|
|
|
|||||||
|
|||||||||
(Received for publication, October 18, 1994; and in revised form, December 12, 1994) From the
The human HOXB2 gene is a member of the vertebrate Hox
gene family that contains genes coding for specific developmental stage
DNA-binding proteins. Remarkably, within the hematopoietic compartment,
genes of the HOXB complex are expressed specifically in
erythro-megakaryocytic cell lines and, for some of them, in
hematopoietic progenitors. Here, we report the study of HOXB2 gene transcriptional regulation in hematopoietic cells, an initial
step in understanding the lineage-specific expression of the whole HOXB complex in these cells. We have isolated the HOXB2 5`-flanking sequence and have characterized a promoter fragment
extending 323 base pairs upstream from the transcriptional start site,
which, in transfection experiments, was sufficient to direct the
tissue-specific expression of HOXB2 in the erythroid cell line
K562. In this fragment, we have identified a potential GATA-binding
site that is essential to the promoter activity as demonstrated by
point mutation experiments. Gel shift analysis revealed the formation
of a specific complex in both erythroleukemic lines K562 and HEL that
could be prevented by the addition of a specific antiserum raised
against GATA-1 protein. These findings suggest a regulatory hierarchy
in which GATA-1 is upstream of the HOXB2 gene in erythroid
cells. The mammalian Hox gene family contains 38 homeobox gene
members located in four independent complexes named HoxA, -B, -C, and -D(1, 2) that
code for proteins with a highly conserved DNA-binding domain of 61
amino acids closely related to the fly Antennapedia(3) . These
genes are expressed during embryonic development, during which they
have a determinant role in the body plan
organization(4, 5) . Hox genes could also be
implicated in the regulation of hematopoietic cell growth and
differentiation. In the mouse and in man, their expression is
restricted to only some of the hematopoietic cell types in which
aberrant expression may have a dramatic effect. Mutations and
translocations involving homeobox genes have also been reported in
cases of human leukemia (for a review, see (6) ). Several lines
of evidence indicate that Hox gene products are candidates for
the regulation of expression of some erythroid markers like the globin
genes, as has recently been suggested for human HOXB6 and
HOXB2(7, 8) . Therefore, Hox genes are
thought to act as specific transcription factors that could control
both the proliferation and the commitment of hematopoietic progenitor
cells in the differentiation pathway. Convergent data are available
showing that human HOXB genes are expressed in
erythro-megakaryocytic cells but not in granulo-monocytic cells. This
striking characteristic is shared by eight of the nine genes in the
complex(9, 10) . Furthermore, nothing is known about
the regulation of homeobox gene expression in hematopoietic cells,
including transcriptional regulation. The aim of our study was to
determine how this specific expression is regulated. We initially
focused our attention on the human HOXB2 gene because of a
previous study based on RT-PCR ( In this paper, we
report our studies of the regulation of HOXB2 expression in
human erythroid cell lines. Recently, Sham et al.(13) have reported the regulation of HoxB2 (Hox2.8) gene by Krox20 in the developing
nervous system of the mouse(13) . This regulation implicates
three sites for the Krox20 protein situated about 2 kb upstream from
the first exon, in the intergenic region between HoxB3 and HoxB2. Krox20, or its human homologue EGR2(14) , is considered to be an early response gene
that can be induced in various cell types, including hematopoietic
cells, by mitogenic agents such as phorbol esters or after serum
deprivation. Because the expression level of human HOXB2 mRNA
was not affected by 12-O-tetradecanoylphorbol-13-acetate
treatment of KG1, K562, and HEL lines, (
RNase
protection was carried out essentially as described (23) with
an antisense RNA probe spanning the region from -114 (EcoRI) to +296 (NarI, in the first exon).
Briefly, a 1.2-kb genomic fragment HindIII/NarI from
the BC206 cosmid was subcloned into the plasmid pBSK RT-PCR
was performed according to the protocol that was previously described (10) using 750 ng of cytoplasmic RNA, and the amplification
step was carried out for 25 cycles. Four sense oligonucleotides were
synthesized (Cyclone Plus DNA synthesizer, Millipore) to be used as
5`-primers in the PCR: primer 1, 5`-CCCAAAATCGCTCCATTACATAAAT-3` from
position +12; primer 2, 5`-TAAAAAAAAAGAGAGACCGAAATCTCCCCCT-3` from
position -19; primer 3, 5`-CGCTTGTATTTATCAGCAATA-3` from position
-62; and primer 4, 5`-TAGAGAGAGTCCCCATACGC-3` from position
-79. The 3`-primer was the antisense oligonucleotide
5`-GCGGGGGAAGGAAGCAGACACTCGG-3` from position +201. The amplified
fragments were revealed with a 35-mer antisense oligonucleotide probe,
5`-end-labeled with [
Figure 1:
Structure of the 5`-flanking region of HOXB2 gene. The map shows the intron-exon organization of the
human HOXB2 gene(20) . A 3.7-kb AsnI fragment
was subcloned from the BC206 cosmid clone into pBlueScript (pB2AsnI)
for restriction mapping and sequencing (Fig. 3). All restriction
sites for the indicated enzyme are shown on the map. A, AsnI; B, BamHI; D, DraII; E, EcoRI; Pv, PvuII; Ps, PstI.
