Human Reproduction, Vol. 15, (Suppl. 2), pp. 1-10, 2000
Origin and persistence of the mitochondrial genome
Robert P.SJansen1
Sydney IVF and the University of Sydney
lr
ro whom correspondence should be addressed at: Sydney IVF, 4 O'Connell Street,
Sydney 2000, Australia. E-mail: [email protected]
The mitochondrial genome comprises a circular, histone-free 'chromosome' of 16.6 kb
of DNA, present in one or more copies in
every mitochondrion. This chromosome has
been tightly conserved for more than half
a billion years, coding in every multicellular
animal so far investigated, both vertebrate
and invertebrate: (i) the same 13 protein
subunits required for oxidative phosphorylation; (ii) a component of each of the
two mitochondrial ribosome subunits; and
(iii) the 22 transport RNAs present within
the mitochondrion. Exons on the circle are
tightly packed, with no spacing introns.
Mitochondrial DNA is histone-free, has limited repair ability, and has a relatively
high mutation-fixation rate. Inheritance is
cytoplasmic and maternal, with epidemiological evidence (namely the familial distribution of polymorphisms) indicating that
recombination with mtDNA of paternal origin is exceedingly rare. Thus the maintenance and evolution of mtDNA (its
remarkably successful symbiotic persistence with the nuclear genome) has been
essentially asexual. The machinery for
homologous recombination is present in
mitochondria of at least some species, however, and it might be surprising if it did not
occur between circles in some circumstances. By bringing together the fields of
© European Society of Human Reproduction and Embryology
mitochondrial biochemistry, evolutionary
genetics, reproductive physiology, and
neuromuscular medicine in focusing on the
inheritance of normal and abnormal human
mtDNA, we can hope to better understand
the forces behind this genome's inheritance
and what might be required of ovarian
function to satisfy its accurate persistence
over millions of years. Clinically we can
hope also for a better understanding of
ooplasmic factors in human fertility and in
the wide manifestations of mitochondrial
genomic disease.
Key words: evolution/mitochondrial DNA/
mitochondrion
Introduction
Molecular geneticists and ovarian physiologists today face the challenge of defining
and reconciling two major biological imperatives that each centre on oogenesis, folliculogenesis and competition between ovarian
follicles: (i) endeavouring to discover what
cytoplasmic factor or factors permit some
oocytes but not others to produce viable
embryos and ongoing pregnancies; and (ii)
defining how the mitochondrial genome
(degradation of which is important in both
ageing and a number of serious mitochondrial
1
R.P.S.Jansen
milion years ago
14,000
4500
3500
milestone
origin of universe
era
solar system condenses
Archean
prokaryotes
2000
Proterozoic
photosynthesis
1000
550
400
Ediacarian
Cambrian
Devonian
karyotes
conodonts
frogs
Pliocene
Pleistocene
australopithicus
homo
Figure 1. A condensed evolutionary scale of life on earth.
Mitochondria evolved during the Proterozoic era, about 2
billion years ago, and are present in all multicellular animals,
fungi, and plants. Since the time of the evolution of the
ancestors to the vertebrates, 500 million years ago, the
mitochondrial genome in animals has been virtually fixed
(see Figure 2).
diseases) is refreshed and purified as it passes,
via the oocyte's cytoplasm, from one generation to the next.
The mitochondrion as symbiont
Phylogenetically, the mitochondrion is
ancient, predating the separation of multicellular animals, fungi and plants (Lang et al.,
1997), perhaps 2 billion years ago (Figure 1).
