 22-2Ca). A specific Fe(II) atom in this cluster, the so-

called Fe

atom, is thought to coordinate the OH group of

the substrate so as to facilitate its elimination. Iron–sulfur

clusters are almost always associated with redox processes

although, intriguingly, not in the case of aconitase. It has

been postulated that the electronic properties of the

[4Fe–4S] cluster permit the Fe

atom to expand its coordi-

nation shell from the four ligands observed in the X-ray

structure of the free enzyme (three S

2⫺

ions and a OH

⫺

ion)

to the six octahedrally arranged ligands observed in the

enzyme–substrate complex. A single metal ion, such as

2⫹

or Cu

2⫹

, is unable to do so, and hence would require

that some of its ligands be displaced on binding substrate.

The second stage of the aconitase reaction is rehydra-

tion of cis-aconitate’s double bond to form isocitrate (Fig.

21-20, bottom). The nonenzymatic addition of H

O across

the double bond of cis-aconitate would yield four stereoiso-

mers.Aconitase, however, catalyzes the stereospecific trans

addition of OH

⫺

and H

⫹

across the double bond to form

only (2R,3S)-isocitrate in the forward reaction and citrate

in the reverse reaction.

Although citrate’s OH group is lost to the solvent in the

aconitase reaction, the abstracted H

⫹

is retained by Ser 642.

Remarkably, it adds to the opposite faces of cis-aconitate’s

double bond in forming citrate and isocitrate. The cis-

aconitate must expose a different face to the sequestered

⫹

for the formation of isocitrate. This is thought to occur

by a 180º flip of the intermediate on the enzyme’s surface

while maintaining its association with Arg 580 or else by

dissociation from the enzyme and replacement by another

cis-aconitate in the “flipped” orientation.

a. Fluorocitrate Inhibits Aconitase

Fluoroacetate, one of the most toxic small molecules

known (LD

⫽ 0.2 mg ⴢ kg

⫺1

of body weight in rats), oc-

curs in the leaves of certain African, Australian, and South

American poisonous plants. Interestingly, fluoroacetate it-

self has little toxic effect on cells; instead, cells enzymati-

cally convert it first to fluoroacetyl-CoA and then to

(2R,3R)-2-fluorocitrate, which specifically inhibits aconi-

tase (see Problem 9):

It is not clear,however,that the inhibition of aconitase fully

accounts for fluorocitrate’s high toxicity. Indeed, fluoro-

citrate also inhibits the transport of citrate across the mito-

chondrial membrane.

C. NAD

ⴙ

-Dependent Isocitrate Dehydrogenase

Isocitrate dehydrogenase catalyzes the oxidative decarboxy-

lation of isocitrate to a-ketoglutarate to produce the citric

acid cycle’s first CO

and NADH (Reaction 3 of Fig. 21-1).

Mammalian tissues contain two isoforms of this enzyme.

Although they both catalyze the same reaction, one iso-

form participates in the citric acid cycle, is located entirely

in the mitochondrion, and utilizes NAD

⫹

as a cofactor.The

other isoform occurs in both the mitochondrion and the

cytosol,utilizes NADP

⫹

as a cofactor,and generates NADPH

for use in reductive biosynthesis.

NAD

⫹

-dependent isocitrate dehydrogenase, which re-

quires an Mn

2⫹

or Mg

2⫹

cofactor, catalyzes the oxidation of

a secondary alcohol (isocitrate) to a ketone (oxalosuccinate)

followed by the decarboxylation of the carboxyl group ␤ to

the ketone (Fig. 21-21). In this sequence, the keto group ␤ to

the carboxyl group facilitates the decarboxylation by acting

as an electron sink.The oxidation occurs with the stereospe-

cific reduction of NAD

⫹

at its re face (A-side addition; Sec-

tion 13-2Aa). Mn

2⫹

coordinates the newly formed carbonyl

group so as to polarize its electronic charge.

COO

Fluoroacetate

Fluoroacetyl-CoA

(2R,3R)-2-

Fluorocitrate

acetate

thiokinase

citrate

synthase

COO

CHO

SCoA

H F

Section 21-3. Enzymes of the Citric Acid Cycle 809

...

COO

–

Isocitrate

+ NADHNAD

–

C C

–

HHO

COO

–

Oxalosuccinate

–

C C

–

...

COO

–

COO

–

␣-Ketoglutarate

–

CHH

Figure 21-21 Probable reaction mechanism of isocitrate dehydrogenase. Oxalosuccinate is

shown in brackets because it does not dissociate from the enzyme.

JWCL281_c21_789-822.qxd 3/17/10 11:34 AM Page 809

Although the intermediate formation of oxalosuccinate

is a logical chemical prediction, evidence for it has been

difficult to obtain because it has only a transient existence

in reactions catalyzed by the wild-type enzyme. However,

an enzymatic reaction rate can be slowed by the mutation

of particular catalytically important residues, resulting in

the accumulation of specific intermediates.Thus, when Lys

230, which facilitates the decarboxylation of the oxalosuc-

cinate intermediate (Fig. 21-21) is mutated to Met in

NADP

⫹

-dependent isocitrate dehydrogenase, the oxalo-

succinate intermediate accumulates. This accumulated in-

termediate was directly visualized in the X-ray structure

of the mutant enzyme, in the presence of a steady-state

flow of substrate, through the use of fast X-ray intensity

measurements.