Figure 4:
Cell-specific expression of the HOXB2 gene 5`-flanking sequence. On the left are
diagrams of HOXB2 promoter-CAT fusion constructs. Restriction
sites indicated are: D, DraII; H, HindIII; B, BstBI. The DraII
fragment containing 2200 bp upstream and 58 bp downstream from the
transcription start site was inserted before the CAT gene in the XbaI site of pBLCAT3. The -865CAT and -323CAT
constructs contain the HindIII-DraII and the BstBI-DraII fragments, respectively, in pBLCAT3. Each
plasmid was transfected into K562 and HeLa cells, and respective CAT
activities were measured 48 h after transfection. In each assay, the
pRSVL plasmid was cotransfected, and CAT assays were normalized
according to the luciferase activity. The promoterless plasmid pBLCAT3
was used as a negative control and measured the experimental
background. On the right, the relative CAT activity of K562
and HeLa cells transfected with each construct, normalized against
pBLCAT3 activity that was taken as 1, is plotted. Each data is the
average of three independent experiments performed in duplicate. The
standard variations are indicated in the
figure.
Figure 3:
Nucleotide sequence of the 5`-flanking
region of the HOXB2 gene. The sequence contains 1143
nucleotides upstream from the ATG codon and the beginning of the first
exon. All numbering is relative to the transcriptional start site
(+1). The HindIII(-865), BstBI(-323), and DraII (58) sites used
for construction of the expression vectors are underlined.
Putative regulatory elements are indicated in boldfacetype; the TATA box is underlined in italics, the GATA element is doubleunderlined, Ets binding elements are underlined,
and the ATTA sites are in italics.
The mutated HOXB2 fragments were obtained by site-directed mutagenesis in the
-48 GATA site, using the polymerase chain reaction as
described(25) , except that we used Pfu DNA Polymerase
(Stratagene) as the enzyme. The plasmid -323CAT was used as
matrix. Briefly, two simultaneous PCR reactions were performed. The
first one used two primers homologous to the pBLCAT3 vector sequence,
5`-AAGTTGGGTAACGCCAGGGT-3` from position 341 (24) and
5`-GAGCTAAGGAAGCTAA-3` from position 471, containing a mismatched
3`-end. The second one used the primer mutated on the -48 GATA
site, 5`-ATACGCTTGTATTTAGAAGCAATATACAATTA-3` from position
-65 in HOXB2 gene (the T and C in the GATA site were
mutated to G and A), and a more external primer in the CAT gene,
5`-GGCATTTCAGTCAGTTGCTC-3` from position 556 in pBLCAT3 vector.
Amplified fragments from each PCR reaction were purified, mixed, and
subjected to another round of PCR with the two most external primers
(341 and 556). The amplified fragments were digested with TaqI
enzyme, end flushed with Klenow fragment, digested with BamHI,
purified, and ligated into the pBLCAT3 vector that was itself first
digested with HindIII, end flushed, and then digested with BamHI. The sequence of the entire insert was verified, and the
insert borders were identical to those in the wild type -323CAT
plasmid.
K562
were transfected by the electroporation method (26) using a
gene pulser (Bio-Rad), set at 960 microfarads, 400 V, in a total volume
of 800 µl. Each assay was done with 40 µg of one of the CAT
constructs. 5 µg of pRSVL plasmid, containing the firefly
luciferase gene under the control of the Rous sarcoma virus promoter (27) were cotransfected with the HOXB2/CAT constructs
for internal control of transfection efficiency. The pRSVCAT plasmid
containing the CAT gene driven by the Rous sarcoma virus promoter was
used as positive control. HeLa cells were transfected by the calcium
phosphate method(28) , each assay containing 10 µg of CAT
constructs and 15 µg of pRSVL.