It is at the heart of what has recently been
called the 'hydrogen hypothesis for the first
eukaryote' (Martin and Miiller, 1998), by
which, in the hydrogen-rich, reducing environment of the earth at that time, a putative
hydrogen and CO2-feeding autotrophic archeobacterium took on board a heterotrophic symbiotic organism that was not only capable of
metabolizing complex carbohydrates but, as
by-products, produced the hydrogen and CO2
the host needed to live. With genomic
rearrangement, and a postulated chimaeric
merger into a third archaic organism to provide
a cytoskeleton (Doolittle, 1998), this trimeric
symbiosis persisted into a more aerobic, oxidizing world, so that, by the time the multicellular plants, fungi and animals took origin from
a line of protists, pyruvate was the mitochondrion's feedstock, and hydrogen, in its oxidized (electron-stripped) form of the hydrogen
ion, was being pumped into an intramembranal
organellar space in such concentrations as to
reverse an ATP-requiring Na + /H + pump, thus
generating ATP for the host cell. While such
symbiotic relationships must have taken place
very many times in primeval times, the evidence is that all of today's extant mitochondrion-containing karyotes are derived from a
single successful symbiosis.
Presumably for reasons of efficiency (and to
rob the symbiont of replicative independence)
most of the proto-mitochondrion's genes were
ceded to the nuclear genome during the establishment of successful symbiosis.
A recently described freshwater protozoon
Reclimonas americana, a heterotrophic flagellate closely related to the amitochondriate
retortamonads (a group of protists that is
thought to have diverged from the main eukaryotic line before the acquisition of mitochondria), has a mitochondrial genome that consists
of 69 034 bp, and codes 97 genes, including
62 protein-coding genes of known function,
far more than its nearest competitor (37 in the
primitive non-vascular land plant Marchantia,
a liverwort; Lang et al., 1997). However, in
all extant multicellular animals, invertebrate
as well as vertebrate, the mitochondrial genome is restricted to coding just 33 genes.
Why this transfer of genomic responsibility
from the cytoplasm to the nucleus was left
incomplete is a fascinating question. It is
known that a slight drift in the transfer RNA
code for amino acids had taken place in proto-
Mitochondrial genome origin and perseverance
mitochondria by the dawn of the Cambrian
era, -550 million years ago (Anderson et aL,
1981), locking into the mitochondria of most
multicellular species a seemingly indelible
genetic independence. This could, in principle,
explain why mitochondria remain partly genetically independent, even though mtDNA has
frequently escaped from mitochondria to enter
the nuclear genome (human nuclear DNA
contains at least several hundred copies of
mtDNA-like sequences) (Fukuda et aL, 1985).
Colonization of eukaryotic organisms by parasites has been a recurring phenomenon through
evolution, however, and another such endosymbiont indispensable to life on earth (and
dating from Proterozoic times, more than a
billion years ago) is the plastid, or chloroplast,
the photosynthetic organelle of green algae
and higher plants. Like the mitochondrion, the
chloroplast also has its own circular chromosome, so that the higher plants, like multicelled
animals, keep a part of the genomic responsibility for proper cytoplasmic metabolic function
in the cell periphery, close to the metabolic
action. However, unlike the mitochondria from
most species, chloroplasts have continued to
share the universal genetic code of the nuclear
genome (and of modern bacteria and viruses).
In consequence, there has been a further drift
of plastid genes to the nucleus among flowering plants, but the number lost has been small
(just 44 among 210 examined genes that
have descended from cyanobacterial genomes)
(Martin et aL, 1998).
One can conclude that there is value in
mitochondrial genes' intraorganellar, cytoplasmic location beyond simply having
become stuck there through their departure
from the universal genetic code: that mitochondria, like chloroplasts, have a locally
active genome for a reason fundamental to
a competitive cytoplasmic economy. What
mitochondria and chloroplasts have in common is highly energetic intramembranal
organellar electron transport chains (including,
respectively, oxidative phosphorylation and
photosynthesis). These organelles are the
power generators of cells. Without carefully
matched and coupled chemistry, 'electrocution' could be said to occur: uncontrolled
electron-donation to chemical species not
meant to be so charged, producing (with
mismatched oxidation and phosphorylation,
for example) reactive oxygen species able and
ready to damage cell components, including
DNA mutation, through inappropriate reduction or (in the case of the hydroxyl radical)
oxidation.
Evidence is accumulating that the evolutionarily important safeguard to uncontrolled
electronation has been to have organelles able
to respond immediately to increased demand
or to electron leakage with synthesis of new
electron transport chain protein, with control
for a crucial rate-limiting step vested at a
genomic level within the organelle (Allen and
Raven, 1996; Race et aL, 1999).
