Scheffler Immo E. Mitochondria

Подождите немного. Документ загружается.

THE ELECTRON TRANSPORT CHAIN 191

details on the structure. It provides answers to many questions, and it suggests

that the evolutionary history may be an explanation for the apparently exces-

sive number of Fe – S centers required for electrons transfer from FMN to ubi-

quinone. Alternatively, there may be an as yet undiscovered rationale linked to

the mechanism of proton translocation by this enzyme.

Understanding the structure and function of this complex has been one of

the challenges for which signiﬁ cant progress can be reported. Two major

unsolved problems remain: (1) to understand the path of the protons pumped

by the complex and (2) to elucidate the mechanism by which electron trans-

port taking place exclusively in the peripheral subcomplex is coupled to proton

translocation that must involve the integral membrane proteins. Another chal-

lenge is to understand the assembly of this complex from 38 subunits imported

from the cytosol and seven subunits made in the mitochondrial matrix. What

regulates the expression of the widely dispersed (unlinked) nuclear genes?

What regulates the level of assembly of this complex? Is there a unique assem-

bly pathway; that is, is it possible to identify assembly intermediates? Are the

ND subunits inserted independently into the inner mitochondrial membrane,

or is their synthesis and membrane integration coupled intimately to the

assembly of subunits imported from the cytosol? Some answers to these ques-

tions have been obtained in recent years.

Figure 5.10 (A) Crystal structure of the peripheral subcomplex of respiratory complex

I (NADH – ubiquinone oxidoreductase) from Thermus thermophilus at 3.3 - Å resolution

(74, 76) . The subunits are labeled following the nomenclature for bacteria. (B) Location

and distances between the 9 Fe – S clusters in this subdomain. (Reproduced from refer-

ence 76 with permission.) See color plates.

192 MITOCHONDRIAL ELECTRON TRANSFER

The ﬁ rst studies to address such questions were reported for Neurospora

crassa (45) . Effectively, gene knockouts can be achieved by site - speciﬁ c inte-

gration of a hygromycin resistance marker, and the consequences of a missing

subunit on assembly can be studied (47, 77) . The mutant nuo21 is of special

interest because in the absence of this peptide the peripheral complex is

assembled completely, while assembly of the membrane arm is only partial.

With this mutant as starting material, the peripheral subcomplex can be pre-

pared and puriﬁ ed conveniently. It can oxidize NADH when ferricyanide is

used as an artiﬁ cial electron acceptor, and it even has a low NADH: ubiqui-

none oxidoreductase activity. The latter is insensitive to piericidin A and thus

believed to be due to ubiquinone binding to a site on the peripheral arm which

is not normally exposed in an intact complex I. Neurospora mutants defective

in a nuclear encoded complex I peptide have a reduced growth rate, but their

mitochondria are not signiﬁ cantly respiration deﬁ cient. The presence of cyto-

chromes and inhibition of respiration by antimycin and cyanide suggests a

functioning electron transport chain downstream from ubiquinone, and the

absence of a piericidin - sensitive complex I is overcome by an alternative

NADH - ubiquinone oxidoreductase.

More recently the assembly of the mammalian complex has been studied

in tissue culture systems. In Schefﬂ er ’ s laboratory (28, 39, 40) Chinese hamster

mutants cells missing either the MWFE or the ESSS subunits in the membrane

arm were shown to lack complex I by Blue Native gel electrophoresis. The

mutants can be complemented with the respective cDNAs; moreover, the

expression of these cDNAs can be regulated from an inducible promoter.

When induced, the MWFE and ESSS subunits reach steady - state levels within

24 hours, but the appearance of the > 900 - kDa complex I on Blue Native gels

is delayed by another ∼ 24 hours, and full respiration measured by an oxygen

electrode requires even more time (28; Potluri, unpublished). Some other

subunits are missing when MWFE or ESSS are absent, presumably because

they are unstable and rapidly turned over in the absence of assembly (their

mRNAs are present at normal levels).