The double-stranded DNA probes used in gel mobility
shift assays were the following: 5`-CTGATGGGCCTTATCTCTTTACCCACCT-3`
from position -83 in the erythroid promoter of human
porphobilinogen deaminase gene (PBGD) used as a GATA-1 standard binding
site(33) ; the HNF-1-binding site from position -92 in
the promoter of human
Figure 2:
Identification of the transcriptional
start site of HOXB2 gene. A, RNase protection
analysis. 5-10 µg of poly(A)
Primer extension
experiments with a 35-mer antisense primer (from position +153 to
+119) allowed us to observe an extension product of approximately
150 nucleotides (data not shown), corresponding to the major band in
the RNase protection assay. For a definitive approach, we performed
a RT-PCR analysis (Fig. 2B). The antisense primer used
in the reverse transcription step was chosen 55 bases downstream from
the ATG codon and was also used in the amplification step. The four
5`-primers were chosen in the region that overlaps the previously
identified start site (from -60 to +36). Each 5`-primer
together with the 3`-primer was able to produce the correct PCR
fragment from genomic DNA. The RT-PCR experiment was performed in
parallel with mRNA from HEL cells and from HL-60 cells as a negative
control(10) . After electrophoresis, the PCR products were
blotted and hybridized with a labeled probe consisting of an
oligonucleotide located within the smallest fragment. A positive signal
was observed with HEL mRNA, only when primer number 1 (from position
+12 to +36) was used and not with the upstream
oligonucleotides. This result was consistent with RNase protection
analysis and allowed us to confirm the localization of the
transcriptional start site. This site is common for HEL, K562, and KG1a
cells and is situated 122 bp upstream from the ATG codon (Fig. 3).
Figure 5:
Effect of point mutation on the GATA
putative binding site in K562 cells. A mutation in the -48 GATA
motif was introduced in the -323CAT construct using a PCR
strategy as described under ``Materials and Methods.'' The
wild type (WT) -323CAT and the mutated (Mut)
-323CAT constructs were introduced into K562 cells for transient
transfection assays. The CAT values obtained with the different
plasmids were expressed relative to the wild type -323CAT, which
was taken as the 100% value. pBLCAT3 was used as a negative control.
Each data is an average of three independent replicates. Errorbars represent standard
deviations.
Thus, this clearly demonstrates that the -48 GATA motif is
functional and essential for the promoter activity of HOXB2 in
K562 cells.
Figure 6:
Gel mobility shift assays of the HOXB2 promoter probe containing the -48 GATA motif. 5 µg of
HeLa, HEL, or K562 nuclear extracts were incubated with the end-labeled HOXB2/-48 GATA probe (lanes 1-11) and
with the PBGD/-75 GATA probe (lanes 12-15) as
described under ``Materials and Methods.'' The binding
reactions were performed in the absence or presence of the indicated
unlabeled double-stranded competitor oligonucleotide (lanes3 and 8, homologous HOXB2/-48GATA; lanes4 and 9, HOXB2/mutated -48GATA; lanes5, 10, and 13, PBGD/-75GATA; lanes6, 11, and 14, non-homologous HNF1. The
mutated oligonucleotide contained two nucleotides changes in the core
GATA motif, the same changes that were used in the mutated construct
listed in Fig. 5. The PBGD oligonucleotide corresponded to the
-75 GATA element in the human porphobilinogen deaminase
gene(33) . The HNF1 oligonucleotide corresponded to the HNF1
element in human
In HEL and K562 cells, it has been
shown that two GATA proteins were expressed: GATA-1 and
GATA-2(17, 41, 42) . Both were therefore
candidates for this interaction. To discriminate between these two
proteins, we used HeLa cells that contain GATA-2 but not GATA-1 (41, 42, 43) . With these extracts, no
complex was observed (Fig. 6, lane1),
suggesting that in erythroid cells, the GATA-binding protein involved
in the interaction with HOXB2 promoter probe is GATA-1. In
addition, we performed a competition experiment with an oligonucleotide
encompassing the -75 GATA element of the PBGD promoter gene that
was used as a standard GATA-1-binding site(33) . The addition
of a 200-fold excess of this oligonucleotide prevented complex
formation with the HOXB2 probe (Fig. 6, lanes5 and 10). The -75 GATA site of PBGD gene
was also used as a probe to compare the electrophoretic mobility of the
two complexes. The identity of the GATA-binding protein to the PBGD
probe was verified using specific antiserum raised to a region of human
GATA-1. This antibody has been shown to bind specifically GATA-1
protein but not GATA-2 protein(35) . When antiserum was added
to the gel shift incubation mixture containing K562 cell nuclear
extracts, no interaction was observed (Fig. 6, lane15). Without competitor (lane12), the
complex formed with GATA-1 protein comigrated with HOXB2 site
complex (Fig. 6). To definitively and directly address the
identity of the GATA protein involved in the interaction with the GATA
site of HOXB2 promoter, we performed mobility shift assays
with the antiserum against human GATA-1 protein. Antiserum was
incubated with K562 or HEL extracts before the addition of the labeled
probe. In the two cell lines, formation of the specific DNA-protein
complex was not affected by the preimmune serum, whereas competition
with the antibody against GATA-1 completely inhibited the formation of
the complex with the DNA probe (Fig. 7). No other band was
visible on the autoradiography except a weak band corresponding to the
supershifted ternary complex.
