4626±4633
Nucleic Acids Research, 2002, Vol. 30 No. 21
ã 2002 Oxford University Press
Human mitochondrial DNA with large deletions
repopulates organelles faster than full-length
genomes under relaxed copy number control
Francisca Diaz1, Maria Pilar Bayona-Bafaluy1, Michele Rana1, Marialejandra Mora2,
Huiling Hao1 and Carlos T. Moraes1,2,*
1
Department of Neurology and 2Department of Cell Biology and Anatomy, University of Miami, School of Medicine,
Miami, FL 33136, USA
Received July 24, 2002; Revised August 15, 2002; Accepted September 11, 2002
ABSTRACT
Partially-deleted mitochondrial DNA (DmtDNA) accumulates during aging of postmitotic tissues. This
accumulation has been linked to decreased metabolic activity, increased reactive oxygen species
formation and the aging process. Taking advantage
of cell lines with heteroplasmic mtDNA mutations,
we showed that, after severe mtDNA depletion,
organelles are quickly and predominantly repopulated with DmtDNA, whereas repopulation with the
wild-type counterpart is slower. This behavior was
not observed for full-length genomes with pathogenic point mutations. The faster repopulation of
smaller molecules was supported by metabolic
labeling of mtDNA with [3H]thymidine during relaxed
copy number control conditions. We also showed
that hybrid cells containing two defective mtDNA
haplotypes tend to retain the smaller one as they
adjust their normal mtDNA copy number. Taken
together, our results indicate that, under relaxed
copy number control, DmtDNAs repopulate mitochondria more ef®ciently than full-length genomes.
INTRODUCTION
The human mitochondrial genome is a 16 569 bp circular
molecule, containing genes that are necessary for the synthesis
of the catalytic components of the oxidative phosphorylation
(OXPHOS) system. Although these components are essential
for mitochondrial respiration and ATP production, they are
intrinsically dependent on factors encoded by the nuclear
DNA, synthesized in cytosolic ribosomes and imported into
the mitochondria. These include factors that regulate mitochondrial DNA (mtDNA) replication, such as the mtDNA
polymerase g. Although several additional protein factors
associated with mtDNA replication have been described, it is
still unclear how mtDNA replication and copy number are
controlled (1,2).
Mitochondrial genomes harboring large deletions are
known to accumulate both in patients with heteroplasmic
mtDNA mutations and in normal individuals during aging,
particularly in postmitotic tissues such as muscle and brain (3).
These observations support the mitochondrial theory of
aging, which states that the slow accumulation of impaired
mitochondria is the driving force of the aging process. This
idea is attractive because it can be reconciled with the free
radical theory of aging, which argues that oxidative damage
plays a key role in senescence. Among the numerous
mechanisms known to generate oxidants, leakage of superoxide anion and hydrogen peroxide from the mitochondrial
electron transport chain are the chief candidates. Increased
damage to mtDNA could exacerbate this leakage of reactive
oxygen species (ROS) (4).
It is not known how mtDNA deletions accumulate during
aging. Although the smaller size of partially-deleted molecules suggested early on that they could have a replicative
advantage (5,6), direct evidence of this phenomenon has been
lacking. In most cases, partially-deleted mtDNAs (DmtDNAs)
contain the same number of replication origins as the wildtype genome but they can be up to 50% shorter (7). We have
previously shown that cells harboring homoplasmic levels of
DmtDNA repopulated their organelles with mtDNA faster
than cells containing wild-type mitochondrial genomes (8). In
these cells, however, there was no competition between
mutated and wild-type genomes, as they were present in a
homoplasmic state. Therefore, we could not rule out that
differences in mtDNA repopulation were due to different
metabolic states of these cells. In the present study, we
addressed this issue by studying heteroplasmic cells. Our
results showed that mtDNA with large deletions, but not with
pathogenic point mutations, repopulates organelles signi®cantly faster than wild-type genomes in the same cell,
particularly during relaxed copy number control.
*To whom correspondence should be addressed at Department of Neurology, University of Miami, 1095 NW 14th Terrace, Miami, FL 33136, USA.
Tel: +1 305 243 5858; Fax: +1 305 243 3914; Email: [email protected]
Present address:
Michele Rana, Azienda Ospedaliera Santa Maria della Misericordia. U.O. di Neurologia, Udine 33100, Italy
The authors wish it to be known that, in their opinion, the ®rst two authors should be regarded as joint First Authors
Nucleic Acids Research, 2002, Vol. 30 No. 21
MATERIALS AND METHODS
Characterization of human cells containing different
pathogenic mtDNA mutations
We have used several characterized cell lines containing the
osteosarcoma 143B(TK±) nuclear background and various
mtDNAs. Cell lines harboring heteroplasmic mtDNA deletions were obtained by fusing 143B/206r° with: (i) an
enucleated HeLa/®broblast hybrid line harboring ~50% of
the 4.9 kb `common deletion' [lines DBH.10.5.n, spanning
mtDNA positions 8482±13 459; (9)]; (ii) a patient-derived
®broblast cell line harboring a 7.5 kb partial deletion in the
mtDNA was used to generate transmitochondrial cybrids
(DHH6.1 and DH2.1, spanning positions 7982±15 504). The
cell line D16.10.40 had the same deletion in a homoplasmic
state (8,10). Two cell lines harboring a heteroplasmic mtDNA
point mutation in the tRNAAsn gene (G5703A, lines W91 and
W76) and two cell lines harboring a heteroplasmic mtDNA
point mutation in the tRNALeu(UUR) gene (C3256T, lines
SUA72 and SUB43) were previously characterized (11,12). A
cell line harboring a homoplasmic 4 bp apocytochrome b gene
deletion was obtained as previously described (13).
