| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Biochemistry and Obstetrics-Gynecology The Cecil H. and Ida Green Center for Reproductive Biology Sciences University of Texas Southwestern Medical Center at Dallas Dallas, Texas 75235-9038
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
100 kDa, is a component of the complex of proteins that
bind the GT box of SP-A2. The finding that only two of the five GT
box-binding complexes were supershifted by incubation with Sp1 antibody
suggests that a factor(s) in type II cell nuclear extracts that is
distinct from Sp1 also interacts with the GT box. By UV
cross-linking and SDS-PAGE/EMSA analysis, we have identified a
55-kDa GT box-binding factor in type II cell nuclear proteins that
preferentially binds the GT box of SP-A2 over the consensus Sp1 GC box
sequence. This 55-kDa factor was able to bind the GT box independently
of Sp1. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
75% of
gestation is complete. Hormones and agents that increase intracellular
cAMP enhance the rate of type II cell differentiation in human fetal
lung tissue in culture and concomitantly increase transcriptional
activity of the SP-A gene (5, 6). The mechanisms whereby cAMP regulates
SP-A gene transcription in human fetal lung have not been defined. The human has two genes encoding SP-A, SP-A1 and SP-A2 (7, 8). The gene encoding SP-A2 is more highly regulated during development and is more sensitive to the inductive effects of cAMP than that encoding SP-A1 (9). We, therefore, have focused our efforts on identifying essential cis-acting elements and the corresponding trans-acting factors that mediate expression of the SP-A2 gene in a type II cell-specific and cAMP-regulated manner in human fetal lung.
Functional analysis of putative regulatory sequences upstream of the human SP-A2 gene in transfected type II cells has revealed that as little as -296 bp of 5'-flanking region directs type II cell-specific and cAMP-stimulated expression of SP-A2 promoter activity (10). By mutational analysis, a cAMP response element (CRE)-like sequence located at -242 bp (TGACCTTA), which we have termed CRESP-A2, was found to be essential for high basal and cAMP-inducible expression (10). A similar element located at -261 bp upstream of the rabbit SP-A gene also was found to be essential for cAMP-mediated induction of the rabbit SP-A promoter activity (11). Although the transcription factor(s) that binds this region has not yet been identified, studies to characterize this factor indicate that it is distinct from the cAMP-response element binding protein (CREB) and may be a member of the nuclear receptor family (12, 10). Additionally, putative binding sites for thyroid transcription factor-1 (TTF-1), which appears to be important for lung morphogenesis and expression of SP-A as well as other surfactant protein genes (13, 14), are also located within this -296 bp upstream region (J. Li, E. Gao, and C. R. Mendelson, unpublished observations).
In the present study, we have characterized a novel G-rich sequence (GT box), located at -61 bp upstream of the SP-A2 structural gene, that is crucial both for cAMP-inducible and basal expression of SP-A2 fusion genes in type II cells. Our studies indicate that the GT box-binding factors are enriched in type II cells as compared with lung fibroblasts, and that the ubiquitous transcription factor Sp1 is a component of a complex of type II cell nuclear proteins that interacts with this sequence. Furthermore, we provide evidence for the existence of an additional GT box-binding activity that clearly is distinct from Sp1 and may work together with Sp1 to regulate SP-A2 gene expression in type II cells.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). One conserved region located at -242 bp
contains a CRE-like sequence termed CRESP-A2. (TGACCTTA)
(Fig. 1). In previous studies we observed that mutagenesis of the
CRE-like sequence in the 5'-flanking region of the human (10) and
rabbit (12) SP-A genes markedly reduced basal and cAMP-induced
expression of SP-A:hGH fusion genes in transfected type II cells. Four
potential binding sites for the transcription factor TTF-1 were
identified in the murine SP-A promoter within a region corresponding to
-175 to -125 bp upstream of the human SP-A2 gene (13); however, the
consensus TTF-1 site at -172 bp in the 5'-flanking sequence of the
human SP-A2 gene appears to be the most highly conserved among
the promoters of the various species (Fig. 1). Another conserved region
located near the TATA box at -61 bp contains a core sequence of
GGGGTGGGG that we have termed the GT box; homologous regulatory
sequences (CA boxes) have recently been characterized in the bovine
papilloma virus promoter and in the locus control region of the
ß-globin gene (19, 20). The GT box sequence of the SP-A promoter has
not been previously studied.
|
).