Figure 7:
Immunological identification of protein
that binds to the -48 GATA motif in HOXB2 promoter. 5
µg (lanes1-3) or 2.5 µg (lanes
4-9) of K562 or HEL nuclear extracts were incubated with the
end-labeled oligonucleotide corresponding to the -48 GATA element
in HOXB2 promoter. Prior to addition of the probe, the
mixtures were incubated 10 min on ice without antiserum (lanes1, 4, and 7) or with 1 µl of
undiluted preimmune serum (lanes2, 5, and 8) or with 1 µl of undiluted antiserum raised to a human
GATA-1 specific peptide (35) (lanes3, 6, and 9).
These data show that GATA-1 is the
GATA protein that binds to the -48 GATA site of the HOXB2 gene in vitro and strongly suggest, together with
transfection experiments, that GATA-1 is the major regulator of the
erythroid-specific expression of HOXB2 gene. In this paper, we present the characterization of the
5`-flanking region of the HOXB2 gene and some insights into
the mechanisms that govern its regulation. The expression pattern of
this gene, as well as other HOX genes, in hematopoietic cell
lines has been established by different
groups(9, 10, 12, 44) . In erythroid
cell lines, a major transcript for HOXB2 of about 1.6 kb was
detected(12) . We identified a major cap site located 121 bp
upstream from the suggested initiation codon. This site is common to
K562 (erythroid), HEL (erythro-megakaryocytic), and KG1a (early
myeloid) cell lines. RNase protection experiments showed that these
cell lines contain similar amounts of HOXB2 mRNA. Three
deletion CAT constructs of 2200, 865, and 323 bp produced comparable
and substantial CAT activities when transfected in K562 cells. We could
observe a slight but consistent decrease between the -865 and
-323CAT constructs that may have functional significance, but we
did not attempt to study it more precisely. Thus, within the
5`-flanking region examined, most of the HOXB2 gene regulatory
sequences are contained in the 323-bp proximal domain. It also contains
erythroid-specific information as it directs only basal expression in
HeLa cells. Whether this sequence can mediate tissue-specific
expression in vivo must be verified in a transgenic mice
model. The examination of the 5`-flanking sequence revealed some
putative regulatory elements. A TATA box is present at -35 bp,
which is not usual for HOX genes. We could identify nine ATTA
or TAAT sites concentrated in 530 bp. This motif is the core consensus
sequence for homeoprotein binding. Similar clusters of homeobox protein
DNA-binding sites are found in cis-regulatory regions of several Drosophila homeotic genes (45) and in human HOXD9 (HOX4C)(46) . Interestingly, four of them are
organized in tandem sites. Actually, Galang and Hauser (47) showed that human HOXA5 (HOX1C) and HOXB5 (HOX2A) homodimers exhibit cooperative DNA
binding on tandem ATTA sites. These elements could be involved in
cross-regulation by other homeoproteins or in autoregulation mechanisms
by the HOXB2 protein, as has been suggested for HOXD9(46) , HOXC5 (HOX3D)(48) , or mouse HoxD4 (Hox4.2)(49) . Six putative binding sites for
members of the Ets family of transcription factors are present in the
-865CAT construct(50) . Some of these factors play a role
in hematopoietic gene expression. Implication of Ets proteins in the
regulation of the megakaryocyte-specific gene for glycoprotein IIb has
been reported (51) . In HOXB2 gene, two of these Ets
recognition sequences are still present in the -323CAT plasmid
and thus are more candidates as cis-acting elements. The small decrease
in CAT activities between the -865CAT and the -323CAT
constructs may be due to the loss of some Ets binding elements or ATTA
sites. We found neither CACCC nor TGAGTCA (AP1-binding site) motifs,
which are often present in erythroid
promoters(33, 52, 53) . A binding site for
the GATA family of transcription factors (54) is located at
position -48. Three members of this family are expressed in
hematopoietic cells. GATA-1 is present in erythroid, megakaryocyte,
mast cell, and eosinophil
lineages(42, 55, 56, 57) , whereas
GATA-2 is expressed in mast cell, basophil, eosinophil, neutrophil,
megakaryocyte, and erythroid lineages. Finally, GATA-3 expression is
restricted to T-cell, mast cell, and eosinophil
lineages(42, 54, 57) . GATA elements play a
central role in erythroid and megakaryocytic gene regulation.