The mitochondrial genome
The human mitochondrial genome comprises
a circular, histone-free molecule composed of
16.6 kb of DNA, present in one or more
copies in every mitochondrion (Clayton, 1991;
2000; Wallace, 1995; Zeviani and Antozzi,
1997). This chromosome has been tightly
conserved (Wolstenholme etaL, 1985), coding
in every multicellular animal so far investigated (vertebrate and invertebrate): (i) the
same 13 protein subunits required for oxidative
phosphorylation (of a total of more than 80,
the remainder encoded by nuclear genes and
imported into the mitochondrion); (ii) a component of each of the two mitochondrial ribosome subunits; (iii) the 22 transfer RNAs
present within the mitochondrion (Figure 2),
and (iv) small RNA primers needed for DNA
replication. Among vertebrates (and hence for
>400 million years) the genes (other than
those for primers and control) (Shadel and
R.P.S.Jansen
nt 13446
13440 WD
8093 WO deletion
Figure 2. Organization of the vertebrate circular mitochondrial genome. The entire circle (16.6 kb, in humans) consists of
exons, except for the D-loop (which includes the heavy-strand replication origin, OH). Protein-coding and ribosomal RNA
genes are shaded and labelled. Small intervening exons code for 22 tRNAs and, in the region of the D-loop, RNA primers.
The area affected by the 'common deletion' is marked with an arc. Drawn by Michele Bon Durant. (For further information,
see Clayton, 2000.)
Clayton, 1997) have remained in the same
order around the circle.
The mitochondrial genome is special in
other ways. Mitochondrial (mt)DNA is not
transmitted through Mendelian (diploid, or
'sexual') principles, but passes from one generation to the next by way of the oocyte's
cytoplasm. An individual's mitochondrial
DNA is consequently entirely derived from
his or her mother. Paternal mtDNA entering
the oocyte with the fertilizing spermatozoon
appears to be expelled from the cleaving
embryo at or soon after the 2-cell stage
(Kaneda et al, 1995; Sutovsky et al., 1996;
Cummins, 2000). The fact that mtDNA replication is switched off in the oocyte by the
time it becomes fertilizable (by suspension of
DNA polymerase y production from nuclear
genes; Jansen, 2000) ought to further discourage any leaked paternal mtDNA replicating in
the zygote.
Mitochondrial DNA mutations
The mitochondrial genome is thought to deteriorate with age, through accumulation of point
mutations and rearrangements (especially
deletions), in most tissues (Wallace, 1995),
including the ovary (Kitagawa et al., 1993),
at a rate much faster than for the nuclear
genome. Partly because of an absence of
histones, partly because replicative intermediates are single-stranded for prolonged
periods during mtDNA replication, partly
because of a lack of double-strand DNA repair
mechanisms, the mutation-fixation rate of
mtDNA is 15-20 times greater than the rate
for nuclear DNA (Wallace et al., 1987).
The mitochondrial genome consists mostly
of tightly packed exons (Figure 2). There are
no introns between genes, so virtually every
point mutation or deletion in a mtDNA circle
is potentially capable of affecting that
Mitochondrial genome origin and perseverance
mitochondrion's ability to support cellular
respiration. Accumulation of such mutations
in post-mitotic somatic cells accounts for at
least some of the well known tissue weakness
that is a normal part of the ageing process,
as well as for a serious group of familial
neuromuscular diseases inherited maternally
(Wallace, 1995; Zeviani and Antozzi, 1997).
For most cells most of the time, that is for
non-embryonic somatic cells, mitochondria
are polyploid, with multiple circles and alleles
present in each organelle (Satoh and Kuroiwa,
1991), which enables mutations to be compensated for as they occur with age (at least
while they fall short of affecting all DNA
circles within the organelle). In addition, despite their discrete, generally elongated appearance in electron micrographs, recent studies
have shown that mitochondria, at least in
somatic cells, can form a dynamic network
(Hayashi et al, 1994; Rizzuto et al, 1998),
with repeated fusion and fission between
organelles, and that mtDNA replication is not
a general or random phenomenon, but occurs
close to the nucleus, with new mtDNA then
finding its way to more peripheral organellar
locations through the network (Davis and
Clayton, 1996).
