Scheffler Immo E. Mitochondria

Подождите немного. Документ загружается.

Figure 6.5 (A,B) Biosynthesis of heme. (C) Heme substituents in type a, b, and c

cytochromes.

COO

Porphobilinogen

COO

HNNH

NH HN

COO

Coproporphyrinogen III

Uroporphyrinogen III

δ-Aminolevulinic acid

Glycine + Succinyl-CoA δ-Aminolevulinic acid

BIOSYNTHESIS OF HEME 311

312 METABOLIC PATHWAYS INSIDE MITOCHONDRIA

Coproporphyrinogen III

Protoporphyrinogen

Protoporphyrin IX

COO

NH HN

Heme

Figure 6.5 (Continued)

among the group of proteins that are associated with iron transport, storage,

and metabolism (see page 324; Section 6.8 .) and that are encoded by mRNAs

containing iron - responsive elements (IRE) in their untranslated regions. When

iron is deﬁ cient, the cytosolic aconitase is in an open conformation (the iron –

sulfur center is dissociated) and capable of blocking the translation of ferritin

and ALA mRNAs by binding tightly to the IRE. This mechanism allows the

biosynthesis of heme to be adjusted to the availability of iron (Fe(II)) in the

cell and mitochondria.

As described in an earlier chapter, the expression of genes encoding mito-

chondrial proteins in yeast is sensitive to oxygen. More speciﬁ cally, the need

for molecular oxygen in the ﬁ nal oxidations in the pathway makes heme syn-

thesis sensitive to oxygen. In turn, heme can induce a number of genes required

for respiration, control of oxidative damage, and sterol synthesis, and it can

repress others (for review see reference 7 ).

As shown in Figures 6.5A and 6.5B , the porphyrin rings have various groups

attached, making the whole structure asymmetrical. In their ﬁ nal form in a - ,

b - , and c - type cytochromes the Fe (II/III) ions are found in distinct microen-

vironments controlling their redox potential, not only because the peptides

differ, but also because the substituents on the rings vary considerably (Figure

COO

a-type

b-type

c-type

CH CH

CHO

S Cys

Figure 6.5 (Continued)

BIOSYNTHESIS OF HEME 313

314 METABOLIC PATHWAYS INSIDE MITOCHONDRIA

6.5C ). In a - type cytochromes the heme is characteristically derivatized with an

isoprenyl side chain. In c - type cytochromes the heme is covalently linked to

cysteine side chains of the peptide.

6.6 CARDIOLIPIN AND LIPID BIOSYNTHESIS/METABOLISM

A comprehensive review of the lipid composition of mitochondria has been

published by Daum (8) . It is generally agreed that the inner membrane has

one of the highest protein : lipid ratios of all biological membranes studied, but

in spite of the dense packing of protein complexes and the various carriers

and import machinery, the membrane nevertheless behaves as a ﬂ uid mosaic

(9, 10) . One of the most noteworthy aspects of the inner mitochondrial mem-

brane is that it is the exclusive location for the lipid diphosphatidylglycerol,

also named cardiolipin. This lipid has long been believed to be absolutely

essential for the functioning of various complexes and integral membrane

proteins. Some doubts were raised from the initial studies of a mutant yeast

strain about the “ absolute ” requirement. In the ﬁ rst report of a yeast mutant

with the cardiolipin synthase gene, CLS1 , knocked out, and no cardiolipin

found in any of the membranes, it may have been a surprise to ﬁ nd the cells

viable in both fermentable and nonfermentable carbon sources. Puriﬁ ed, func-

tional preparations of such complexes have inevitably contained cardiolipin,

suggesting that this lipid has a special afﬁ nity for such proteins. More discus-

sion on the need of this lipid for proper function of these complexes is found

in Chapter 5 . A comprehensive and expert review of the biosynthesis and

biological signiﬁ cance of cardiolipin has been written by Schlame et al. (11)

The proposed functions of this unique lipid have also been broadened to

include a participation in mitochondrial protein import, protein folding, and

assembly — that is, in the assembly and maintenance of the inner membrane.

However, most of these roles of cardiolipin must remain highly speculative. In

this discussion the focus will be on the synthesis of this lipid and some other

characteristics of its composition.

Cardiolipin is found in the membranes of eubacteria such as Escherichia

coli , as well as in the mitochondrial membranes of all eukaryotes studied so

far, but not in archaebacteria (12 – 14) . It is an unusual phospholipid with four

acyl chains. The synthesis in bacteria involves the combination of two mole-

cules of phosphatidyl glycerol with the elimination of one glycerol molecule.