Functional GATA elements have been described in the regulatory regions
of the vast majority of promoters of erythroid-expressed genes
including the genes for globins(58) , porphobilinogen
deaminase(51, 52) , and erythropoietin
receptor(59) , as well as several megakaryocytic expressed
genes like the genes for platelet glycoprotein IIb(35) ,
platelet factor 4(60) , and P-selectin (61) . In this
paper, we demonstrated that the GATA element upstream from HOXB2 gene is functional since its mutation reduced promoter activity by
85% and abolished its protein binding properties in gel mobility shift
assays. We showed that GATA-1 binds in vitro to this sequence
as indicated by three experiments: 1) the use of a negative control
cell line (HeLa) that expresses GATA-2 but not GATA-1 protein, 2) the
DNA-protein complex electrophoretic mobility and the competition with a
DNA-binding site for GATA-1, and 3) the use of a specific antibody
against GATA-1. Further investigation is necessary to determine if the
other HOXB genes that are essentially expressed in the
erythro-megakaryocytic compartment are also GATA-1 regulated. From the
sequences available in the literature, we could identify GATA consensus
sequences in mouse HoxB7 (Hox2.3) (62) and HoxB9 (Hox2.5) (63) promoters. Therefore, it
is possible that GATA-1 mediates erythro-megakaryocytic expression of HOXB genes through cis-elements in each promoter. In
conclusion, the study presented here is the first characterization of HOX gene promoter in hematopoiesis. We identified the
transcription factor GATA-1 as one of the major regulators of HOXB2 gene expression in erythroid cell lines. GATA-1, strongly
expressed in erythroid and megakaryocytic cell lineages, is likely to
be a major component of HOXB2-specific expression in the
hematopoietic compartment. The nucleotide sequence(s) reported in this paper has been submitted
to the GenBank(TM)/EMBL Data Bank with accession number(s)
X78978[GenBank].
Volume 270,
Number 9,
Issue of March 3, 1995 pp. 4544-4550
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
)analysis of
homeobox-containing transcripts, in which we established that the HOXB2 (HOX2H) gene was expressed not only in the
erythroid line K562, the erythro-megakaryocytic line HEL, and the
megakaryocytic line Meg01, as are the other members of the HOXB complex, but in two early myeloid cell lines, KG1a and KG1, as
well(10) . This characteristic led us to believe that HOXB2
could be an early regulator of both erythroid and megakaryocytic
lineages. HOXB2 gene is also expressed in normal human bone
marrow (9) as well as in progenitor cells(11) . Other
authors (12) have reported the existence of a transcript for HOXB2 of about 1.6 kb in hematopoietic cells. Very recently,
using specific antibodies, Sengupta et al.(7) have
demonstrated the presence of the HOXB2 protein in two
erythroid cell lines, K562 and HEL. The same authors have suggested
that the HOXB2 protein binds to specific sites within the
A
globin-regulating sequences(7) .
)whereas, under the
same conditions, 12-O-tetradecanoylphorbol-13-acetate modifies
expression of different genes coding for erythroid and megakaryocytic
markers(15) , we thought that HOXB2 gene could be
regulated by different pathways during brain ontogenesis and
erythropoiesis. Furthermore, the specificity of HOXB2 expression in early myeloid cell lines suggested the existence of
specific regulating regions for HOXB2 gene expression. To
study this, we isolated and characterized the 5`-flanking region of the
gene. Because the nucleotidic sequence contained a consensus site for
the erythroid factor GATA-1, which has been demonstrated to play a
major role in the regulation of expression of various specific
erythroid genes and in the terminal erythroid differentiation in
vivo(16, 17, 18, 19) , we
investigated the possible regulation of HOXB2 by GATA-1
factor. We report here the evidence that, in erythroid cell lines, the
expression of HOXB2 gene is under major control of GATA-1
protein, suggesting that HOXB2 is a part of the cascade of regulating
events that lead to the establishment of the erythroid phenotype.