Conventionally it is thought that all of
an individual's mtDNA will have the same
nucleotide sequence (at least until age-related
mutations occur in various tissues), a condition
termed homoplasmy. When a non-lethal mutation occurs in the mtDNA of the germline,
the affected individual will have more than
one mtDNA species, termed heteroplasmy.
Heteroplasmy occurs within cells (and probably within somatic cell mitochondria) as well
as within and between tissues, and so could
be considered to be the polyploid version of
what for diploid genomics is termed heterozygosity. A feature of mitochondrial inheritance is that heteroplasmy can revert to
homoplasmy quickly, within a small number
of generations. This rapid genetic drift is
attributed to a tight restriction event (or
'bottleneck') in the population of mitochondria
at some point or series of points in passage
through the female germ line (Hauswirth and
Laipis, 1985) - the phenomenon that is a
major focus of the present volume.
Meiotic mammalian oocytes appear to have
a mtDNA copy number just slightly in excess
of the number of mitochondria (Piko and
Matsumoto, 1976), i.e., we should regard at
least mature oocyte mitochondrial chromosomes as single or haploid (Lynch et al,
1993), distinct both from the diploid state that
characterizes the paired nuclear chromosomes
and from the polyploid mitochondrial chromosomes of somatic cell mitochondria. Because
it is not subjected to methodical synapsis,
recombination and cross-over at the time of
nuclear meiosis (or mitosis), there should
be another way to ensure that germ plasm
mitochondria with mutant DNA (Medvedev,
1981) are not transmitted to, or do not survive
indefinitely in, the offspring.
Mitochondrial DNA mutations in primary
oocytes in primordial follicles might be
expected to increase with time, that is with
maternal age. Empirical data on this question
are incomplete and conflicting, being mostly
confined so far to searches for deletions rather
than point mutations (Chen et al, 1995; Keefe
et al, 1995; Brenner et al, 1998).
Muller's ratchet and the non-diploid
evolution of DNA
In any highly ordered information set, such
as a code (Shannon and Weaver, 1949), the
great majority of random changes will, as
'noise', cause a degradation of the information: a chance improvement is a fluke. As
the process continues, sooner or later the
information set becomes unusable.
This restating of the phenomenon of entropy
can also be articulated as Newton's second
law of thermodynamics: that without a focused
R.P.SJansen
input of energy all systems tend inexorably to
a state of disorder. In the framework of the
genome, entropy implies that the net effect of
random mutations is to produce inferior gene
products, and that sooner or later a 'mutational
meltdown' (Lynch et al., 1993) will occur,
after which the information set is essentially
useless. This process has been named Muller's
ratchet (Felsenstein, 1974), after W.H.Muller
(Muller, 1964), who first suggested that mutational degradation and then extinction would
be the inevitable consequence of a species
adopting asexual replication as a strategy for
reproduction. By having two copies of every
gene, diploids are virtually assured of producing at least one normal copy of each protein.
Although diploids carry twice as many alleles
as haploids, therefore experiencing higher
mutation pressure, diploids carry the advantage of being able to compensate for and
hence to mask deleterious alleles (Otto and
Goldstein, 1992). Polyploidy, which is how
mtDNA circles populate somatic cell mitochondria, extends the benefits of diploid compensation, increasing the number of mutations
still consistent with normal mitochondrial
function.
If mutations occur, it would be advantageous if there were to be a reliable physiological mechanism: (i) for giving an opportunity
for back-mutations to occur (thus preserving
the genome); (ii) for selection in favour of
those back-mutations (and of rare advantageous mutations); and (iii) for limiting the
spread of persistent harmful mutations through
the population, mutations that are too slight (or
too late in origin) to have escaped intraovarian
culling (whatever form that might take). The
conservation of the 16.6 kb vertebrate mitochondrial genome for the last 400 million years
or more, despite a mutation rate estimated at
15- 20 times that of nuclear DNA (see above),
suggests that such a physiological selection
process must exist.