In contrast, in eukaryotes (liver, Saccharomyces cerevisiae, Neurospora crassa ,

bean sprouts, and mussels) the two starting substrates have been shown to be

phosphatidylglycerol (PG) and CDP - diacylglycerol (see references 11 , 13 , and

15 for reviews and listing of additional references). The phosphatidyl group

from CDP - diacylglycerol is transferred to PG by the enzyme cardiolipin syn-

thase (Figure 6.6 ). The evolution of the reaction in eukaryotes from the bio-

synthetic pathway in bacteria has been postulated to be a consequence of the

different levels of the precursor present in bacteria and mitochondria (15) . In

Figure 6.6 (A,B) Cardiolipin biosynthesis.

CTP

Glycerol-3-

phosphatidylglycerol

CHOH

OCH

Cytidine

CDP-diacylglycerol

phosphatidylglycerol phosphate

CHOH

OCH

phosphatidic acid

the former, phosphatidyl glycerol is abundant, and a simple, reversible trans-

esteriﬁ cation can be driven toward cardiolipin by substrate availability. In

mitochondria the precursor may be limited, and the forward reaction is driven

by the irreversible cleavage of a high - energy anhydride bond in phosphatidyl -

CMP (CDP - diacylglycerol). Hence, the mitochondrial reaction is highly exer-

gonic. Properties of this enzyme from various mitochondrial preparations have

been described (13, 15, 16) . Although it had long been suspected to be present

CARDIOLIPIN AND LIPID BIOSYNTHESIS/METABOLISM 315

316 METABOLIC PATHWAYS INSIDE MITOCHONDRIA

exclusively in mitochondria, a detailed study has localized it to the matrix side

of the inner membrane in rat liver mitochondria (17) ; that is, it is an integral

membrane protein in this membrane with a large hydrophilic domain exposed

to the matrix. The enzyme has been solubilized with the help of detergents

and puriﬁ ed to a signiﬁ cant degree from yeast and mammalian sources. Several

of the properties of this enzyme appear to be similar in all eukaryotes exam-

ined, in support of the monophyletic origin of mitochondria (13) , but there are

also some distinctions between the yeast enzyme and the mammalian enzyme

of as yet uncertain signiﬁ cance. Most notably, the molecular mass of the yeast

enzyme has been reported to be 20 – 30 kDa, in contrast to a molecular mass

of ∼ 50 kDa for the mammalian enzyme.

Schlame and co - workers (11, 13, 15, 16) also addressed the problem of the

molecular acyl species found in cardiolipin of various species. In all organisms

a high percentage of symmetrical acyl species was found — that is, species with

identical diacylglycerol moieties. Acyl chains either were identical or differed

only by two carbons or one double bond, respectively. Minimum energy confor-

mations were calculated by these authors, but they point out that the conforma-

tion in a real membrane and in association with protein complexes may be quite

different. Not surprisingly, the models suggest all acyl chains to be arranged

more or less in parallel, and in a highly cartoon - like fashion one could represent

cardiolipin as a single polar head group with four aliphatic chains (Figure 6.6B ).

Another notable property of cardiolipin is that it is an anionic phospholipid.

OCH

Cardiolipin

Figure 6.6 (Continued)

The Barth syndrome is an X - linked cardiomyopathy due to a defect in a

protein called tafazzin (gene: TAZ). This enzyme is involved in a remodeling

reaction of cardiolipin following the de novo biosynthesis of this lipid. Remod-

eling leads to cardiolipin highly enriched in linoleic acid. In the absence of

tafazzin, cardiolipin levels are signiﬁ cantly reduced, and the remaining cardio-

lipin has a more diverse set of acyl chains (18, 19) . Model systems for the study

of tafazzin have been established in yeast (20) and in Drosophila (21) , where

the tafazzin mutation causes reduced motor activity and morphologically

abnormal mitochondria in ﬂ ight muscles.

Since cardiolipin is synthesized on the inside of the inner membrane and is

therefore most likely found in the inner leaﬂ et of the bilayer, one has to ask

whether it can be distributed beyond this particular location. A “ ﬂ ippase ”

would be required to translocate it to the outer leaﬂ et. On the other hand, in

the immediate vicinity of the multisubunit integral membrane protein com-

plexes where cardiolipin is found, the distinction between outer and inner

leaﬂ ets may be meaningless. Movement to the outer membrane is even less

likely but still not deﬁ nitely excluded by experiments. Since mitochondria may

fuse with each other, but not (normally) with other membranous organelles

in a cell, the exclusive presence of cardiolipin in mitochondria can be

explained.