Cell Culture
Cell lines were obtained
from the American Type Culture Collection. HEL (erythroleukemia), K562
(erythroid), HeLa (epithelial carcinoma), and KG1a cells were
maintained in RPMI 1640 medium (Life Technologies, Inc.) supplemented
with 10% fetal calf serum (Boehringer Mannheim).Isolation of the 5`-Flanking Region of HOXB2
Gene
A human genomic clone in a pcos2EMBL cosmid vector,
BC206, containing 36 kb of HOXB complex, was generously
provided by E. Boncinelli (20) and used as starting material. A
3.7-kb AsnI restriction fragment, including a large intergenic
region between the HOXB3 last exon and the HOXB2 first exon, was subcloned and inserted into the SmaI site
of the Bluescript I SK
plasmid vector
(pBSK, Stratagene) for restriction enzyme mapping and
DNA sequencing. All sequencing was carried out by the chain termination
procedure on both strands (21) using Sequenase (U. S.
Biochemical Corp.).Determination of Transcriptional Start
Site
Poly(A) RNAs were used for RNase
protection experiments. They were prepared from HEL, K562, and KG1a
cells by the thiocyanate guanidinium method (22) and purified
using the mRNA purification kit (Pharmacia Biotech Inc.). For RT-PCR
studies, RNA from HEL and HL60 were prepared by Nonidet P-40 lysis and
phenol extractions as previously described(10) . and linearized with EcoRI. The antisense RNA probe (472
bases) was generated with T3 RNA polymerase (Boehringer Mannheim) and
[
-
P]UTP (Amersham Corp.). 5-10 µg
of poly(A) RNA were hybridized overnight with 0.5
10
cpm of probe at 45 °C in standard buffer and
digested with 13 µg of RNase A (Boehringer Mannheim) and 2700 units
of RNase T1 (Life Technologies, Inc.). The RNase-resistant products
were subsequently analyzed on a 6% acrylamide sequencing gel.-P]ATP (DuPont NEN),
corresponding to the first 11 amino acids in the cDNA sequence (20) (5`-AACCCAATCTCCCTCTCAAATTCAAAATTCATGGC-3` from position
153).Construction of CAT-expressing
Vectors
The chloramphenicol acetyltransferase (CAT) plasmid
used was pBLCAT3(24) . It contains the coding region of the
bacterial CAT gene and the polyadenylation signal from SV40. The DraII genomic segment from pB2AsnI (Fig. 1), containing
the cap site and 2.2 kb of 5`-flanking sequences, was inserted into the XbaI site of pBLCAT3. The construct (named -2200CAT) was
sequenced in the 3`-side of the insert for confirmation. Two shorter
constructs were derived as follows: the -865CAT was prepared by
excision of the 1.3-kb HindIII fragment from the
-2200CAT and recircularization, and the -323CAT was
prepared by excision of the 1.9-kb HindIII/BstBI
fragment from the -2200CAT, followed by Klenow treatment and
recircularization (Fig. 4).
DNA Transfections
All plasmids used for
transfection were prepared with the Qiagen columns protocol.Luciferase and CAT Assays
Cells were
harvested 48 h after transfection, and cell extracts were obtained by
three cycles of freeze and thaw lysis. Luciferase activity of the
extracts was measured using the luciferase assay system (Promega
Biotec). CAT assays were performed essentially as described by Gorman et al.(29) . The amount of cellular extracts used in
the CAT assay was normalized according to luciferase activities.Nuclear Extracts
The nuclear extracts
were prepared from HEL, K562, and HeLa cells according to the method of
Dignam et al.(30) from exponentially growing cells.Gel Mobility Shift Assays
Gel mobility
shift assays were performed by a combination of the procedures of
Halligan and Desiderio (31) and of Singh et
al.(32) . For the binding reaction, 0.5-0.9 ng of
radiolabeled DNA fragment (2 10
cpm) were mixed
with 2.5-5 µg of nuclear extracts in a final volume of 10
µl containing 10 mM Tris-HCl (pH 7.5), 25 mM KCl,
1 mM EDTA, 1 mM dithiothreitol, 5% glycerol, and 2
µg of poly(dI-dC)
poly(dI-dC) (Pharmacia) used as nonspecific
competitor. Specific competitors and antisera were added to the
reaction mixture before the addition of the end-labeled probe as
described in individual experiments and incubated 10 min at +4
°C. Finally, samples were incubated for 15 min at room temperature
and analyzed on 5% polyacrylamide, 2.5% glycerol gels that were run in
0.5 TBE buffer (1 TBE buffer = 0.089 M Tris-HCl, 0.089 M boric acid, 0.002 M EDTA) for
1 h and 30 min at 200 V and then dried prior to overnight
autoradiography.
fibrinogen gene,
5`-AAAATTAAATATTAACTAAGGA-3`(34) ; the wild type and the
mutated GATA-binding sites from position -67 in the human HOXB2 gene promoter, respectively,
5`-CCATACGCTTGTATTTATCAGCAATATAC-3` and
5`-CCATACGCTTGTATTTAGAAGCAATATAC-3`.Antiserum
Specific rabbit antiserum
directed against the human protein GATA-1 was generously provided by F.