SP-A2-296:hGH fusion genes containing the wild type and
mutated GT box were incorporated into replication-defective human
adenoviruses and introduced into primary cultures of human type II
cells by infection. Mutagenesis of the GT box within the -296 bp
fusion gene dramatically reduced basal (by >90%) and abolished
cAMP-induced expression. Expression of SP-A2-296GTmut:hGH
was similar to that of the minimal promoter construct,
SP-A2-47:hGH, which lacks the GT box sequence and includes
only the TATA motif (Fig. 2A). Basal expression of the
SP-A2-62:hGH construct, which just includes the GT box,
was >3-fold higher than the -47 bp minimal promoter construct (Fig. 2A). These data indicate that the GT box is essential for elevated
levels of basal and cAMP induction of SP-A2 promoter activity. In
previous studies, we observed that mutagenesis of the
CRESP-A2 sequence within the context of -296 bp fusion
gene reduced basal and abolished cAMP-inducible expression of the
fusion gene construct (10). In the present study, this construct was
transfected into type II cells for comparison with the GT box mutant.
As can be seen, expression of the SP-A2-296CREmut:hGH
fusion gene was decreased to levels comparable to those of the
SP-A2-62:hGH fusion gene, which includes only the GT and
TATA boxes. These findings suggest that both CRESP-A2 and
the GT box are necessary for basal and cAMP-inducible SP-A2 gene
expression and that the GT box plays a more crucial role in the
regulation of the SP-A2 promoter than does CRESP-A2.
|
). In each cell line, expression of
SP-A2-296GTmut:hGH was reduced to levels comparable to
those of the minimal promoter construct. These findings indicate that
the GT box contributes significantly to basal expression and cAMP
induction of SP-A2 promoter activity in primary cultures of type II
cells and to basal levels of SP-A2 promoter activity in pulmonary
epithelial cell lines H441 and A549.
GT Box of SP-A2 Binds Lung Nuclear Proteins; Protein Binding Is
Enriched in Type II Cells as Compared with Lung Fibroblasts
To investigate the interactions of putative
trans-acting factors with the GT box-regulatory sequence, we
performed electrophoretic mobility shift assays (EMSAs) using nuclear
proteins prepared from primary cultures of human lung type II cells,
primary cultures of human lung fibroblasts, and the two lung cell
lines, A549 and H441. Nuclear extracts isolated from each cell type
were incubated with a 32P-labeled oligonucleotide
containing the core sequence of the GT box of SP-A2 plus flanking
nucleotides. The DNA-protein complexes subsequently were resolved on a
nondenaturing acrylamide gel. The radiolabeled probe containing the GT
box of SP-A2 bound nuclear proteins isolated from type II cells as five
distinct complexes (lane 2, Fig. 3A). Each complex was
effectively competed with 100-fold excess of self but not with 100-fold
excess of an oligonucleotide containing the same five mutated residues
tested in transfection experiments (-296GTmut) (Fig. 3A, lanes 3 and 4, respectively). Because the GT box was found to be
functionally important for basal expression of SP-A2 promoter activity
in lung cell lines, we also examined the interaction of nuclear
proteins isolated from A549 (lane 4) and H441 (data not shown) cells
with the radiolabeled GT box. The binding pattern and activity of A549
and H441 nuclear proteins for the GT box were very similar to proteins
isolated from primary cultures of human fetal type II cells (Fig. 3B, lanes 3 and 4, and data not shown). By contrast, the protein binding
activity of the GT box was markedly reduced in nuclear extracts
prepared from primary cultures of human fetal lung fibroblasts as
compared with type II cells isolated from the same preparation of
cultured fetal lung tissue (Fig. 3B, lane 2). To assess the integrity
of the nuclear extracts prepared from the different cell types, we also
analyzed their binding activity for an oligonucleotide containing a
consensus CRE (TGACGTCA) sequence present in the rat somatostatin gene
(known to bind the ubiquitously expressed transcription factor CREB).