Among an asexually reproducing species
with a single or haploid genome, a tight
restriction or bottleneck in replicating DNA
unit copy number, followed first by rapid
amplification into a tolerant environment, then
ultimately bringing to bear a competitionfostering environmental constraint, constitutes
a strong force for survival of corrective back
mutations and advantageous novel mutations
(Chao et al., 1997; Burch and Chao, 1999).
It might help in understanding the informational persistence of the mitochondrial genome
to examine in some detail the experiments
credited with first demonstrating an advantage
for recombination in halting or reversing
Muller's ratchet for an asexually reproducing
organism. Chao et al. demonstrated in the
RNA bacteriophage 0 6 that selfing and mass
selection, a repeated sequence of restriction
event (bottleneck selfing, see below), amplification, and then an environmental constraint
to confer advantage to the most competitive
species, could reverse Muller's ratchet without
genetic recombination or reassortment (Chao
et al. 1997).
Chao demonstrated the power of selfing and
mass selection empirically, but also serendipitously; he was studying RNA phage viruses that
normally carry out limited genetic exchange
through genome segmentation and segment
reassortment (the (j) 6 virus has three segments
that can separate and recombine, and it was
shown that recombination improved the efficiency of the process). The virus strains used
had been subjected to intensified genetic drift
and fixation through repeated tight bottlenecking, achieved by reinfecting a new bacterial
cell with just one viral particle as soon as it
began to reproduce; after 40 such cycles, all
20 new strains had suffered a decline in fitness
from mutational deterioration in the genome
in comparison with the parent strain. He then
subjected these new pure strains (strains
'selfed' this way) and crosses of such strains
to mass selection, achieved by propagating
the pure or the hybrid populations through
Mitochondrial genome origin and perseverance
several growth cycles with larger bottlenecks
(tight enough to reward the fittest but wide
enough to prevent significant random genetic
drift and fixation). After 30 generations of
mass selections the pure degraded strains had
improved their fitness by 21%. The hybrid
populations had improved their fitness by
30%. The additional 9% reflected the value
of recombination of segments with those of a
different strain, and thus revealed the advantage of recombination. In other words, genetic
fitness was recovered (Muller's ratchet was
reversed) by simple mass selection: improvement from mutations alone (compensatory
and back-mutations) through selfing and mass
selection in this experiment was more than
twice as high as the additional improvement
conferred by segmental reassortment.
The parallels between Chao's viruses and
intergenerational mitochondrial genomes is
striking in more ways than one. In a computer
simulation of an asexually replicating genome
regarded as modelling germline mitochondrial
bottlenecks, it has been predicted (Bergstrom
and Pritchard, 1998) that the tighter the
bottleneck the more effectively a mutation
will be eliminated, either through good luck
(when a non-mutated species or a small group
of species is sampled by chance and then fixed,
or 'selfed'), or through good management (by
concentrating the mutation and thus increasing
the vulnerability of the host egg, in anticipation
of the subsequent application of selection
pressure on eggs). Bergstrom and Pritchard's
simulations predict that a tight bottleneck will
hasten the degrading ratchet for mitochondria
of individual germ cell lineages, but by intensifying competition between germ cell lineages
it should reduce the population-level (i.e. the
host-level) ratchet.
Before the primeval evolution of diploidy,
mass selection (the sequence of restriction,
amplification and constraint, with or without
recombination between DNA circles) was presumably the chief mechanism through which
evolution produced biological success and
diversification. Added to mass selection, even
the simplest form of recombination, namely
segment reassortment, exposes mutations to
stronger selective forces (by increasing the
mutation load variance). Recombination offers
an organism the chance of reconstructing a
mutation-free genome from two genomes that
contain different mutations, while also constructing mutation-packed genomes (again by
chance) headed nowhere. Any form of
recombination thus gives an advantage in
the race to improve fitness. Since catananes
(dimeric to hexameric mtDNA circles) are
known to account for -30% of mtDNA forms
in both sea urchin oocytes (Matsumoto et ai,
1974) and mouse oocytes (Piko and
Matsumoto, 1976), but for < 1 % of mtDNA
circles in somatic cell mitochondria
(Matsumoto et al., 1976), there is a need
to learn whether these germ cell mtDNA
catananes could represent physiological
opportunities for recombinative reassortment
within mitochondrial genomes of a lineage.