Martin. Anti-GATA-1 antibody was prepared by injecting a synthetic
peptide corresponding to a hydrophilic region of GATA-1 having no
homology with any sequence of other GATA proteins. The specificity of
the antibody has already been established(35) .
Isolation of the 5`-Flanking Region of the Human
HOXB2 Gene
To identify the promoter and the regulatory
sequences of HOXB2, we subcloned the HOXB2 upstream
genomic sequences from the human genomic clone BC206(20) . This
clone contains 36 kb of the HOXB complex from chromosome 17,
encompassing exons 2, 3, and 4 of HOXB3 gene, the entire HOXB2 gene, and the two exons of HOXB1, which is the
most 3`-gene in this complex. A 3.7-kb AsnI restriction
fragment containing the 5`-flanking region of HOXB2 was
subcloned into pBSK and analyzed by restriction
mapping (Fig. 1); 1 kb of proximal sequence was determined (Fig. 3).Determination of the Transcriptional Start
Site
To define the transcriptional start site of HOXB2 gene, three assays were used. The RNase protection assay was
performed using an antisense RNA EcoRI/NarI probe. A
strong protected band of 300 nucleotides in length was produced from
mRNA from K562, HEL, and KG1a cells but not from control tRNA (Fig. 2A). A band of smaller size (216 nucleotides) was
also weakly detected with the three cell lines.
RNA from HEL,
K562, KG1a (lanes1 and 2, 5 µg; lane3, 10 µg), or 10 µg of tRNA (lane4) were hybridized with the antisense RNA probe. After
treatment with RNase, the protected products were run on a 6%
sequencing gel. Sequencing reaction products of a non-related DNA were
used as size markers. As a control of the whole experiment, the 2
microglobulin (2m)-protected products were run in
parallel (amounts of RNA were as follows: lanes5 and 6, 0.25 µg; lane7, 0.5 µg; lane8, 10 µg). The major band corresponding to the HOXB2 protected fragment is indicated with a closedarrow, and the band corresponding to 2m is indicated with an openarrow. B,
RT-PCR analysis. Relative positions of primers (arrows) and probe (hatchedbox) used in the
analysis are indicated in the diagram. On the leftpanel, the control of 5`-primers efficiency tested with
DNA from HeLa cells is shown. Sizes of amplified fragments were,
respectively, 280, 263, 220, and 190 bp. On the rightpanel, the RT-PCR experiment performed with 750 ng of
cytoplasmic RNA from HEL and HL60 as negative control is reported.
Appropriate 5`-primer is indicated for each amplification (primer 1
(+12/+36), primer 2 (-19/+11), primer 3
(-62/-42), and primer 4 (-79/-60). The
+201/+177 antisense oligonucleotide was the 3`-primer. After
25 cycles of amplification following the reverse transcription step,
the PCR products were electrophoresed, blotted, and hybridized with the P end-labeled 35-mer antisense (+153/+119)
oligonucleotide. Plus (+) indicates samples treated with
reverse transcriptase, and minus(-) indicates non-treated samples
(see reference 10).
