Scheffler Immo E. Mitochondria

Подождите немного. Документ загружается.

Fatty acid transport is invoked to provide a mechanism for exporting free

fatty acid anions from the matrix, if the supply exceeds the capacity for oxida-

tion. Alternatively, excess fatty acids “ in the cytosol ” can be protonated and

“ ﬂ ipped ” into mitochondria. These simple explanations still lack convincing

experimental support, and they rely on the postulated capacity of UCPs to

transport fatty acids as opposed to being activated by fatty acids. It is also not

clear what is meant by excess and free (?) fatty acids. Normal import of fatty

acids into mitochondria occurs via the well - known carnitine shuttle.

In summary, there are plausible suggestions for the functions of UCP2/

UCP3 in human physiology and pathology, but in view of the relatively minor

phenotypes shown by UCP2 or UCP3 knockout mice, the effects in humans

may also be subtle, and it remains to be seen whether the treatment of obesity

or a signiﬁ cant protection from ROS damage by activation of UCPs can be

achieved.

5.6.3 Uncoupling in Other Organisms

5.6.3.1 In Saccharomyces cerevisiae The growth of yeast on a variety of

carbon sources (glucose, glycerol, acetate, ethanol) requires a large number of

adjustments to optimize growth rates under conditions that may be constitute

a carbon limitation or an energy limitation. Theoretical considerations of

the need to maintain redox balance, along with the observed discrepancy

between ATP requirements for biomass formation with expected yields of

ATP production from substrate oxidation, has suggested that oxidative pro-

cesses uncoupled from ATP production must be operative (347) . Among the

mechanisms considered have been futile cycles (see above) and increased

proton leakage. In a series of papers, Prieto et al. (347 – 349) have provided

details on the activation of a proton - conductance pathway by ATP. At low

[ATP]/[Pi] ratios, binding of ATP to cytochrome oxidase alters its kinetic

properties; but at higher ratios, ATP binding to the outer side of the inner

membrane can increase membrane permeability and cause the collapse of

Δ pH or ΔΨ (348) .

Understanding the properties of the endogenous ATP - induced proton

channel in yeast mitochondria was an important prerequisite for the successful

expression and study of the mammalian UCP - 1 in yeast (342) . In the presence

of phosphate, the two uncouplers could be clearly distinguished. One can also

note that the identity of the newly isolated UCP - 2 gene product was conﬁ rmed

by its expression in yeast, where its activity as an uncoupler was even more

pronounced than that of UCP - 1 (333) .

5.6.3.2 In Plants A previous section has emphasized that plant mitochon-

dria have all of the components found in animal mitochondria, making oxida-

tive phosphorylation in plant mitochondria a very similar process. At the same

time, plant mitochondria have several unique features, most notably a cyanide -

insensitive electron pathway not coupled to phosphorylation. Therefore, there

CONTROL OF RESPIRATION AND OXIDATIVE PHOSPHORYLATION 271

272 MITOCHONDRIAL ELECTRON TRANSFER

appears to be no need for a mechanism requiring a speciﬁ c uncoupling protein,

when oxidative reactions have to be speeded up at certain developmental

stages, in speciﬁ c tissues, or under deﬁ ned physiological conditions. However,

a ﬁ rst report on the cloning of a potential uncoupling protein from a potato

ﬂ ower has recently appeared (350) . It is 44% and 47% identical to the human

UCP1 and UCP2, respectively, and is induced after exposure to cold tempera-

tures. Although there is still uncertainty and controversy about the physiologi-

cal signiﬁ cance of cyanide - resistant respiration (see above), there is one class

of examples where the operation of this uncoupled process for the apparently

sole purpose of heat production is understood in physiological and functional

terms. In the male reproductive structure of cycads and in the ﬂ owers or inﬂ o-

rescences of species belonging to the families Annonaceae, Araceae, Aristolo-

chiaceae, Cyclanthaceae, and Nympheaceae , an accumulated supply of starch

is metabolized (glycolysis) and oxidized by the alternate pathway in at times

a matter of hours. The temperature in the structures involved can reach 15 ° C

above the ambient temperature. This thermogenesis is useful in at least two

ways. In some instances it simply permits the plant to ﬂ ower under climacti-

cally adverse conditions — that is, early in a season; in other examples it has

been clearly established that the elevated temperatures help to volatilize

insect attractants (e.g, the putrescent odor of skunk cabbage) to gain the atten-

tion of pollinating insects (191, 193, 351) .

