ABSTRACT
The DRE/DREF system plays an important role in transcription of DNA replication
genes such as those encoding the 180 and 73 kDa subunits of DNA polymerase
[alpha]
as well as that for encoding PCNA. In this study, we found two sequences
homologous to DRE (5
'
-TATCGATA-3
'
) in the 5
'
-flanking region (-370 to -357 with respect to the transcription initiation site) of
the
D-raf
gene and confirmed transcriptional activity through gel mobility shift assays,
transient CAT assays, and spatial patterns of
lac
Z expression in transgenic larval tissues carrying
D-raf
and
lac
Z fusion genes. Further, we demonstrated that the
D-raf
gene is another target of the Zerknüllt (Zen) protein with observation of
D-raf
repression by Zen protein in cultured cells and its ectopic expression in the
dorsal region of the homozygous
zen
mutant embryo. The evidence of DRE/DREF involvement in regulation of the
D-raf
gene obtained in this study strongly supports the idea that the DRE/DREF system
is responsible for the coordinated regulation of cell proliferation-related genes in
Drosophila
.
Raf-1, a protein serine/threonine kinase located primarily in the cytosol (
1
,
2
), serves as central intermediate in many signaling pathways, ultimately
regulating cell proliferation, differentiation, and development (
3
,
4
) by connecting upstream tyrosine kinases with downstream serine/threonine kinases such as mitogen-activated protein kinase (MAPK) and MAPK kinase (MAPKK) (
3
,
5
). A
Drosophila
homolog of the human
c-raf-1
,
D-raf
has been cloned and mutants defective for this gene have been identified (
6
-
8
). It has thereby been shown to be required for regulation of cell proliferation and differentiation (
6
,
7
,
9
,
10
). However, little is known about the control of
raf
proto-oncogene expression in
Drosophila
.
The DNA replication-related element (DRE) consisting of an 8 base pair (bp) palindrome,
TATCGATA, is responsible for activating promoters of the
Drosophila melanogaster
PCNA (proliferating cell nuclear antigen) and DNA polymerase [alpha]-encoding genes, both in cultured cells (
11
) and in transgenic flies (
12
). Furthermore, a specific DRE-binding factor (DREF) consisting of an 80 kDa polypeptide homodimer has been purified (
11
), and a corresponding cDNA has recently been cloned (
13
).
Promoters of
Drosophila
DNA replication-related genes are repressed by the product of
zerknüllt
(
zen
) (
14
), a homeobox gene which regulates the differentiation of the optic lobe and the
amnioserosa in the dorsal region of the
Drosophila
embryo (
15
-
17
). Repression of promoter activities by the Zen protein has been observed not
only in cultured Kc cells but also in transgenic flies carrying the PCNA gene
promoter-directed
lac
Z gene (
14
,
18
). Overexpression of Zen results in reduction of DREF activities in the cell (
18
). Therefore, DREF may be one of the key regulatory factors involved in
proliferation- and differentiation-related control of DNA replication related genes (
18
).
In this study, we found two DRE-like sequences in the 5'-flanking region of the
D-raf
gene and have examined their role in promoter activity. The obtained results
indicate that
D-raf
, which functions as a signal transducer, is indeed under the control of the
DRE/DREF system, like DNA replication-related genes. We also report that the promoter activity of the
D-raf
gene is negatively regulated by the Zen protein, both in cultured cells and in
living organisms.
A 1233 bp DNA fragment containing the
D-raf
promoter region (-878~+358 with respect to the transcription initiation site) (
19
) was isolated from plasmid pGEM-
Draf
4.3 bearing a genomic 4.3 kb
Bam
HI fragment (
6
) by digestion with
Bam
HI and
Pst
I. The fragment was blunt-ended with T4 DNA polymerase and subcloned into the
Sma
I site of pGEM-3. The direction of insert was examined by digestion with
Fok
I. The resultant plasmid was named pGEM-
Draf
1.23. The
D-raf
promoter region obtained from the plasmid pGEM-
Draf
1.23 by digestion with
Xba
I and
Sac
I was then inserted into the
Xba
I
and
Sac
I sites of pSKCAT (
14
). The resultant plasmid was named p5'-878
Draf
CAT.
To construct a 2 bp insertional mutation, the plasmid pGEM-
Draf
1.23 was digested with
Cla
I targeting a site at the center of the
Draf-
DRE-like sequence. The digested DNA fragment was blunt-ended using T4 DNA polymerase and then self ligated with T4 DNA
ligase. The resultant plasmid was confirmed to have an additional 2 bp GC
sequence, ATC
The expression plasmid pAct5C-zen (
20
) contains
zen
cDNA placed under the control of the
Drosophila
actin 5C gene promoter (-2500 to +88) (
21
). The expression plasmid of mutant
zen
, pAct5C-zen-[Delta]1 contains an internal deletion from amino acids 137 to 236
of the Zen protein (
20
).
