Download DmTTF, a novel mitochondrial transcription termination factor that

Nucleic Acids Research, 2003, Vol. 31, No. 6 1597±1604
DOI: 10.1093/nar/gkg272
DmTTF, a novel mitochondrial transcription
termination factor that recognises two sequences of
Drosophila melanogaster mitochondrial DNA
Marina Roberti1, Paola Loguercio Polosa1, Francesco Bruni1, Clara Musicco2,
Maria Nicola Gadaleta1,2 and Palmiro Cantatore1,2,*
1
Dipartimento di Biochimica e Biologia Molecolare, UniversitaÁ di Bari and 2Istituto di Biomembrane e Bioenergetica,
CNR, Via Orabona 4, 70125 Bari, Italy
Received December 18, 2002; Revised and Accepted January 28, 2003
ABSTRACT
Using a combination of bioinformatic and molecular
biology approaches a Drosophila melanogaster
protein, DmTTF, has been identi®ed, which exhibits
sequence and structural similarity with two mitochondrial transcription termination factors, mTERF
(human) and mtDBP (sea urchin). Import/processing
assays indicate that DmTTF is synthesised as a
precursor of 410 amino acids and is imported into
mitochondria, giving rise to a mature product of 366
residues. Band-shift and DNase I protection experiments show that DmTTF binds two homologous,
short, non-coding sequences of Drosophila mitochondrial DNA, located at the 3¢ end of blocks of
genes transcribed on opposite strands. The location
of the target sequences coincides with that of two of
the putative transcription termination sites previously hypothesised. These results indicate that
DmTTF is the termination factor of mitochondrial
transcription in Drosophila. The existence of two
DmTTF binding sites might serve not only to stop
transcription but also to control the overlapping of
a large number of transcripts generated by the
peculiar transcription mechanism operating in this
organism.
INTRODUCTION
The regulation of the expression of mitochondrial genes in
animals has been studied carefully, especially in mammals (1).
In these organisms, mitochondrial DNA (mtDNA) transcription is regulated at the level of initiation and termination.
Three protein factors, TFAM (2), TFB1M and TFB2M (3), are
required for the initiation of mitochondrial transcription at the
two promoters (HSP and LSP) placed in the main non-coding
region (NCR); the DNA-binding protein mTERF (4) mediates
termination of transcription of the ribosomal unit, accounting
for the production of the 16S and 12S rRNAs and of two
DDBJ/EMBL/GenBank accession no. AY196479
tRNAs. Such a mode of regulation does not seem to be
conserved during evolution due to remarkable variations in the
gene order of mitochondrial genomes in the various phyla (5).
In sea urchins (echinoderms) the two ribosomal genes are not
adjacent; the main NCR has a reduced size and is placed
downstream of the small ribosomal gene; ®ve short AT-rich
regions probably act as bidirectional promoters (6). In this
organism, mitochondrial transcription takes place via multiple
and partially overlapping transcription units (7). Control at the
level of transcription initiation, depending on a TFAM-like
protein, seems to be absent, as no such factor was ever
detected in sea urchins. On the contrary, there is evidence that
transcription is regulated at the level of termination. In fact, a
DNA-binding protein, mtDBP, identi®ed in Paracentrotus
lividus mitochondria, is able to arrest elongating RNA
polymerase by contacting two regions of mtDNA, located in
the NCR and at the boundary of ND5 and ND6 genes,
respectively (8±10). This can allow the control of the
overlapping of the large number of transcripts starting from
the multiple transcription initiation sites. Although mtDBP
and mTERF contact mtDNA in different regions, they share
the same role as transcription termination factor and also show
a signi®cant sequence similarity that suggests a common
evolutionary origin for the two proteins.
Drosophila melanogaster is an interesting system for
studying mitochondrial biogenesis due to the knowledge of
the entire genome sequence (11) and to the possibility of
testing the effect of altered expression of genes at the
phenotypic and molecular levels. mtDNA gene organisation
of D.melanogaster (12) is different from that of echinoderms
and vertebrates: in particular, differences between mammalian
and Drosophila gene order concern the translocation and/or
inversion of two blocks of genes (ND4L/ND4/ND5 and 12S/
16S/ND1) and of the ND6 gene. As a consequence,
Drosophila mitochondrial genes form clusters alternatively
distributed on the two strands and, contrary to sea urchins, the
two ribosomal genes map in adjacent positions. Information
on the transcription mechanism is still very limited. An early
paper by Berthier et al. (13) suggested the existence of one
transcription unit and one transcription termination site for
*To whom correspondence should be addressed at Dipartimento di Biochimica e Biologia Molecolare, UniversitaÁ di Bari, Via Orabona 4, 70125 Bari, Italy.
