Báo cáo khoa học: A truncated form of DNA topoisomerase IIb associates with the mtDNA genome in mammalian mitochondria doc

A truncated form of DNA topoisomerase IIb associates
with the mtDNA genome in mammalian mitochondria
Robert L. Low
1
, Shayla Orton
1
and David B. Friedman
2
1
Department of Pathology and
2
Department of Cellular and Structural Biology, University of Colorado Health Sciences Center,
Denver, CO, USA
Despite the likely requirement for a DNA topoisomerase II
activity during synthesis of mitochondrial DNA in mam-
mals, this activity has been very difficult to identify convin-
cingly. The only DNA topoisomerase II activity conclusively
demonstrated to be mitochondrial in origin is that of a type
II activity found associated with the mitochondrial, kineto-
plast DNA network in trypanosomatid protozoa [Melendy,
T., Sheline, C., and Ray, D.S. (1988) Cell 55, 1083–1088;
Shapiro, T.A., Klein, V.A., and Englund, P.A. (1989) J. Biol.
Chem. 264, 4173–4178]. In the present study, we report the
discovery of a type DNA topoisomerase II activity in bovine
mitochondria. Identified among mtDNA replicative pro-
teins recovered from complexes of mtDNA and protein, the
DNA topoisomerase relaxes a negatively, supercoiled DNA
template in vitro, in a reaction that requires Mg
2+
and ATP.
The relaxation activity is inhibited by etoposide and other

inhibitors of eucaryotic type II enzymes. The DNA topo-
isomerase II copurifies with mitochondria and directly
associates with mtDNA, as indicated by sensitivity of some
mtDNA circles in the isolated complex of mtDNA and
protein to cleavage by etoposide. The purified activity can be
assigned to a  150-kDa protein, which is recognized by a
polyclonal antibody made against the trypanosomal mito-
chondrial topo II enzyme. Mass spectrometry performed on
peptides prepared from the  150-kDa protein demonstrate
that this bovine mitochondrial activity is a truncated version
of DNA topoisomerase IIb, one of two DNA topoisomerase
II activities known to exist in mammalian nuclei.
Keywords: mitochondrial DNA topoisomerase; mito-
chondrial DNA; mtDNA replication; type II DNA topo-
isomerase.
Mitochondria in mammalian cells contain multiple copies
of a small ( 16 kb) circular duplex DNA genome
(mtDNA) that is produced within mitochondria through
repeated cycles of DNA synthesis [1]. The mtDNA genome
encodes 13 polypeptides, each of which is an essential
component of one of the enzyme complexes of the
respiratory chain [2]. Consequently, all of the enzymes and
DNA binding proteins required for the replication of
mtDNA are encoded on nuclear chromosomes, and
imported into the organelle. Despite progress made in
characterizing the mtDNA replicative polymerase (DNA
pol c) [3–6], efforts to isolate and study some other
components of the DNA replicative complex responsible
for mtDNA synthesis has proved to be exceedingly difficult.
This continues to limit our ability to understand the

biochemistry of how mtDNA replication is carried out.
This problem is due both to the low abundance of mtDNA
replicative enzymes in tissues, and to the presence of potent
nuclease activity, and other, ill-defined inhibitors in protein
extracts of mitochondria that block mtDNA replication
assays in vitro. Furthermore, the presence of small frag-
ments of nuclear DNA in standard preparations of
mitochondria has also raised concerns that DNA replica-
tion activities attributed to mitochondria could in fact
represent nuclear contaminants.
One class of enzyme activity likely essential for the
successful synthesis of mtDNA is DNA topoisomerase.
Widely distributed throughout nature, DNA topoiso-
merases promote the passage of DNA strands through
one another, and relieve the torsional stress in DNA
produced for example, during progression of the DNA
replication fork, during transcription, and when newly
replicated DNA genomes need to disentangled from one
another [7]. Different types of DNA topoisomerase
activity have been identified in prokaryotes and in nuclei
of eucaryotes. The type I and III activities are each ATP-
independent and break single DNA strands during
catalysis [7]. They alter the number of times the two
strands of DNA revolve around one another (the linking
number), in steps of one. In contrast, the type II activities
require ATP, produce double-strand breaks during cata-
lysis, and change the DNA linking number in steps of two
[8,9]. Mammalian nuclei contain two different type II
activities, named DNA topoisomerase IIa and IIb,which
are encoded by separate genes ([10,11]). The type II

enzymes are inhibited by novobiocin, and by a variety of
useful anticancer drugs including adriamycin, 4¢-(9-acrid-
inylamine)methanesulfon-m-anisidide (m-AMSA), etopo-
side (VP-16), and ellipticine [12].
Correspondence to R. L. Low, Department of Pathology B-216,
University of Colorado Health Sciences Center, 4200 East Ninth
Avenue, Denver, CO 80262, USA.
Fax: +1 303 315–6721, Tel.: +1 303 3158024,
E-mail:
Abbreviations: BSA, bovine serum albumin; m-AMSA, 4¢-(9-acridi-
nylamine)methanesulfon-m-anisidide.
(Received 15 April 2003, revised 29 August 2003,
accepted 2 September 2003)
Eur. J. Biochem. 270, 4173–4186 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03814.x
In contrast to the nuclear DNA topoisomerases, the
DNA topoisomerase activities present in mitochondria
have been very difficult to purify and identify, especially in
vertebrates. Recently, human mitochondria have been
shown to possess a specific DNA topoisomerase I. This
mitochondrially targeted DNA topo I activity (TOP1mt)
is encoded by a unique gene on chromosome 8q24.3 [13].
The mitochondrial enzyme is highly homologous to
nuclear topo I, and based on its size and properties, no
doubt corresponds to the well-known nuclear-like topo I
activity that had previously been reported in mitochondria
of a variety of cell types [14–19]. Recently, human
mitochondria have also been shown to import nuclear
DNA topoisomerase IIIa activity [20]. This results from
the use of an alternate transcription start site in the
topo IIIa gene that incorporates a mitochondrial targeting

sequence onto the N-terminus of the protein. Much less is
known about mitochondrial type II enzymes. Thus far,
the only type II DNA topoisomerase conclusively dem-
onstrated to be mitochondrial in origin is an enzyme
denoted DNA topoIImt found within the mitochondrion
of protozoa Crithidia fasciculate and Trypanosoma brucei
[21,22]. This enzyme likely plays a role in mtDNA (or
ÔkinetoplastÕ DNA) replication and catenating/decatenat-
ing DNA circles from the kinetoplast network [22]. The
topoIImt has been localized at the periphery of the
kinetoplast network by immunohistochemistry [23], and
epipodophyllotoxins and related drugs that promote
cleavage of DNA by topoIImt and other type II enzymes,
have been shown to promote cleavage of kinetoplast
DNA [24,25]. Recently, suppression of the trypanosome
topoIImt by RNAi has been shown to cause loss of
kinetoplast DNA [26]. In addition to trypanosomes, there
is also evidence for a topoIImt in Dictyostelium discoideum
[27] and Plasmodium falciparum [28]. In mammalian
mitochondria, potent endonuclease and topoImt activities
have made it exceedingly hard to detect any type II
activity. Several years ago, a putative type II activity,
identified from catenation/decatenation and unknotting
assays, was reported in mitochondria from human
leukemic cells [29] and calf thymus [30], and partially
purified. Unfortunately, neither activity could be purified
to near homogeneity, nor shown to relax a supercoiled
DNA substrate in an ATP-dependent manner. Despite the
difficulties even finding DNA topoisomerase II in mam-
malian mitochondria, such an activity has been suspected

to be responsible for the cleavage of mtDNA seen in
ciprofloxacin-treated cells [31], and for producing a
common deletion seen in human mtDNA, which accu-
mulates with aging. In the case of this deletion, nucleotide
sequences where the mtDNA is deleted seem to resemble
a nucleotide consensus sequence often targeted by verte-
brate type II DNA topoisomerase activities [32,33].
In the present study, we report that a eucaryotic-like type
II DNA topoisomerase activity is associated with bovine
mtDNA. This activity was recovered from insoluble com-
plexes of mtDNA and mtDNA replicative factors that were
gently isolated from disrupted heart mitochondria. An
analysis of tryptic peptides prepared from the purified
enzyme using mass spectrometry indicates that this
topoIImt activity is a truncated form of DNA topoiso-
merase IIb.
Materials and methods
Antibodies
Polyclonal antibodies prepared against the trypanosome
topoIImt, human DNA topoisomerase IIa, and human
DNA topoisomerase IIb were generous gifts of D. Ray, Mol.
Biol.Inst.,UCLA,LosAngeles,CA,USA;J.Holden,Dept.
of Pathology, Utah Health Sciences Center, Salt Lake City,
UT, USA; C. Austin, School of Cell and Mol. Biosci.,
University of Newcastle upon Tyne, UK, respectively.
Isolation of the complex of mitochondrial DNA
and its associated proteins
All procedures were carried out at 0–4 °C, unless otherwise
stated. Mitochondria were isolated from fresh ventricular
muscle of adult bovine-heart obtained from a local meat

processing plant (Hyclone, Greeley, CO, USA). The mito-
chondria were recovered from the disrupted heart tissue
essentially as described [34], except that minced tissue was
ruptured by shearing for 30 s at the lowest (not the highest)
speed setting in the Waring blender. To isolate the mtDNA–
protein complex, each 40 mL aliquot of mitochondrial
suspension was diluted with 140 mL of 30 m
M
Tris/HCl
(pH 8), 4 m
M
EDTA, 100 m
M
NaCl, 20 m
M
potassium
glutamate, 10% (w/v) glycerol (buffer A), and gently
disrupted with the addition of 0.5% (w/v) Triton X-100.
After 30 min, the lysate was centrifuged at 145 000 g for
60 min in a Ti50.2 rotor (Beckman). The supernatant
fraction was discarded, and the pellets were pooled and
re-suspended in 35 mL of buffer A, without Triton X-100, by
repeated Dounce homogenization. After 60 min, this sus-
pension was centrifuged at 3000 g for 10 min in a JA20 rotor.
The loose, tan pellet was discarded and the supernatant was
carefully decanted. The supernatant was similarly clarified
once more. The final supernatant was then centrifuged
145 000 g for 60 min in a Ti50.2 rotor. For experiments
requiring intact mtDNA–protein complexes, the dark brown
pellet (containing complexes of mtDNA and protein) was