All double-stranded oligonucleotides contained a 6 bp linker sequence recognizable by
Bgl
II and
Bam
HI and were chemically synthesized using a Applied Biosynthesis DNA Synthesizer.
The
Draf
-DRE,
Draf
-DRE-mut1 and
Draf
-DRE-In2 oligonucleotides being as follows:
Draf
-DRE
5'-gatccTTTATCGTTATCGATTGGTACAGCa-3'
3'-gAAATAGCAATAGCTAACCATGTCGtctag-5'
Draf
-DRE-mut1
5'-gatccTT
3'-gAA
Draf
-DRE-In2
5'-gatccTTTATCGTTATC
3'-gAAATAGCAATAG
where mutated bases are underlined and lower-case letters indicate the linker sequence. The double-stranded 30 bp oligonucleotides for
Draf
-DRE contain the 24 bp DRE like-containing sequence of the
D-raf
gene promoter and the 6 bp linker sequence, while the
Draf
-DRE-mut1 contains two 3 bp substitutions in this DRE sequence.
Draf
-DRE-In2, a double-stranded 32 bp oligonucleotide, contains a 2 bp insertion in the DRE sequences of
Draf
-DRE. Control double-stranded oligonucleotides, DRE-P and DRE-PM, being as follows:
DRE-P
5'-gatccCTGCCTGCTATCGATAGATTCAGGa-3'
3'-gGACGGACGATAGCTATCTAAGTCCtctag-5'
DRE-PM
5'-gatccCTGCCTGCTTACGATAGATTCAGGa-3'
3'-gGACGGACGAATGCTATCTAAGTCCtctag-5'
were also generated as in Hirose
et al.
(
11
).
The gel mobility shift analysis was performed as reported previously (
11
). Kc cell nuclear extracts and
Escherichia coli
lysates containing GST-DREF(16-608) fusion protein were prepared as described elsewhere (
13
). These were then added to reaction mixtures containing 15 mM HEPES (pH 7.6), 60 mM KCl, 0.1 mM EDTA, 1 mM DTT, 12% glycerol, 0.5 [mu]g of sonicated calf thymus DNA (average size of 0.2 kb) and double-stranded 32
P-labeled synthetic oligonucleotides (10 000 c.p.m.) and incubated for 15 min on ice. For this step, unlabeled DNA fragments were added as competitors. DNA-protein complexes were electrophoretically resolved on 6% nondenaturing polyacrylamide gels in 50 mM Tris/borate (pH
8.3), 1 mM EDTA and 2% glycerol at 25oC. Gels were dried and autoradiographed. The gel shift assay was also performed with anti-DREF monoclonal antibody No. 1 and anti-DREF monoclonal antibody No. 4 (
13
). Kc cell nuclear extracts were mixed with each antibody, incubated for 2 h on
ice, added to mixtures containing 32
P-labeled synthetic oligonucleotides (10 000 c.p.m.) and 0.5 [mu]g poly(dI-dC), and then incubated for 15 min on ice as described above.
Drosophila
Kc cells (
22
) were grown at 25oC in M3 (BF) medium (Sigma) supplemented with 2% fetal bovine serum and 0.5% penicillin-streptomycin (GIBCO-BRL). Kc cells (5 * 106
/dish) were plated into 60 mm plastic dishes 24 h before DNA transfection by the
calcium phosphate coprecipitation method, as described elsewhere (
23
). Each transfection also included 10 [mu]g of the
D-raf
gene promoter-CAT plasmid as a reporter. For cotransfecting Zen expression plasmids, unless otherwise specified, 2 [mu]g of p5'-878
Draf
CAT or 0.5 [mu]g of p5'-168DPCNACAT and 1-8 [mu]g of expression plasmid were transfected. The total
amount of DNA for transfection was adjusted to 10 [mu]g/dish with pGEM-3 plasmid DNA. Cells were harvested at 48 h after DNA transfection. Cell extracts for determination of CAT activites were prepared as
described (
24
). The spots corresponding to acetylated [14
C]chloramphenicols were taken from thin layer plates and radioactivities counted in a toluene-based scintillator. CAT activities were normalized to protein amounts,
determined using a BioRad protein assay kit (
25
).
P-element-mediated transformation and establishment of homozygous transformant
stocks were performed as described previously (
26
,
27
). Three independent transformant lines were established for p5'-878
Draflac
Z (
28
).