mtDNA depletion and repopulation analysis
The various cell lines were grown in complete medium
(DMEM supplemented with 4.5 mg/ml glucose, 50 mg/ml
uridine, 10% FCS, 1 mM pyruvate and 10 mg/ml gentamycin)
in the presence of 50 ng/ml ethidium bromide (EtBr). EtBr
was kept in the media for 15 days, after which cells were fed
with complete media without EtBr. Cells were kept between
30 and 80% con¯uence with excess fresh medium to assure
exponential growth. At selected days (0, 6, 15, 22, 30 and 45)
cells were harvested for DNA extraction.
For each cell line (obtained from all selected treatment
days) DNA samples (~5 mg) were digested with PvuII and
analyzed by Southern blot using two probes. The ®rst probe
corresponded to a non-deleted region (mtDNA position 3305±
4261), whereas the second probe corresponded to the nuclear
18S rDNA gene (14). All probes were prepared by the random
primer method. Hybridizing fragments were quantitated in a
Cyclone Phosphoimager (Packard Instruments). Signals were
used to determine the ratios between mtDNA/nDNA corresponding to the three concentrations. The intensity of the signal
was normalized to day zero in the repopulation analyses.
mtDNA synthesis analysis
Selected hybrid lines were labeled with [3H]thymidine
(20 mCi/ml; Perkin Elmer) for the times indicated. Mitochondria were puri®ed by nitrogen cavitation as described (15)
and DNA puri®ed by phenol/chlorophorm extraction. Puri®ed
mtDNA was digested with PvuII and separated by electrophoresis on a 0.8% agarose gel. The gel was stained with EtBr
and the fragments corresponding to mtDNA were visualized
under UV light. Approximately 0.2 cm segments were sliced
from the agarose lane and dissolved with 200 ml of 7 M
NaClO4 for 30 min at 50°C. The solubilized agarose slice was
added to scintillation liquid (10 ml) and c.p.m. measured in a
scintillation counter. Cells were either directly labeled in
medium containing 10% dialyzed serum or pre-treated with
50 ng/ml of EtBr in complete medium for 3 days, allowed to
4627
recover for 24 h and pulse labeled for the times indicated.
When necessary, the label was chased for 4 additional days in
complete medium. The incorporated c.p.m. were normalized
by the number of thymidines in each molecule and a
correction factor (1.87) was applied to the DmtDNA counts
before the percentage labeled mutated mtDNA was calculated.
mtDNA haplotype analysis
Total DNA from different cell lines was puri®ed as described
(14). For the differentiation between partially-deleted or
partially-duplicated mtDNA, Southern blots of PvuII and
SnaBI digests were performed as described above, with the
exception that the probe outside the deleted region corresponded to positions 140±585. The analyses of point mutations
in the tRNALeu(UUR) and tRNAAsn genes were performed by
`last cycle hot' PCR as previously described (11,12). For the
4 bp deletion in the apocytochrome b gene (D4bp cyt b)
analysis, ~250 ng of DNA were used to amplify a 93 bp
fragment by PCR. We used oligonucleotide primers corresponding to mtDNA position 14 747±14 770 (F) and 14 840±
14 817 (B). The forward primer was end-labeled with
[g-32P]ATP. Labeled products were separated by electrophoresis on a 6% denaturing polyacrylamide gel, dried and
exposed to a phosphoimager screen. Radiolabeled fragments
were identi®ed and quantitated in a Cyclone (Packard
Instruments) phosphoimager storage system (13).
Establishment and genetic characterization of hybrid
lines
The cell lines 143B, D16.10.40 and D4cytb3.E were
transfected with the respective plasmids pIRESneo, pZEO
and pIRESpuro, and stable transfectants isolated by selection
with the respective drugs. Transfection was performed with
lipofectamine as suggested by the manufacturer (Life
Technologies). These cells were fused to each other with
polyethylene glycol as described (9). Fusion products were
subjected to double selection for the nuclear markers.
Individual hybrid clones were isolated by the cloning ring
method, expanded and their mtDNA analyzed for (i) the D4cyt
b by PAGE of PCR-ampli®ed fragments as described above,
(ii) large deletions by Southern blot also as described above.