re-suspended in 10 mL of Buffer A using a Dounce
homogenizer, and stored at 3 °C. When replication proteins
were recovered, the dark brown ÔmtDNA–proteinÕ pellet was
re-suspended in 20 mL of 300 m
M
Tris/HCl (pH 8.8),
900 m
M
NaCl, 20 m
M
EDTA, 10 m
M
dithiothreitol by
Dounce homogenization. After > 2 h, the suspension was
centrifuged 175 000 g for 60 min in a Ti80 rotor, and the
supernatant fraction containing soluble mtDNA replication
factors (fraction II) was recovered and stored at 3 °C.
Relaxation and catenation assays for DNA
topisomerase II activity
Each relaxation reaction contains in 40 lL: 50 m
M
Tris/
HCl (pH 7.9), 125 m
M
NaCl, 7.5 m
M
Mg(OAc)
2
,
0.25 mgÆmL

)1
bovine serum albumin (BSA), 5 m
M
dithio-
threitol, 1.5 m
M
ATP, 500 ng of pUC19 DNA, and
0.5–4 lL of the fraction being assayed. Relaxation reactions
are incubated 60 min at 37 °C, unless otherwise indicated,
and stopped by the addition of 1% (w/v) sodium dodecyl
sulfate (SDS). The terminated reactions are then applied to
a 150-mL 0.8% (w/v) agarose-gel cast and run in 40 m
M
4174 R. L. Low et al. (Eur. J. Biochem. 270) Ó FEBS 2003
Tris/acetate, 1 m
M
EDTA (pH 8). After electrophoresis at
1.5 VÆcm
)1
overnight, the gel is stained in 0.5 lgÆmL
)1
of ethidium bromide and photographed under UV
illumination. One unit of DNA topoisomerase II relaxation
activity is defined as the amount of enzyme that relaxes 50%
(500 ng) of the input supercoiled DNA template in 60 min.
Activity is estimated from visual inspection of the gel or by
scanning densitometry.
Each catenation reaction contains in 40 lL: 40 m
M
Tris/

HCl(pH7.9),125m
M
NaCl, 7.5 m
M
Mg(OAc)
2
,
0.25 mgÆmL
)1
of BSA, 5 m
M
dithiothreitol, none or
1.5 m
M
ATP as indicated, 500 ng of topologically relaxed
pUC19 DNA, and added enzyme. Reactions are run 30 min
at 37 °C, and stopped by the addition of 1% SDS, 20 m
M
EDTA, and 400 m
M
NaCl. After heating terminated reac-
tions to 85 °C for 10 min, reactions are applied to a 150-mL
0.8% agarose gel. Electrophoresis and photography of the
ethidium stained gel is carried out as described above.
Purification of the mitochondrial DNA topoisomerase II
All steps were carried out at 4 °C. All buffers contained
0.5 m
M
phenylmethylsulfonyl fluoride, 1 m
M

sodium
metabisulfite, 0.5 lgÆmL
)1
leupeptin, and 0.01 m
M
pepsta-
tin, unless otherwise stated. mtDNA–protein complexes
were isolated from an 80-mL suspension of purified bovine
heart mitochondria, and the soluble proteins subsequently
released from the mtDNA–protein complexes at 900 m
M
NaCl were collected, as described above (fraction II, 40 mL;
44 mg total protein). Three milliliters of DEAE-Sepharose
resin was added to the fraction II protein concentrate. After
diluting the suspension fourfold in 5 m
M
dithiothreitol and
gently mixing for 10 min, the DEAE-Sepharose was
collected at 12 100 g for 10 min in a JA20 rotor (Beckman),
and the supernatant, containing DNA topoisomerase II
activity, was carefully decanted and saved (fraction III,
155 mL; 36 mg protein). Fraction III was applied to a
20-mL (12 · 1.7 cm
2
) hydroxylapatite column equilibrated
in 30 m
M
Tris/HCl (pH 7.9), 20 m
M
potassium glutamate,

5m
M
dithiothreitol, and 20% (w/v) glycerol (buffer B).
Activity was eluted with a linear 200 mL gradient of
0–1.2
M
potassium phosphate (pH 8.0) in buffer B. Active
fractions of DNA topoisomerase II activity eluted near
300 m
M
potassium phosphate, just prior to those of the
mitochondrial DNA topoisomerase I activity, and were
pooled (fraction IV, 8 mL; 4 mg of protein). Fraction IV
was dialyzed against 1 L of buffer B for 2.5 h, and applied
to a 1.5-mL (2 cm · 0.75 cm
2
) heparin agarose column.
Activity was eluted using a linear 15 mL gradient of 0–1
M
NaCl in buffer B. Active fractions of DNA topoisomerase II
activity eluted near 500 m
M
NaCl and were pooled (fraction
V, 1.3 mL; 0.2 mg of protein). Fraction V was diluted with
6.5 mL of 10 m
M
dithiothreitol and applied to a 1-mL
column of native DNA-cellulose equilibrated with buffer B.
Once loaded, the column was washed with 10 of buffer B,
and eluted with a 10-mL linear gradient of 0–1.1

M
NaCl.
DNA topoisomerase II activity eluted near 350 m
M
NaCl.
Active fractions were pooled (fraction VI, 0.6 mL; 0.024 mg
of protein) and concentrated in a Centricon 30 filter. The
fraction VI concentrate ( 60 lL) was diluted threefold
with 10 m
M
dithiothreitol, and layered onto a linear 4.2 mL
gradient of 15–42% (w/v) glycerol containing 30 m
M
Tris/
HCl(pH7.9),0.1m
M
EDTA, 10 m
M
Mg(OAc)
2
,5m
M
dithiothreitol, 0.1% N-octylglucopyranoside, 1
M
NaCl.
Sedimentation was carried out at 299 000 g in an SW60
rotor (Beckman) for 20 h. Twenty-four fractions were
collected dropwise from the bottom of the tube. Peak
fractions of activity were saved (fraction VII, 0.3 mL;
 2 lg of protein, as estimated from intensity of bands

observed on silver-stained SDS/PAGE gels relative to that
of marker proteins). The specific activity of the fraction VII
enzyme is  1.7 · 10
5
UÆmg
)1
.
Western blot analysis
A purified enzyme fraction containing about 400 ng of total
protein was resolved by electrophoresis on a 7.5% reducing
SDS/PAGE gel at 100 V in a Bio-Rad mini-PROTEAN II
apparatus with a Tris/glycine buffer system [35]. Proteins
were transferred by electrophoresis to Immobilon-P mem-
brane (Millipore) in the Mini Trans-Blot Cell (Bio-Rad) for
180 vH in 30 m
M
Tris/HCl (pH 8.3), 0.02% (w/v) SDS,
0.014% (w/v) glycine, 20% (v/v) methanol. Blots were
briefly stained with 0.2% (w/v) Ponceau S to confirm
efficient protein transfer. The blots were blocked with
3%(w/v) BSA in phosphate-buffered saline (NaCl/P
i
), 0.5%
(w/v) Tween 20 for 1 h at ambient temperature. Incubation
with the primary antibody (diluted 1 : 500–1 : 25 000 in
blocking buffer) was carried out overnight at 4 °C. Sub-
sequently, blots were washed extensively with frequent
changes of NaCl/P
i
containing 0.5% (w/v) Tween 20, at

ambient temperature for 2 h. The secondary antibody,
which was an anti-rabbit, peroxidase-labeled antibody, was
then applied, at a dilution of 1 : 1000. After final washes in
NaCl/P
i
for 2 h at ambient temperature, protein-immune
complexes were visualized by chemiluminescence, according
to the procedure recommended by the manufacturer (ECL,
Amersham Life Science).
Protein identification by mass spectrometry
Proteins were separated by 1D SDS/PAGE and stained
with a low-fixation silver stain [36]. Protein bands were
individually excised and the silver was removed. Gel slices
were equilibrated in 100 m
M
NH
4
HCO
3
and dehydrated
with acetonitrile and vacuum centrifugation. Dehydrated
gel slices were then rehydrated with 15 lL25m
M
NH
4
HCO
3
containing 0.01 lgÆlL
)1
modified trypsin

(Promega), and trypsin digestion was carried out for > 3 h
at 30 °C. Peptides were extracted with 60% acetonitrile,
0.1% trifluoroacetic acid, dried by vacuum centrifugation,
and reconstituted in 8 lL 0.1% trifluoroacetic acid. Pep-
tides were then desalted and concentrated into 2 lL60%
acetonitrile, 0.1% trifluoroacetic acid using ZipTipC18
pipette tips (Millipore). 0.2 lLwasappliedtoaMALDI
target and overlayed with 0.2 lL a-cyano-4-hydroxycin-
namic acid matrix. MALDI-TOF mass spectrometry was
carried out using a Voyager DE-PRO mass spectrometer
(Applied Biosystems) operated in reflectron mode. Ions
[M + H] corresponding to peptide masses were entered
into the MS-FIT database search algorithm (http://
prospector.ucsf.edu/) and the SWISS-PROT, NCBInr and
pdbEST databases were searched, allowing for complete
carbamidomethylation of cysteine and partial oxidation of
Ó FEBS 2003 DNA Topo IIb mammalian mtDNA (Eur. J. Biochem. 270) 4175
methionine. Peptide mass errors of up to 50 p.p.m. were
considered during the search.
Isolation of the mtDNA–protein complex from mitoplasts
Mitoplasts were prepared from bovine heart mitochon-
dria, essentially as described [37] unless otherwise stated.
Briefly, a 20 mL suspension of bovine-heart mitochondria
(25 mgÆmL
)1
protein) was supplemented with 0.5 mgÆmL
)1
bovine serum albumin and 0.1% (w/v) digitonin. After gently
stirring the suspension for 15 min at 0 °C, the mitochondria
were diluted with 75-mL of 5 m