Would mitochondria have benefited from
diploidy? In principle, diploidy doubles the
exposure to mutations while halving the effective reproductive rate, so carries a substantial
cost. Computer simulations reveal that diploids
can only invade haploid population models
when there is both a relatively strong heterozygous masking of deleterious alleles by wild-type
alleles (compensation) and a high rate of homologous recombination (Otto and Goldstein,
1992), increasing population variance for the
distribution of mutations among individuals
and enabling two advantageous mutations to
aggregate much more often than they would by
sequential mutation in transmission of a haploid
genome, and thus to more rapidly survive or
exploit environmental change.
Arguably, however, the mitochondrial genome's job in metabolism is a very conservative
one (i.e. the prevention of damaging electronation during ATP production, see above),
7
R.P.S.Jansen
not an inventive, new niche-seeking one.
Mitochondrial DNA has less need for mutational preadaptation to exploit new environments than simply to remain fully effective
for safe, efficient mitochondrial function in a
eukaryotic intracellular environment that has
changed little in half a billion years or more.
At first sight this might imply the need for a
genome resistant to mutation. Yet the opposite
seems to be the case: the mitochondrial mutation fixation rate is 15-20 times the mutation
rate of the nuclear mutation rate. In effect,
what Chao et al. demonstrated with the restriction, amplification, constraint sequence for
improvement of single, haploid viral genomes
reproducing asexually (Chao et al. 1997) is
that not only should mitochondria need a high
mutation rate to provide enough opportunity
for back mutations to return enough genomes
to normal, but degradation of the genome will
be rapid unless clonal mitochondrial genome
populations frequently expand to very large
sizes, numbers that should, at least in the long
run, exceed the reciprocal of the back-mutation
rate (Lynch, 1996).
For mtDNA species from one generation of
the host individual to the next, this population
size increase could need to be, and probably
is (Jansen and De Boer, 1998), greater than
the 50-fold expansion from 'four or five rounds
of replication' previously proposed to explain
mitochondrial bottleneck phenomena in
humans (Poulton et al., 1998).
Selfing and mass selection might be enough
to stop rapid mitochondrial genomic deterioration, but when winning mtDNA species are
rewarded with survival irrespective of their
absolute qualities there is still the challenge
of preventing a longer-term, gradual
weakening of the mitochondrial genome. For
free-living haploid species (as for the diploids)
nature can achieve this from time to time
by a major change in the environment that
provides the opportunity for competition
between species under extraordinary condi-
tions, conditions that will reward only
extraordinary genomes, and thus provide periodic rejuvenation of germ lines (Medvedev,
1981).
The mitochondrial genome of eukaryotic
organisms is protected from environment-led
purgings (short of the extinction of a species)
by its indispensable, intracellular location,
existing in symbiosis with an evolving nuclear
genome. For the mitochondrial genome to
evolve (or at least not to degrade), pressure
must come from the species' nuclear genome,
which 'has an interest' in mitochondria
remaining efficient.
The idea behind this symposium is that the
manifestations that the organism's elaboration
of tests and punishments can take to force
periodic rejuvenation of the mitochondrial
genome will have clinically important consequences. These considerations and consequences form the basis of many of the
papers that follow.
References
Allen, J.F., and Raven, J.A. (1996) Free-radical-induced
mutation vs redox regulation: costs and benefits of
genes in organelles. J. Mol. Evol, 42, 482-492.
Anderson, S., Bankier, A.T., Barrell, B.G. et al. (1981)
Sequence and organization of the human
mitochondrial genome. Nature, 290, 457-465.