Expression patterns for
lac
Z were analyzed by X-gal staining as described earlier (
29
). Larval tissues were dissected, immersed in fixative (12 mM sodium cacodylate buffer, pH 7.3/26% glutaraldehyde) for 15 min at room temperature, and then incubated with a staining solution
containing 0.2% X-gal in the dark at 37oC for 5-16 h.
Whole-mount
in situ
hybridization for detecting expression of endogenous
D-raf
gene in wild-type
Drosophila
and the homozygous
zenw36
mutant (
30
) was conducted essentially as described by Tautz and Pfeifle (
31
). Embryos were collected, aged at 25oC, dechorinated and fixed, then were stored in 70% ethanol at -70oC and rehydrated when needed. As a probe, The 2.3 kb
Hin
cII-
Sma
I fragment from plasmid pGEM-
Draf
4.3 was labeled by random priming with a Digoxigenin Non-radioactive DNA Labeling and Detection Kit (Boeringer Mannheim). Embryos were developmentally staged using criteria described by Campos-Ortega and Hartenstein (
17
).
In the 5'-flanking region of the
D-raf
gene, we found two adjoining sequences homologous to DRE (5'-TATCGATA-3') TATCGTTATCGATT,
Draf-
DRE extending from positions -370 to -357 with respect to the transcription initiation signal site (
32
-
34
) of the
D-raf
gene (
6
,
8
,
10
) (Fig.
1
). Each of these sequences matches 7 bp out of the 8 bp DRE sequence. We also
found a conserved downstream basal promoter element (DPE) consensus sequence (
19
) in the +30 to +33 region downstream of the transcription initiation site of
the
D-raf
promoter (Fig.
1
). The DPE consensus sequence is usually located ~30 nucleotides downstream of the RNA start site of the
Drosophila
TATA-box-deficient (TATA-less) promoters (
19
).
For confirmation of the role of the DRE/DREF system in transcription of the
D-raf
gene, we examined whether
Draf
-DRE sequences can be recognized by DREF, the DRE-binding factor identified previously (
11
). Gel mobility shift assays were thus performed using Kc cell nuclear extracts
as the control source of DREF. Specific DNA-protein complexes could thereby be detected using a chemically
synthesized oligonucleotide carrying the
Draf-
DRE sequence as a probe (Fig.
2
, lanes 1 and 14). Two shifted bands on the gel suggest that there are at least
two complexes, although it is not clear yet if these reflect occupation of one
site and both sites. The complex with 32
P-labeled
Draf-
DRE was diminished by adding excess amounts of unlabeled
Draf
-DRE (Fig.
2
, lanes 2-5) or DRE-P (Fig.
2
, lanes 15-18), an oligonucleotide containing the DRE sequence from the
Drosophila
PCNA gene, as a competitor (
11
). However,
Draf-
DRE-mut1 carrying the multi-base-substitution inside the DRE sequence (Fig.
2
, lanes 6-9) and DRE-PM (Fig.
2
, lanes 19-22) did not diminish the complex formation.
Draf
-DRE-In2 carrying the 2 bp insertion more or less diminished the complex
formation (Fig.
2
, lanes 10-13), suggesting that the DRE-related sequence TATCGTTA, can also function as DRE.
Promoters of
Drosophila
DNA replication-related genes are repressed by the product of the
zen
gene (
12
,
18
). Whether Zen protein can affect transcription of the
D-raf
gene was therefore examined in cultured cells and in living organisms.
In cultured cells, cotransfection assays were carried out with the plasmid p5'-878
Draf
CAT and plasmids bearing wild-type or mutant Zen under the direction of the
Drosophila
actin 5C promoter (
21
), which is highly active in
Drosophila
cells. As a control, the plasmid p5'-168DPCNACAT carrying the upstream region (-168 to +24) of the PCNA gene was cotransfected with wild-type or mutant Zen expression plasmids.
Wild-type Zen repressed the activity of the
D-raf
gene promoter, with the degree of decrease being progressively augmented by increasing the amount of effector plasmid (Fig.
6
A). The plasmid pAct5C-zen-[Delta]1 carrying an internal in frame deletion (99 amino acid
residues including 13 carboxyl-terminal amino acid residues of the homeobox) only slightly affected the
CAT expression by p5'-878
Draf
CAT or p5'-168DPCNACAT (Fig.
6
A and B). The results obtained indicate that the active Zen protein can
specifically repress the
D-raf
promoter activity, as was the case with the PCNA gene promoter as well as the
DNA polymerase [alpha] promoter (
14
,
18
). The extent of repression of the
D-raf
promoter activity by Zen protein was similar to that of the PCNA promoter
activity (Fig.
6
A and B).