M
Hepes (pH 8),
0.5 mgÆmL
)1
bovine serum albumin, 70 m
M
sucrose,
220 m
M
mannitol (buffer C) and disrupted using a 40-mL
Dounce homogenizer (four to-and-fro passes with the ÔtightÕ
pestle). The homogenate was then centrifuged at 15 000 g for
10 min in a JA-14 rotor (Beckman). The soft, brown inner-
membrane (mitoplast) pellet was collected, and resuspended
in 100-mL buffer C using Dounce homogenization, and
centrifuged 15 000 g for 10 min as before. The washed
mitoplast pellet was resuspended with 10 mL of buffer B, and
Triton X-100 was added to 0.5%. After 20 min at 0 °C, the
mtDNA–protein complex was collected at 30 000 g for
30 min, and resuspended in 1 mL of buffer B minus glycerol.
Isolation of the mtDNA–protein complex from
mitochondria sequentially treated with DNase I and
proteinase K
A 20-mL sample of a freshly prepared suspension of
bovine-heart mitochondria was diluted to 220 mL with
30 m
M
TrisÆHCl(pH7.7),50m
M
sodium glutamate, 10%

(w/v) sucrose (buffer D), and the mitochondria collected at
15 300 g for 15 min in a JA14 rotor (Beckman). The
brown mitochondrial pellet was resuspended in 25 mL of
buffer D by Dounce homogenization and 4 m
M
Mg(OAc)
2
and 0.2 mgÆmL
)1
pancreatic DNase I were added. DNase
digestion was carried out for 30 min at ambient tempera-
ture, and terminated by the addition of 10 m
M
EDTA. The
DNase I-treated mitochondria were diluted to 220 mL
with buffer D plus 5 m
M
EDTA, and then collected at
15 300 g for 15 min in a JA14 rotor. This step is intended
to facilitate removal of DNA fragment debris and residual
DNase.Thewashstepwasrepeatedtwicemore.Thefinal
pellet of mitochondria was resuspended in 200 mL of
5m
M
Mops (pH 7.4), 5 m
M
KH
2
PO
4

,1m
M
EDTA, 0.3
M
sucrose, 0.1% BSA (buffer E), then collected at 15 300 g for
15 min in a JA14 rotor. The mitochondria were resuspended
in 20 mL of buffer E, and 40 mL of 10 m
M
Hepes (pH 7.4),
0.6
M
mannitol containing 45 lgÆmL
)1
of proteinase K was
added, as otherwise modified [20]. Protein digestion was
carried out 30 min at 0 °C, and stopped by the addition of
4m
M
phenylmethylsulfonyl fluoride. After 10 min at 0 °C,
the mitochondria were collected at 12 100 g for 15 min in a
JA20 rotor. The mitochondrial pellet was resuspended in
220 mL of buffer E plus 0.1 m
M
phenylmethylsulfonyl
fluoride, recentrifuged at 15 300 g for 15 min in a JA14
rotor, and the washed pellet resuspended in 220 mL of buffer
E plus 0.1 m
M
phenylmethylsulfonyl fluoride. This wash step
was repeated four times. The final pellet of DNase I/

proteinase K treated mitochondria was resuspended in
90 mL of buffer A and the mitochondria were disrupted with
the addition of 0.5% Triton X-100. Complexes of mtDNA
and protein were isolated as described above.
Results
Recovery of a DNA topoisomerases II activity
from isolated complexes of mtDNA and protein
When preparations of purified bovine-heart mitochondria
are disrupted with the addition of 0.5% (w/v) Triton X-100,
the mtDNA and its associated proteins are found to reside
in an insoluble complex, which can be recovered from the
mt lysate through a series of differential centrifugation steps.
These low and high-speed centrifugation steps eliminate
fragments of nuclear DNA–protein complexes that invari-
ably contaminate the mtDNA–protein complex. As well,
they separate the insoluble complex of mtDNA–protein
from > 95% of the mitochondrial protein, which is soluble.
Subsequent treatment of the isolated complex of mtDNA
and protein with 900 m
M
NaCl releases a fraction of
mtDNA replicative proteins and DNA binding proteins
from the mtDNA. These proteins (now soluble) are
recovered in the supernatant following centrifugation of
the high salt extract at 30 000 g for 30 min. In contrast, the
mtDNA, which nearly all remains insoluble, is still pelleted.
During purification of DNA polymerase c released from
the mtDNA using successive steps of hydroxylapatite and
native DNA cellulose chromatography, and glycerol gradi-
ent velocity sedimentation, we identified a eucaryotic type

ATP-dependent DNA topoisomerase activity. This activity
partially copurifies with the DNA pol c activity. The
topoisomerase activity is evident from its relaxation of a
negatively supercoiled plasmid DNA in a fairly nonproces-
sive fashion (Fig. 1A). The rate of DNA relaxation appears
constant for 30 min at 37 °C and the extent of DNA
relaxation proportional to added enzyme is in the range of
5–20 ng of protein. The maximal rate of template relaxation
requires 1.5 m
M
ATP with one-half maximal relaxation
occurring at 0.25 ± 0.05 m
M
ATP. As observed with other
type II activities, a trace level of DNA relaxation occurs in
the absence of ATP, presumably due to ATP copurified
with the enzyme. While addition of 1.5 m
M
dATP can
substitute for ATP, neither 1.5 m
M
CTP, UTP, nor GTP
supports activity (Fig. 1B). The relaxation activity is
inhibited by novobiocin. In assays containing 0.5 m
M
ATP, levels of novobiocin above 200 l
M
are completely
inhibitory. Additional titration experiments indicate that
50% inhibition occurs at a concentration of about 70 l

M
.
The mt topoisomerase II lacks DNA gyrase activity as
indicated by the failure of the enzyme to supercoil 500 ng of
a relaxed DNA template in standard assays that contained
ATP in the range of 0.1–5 m
M
(data not shown). The
fraction V enzyme also lacks detectable DNA ligase activity
(< 0.05 UÆlL
)1
) as assessed using a DNase I nicked
plasmid DNA template (data not shown).
The DNA topoisomerases II activity copurifies
with mitochondria
To rule out the possibility that the observed DNA
topoisomerase II activity was a contaminant, we assessed
whether this activity copurified with mitochondria collected
4176 R. L. Low et al. (Eur. J. Biochem. 270) Ó FEBS 2003
through two successive linear 0.5–2
M
sucrose gradients for
2 h each at 22 000 r.p.m. in an SW28 rotor (Materials and
methods). As expected, the mitochondria recovered from
peak fractions of the second gradient appear relatively free
of nuclear DNA contamination that could provide a source
of topoisomerase activity. This can be seen from the
prominent 7.3-, 4.8-, and 4.3-kb EcoRI restriction fragments
of bovine mtDNA that are seen when DNA extracted from
Fig. 1. Identification of an ATP(dATP)-dependent mitochondrial DNA topoisomerase and demonstration that the activity cosediments with mito-

chondria. Photographs of agarose-gel assays are shown. Relaxed and supercoiled (sc) forms of the plasmid DNA are as labeled. (A) Time course.
Standard agarose-gel, relaxation assays contained 20 ng of fraction VI enzyme, without or with 1.5 m
M
ATP, and were run 5, 10, 15, or 30 min, as
indicated. (B) Activity requires ATP(dATP). Standard relaxation assays were carried out either with none, or 1.5 m
M
ATP, GTP, CTP, UTP, or
dATP, as indicated. Reactions were run for 30 min at 37 °C. (C) ATP-dependent DNA topoisomerase II cosediments with mitochondria. Two
12-mL samples of fresh bovine heart mitochondria (40 mgÆmL
)1
protein) were sedimented through 30 mL, linear 0.5–2
M
sucrose gradients
(preparedin30m
M
Tris/HCl (pH 8), 75 m
M
NaCl). Sedimentation was for 2 h at 70 000 g in an SW28 rotor (Beckman), run at 3 °C. The visible
band of mitochondria from each gradient was removed laterally from the tube using a 16-gauge needle. The mitochondria were pooled and diluted
to250mLin30m
M
Tris/HCl (pH 8), 75 m
M
NaCl (buffer C). The mitochondria were collected at 11 000 r.p.m. for 20 min in a JA14 rotor,
resuspended in 10 mL of buffer C and layered onto a second, 30 mL linear 0.5–2
M
sucrose gradient which was centrifuged for 2 h at 70 000 g,as
described above. Following centrifugation, 15–2.3 mL fractions were collected dropwise from the bottom of the tube. The visible band of
mitochondria eluted in fractions 8 and 9. DNA was phenol extracted from a 0.5-mL aliquot of fractions 5 through 14 that were supplemented with
1% SDS. After a treatment with 0.1 mgÆmL