RESULTS
Observations that mtDNA with large-scale deletions accumulate with age prompted us to study the relative organelle
repopulation rate of such molecules. To better understand this
process we analyzed several cell lines containing different
heteroplasmic mtDNA mutations but having a similar nuclear
background, derived from the human osteosarcoma
143B(TK±) cell line. These cell lines were obtained by fusing
the mtDNA-less 143B derivative (206r°) to cytoplasts
obtained from ®broblasts of patients with mitochondrial
disorders. Figure 1 shows the location of the mutations in
the mitochondrial genome and their level of heteroplasmy. We
investigated for the presence of deletion dimers or partial
duplications in the cell lines described above by a series of
Southern blots. PvuII and SnaBI digests using probes either
inside or outside the deletion region showed that the cell lines
used in these experiments contained essentially only the
wild-type mtDNA and deletion monomers (not shown).
4628
Nucleic Acids Research, 2002, Vol. 30 No. 21
Figure 1. Characterization of cell lines with mtDNA mutations used in this
study. (A) Diagram illustrating the location of two point mutations, two
large deletions and a 4 bp deletion in the apocytochrome b gene in the
human mtDNA. (B) Southern blot showing the heteroplasmic mtDNA deletions in some of the cell lines used. Total DNA extracted from the different
cell lines was digested with PvuII and analyzed by Southern hybridization
to a probe corresponding to the ND1 gene. (C) PCR/RFLP analyses of the
two pathogenic point mutations and PCR/LP of the apocytochrome b 4 bp
deletion (A). The cell lines with tRNA mutations used in this study were
heteroplasmic. Cell lines harboring homoplasmic levels of apocytochrome b
4 bp deletion, a 7.5 kb deletion and wild-type genomes were used in the hybrid experiments described below.
mtDNA harboring deletions repopulate mitochondria
faster than the full-length counterparts in the same cell
EtBr has been used extensively to reduce the mtDNA copy
number in proliferating cells (16). We treated osteosarcoma
cell lines with 50 ng/ml EtBr for 15 days, after which the drug
was removed from the medium and the cells allowed to grow
for 30 additional days. Throughout the experiment, the
medium was supplemented with pyruvate and uridine to
avoid a strong selection against cells defective in respiration,
which can be auxotrophic for these substrates (17). DNA was
obtained at different days and analyzed for the ratio mtDNA/
nuclear gene. Figure 2A shows that the EtBr treatment led to a
marked depletion of mtDNA followed by a repopulation
phase.
Selected cell lines were divided into two groups, one
containing heteroplasmic mtDNA large deletions and one
containing heteroplasmic mtDNA point mutations. In all cases
they were treated with EtBr for 15 days at which point the drug
was removed and the cells allowed to grow for 30 additional
days. Parallel cultures that were not treated with EtBr were
also analyzed to assess ¯uctuations in mtDNA heteroplasmy
that were independent of EtBr treatment. The EtBr-induced
mtDNA depletion was con®rmed for all the treated samples by
Southern blot (not shown). The percentages of mutated and
wild-type mtDNA were not calculated during days 6 and 15 of
EtBr treatment because mtDNA levels were markedly
decreased, affecting the reliability of the Southern blot assay.
In the heteroplasmic mtDNA large deletions group,
untreated cells showed a small drift towards lower levels of
mtDNA mutations (Fig. 2B±D, open triangles). After 45 days
in culture (under non-selective conditions) the percentage of
mutated mtDNA decreased by 10±20% depending on the cell
line. In the EtBr-treated group, there was an increase in the
mutated mtDNA population immediately following the
removal of EtBr from the medium (Fig. 2B±D, solid triangles).
Seven days after EtBr removal, the percentage of DmtDNA
increased by 6.2 6 4.8% (mean 6 SD) whereas during the
same time in the untreated group the percentage of DmtDNA
decreased by 9.1 6 8.8%. Therefore the mean difference in
the percent mutated mtDNA between treated and untreated,
7 days after EtBr withdrawal was 15.3% (P = 0.011). The
increase in percent of DmtDNA observed in the EtBr treated
cells was followed by a drift towards slightly lower levels,
con®rming the trend observed in untreated cells (Fig. 2B±D).
In contrast to the mtDNA large deletions group, increases in
the percentage of mutated mtDNA after EtBr treatment/
withdrawal could not be observed in cells heteroplasmic for
pathogenic point mutations (Fig. 3). Overall, there was also a
trend for a small reduction in the percentage of point mutated
mtDNA with time in culture without treatment. EtBr treatment/withdrawal was not associated with an increase in the
percentage of mutated mtDNA. With exception of the cell line
SUB43, the other three cell lines showed a decrease in the
percentage of mutated mtDNA following the EtBr treatment/
withdrawal (Fig. 3).
mtDNA harboring large deletions are preferentially
replicated under relaxed copy number control
We measured the rate of [3H]thymidine incorporation
into cell line DDH2.1, which harbors a heteroplasmic
7.5 kb mtDNA deletion. Cells were pulse-labeled with
[3H]thymidine for 20 h both without EtBr and after 3 days
EtBr (50 ng/ml) treatment. The EtBr treatment was shorter
than in other experiments because a relatively high level of
mtDNA was necessary to obtain reliable results with this
method. mtDNA puri®ed from labeled cells was digested
with PvuII and fractionated by agarose gel electrophoresis.