Drosophila
DRE/DREF
system plays an important role in the regulation of DNA replication-related genes such as those encoding the 180 kDa (
11
) and 73 kDa (
36
) subunits of DNA polymerase [alpha], PCNA (
37
) and cyclin A (
38
).
D-raf
has been demonstrated to bear multiple functions in the regulation of both
proliferation and differentiation of cells during development (
6
-
8
). It is expressed throughout development in a wide range of tissues with higher
levels of expression in the ovary and in tissues containing rapidly
proliferating cells (
6
,
28
). Multiple regulatory elements should participate in the expression of
D-raf
, and here we have demonstrated that DRE is one of them.
Two overlapping DRE-like sequences are found in the 5'-flanking region (-370 to -357 with respect to the putative transcription
initiation site) of
D-raf
(Fig.
1
). A gel mobility shift assay using Kc cell extracts and bacterially produced
GST-DREF fusion protein clearly demonstrated that the sequences are indeed the
target for the binding of DREF (Figs
2
,
3
and
4
). Disruption of one of the DRE elements results in a significant reduction of
CAT activity in the cells transiently expressing the
CAT
gene fused to the 5'-flanking sequence of
D-raf
(Fig.
5
). These observations strongly suggest that the expression of
D-raf
is under the control of the DRE/DREF system as are the DNA replication-related genes. A reporter
lacZ
gene fused to the 5'-flanking sequence of
D-raf
(-878 to +525 with respect to the transcription initiation site) is
expressed in the tissues containing proliferating cells such as the imaginal
rings of the salivary glands and the imaginal discs (
28
). This indicates that
D-raf
is expressed at higher levels in tissues with rapidly proliferating cells and
that the DRE/DREF system would be for the activation.
It has been demonstrated that the transcription from the gene for PCNA is
repressed by Zerknüllt in the embryonic dorsal region including the amnioserosa (
14
). Although the precise mechanism for this repression remains to be elucidated,
it has been demonstrated that the DRE sequence in the 5'-flanking region of the
PCNA
gene is responsible for the repression and that Zerknüllt may affect the amount or activity of DREF (
14
).
D-raf
was demonstrated to be similarly under the regulation of Zerknüllt both
in vitro
(Fig.
6
) and
in vivo
(Fig.
7
). These observations further support the idea that the expression of
D-raf
is under the control of the DRE-DREF system in concert with the DNA replication-related genes.
The major role of
D-raf
in proliferation would be the transduction of transmembrane growth-stimulating signals into the nucleus in the G0
/G1
transition and G1
phase as has been demonstrated with mammalian Raf-1 (
39
,
40
), and the activation of the DRE/DREF system should follow this signaling
process. Then, what is the significance of the coordinated expression of
D-raf
with the DNA replication-related genes? There is no evidence for the participation of
D-raf
in DNA replication, but it has been reported that Raf-1 is activated during M phase (
41
). On the other hand, we have observed no significant accumulation of neuroblast
cells arrested in M phase in temperature-sensitive mutant larvae of
D-raf
at non-permissive temperature (
7
), and no abberation of mitosis in the cleavage division stage embryos lacking
both maternal and zygotic
D-raf
(L. Tsuda, H.-Y. Ha and Y. Nishida, manuscript in preparation). Thus, it is possible
that
D-raf
may function both in G1
and M phase and its redundant function in M phase is dispensable. Expressions
of DNA polymerase [alpha], PCNA and cyclin A are regulated by the cell cycle-dependent transcription and degradation of their proteins (
42
,
43
). In contrast,
D-raf
seems to be quite stable, since the maternally-provided
D-raf
is sufficient to support the development of animals hemi- or homozygous for the null functional mutation of
D-raf
until late third instar larval or early pupal stages (
6
,
8
). The DRE/DREF system-dependent expression of
D-raf
may result in a persistent increase of
D-raf
in the progenitor cells, and this would elevate their competence to growth-stimulating signals allowing their rapid and continuous proliferation as
observed in the imaginal discs. It is of interest to learn whether the
expression of
Dsor1
and
rolled (rl)
encoding the homologs of MAP kinase kinase and MAP kinase, respectively (
44
,
45
), are also under the control of the DRE/DREF system.
We are grateful to Dr C. Nüsslein-Volhard for
zen
mutant fly stocks and to Dr M. Moore for critical reading of the manuscript.
This work was supported by grants from the Korean Ministry of Education
(Genetic Engineering Research) to M.A.Y., and by grant-in-aid from the Ministry of Education, Science and Culture, Japan.
*To whom correspondence should be addressed. Tel: +82 51 510 2278; Fax: +82 51
513 9258; Email: mayoo@hyowon.cc.pusan.ac.kr
REFERENCES
Return