Tel: +39 080 5443378; Fax: +39 080 5443403; Email: [email protected]
1598
Nucleic Acids Research, 2003, Vol. 31, No. 6
MATERIALS AND METHODS
a Protease Inhibitor's Cocktail (Sigma). The suspension was
pulse-sonicated for 90 s, centrifuged for 15 min at 15 000 g at
4°C; the supernatant was adjusted to a glycerol concentration
of 15% and stored at ±80°C. The extract was diluted with
100 mM KCl, 50 mM Tris±HCl pH 8.0, 1% Tween-20, 20%
glycerol, 1 mM EDTA, 1 mM DTT, and used in DNA-binding
reactions. For footprinting assays, bacterial DmTTF was
puri®ed by heparin±Sepharose chromatography. For this
purpose, a protein extract from 200 ml of bacterial culture
was prepared as above. The supernatant was diluted to 150 mM
NaCl and loaded onto 6 ml of heparin±Sepharose CL-6B
(Amersham Pharmacia Biotech) equilibrated with a buffer
containing 150 mM KCl, 10 mM Tris±HCl pH 8.0, 10 mM
MgCl2, 1 mM EDTA, 10% glycerol, 1 mM DTT and protease
inhibitors (Sigma). DmTTF was eluted with 500 mM KCl
buffer.
Cloning, expression and puri®cation of DmTTF
Import of DmTTF into rat mitochondria
DmTTF cDNA was obtained by means of PCR ampli®cation
on a D.melanogaster cDNA library from 2±14-h embryos.
Primers were BG-For (ATGATTAGAAGCCTTCTGCGCAGCT) and BG-Rev (TCATCCTTCTGATACACTTTGCGGA), nucleotide positions 118 039±118 015 and 116 688±
116 712 of the genomic sequence (AE003643.2), respectively.
The ampli®cation product (1233 bp) was cloned into the
vector pGEMT (Promega) and sequenced.
To produce in vitro the full-length precursor form of
DmTTF (DmTTFp), which includes the mitochondrial leader
sequence, and the mature truncated version, which lacks the
N-terminal 44 amino acids, two versions of the cDNA were
cloned into pBluescript II KS vector between XhoI and BamHI
sites. To generate the construct DmTTFp, the forward primer
was DmTTFp-For (GCCGCTCGAGACCATGGTTAGAAGCCTTCTGCGCAGCT), containing an XhoI site (underlined) and the ATG codon (bold) in the optimal consensus
context; the reverse primer was DmTTF-Rev (GCGCGGATCCTCATCCTTCTGATACACTTTG), containing a
BamHI site (underlined) and the stop codon (bold). To
generate the construct DmTTF (lacking the presequence), the
forward primer was DmTTF-For (GCCGCTCGAGACCATGGCCCCCAGTGGCACTTTAGGTAGC), containing
an XhoI site (underlined) and an arti®cial ATG codon (bold) in
the optimal consensus context; the reverse primer was
DmTTF-Rev. These two cDNA constructs were used to
programme the in vitro reticulocyte transcription-translation
system TNTâ (Promega) in the presence of [35S]methionine,
according to the supplier's instructions.
The mature version of DmTTF was expressed in bacteria
using the pCRâ T7 TOPO TA Expression Kit (Invitrogen).
The DmTTF construct was obtained by PCR with primers BG
TOPO-For (ATGACCCCCAGTGGCACTTTAGGT), containing an arti®cial initiation codon (bold), and BG TOPORev
(GCCGCTCGAGTCATCCTTCTGATACACTTTG),
containing the stop codon (bold); the product was cloned
into the pCRâ T7/CT TOPO vector. The recombinant plasmid
was used to transform Escherichia coli BL21(DE3)pLys S
cells. Bacteria were grown at 37°C to an OD600 of 0.6 and
induced with isopropyl-b-D-thiogalactopyranoside for 3 h.
The protein extract was prepared by resuspending cells from
30 ml of culture in 2.5 ml of 10 mM phosphate buffer, pH 7.4,
500 mM NaCl, 2 mM EDTA, 1 mM DTT containing 120 ml of
The in vitro produced 35S-labelled DmTTFp was incubated for
import by rat liver mitochondria, prepared according to the
procedure of Barile et al. (17). The radioactive protein present
in 7.5 ml of the reticulocyte system was incubated with 0.4 mg
of mitochondrial proteins in 70 mM sucrose, 220 mM
mannitol, 2 mM HEPES, pH 7.4 (HMS buffer), supplemented
with 2 mM MgCl2, 5 mM succinate, 2 mM phosphate, 1 mM
methionine, in a ®nal volume of 50 ml, at 30°C for 30 min. In
some experiments carbonylcyanide-p-tri¯uoromethoxyphenyl
hydrazone (FCCP) was included in the reaction mixture to
1.25 mM. After import/processing reactions, trypsin was added
to the incubation mixture at 0.05 mg/ml and the samples were
incubated on ice for 15 min, then the addition of 1.5 mg/ml of
soybean trypsin inhibitor followed. After 5 min of incubation,
mitochondria were diluted with 1.5 ml of HMS buffer and
centrifuged; pellets were lysed in 13 Laemmli sample buffer.
In a control experiment, the mitochondrial pellet was
resuspended in 12 ml of HMS buffer and Triton X-100 was
added to a ®nal concentration of 3.5%; trypsin treatment was
followed as described above.
Proteins were analysed by SDS±PAGE on 12% polyacrylamide gels and the radioactive bands were visualised with a
Typhoon Phosphor Imaging System (Molecular Dynamics).
each gene cluster. More recently, protein factors homologous
to mammalian TFAM and TFBM have been identi®ed
(14±16).
Here, in order to gain further information on mtDNA
transcription in D.melanogaster, we investigated the existence
of a transcription termination factor. Taking advantage of the
structural and functional similarity of human and sea urchin
transcription termination factors mTERF and mtDBP, we
identi®ed a Drosophila gene product homologous to both
factors. The molecular characterisation of this product, that
recognises two target sites on mtDNA, strongly suggests that it
acts as transcription termination factor in Drosophila
mitochondria.
Assay of DmTTF DNA-binding activity
The DNA-binding activity of DmTTF was studied by means
of electrophoretic mobility-shift assays (EMSA) using the
E.coli protein extract containing the recombinant protein.
Probes were DNA fragments obtained by PCR and labelled
with T4 polynucleotide kinase and [g-32P]ATP. The protein
extract was incubated with ~20 fmol of probe at 25°C for
30 min, in 20 ml of reaction mixture containing 20 mM
HEPES±KOH pH 7.9, 5 mM MgCl2, 75 mM KCl, 1 mM DTT,
0.1 mM EDTA, 2 mg of BSA and 2 mg of poly(dI±dC)´
poly(dI±dC). Reaction mixtures were then loaded on native
6% polyacrylamide gels and run in 0.53 TBE at 300 V at
4°C. EMSA were also used to determine the stability of
DNA±DmTTF complexes. Radioactive bands were visualised
as before; quantitative analysis of data was performed using
the Image Quant software (Molecular Dynamics).
For DNase I footprinting experiments, DNA fragments
were 5¢ labelled and incubated with the heparin±Sepharose
protein fraction in the same conditions used for EMSA, except
Nucleic Acids Research, 2003, Vol. 31, No. 6
that samples were 2.5-fold scaled up and poly(dI±dC)´
poly(dI±dC) was 10-fold lowered. After incubation at room
temperature for 10 min, reactions were added with an equal
volume of 5 mM CaCl2, 10 mM MgCl2, and with 0.5 U/ml
of DNase I (Roche). After 1 min of incubation at room
temperature, reactions were stopped by adding an equal
volume of 1% SDS, 200 mM NaCl, 20 mM EDTA, 250 mg/ml
tRNA. Samples were phenol extracted, ethanol precipitated
and loaded onto a 6% sequencing gel.
RESULTS
Identi®cation of a D.melanogaster mitochondrial protein
homologous to mtDBP and mTERF
A BLASTP analysis using as query the sequence of the sea
urchin DNA-binding protein mtDBP against a database of
translated D.melanogaster nuclear coding sequences revealed,
as the most signi®cant, a match with the gene product
BG:DS01068.4. The importance of this similarity is supported
by the observation that, when the sequence of this gene
product was tested versus the available protein databases, the
highest scores were obtained for mtDBP and mTERF. These
data suggest that the Drosophila gene product could be the
homologue of mtDBP and mTERF. Therefore, we termed the
protein DmTTF for Drosophila mitochondrial transcription
termination factor. FlyBase GadFly Genome Annotation
Database reports that the DmTTF gene is 1587 bp long; it is
composed of three exons and two introns and generates a
transcript of 1468 nt. The cDNA of DmTTF was cloned by
means of PCR on a cDNA library of D.melanogaster. The
sequence of the ORF (accession no. AY196479), which is
1233 nt long and codes for a polypeptide of 410 residues,
differs from that reported in the Drosophila database because
of 20 nucleotide substitutions giving rise to four amino acid
changes (P, A, I and T instead of S, T, V and N, in positions 46,
72, 111 and 223, respectively). The analysis of the DmTTF
protein sequence with programs predicting the cellular
localisation, such as MitoProt II, Target P and Predotar,
resulted in a very high probability for the mitochondrial
localisation, with a putative signal sequence of 44 residues
according to the (R ± 2) rule. Therefore, the mature DmTTF
(366 amino acids long) should have a calculated molecular
weight of 43 100 Da and an isolectric point of 9.43.