)1
of RNase A for 40 min at 37 °C, each DNA sample was digested with 20 U of EcoR1 for 5 h at
37 °C, analyzed by agarose-gel electrophoresis on a 0.8% agarose gel. A photograph of the ethidium-stained gel is shown in (C). The mitochondria
in the remainder of fractions 8 and 9 were pooled, diluted to 32 mL with buffer A and disrupted in the presence of 0.5% Triton X-100. After 30 min
at 3 °C, the mtDNA–protein complex was collected by centrifugation at 175 000 g in a Ti80 rotor (Beckman) for 30 min. Proteins released from the
isolated mtDNA–protein complex at 900 m
M
NaCl were then prepared, concentrated to 0.2 mL using a Centricon-10 filter. The concentrate was
then layered onto a 4-mL, linear 15–42% (v/v) glycerol gradient. Sedimentation was carried out for 20 h at 299 000 g in a SW60 rotor (Beckman).
Fractions were collected dropwise from the bottom of the tube and assayed for type II DNA topoisomerase activity. Standard, agarose-gel
relaxation assays performed on even numbered fractions between 2 and 12, without and with 1.5 m
M
ATP are shown in (D). These reactions
contained 2 lL aliquots of each fraction assayed and were carried out 30 min at 37 °C.
Ó FEBS 2003 DNA Topo IIb mammalian mtDNA (Eur. J. Biochem. 270) 4177
samples of the mitochondria are digested by EcoRI and
analyzed by agarose-gel electrophoresis (Fig. 1C). To assess
whether the purified mitochondria still contain topoiso-
merase II activity, the remainder of the mitochondria were
disrupted with 0.5% (w/v) Triton X-100, complexes of the
mtDNA–protein were collected, and a concentrate of
proteins released from the mtDNA at 900 m
M
NaCl was
prepared and sedimented through a linear glycerol gradient.
This velocity sedimentation step has proved to be quite
effective in separating the topoisomerase II activity from the
mitochondrial topoisomerase I and endonuclease G acti-
vities, which both strongly inhibit type II topo assays. As
seen in Fig. 1D, ATP-dependent topoisomerase activity is

evident in the gradient, in those fractions collected near the
bottom (fractions 5 and 6), where the peak of DNA
polymerase c activity is also found (data not shown). Thus it
is unlikely that the topoisomerase II activity is a nuclear or
cytoplasmic contaminant.
The purified mitochondrial DNA topoisomerase II
activity is sensitive to known inhibitors of eucaryotic type
II enzymes and can catalyze catenation of plasmid DNA
circles. As determined using the fraction V enzyme, DNA
relaxation activity occurs over a fairly narrow range of
Mg
2+
, NaCl, and pH with maximal rates occurring at
7.5 m
M
Mg
2+
,100 m
M
NaCl, and pH 8.0–8.5, respectively.
Relaxation activity shows a strict requirement for Mg
2+
,
but not Ca
2+
or Mn
2+
, in the range of 0.1–15 m
M
supports

activity. In addition to novobiocin, etoposide and m-AMSA
(inhibitors of eucaryotic type II enzymes), inhibit the
mitochondrial DNA topoisomerase II activity (Table 1).
Concentrations of each substance that cause 50% inhibition
of the relaxation activity of the fraction V enzyme are listed.
In addition to being able to relax negatively supercoiled
DNA templates in the presence of ATP, the mitochondrial
DNA topoisomerase II activity can also promote catenation
of topologically relaxed plasmid DNA circles in the presence
of a DNA crowding agent. As shown in Fig. 2, this
catenation activity requires ATP, as expected, and produces
huge networks of interlocked DNA circles that fail to enter
the agarose gel during electrophoresis.
The mitochondrial DNA topoisomerases II activity
appears to be associated with mtDNA
An association of DNA topoisomerase II with mtDNA has
been demonstrated by showing that treatment of the
isolated mtDNA–protein complex with etoposide in the
presence of SDS promoted cleavage of mtDNA circles into
full-length linear mtDNA [38]. In this experiment, small
samples of a suspension of the isolated mtDNA–protein
complex were incubated with or without etoposide in the
presence of 1% SDS for 15 min at 37 °C. After the addition
of proteinase K, the mtDNAs were phenol purified,
resolved by agarose-gel electrophoresis and transferred to
nitrocellulose paper. The extent of drug mediated cleavage
of the mtDNA was then assessed by Southern blot analysis
usinga[
32
P]BamH1-Hpa1 restriction fragment of bovine

mtDNA as probe. As seen in Fig. 3, treatment with
etoposide, in the range of 50–500 l
M
, converted some
mtDNA circles to full-length linear DNA. However, so far,
we have not seen more than about 15% of the input circles
linearized (as indicated from scanning densitometry, data
not shown), even if the incubation periods at 37 and 64 °C
were extended. Maximal conversion occurs at about 100 l
M
of drug. This cleavage reaction required 10 m
M
Mg
2+
but
not the addition of ATP. No DNA cleavage was observed
using drug vehicle (dimethylsulfoxide) alone.
The purified mitochondrial type II activity can be
assigned to a polypeptide of  150 kDa
In order to further identify the mitochondrial DNA
topoisomerase II, the enzyme recovered from the isolated
mtDNA–protein complex was purified using successive
steps of hydroxylapatite, native DNA-cellulose, and heparin
agarose chromatography, followed by glycerol gradient
Table 1. Known inhibitors of eucaryotic type II DNA topoisomerases
inhibit the bovine mitochondrial DNA topoisomerases II activity.
Standard ATP-dependent relaxation assays were carried out with the
fraction VI enzyme as detailed in Materials and methods. Novobiocin
assays contained 0.25 m
M

ATP. N-ethylmaleimide assays performed
without added dithiothreitol.
Inhibitors
Concentration that gives 50%
inhibition of relaxation activity
1. Etoposide 70 l
M
2. m-AMSA 3 l
M
3. Novobiocin 70 l
M
4. N-Ethylmaleimide 60 l
M
5. Ethidium bromide 0.6 lgÆmL
)1
Fig. 2. The purified mitochondrial DNA topoisomerase promotes ATP-
dependent catenation of relaxed plasmid DNA circles. Catenation assays
were performed for 30 min at 37 °C as detailed in Materials and
methods. Assays contained 0.5, 1.0, or 2.0 lL of the glycerol gradient
pool of mitochondrial DNA topoisomerase II (fraction VII), 5 U of
purified mitochondrial DNA topoisomerase I activity, and none or
1.5 m
M
ATP, as indicated. A photograph of the ethidium stained
agarose gel is shown; C, control DNA minus enzyme; rc, relaxed cir-
cular form DNA; sc, supercoiled form DNA.
4178 R. L. Low et al. (Eur. J. Biochem. 270) Ó FEBS 2003
velocity sedimentation. When the fractions of the glycerol
gradient spanning the peak of DNA topoisomerase II
activity were analyzed using a silver-stained SDS/PAGE, a

band ÔdoubletÕ of  150 kDa, within the peak of highest
activity (fraction 6), correlated with the DNA topoiso-
merase II activity (Fig. 4A,B). Western blot analysis using
antibodies against trypanosome topoIImt, and human
nuclear topoIIa and topoIIb enzymes provided further
evidence that this 150 kDa polypeptide corresponds to a
DNA topoisomerase II activity (Fig. 4C). As shown, the
bovine mitochondrial  150 kDa polypeptide was recog-
nized by antibody prepared against the purified trypano-
somal DNA topoIImt (left most gel, Fig. 4C), however, not
by a rabbit nonimmune serum (not shown). As seen in gel
panel 2, this bovine  150 kDa band was also recognized by
antibody made against human nuclear DNA topoIIa,albeit
less well than that seen with the antigen control (panel 3). In
contrast, as evident in panel four, it failed to be recognized
at all by the antinuclear topoIIb antibody. Positive control
blots for antinuclear topoIIa and topoIIb antibodies with
human topoIIa and topoIIb antigens are shown in gel
panels three and five, respectively.
Identification of the mitochondrial type II activity
as a truncated form of DNA topoisomerases IIb
Mass spectrometry was used to confirm that the mito-
chondrial proteins representing the 150 kDa bands were
topoisomerases. Proteins in the activity peak from a glycerol
gradient were separated by SDS/PAGE, silver stained and
digested in-gel with trypsin protease as described in
Materials and methods. Peptide masses acquired by mat-
rix-assisted laser desorption/ionization, time of flight mass
spectrometry (MALDI-TOF MS) were used in database
search algorithms and led to an unambiguous match to

human topoisomerase IIb (Fig. 5). The bovine homologue
was not present in the databases searched. However, a
bovine cDNA was found that contained 100% identity with
the human sequence (Fig. 5).
Human topoisomerase IIb,aswellasseveralother
mammalian homologues, has a predicted molecular weight
 182 kDa, well above the  150 kDa mobility observed
for the mitochondrial proteins. Furthermore, peptide cov-
erage was not found past residue 1250 in the human
sequence (out of a total of 1621 or 1626, depending on the
splice variant), despite nearly all ions being accounted for in
the mass spectrum (Fig. 5). These findings are consistent
with the hypothesis that the mitochondrial enzyme identi-
fied from the gel slice is truncated. Although the spectra
contain peptides unique to topo IIb, no peptides unique to
topo IIa have been encountered. This finding suggests that
the mitochondria topo II activity is likely a form of
topo IIb, although the spectra do not exclude the possibility
that a fragment of topo IIa could also be present.
Further evidence that the truncated topo IIb is
mitochondrial in origin and not a nuclear contaminant
As the mitochondrial DNA topoisomerase II activity
corresponds to a truncated form of the DNA topoisomerase
IIb found in nuclei, we decided it was imperative to
re-evaluate whether the mitochondrial activity could simply
be a nuclear contaminant. Had the topoisomerase II
recovered from the purified mitochondria originated from
fragments of nuclear DNA that adhere to mitochondria?
Two additional experiments carried out indicate that this is
not likely. In the first experiment, samples of mitochondria

were treated with 0.1% digitonin to strip away outer
membranes and remove nuclear DNA debris that could be
adherent to mitochondria. The resultant mitoplasts were
then collected, disrupted with 0.5% Triton X-100, and the
complex of mtDNA–protein recovered by differential
centrifugation. Proteins released from this mtDNA complex
at 600 m
M
NaCl were then fractionated by glycerol gradient
velocity sedimentation and assayed for DNA topoisomerase
II activity. As shown in Fig. 6A, agarose-gel electrophoresis
of the mtDNA purified from the mitoplast mtDNA–protein
complex and digested with EcoR1 reveals the prominent
7.3-, 4.8-, and 4.3-kb bands characteristic of bovine
mtDNA. As seen, the mitoplast preparation appears
essentially free of nuclear DNA contaminants. In spite of
Fig. 3. Etoposide promotes cleavage of some mtDNA circles of the
mtDNA–protein complex [38]. Ten-microliter samples of the suspen-
sion of mtDNA–protein complex were each diluted into 150 lLof
30 m
M
Tris/HCl (pH 7.9), 125 m
M
NaCl, 2 m
M
dithiothreitol, 7.5 m
M
Mg(OAc)
2
,1.5m