D. ␣-Ketoglutarate Dehydrogenase

a-Ketoglutarate dehydrogenase catalyzes the oxidative de-

carboxylation of an a-keto acid (a-ketoglutarate), releasing

the citric acid cycle’s second CO

and NADH (Reaction 4

of Fig.21-1).The overall reaction, which chemically resem-

bles that catalyzed by the PDC (Fig. 21-6), is mediated

by a homologous multienzyme complex consisting of

␣-ketoglutarate dehydrogenase (E1o),dihydrolipoyl trans-

succinylase (E2o), and dihydrolipoyl dehydrogenase (E3)

in which the E3 subunits are identical to those in the PDC

(Section 21-2A).

Individual reactions catalyzed by the complex occur by

mechanisms identical to those of the pyruvate dehydroge-

nase reaction (Section 21-2A), the product likewise being a

“high-energy” thioester, in this case succinyl-CoA. There

are no covalent modification enzymes in the ␣-ketoglu-

tarate dehydrogenase complex, however.

E. Succinyl-CoA Synthetase

Succinyl-CoA synthetase (also called succinate thiokinase)

hydrolyzes the “high-energy” compound succinyl-CoA with

the coupled synthesis of a “high-energy” nucleoside triphos-

phate (Reaction 5 of Fig. 21-1). (Note: Enzyme names can

refer to either the forward or the reverse reaction; in this

case, succinyl-CoA synthetase and succinate thiokinase re-

fer to the reverse reaction.) GTP is synthesized from

GDP ⫹ P

by the mammalian enzyme; plant and bacterial

enzymes utilize ADP ⫹ P

to form ATP. These reactions

are nevertheless equivalent since ATP and GTP are rapidly

interconverted through the action of nucleoside diphos-

phate kinase (Section 16-4C):

a. The Succinyl-CoA Thioester Bond Energy Is

Preserved through the Formation of a Series of

“High-Energy” Phosphates

How does succinyl-CoA synthetase couple the exergonic

hydrolysis of succinyl-CoA (⌬G°¿ ⫽⫺32.6 kJ ⴢ mol

⫺1

)to

the endergonic formation of a nucleoside triphosphate

(⌬G°¿ ⫽ 30.5 kJ ⴢ mol

⫺1

)? This question was answered

GTP ⫹ ADP Δ GDP ⫹ ATP

¢G°¿ ⫽ 0

through the creative use of isotope tracers. In the absence

of succinyl-CoA, the spinach enzyme (which utilizes ade-

nine nucleotides) catalyzes the transfer of ATP’s ␥ phos-

phoryl group to ADP as detected by

C-labeling ADP and

observing the label to appear in ATP. Such an isotope ex-

change reaction (Section 14-5D) suggests the participation

of a phosphoryl–enzyme intermediate that mediates the

reaction sequence:

Indeed, this information led to the isolation of a kineti-

cally active phosphoryl–enzyme in which the phosphoryl

group is covalently linked to the N3 position of a His

residue.

When the succinyl-CoA synthetase reaction, which is

freely reversible, is run in the direction of succinyl-CoA

synthesis (opposite to its direction in the citric acid cycle)

using [

O]succinate as a substrate,

O is transferred from

succinate to phosphate. Evidently, succinyl phosphate, a

“high-energy” mixed anhydride, is transiently formed dur-

ing the reaction.

These observations suggest the following three-step

sequence for the mammalian succinyl-CoA synthetase

reaction (Fig. 21-22):

1. Succinyl-CoA reacts with P

to form succinyl phos-

phate and CoA (accounting for the

O-exchange reaction).

2. Succinyl phosphate’s phosphoryl group is transferred

to an enzyme His residue, releasing succinate (accounting

for the 3-phosphoHis residue).

3. The phosphoryl group on the enzyme is transferred

to GDP, forming GTP (accounting for the nucleoside

diphosphate exchange reaction).

Note how in each of these steps the “high-energy” succinyl-

CoA’s free energy of hydrolysis is conserved through the

successive formation of “high-energy” compounds: First

succinyl phosphate, then a 3-phosphoHis residue, and

finally GTP. The process is reminiscent of passing a hot

potato.

b. A Pause for Perspective

Up to this point in the cycle, one acetyl equivalent has

been completely oxidized to two CO

. Two NADHs and

one GTP (in equilibrium with ATP) have also been gener-

ated. In order to complete the cycle, succinate must be con-

verted back to oxaloacetate.This is accomplished by the cy-

cle’s remaining three reactions.

PP+

A +

Step 1

P P P E

ATP

P P E

ADP ATP*

ADP*

Phosphoryl-

enzyme

A +

Step 2

P P E

A P P E

810 Chapter 21. Citric Acid Cycle

JWCL281_c21_789-822.qxd 7/2/10 9:06 AM Page 810