As can be seen, binding to the oligonucleotide containing the consensus
CRE was similar in type II cells, A549 cells, and lung fibroblasts
(Fig. 3C). These results indicate that factors that bind the GT box
element of SP-A2 are enriched in type II cells as compared with lung
fibroblasts and suggest that they may play an important role in
activating genes of pulmonary epithelial cells.
|
, lane 2); each complex was effectively competed with
200-fold excess of self (lane 3) but not with up to 200-fold excess of
an oligonucleotide containing a mutated GT box (lane 5) or with one
containing a consensus-binding site for transcription factor AP2 (lane
6), a protein also known to bind a GC-rich sequence (22). This provides
additional evidence that factors that comprise complexes 1–5 are
specific for the GT box of SP-A2. Interestingly, we observed that
excess nonradiolabeled Sp1 consensus oligonucleotide effectively
competed for the two lowest mobility complexes (complex 1 and 2) but
not the higher mobility complexes, 3, 4, or 5 (Fig. 4A, lane 4). In
parallel, we performed studies using an oligonucleotide containing the
Sp1 consensus site as the radiolabeled probe (Fig. 5A).
The radiolabeled GC box bound type II cell nuclear protein as only
three specific bands labeled a, b, and c (Fig. 5A, lane 2,); each
protein-DNA complex was competed by 200-fold excess of nonradiolabeled
GC box oligonucleotide (lane 3). Whereas the GT box oligonucleotide
effectively competed with the radiolabeled GC box for binding to
complexes b and c, it competed less well for binding to complex a (lane
4). On the other hand, an oligonucleotide containing the AP2 consensus
site failed to compete for binding (lane 5). It is unlikely that the
highest mobility GC box-binding complex, complex c, represents complex
3 observed in experiments using the GT box as probe because complex c
was effectively competed both by nonradiolabeled GC box and GT box,
whereas complex 3 was competed only by the GT box. These studies
suggest that the type II cell nuclear proteins that comprise complexes
3, 4, and 5 observed in EMSA using the radiolabeled GT box of SP-A2
failed to bind to the GC box under the conditions used.
|
|
, lane
4). By contrast, in experiments using the canonical GC box as the
labeled probe, the addition of an equivalent amount of antibody (0.1
µg) altered the mobility of complexes a and c, but had no detectable
effect on complex b (Fig. 5B, lane 4). The antibody itself did not
interact with either the radiolabeled GT box or GC box probes in the
absence of nuclear proteins (Figs. 4B and 5B, lane 2 of each). In
studies to directly analyze the interaction of Sp1 and the GT box, we
found that recombinant Sp1 bound both to the GT and the GC box as a
single complex; however, binding activity of Sp1 at low concentrations,
as determined by intensity of the shifted complex, was higher for the
GC box than for the GT box (data not shown). As a control, the binding
of purified AP2 protein to the radiolabeled GT box was also tested; no
interaction of AP2 and the GT box of SP-A2 could be detected (data not
shown). Additionally, as shown in Fig. 4, an AP2 consensus sequence
failed to compete for binding to any of the five GT box-nuclear protein
complexes in EMSA. Together, these studies suggest that Sp1 is a
component of a complex of proteins that interact with the GT box.
Furthermore, unlike the canonical Sp1 sequence, the GT box of SP-A2
also interacted specifically with another factor(s) in type II cell
nuclear proteins that is distinct from Sp1.
The GT Box of SP-A2 Interacts with a Factor(s) Distinct from
Sp1
In an attempt to further characterize the lung nuclear protein(s)
that comprises complexes 3, 4, and 5 and to begin to understand the
relationship between this factor(s) and Sp1, we performed UV
cross-linking. The GT box of SP-A2 was uniformly labeled and incubated
with type II cell nuclear proteins. The reaction was subjected to long
wave UV irradiation to covalently cross-link the proteins in contact
with the radiolabeled DNA, and the products were analyzed by SDS-PAGE.
Results of these experiments demonstrate that a protein of 55 kDa in
type II cell nuclear extracts specifically bound the radiolabeled GT
box probe (Fig. 6A, lane 2). This band was effectively
competed by 1000-fold excess of self (lane 4) but was more modestly
diminished by 1000-fold excess of the canonical GC box (Fig. 6A, lanes
7 and 8). The intensity of the 55-kDa band was unaffected by 1000-fold
excess of an oligonucleotide containing a mutated GT box (Fig. 6A, lanes 5 and 6). These findings suggest that the 55-kDa protein that
interacts with the GT box of SP-A2 has reduced binding activity for the
consensus Sp1-binding site, the GC box. A high molecular mass band
corresponding in size to Sp1 was not detected as a cross-linked product
using either the GT box or the canonical GC box as labeled probe. This
may be due to the low sensitivity of the UV cross-linking assay or,
alternatively, the amino acids in the DNA-binding domain of Sp1 may not
cross-link effectively to DNA. In the autoradiogram shown in Fig. 6A, specific binding of a number of proteins of <35 kDa also was apparent.