M
ATP, without or with 50, 100, 250, or 500 l
M
etoposide, at 3 °C.SDSwasimmediatelyaddedto1%(w/v),andthe
reactions were incubated 30 min at 37 °C. Proteinase K was then
supplemented to 0.1 mgÆmL
)1
, and the reactions were further incu-
bated 20 min at 64 °C. DNAs were phenol and chloroform extracted
once, ethanol precipitated, and the precipitates collected at 31 000 g
for 30 min in a JA20 rotor. Each precipitated DNA sample was
re-suspended in 40 lLof40m
M
Tris/acetate, 1 m
M
EDTA (pH 8)
(TAE), plus 1% SDS, and applied to a 150-mL 0.8% agarose gel that
was then run at 1.5 VÆcm
)1
overnight in a TAE buffer system. DNAs
in the gel were blotted onto nitrocellulose membrane by capillary
transfer in 1.5
M
NaCl, 0.15
M
sodium citrate (pH 7) (10 · NaCl/Cit),
as described [69]. Hybridization was carried out overnight at 68 °Cin
6 · NaCl/Cit, 0.25% (w/v) nonfat dried milk, with a heat-denatured
[5¢-
32

P]BamH1-Hpa1 restriction fragment of the D-loop region
sequence of bovine mtDNA ( 10
8
dpmÆlg
)1
), as probe. After
extensive washing of the filter, as detailed [69], autoradiography was
carried out for 1 h at )80 °C using Kodak XR5 film. A photograph of
the autoradiogram is shown. The full-length linear form of bovine
mtDNA was identified using a Hpa1-digested sample of purified
bovine mtDNA, as shown.
Ó FEBS 2003 DNA Topo IIb mammalian mtDNA (Eur. J. Biochem. 270) 4179
this, the topoisomerase assays performed on the glycerol
gradient reveal a vigorous topoisomerase II activity, peak-
ing in fractions 8 and 9. This indicates strongly that there is
DNA topoisomerse II activity associated with mtDNA as
previous experiments suggested. Furthermore, as shown in
Fig. 6B, Western blot analysis of the topoisomerase II peak
Fig. 4. The mitochondrial DNA topoisomerase II activity is associated with a » 150-kDa polypeptide that is recognized by an anti-trypanosome
topoIImt Ig. (A) Silver-stained SDS/PAGE gel of the active, mitochondrial DNA topoisomerase II fraction of the glycerol gradient velocity
sedimentation purification step. A 200-lL sample of the fraction V enzyme (200 l) was layered onto a 4-mL linear 15–42% (v/v) glycerol gradient
containing 30 m
M
Tris/HCl (pH 7.9), 300 m
M
NaCl, 10 m
M
Mg(OAc)
2
,5m

M
dithiothreitol, 0.05% (w/v) n-octylglucoside. Sedimentation was
carriedoutat299000g in a SW60 rotor (Beckman) for 20 h at 3 °C. Twenty, five-drop fractions were collected from the bottom of the tube.
Proteins in a 100-lL aliquot of fractions 4, 6, and 8 were each precipitated in 10% (w/v) trichloroacetic acid, the protein precipitants collected at
31 000 g for 30 min in a JA20 rotor, were resuspended in 20 lLof50m
M
Tris/HCl (pH 6.8), 10% (v/v) glycerol, 2% (w/v) SDS, 0.7
M
2-mercaptoethanol, 0.05% (w/v) bromophenol blue. Samples were heat-denatured 3 min at 94 °C, and run through a 7.5% reducing, SDS/PAGE
gel, at 100 V in a Tris/glycine buffer system [34]. A photograph of the gel stained with silver [34], is shown in (A). Marker, molecular size standards:
myosin (200 kDa), b-galactosidase (116 kDa), phosphorylase b (97 kDa), BSA (66 kDa), ovalbumin (45 kDa), carbonic anhydrase (31 kDa). (B)
Standard agarose-gel relaxation assays of glycerol gradient contained in 40 lL: a 2-lL aliquot of fraction 4, 6, or 8, without or with 1.5 m
M
ATP, as
indicated. Assays were carried out for 60 min at 37 °C. A photograph of the agarose gel is shown. (C) Western analysis using antibodies against
trypanosomal topoIImt, and human nuclear DNA topoisomerases II a and b reveals cross reactivity with the bovine mitochondrial enzyme. Blots
were probed with diluted antibodies and were developed by chemiluminescence, and exposed to X-ray film. Primary antibodies were used at the
following dilutions: anti-trypanosome (1 : 500); anti-topoIIa (1 : 500) and anti-topoIIb (1 : 25 000). The blot prepared with the fraction V of the
bovine mitochondrial DNA topoisomerase II (400 ng) was probed with either the anti-trypanosome Ig (SDS-gel panel 1), the anti-topoIIa Ig (SDS-
gel panel 2), or the anti-topoIIb Ig (SDS-gel panel 4), respectively, as shown. The anti-topoIIa and topoIIb Igs were tested with a positive control of
either recombinant human topoIIa ( 150 ng) or topoIIb antigen ( 200 ng) provided by the supplier of the antibody (SDS-gel panels 3 and 5,
respectively). Relative sizes of prestained standards run in an adjacent lane are shown at the left of the gel.
4180 R. L. Low et al. (Eur. J. Biochem. 270) Ó FEBS 2003
of the glycerol gradient reveals a faint, immunoreactive
band of  150 kDa, in agreement with size of the purified
activity. This result further suggests that the apparent
truncation of the topoisomerase IIb does not simply result
from in vitro proteolysis during the several-step enzyme
purification. In contrast to the mtDNA–protein complexes
recovered from the mitoplasts, the complexes of nuc-

lear protein and DNA fragments removed from the
mitochondria by digitonin fail to yield any detectable
DNA topoisomerase II activity, when proteins bound to
this DNA are released by high salt, fractionated by glycerol
gradient velocity sedimentation, and assayed (Fig. 6C).
In a second experiment, a 20-mL sample of bovine-heart
mitochondria was treated successively with DNase I and
Fig. 5. Tryptic peptides from bovine topoisomerases IIb identifiedbymassspectrometry.Protein bands were excised after 1D SDS/PAGE and
digested in-gel with trypsin protease as described in Materials and methods. (A) Silver-stained SDS/PAGE gel separating proteins present in
fraction 6 from the second glycerol gradient. Arrows indicate the 150 kDa proteins that were individually subjected to in-gel digestion with trypsin
protease as described in Materials and methods. (B) Matrix-assisted laser desorption/ionization, time of flight (MALDI-TOF) mass spectrum of
tryptic peptides isolated from the lower 150 kDa protein band. Asterisked ions (m/z ¼ 1070.48, 1115.64, 1122.59, 1128.64, 1131.57, 1159.58,
1264.60, 1270.65, 1296.66, 1315.70, 1331.66, 1357.74, 1436.68, 1461.78, 1500.80, 1640.85, 1830.98, and 2423.19 Da – left to right) match predicted
tryptic peptide masses (plus one amu) from human topoisomerase IIb (MOWSE Score ¼ 1.35 · 10
9
).IonslabeledTmatchexpectedtrypsin
autolytic peptides (m/z ¼ 842.50, 1045.56 and 2211.09 Da) and were used to internally calibrate the mass spectrum to a mass accuracy of within
50 p.p.m. Ions labeled K match background peptides derived from keratin that were also present in controls. x-axis, mass-to-charge ratio (m/z);
y-axis, relative ion intensity. (C) Amino acid sequence of human topoisomerases IIb. Residues contained within the predicted tryptic peptides
matched by the MALDI-TOP MS data are indicated in boldface. The only bovine sequence found to significantly match the MALDI-TOF MS
data was the cDNA 211850 MARC 2BOV, which encodes a peptide containing a 100% match to the human amino acid sequence (shaded in gray,
amino acids 725–904).
Ó FEBS 2003 DNA Topo IIb mammalian mtDNA (Eur. J. Biochem. 270) 4181
proteinase K to degrade any DNA topoisomerase II that
could be adherent to mitochondria. After several cycles of
washing to remove proteolytic debris and any trace
proteinase K activity, the mitochondria were disrupted
with the addition of 0.5% Triton X-100, and the
mtDNA–protein complexes recovered, and proteins
released from the mtDNA fractionated by glycerol gradient

velocity sedimentation. DNA topoisomerase II activity,
measured either by ATP-dependent catenation or relaxation
assays, was identified in glycerol gradient fractions 7, 8, and
9, as expected. The amount of activity recovered was about
75% that obtained in the mitoplast experiment (see Fig. 6).
In contrast, no topoisomerase II activity could be recovered
if the mitochondria were first disrupted with 0.5% Triton
X-100 prior to the addition of the proteinase K.
Discussion
In this study, we present biochemical evidence that mam-
malian mitochondria contain a catalytically active, trun-
cated form of DNA topoisomerase IIb. This activity
copurifies with mitochondria collected over successive
sucrose gradients, and the activity is associated with purified
complexes of mtDNA and protein that are recovered from
isolated mitochondria and digitonin-treated mitoplasts
using steps that eliminate nuclear DNA contaminants.
Unlike the well-characterized nuclear form of DNA topo-
isomerase IIb that consists of a 180-kDa polypeptide and
relaxes DNA in a processive fashion in vitro [39,40], the
polypeptide of the mitochondrial DNA topoisomerase IIb is
 150 kDa in size, and its relaxation activity acts fairly
nonprocessively. The mitochondrial form of DNA topo-
isomerase IIb retains sensitivity to well-known, clinically
useful inhibitors of DNA topoisomerase II activity, but the
enzyme fails to be recognized by a topoisomerase IIb-
specific antibody prepared against C-terminal epitopes not
present in DNA topoisomerase a. Furthermore, mass
spectrometric analysis of the mitochondrial polypeptide
shows an absence of peptides predicted from the C-terminal

sequence. These findings suggest that the mitochondrial
activity lacks the  30-kDa C-terminal domain of the
nuclear enzyme.
This C-terminal truncation does not appear to be an
in vitro proteolytic artifact, as several other polypeptides
identified in these fractions of mitochondrial protein,
including the polypeptides for DNA polymerase c and its
accessory factor, adenylate kinase, apoptosis inducing
factor, and endonuclease G, each has a size that is what
should be expected (R. Low, K. Fang and D. Friedman,
unpublished data). Furthermore, we have been able to
detect this  150-kDa polypeptide (as shown in Fig. 6), but
not any protein band of  180-kDa on a Western blot at an
early step in purification suggesting that this form of the
enzyme is not a proteolytic artifact but what resides in
mtDNA–protein complexes.
Proteolytic degradation of DNA topoisomerase II during
its purification can be a major problem with some types of
tissue. For example, purification of DNA topoisomerase II
isoenzyme forms from calf thymus that is notoriously rich in
protease activity typically yields active 120- and 140-kDa
fragment artifacts of DNA topoisomerase II activity [41,42].
However, with heart tissue and isolated heart mitochondria
used in this study, proteolysis during enzyme isolation
appears much less problematic. In contrast to thymic and
other types of cells, adult myocytes contain relatively few
lysosomes [43], a major source of proteolysis. Furthermore,
Fig. 6. The truncated topo IIb arises from mtDNA not from contami-
nants of nuclear DNA fragments. (A) Identification of DNA
topoisomerase II activity among proteins released from mitoplast

mtDNA–protein complexes and fractionated by glycerol gradient
velocity sedimentation. Proteins released at 600 m
M
NaCl from
mtDNA–protein complexes of mitoplasts (see Materials and methods)
were concentrated in a Centricon 10 filter (Amicon). This concentrate
(150 mL) was then run through a 15–42% glycerol gradient as des-
cribed in the Fig. 4 legend, except that the glycerol gradient contained
1
M
, not 300 m
M
NaCl. Topo II assays plus and minus ATP on active
fractions were carried out, are shown. An EcoR1 digest of DNA
phenol-purified from a sample of the mtDNA–protein complex (25%
of total) is also shown. (B) SDS/PAGE and Western blot analysis of
glycerol gradient fractions 8 plus 9. One-half of this pool was used for
each gel analysis. See Fig. 4 legend for details. (C) Topo II assays,
minus and plus ATP, performed on glycerol gradient fractionation of
proteins released at 600 m
M
NaCl from insoluble Ôouter membraneÕ
complexes of nuclear DNA–protein that were released from mitoplasts
with digitonin, and recovered at 16 000 g for 20 min. Velocity sedi-
mentation was carried out as described in the legend of Fig. 4.
Agarose-gel analysis of an EcoR1 digest of 4 lg nuclear (nuc) DNA
recovered from the Ôouter membraneÕ fraction is shown. Relative
positions of mtDNA EcoR1 fragments in a far lane (not seen) are
indicated.
4182 R. L. Low et al. (Eur. J. Biochem. 270) Ó FEBS 2003

adult heart tissue, in contrast to thymus, contains very few
monocytes/macrophages that are enriched in lysosomes,
and cells undergoing apoptosis, and engaged in proteolysis
and active cellular degradation [44].
The C-terminal domain of DNA topoisomerase IIb
apparently deleted from the mitochondrial enzyme consti-
tutes one of three structural and functional domains that are
present within the polypeptide sequence of all eucaryotic
type II enzymes [7]. The two other domains, the N-terminal
and central domains, contain the ATPase activity and active
site tyrosine residue, respectively. These latter motifs are
highly conserved among all of the eucaryotic type II
enzymes and crucial for activity. In contrast, the C-terminal
region is the only domain that is not well conserved. The
amino acid sequences in the C-terminal quarter of the
a-andb-isoform polypeptides, for example, only share
about 34% identity. In contrast, those from the remaining
three-quarters of the polypeptides, that include the
N-terminal and central domains, are 78% identical
[7,10,11,45]. Although the function of this domain is still
not fully characterized, the C-terminal domain region
contains phosphorylation sites and sequences that probably
play a role in the regulation of enzyme activity in vivo,andin
directing the enzyme to specific protein partners and tasks in
the nucleus required to maintain chromatin topology [10,11]
In yeast [46], Drosophila [47], and humans [10,11], the
C-terminal region has been shown to contain the signal for
nuclear localization. In vivo,thea-andb-isoforms localize
to different sites in the nucleus and play different roles in
orchestrating DNA topology [10,11,48]. Sequence differ-

ences in the C-terminal regions of the a-andb-isoforms are
likely what makes this possible. Consequently, truncation of
the C-terminal domain may be essential for the b isoform to
be targeted to the mitochondrion and to assume a role in
mtDNA replication.
The presence of DNA topoisomerase IIb within the
mitochondrion, adds to the growing list of enzymes active in
DNA metabolism that are shared by both nuclear and
mitochondrial compartments. Although many enzymes
involved with mtDNA such as DNA polymerase c and
the mitochondrial RNA polymerase likely act exclusively in
mitochondria, other enzymes and DNA binding factors,
including human N-glycosylase hOGG1 [49,50], DNA
ligase III [51], DNA topoisomerase IIIa [20], Ku80 [52],
MDDX28 RNA helicase [53], RXRa [54] and the yeast
N-glycosylase Ntg1p [55], apparently function in both mito-
chondria and nuclei. In the case of the hOGG1 N-glycosy-
lase, alternative splicing of its gene transcript can produce
enzyme forms that possess either nuclear or mitochondrial
targeting signals. For DNA ligase III [51] and DNA
topoisomerase IIIa [20], this is achieved by choosing
alternative transcription start sites. In the case of RXRa,
yeast DNA helicase Hmi 1p [56], and Ku80 [52] proteolytic
cleavage of the full-length nuclear protein creates a
truncated form that can selectively translocate into mito-
chondria. For RXRa, this cleavage is carried out by a
cytoplasmic or mitochondrial-associated protease, m-cal-
pain that proteolytically removes a segment of the
N-terminus, thereby apparently exposing a mitochondrial
targeting sequence. Although most mitochondrial prepro-

teins possess cleavable targeting signals at the N-terminus
and translocate into mitochondria in the N- to C-terminal
direction, yeast DNA helicase Hmil 1p [56], and likely Ku80
[52] transport into mitochondria via cleavable, C-terminal
targeting signals and translocate in the C- to N-terminal
direction [57].
The mechanism for targeting DNA topoisomerase IIb to
mitochondria is at this point speculative. We suspect that
removal of the C-terminal region, needed to at least
eliminate the nuclear localization signal, probably plays an
essential role in the mechanism. Whether this occurs by
alternative splicing or proteolytic processing by a calpain-
like protease or other activity prior to import remains
unclear. The human DNA topoisomerase IIb gene, TOP2B,
contains 36 exons [58,59]. Two major splice variants,
producing full-length polypeptides of 1621 and 1626 amino
acids, have been defined. If there is an additional splice
variant for the mitochondrial-targeted enzyme, we expect
that it may be exceedingly difficult to identify on standard
Northern blots as only very low levels of DNA topo IIb
appear directed to mitochondria. In cardiac myocytes, we
estimate that there is only one topo IIb enzyme molecule
per several mtDNA circles. This equates to a few hundred to
perhaps no more than a thousand enzyme molecules per
cell. As mentioned, most, but not all proteins targeted to
mitochondria contain a discrete mitochondrial targeting
sequence, usually near the N-terminus but occasionally at
the C-terminus. Targeting sequences usually comprise a
helices in which positive charged and hydrophobic residues
align on opposite faces of the helical axis [60,61]. We have

been unable to find any obvious amphipathic a-helix
targeting signal in the N-terminal region of human DNA
topoisomerase IIb.
MITOPROT
II, a program to find such
N-terminal signals, predicts only a low probability that
topo IIb is mitochondrial (0.02) [62]. Similar results have
been obtained with the
PREDOTAR
program (http://www.
inra.fr/Internet/Produits/Predotar/) for identifying putative
mitochondrial and plastid targeting sequences [63]. How-
ever, analysis of the entire protein sequence of human
topo IIb using the
STABLECOIL
program (http://biomol.
uchsc.edu/researchFacilities/computationalCore/stablecoil/
index.html) [64] indicates that topo IIb enzyme does possess
a strong amphipathic a helix. This structure lies between
amino acid residues 1034 and 1060, within 100–200 residues
from what is the C-terminus of the truncated, mitochondrial
enzyme. Of interest, a near identical amphipathic a-helix
structure can also be identified at this site in the amino acid
sequence of topo IIa. Unfortunately, we do not know yet
whether this amphipathic a-helix serves as an authentic
mitochondrial targeting signal in vivo. Possibly, removal of
the C-terminal domain from the nuclear enzyme exposes
this amphipathic helix and this allows the enzyme to be
imported into mitochondria, in the C- to N-terminal
direction. Of interest, removal of the C-terminal domain

of the nuclear topo IIb also raises the pI of the enzyme from
8.8 to about 9.7 (assuming the truncated enzyme is 1378
amino acids in length). Possibly, this could be important for
mitochondrial targeting as well. For many proteins impor-
ted into mitochondria, the pI of the mitochondrial form of
the protein is higher than that of the cytosolic protein form
[65,66].
It will be surprising if the mitochondrial form of DNA
topoisomerase IIb does not assume an important role in
mtDNA replication. By analogy to that of other eucaryotic
Ó FEBS 2003 DNA Topo IIb mammalian mtDNA (Eur. J. Biochem. 270) 4183
type II enzymes, the mitochondrial type IIb activity may
serve to de-catenate newly replicated mtDNA circles from
one another at the end of a cycle of mtDNA synthesis. In
addition, the enzyme could support a structural role, serving
to help attach the mtDNA replication complex on mtDNA
to a specific site on the inner membrane. Although only a
low level of the type II b-isozyme is likely targeted to
mitochondria, an involvement of DNA topoisomerase IIb
in mtDNA replication reveals yet a new role for the b
isozyme within the cell, and could help explain why the
enzyme is widely distributed among different types of
tissues, and not like the a isozyme that is only restricted to
cells that are proliferating [7,10,11]. Whether the a iso-
enzyme is targeted to mitochondria in proliferating cells is
an important question that needs to be answered. Recently,
Holt and colleagues [67] have challenged the classic model
of mtDNA replication [1], in which replication of the
leading strand was thought to begin first, producing an
expanding displacement (D)-loop prior to the initiation of

lagging strand synthesis. Instead, synthesis of leading and
lagging strands may proceed in a tightly coupled fashion
[68]. The method to identify and isolate DNA topo IIa from
isolated complexes of mtDNA and mtDNA replicative
factors that we describe in present study should guide the
search for additional DNA replicative enzymes needed to
reconstitute faithful mtDNA replication in vitro. This
should help us understand how the distinct modes of
mtDNA replication occur and the role that DNA topo IIa
serves in this process.
Acknowledgments
We thank Dr Mark Duncan and the Biochemical Mass Spectrometry
Facility for advice and providing resources, Lisa Litzenberger for help
preparing figures, Carole Witka for typing the manuscript and Drs
Robert Hodges, Brian Tripet, and Stephen Lu for assessing topo II for
amphipathic helical structure using
STABLECOIL
and
MITOPROT
pro-
grams.
References
1. Clayton, D.A. (1991) Replication and transcription of vertebrate
mitochondrial DNA. Ann. Rev. Cell Biol. 7, 453–478.
2. Chomyn, A., Mariottini, P., Cleeter, M.W., Ragan, C.I., Mat-
suno-Yagi,A.,Hatefi,Y.,Doolittle,R.F.&Attardi,G.(1985)Six
unidentified reading frames of human mitochondrial DNA encode
components of the respiratory-chain NADH dehydrogenase.
Nature 314, 592–597.
3. Farr, C.L., Yang, Y. & Kaguni, L.S. (1999) Functional inter-

actions of mitochondrial DNA polymerase and single-stranded
DNA-binding protein: template-primer DNA binding and initia-
tion and elongation of DNA strand synthesis. J. Biol. Chem. 274,
14779–14785.
4. Carrodeguas, J.A., Kobayashi, R., Lim, S.E., Copeland, W.C. &
Bogenhagen, D.E. (1999) The accessory subunit of Xenopus laevis
mitochondrial DNA polymerase gamma increases processivity of
the catalytic subunit of human DNA polymerase gamma and is
related to class II aminoacyl-tRNA synthetases. Mol. Cell Biol. 19,
4039–4046.
5. Graziewicz, M.A., Day, B.J. & Copeland, W.C. (2002) The
mitochondrial DNA polymerase as a target of oxidative damage.
Nucleic Acids Res. 30, 2817–2824.
6. Johnson, A.A., Tsai, Y.C., Graves, S.W. & Johnson, K.A. (2000)
Human mitochondrial DNA polymerase holoenzyme: recon-
stitution and characterization. Biochemistry 39, 1702–1708.
7. Wang, J.C. (1996) DNA topoisomerases. Annu. Rev. Biochem. 65,
635–692.
8. Brown, P.O. & Cozzarelli, N.R. (1979) A sign inversion mechanism
for enzymatic supercoiling of DNA. Science 206, 1081–1083.
9. Liu, L.F., Liu, C C. & Alberts, B.M. (1980) Type II DNA
topoisomerases: enzymes that can unknot a topologically knotted
DNA molecule via a reversible double-strand break. Cell 19,
697–707.
10. Watt, P.M. & Hickson, I.D. (1994) Structure and function of type
II DNA topoisomerases. Biochem. J. 303, 681–695.
11. Austin, C.A. & Marsh, K.L. (1998) Eukaryotic DNA topo-
isomerase II beta. Bioessays 20, 215–226.
12. Chen, A.Y. & Liu, L.F. (1994) DNA topoisomerases: essential
enzymes and lethal target. Annu. Rev. Pharmacol. Toxicol. 34,

191–218.
13. Zhang, H., Barcelo
´
, J.M., Lee, B., Kohlhagen, G., Zimonjic, D.B.,
Popescu, N.C. & Pommier, Y. (2001) Human mitochondrial
topoisomerase I. Proc. Natl Acad. Sci. USA 98, 10608–10613.
14. Brun, G., Vannier, P., Scovassi, I. & Callen, J.C. (1981) DNA
topoisomerase I from mitochondria of Xenopus laevis oocytes.
Eur. J. Biochem. 118, 407–415.
15. Fairfield, F.R., Bauer, W.R. & Simpson, M.V. (1985) Studies on
mitochondrial type I topoisomerase and on its function. Biochim.
Biophys. Acta 824, 45–57.
16. Lin, J.H. & Castora, F.J. (1995) Response of purified mitochon-
drial DNA topoisomerase I from bovine liver to camptothecin and
m-AMSA. Arch. Biochem. Biophys. 324, 293–299.
17.Lazarus,G.M.,Henrich,J.P.,Kelly,W.G.,Schmitz,S.A.&
Castora, F.J. (1987) DNA topoisomerase I from calf thymus
mitochondria is associated with a DNA binding, inner membrane
protein. Biochemistry 26, 6195–6203.
18. Tua, A., Wang, J., Kulpa, V. & Wernette, C.M. (1997)
Mitochondrial DNA topoisomerase I of Saccharomyces cerevi-
siae. Biochemie 79, 341–350.
19. Kosovsky, M.J. & Soslau, G. (1991) Mitochondrial DNA topo-
isomerase I from human platelets. Biochim. Biophys. Acta 1078,
56–62.
20. Wang, Y., Lyu, Y.L. & Wang, J.C. (2002) Dual localization of
human DNA topoisomerase IIIa to mitochondria and nucleus.
Proc.NatlAcad.Sci.USA99, 12114–12119.
21. Melendy, T. & Ray, D.S. (1989) Novobiocin affinity purification
of a mitochondrial type II topoisomerase from the trypanoso-

matid Crithidia fasciculate. J. Biol. Chem. 264, 1870–1876.
22. Shapiro, T.A. & Englund, P.A. (1995) The structure and rep-
lication of kinetoplast DNA. Annu. Rev. Microbiol. 49, 117–
143.
23. Melendy, T., Sheline, C. & Ray, D.S. (1988) Localization of a type
II DNA topoisomerase to two sites at the periphery of the kine-
toplast DNA of Crithidia fasciculate. Cell 55, 1083–1088.
24. Shapiro, T.A., Klein, V.A. & Englund, P.T. (1989) Drug-pro-
moted cleavage of kinetoplast DNA minicircles: evidence for type
II topoisomerase activity in trypanosome mitochondria. J. Biol.
Chem. 264, 4173–4178.
25. Ray, D.S., Hines, J.C. & Anderson, M. (1992) Kinetoplast-asso-
ciated DNA topoisomerase in Crithidia fasciculate: crosslinking of
mitochondrial topoisomerase II to both minicircles and maxi-
circles in cells treated with the topoisomerase inhibitor VP16.
Nucleic Acids Res. 20, 3353–3356.
26. Wang, Z. & Englund, P.T. (2001) RNA interference of a trypa-
nosome topoisomerase II causes progressive loss of mitochondrial
DNA. EMBO J. 20, 4674–4683.
4184 R. L. Low et al. (Eur. J. Biochem. 270) Ó FEBS 2003
27. Komori, K., Kuroe, K., Yanagisawa, K. & Tanaka, Y. (1997)
Cloning and characterization of the gene encoding a mitochond-
rially localized DNA topoisomerase II in Dictyostelium dis-
coideum. Biochim. Biophys. Acta 1352, 63–72.
28. Chavalitshewinkoon-Petmitr, P., Worasing, R. & Wilairat,
P. (2001) Partial purification of mitochondrial DNA topoiso-
merase II from Plasmodium falciparum and its sensitivity to
inhibitors. Southeast Asian J. Trop. Med. Public Health 32, 733–
738.
29. Castora, F.J., Lazarus, G.M. & Kunes, D. (1985) The presence of

two mitochondrial DNA topoisomerases in human acute leuke-
mia cells. Biochem. Biophys. Res. Commun. 130, 854–866.
30. Lin, J H. & Castora, F.J. (1991) DNA topoisomerase II from
mammalian mitochondria is inhibited by the antitumor drugs,
m-AMSA and VM-26. Biochem. Biophys. Res. Commun. 176,
690–697.
31. Lawrence, J.W., Claire, D.C., Weissig, V. & Rowe, T.C. (1996)
Delayed cytotoxicity and cleavage of mitochondrial DNA in
ciprofloxacin-treated mammalian cells. Mol. Pharmacol. 50, 1178–
1188.
32. Mita,S.,Rizzuto,R.,Moraes,C.T.,Shanske,S.,Arnaudo,E.,
Fabrizi, G.M., Koga, Y., DiMauro, S. & Shon, E.A. (1990)
Recombination via flanking direct repeats is a major cause of
large-scale deletions of human mitochondrial DNA. Nucleic Acids
Res. 18, 561–567.
33. Blok, R.B., Thorburn, D.R., Thompson, G.N. & Dahl, H H.M.
(1995) A topoisomerase II cleavage site is associated with a novel
mitochondrial DNA deletion. Hum. Genet. 95, 75–81.
34. Nielsen-Preiss, S.M. & Low, R.L. (2000) Identification of a beta-
like DNA polymerase activity in bovine heart mitochondria. Arch.
Biochem. Biophysc. 374, 229–240.
35. Laemmli, U.K. (1970) Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature 227,
680–685.
36. Shevchenko, A., Wilm, M., Vorm, O. & Mann, M. (1996) Mass
spectrometric sequencing of proteins from silver-stained poly-
acrylamide gels. Anal. Chem. 68, 850–858.
37. Schnaitman, C. & Greenawalt, J.W. (1968) Enzymatic properties
of the inner and outer membranes of rat liver mitochondria. J. Cell
Biol. 38, 158–175.

38. Tewey, K.M., Chen, G.L., Nelson, E.M. & Liu, L.F. (1984)
Intercalative antitumor drugs interfere with the breakage-reunion
reaction of mammalian DNA topoisomerase II. J. Biol. Chem.
259, 9182–9187.
39. Drake, F.H., Zimmerman, J.P., McCabe, F.L., Bartus, H.F., Per,
S.R., Sullivan, D.M., Ross, W.E., Mattern, M.R., Johnson, R.K.,
Crooke, S.T. & Mirabelli, C.K. (1987) Purification of
topoisomerase II from amsacrine-resistant P388 leukemia cells:
evidence for two forms of the enzyme. J. Biol. Chem. 262,
16739–16747.
40. Drake, F.H., Hofman, G.A., Bartus, H.F., Mattern, M.R.,
Crooke, S.T. & Mirabelli, C.K. (1989) Biochemical and pharma-
cological properties of p170 and p180 forms of topoisomerase II.
Biochemistry 28, 8154–8160.
41. Halligan, B.D., Edwards, K.A. & Liu, L.F. (1985) Purification
and characterization of a type II DNA topoisomerase from bovine
calf thymus. J. Biol. Chem. 260, 2475–2482.
42. Austin, C.A., Barot, H., Margerrison, E.E.C., Hayes, M.V. &
Fisher, L.M. (1989) Biochemical and immunological character-
ization of mammalian DNA topoisomerase II. Biochem. Soc.
Trans. 17, 528–529.
43. Beaufay, H. (1972) Lysosomes, pp. 1–45. Elsevier, North-Holland,
New York.
44. Quaglino, D. & Ronchetti, I.P. (2001) Cell death in the rat thymus:
a minireview. Apoptosis 6, 389–401.
45. Austin, C.A., Marsh, K.L., Wasserman, R.A., Willmore, E.,
Sayer, P.J., Wang, J.C. & Fisher, L.M. (1995) Expression, domain
structure, and enzymatic properties of an active recombinant
human DNA topoisomerase II beta. J. Biol. Chem. 270, 15739–
15746.

46. Shiozaki, K. & Yanagida, M. (1992) Functional dissection of the
phosphorylated termini of fission yeast DNA topoisomerase II.
J. Cell. Biol. 119, 1023–1036.
47. Crenshaw, D.G. & Hsieh, T S. (1993) Function of the hydro-
philic carboxyl terminus of type II DNA topoisomerase from
Drosophila melanogaster. I. In vitro studies. J. Biol. Chem. 268,
21328–21334.
48. Christensen, M.O., Larsen, M.K., Barthelmes, H.U., Hock, R.,
Andersen,C.L.,Kjeldsen,E.,Knudsen,B.R.,Westergaard,O.,
Boege, F. & Mielke, C. (2002) Dynamics of human DNA
topoisomerases IIa and IIa in living cells. J. Cell Biol. 157, 31–44.
49.Nishioka,K.,Ohtsubo,T.,Oda,H.,Fujiwara,T.,Kang,D.,
Sugimachi, K. & Nakabeppu, Y. (1999) Expression and differ-
ential intracellular localization of two major forms of human
8-oxoguanine DNA glycosylase encoded by alternatively spliced
OGG1 mRNAs. Mol. Biol. Cell 10, 1637–1652.
50. Takao, M., Aburatani, H., Kobayashi, K. & Yasui, A. (1998)
Mitochondrial targeting of human DNA glycosylases for repair of
oxidative DNA damage. Nucleic Acids Res. 26, 2917–2922.
51. Lakshmipathy, U. & Campbell, C. (1999) The human DNA ligase
III gene encodes nuclear and mitochondrial proteins. Mol. Cell
Biol. 19, 3869–3876.
52. Coffey, G. & Campbell, C. (2000) An alternate form of Ku80
is required for DNA end-binding activity in mammalian
mitochondria. Nucleic Acids Res. 28, 3793–3800.
53. Valgardsdottir,R.,Brede,G.,Eide,L.G.,Frengen,E.&Prydz,H.
(2001) Cloning and characterization of MDDX28, a putative
DEAD-box helicase with mitochondrial and nuclear localization.
J. Biol. Chem. 276, 32058–32063.
54. Casas,F.,Daury,L.,Grandemange,S.,Busson,M.,Seyer,P.,

Hatier,R.,Carazo,A.,Cabello,G.&Wrutniak-Cabello,C.
(2003) Endocrine regulation of mitochondrial activity: involve-
ment of truncated RXRa and c-Erb Aa1proteins.FASEB J. 17,
426–436.
55. You, H.J., Swanson, R.L., Harrington, C., Corbett, A.H., Jinks-
Robertson, S., Senturker, S., Wallace, S.S., Boiteux, S., Dizdaro-
glu, M. & Doetsch, P. (1999) Saccharomyces cerevisiae Ntg1p and
Ntg2p: broad specificity N-glycosylases for the repair of oxidative
DNA damage in the nucleus and mitochondria. Biochemistry 38,
11298–11306.
56. Lee, C.M., Sedman, J., Neupert, W. & Stuart, R.A. (1999) The
DNA helicase, Hmi1p, is transported into mitochondria by a
C-terminal cleavable targeting signal. J. Biol. Chem. 274, 20937–
20942.
57. Folsch,H.,Baume,B.,Brunner,M.,Neupert,W.&Stuart,R.A.
(1998) C- to N-terminal translocation of preproteins into
mitochondria. EMBO J. 17, 6508–6515.
58. Lang, A.J., Mirski, S.E.L. & Cummings, H.J., YuQ., Gerlach,
J.H. & Cole, S.P.C. (1998) Structural organization of the human
TOP2A and TOP2B genes. Gene 21, 255–266.
59.Sng,J H.,Heaton,V.J.,Bell,M.,Maini,P.,Austin,C.A.&
Fisher, L.M. (1999) Molecular cloning and characterization of the
human topoisomerase IIa and IIb genes: evidence for isoform
evolution through gene duplication. Biochim. Biophys. Acta 1444,
395–406.
60. von Heijne, G. (1986) Mitochondrial targeting sequences may
form amphiphilic helices. EMBO J. 5, 1335–1342.
61. Roise, D., Theiler, F., Horvath, S.J., Tomich, J.M., Richards,
J.H., Allison, D.S. & Schatz, G. (1988) Amphiphilicity is essential
for mitochondrial presequence function. EMBO J. 7, 649–653.

Ó FEBS 2003 DNA Topo IIb mammalian mtDNA (Eur. J. Biochem. 270) 4185
62. Claros, M.G. & Vincens, P. (1996) Computational method to
predict mitochondrially imported proteins and their targeting
sequences. Eur. J. Biochem. 241, 779–786.
63. Peeter, N. & Small, I. (2001) Dual targeting to mitochondria and
chloroplasts. Biochim. Biophys. Acta 1541, 54–63.
64. Zhou, N.E., Monera, O.D., Hodges, C.M. & Hodges, R.S. (1994)
Alpha-helical propensities of amino acids in the hydrophobic face
of an amphipathis alpha-helix. Protein Pept. Lett. 1, 114–119.
65. Hartmann, C. & Christen, P. (1991) Mitochondrial protein
charge. Nature 352, 762–763.
66. Jaussi, R. (1995) Homologous nuclear-encoded mitochondrial and
cytosolic isoproteins: a review of structure, biosynthesis, and
genes. Eur. J. Biochem. 228, 551–561.
67. Holt, I.J., Lorimer, H.E. & Jacobs, H.T. (2000) Coupled leading-
and lagging-strand synthesis of mammalian mitochondrial DNA.
Cell 100, 515–524.
68. Yang,M.Y.,Bowmaker,M.,Reyes,A.,Vergani,L.,Angeli,P.,
Gringeri, E., Jacobs, H.T. & Holt, I.J. (2002) Biased incorpora-
tion of ribonucleotides on the mitochondrial 1-strand account
for apparent strand-asymmetric DNA replication. Cell 111,
495–505.
69. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular
Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor
Laboratory Press, Cold Spring Harbor, New York.
4186 R. L. Low et al. (Eur. J. Biochem. 270) Ó FEBS 2003