Bergstrom, C.T. and Pritchard, J. (1998) Germline
bottlenecks and the evolutionary maintenance of
mitochondrial genomes. Genetics, 149, 2135-2146.
Brenner, C.A., Wolny, Y.M., Barritt, J.A. et al. (1998)
Mitochondrial DNA deletion in human oocytes and
embryos. Mol. Hum. Reprod., 4, 887-892.
Burch, C.L. and Chao, L. (1999) Evolution by small
steps and rugged landscapes in the RNA virus (j) 6.
Genetics, 151, 921-927.
Chao, L., Tran, T.T., and Tran, T.T. (1997) The advantage
of sex in the RNA virus (j) 6. Genetics, 147, 953-959.
Chen, X., Prosser, R., Simonetti, S. et al. (1995)
Rearranged mitochondrial genomes are present in
human oocytes. Am. J. Hum. Genet., 57, 239-247.
Clayton, D.A. (1991) Replication and transcription of
vertebrate mitochondrial DNA. Ann. Rev. Cell BioL,
7, 453^78.
Mitochondrial genome origin and perseverance
Clayton, D.A. (2000) Transcription and replication of
the mitochondrial genome. Hum. Reprod., 15 (Suppl.
2), 11-17.
Cummins, J.M. (2000) Fertilization and elimination of
the paternal mitochondrial genome. Hum Reprod., 15
(Suppl. 2), 92-101.
Davis, A.F. and Clayton, D.A. (1996) In situ localization
of mitochondrial DNA replication in intact
mammalian cells. J. Cell Biol, 135, 883-893.
Doolittle, R.F. (1998) Microbial genomes opened up.
Nature, 392, 339-342.
Felsenstein, J. (1974) The evolutionary advantage of
recombination. Genetics, 78, 737-756.
Fukuda, M., Wakasugi, S., Tsuzuki, T. et al. (1985).
Mitochondrial DNA-like sequences in the human
nuclear genome. Characterization and implications in
the evolution of mitochondrial DNA. J. Mol. Biol,
186, 257-266.
Hauswirth, W.W. and Laipis, P.J. (1985) Transmission
genetics of mammalian mitochondria: a molecular
model and experimental evidence. In Quagliariello, E.,
Slater, E.C., Palmieri, F. et al. (eds). Achievements
and Perspectives of Mitochondrial Research. Vol. II:
Biogenesis. Elsevier Science, Amsterdam, pp. 49-59.
Hayashi, J.-L, Takemitsu, M., Goto, Y., and Nonaka, I.
(1994) Human mitochondria and mitochondrial
genome function as a single dynamic cellular unit.
J. Cell,. Biol, 125, 43-50.
Jansen, R.P.S. and De Boer, K. (1998) The bottleneck:
mitochondrial imperatives in oogenesis and ovarian
follicular fate. Mol. Cell Endocrinol, 145, 81-88.
Jansen, R.P.S. (2000) Germline passage of mitochondria:
quantitative considerations and possible embryological sequelae. Hum. Reprod., 15 (Suppl. 2), 112128.
Kaneda, H., Hayashi, J.-L, Taya, C. et al. (1995)
Elimination of paternal mitochondrial DNA in
intraspecific crosses during early mouse embryogenesis. Proc. Natl Acad. Sci. USA, 92, 4542^546.
Keefe, D.L., Niven-Fairchild, T, Powell, S. and
Buradagunta, S. (1995) Mitochondrial deoxyribonucleic acid deletions in oocytes and reproductive
aging in women. Fertil Steril, 64, 577-583.
Kitagawa, T, Suganuma, N., Nawa, A. et al. (1993)
Rapid accumulation of deleted mitochondrial
deoxyribonucleic acid in postmenopausal ovaries.
Biol. Reprod., 49, 730-736.
Lang, B.F., Burger, G., O'Kelly, C.J. et al. (1997) An
ancestral mitochondrial DNA resembling a eubacterial
genome in miniature. Nature, 387, 493-497.