5.7 REACTIVE OXYGEN SPECIES

Entering the keyword ROS (reactive oxygen species) in a PubMed search

leads to the recovery of ∼ 13,600 citations (October 2006). Of these, ∼ 2200

papers were published in 2006. ROS can be blamed for a lot of problems, and

they may have some positive functions (352) . Oxygen sensing itself has been

proposed to depend on mitochondria - generated reactive oxygen species

(ROS) but independent of oxidative phosphorylation (353, 354) . We know

what they are. We have a good understanding of how and where they are

produced: as an inevitable byproduct of respiration in mitochondria, but in

lesser amounts also at other sites in a cell. However, knowing that some of

them are very reactive and potentially destructive, we do not yet understand

very well in detail how they exert their pathological effects. An expert and

comprehensive treatment on free radicals in biology and medicine can be

found in the book by Halliwell and Gutteridge (355) . It is clear that no attempt

can be made here to summarize the content of approximately 900 pages of

this book, let alone the additional work from the past 7 – 8 years. There is

progress and optimism (356) .

The major reactive oxygen species (ROS) are the superoxide radical, O

−

hydrogen peroxide, H

, and the hydroxyl radical, OH

. When protonated,

the superoxide radical becomes a hydroperoxyl radical (p K

= 4.8). In fact, the

only very reactive species among them is the hydroxyl radical, which is believed

to be responsible for most of the actual damage done to biological macromol-

ecules. It is so reactive that damage is restricted to a small radius limited by

diffusion of the radical. The risk posed by the other two species comes primar-

ily from their spread and subsequent reactions, which ultimately lead to the

formation of the hydroxyl radical. Attempts to react the superoxide in aqueous

solution directly with biologically relevant molecules have been frustrated.

Therefore, the proposal has been made that a series of reactions ultimately

result in hydroxyl radical formation (357, 358) :

22OHOHO

222

⋅

−

+⇒+

(reaction 1)

O Fe(III) O Fe(II)

⋅

−

+⇒+

(reaction 2)

Fe(II) H O Fe(III) OH OH

+⇒ ++

− ⋅

(reaction 3)

Summation of reactions 2 and 3 yields

O HOH OHOOH

222

⋅

−

++⇒++

(reaction 4)

Reactions 2 and 3 have been postulated as intermediate reactions because

reaction 4 has been shown to depend on the presence of redox metals. Reac-

tion 3 is the Fenton reaction, and it emphasizes again that hydrogen peroxide

is by itself less reactive; but upon encountering the ubiquitous metal ions in a

cell, it generates the highly reactive hydroxyl radical. Instead of Fe(II), reac-

tion 3 can also take place with Cu(I) as the reducing agent. It should be noted

that ligands of these metal ions can affect the reaction rate signiﬁ cantly

(355) .

Oxygen is a true substrate of complex IV of the mitochondrial electron

transport chain, and naively it may be assumed that it is the site in a cell where

oxygen can be activated to the superoxide radical. Instead, it is widely accepted

that electrons “ leaking ” from the electron transport chain to oxygen at high

potential upstream sites are responsible for the majority of ROS produced in

mitochondria (358 – 361) . Thus, complexes I ( (362, 363) and III (364, 365) , and

recently complex II (366, 367) , have been implicated in the generation of ROS.

The proposed mechanism has mutated over the years. In consideration of the

small size and zero charge of the oxygen molecule, it can potentially penetrate

deeply into protein domains of these complexes. But where does the errant

electron originate from? Among the suspects are the free radical semiquinone

intermediate of ubiquinone (coenzyme Q) and the free radical intermediates

formed by the ﬂ avin moieties in complex I (and II). Alternatively, it has been

proposed that electrons may escape from the reduced [Fe – S] center N2 (368) .

It is quite likely that a semiquinone intermediate of the Q cycle is the source

of electrons in ROS production by complex III (364) . Recent experiments and

arguments in favor of complex I as a major source of ROS tend to favor the

ﬂ avin intermediate as the donor (366, 369, 370) . McLennan and Degli also

implicate complex II, which had long been ignored as a source of ROS.