The problem is less settled for the precursor phosphatidylglycerol (PG). It

is derived from the precursor phopatidylglycerophosphate (PGP) (catalyzed

by a phosphatase). PGP is made from CDP - diacylglycerol, and PGP synthase

has also been localized in mitochondria. Such a ﬁ nding would explain the

presence of cardiolipin and PG in mitochondria, but PG appears to be a minor

constituent of other intracellular membranes of mammalian cells, and it may

be necessary to postulate a second, cytoplasmic isozyme for PGP synthase

(22) .

The biosynthesis of the other common phospholipids found in mitochon-

dria is not exclusive, and detailed reviews on phospholipid metabolism are

beyond the scope of this treatise (8, 12, 14) . If the synthesis of these lipids takes

place in the smooth ER, a mechanism must be found for distributing these

lipids to other cellular locations. Vesicular trafﬁ c from the ER via the Golgi

to the plasma membrane, as well as vesicular trafﬁ c in both directions between

these membranes and various intermediate compartments, can be used to

rationalize the distribution of lipids within a cell. Nevertheless, mechanisms

must be postulated to explain the nonuniform distribution of lipids such as

cholesterol in these membranes. Less attention appears to have been devoted

to the delivery of lipids to mitochondria: Are specialized shuttle vesicles

involved? Are lipids delivered individually by means of specialized carrier

proteins? “ Conﬂ uencies ” between the outer mitochondrial membrane and

the endoplasmic reticulum have been observed (see reference 23 for a review).

It is not clear whether it is a transient state, and whether it serves to

import components into mitochondria, or whether it represents an early step

in degradation. When mitochondria are vitally stained with the cyanine dye

DiOC

, the dye is ﬁ rst accumulated by mitochondria, but subsequently it can

CARDIOLIPIN AND LIPID BIOSYNTHESIS/METABOLISM 317

318 METABOLIC PATHWAYS INSIDE MITOCHONDRIA

be observed in the ER after the mitochondria have lost their elongated

shape (23) .

A few generalizations and basic principles are expertly reviewed by Daum

(8) , with special reference to mitochondria, and more recently by Kent (24)

and Dowhan (14) from a more general perspective. While the lipid content of

mitochondria is somewhat variable as expected, differing between tissues as

well as between organisms, it is clear that the major phospholipids of mito-

chondria in addition to cardiolipin (10 – 20%) are phosphatidylethanolamine,

PE (20 – 40%), and phosphatidylcholine, PC (35 – 50%). The other phospholip-

ids are present at a level generally less than 5%, and cholesterol is present

only in traces. These generalizations apply to vertebrate tissues, as well as to

a variety of microorganisms and plant mitochondria. The latter also contain a

variety of free and derivatized sterols (see reference 8 for review and original

references). A further distinction can be made between the outer and inner

membranes of mitochondria, and a few attempts have been made to character-

ize the differences. The presence of cardiolipin in the outer membrane is still

controversial, since contamination is a possibility. Other differences in the

ratio of PE : PC, or the relative amounts of phosphatidylinositol, PI, have been

observed, but the functional signiﬁ cance is still obscure.

The origin and site of synthesis of mitochondrial lipids are areas still under

investigation and in a state of evolution (25) . Conclusions are somewhat

subject to the success of fractionation schemes — most notably, fractionation

of mitochondria from microsomes and other vesicular structures, identiﬁ ed by

marker enzymes. Phosphatidylethanolamine is made in mitochondria from the

decarboxylation of phosphatidylserine, PS. A search for the cellular location

(in yeast) of phosphatidylserine synthase has identiﬁ ed a microsomal subfrac-

tion, but the absence of typical microsomal marker enzymes has led the authors

to propose a novel particle population. One is forced to conclude that PS is

imported into mitochondria, but some of the PE made must also be exported

to other cellular membranes where it is found. Alternatively, a second PS

decarboxylase may exist in a cytosolic membrane, and indications for its exis-

tence are found from the study of a yeast mutant defective in the mitochon-

drial PS decarboxylase (see reference 24 for discussion).

While cell fractionation studies have proved their worth over the decades,

the fractionation of subcellular membranous structures has been a technical

challenge because of microheterogeneity, and results are subject to misinter-

pretation when contamination cannot be ruled out. In the future, one can

expect that model organisms like yeast will allow systematic gene disruptions

to explore how the absence of a particular enzyme (and hence the product

from this pathway) affects growth and other cellular activities. An example of

the unexpected type of result derived from a genetic approach is the ﬁ nding

that unsaturated fatty acids are essential for the distribution of mitochondria

to the daughter cells in the yeast Saccharomyces cerevisiae , presumably because

the necessary shape changes (elongation) are dependent on the ﬂ uidity of the

mitochondrial membranes (26) .