Nucleic Acids Research, 2002, Vol. 30 No. 21
4629
Figure 2. DmtDNA heteroplasmy ¯uctuations after induced mtDNA depletion. (A) An osteosarcoma cell line containing high levels of a DmtDNA was treated
with 50 ng/ml EtBr for 15 days and allowed to recover for 30 additional days. Cells were harvested at the indicated times and their DNA puri®ed and
analyzed for relative levels of mtDNA (mutated or wild-type, closed and open circles, respectively) / nDNA as described in Materials and Methods.
(B±D) Three cell lines harboring DmtDNA were subjected to a 15 day EtBr treatment followed by a 30 day recovery period. Parallel cultures growing in the
same medium, but without EtBr were also analyzed to assess heteroplasmy ¯uctuations related to time in culture. Total DNA extracted at different times was
analyzed by Southern blot and the percentage mutated mtDNA represented as ®lled triangles in the EtBr treated series and as open triangles in the control
series. The mtDNA levels during EtBr treatment (days 7 and 15) were very low and did not allow a reliable estimation of heteroplasmy. There was a reduction in the percentage mutated mtDNA during normal growth conditions, but EtBr treatment was consistently associated with an increase in percentage
mutated mtDNA immediately after the treatment.
Figure 4A shows that the relative rate of [3H]thymidine
incorporation was signi®cantly increased in the partiallydeleted molecule after EtBr treatment. Bars on Figure 4B
represent the percentage DmtDNA calculated from the
[3H]thymidine-labeled fraction after correction for the number
of thymidines (mutated molecule values were multiplied by
1.87). Twenty hours labeling followed by a 4 day chase
showed a 3H-labeled fraction of DmtDNA that was similar to
the one obtained by Southern blot data (i.e. ~89±92% deleted,
horizontal grey bar on Fig. 4B). The EtBr pre-treatment led to
an over-representation of 3H-labeled fraction of DmtDNA
(96% of total, Fig. 4B).
mtDNA harboring large deletions are preferentially
maintained in hybrids with competing mitochondrial
genomes
In order to study mtDNA heteroplasmic dynamics in an
independent system, we created hybrids containing different
types of mitochondrial genomes and assessed if there were
speci®c haplotypes that were preferentially maintained. To
do so, we used three cell lines that had 143B-derived
nuclear background, but each one tagged with a different
dominant nuclear marker, namely neomycin-, zeocin- or
puromycin-resistance. Figure 5A illustrates the procedure and
describes the following cell lines: (i) a neo-resistant cell line
with wild-type mtDNA; (ii) a zeo-resistant cell line with a
large mtDNA deletion (D7.5 kb mtDNA deletion, D16.10.40);
(iii) a puro-resistant cell line with a small 4 bp deletion in the
apocytochrome b gene. In contrast to the cell lines used in the
previous experiments, these cell lines were homoplasmic for
the respective mtDNA haplotypes. Cell fusions were performed and hybrid clones were isolated by double selection
according to the nuclear markers present in their nuclei. All
media were supplemented with pyruvate and uridine. The
expanded clones had their DNA extracted and characterized as
described below.
To differentiate mtDNA haplotypes in PN hybrids [i.e. puro
(harboring D4 bp) and neo (harboring wt mtDNA)], we
ampli®ed a 93 bp fragment ¯anking the 4 bp deletion and
analyzed the amplicons by PAGE. Surprisingly, PN clones
were homoplasmic for the wild-type mtDNA, losing the D4 bp
mtDNA (Fig. 5B). As expected, ampli®cations from ZP clones
[i.e. zeo (harboring mtDNA with a D7.5 kb deletion) and puro
(harboring D4 bp)] gave rise exclusively to the D4 bp cyt b in
4630
Nucleic Acids Research, 2002, Vol. 30 No. 21
Figure 3. Point mutant mtDNA heteroplasmy ¯uctuations after induced mtDNA depletion. Four cell lines harboring two different pathogenic mtDNA point
mutations in tRNA genes were subjected to a 15 day EtBr treatment followed by a 30 day recovery period. Parallel cultures growing in the same medium, but
without EtBr were performed to assess heteroplasmy ¯uctuations during time in culture. Total DNA was extracted at different times and analyzed by
PCR/RFLP. The percentage mutated mtDNA is represented as ®lled circles in the EtBr treated series and as open circles in the control series (A±D). With the
exception of one cell line (B), the percentage of point mutated mtDNA either did not change signi®cantly or decreased from pretreatment levels. The
treatment with EtBr did not seem to alter this natural trend.
this test, as the large deletion present in the zeo-resistant cell
line spanned the PCR target (i.e. the D4 bp cyt b region;
Fig. 5B). mtDNA with a large 7.5 kb deletion were
distinguished from the wild type and the D4 bp cyt b by
Southern analysis. With the exception of one clone (NZ1) NZ
clones [i.e. neo (harboring wt mtDNA) and zeo (harboring
D7.5 kb mtDNA)] showed a higher percentage of wild type
than DmtDNA. In contrast, ZP clones [i.e. zeo (harboring D7.5
kb mtDNA) and puro (harboring D4 bp)] had consistently high
levels of the mitochondrial genome harboring a large deletion
(Fig. 5C). Taken together, these results suggest that the hybrid
lines tended to preserve the functional mitochondrial genome,
probably because of better growth performance. In the
cases where both mitochondrial genomes were defective,
the smaller one prevailed.