The alignment of the putative mature version of DmTTF
with mtDBP and mTERF (Fig. 1) shows that the similarity
among the three proteins is rather uniformly distributed along
the molecule with a higher conservation in the C-terminal
region. In particular, DmTTF displays 17.8% amino acid
identity and 43.6% amino acid similarity with mtDBP, and
18.8% amino acid identity and 36.5% amino acid similarity
with mTERF. Sequence analysis of DmTTF shows that the
protein contains a main basic domain placed between residues
210 and 221. Moreover, by using the program COILS (18), we
detected four regions that could form coiled coils. Coiled-coil
structures are also predictable in mtDBP and mTERF, as
revealed by a similar analysis. As shown in Figure 1, the
position of some of these motives is conserved in the three
proteins; in particular this concerns the fourth motif of
DmTTF and mtDBP, and the third motif of mTERF, which all
fall in the most conserved portion of the proteins. Similarly,
1599
the ®rst motif of DmTTF and mtDBP, and the third motif of
DmTTF and the second of mtDBP, also lie in similar
positions.
Mitochondrial localisation of DmTTF
To obtain direct evidence on the mitochondrial localisation of
DmTTF, we set up an import/processing assay using isolated
rat liver mitochondria. For this purpose, two versions of the
cDNA (DmTTFp and DmTTF, with and without the putative
signal sequence, respectively; Fig. 2A) were cloned in the
vector pBluescript II and used as templates in an in vitro
reticulocyte transcription/translation system. When the entire
ORF was translated (Fig. 2B, lane 1), one main product,
DmTTFp, starting at the ®rst AUG codon, was obtained.
Figure 2B (lane 2) shows that incubation of the in vitro
translated protein with isolated mitochondria, followed by
trypsin treatment, produced a faster migrating band. The
polypeptide, which was resistant to trypsin, comigrated with
the putative mature DmTTF synthesised in vitro that is 366
amino acids long (Fig. 2B, lane 5). The processed product
became protease sensitive after solubilisation of the membranes with Triton X-100 (Fig. 2B, lane 3). Moreover, the
import-processing reaction was inhibited by the uncoupler
FCCP (Fig. 2B, lane 4), as expected for a mitochondrial
import depending on membrane potential. The overall results
are typical for a mitochondrial import reaction and are
consistent with the bioinformatic prediction that DmTTF is a
mitochondrial protein of 366 amino acids.
Target sites of DmTTF on D.melanogaster mtDNA
In order to show that DmTTF is able to bind DNA, and to
identify the target sites, the protein was expressed in E.coli and
used in gel-shift experiments. For this analysis we chose four
regions of Drosophila mtDNA (see Fig. 3A) that, on the basis
of the gene organisation and transcription map (13), could
contain transcription termination sites. The region containing
a few nucleotides of the 3¢ end of the 16S rRNA gene, the
tRNALeu(CUN) gene and the 5¢ end of ND1 gene was selected by
analogy to the position of the mTERF binding site on human
mtDNA; the region comprising the tRNA genes for Met, Gln,
Ile and a few nucleotides of the adjacent ND2 gene might
contain the termination site of the complete transcript coded
by the rRNA template strand; the region comprising the tRNA
cluster and the 3¢ end of ND3 and ND5 genes, and that
comprising the tRNASer(UCN) gene and the 3¢ end of ND1 and
cyt b genes, might both contain bidirectional termination sites
of gene clusters transcribed on opposite strands. The ND3/
ND5 region was divided into four fragments. Mobility-shift
experiments with fragments containing the ®rst two regions
did not produce any DNA±protein complex (data not shown).
When the four fragments of the ND3/5 region were used, a
slower migrating band was observed with fragment ND3/5-3
(Fig. 3B, lanes 1±3). The fragment, 261 bp long, contained
most of the tRNAAsn gene, the tRNASer and tRNAGlu genes,
an 18-nt-long non-coding sequence, and almost the entire
tRNAPhe gene. The protein±DNA complex was speci®c, as it
was unaffected by the addition of a heterologous competitor,
that is, the fragment containing the 3¢ region of 16S rRNA
gene (lanes 4 and 5). The displaced band disappeared when the
binding reaction contained an excess of cold ND3/5-3
fragment (lanes 6 and 7). The protein±DNA complex was
1600
Nucleic Acids Research, 2003, Vol. 31, No. 6
Figure 1. Clustal W (22) alignment of the putative mature form of DmTTF (AY196479) with sea urchin mtDBP (AJ011076) and human mTERF (Y09615).