Other laboratories have examined complex I assembly and assembly inter-

mediates in human cells from patients with isolated complex I deﬁ ciencies (see

Chapter 7 ) (29, 30, 78 – 80) . Such patients are alive because there is a residual

activity; and in all cases examined so far, it appears that the reduced activity

is due to a reduced steady - state level of intact complex I. The interpretation

is that assembly is slowed down due to mutations in proteins encoded by

nuclear genes (most of those are in the peripheral complex). Two precaution-

ary statements may be in order. First, the assembly intermediates in such

patient/mutant cell lines may not be exactly the same as those formed under

normal conditions. Second, the analysis requires the use of Blue - Native gels

and the solubilization of mitochondrial membranes by selected detergents.

Mutations may introduce weakened associations in the complex that lead to

dissociation into subcomplexes under the conditions of the experiment, but

not in vivo .

THE ELECTRON TRANSPORT CHAIN 193

Over the years the analysis of this complex and of its activity in mitochon-

dria has been aided by the discovery of a variety of highly speciﬁ c, naturally

occurring inhibitors (Figure 5.11 ). The best known of these is rotenone, used

as a ﬁ sh poison; another is the antibiotic piericidin A. Some of these are being

developed as potential insecticides and for various other pharmacological

purposes (81) . As discussed elsewhere in this volume, the “ designer drug ”

MPTP (1 - methyl - 4 - phenyl 1,2,3,6 tetrahydropyridine) can be converted to the

active metabolite 1 - methyl - 4 - phenyl prydinium (MPP

) by monoamine oxidase

and transported into dopaminergic neurons. At sufﬁ ciently high concentra-

tions (mM) it is a speciﬁ c inhibitor of complex I. This mechanism is thought

to be responsible for the death of neurons in the substantia nigra and hence

symptoms of Parkinsonism in humans and experimental primates (82, 83) . A

large and diverse series of speciﬁ c inhibitors of complex I, some occurring

naturally and some synthetics, have been discussed and investigated systemati-

cally by Friedrich et al. (84) . On the basis of detailed kinetic analyses, these

authors were able to divide the inhibitors into two classes: class I inhibitors

(piericidin A, annonin VI, phenalamid A2, aurachins A and B, thiangazole, and

fenpyroximate (synthetic)) inhibit complex I in a partially competitive manner

with respect to ubiquinone; class II inhibitors (rotenone, phenoxan, aureothin,

and benzimidazole (synthetic)) act in a noncompetitive manner. None of the

inhibitors inhibit the NADH - ferricyanide reductase — that is, the peripheral

membrane arm of the complex extending into the matrix. Instead, all inhibitors

appear to arrest the transfer of electrons from the high - potential iron – sulfur

cluster (N - 2) to ubiquinone. Since the Fe – S clusters have not been completely

characterized in this complex, it is not possible to be more speciﬁ c. However,

photoafﬁ nity labeling studies with a rotenone derivative have suggested that

this binding site is close to the peptide encoded by the mitochondrial ND1

gene. A more recent investigation of rotenone binding raising several still

controversial issues related to conformational changes within the complex has

been published by Grivennikova et al. (85) . Speciﬁ cally, these authors propose

that one rotenone - speciﬁ c site in complex I affects NADH oxidation by ubi-

quinone, while a second site is active in the ubiquinol – NAD

reductase reac-

tion. This apparent violation of the principle of microscopic reversibility clearly

deserves further clariﬁ cation.

Detailed studies on the mechanism of inhibition by MPP+ analogues by

Singer and colleagues (86) have led to the interpretation that there are two

MPP+ - binding sites, both of which must be occupied for complete inhibition.

A “ hydrophilic site ” is deﬁ ned by accessibility to relatively hydrophilic

analogues, while a “ hydrophobic site ” may be less exposed to the aqueous

environment. This interpretation is consistent with ﬁ ndings that the complex

I can contain two ubisemiquinones. Miyoshi and colleagues have reexamined

this question with the help of a new series of N - methyl - pyridinium and N -

methyl - quinolinium analogues, with the goal of ﬁ nding a more potent inhibitor

effective at lower concentrations and, perhaps more signiﬁ cantly, the aim of

ﬁ nding an inhibitor which exhibits selectivity for only one of the two sites (87) .