DISCUSSION
mtDNA heteroplasmy under tight copy number control
A preferential maintenance of DmtDNA was not observed
during normal cell growth conditions (Fig. 2) (18). In
heteroplasmic cells harboring DmtDNA, the fraction of
partially-deleted genomes actually showed a small decrease
(10±20%) with time under non-selective culture conditions.
This reduction was probably observed because cells with
relatively higher levels of the wild-type mtDNA have slightly
better growth performance, even in rich medium (i.e. media
supplemented with high glucose, pyruvate and uridine).
Osteosarcoma cybrids with homoplasmic levels of DmtDNA
have been shown to grow slightly more slowly than their wildtype counterparts (18). Hayashi et al. showed that HeLa
cybrids with different levels of DmtDNA also had slightly
slower population doubling times, with the slowest being the
cell line with the highest percentage of mutated mtDNA (19).
Tang et al. (18) found that the DmtDNA fraction decreased by
only ~10% during prolonged culture, and found it to be not
signi®cant. Hayashi et al. found that, after producing cybrids,
there was a bias towards an increase in the DmtDNA
population after which the levels `stabilized'. However,
examination of their data showed a consistent small reduction
(5±15%) in the percentage of DmtDNA after 10 weeks
post-fusion (19). Differences in nuclear background may
affect the balance between mutated and wild-type mtDNA in
non-selective conditions.
mtDNA heteroplasmy under relaxed copy number
control
Treatment with EtBr, which reduced the mtDNA copies to
very low levels, had a signi®cant effect on the mtDNA
repopulation rates within organelles. When EtBr was removed
from the media, mtDNA replication ensued at possibly
Nucleic Acids Research, 2002, Vol. 30 No. 21
Figure 4. Metabolic labeling of mtDNA with [3H]thymidine. The cell line
DDH2.1 was incubated with 20 mCi/ml of [3H]thymidine for the times
indicated. In one group, cells were pretreated with EtBr for 3 days, allowed
to recover without the drug for 24 h and labeled with [3H]thymidine for the
periods indicated. mtDNA from labeled cells was puri®ed from mitochondria isolated by N2 cavitation as described in Materials and Methods.
mtDNA was digested with PvuII and separated by electrophoresis on a
0.8% agarose gel. The gel lane was sliced and the gel fractions analyzed in
a scintillation counter (A). Percentage mutated mtDNA was calculated from
the incorporated 3H after correction for the number of thymidines in each
molecule. (B) Comparison of these values with the percentage mutation
obtained from Southern blots (grey horizontal area).
maximum speed, as the still poorly characterized copy number
control mechanism attempted to repopulate organelles with the
normal copy number or mtDNA mass (20). Under these
conditions of fast replicative activity, DmtDNAs fared signi®cantly better, repopulating organelles ~2±3 times faster than
their wild-type counterparts. The present results suggest that
increased repopulation rates were not due to an OXPHOS
response, as the mutated mtDNA harboring point mutations
(but still having an OXPHOS defect) did not show changes
in repopulation rates relative to the wild-type mtDNA.
Differences in cellular growth were also minimized by the
fact that mutated and wild-type genomes were present in the
same cell. [3H]Thymidine incorporation was signi®cantly
increased in the partially-deleted genomes after mtDNA copy
number reduction, corroborating the concept that fast replicative rates favor smaller molecules. These results are compatible with a previous report showing that the rates of
[3H]thymidine incorporation were similar between partiallydeleted and wild-type mtDNA when cells were labeled without
previous mtDNA depletion (21). The replicative advantage of
4631
Figure 5. Heteroplasmy levels of somatic hybrids generated without selection for OXPHOS function. Three different cell lines were fused to each
other as depicted in (A). One contained wild-type mtDNA and a neomycinresistance nuclear marker (N), a second was homoplasmic for a mtDNA
7.5 kb deletion containing a zeocin-resistance nuclear marker (Z) and a
third was homoplasmic for a pathogenic 4 bp deletion in cytochrome b gene
containing a puromycin-resistance nuclear marker (P). Fusion products were
grown in high glucose media containing two of the respective selection
drugs (i.e. G418, zeocin or puromycin) and supplemented with uridine.
Surviving clones were isolated and their DNA analyzed by Southern blot as
described in Materials and Methods. (B) Analysis of the cytochrome b 4 bp
deletion in hybrids of PN cell lines. A 32P-labeled PCR amplicon, corresponding to the 5¢ end of the apocytochrome b gene, was separated by
electrophoresis on a 6% denaturing polyacrylamide gel and analyzed in a
phosphoimager. (C) Phosphoimager signal from a Southern blot of DNA
extracted from the hybrids between cells containing the 7.5 kb deletion and
wt (NZ lines) or the 4 bp deletion in apocytochrome b (ZP lines).
smaller mitochondrial genomes may be masked by the slow
replication rates in copy number-controlled cells as well as by
the cellular selection observed in proliferating cells.