REACTIVE OXYGEN SPECIES 273

274 MITOCHONDRIAL ELECTRON TRANSFER

The superoxide radical is also generated “ accidentally ” by reactions of

oxygen at microsomal cytochromes P450, as well as by less well deﬁ ned or

understood reactions (auto - oxidation reactions) involving catecholamines,

ascorbic acid, and reduced ﬂ avins, often with the participation of a metal ion.

The toxicity of such compounds has been exploited in some physiologically

useful reactions. The best - known example is the generation of superoxide,

radicals, hydrogen peroxide, and HOCl by activated macrophages, presumably

to kill their target cells (bacteria, fungi, etc.) before phagocytosis. Several per-

oxisomal oxidases generate hydrogen peroxide — for example, d - amino acid

oxidase, xanthine oxidase, and the dehydrogenase in the ﬁ rst step of very long

chain fatty acid oxidation. In some systems the superoxide radical has been

proposed as an intercellular signal. In a more indirect way, the O

−

ion can

remove nitric oxide, another free radical, and an intercellular messenger in

neurotransmission, in the immune response, and in the regulation of smooth

muscle contraction (see below). Finally, reactive oxygen species have been

assigned a central role in the mechanisms of apoptosis (see Chapter 7 ), with

arguments still unresolved about whether they are responsible for signaling

and cell killing or whether their increased generation is the consequence of

another damaging event affecting mitochondria and perhaps other cellular

constituents.

A variety of protective enzymes have evolved to guard against damage by

ROS, or even to repair damage (e.g., in DNA) resulting from their reactivity.

Present - day aerobic organisms presumably arose from anaerobes by evolving

antioxidant defense mechanisms among others, as the oxygen content of the

atmosphere on earth increased to the present 21%. The existence of antioxidant

defense mechanisms in aerobic organisms is therefore crucial for survival or

longevity. Activity in the ﬁ eld concerned with ROS and antioxidants continues

to be high, because many fundamental questions remain unanswered (359) .

There are various controls to maintain the mitochondrial membrane poten-

tial below dangerous levels by uncoupling proteins and other mechanisms to

prevent leakage of electrons from high potential sites to oxygen. Another pro-

tective mechanism in cells is to sequester iron and copper ions in such a way that

they cannot react readily with superoxide or hydrogen peroxide. Thus, avail-

ability and location of “ free ” metal ions in a cell thus can determine the site of

damage, because the hydroxyl radical formed is too short - lived to diffuse far

from its site of origin. For example, metal ions bound to DNA can initiate free

radical damage in the nucleus. The discovery of superoxide dismutase (SOD)

by McCord and Fridovich (371) represented a major beakthrough in the ﬁ eld.

These enzymes greatly accelerate the dismutation reaction:

22OHHOO

2222

⋅

−+

+→ +

Catalases (see below) then convert the hydrogen peroxide to water and

oxygen. Two SODs have since been characterized in some detail. A cytosolic

form (SOD1) is distinguished by having copper and zinc at the active site

(Cu,Zn - SOD), while a mitochondrial form (SOD2) contains manganese (Mn -

SOD). SOD1 may also be localized in the IMS. The cytosolic SOD is encoded

by a gene on human chromosome 21, and tissue levels of Cu,Zn - SOD are

elevated ∼ 50% in Down ’ s syndrome patients. Since transgenic mice overex-

pressing human SOD also have some neuromuscular abnormalities reminis-

cent of Down ’ s syndrome, it has been concluded that too much of this protective

enzyme can also be deleterious. At the same time, certain mutated forms of

SOD1 are associated with amyotrophic lateral sclerosis (ALS) (372) . It remains

puzzling why such mutant SOD1 proteins affect primarily the mitochondrial

capacity of motoneurons (373 – 375) . A further discussion of the role of SOD

and other scavenging enzymes in neurodegenerative diseases and in aging will

be deferred to Chapter 7 .

Catalases are enzymes deﬁ ning peroxisomes, where they remove hydrogen

peroxide generated by peroxisomal oxidations. However, the bulk of the H

in mammalian cells is removed by glutathione peroxidases. A selenocysteine

residue is required at their active site, where H

is used to oxidize glutathi-

one (GSH) to oxidized glutathione (GSSG). GSH can be regenerated from

GSSG by the oxidation of NADPH, catalyzed by glutathione reductase. Thus,

levels can inﬂ uence the mitochondrial ratio of GSH/GSSG (a measure

of the redox potential), which in turn may control sulfhydryl and disulﬁ de

bond formation and hence activity of certain key enzymes.