Structural Features of the 5`-Flanking
Region
The sequence of part of the first exon and about 1.1
kb upstream from the ATG codon is shown in Fig. 3. Two
independently isolated clones were sequenced, and no difference was
observed. A canonical TATA box was found 35 bp upstream from the
transcription start point but no CCAAT box(36) . A number of
potential regulatory elements are present, including at position
-48 in an inverted orientation, a putative binding site for the
GATA family of zinc finger transcription factors which consensus is
5`-(A/T)GATA(A/G)-3`(37) , six elements at positions
-303, -312, -381, -467, -498, and
-776 containing the sequences 5`-AGGAA(A/G)-3` or 5`-GAGGAA-3`,
which are recognized by factors belonging to the Ets class of
oncoproteins(38) , and nine elements at positions -21,
-37, -82, -105, -178, -189, -400,
-411, and -543 containing an ATTA core sequence that is
similar to the binding site (5`-TCAATTAAAT-3`) for Antennapedia class
of homeoproteins(39, 40) . Among them, four are
organized in tandem sites in an inverted orientation and are separated
by exactly 7 bp (-178 and -189 sites, -400 and
-411 sites).A 381-bp Fragment from the 5`-Flanking Region
Contains the Promoter Region and Is Sufficient for Tissue-specific
Expression in Erythroid Cells
To test whether the
5`-flanking region of the HOXB2 gene had promoter activity, we
inserted the HOXB2 5`-region upstream from a promoterless CAT
gene in the plasmid pBLCAT3. Three constructs of different sizes
(-2200CAT, -865CAT, and -323CAT) were analyzed in
transient CAT assay, both in erythroid cells K562, which express HOXB2, and in HeLa cells, which do not normally synthesize HOXB2(10) . The three constructs allowed the
expression of the reporter gene in K562 cells, whereas very weak
expression was detected in HeLa cells (Fig. 4). Although the
-323CAT was slightly less active than the other two in K562, it
contained sufficient elements to promote the specific activity of the
gene in erythroid cells. This characteristic feature allowed us to
think that the GATA motif at -48 could be a functional binding
site for erythroid factors belonging to the GATA family.The GATA Element at -48 Is Required for Optimal
Function of the HOXB2 Gene
The GATA motif at position
-48 is located 15 bp upstream of the TATA box and exhibits the
strict consensus 5`-(A/T)GATA(A/G)-3` in an inverted orientation (Fig. 3). To characterize the contribution of this motif to the
promoter activity, we mutated the wild type sequence TTATCA to TTAGAA
in the -323CAT construct. This mutation is known to disrupt the
binding activity of GATA proteins. The mutation resulted in an 85%
decrease in activity of the -323CAT construct (Fig. 5).
The Function of the -48 GATA Element Correlates
with Its Ability to Bind GATA-1 Protein
To determine
whether the -48 GATA element bound nuclear proteins, we
synthesized a 29-bp double-stranded oligonucleotide probe encompassing
this sequence. Incubation of the P end-labeled probe with
HEL and K562 nuclear extracts resulted in the formation of a single
complex (Fig. 6). This complex was sequence specific, as it was
prevented by addition of a 200-fold molar excess of the unlabeled
oligonucleotide (lanes3 and 8) but not by a
200-fold excess of a heterologous oligonucleotide (lanes6 and 11). Furthermore, the competition with 200-fold
excess of the HOXB2 oligonucleotide mutated on the -48
GATA site (same mutation than in Fig. 5, the TTATCA motif
mutated in TTAGAA) did not prevent the complex formation (lanes4 and 9), suggesting that the core GATA sequence
is required for the binding.
fibrinogen gene(34) . In lane15, the K562 nuclear extract was incubated with 1 µl
of antiserum against GATA-1 protein (35) before the addition of
the PBGD probe.
))
We thank E. Boncinelli for the generous gift of BC206
genomic clone and F. Martin for providing the GATA-1 antiserum. We are
grateful to F. Aubouy for artwork, to G. Uzan for useful discussions,
and to E. Dejana and D. DiMichele for careful reviewing of the
manuscript.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  E. Prabagaran, A.H. Bandivdekar, V. Dighe, and V.P. Raghavan HOXBES2: A Novel Epididymal HOXB2 Homeoprotein and Its Domain-Specific Association with Spermatozoa Biol Reprod, February 1, 2007; 76(2): 314 - 326. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. Lawrence, J. Christensen, S. Fong, Y.-L. Hu, I. Weissman, G. Sauvageau, R. K. Humphries, and C. Largman Loss of expression of the Hoxa-9 homeobox gene impairs the proliferation and repopulating ability of hematopoietic stem cells Blood, December 1, 2005; 106(12): 3988 - 3994. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. I. Sbodio and N.-W. Chi Identification of a Tankyrase-binding Motif Shared by IRAP, TAB182, and Human TRF1 but Not Mouse TRF1. NuMA CONTAINS THIS RXXPDG MOTIF AND IS A NOVEL TANKYRASE PARTNER J. Biol. Chem., August 23, 2002; 277(35): 31887 - 31892. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Nakagawa, S. Sugioka, Y. Kaneko, and S. Harashima O2R, a Novel Regulatory Element Mediating Rox1p-Independent O2 and Unsaturated Fatty Acid Repression of OLE1 in Saccharomyces cerevisiae J. Bacteriol., January 15, 2001; 183(2): 745 - 751. [Abstract] [Full Text]
|
|
|
|
|
|
L. B. Ludlow, B. P. Schick, M. L. Budarf, D. A. Driscoll, E. H. Zackai, A. Cohen, and B. A. Konkle Identification of a Mutation in a GATA Binding Site of the Platelet Glycoprotein Ibbeta Promoter Resulting in the Bernard-Soulier Syndrome J. Biol. Chem., September 6, 1996; 271(36): 22076 - 22080. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |