Expression of the
Drosophila melanogaster
ATP synthase a subunit
gene is regulated by a transcriptional element containing GAF
and Adf-1 binding sites
Ana Talamillo
1,
*, Miguel Angel Ferna
´
ndez-Moreno
1
, Francisco Martı
´
nez-Azorı
´
n
1
,Bele
´
n Bornstein
1,2
,
Pilar Ochoa
1
and Rafael Garesse
1
1
Departamento de Bioquı
´
mica, Instituto de Investigaciones Biome
´
promoters i n an o rientation-independent manner. In addi-
tion, Northern blot and RT-PCR analysis identified two
a-F1-ATPase mRNAs that differ in the length of the 3¢
untranslated region due to the selection of alternative
polyadenylation sites.
Keywords: mitochondria; a-F
1
-ATPase; GAGA; Adf-1;
transcription regulation.
The bulk of cellular ATP is synthesized through oxidative
phosphorylation (OXPHOS) that takes place in the mito-
chondria. The OXPHOS system is composed of five
multisubunit complexes embedded in the inner mitochond-
rial membrane and two small electron carriers, ubiquinone
and cytochrome c [1]. The O XPHOS system is generated in a
unique manner. The majority of the more than 80 OXPHOS
subunits are encoded by genes in the nuclear DNA (n-DNA),
while 13 essential subunits are encoded i n the mitochondrial
DNA (mtDNA), contributing to four out of the five
OXPHOS complexes. The mtDNA consists of a small,
double-stranded, circular DNA molecule that is transcribed
and translated within this organelle. However, all of the
components involved in the replication, maintenance and
expression of the mtDNA, as well as the factors that
participate in t he assembly of the respiratory complexes, are
encoded in the nucleus. Therefore, correct OXPHOS func-
tion relies on the co ordinated expression of numerous genes
encoded in two physically separated genetic systems [2,3].
The multisubunit enzyme ATP synthase (complex V of the
OXPHOS system) is p resent in the membranes of eubacteria,
in this process, including transcriptional and post-transcrip-
tional regulation of gene expression [5,6], changes in Ca
2+
Correspondence to R. Garesse, Departamento de Bioquı
´
mica, Insti-
tuto de Investigaciones Biome
´
dicas ‘Alberto Sols’ CSIC-UAM, Fac-
ultad de Medicina, Universidad Auto
´
noma de Madrid, C /Arzobispo
Morcillo 4, 28029 Madrid, Spain. Fax: +34 91 5854001,
Tel.: +34 91 4975453, E-mail:
Abbreviations: Adf-1, alcohol dehydrogenase distal f actor; GAF,
GAGA factor; OXPHOS, oxidative phosphorylation; n-DNA, nuc-
lear DNA; mtDNA, mitochondrial DNA; NRF, nuclea r respiratory
factor; RACE, rapid amplification of cDNA ends; AEL, after egg
laying; UTR, untranslated region; DPE, downstream promoter
element.
*Present a ddress: Departamento de Anatomı
´
a y Biologı
´
aCelular,
Facultad de Medicina, Universidad de Cantabria, Santander, Spain.
(Received 2 June 2004, revised 6 August 2004,
accepted 18 August 2004)
Eur. J. Biochem. 271, 4003–4013 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04336.x
concentration [7], control of the mitochondrial dNTP pool
tory elements have been identified in the promoter of several
genes involved in mitochondrial biogenesis, such as
OXBOX/REBOX [21], Mts [22] or GRBOX [23]. However
the putative transcription factors that recognize these DNA
motifs remain to be identified.
In contrast, less is known about the m echanisms
controlling mitochondrial biogenesis in o ther animal sys-
tems or in invertebrates. We previously described how the
transcription of several Drosophila melanogaster genes
encoding components of the mtDNA replication machinery
was r egulated. These included the mitochondrial single-
stranded binding protein (mtSSB), and the catalytic (a)and
accessory (b) subunits of the DNA polymerase c (P ol c)
[24,25]. Interestingly, the expression of the genes enco ding
mtSSB and Pol c-b is transcriptionally regulated by the
DNA replication-related-element binding factor (DREF).
Indeed, in Drosophila this transcription factor regulates the
expression of genes that are essential for the cell-cycle and
for the nuclear DNA replication machinery [26], establish-
ing a link b etween mitochondrial and nuclear DNA
replication [24,25]. Here, we have identified essential
elements that participate in the transcriptional regulation
of the g ene encoding the a subunit of the H
+
ATP synthase
(a-F1-ATPase)inD. melanogaster.
Materials and methods
Library screenings
We screened a D. melanogaster genomic library prepared in
the vector k-DASH u sing the previously described a-F1-
GCG
programme (University of Wisconsin) [28].
Mapping of transcriptional initiation sites
Identification o f the a-F1-ATPase transcription start site
was achieved by three different method s: primer extension,
high-resolution S1 mapping and rapid amplification of
cDNA ends (RACE).
Primer extension analysis. Two different oligonucleotides
were used: a-PE1 (5¢-ACGGCC GGTCTCCTCCAGA
TC-3¢) from bp 216–195 and a-PE2 (5¢-GGACGC
CAGGCGGGCGGAAAAAATCG-3¢) from bp 30–4
from the A TG start codon in the a-F1-ATPase cDNA
sequence, respectively [27] (accession number Y07894). In
the assay, 50 pmol of the oligonucleotides were labelled with
50 lCi of [
32
P]ATP[cP] and polynucleotide kinase. Total
RNA ( 30 lg) from adults or from embryos obtained 0–18 h
after egg laying (AEL), and 8 lCi
1
of
32
P-labelled primer
were used in each experiment. Annealing and reverse
transcription were carried out as described previously [29],
and the extended products were analysed in 8% (w/v)
polyacrylamide/7
M
urea gels. Sequencing reactions using
the same oligonucleotides were run in parallel.
M
ZnSO
4
), the
sample was incubated with 150 units of S1 nuclease
(Pharmacia) for 60 min. The reaction was stopped with
4
M
ammonium acetate and 0.1
M
EDTA, and the nucleic
acids extracted with phenol and precipitated with ethanol.
The p ellet was resuspended in 75 m
M
NaOH, a nd after
incubating for 15 min at 90 °C it was precipitated w ith
ethanol, resuspended in 98% (v/v) formamide, 25 m
M
EDTA, 0.02% (w/v) bromophenol blue, 0.02% (w/v)
xylene cyanol, and analysed in 8% (w/v) polyacrylamide/
7
M
urea gels. Sequencing reactions were run in parallel.
5¢ RACE experiments. We used the RLM-R ACE kit from
Ambion Inc. (cat. # 1700), following the manufacturer’s
instructions. We used a-PE1 as the outer primer oligo-
nucleotide for the a-F1-ATPase cDNA (see Primer exten-
sion analysis) and the inner primer was 5¢-TCTCCTCCA
GATCAGCCTTGGGGG-3¢.
RT-PCR analysis of a
generated either by Ex oIII digestion and blunt-end cloning,
by restriction endonuclease-based cloning, or by PCR
amplification and cloning of selected DNA fragments.
Finally, we obtained the constructs shown below, where +1
represents the transcription start point according to the data
presented here.
Mutagenesis of t he GAGA element was achieved by
PCR and subcloning of the amplified fragment. The
oligonucleotides used for P CR were the l uciferase gene
internal primer 5¢-GGCGTCTTCCATTTTACC-3¢ and
the oligonucleotide 5 ¢-CCGTCGACATTAATTTAATTT
ccccAATTATATTGCGTCG-3¢ in which the SalI recogni-
tion site is in bold and th e GAGA element is replaced by the
sequence underlined (lowercase letters show the nucleo tide
changes). The )146/+79 construct was used as a template
and the amplified fragment was cloned into the pXp2
5
vector.
This strategy was also used to mutate the alcohol dehy-
drogenase distal factor (Adf-1) element using the specific
primer 5¢-CCGTCGACATTAATTTGAGAAATTATAT
TGCGTCGCccg
ccggcCgcCacgGAGGGTGAC-3¢ (again
the SalI r ecognition site is i n bold, the location of the
Adf-1 element is underlined, and nucleotides in lowercase
have been changed). A similar strategy was carried out to
construct the GAGA or/and Adf-1 mutants in the hybrid
promoters (see Results).
Cell transfection assays
The pXp constructs (5 lg) were transiently transfected into
and the grey boxes represent UTRs. The line underneath the gene
shows the position of several restriction endonucleases. E: Ec oRI; K:
KpnI; C: ClaI; Ev: EcoRV; S: SalI; P: PstI; B: BamHI.
Ó FEBS 2004 Drosophila a-F
1
-ATPase gene expression (Eur. J. Biochem. 271) 4005
exon encodes the 5¢ untra nslated region (5¢-UTR) as well as
the first 22 amino acids of the 23 residues which form the
targeting sequence. The last amino acid of the presequence,
the complete mature protein and the 3¢-UTR re gion are
encoded in e xons 2–4. To determine the transcriptional
initiation sites of the a-F1-ATPase gene we first carried out
primer extension analysis using total RNA extracted from
adults or embryos 0–18 h AEL, and two different
32
P-labelled oligonucleotide primers (a-PE1 and a-PE2).
Both primers produced identical results in embryos and
adults, three transcriptional initiation sites being detected at
positions )86, )91 (the majority) and )120, considering
position +1 as the first nucleotide of the translation
initiation codon ATG (Fig. 2A,C).
This result was confirmed by high resolution S1 mapping
using a 506 bp probe that extended from the coding region
(position +30) to 477 bp upstream of the ATG. In th is
analysis, several DNA fragments were protected (Fig. 2B),
with the strongest signal corresponding to position )91, the
prominent position detected by primer extension. The
position )91 is 22 nucleotides upstream of the transcription
startpoint previously described for this gene [27] (Gen Bank
accession number Y07894). Additionally, more weakly
sequence of a-F1-ATPase 5¢ upstream region. Black arrows show the transcription start poin ts identified in the primer extension assays. Wh ite
arrows represent transcrip tion start points from S1 mapping. The thickn ess of the b lack and wh ite arrows is r elated to the i ntensity of the b and.
Asterisks represent transcription start points from the 5¢ end amplification experiment s (see Materials and methods). PCR prod ucts were cloned and
six of them were sequenced, identifying the nucleotide shown by asterisk as the 5¢ end of a-F1-ATPase cDNA.
4006 A. Talamillo et al.(Eur. J. Biochem. 271) Ó FEBS 2004
transcriptional apparatus t o the a-F1-ATPase promoter.
Furthermore, several short sequences commonly found
downstream of transcriptional initiation sites in Drosophila
promoters include ACGT, ACAA, ACAG, and AACA
[32], and these were detected at )17, )18, )36 and )103
positions of the a-F1-ATPase gene (position r elative t o
ATG). However, the region did not contain a canonical
downstream promoter element (DPE), which is recognized
by the TAFII60 factor [33]. Indeed, the short e lements
described above probably substitute for t he DPE motif in
the Drosophila a-F1-ATPase promoter.
Functional analysis of the a-F1-ATPase promoter region
The function of t he Drosophila a-F1-ATPase promoter
region was characterized by transient transfection into
Schneider SL2 cells. A series of deletions of the 5¢ upstream
region of the gene were cloned in the pXp2 vector that
contains the luciferase reporter gene. A construct containing
the )823/+86 region (position +1 corresponds to the main
transcriptional initiation site located 91 nucleotides
upstream of the ATG initiation codon) promoted substan-
tial luciferase activity in Schne ider cells, 2 ,300-fold higher
than the native p XP2 vector (Fig. 3). This activ ity was
orientation-dependent and indicates that t he a-F1-ATPase
5¢ proximal upstream region contains a strong promoter,
with similar activity in Schneider cells to the promoter of
directed mutagenesis and examined the activity of the
mutated constructs in cell transfection assays. M utating
the GAGA or Adf-1 elements individually significantly
reduced promoter activity by up to 40–60%, whereas
when both sites were abolished, the a ctivity of t he
promoter was reduced by 75% (Fig. 4B). In addition, we
carried out cotransfection studies in Schneider cells using
different a-F1-ATPase promoter constructs and a plasmid
that express GAF under the control of the actin 5C
promoter. The GAGA factor stimulated at least threefold
the activity of the promoter in constructs )397/+86 and
)146/+86, but had no effect on the activity of the
construct )93/+86, which does not contain the potential
GAF binding site (Fig. 4C).
The combination of GAF/Adf-1 h as been shown t o
activate transcription in a variety of promoter contexts.
Hence, we generated a construct containing a 56 b p
DNA fragment ()144/)89) that included the GAGA and
Adf-1 elements linked to the basal promoters of the
d-aminolevulinate synthase (ALAS)andb-F1-ATPase
genes [29,34]. In both constructs there was a s ubstantial
Fig. 3. Functional analysis of the D. melanogaster a-F1-ATPase promoter. Sche me of the a-F1-ATPase promoter constructs used fo r transient
transfection assays in Schneider cells (see Materials and methods). The promoter regions are represented by solid lines and the luciferase reporter
gene is shown as a solid arrow. The numbers to th e left of each construct indicate the limit of the promoter fragment with reference to the
transcription start point as established in this study. The relative promoter activities of the constructs measured in the luciferase assay are indicated
on the right by black boxes. The luciferase activity of the vector with no insert was defined as being equal to one. Luciferase activity was normalized
to the b-galactosidase activity of cotransfected control plasm id. Valu es are the mean s ± SD of at least five independent exp eriments.
Ó FEBS 2004 Drosophila a-F
1
-ATPase gene expression (Eur. J. Biochem. 271) 4007
different sizes (roughly 2.2 and 1.8 kb) can be detected at all
developmental stages, as well as in adults [27] (Fig. 6A).
Furthermore, we previously isolated two cDNAs that differ
in their 3¢ region, suggesting that the difference in size may
be due to the alternative selection of polyadenylation sites
[27]. In order to precisely determine the origin of the two
transcripts, we carried out RT-PCR experiments on total
RNA e xtracted from embryos 0–18 h AEL or adults. In
these reactions, we used two different oligonucleotide
primers to map the 3¢ regionofthegene(a-RT1 and
a-RT2; see Materials and methods) and an oligo(dT)
primer. With t he combination o f oligo(dT)/a-RT1 primers
two 700 and 320 bp fragments were amplified both f rom
embryos and adults, while the fragments amplified by the
Fig. 4. The transcription of a-F1-ATPase is controlled by a cassette containing the GAGA and Adf-1 elements. (A) Sequence of the )146/+94
(relative to the main transcription start point) a-F1-ATPase promoter region. The 56 bp region essential for promoter activity is boxed, and the
GAGA and Adf-1 elements in the DNA are underlined. The translation start codon and the main transcription start point are larger and in bold.
(B) Mutations either in the G AGA or Adf-1 binding sites significantly reduced a-F 1-ATPase promoter activity. Each DNA element in the )146/
+86 promoter construct (whose lucifer ase a ctivity w as considered as 100%) was mutated and the activity of the m utant constructs was asse ssed.
Wild type an d mutated GAGA and Adf-1 binding sites are s hown by open and filled symbols, respectively. (C) Activation of the a-F1-ATPase
promoter was induced in presence of GAF. Different a-F1-ATPase promoter constructs were cotransfected with a plasmid expressing GAF and the
activity of constructs including the GAGA elem ent increased significantly. The data show the luciferase activity of the a-F1-ATP ase promoter
constructs cotransfected with GAF relative to that cotransfected with the non-GAF containing vector. Values are the means ± SD of at least three
independent experiments. Wild type an d mutated GAGA and Adf-1 binding sites are shown by open and fi lled symbols, respectively.
4008 A. Talamillo et al.(Eur. J. Biochem. 271) Ó FEBS 2004
oligo(dT)/RT-2 primers were approximately 600 and
230 bp in siz e ( Fig. 6B). We sequenced the four DNA
fragments an d confirmed that they corresponded to cDNAs
originated by the a lternative selection o f polyadenylation
signals located at position 101 and 462 downstream of the
mechanisms involved in regulating the activity of individual
gene promoters [42].
We are interested in und erstanding the transcriptional
regulation of genes encoding essential components of the
mitochondrial respiratory chain in Drosophila melanogaster.
The energetic demands of the different tissues in an
organism vary depending on their physiology, and can
respond to both environmental and developmental signals
[2,43]. Tissues with large energy demands contain mito-
chondria that are well differentiated, both structurally and
functionally, with highly developed c ristae full of ATP
synthase complexes. For this reason, ATP synthase and in
particular the a-F1-ATPase and b-F1-ATPase catalytic
subunits have been often used as markers for mitochondrial
biogenesis [6,31,44,45]. The Drosophila a-F1-ATPase gene is
organized into four exons that are separated by three
introns, and that map in the 59AB region of the 2R polytene
chromosome. Two mutations have been mapped to the
aF1-ATPase gene, bellwether and colibri,whichwere
isolated in screens to identify lethal mutations in the 59AB
region [46] and larval growth defects [47], respectively.
We have characterized some of the elements involved in
regulating the transcription of the Drosophila aF1-ATPase
gene. Initially, we mapped the transcription initiation sites
by primer extension, S1 mapping analysis and mRNA 5¢-
end a mplification. In this way, we localized a small region
containing several sites where RNA synthesis commences,
the strongest located 9 1 nucleotides upstream of t he
translation i nitiation codon. The structural organization
of the transcription initiation region of the Drosophila a-F1-
gene to a 93 bp proximal region and identify a region critical
for the transcriptional r egulation of the gene. Full promoter
activity is strictly dependent on a 56 bp region located
between position )89 and )144 (relative to the main
transcription start point), that contains binding sites for the
GAGA factor and Adf-1. This small activator region is also
active on heterologous promoters in an orientation-inde-
pendent manner. Transfection assays unequivocally dem-
onstrate that both sites are functional and suggest that the
activity of this fragment is dependent of the GAF and Adf-1
regulatory proteins.
Both GAF and Adf-1 are ubiquitous transcription factors
critical for Drosophila viability. GAF is encoded by the
trithorax-like locus [50] which is required for the correct
expression of several homeotic genes. It was first identified
as an activator of the engrailed and Ultrabithorax promot-
ers, and was later shown to regulate the activity of several
constitutive, inducible and developmentally regu lated genes
(reviewed i n [51]). GAF binds to GA-rich sequences and
specifically interacts w ith the trinucleotide GGA via a single
zinc-finger DNA binding domain [52]. GAF does not
directly regulate the RNA polymerase II basal machinery,
but it does a ctivate transcription by relieving the repression
of histone in chromatin structure, permitting the access of
transcription factors to promoter or enhancer regulatory
regions [35]. Although d irect stimulation by GAF has only
been demonstrated in roughly a dozen promoters, the
number of genes transcriptionally regulated by GAF is
likely to be much larger as G AF binds to hundreds of
euchromatic regions in salivary polytene chromosomes [53].
strated in vivo [38]. Interestingly, GAF or Adh-1 binding
alone activates transcription but the establishment of
DNaseI hypersensitivity is dependent on the binding of
both factors [38].
Our results suggest that GAF and Adf-1 also act together
to regulate the expression of the Drosophila a-F1-ATPase
gene. Indeed, this is highlighted by the presence of similar
combinations of the two tr anscription factors in the
promoter proximal region of the orthologous genes of
different insects that diverged s everal million years ago,
including three Drosophila species and Anopheles gambiae.
We also have identified two a-F1-ATPase mRNAs,
which differ in t he length of their 3¢ UTRs, b y selection of
alternative polyadenylation sites. T he ratios of the two
a-F1-ATPase mRNAs are similar at different developmen-
tal stages, as well as in different parts of the body of the
adult flies. Although this may reflect the random selection of
two poly(A) sites with different inherent pr operties, the
possibility that the tissue-specific regulation of polyadeny-
lation plays an important physiological role can not be ruled
out. Polyadenylation of mRNA is essential for its transport
from the nucleus to the cytoplasm, for the stability of the
transcript, a nd for the efficiency of translation [58], and thus
poly(A) site selection is important for the control of gene
expression. In the last few years, a variety of transcription
units containing two or more poly(A) sites in their 3¢
terminal exons have been described (reviewed in [59]).
Furthermore, in some cases it has been shown that the ratio
of the transcripts containing different 3¢-UTRs determines
the amount of protein in specific tissues [59]. Experiments
2+
signaling in C2C12
skeletal myocytes in response to mitochondrial genetic an d
metabolic stress: a novel mode of inter-organelle crosstalk. EMBO
J. 18, 522–533.
8. Nishino, I., Spinazzola, A. & Hirano, M. (1999) Thymidine
phosphorylase gene m utations in MNG IE, a hum an mitochon-
drial disorder. Science 283, 689–692.
9. Rampazzo, C., Gallinaro, L., Milanesi, E., Frigimelica, E.,
Reichard, P. & Bianchi, V. (2000) A deoxyribonucleotidase in
mitochondria: involvement in regulation of dNTP pools an d
possible link to genetic disease. Proc.NatlAcad.Sci.USA97,
8239–8244.
10. Davis, A.F. & Clayton, D.A. (1996) In situ localization o f
mitochondrial DNA replicationinintactmammaliancells.J. Cell
Biol. 135, 883–893.
11. Enriquez, J.A., Fernandez-Silva, P. & Montoya, J. (1999)
Autonomous regulation in mammalian mitochondrial DNA
transcription. Biol. Chem. 380, 737–747.
12.Lenka,N.,Vijayasarathy,C.,Mullick,J.&Avadhani,N.G.
(1998) Structural organization and transcription regulation of
nuclear genes encod ing the mamma lian cytochrome c oxidase
complex . Prog. Nucle ic Acid Res. Mol . Biol. 61, 309–344.
13. Zaid, A., Li, R., Luciakova, K., Barath, P., Nery, S. & Nelson,
B.D. (1999) On the role of the general transcription factor Sp1 in
the activation and repression of diverse mammalian oxidative
phosphorylation genes. J. Bioenerg. Biomembr. 31 , 129–135.
14. Kelly, D.P. & Scarpulla, R.C. (2004) T ranscriptional regulatory
circuits controlling mitochondrial biogenesis a nd function. Genes
Dev. 18, 357–368.
regulatory regions o f human nuclear genes and mitochondrial
gene for the oxidative phosphorylation system. J. Biol. Chem. 266,
2333–2338.
23. Giraud, S., Bonod, C., Wesolowski B., L ouvel, M. & Stepien, G.
(1998) Expression of human ANT2 gene in highly proliferative
cells: GRBOX, a new transcriptional element, is involved in the
regulation of glyco lytic ATP import into m itoch ondria. J. Mol.
Biol. 281, 409–418.
24. Ruiz de Mena, I., Lefai, E., Garesse, R. & Kaguni, L.S. (2000)
Regulation of mitochondrial single-stranded DNA-binding pro-
tein gene e xpression links nuclear and mitoch ondrial D NA
replication in Drosophila. J. Biol. Chem. 275, 13628–13636.
25. Lefai, E., Fernandez-Moreno, M.A., Alahari, A., Kaguni, L.S. &
Garesse, R. (2000) Differential regulation of the catalytic and ac-
cessory subunit genes of drosophila mitochondrial DNA poly-
merase. J. Biol. Chem. 275, 33123–33133.
26. Yamaguchi, M., Hayashi, Y., Nishimoto, Y., Hirose, F. &
Matsukage, A . (1995) A nu cleotide sequence essen tial for the
function of DRE, a common promoter element for Drosophila
DNa replication-related genes. J. Biol. Chem. 270, 15808–15814.
27. Talamillo,A.,Chisholm,A.A.,Garesse,R.&Jacobs,H.T.(1998)
Expression of the nuclear gene encoding mitochondrial ATP
synthase subunit alpha in early development of Drosophila and sea
urchin. Mol. Biol. Report 25, 87–94.
28. Devereux, J., Haeberli, P. & Smithies, O. (1984) A comprehensive
set of sequence analysis programs for the VAX. Nucleic Acids Res.
12, 387–395.
29. Ruiz de Mena, I., Fernandez-Moreno, M.A., Bo rnstein, B.,
Kaguni, L .S. & Garesse, R. ( 1999) Structure and re gulated
expression of the delta-aminolevulinate synthase gene from Dr o-
36. England, B.P., Heberlein, U. & Tjian, R. (1990) Purified Droso-
phila transcription factor, Adh distal factor-1 (Adf-1), binds to
sites in several Drosophila promoters and activates transcription.
J. Biol. Chem. 265, 5086–5094.
37. Han,W.,Yu,Y.,Su,K.,Kohanski,R.A.&Pick,L.(1998)A
binding site for multiple tran scriptional activators in the fushi
tarazu proximal enhancer is essential for gene expression in vivo.
Mol. Cell. Biol. 18, 3384–3394.
38.Pile,L.A.&Cartwright,I.L.(2000)GAGAfactor-dependent
transcription and estab lishment o f DNase hyperse nsitivity are
independent and unre lated even ts in vivo. J. Biol. Chem. 275, 1398–
1404.
39. Konig, S., Burkman, J., Fitzgerald, J., Mitchell, M., Su, L. &
Stedman, H. (2002) Modular organizat ion o f phyloge netica lly
conserved domains contro lling develop mental regulat ion of the
human skeletal myosin heavy chain gene family. J. Biol. Chem.
277, 27593–27605.
40. Mas, J.A., Garcia-Zaragoza, E . & Cervera, M. (2004) Two func-
tionally identical modular enhancers in Drosophila troponin T
gene establish the correct protein levels in different muscle types.
Mol. Biol. Cell 15 , 1931–1945.
41. Sandelin,A.&Wasserman,W.W.(2004) Constrained binding site
diversity within families of tran scription factors enhances pattern
discovery bioinformatics. J. Mol. Biol. 338, 207–215.
42. Lemo n, B. & Tjian, R. (2000) Orchestrated re sponse : a symphony
of transcription factors for gene control. Genes Dev. 14, 2551–
2569.
43. Moyes, C.D., Mathieu-Costello, A., Tsuchiya, N., Filburn, C. &
Hansford, R.G. (1997) Mitochondrial biogenesis during cellular
differentiation. Am. J. Physiol. 272, C1345–51.
2672–2678.
53. Benyajati, C., Mueller, L., Xu,N.,Pappano,M.,Gao,J.,Mosa-
mmaparast, M., Conklin, D., Granok, H., Craig, C. & Elgin, S.
(1997) Multiple isoforms of GAGA factor, a critical component of
chromatin structure. Nucleic Acids Res. 25, 3345–3353.
54. Bhat,K.M.,Farkas,G.,Karch,F.,Gyurkovics,H.,Gausz,J.&
Schedl, P. (1996) The GAGA factor is required in the ea rly Dro-
sophila embryo not only for transcriptional regulation but also for
nuclear division. Development 122, 1113–1124.
55. Pile, L .A., Spellman, P .T., Katzenb erger, R.J. & Wassarman,
D.A. (2003) The SIN3 deacetylase complex represses genes en-
coding mitochondrial p roteins: implications for the regulation of
energy metabolism. J. Biol. Chem. 278, 37840–37848.
56. DeZazzo, J., Sandstrom, D., de Belle, S., Velinzon, K., Smith, P.,
De Gr ady, L.I., Vecchio, M., Ramaswami, M. & Tully, T. (2000)
nalyot, a mutation of the Dro sophila myb-related Adf1
4012 A. Talamillo et al.(Eur. J. Biochem. 271) Ó FEBS 2004
transcription factor, disrupts synapse formation and olfactory
memory. Neuron 27, 145–158.
57. Gao, J. & Benyajati, C. ( 1998) S pecific local histone-DNA
sequence contacts facilitate high-affinity, non-cooperative
nucleosome binding of both adf-1 and GAGA f actor. Nucleic
Acids Res. 26, 5394–5401.
58. Colgan, D.F. & Manley, J.L. (1997) Mechanism and regulation of
mRNA polyadenylation. Genes Dev. 11, 2755–2766.
59. Edwalds-Gilbert, G., Veraldi, K.L. & Milcarek, C. (1997) Alter-
native poly(A) site selection in comp lex transcription units: means
to an end? Nucleic Acids Res. 25, 2547–2561.
Ó FEBS 2004 Drosophila a-F
1