It is generally considered that the generation of ROS and the activity of

antioxidant defenses may have to be carefully balanced in vivo . A very slight

imbalance in favor of ROS may be responsible for normal aging. A more

severe imbalance is said to result in oxidative stress. Oxidative stress may

result in the induction of antioxidant defense enzymes and mechanisms, but

beyond a certain level it can increase intracellular calcium levels and free iron

levels, as well as mtDNA mutations, escalating the problem to catastrophic

proportions. A major unsolved and still controversial issue is whether increased

ROS production is a primary consequence of mitochondrial mutations and

dysfunction, or whether a primary defect in ROS scavenging activity is respon-

sible for an abnormal respiratory activity. Clearly, speciﬁ c mutations in struc-

tural genes for subunits of the electron transport complexes could increase the

resistance to electron ﬂ ow and hence cause the escape of electrons from ﬂ avin

or semiquinone radicals. However, the electron leakage is likely to occur

upstream from the mutated complex; in other words, the mutation is not

directly responsible for electron leakage from that complex. For example, iso-

lated bc1 complexes from bovine heart and from wild - type and mutant yeast

strains (in which the midpoint potentials of the cytochrome b hemes and the

Rieske iron – sulfur cluster were altered) were examined by Sun and Trumpower

(376) . In the absence of inhibitors, about 3 – 5% of the electrons escaped to

oxygen to form superoxide anion instead of reducing cytochrome c. The

authors concluded that changes in the midpoint potentials of the redox com-

ponents that accept electrons during ubiquinol oxidation may slow down the

activity of the enzyme but have only small effects on the rate of superoxide

REACTIVE OXYGEN SPECIES 275

276 MITOCHONDRIAL ELECTRON TRANSFER

formation. It should also be noted that stigmatellin, binding at the ubiquinol

oxidation site of the bc1 complex, prevented superoxide formation by the

isolated complex. When the bc1 complex is inhibited by antimycin in intact

mitochondria, increased ROS production is probably due to a back - up and

leakage from complex I. Speciﬁ c inhibitors of the complexes of the electron

transport chain have been shown to cause increased ROS production. When

rotenone occupies the quinone binding site of complex I, the ﬂ avin semiqui-

none is likely to become an electron donor. Similarly, the complex II inhibitor

alpha - tocopheryl succinate has been shown to stimulate ROS production fol-

lowed by apoptosis (367) . The rate of ROS production is increased when oli-

gomycin is used to inhibit the ATP synthase. This inhibition hyperpolarizes the

membrane. An increased ΔΨ will effectively increase the resistance to electron

ﬂ ow in the electron transport chain, stalling electrons at upstream sites from

which they can escape to oxygen.

Reports on the relationship of ROS production to apoptosis are too numer-

ous to summarize here (see also Section 7.4). Are they a signal to, or a side

effect of, the apoptosis pathway? To cite just one example (377) : Depriving

neurons of nerve growth factor triggers apoptosis by Bax - induced cytochrome

c release and activation of a caspase cascade. The authors claim that “ a Bax -

dependent increase in mitochondrial - derived ROS is an important component

of the apoptotic cascade. ” How does Bax induce ROS production? Does the

loss of cytochrome c lead to membrane hyperpolarization? Are the ROS

needed to promote further steps in the progression of apoptosis, or is their

appearance incidental?

A very clever and stimulating experimental model system has recently been

reported by two groups (378 – 380) . An error - prone DNA polymerase γ in a

mouse model was shown to increase mutations in mtDNA quite dramatically,

with the result that the life span of such mice was shortened signiﬁ cantly. At

the same time, there was no apparent increase in ROS production, and the

typical symptoms of oxidative stress were not detectable (379, 380) . These

results argue against an “ error catastrophe ” hypothesis that postulates that a

steady increase in mtDNA mutations would in turn lead to increased oxidative

stress (ROS production) causing the accelerated aging process.