The alignment was performed at the EMBnet site using the GONNET matrix. Amino acids that are identical or similar in two sequences out of three are
shaded in black and grey, respectively. Shaded bars show the regions that could form coiled-coil structures.
produced by the protein in the mature form, whereas the
precursor form showed a weak binding activity (data not
shown). This is in agreement with the behaviour of other
mtDNA-binding proteins which are unable to bind DNA as
precursor forms (9). The mtDNA sequence contacted by
DmTTF was determined by DNase I footprinting analysis
using recombinant DmTTF puri®ed by heparin±Sepharose
chromatography. As shown in Figure 3C, the pattern of DNase
I digestion of the ND3/5-3 fragment revealed a protected
region of 28 bp, from positions 6314 to 6341. This sequence,
which can be considered the DmTTF target site in this region,
contains the last 5 nt of the tRNAGlu gene, the entire small
non-coding sequence (nucleotides 6319±6336) and the last 5 nt
of the tRNAPhe gene. It is noteworthy that this sequence is
located at the boundary of two clusters of genes transcribed
on opposite strands, an ideal position for a transcription
termination factor to bind DNA and exert its function.
DmTTF was able to also bind the cyt b/ND1 fragment
(360 bp) (Fig. 3D, lanes 1±3). As expected, the binding was
abolished by the addition of an excess of cold ND3/5-3
fragment, which contains the DmTTF target site located
between tRNAGlu and tRNAPhe (lanes 4 and 5). Similarly, the
binding to ND3/5-3 probe was abolished by the addition of an
excess of cold cyt b/ND1 fragment (lanes 6±10). Interestingly,
also the cyt b/ND1 region contains a short non-coding
sequence (nucleotides 11 703±11 719), located between the
genes for tRNASer(UCN) and ND1, which displays a signi®cant
homology with that between tRNAGlu and tRNAPhe genes. In
order to test whether DmTTF is able to recognise this 17 nt
sequence, DNase I footprinting analysis was carried out on the
cyt b/ND1 fragment (Fig. 3E). The protected region (28 bp)
spans from positions 11 698 to 11 725, and contains the last
5 nt of the tRNASer(UCN) gene, the entire non-coding sequence
and the last 6 nt of the ND1 gene. These results show that
Nucleic Acids Research, 2003, Vol. 31, No. 6
1601
DISCUSSION
Figure 2. Import into mitochondria and processing of DmTTF precursor.
(A) Schematic representation of the DmTTF precursor (DmTTFp) and
mature form (DmTTF) constructs. The shaded region represents the
mitochondrial presequence. (B) SDS±PAGE analysis of the translation
product of the two constructs, analysed directly (lanes 1 and 5) or after
incubation with rat liver mitochondria in different conditions (lanes 2±4).
DmTTF is able to contact two homologous sequences of
D.melanogaster mtDNA, located in two distinct regions.
To compare the relative strength of binding at the two target
sites, the dissociation rate of the preformed DNA±DmTTF
complexes was estimated. In this experiment, the radioactive
probe was ®rst incubated with the induced bacterial extract,
then a 100-fold molar excess of unlabelled homologous
competitor was added to each sample. After different times of
incubation, reactions were loaded on a non-denaturing gel and
the dissociation rate of the complex was monitored by
determining the rate of disappearance of the radioactive
retarded band, as shown in Figure 4. The half-life of the two
complexes was determined by analysing the results according
to the integrated pseudo-®rst-order equation (19): ln[CP] /
[CPo] = ±Kdiss 3 t, where [CP] represents the concentration of
the protein±DNA complex at time t and [CPo] that of the
complex before the addition of the cold competitor. The halflife of the complex with fragment ND3/5-3 is ~15 min,
whereas that with fragment cyt b/ND1 is ~7 min, corresponding to dissociation rates of 0.46 3 10±1 s±1 and 1.03 3 10±1 s±1,
respectively.
The termination of mtDNA transcription has a relevant role in
regulating the expression of the mitochondrial genome in
animal organisms. This process requires protein factors that
bind speci®cally DNA. Two transcription termination factors,
mTERF in mammals and mtDBP in sea urchins, have been
described to date (4,9,10). In the present work, we report the
characterisation of a D.melanogaster polypeptide, DmTTF,
identi®ed on the basis of its similarity with mTERF and
mtDBP. DmTTF can be imported and processed by rat liver
mitochondria, giving rise to a mature protein migrating as a
366 amino acid polypeptide. The factor speci®cally binds two
short, highly homologous non-coding sequences, located
between tRNAGlu and tRNAPhe, and tRNASer(UCN) and ND1
genes, respectively (Fig. 5). DmTTF±DNA complexes exhibit
a similar half-life of ~10 min at both binding sites. Data on the
stability of the protein±DNA complex for other mitochondrial
termination factors are limited to sea urchin mtDBP. In this
case, the stability of the complex is remarkably different, with
a value of 25 min at the ND5/ND6 binding site and of 150 min
at the NCR binding site (8).
The comparison between mTERF, mtDBP and DmTTF
indicates that they could share the mode of binding to DNA, as
suggested by the presence of coiled-coil structures in all of
them. It is likely that the coiled-coil heptads identi®ed in these
three proteins serve to establish intra-molecular interactions
needed to place the DNA-binding domain in the right position
for contacting DNA. This possibility is supported by the
observation that both mtDBP and mTERF bind DNA as a
monomer and not as a dimer, as it would occur if these
structures were involved in intermolecular interactions.
Drosophila mitochondrial genes are almost equally distributed on the two mtDNA strands and are organised as two pairs
of convergent blocks (Fig. 3A). An early study (13) on the
mapping of D.melanogaster mtRNAs suggested the existence
of multiple sites of transcription initiation and termination. It
is remarkable that DmTTF binding sites are located just at the
3¢ end of the convergent gene units and coincide with two of
the hypothesised transcription termination sites. These observations, all together, strongly suggest that DmTTF is the
termination factor of Drosophila mitochondrial transcription.
Similarly to mtDBP, and contrary to mTERF, DmTTF does
not bind immediately downstream of the 16S rRNA gene.
Therefore, Drosophila, along with the sea urchin, would
represent a further organism where transcription termination
mediated by a protein occurs at positions different from the 3¢
Figure 3. (Next page) DNA-binding activity of DmTTF. (A) Location on the D.melanogaster mtDNA map of the fragments used as probes in EMSA.
Positions of fragment ends on mtDNA sequence (NC_001709) are the following: AT/ND2, 1±325; ND3/5-1, 5811±6074; ND3/5-2, 6011±6200; ND3/5-3,
6139±6399; ND3/5-4, 6335±6558; cyt b/ND1, 11496±11855; ND1/16S, 12665±12796. Black boxes indicate the two fragments that are bound by DmTTF.
Arrows indicate the transcription direction of the two strands. (B) Binding of DmTTF to the ND3/5-3 fragment. The mobility-shift assay was performed using
30 000 c.p.m. (20 fmol) of 32P-labelled fragment and 5 ml of a 100-fold dilution of the protein extract, from not induced (lane 2) and induced (lanes 3±7)
bacteria, prepared as described in Materials and Methods. (C) DNase I protection analysis of the DmTTF±ND3/5-3 complex. The plus (+) strand of the
fragment was labelled with [g-32P]ATP. The fragment (30 fmol, 30 000 c.p.m.) was incubated with (+DmTTF) and without (±DmTTF) a 5-fold dilution of the
heparin±Sepharose protein fraction, in the same conditions used for EMSA, and treated with DNase I as described in Materials and Methods. The sequence
showing (A, T) refers to the plus (+) strand. (D) Binding of DmTTF to the cyt b/ND1 fragment and cross-competition with the ND3/5-3 fragment. Mobilityshift assay was performed using 30 000 c.p.m. (20 fmol) of the 32P-labelled fragment indicated as probe and 5 ml of a 100-fold dilution of the protein extract,
from not induced (lanes 2 and 7) and induced (lanes 3±5 and 8±10) bacteria. (E) DNase I protection analysis of the DmTTF±cyt b/ND1 complex. The plus
(+) strand of the fragment was labelled with [g-32P]ATP. The fragment (30 fmol, 30 000 c.p.m.) was incubated with (+DmTTF) and without (±DmTTF) a
10-fold dilution of the heparin±Sepharose protein fraction, in the same conditions used for EMSA, and treated with DNase I as described in Materials and
Methods. The sequence showing (A, T) refers to the plus (+) strand.
1602
Nucleic Acids Research, 2003, Vol. 31, No. 6
Nucleic Acids Research, 2003, Vol. 31, No. 6
Figure 4. Stability of DmTTF±DNA fragment complexes. (A) Dissociation
rate of the protein±ND3/5-3 complex. (B) Dissociation rate of the protein±cyt b/ND1 complex. To load all samples simultaneously on the gel, the
binding reactions for each fragment were set up at different times. Labelled
fragments (20 000 c.p.m., 10 fmol) were pre-incubated with 5 ml of a 100fold dilution of the induced bacterial protein extract under standard conditions in a total volume of 20 ml. Then, a 100-fold molar excess of the unlabelled homologous fragment was added to each reaction. After the indicated
times, samples were loaded on a non-denaturing 6% polyacrylamide gel in
0.5% TBE. The ®rst lane to the left reports the migration of the fragment in
the absence of the protein extract. The linear relationship between ln[CP]/
[CPo] and time for the two complexes is shown to the right of the respective
gel picture. The reported values are normalised with respect to the sum of
the intensity of bound and unbound DNA at each time point.