Our observation that replication between the two mtDNA
haplotypes was not synchronized suggests that, even though
human mitochondrial genomes are associated with proteins
and membranes (22), possibly resembling the yeast nucleoids
(23), they do not appear to behave as a group of molecules
segregating as a unit as previously suggested (24). Therefore,
in a putative mammalian nucleoid, individual molecules have
the potential to replicate independently from others. The
nucleoid concept described above (24) was proposed to
explain mtDNA heteroplasmy stability. However, our results
show that heteroplasmy is not stable during fast mtDNA
replication.
4632
Nucleic Acids Research, 2002, Vol. 30 No. 21
mtDNA heteroplasmy under copy number re-setting
The proliferating hybrid results suggest that when a double
dose of mtDNA is introduced in cells, they tend to maintain
functional mitochondrial genomes. This conclusion is based
on the observation that wild-type genomes were preferentially
maintained in hybrids of 143B and either D16.10.40 or D4cyt
b. In contrast, in hybrids between D16.10.40 and D4cyt b, the
mtDNA with the largest deletion was preferentially maintained. It is unclear why the wild-type mitochondrial genomes
were retained at such high relative levels. In particular, the
absence of genomes with the cyt b 4 bp deletion in the PN
(wt + D4cyt b) clones was surprising. Increased ROS
previously associated with this cyt b mutation may make
this mtDNA haplotype particularly unfavorable to organelles
or cells during the resetting of mtDNA copy number (13).
Factors in¯uencing mtDNA repopulation rates
It is unlikely that the processivity of the DNA polymerase g
differs between mutated and wild-type mtDNA. A `normal'
replication rate has been reported for mtDNA with point
mutations (25), and it is likely to be unchanged in partiallydeleted mitochondrial genomes as well. It is more likely that
the increased repopulation rate is due to faster completion of
the relatively slow mtDNA replication process (26).
Seidel-Rogol and Shadel recently showed that during the
EtBr treatment recovery phase, mtTFA (which is known to be
unstable in the absence of mtDNA) and mtRNA polymerase
increased to normal levels at a signi®cantly slower rate than
mtDNA (27). Transcripts for these genes have been previously
shown to be unaffected by mtDNA depletion (8). It is likely
that such replication factors will be limiting during the
repopulation phase and the small amounts of replication
factors available after depletion of mtDNA can be re-engaged
only after a mtDNA molecule ®nishes its replication cycle.
Therefore, replication factors would be available to replicate
another molecule sooner when the target molecule is smaller.
Insights into DmtDNA accumulation during aging
The results described in this report provide direct evidence
that under conditions of relaxed copy number control,
DmtDNAs have an advantage over wild-type molecules in
the same cell. This phenomenon, previously suggested based
on indirect evidence (28,29), has important implications for
the mechanisms of mtDNA deletions accumulation during the
aging and disease processes.
Although no consensus has been reached, it is generally
believed that accumulation of mtDNA deletions is more
prevalent than accumulations of mtDNA point mutations
during aging of postmitotic tissues (30). Recent models
showed that relaxed replication of mtDNA alone could lead,
through random genetic drift, to the clonal expansion of single
mutant events during human life (31). This mechanism could
lead to the accumulation or disappearance of a speci®c
mtDNA haplotype in some cases, but in our view, genetic drift
does not explain the consistent increase of DmtDNA nor
excludes the contributing effect of faster repopulation of
organelles with smaller genomes during active replication.
There is also evidence that mtDNA with partial duplications
may accumulate with age (32±34). The mechanism for the
accumulation of partial duplications is unclear and cannot be
explained by our results. Although mtDNA molecules with
partial duplications have additional replication origins that
could compensate for the increased length, earlier work
showed that mtDNA dimers replicate slower than monomers
(35).
Although there may be important differences between
postmitotic tissues and our culture cell system, the observation
of heteroplasmy ¯uctuations during rapid mtDNA repopulation allows us to draw some conclusions regarding the
molecular aspect of differential repopulation rates. Our results
are in agreement with previous in situ hybridization experiments that showed that most age-related mtDNA deletions in
muscle are caused by clonal expansion of deletions (36,37). In
muscle, mitochondria with defective function are stimulated
to proliferate, and that may increase mtDNA replication,
mimicking a relaxed copy number control situation. It also
strengthened the view that age-related mtDNA deletions are
probably generated at random but their levels gradually
increase with time. Our results also raise the possibility that
the accumulation of DmtDNAs may be accelerated by
metabolic or environmental changes leading to either a
transient reduction in mtDNA levels or a relaxation in copy
number control.