The whole ﬁ eld of mitochondria, oxidative stress, and disease deserves a

book of its own. It continues to be a challenge in many pathological cases to

establish clear cause - and - effect relationships: (1) What is the primary cause or

mechanism resulting in oxidative stress? (2) If oxidative stress means increased

levels of reactive oxygen species, what are the targets: DNA, proteins, or lipids,

or all of the above? (3) What kind of damage (faulty protein) can then act

“ catalytically ” and increase the rate of accumulation of further damage by

ROS - independent mechanisms? From a practical point of view, there are

several challenges. The hydroxyl radical is too short - lived to be measured.

Hydrogen peroxide can be measured in a cellular extract, but such an assay

obscures the intracellular localization of this ROS. Superoxide anions are also

too short - lived to be determined in extracts, but their production can be

assessed in situ in cultured cells by introducing a reagent (e.g., hydroethidine)

that will be oxidized by the superoxide to produce a ﬂ uorescent product. In

such an assay, one measures a rate of accumulation of ﬂ uorescent product

(i.e., the rate of superoxide production) and not its steady - state level. A precise

subcellular localization of the source of O

−

production represents an addi-

tional challenge.

From a medical point of view, there have been numerous efforts to reduce

mitochondrial oxidative damage by targeting antioxidants speciﬁ cally to

mitochondria. The general strategy is to use lipophilic, membrane - permeable

cations (e.g., hexyltriphenylphosphonium cation) similar to those used for

staining mitochondria to promote uptake and accumulation in the organelle.

These cations are linked covalently to organic molecules that can act as free

radical scavengers (MitoQ, MitoE, MitoBPN, etc.). Pioneering work has been

done by Murphy (381 – 383) , and a recent review by Reddy (384) can also be

consulted for entry into the relevant literature.

5.8 NITRIC OXIDE (NO)

In 1987 the ﬁ rst physiological function for NO (nitric oxide) was identiﬁ ed as

a vasodilator of blood vessels, and in 1992 it gained the distinction of being

declared the “ molecule of the year ” in recognition of its many diverse and

important roles. The original discovery earned a Nobel Prize for three out of

the four principal early investigators. In the meantime, NO has been shown to

act also on mitochondria in very signiﬁ cant direct and indirect ways that will

be summarized brieﬂ y in the following paragraphs.

Nitric oxide is formed by the enzyme nitric oxide synthase (NOS) using

arginine as a substrate and producing NO and citrulline. Required cofactors

are: tetrahydrobiopterin (BH4), FAD, FMN, NADPH, and oxygen. NOS con-

stitute a family of enzymes comprised of neuronal NOS ( n NOS or NOS - 1),

inducible NOS (iNOS or NOS - 2), and endothelial NOS (eNOS or NOS - 3).

There has been a long - lasting controversy about the existence of a distinct

mitochondrial NOS (mNOS). One authoritative and exhaustive review dis-

cusses the contentious issues regarding mNOS and comes to a negative conclu-

sion (385) , although publications afﬁ rming the existence of a distinct mNOS

also continue to appear (e.g., 386 – 388). Support for a mitochondrial NOS has

been provided in plants (Arabidopsis) by Guo and Crawford (389) . These

authors made a NOS - GFP fusion protein that could be localized convincingly

in mitochondria. Sequence analyses of eukaryotic genomes have identiﬁ ed

unique genes for the three isoforms of NOS (NOS - 1, NOS - 2, NOS - 3) and

failed to ﬁ nd another that could encode a mitochondrial isoform. NOS co -

puriﬁ ed with mitochondria (contaminant?) “ looks like ” nNOS (386) , but this

problem has not yet been deﬁ nitely settled. Colocalization by immunocytology

cannot distinguish whether the detected NOS is in the inside (matrix) or

bound on the outside. Post - translational modiﬁ cations such as acylation have

NITRIC OXIDE (NO) 277

278 MITOCHONDRIAL ELECTRON TRANSFER

been reported (386) which may explain the attachment to mitochondrial mem-

branes, but do not resolve the problem of localization. Various stimuli are

known to affect the expression of the NOS isoforms, but no unique and speciﬁ c

physiological signal has been discovered targeting exclusively mNOS expres-

sion. As Giulivi et al. (388) put it, “ NO production in mitochondria is spatially

regulated by mechanisms that determine subcellular localization of mtNOS,

likely acylation and protein - protein interactions, in addition to transcriptional

regulation as nNOS. ” They also claim that NO rapidly decays in mitochondria.