end of the 16S rRNA gene. This is surprising given that
ribosomal genes keep in Drosophila the same arrangement
described in mammals. Previously, Valverde et al. (20)
observed that a sequence homologous to the tridecamer
termination signal found in human downstream of the 16S
rRNA gene was well conserved, in equivalent position, in
Drosophila and in a large number of animals. This suggested
that a protein-dependent transcription termination mechanism
at the 3¢ end of the 16S rRNA gene was largely maintained
during evolution. Our ®ndings in Drosophila, as well as in the
sea urchin (21), indicate that the presence of the conserved
sequence does not necessarily imply the existence of a protein
factor able to bind such a target. Unless this sequence
represents a protein-independent termination signal, the most
likely hypothesis is that in Drosophila and in the sea urchin the
3¢ end of 16S rRNA is generated by a post-transcriptional
processing event.
The different number and location of the binding sites for
mtDBP, mTERF and DmTTF probably re¯ects remarkable
differences in transcription mechanisms. The existence of two
binding sites for DmTTF, as well as for mtDBP, as opposed to
the single mTERF target site, could be related to the peculiar
function played by these factors in the scenario of mtDNA
transcription. While in mammals mTERF controls the relative
abundance of rRNAs versus mRNAs (4), in Drosophila and in
sea urchins, where mitochondrial transcription proceeds via
multiple transcription units (7,13), DmTTF and mtDBP could
exert more than a simple stop of transcription. In fact, they
could also be involved in controlling the overlapping of the
transcripts generated by those units and in avoiding the
1603
Figure 5. Comparison of DmTTF binding sites. Genes are depicted in grey
and the two short non-coding sequences as stripped boxes. Black arrows
indicate the transcription direction. The black boxes represent the footprinted sequences (Fig. 3C and E). Square brackets on the nucleotide sequences indicate the gene boundaries.
collision between the transcription machineries moving on
opposite strands. In light of this, the termination of
mitochondrial transcription appears to constitute in these
organisms a crucial step in the control of the expression of
mitochondrial genes. In conclusion, our ®ndings support the
hypothesis that, during animal evolution, the role of the
transcription termination factor has been progressively
adjusted to meet changes in mitochondrial gene organisation
and expression which occurred in the various phyla.
ACKNOWLEDGEMENTS
We are very grateful to H. T. Jacobs for his valuable
suggestions. The D.melanogaster embryos cDNA library was
kindly provided by S. Ban® from TIGEM, Telethon Institute
of Genetics and Medicine, Milan. We are greatly indebted to
M. Barile and C. Brizio for their help in mitochondrial import
experiments. We thank S. Alziari and R. Garesse for the kind
gift of D.melanogaster mtDNA. This work was supported by
grants from `Ministero dell'Istruzione, dell'UniversitaÁ e della
Ricerca': COFIN PRIN 2002; Centro di Eccellenza di
Genomica in Campo Biomedico e Agrario; Piani di
Potenziamento della Rete Scienti®ca e Tecnologica, Legge
488/92 Cluster C03; from `European Union': Concerted
Actions on Mitochondrial Biogenesis and Disease (Contract
N. QLG1-CT-2001-00966); and from `UniversitaÁ di Bari':
Progetto di Ricerca di Ateneo.
REFERENCES
1. Taanman,J.-W. (1999) The mitochondrial genome: structure,
transcription, translation and replication. Biochim. Biophys. Acta, 1410,
103±123.
2. Larsson,N.G., Wang,J., Wilhelmsson,H., Oldfors,A., Rustin,P.,
Lewandoski,M., Barsh,G.S. and Clayton,D.A. (1998) Mitochondrial
transcription factor A is necessary for mtDNA maintenance and
embryogenesis in mice. Nature Genet., 18, 231±236.
3. Falkenberg,M., Gaspari,M., Rantanen,A., Trifunovic,A., Larsson,N.G.
and Gustafsson,C.M. (2002) Mitochondrial transcription factors B1 and
B2 activate transcription of human mtDNA. Nature Genet., 31, 289±294.
1604
Nucleic Acids Research, 2003, Vol. 31, No. 6
4. Fernandez-Silva,P., Martinez-Azorin,F., Micol,V. and Attardi,G. (1997)
The human mitochondrial transcription termination factor (mTERF) is a
multizipper protein but binds to DNA as a monomer, with evidence
pointing to intramolecular leucine zipper interactions. EMBO J., 5,
1066±1079.
5. Boore,J.L. (1999) Animal mitochondrial genomes. Nucleic Acids Res.,
27, 1767±1780.
6. Cantatore,P., Roberti,M., Rainaldi,G., Gadaleta,M.N. and Saccone,C.
(1989) The complete nucleotide sequence, gene organization and genetic
code of the mitochondrial genome of Paracentrotus lividus. J. Biol.
Chem., 264, 10965±10975.
7. Cantatore,P., Roberti,M., Loguercio Polosa,P., Mustich,A. and
Gadaleta,M.N. (1990) Mapping and characterization of Paracentrotus
lividus mitochondrial transcripts: multiple and overlapping transcription
units. Curr. Genet., 17, 235±245.
8. Roberti,M., Mustich,A., Gadaleta,M.N. and Cantatore,P. (1991)
Identi®cation of two homologous mitochondrial DNA sequences, which
bind strongly and speci®cally to a mitochondrial protein of Paracentrotus
lividus. Nucleic Acids Res., 19, 6249±6254.
9. Loguercio Polosa,P., Roberti,M., Musicco,C., Gadaleta,M.N.,
Quagliariello,E. and Cantatore,P. (1999) Cloning and characterisation
of mtDBP, a DNA-binding protein which binds two distinct
regions of sea urchin mitochondrial DNA. Nucleic Acids Res., 27,
1890±1899.
10. Fernandez-Silva,P., Loguercio Polosa,P., Roberti,M., Di Ponzio,B.,
Gadaleta,M.N., Montoya,J. and Cantatore,P. (2001) Sea urchin mtDBP is
a two-faced transcription termination factor with a biased polarity
depending on the RNA polymerase. Nucleic Acids Res., 29,
4736±4743.
11. Adams,M.D. Celniker,S.E., Holt,R.A., Evans,C.A., Gocayne,J.D.,
Amanatides,P.G., Scherer,S.E., Li,P.W., Hoskins,R.A., Galle,R.F. et al.
(2000) The genome sequence of Drosophila melanogaster. Science, 287,
2185±2195.
12. Lewis,D.L., Farr,C.L. and Kaguni,L.S. (1995) Drosophila melanogaster
mitochondrial DNA: completion of the nucleotide sequence and
evolutionary comparisons. Insect Mol. Biol., 4, 263±278.
13. Berthier,F., Renaud,M., Alziari,S. and Durand,R. (1986) RNA mapping
on Drosophila mitochondrial DNA: precursors and template strands.
Nucleic Acids Res., 14, 4519±4533.
14. Takata,K., Yoshida,H., Hirose,F., Yamaguchi,M., Kai,M., Oshige,M.,
Sakimoto,I., Koiwai,O. and Sakaguchi,K. (2001) Drosophila
mitochondrial transcription factor A: characterization of its cDNA and
expression pattern during development. Biochem. Biophys. Res.
Commun., 287, 474±483.
15. Goto,A., Matsushima,Y., Kadowaki,T. and Kitagawa,Y. (2001)
Drosophila mitochondrial transcription factor A (d-TFAM) is
dispensable for the transcription of mitochondrial DNA in Kc167 cells.
Biochem J., 354, 243±248.
16. McCulloch,V., Bonnie,L., Seidel-Rogol,B.L. and Shadel,G.S. (2002) A
human mitochondrial transcription factor is related to RNA adenine
methyl transferases and binds S-adenosylmethionine. Mol. Cell. Biol., 22,
1116±1125.
17. Barile,M., Brizio,C., De Virgilio,C., Del®ne,S., Quagliariello,E. and
Passarella,S. (1997) Flavin adenine dinucleotide and ¯avin
mononucleotide metabolism in rat liver. The occurrence of FAD
pyrophosphatase and FMN phosphohydrolase in isolated mitochondria.
Eur. J. Biochem., 249, 777±785.
18. Lupas,A., Van Dyke,M. and Stock,J. (1991) Predicting coiled coils from
protein sequences. Science, 252, 1162±1164.
19. Fried,M,G. and Crothers,D.M. (1984) Kinetics and mechanism in the
reaction of gene regulatory proteins with DNA. J. Mol. Biol., 172,
263±282.
20. Valverde,J.R., Marco,R. and Garesse,R. (1994) A conserved heptamer
motif for ribosomal RNA transcription termination in animal
mitochondria. Proc. Natl Acad. Sci. USA, 91, 5368±5371.
21. Roberti,M., Loguercio Polosa,P., Musicco,C., Milella,F., Qureshi,S.,
Gadaleta,M.N., Jacobs,H.T. and Cantatore,P. (1999) In vivo
mitochondrial DNA±protein interactions in sea urchin eggs and embryos.
Curr. Genet., 34, 449±458.
22. Thompson,J.D., Higgins,D.G. and Gibson,T.J. (1994) CLUSTAL W:
improving the sensitivity of progressive multiple sequence alignment
through sequence weighting, position-speci®c gap penalties and weight
matrix choice. Nucleic Acids Res., 22, 4673±4680.