CONCLUSION
In summary, our results provide evidence that under proliferating conditions, cells harboring relatively high levels of
DmtDNAs have a slight reduction in the mutated fraction,
probably because of better growth of cells with lower mtDNA
mutation loads. This is compatible with the relatively low
percentage of mutated mtDNA in proliferating peripheral
blood cells and ®broblasts of patients with mitochondrial
diseases (38). On the other hand, at the molecular level,
smaller mtDNAs containing the major two replication origins
have a repopulating advantage over the wild-type counterpart
that operates most ef®ciently during relaxed copy number
control. This situation parallels the accumulation of largescale mtDNA deletions in postmitotic tissues, where selection
based on cellular growth or survival does not take place, and
abnormal organelle proliferation would lead to an increase in
mtDNA replication rates (36,37).
ACKNOWLEDGEMENTS
This work was supported by National Institutes of Health
Grants EY10804 and GM55766 and by the Muscular
Dystrophy Association.
REFERENCES
1. Shadel,G.S. and Clayton,D.A. (1997) Mitochondrial DNA maintenance
in vertebrates. Annu. Rev. Biochem., 66, 409±435.
2. Moraes,C.T. (2001) What regulates mitochondrial DNA copy number in
animal cells? Trends Genet., 17, 199±205.
3. Wallace,D.C. (1997) Mitochondrial DNA in aging and disease. Sci. Am.,
277, 40±47.
4. Raha,S. and Robinson,B.H. (2000) Mitochondria, oxygen free radicals,
disease and ageing. Trends Biochem. Sci., 25, 502±508.
5. Corral-Debrinski,M., Horton,T., Lott,M.T., Shoffner,J.M., Beal,M.F. and
Wallace,D.C. (1992) Mitochondrial DNA deletions in human brain:
regional variability and increase with advanced age. Nature Genet., 2,
324±329.
Nucleic Acids Research, 2002, Vol. 30 No. 21
6. Cortopassi,G.A., Shibata,D., Soong,N.W. and Arnheim,N. (1992) A
pattern of accumulation of a somatic deletion of mitochondrial DNA in
aging human tissues. Proc. Natl Acad. Sci. USA, 89, 7370±7374.
7. Mita,S., Rizzuto,R., Moraes,C.T., Shanske,S., Arnaudo,E., Fabrizi,G.M.,
Koga,Y., DiMauro,S. and Schon,E.A. (1990) Recombination via ¯anking
direct repeats is a major cause of large-scale deletions of human
mitochondrial DNA. Nucleic Acids Res., 18, 561±567.
8. Moraes,C.T., Kenyon,L. and Hao,H. (1999) Mechanisms of human
mitochondrial DNA maintenance: the determining role of primary
sequence and length over function. Mol. Biol. Cell, 10, 3345±3356.
9. Sancho,S., Moraes,C.T., Tanji,K. and Miranda,A.F. (1992) Structural and
functional mitochondrial abnormalities associated with high levels of
partially deleted mitochondrial DNAs in somatic cell hybrids. Somat.
Cell Mol. Genet., 18, 431±442.
10. King,M.P. (1996) Use of ethidium bromide to manipulate ratio of
mutated and wild-type mitochondrial DNA in cultured cells.
Methods Enzymol., 264, 339±344.
11. Hao,H. and Moraes,C.T. (1996) Functional and molecular mitochondrial
abnormalities associated with a C ® T transition at position 3256 of the
human mitochondrial genome. The effects of a pathogenic mitochondrial
tRNA point mutation in organelle translation and RNA processing.
J. Biol. Chem., 271, 2347±2352.
12. Hao,H. and Moraes,C.T. (1997) A disease-associated G5703A mutation
in human mitochondrial DNA causes a conformational change and a
marked decrease in steady-state levels of mitochondrial tRNA(Asn).
Mol. Cell. Biol., 17, 6831±6837.
13. Rana,M., de Coo,I., Diaz,F., Smeets,H. and Moraes,C.T. (2000) An outof-frame cytochrome b gene deletion from a patient with parkinsonism is
associated with impaired complex III assembly and an increase in free
radical production. Ann. Neurol., 48, 774±781.
14. Moraes,C.T., Shanske,S., Tritschler,H.J., Aprille,J.R. Andreetta,F.,
Bonilla,E., Schon,E.A. and DiMauro,S. (1991) mtDNA depletion with
variable tissue expression: a novel genetic abnormality in mitochondrial
diseases. Am. J. Hum. Genet., 48, 492±501.
15. Gottlieb,R.A. and Granville,D.J. (2002) Analyzing mitochondrial
changes during apoptosis. Methods, 26, 341±347.
16. Desjardins,P., de Muys,J.M. and Morais,R. (1986) An established avian
®broblast cell line without mitochondrial DNA. Somat. Cell Mol. Genet.,
12, 133±139.
17. King,M.P. and Attardi,G. (1989) Human cells lacking mtDNA:
repopulation with exogenous mitochondria by complementation. Science,
246, 500±503.
18. Tang,Y., Manfredi,G., Hirano,M. and Schon,E.A. (2000) Maintenance of
human rearranged mitochondrial DNAs in long-term cultured
transmitochondrial cell lines. Mol. Biol. Cell., 11, 2349±2358.