On the other hand, it was recently reported that mNOS activity is regulated

by the membrane potential; that is, mNOS is a voltage - dependent enzyme

(387) . Brookes (385) also addresses various other relevant issues including

substrate availability, with a special focus on the intersection of the reactions

catalyzed by NOS with reactions of the urea cycle, and the conditions in the

matrix (pH and pO

Why the high level of interest? What are the actions of NO on mitochon-

dria? At a physiological level, NO regulates blood ﬂ ow and hence the supply

of substrates to mitochondria. It has also been proposed that NO directly

regulates the afﬁ nity of oxygen to hemoglobin and thus the availability of

oxygen. These mechanisms will not be further discussed here.

NO is an inhibitor of cytochrome oxidase (COX), and a regulated synthesis

of NO in the matrix could be a potential mechanism for controlling respira-

tion. The most direct action is the competition of NO for the oxygen binding

site on COX (complex IV) (390, 391) . However, as pointed out by Brookes,

there is a paradox: At low [O

], NO can compete with oxygen at the binuclear

site of COX, but this [O

] is too low for NOS to produce NO (if the reported

of mNOS for O

is correct). When [O

] is high enough for NOS activity, it

will also outcompete NO at the COX binding site. Thus, it is likely that if NO

is to play a role in the control of COX activity, it will have to be produced by

enzymes outside of the mitochondria (with lower K

values for O

The inhibition is freely reversible and dependent on the relative concentra-

tions of these two gases in the matrix. Competitive inhibition has been dem-

onstrated with the isolated enzyme, with mitochondria, nerve terminals, cells

in culture, and tissues. NO effectively increases the apparent K

of the enzyme

for oxygen, and the NO/COX system has been proposed to serve as the acute

oxygen sensing system in cells (392) . Difﬁ culties in measuring NO concentra-

tions in vivo have made it problematic to state whether NO concentrations

reach the level required for effective inhibition (see reference 391 for further

discussion). A mathematical modeling approach by Antunes et al. (393) leads

to the suggestion that there is a “ moderate, but not excessive, overlap between

the roles of NO in COX inhibition and in cellular signaling ” under normal

conditions, but under conditions of COX deﬁ ciency a signiﬁ cant inhibition of

COX activity/respiration is predicted. A recent reevaluation of the interaction

of NO with COX has concluded that NO interacts either with ferrous heme

(competitive) or with the oxidized copper (noncompetitive) where it is reduced

to nitrite (394) . In this context an interesting recent study reports that COX

(from yeast and rat liver) can produce NO from nitrite under hypoxic condi-

tions (395) . These authors therefore suggest that mitochondrially produced

NO is responsible for hypoxic signaling.

It should be noted that NO is a hydrophobic gaseous free radical that is

relatively stable and capable of diffusing across large distances within a cell

and even between cells. It binds to ferrous ions within heme proteins (COX,

hemoglobin, and guanylate cyclase) and reacts rapidly with superoxide to

produce peroxynitrite. The latter reacts readily with thiol groups on proteins

(or glutathione) to produce nitrosothiols (391) . The known cytotoxicity of NO

may in large part result from its conversion to peroxynitrite. It is assumed that

such covalent modiﬁ cations are generally irreversible and hence may lead to

irreversible inhibition of respiratory complexes (396) . A discussion of the

consequences of the modiﬁ cation of thiol groups in mitochondrial proteins is

a subject by itself and the focus of considerable attention (383, 397 – 399) .

The interactions of NO with mitochondria and COX considered so far

describe short - term and direct effects. A long - term effect has now become

apparent that may surpass the short - term effects in signiﬁ cance. In a diverse

group of cells (brown adipocytes, 3T3, HeLa), exposure to the NO donor S -

nitrosoacetyl penicillamine (SNAP) caused a three - to sixfold increase in

mtDNA/cell and an increase in the biogenesis of functionally active mitochon-

dria (390, 400) . NO was found to activate guanylate cyclase and hence elevate

cGMP levels. This signal induced the peroxisome proliferator - activated recep-

tor γ coactivator 1 α (PGC - 1 α ) and PGC - 1 β , followed by expression of the

nuclear respiration factors NRF - 1 and NRF - 2. These transcription factors play

crucial roles in the transcription of nuclear genes encoding subunits of the

electron transfer chain complexes and the mitochondrial transcription factor

A (Tfam alias mtTFA) (401 – 403) . Details on the mechanisms remain to be

elucidated, but a very plausible signaling pathway now includes eNOS as a

master regulator of mitochondrial biogenesis, with important implications for

understanding physiological phenomena such as thermogenesis, obesity, dia-

betes, response to aerobic exercise, calorie restriction, senescence, and more.

In this context the emphasis so far has been on eNOS, since in experiments

with eNOS - negative ( − / − ) mice the anticipated responses to calorie restriction

were not observed (404) .

5.9 THE ROLE OF SPECIFIC LIPIDS

Over the years the puriﬁ cation of various complexes from the electron trans-

port chain or complex V, and also of individual carriers such as the nucleotide

transporter, has yielded preparations that not unexpectedly contained various

lipids, since efforts were made to retain biological activity and native confor-

mations. As higher - resolution structures were obtained by X - ray crystallogra-

phy, speciﬁ c lipids were consistently localized in association with these

complexes, suggesting more than a simple, random role in the integration of

THE ROLE OF SPECIFIC LIPIDS 279

280 MITOCHONDRIAL ELECTRON TRANSFER

the protein complex in a lipid bilayer. Again and again, such preparations

contained cardiolipin (e.g., 405 – 408 ). Cardiolipin has become the “ signature

lipid of mitochondria ” (see Section 6.6 ), and it became “ conventional wisdom ”

to postulate that cardiolipin may be essential for either the assembly of these

complexes in the membrane or their maintenance in their functional confor-

mation (e.g., 409, 410 ).

An experimental test for this idea became possible when genetic approaches

were used to eliminate cardiolipin by mutating the speciﬁ c enzymes required

for its biosynthesis (see Section 6.6). In mammalian cells in tissue culture, this

has been achieved by Ohtsuka et al. (411, 412) , who isolated Chinese hamster

mutants with a defect in phosphotidylglycerophosphate synthase. Such cells had

depleted cardiolipin levels and exhibited mitochondrial dysfunction. Complex

I appeared to be the most seriously affected. In contrast, the initial studies with

a yeast mutant containing a defective cardiolipin synthase gene, cls1 , gave the

surprising result that such mutants were viable and capable of respiration;

that is, they grew in the presence of either nonfermentable or fermentable

carbon sources. These molecular – genetic studies in yeast have now been greatly

expanded. The relevant enzymes and genes required for the biosynthesis have

been identiﬁ ed to be used for genetic manipulations (413) . It emerges that the

biosynthesis of cardiolipin is highly regulated by several mechanisms. Some of

these could be expected to be linked to mitochondrial biogenesis in general, but

there are also speciﬁ c controls mediated by inositol and by the mitochondrial

pH (414, 415) . Thus, mutant mitochondria with a defective assembly of the elec-

tron transport chain exhibit decreased cardiolipin synthesis. While cardiolipin

may not be absolutely essential for the assembly and activity of the electron

transport chain, experiments suggest that it plays a role in stabilizing respiratory

chain supercomplexes (36, 416) . Needless to say, its implication in the process of

apoptosis has also been investigated (417 – 419) . One hypothesis is that the oxi-

dation of cardiolipin by ROS reduces cytochrome c binding to the inner mem-

brane and thus enhances cytochrome c release.

A combination of HPLC, mass spectroscopy, and selective enzymatic cleav-

age has shown that cardiolipin from different organisms and different tissues

contains only one or two types of fatty acids (out of four per molecule of CL),

suggesting a “ high degree of structural uniformity and molecular symmetry ”

(420) . Is this critical? One afﬁ rmative answer comes from the elucidation of

the defect in the Barth syndrome in humans, an X - linked cardio - and skeletal

myopathy (415, 416, 421 – 423) . Progress in our understanding of the molecular

basis for this syndrome came from the identiﬁ cation/cloning of the human

gene (TAZ, encoding tafazzin), followed by the ﬁ nding, characterization, and

manipulation of homologous genes in yeast and Drosophila . As discussed

further in Section 6.6, cardiolipin is synthesized de novo , but is then “ remod-

eled ” to contain an increased proportion of unsaturated fatty acids (e.g.,

linoleic acid). Tafazzin is required for this remodeling of cardiolipin. It has

been characterized as a 1 - palmitoyl - 2 - [14C]linoleoyl - phosphatidylcholine:

monolysocardiolipin linoleoyltransferase (424) . Drosophila melanogaster