194 MITOCHONDRIAL ELECTRON TRANSFER

Figure 5.11 A and B: Structures of speciﬁ c inhibitors of complex I frequently used in

the study of this complex (modiﬁ ed from references 84 and 87 ). The compounds shown

in part B are N - methyl - pyridinium analogues in which various side chains can be sub-

stituted at positions 2, 3, or 4.

MTPT -------> MPP

Rotenone

Annonin VI

Phenoxan

Piericidin A

OCH

Amytal

THE ELECTRON TRANSPORT CHAIN 195

The results clearly support the two - site model, and signiﬁ cant progress toward

achieving the second aim was also made. The presence of two ubiquinone

binding sites is in good agreement with the acceptance of ubisemiquinone as

an obligatory intermediate in the two - electron transfer from NADH to ubi-

quinone (88, 89) .

5.2.5 Complex II

Complex II (succinate:ubiquinone oxidoreductase) consists of only four pep-

tides and is thus the simplest of all the complexes of the ETC. The two largest

peptides constitute the peripheral portion of the complex and function as the

enzyme succinate dehydrogenase in the Krebs cycle. They are associated with

the membrane through two integral membrane proteins, also referred to as the

“ anchor ” proteins (see Figure 5.12 ). Electrons from the oxidation of succinate

to fumarate are channeled through this complex to ubiquinone. Thus, complex

II links the Krebs cycle directly to the ETC (90 – 92) . A highly conserved and

similar complex is also found as part of the electron transport system in bacte-

rial membranes. The corresponding genes were ﬁ rst cloned in bacteria, and the

N-Methyl-pyridinium

N-Methyl-quinolinium

(CH

)

Figure 5.11 (Continued)

196 MITOCHONDRIAL ELECTRON TRANSFER

bacterial systems ( Bacillus subtilis and Escherichia coli ) have been very useful

model systems in the study of the structure, function, and assembly of this

complex. Excellent authoritative reviews have been written by several authors

(93 – 95) .

Cofactors and metal ions make up the additional components of complex

II. A ﬂ avin is linked covalently to the largest peptide (70 kDa), yielding the

ﬂ avoprotein subunit (Fp). Singer and his colleagues established the chemical

linkage in the early 1970s, and they also identiﬁ ed a histidine as the point of

attachment (96 – 98) . The timing and mechanism of this conjugation reaction

was an open question until recently when the Fp gene (SDH1) was cloned in

yeast. Conversion of His90 to Ser90 by site - directed mutagenesis prevented

covalent attachment of FAD, but the assembly of the complex II in mitochon-

dria was not affected (99 – 101) . It was clearly shown that ﬂ avin attachment to

the normal Fp peptide occurred in the mitochondrial matrix, after the removal

of the targeting sequence (99) . Additional experiments were interpreted to

indicate that the enzyme responsible for this modiﬁ cation has to recognize a

folded Fp peptide as one of its substrates, rather than a short peptide sequence

surrounding the targeted histidine side chain. Nevertheless, this region of the

peptide is highly conserved in evolution (see reference 102 for a review). When

complex II was assembled in yeast with a noncovalently bound ﬂ avin, the

complex was active only as a fumarate reductase, not as a succinate dehydro-

genase (101) . This result supports the idea that the covalent attachment of the

ﬂ avin raises its midpoint potential to the level required for the oxidation of

succinate (103) .

The Fp subunit is intimately associated with the iron – protein subunit (Ip),

made up of a peptide of 27 kDa containing three nonheme Fe – S centers:

[2Fe – 2S], [3Fe – 4S], [4Fe – 4S] (104) . Each has a characteristic redox potential,

and together these centers serve in the electron transport through the subunit

(93, 98, 105) . The Ip peptides from all species examined to date, and even from

prokaryotes, have three highly conserved cysteine clusters (CI, CII, CIII)

within the peptide. The ﬁ rst cluster is used to make the [2Fe – 2S] center, and

it resembles the [2Fe – 2S] cluster also found in plant ferredoxins. The [4Fe – 4S]

center is made from the ﬁ rst, second, and third cysteines of cluster II and the

third cysteine of cluster III, while the [3Fe – 4S] center is made from the ﬁ rst

and second cysteines of cluster III and the fourth cysteine of cluster II. These

deductions were made from EPR studies made with the help of site - directed

mutagenesis of bacterial sdh and frd iron – proteins and were subsequently

conﬁ rmed in the crystal structures. Thus, the assembly of the clusters depends

on the secondary and tertiary structure of the Ip subunit rather than the local

peptide sequence. It should be noted here that the Fe – S clusters are assembled

ﬁ rst by an independent mechanism (see Section 6.8) and then delivered to

their ﬁ nal destination in Ip and various other proteins.

The Fp – Ip complex (SDH) can be dissociated from an isolated complex II

by treatment with chaotropic ions as ﬁ rst shown by Hateﬁ and colleagues (92) .

THE ELECTRON TRANSPORT CHAIN 197

It functions as a succinate dehydrogenase, when an artiﬁ cial electron acceptor

such as ferricyanide or tetrazolium is included in the assay, but cannot interact

directly with ubiquinones. To reconstitute ubiquinone reduction, the SDH

complex must be combined with the two integral membrane proteins of

complex II, C

II - 3

and C

II - 4

, also referred to as QPs - 1 and QPs - 2 (or CybL and

CybS). These “ anchor ” proteins are small peptides of molecular weight 15 and

12 – 13 kDa, respectively (in mammals), and sequence analysis suggested that

each has three transmembrane segments, along with an N - terminal domain

extending into the matrix for interaction with the SDH enzyme (106, 107) .

Solubilization and reconstitution experiments, sequence analysis and

hydropathy plots, and some earlier radiolabeling experiments by Merli et al.

(108) suggested a domain structure for complex II that has been largely con-

ﬁ rmed by X - ray crystallography. The ﬁ rst structure to be solved was the struc-

ture of the E. coli fumarate reductase (109) . For good reasons (see below) the

overall structure served as an excellent model for the structure of complex II

in prokaryotes as well as eukaryotes. Since then the E. coli succinate dehydro-

genase complex has also been solved (110 – 113) , revealing in great detail the

path of electrons from succinate via the ﬂ avin, three Fe – S cluster to ubiqui-

none (see Figure 5.13 ). Finally, a high - resolution crystal structure of a mam-

malian complex II (porcine heart) at 2.4 - Å resolution was published by Sun

Figure 5.12 Schematic representation of complex II, succinate – ubiquinone oxidore-

ductase. Two integral membrane proteins serve to anchor the iron – protein (Ip) and the

ﬂ avoprotein (Fp) of SDH.

FAD

[2Fe-2S]

[4Fe-4S]

[3Fe-4S]

Succinate Fumarate

cytochrome

MATRIX

198 MITOCHONDRIAL ELECTRON TRANSFER

et al. (114) . This structural determination has permitted a detailed analysis of

binding sites for phospholipids, ubiquinone (two sites), and various inhibitors

of the enzyme. A comparison with the E. coli enzyme has revealed many con-

served structural components as well as informative differences.

There is some analogy with complex I, but on a much simpler scale. The

peripheral subcomplex (Fp (SdhA), Ip (SdhB)) has the substrate binding site

(succinate), the hydrogen acceptor (FAD), and 3 [Fe – S] centers for the con-

ductance of electrons to the second and relatively simple membrane subcom-

plex consisting of the two anchor proteins. This membrane complex also

contains a b - type cytochrome and the binding site(s) for ubiquinone. The

involvement of the b - type cytochrome with electron transport to ubiquinone

is still under investigation; it is attached to both anchor proteins. The C

II - 3

II - 4

peptides have highly conserved histidines linking the heme b to the simple

membrane complex (102) .

The reaction catalyzed by complex II is

succinate Q fumarate QH

+→ +

In most eukaryotic cells the reaction proceeds from left to right, not only

because of the relative concentrations of substrates and products, but also

because the enzyme has diode - like properties that prevent electron ﬂ ow in the

opposite direction (115, 116) . Understanding the biophysical basis for this

behavior remains one of the challenges.

Speciﬁ c inhibitors continue to play an important role in biochemical studies

of electron transport and the analysis of mitochondrial functions. For complex

II the most useful inhibitor has been the substrate analogue malonate, which

binds competitively and speciﬁ cally at the active site on the Fp subunit.

There are examples of organisms or conditions in which the reduction of

fumarate becomes a physiologically relevant reaction. Facultative bacteria

Figure 5.13 X - ray structure of succinate dehydrogenase (complex II) from E. coli .

(A) The trimer viewed parallel to the membrane. (B) Perpendicular view of the trimer

viewed from the cytoplasm. (C) View of a monomer showing the redox centers and

the distances between them. (From reference 113 with permission.)

125 Å

60 Å

70 Å

cytoplasm

periplasm

SdhC

SdhA

OAA

4.6(2.5)

16.0(11.1)

12.4(9.3)

11.9(9.1)

[3Fe-4S]

[4Fe-4S]

[2Fe-2S]

18.5

(11.4)

11.0

(7.6)

Heme b

6.5(9.8)

FAD

SdhB

SdhD

120

THE ELECTRON TRANSPORT CHAIN 199

such as E. coli can proliferate under aerobic conditions and with a minimal

supply of glucose when complete combustion to carbon dioxide and the most

efﬁ cient exploitation of available energy resources is achieved by operating

the Krebs cycle and oxidative phosphorylation. The interconversion of essen-

tial metabolites also requires the Krebs cycle. E. coli and other bacteria can

also grow under anaerobic conditions when energy metabolism has to be

altered drastically. Under such conditions the reduction of fumarate to succi-

nate becomes an essential reaction. However, the reverse reaction is not

achieved by the same SDH enzyme induced under aerobic conditions, but is

instead catalyzed by a completely distinct complex named fumarate reductase,

FRD. The two complexes are similar with respect to protein composition (Fp,

Ip, and anchor proteins), but these peptides are encoded by two distinct

operons. SdhA,

sdhB, sdhC, and sdhD are represented by four cistrons on a

polycistronic transcript for succinate:ubiquinone oxidoreductase, and frdA,

frdB, frdC, and frdD are encoded by four cistrons for fumarate:ubiquinone

oxidoreductase). In B. subtilis there is only one anchor protein, but it is almost

twice the size of the two E. coli peptides and may represent a gene fusion. The

regulation of the expression of these operons by oxygen and glucose has been

studied extensively in bacteria (94, 117, 118) . It remains to understand in some

detail how some changes in the primary sequence of these peptides can alter

the microenvironment of the redox couples (Fe – S, FAD) to favor reactions in

one direction or the other. Another noteworthy aspect of the SDH and FRD

activities in bacteria is that ubiquinone serves as the electron acceptor from

SDH under aerobic conditions, while reduced menaquinone serves as the

electron donor via the FRD in the reduction of fumarate to succinate under

anaerobic conditions. For a more detailed discussion of issues, the reader is

referred to the authoritative review of Ackrell et al. (93) , in which both pro-

karyotic and eukaryotic enzymes are discussed.

Another interesting situation arises for eukaryotic organisms, which spend

part of their developmental cycle under relatively anaerobic conditions and

another part of their life under aerobic conditions. For example, the sheep

nematode Haemonchus contortus switches to a fermentative, predominantly

anaerobic metabolism of the parasitic (adult) stages from the aerobic metabo-

lism found in the free - living larvae (119) . It appears that isozymes of the

iron – protein of complex II are differentially expressed in larvae and adults,

and it has been argued that a switch in isozymes is responsible for the change

in complex II from functioning predominantly as a fumarate reductase to the

conventional succinate:ubiquinone oxidoreductase under aerobic conditions.

In a related species, Ascaris suum , a similar switch and the existence of iso-

zymes for the ﬂ avoprotein have been demonstrated (120) . The existence of

multiple genes (isozymes) for complex II may be a general observation for

The genes for complex II have been designated in different ways by different authors for various

organisms. The designation SDH1, SDH2, SDH3, and SDH4 is proposed for the four genes encod-

ing the subunits in the order of decreasing molecular mass of the peptides in yeast. sdhA, sdhB,

sdhC, and sdhD have been used for the bacterial genes.

200 MITOCHONDRIAL ELECTRON TRANSFER

some protozoa, parasitic helminths, and marine organisms such as annelids

and mussels, which spend their different life cycles under different conditions

(121) . It is not yet clear whether one or all of the subunits of complex II have

to be changed. The problem is not only to change the direction of electron

ﬂ ow between succinate and fumarate, since the redox potential of all the par-

ticipants must be considered. Hence, ubiquinone (UQ/UQH

; E

′ = +100 mV)

is generally the acceptor in succinate oxidation, while menaquinone is the

donor in bacteria. Eukaryotes engaged in mitochondrial fumarate reduction

must synthesize a closely related rhodoquinone (RQ) (122) . Since these qui-

nones interact with the anchor proteins, a switch in energy metabolism during

development may require not only a different set of anchor proteins, but also

a new pathway for the synthesis of a different quinone. Since UQ and RQ

differ only in one substituent on the quinone ring (the 5 - methoxy group is

replaced by an amino group in RQ), the branchpoint can be postulated to

occur at the end of the pathway of UQ or RQ biosynthesis. In all eukaryotes

examined to date, the ﬁ nding of RQ has been indicative of a potential for

fumarate reduction. For example, the detection of a signiﬁ cant quantity of RQ

in the sporocysts of Schistosoma mansoni has been correlated with fumarate

reduction as an essential pathway during the parasitic stage in the intermediate

hosts (snails) where sporocysts behave as facultative anaerobes (121) . On the

other hand, in Leishmania infantum promastigotes it had also been suggested

that fumarate reduction to succinate constitutes an electron sink during anoxia.

However, the absence of RQ in these organisms suggested that this reaction

was absent, and studies have conﬁ rmed that these organisms produce succi-

nate exclusively via the Krebs cycle under aerobic conditions and go into a

metabolic arrest during oxygen deprivation (123) .

It has been speculated that fumarate reductase was the ﬁ rst activity in pro-

karyotes in evolution, with menaquinone as the cofactor. When oxygen con-

centrations were raised in the atmosphere, an oxidative metabolism evolved

which included the Krebs cycle and speciﬁ cally required the conversion of an

FRD activity to an SDH activity. This was achieved by gene duplication, cova-

lent attachment of the ﬂ avin, a raise in the standard redox potentials of the

iron – sulfur clusters, and the utilization of ubiquinone instead of menquinone

(124) . The FRD activity was subsequently lost in eukaryotic mitochondria.

However, it is conceivable that in a few facultative anaerobic eukaryotes a

“ reversed ” evolution took place. A novel activity allowed the diversion of

5 - hydroxy - 6 - methoxy - 3 - methyl - 2 - polyprenyl - 1,4 - benzoquinone to rhodoqui-

none instead of ubiquinone, while gene duplication and evolution yielded

isopeptides that could be assembled to function as fumarate reductase again.

A detailed examination of cloned genes and deduced peptide sequences will

be required to resolve and support such a hypothesis.

5.2.5.1 Nuclear Versus Mitochondrial Location of Complex II Genes For

a while it was correct to state that all complex II proteins were encoded by

nuclear genes. This generalization has to be abandoned, because several dif-