6.7 BIOSYNTHESIS OF UBIQUINOL (COENZYME Q)

The biosynthesis of ubiquinol was elucidated some time ago from studies in

Escherichia coli and yeast Saccharomyces cerevisiae . Genetic studies were

critical, since CoQ - deﬁ cient strains of each organism could be isolated, and

intermediates accumulating at different stages in the biosynthetic pathway

gave clues about the sequence of reactions. Although the pathways differ

slightly in some intermediate steps in prokaryotes and yeast, 10 genes have

been deﬁ ned in each organism by the isolation of eight complementation

groups ( coq1 – coq10 in yeast). The CoQ - deﬁ cient yeast strains are respiration -

deﬁ cient and fail to grow on nonfermentable carbon sources, but they do grow

on glucose (27) . The laboratory of Clarke has made a number of signiﬁ cant

contributions to the detailed functional analysis of a number of different coq

mutants in yeast (28 – 32) . The deﬁ ciency can in many cases be overcome by

the addition of exogenous Q

The reactions deﬁ ned by mutations in CoQ/ubiquinone - deﬁ cient bacteria

or yeast are the ﬁ nal reactions unique for the biosynthesis of ubiquinone. Key

starting materials in this pathway are p - hydroxybenzoic acid and polyisoprene

diphosphate. The ﬁ nal polyprenyl tail contains n isoprene units, where n varies

between organisms. In mammalian mitochondria n = 10, and hence the com-

pound is also referred to as Q

. The Q

derivative is extremely insoluble in

water, and in in vitro experiments as well as in therapeutic applications deriva-

tives with shorter tails are often used.

The synthesis of the polyprene tail begins in the well - known pathway

from acetyl - CoA via hydroxymethylglutaryl - CoA , mevalonate, geranyl -

pyrophosphate, and farnesyl - pyrophosphate, the initial reactions established

ﬁ rst by K. Bloch in the biosynthesis of cholesterol. It has now become clear

that farnesyl - pyrophosphate is a key intermediate and branch - point in several

very important pathways leading to cholesterol synthesis, dolichol synthesis,

ubiquinone synthesis and protein isoprenylation (see references 33 and 34 for

recent reviews). The reactions up to the synthesis of farnesyl - pyrophosphate

are also referred to as the mevalonate pathway, and they are illustrated in

Figures 6.7A and 6.7B . Three acetyl - CoA molecules are ﬁ rst combined succes-

sively to form hydroxymethylglutaryl - CoA, the precursor for mevalonate.

Phosphorylations and a decarboxylation yield isopentenyl - pyrophosphate

(IPP), which is reversibly isomerized to dimethylallyl - pyrophosphate (DMA -

PP). A condensation of IPP and DMA - PP yields geranyl - pyrophosphate

(GPP), and the addition of another isopentene unit leads to farnesyl - pyro-

phosphate (FPP). The enzymes are soluble cytosolic enzymes, with one excep-

tion: The enzyme HMG - CoA reductase is an integral membrane protein in

the endoplasmic reticulum. It has received a considerable amount of attention

as a key regulatory enzyme in cholesterol biosynthesis. It may therefore also

inﬂ uence the rate of synthesis of the other products at the end of the various

branches, but while its regulatory role in cholesterol biosynthesis is undis-

puted, modulating its activity by dietary conditions or inhibitors has been

BIOSYNTHESIS OF UBIQUINOL (COENZYME Q) 319

320 METABOLIC PATHWAYS INSIDE MITOCHONDRIA

Figure 6.7 (A – D) Ubiquinone, cholesterol, and Dolichol synthesis.

SCoA

HMG-CoA reductase

mevalonate-5-phosphotransferase

phospho-mevalonate kinase

pyrophospho-mevalonate

decarboxylase

HMG-CoA

mevalonate

isopentenyl pyrophosphate

OCH

CCH

O P

CCH

OHCH

CCH

found to have no signiﬁ cant inﬂ uence on the biosynthesis of dolichol, ubiqui-

none, or the isoprenylation of proteins. This apparent dilemma has been

addressed by several authors, and a plausible model referred to as the “ ﬂ ow

diversion hypothesis ” has been postulated. According to this hypothesis, most