19. Hayashi,J., Ohta,S., Kikuchi,A., Takemitsu,M., Goto,Y. and Nonaka,I.
(1991) Introduction of disease-related mitochondrial DNA deletions into
HeLa cells lacking mitochondrial DNA results in mitochondrial
dysfunction. Proc. Natl Acad. Sci. USA, 88, 10614±10618.
20. Tang,Y., Schon,E.A., Wilichowski,E., Vazquez-Memije,M.E.,
Davidson,E. and King,M.P. (2000) Rearrangements of human
mitochondrial DNA (mtDNA): new insights into the regulation of
mtDNA copy number and gene expression. Mol. Biol. Cell, 11,
1471±1485.
21. Moraes,C.T. and Schon,E.A. (1995) In Palmieri,F., Papa,S., Saccone,C.
and Gadaleta,M.N. (eds), Progress in Cell Research: Symposium on
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
4633
`Thirty years of Progress in Mitochondrial Bioenergetics and Molecular
Biology'. Elsevier, Vol. 5, pp. 209±215.
Albring,M., Grif®th,J. and Attardi,G. (1977) Association of a protein
structure of probable membrane derivation with HeLa cell mitochondrial
DNA near its origin of replication. Proc. Natl Acad. Sci. USA, 74,
1348±1352.
Miyakawa,I., Sando,N., Kawano,S., Nakamura,S. and Kuroiwa,T. (1987)
Isolation of morphologically intact mitochondrial nucleoids from the
yeast, Saccharomyces cerevisiae. J. Cell Sci., 88, 431±439.
Jacobs,H.T., Lehtinen,S.K. and Spelbrink,J.N. (2000) No sex please,
we're mitochondria: a hypothesis on the somatic unit of inheritance of
mammalian mtDNA. Bioessays, 22, 564±572.
Emmerson,C.F., Brown,G.K. and Poulton,J. (2001) Synthesis of
mitochondrial DNA in permeabilised human cultured cells. Nucleic
Acids Res., 29, e1.
Clayton,D.A. (1982) Replication of animal mitochondrial DNA. Cell, 28,
693±705.
Seidel-Rogol,B.L. and Shadel,G.S. (2002) Modulation of mitochondrial
transcription in response to mtDNA depletion and repletion in HeLa
cells. Nucleic Acids Res., 30, 1929±1934.
Chen,X., Prosser,R., Simonetti,S., Sadlock,J., Jagiello,G. and Schon,E.A.
(1995) Rearranged mitochondrial genomes are present in human oocytes.
Am. J. Hum. Genet., 57, 239±247.
Larsson,N.G., Holme,E., Kristiansson,B., Oldfors,A. and Tulinius,M.
(1990) Progressive increase of the mutated mitochondrial DNA fraction
in Kearns-Sayre syndrome. Pediatr. Res., 28, 131±136.
Pallotti,F., Chen,X., Bonilla,E. and Schon,E.A. (1996) Evidence that
speci®c mtDNA point mutations may not accumulate in skeletal muscle
during normal human aging. Am. J. Hum. Genet., 59, 591±602.
Elson,J.L., Samuels,D.C., Turnbull,D.M. and Chinnery,P.F. (2001)
Random intracellular drift explains the clonal expansion of mitochondrial
DNA mutations with age. Am. J. Hum. Genet., 68, 802±806.
Bodyak,N.D., Nekhaeva,E., Wei,J.Y. and Khrapko,K. (2001)
Quanti®cation and sequencing of somatic deleted mtDNA in single cells:
evidence for partially duplicated mtDNA in aged human tissues.
Hum. Mol. Genet., 10, 17±24.
Kajander,O.A., Rovio,A.T., Majamaa,K., Poulton,J., Spelbrink,J.N.,
Holt,I.J., Karhunen,P.J. and Jacobs,H.T. (2000) Human mtDNA
sublimons resemble rearranged mitochondrial genoms found in
pathological states. Hum. Mol. Genet., 9, 2821±2835.
Moore,C.A., Gudikote,J. and Van Tuyle,G.C. (1998) Mitochondrial DNA
rearrangements, including partial duplications, occur in young and old rat
tissues. Mutat. Res., 421, 205±217.
Bogenhagen,D., Lowell,C. and Clayton,D.A. (1981) Mechanism of
mitochondrial DNA replication in mouse L-cells. Replication of
unicircular dimer molecules. J. Mol. Biol., 148, 77±93.
Johnston,W., Karpati,G., Carpenter,S., Arnold,D. and Shoubridge,E.A.
(1995) Late-onset mitochondrial myopathy. Ann. Neurol., 37, 16±23.
Moslemi,A.R., Melberg,A., Holme,E. and Oldfors,A. (1996) Clonal
expansion of mitochondrial DNA with multiple deletions in autosomal
dominant progressive external ophthalmoplegia. Ann. Neurol., 40,
707±713.
Moraes,C.T., DiMauro,S., Zeviani,M., Lombes,A., Shanske,S.,
Miranda,A.F., Nakase,H., Bonilla,E., Werneck,L.C., Servidei,S. et al.
(1989) Mitochondrial DNA deletions in progressive external
ophthalmoplegia and Kearns-Sayre syndrome. N. Engl. J. Med., 320,
1293±1299.