Lynch, M. (1996) Mutation accumulation in transfer
RNAs: molecular evidence for Muller's ratchet in
mitochondrial genomes. Mol. Biol. Evol, 13, 209-220.
Lynch, M., Burger, R., Butcher, D., and Gabriel, W.
(1993) The mutational meltdown in asexual
populations. J. Hered., 84, 339-344.
Martin, W. and Miiller, M. (1998) The hydrogen
hypothesis for the first eukaryote. Nature, 392, 37-41.
Martin, W., Stoebe, B., Goremykin, V. et al. (1998)
Gene transfer to the nucleus and the evolution of
chloroplasts. Nature, 393, 162-165.
Matsumoto, L., Kasamatsu, H., Piko, L., and Vinograd, J.
(1974) Mitochondrial DNA replication in sea urchin
oocytes. J. Cell Biol, 63, 146-159.
Matsumoto, L., Piko, L., and Vinograd, J. (1976)
Complex mitochondrial DNA in animal thyroids.
A comparative study. Biochim. Biophys. Ada, 432,
257-266.
Medvedev, Z.A. (1981) On the immortality of the germ
line: genetic and biochemical mechanisms. A review.
Mech. Ageing Develop., 17, 331-359.
Muller, H.J. (1964) The relation of recombination to
mutational advance. Mutat. Res., 1, 2-9.
Otto, S.P. and Goldstein, D.B. (1992) Recombination
and the evolution of diploidy. Genetics, 131, 745-751.
Piko, L. and Matsumoto, L. (1976) Number of
mitochondria and some properties of mitochondrial
DNA in the mouse egg. Dev. Biol, 49, 1-10.
Poulton, J., Macaulay, V., and Marchington, D.R. (1998)
Is the bottleneck cracked? Am. J. Hum. Genet., 62,
752-757.
Race, H.L., Herrmann, R.G., and Martin, W. (1999) Why
have organelles retained genomes? Trends Genet., 15,
364-370.
Rizzuto, R., Pinton, P., Carrington, W. et al. (1998)
Close contacts with the endoplasmic reticulum as
determinants of mitochondrial Ca 2+ responses.
Science, 280, 1763-1766.
Satoh, M. and Kuroiwa, T. (1991) Organization of
multiple nucleoids and DNA molecules in
mitochondria of a human cell. Exp. Cell. Res., 196,
137-140.
Shadel, G.S., and Clayton, D.A. (1997) Mitochondrial
DNA maintenance in vertebrates. Annu. Rev.
Biochem., 66, 409^35.
Shannon, C.E. and Weaver, W. (1949) The mathematical
theory of communication. University of Illinois Press,
Urbana. (Cited by Davies, P. The Fifth Miracle. The
Search for the Origin of Life. Allen Lane, Ringwood,
Victoria, 1998)
Sutovsky, P., Navara, C.S., and Schatten, G. (1996) Fate
of the sperm mitochondria, and the incorporation,
conversion, and disassembly of the sperm tail
structures during bovine fertilization. Biol. Reprod.,
55, 1195-1205.
Wallace, D.C. (1995) Mitochondrial DNA variation in
human evolution, degenerative disease, and aging.
Am. J. Hum. Genet, 57, 201-223.
Wallace, D.C, Ye, J., Neckelmann, S.N. et al. (1987)
Sequence analysis of cDNAs for the human and
bovine ATP synthase subunit: mitochondrial DNA
genes sustain seventeen times more mutations. Curr.
Genet., 12, 81-90.
R.P.S.Jansen
Wolstenholme, D.R., Clary, D.O., MacFarlane, J.L. etal.
(1985) Organization and evolution of invertebrate
mitochondrial genomes. In Quagliariello, E., Slater,
E.C., Palmieri, F. et al. (eds). Achievments and
Perspectives of Mitochondrial Research. Vol. II:
Biogenesis. Elsevier Science, Amsterdam, pp. 61-69.
Zeviani, M. and Antozzi, C. (1997) Mitochondrial
disorders. Mol. Hum. Reprod., 3, 133-148.
10