This was not a consistent finding from one experiment to another, and
we considered that they may be degradation products of the 55-kDa GT
box-binding protein. However, it should be noted that binding of low
molecular mass proteins distinct from Sp1 to the GT box of the human
c-myc gene promoter has been described; one of these is the
17-kDa PuF (nm23-H2 nucleoside diphosphate kinase) (23).
|
). Binding of the 55-kDa factor was barely detectable in lung
fibroblasts (Fig. 6B, lane 3); by contrast, binding activity in A549
cells was comparable to that in primary cultures of human fetal type II
cells (Fig. 6B, lanes 2 and 4).
To confirm that a 55-kDa nuclear factor did indeed interact with the GT
box of SP-A2, we size fractionated type II cell nuclear proteins by
SDS-PAGE and analyzed binding activity of the eluted proteins by EMSA.
Type II cell nuclear proteins were heated at 100 C in SDS-containing
sample buffer and resolved by SDS-PAGE. The gel was then cut into
slices, each slice spanning a molecular mass range as shown in Fig. 7. The proteins from each fragment were eluted in a
buffer containing Triton X-100, which serves to renature the proteins
by extracting the SDS. An aliquot of eluted proteins from each
fraction, which represented a distinct molecular mass range, was used
in two EMSAs, one using the GT box and the second using the consensus
GC box as radiolabeled probes. The GT box was found to bind strongly to
a factor in the molecular mass range 50–60 kDa as a single band on
EMSA (Fig. 7A). Intriguingly, in three of four independent experiments,
this factor failed to bind the radiolabeled GC box as determined by
EMSA (Fig. 7B). Very weak binding of a 50–60 kDa protein to the GC box
was detected in a fourth experiment (data not shown). On the other
hand, the GC box bound a factor in the size range of 90–150 kDa, which
is within the appropriate size range for Sp1 (100 kDa) (Fig. 7B). No
binding activity within this size range for the GT box was evident in
the experiment shown. In other experiments, we were able to detect some
binding activity of a high molecular mass factor(s); however, binding
activity of a 90–150 kDa protein(s) for the GT box was markedly
reduced as compared with the GC box (data not shown). Low levels of
binding activity for the GC box also were detected in the fraction
containing proteins of >150 kDa; however, this was not consistently
observed in all experiments. Neither of the radiolabeled probes was
found to bind nuclear proteins below 35 kDa (data not shown). These
findings support those of the UV cross-linking studies, which indicated
that a factor of 55 kDa in type II cell nuclear extracts interacts
with the GT box of SP-A2. By contrast, the 55 kDa protein has
dramatically reduced binding activity for the GC box. The SDS-PAGE/EMSA
experiments also indicate that this factor is able to interact directly
with the GT box, independent of Sp1.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
22 weeks and increased levels of SP-A protein are detected
in amniotic fluid after 30 weeks of gestation (24). Cyclic AMP
treatment of human fetal lung in culture increases the rate of
differentiation of type II cells and SP-A gene expression (25). In
previous studies, we observed that the human SP-A2 gene is more highly
regulated during development and by cAMP than that encoding SP-A1
(9).
In previous studies to map important cis-acting regulatory
elements of the human SP-A2 gene, we found that the sequences between
-47 and -296 bp are required for cAMP-induced transcription of
SP-A2:hGH fusion genes in primary cultures of type II cells. We found
that mutagenesis of a CRE-like element at -242 bp markedly reduced
basal levels of expression and abolished cAMP-induction of SP-A2
promoter activity (10). In the present study, we found that a 9-bp
G-rich sequence (GGGGTGGGG) present between -61 and -53 bp also acts
as an essential regulatory element. This element is conserved in the
promoter regions of the SP-A genes of a number of species thus far
studied. The functional significance of the GT box element in SP-A2
gene transcription was clearly demonstrated by the finding that
mutagenesis of this element within the context of the
SP-A2-296:hGH fusion gene reduced basal promoter activity
by >90% and abolished cAMP responsiveness in primary cultures of
human type II cells. A similar reduction of basal expression of SP-A2
promoter activity was observed in lung epithelial cell lines A549 and
H441. Furthermore, the SP-A2-62:hGH fusion gene construct,
which contains the TATA box and the GT box, was expressed at levels
that were 3-fold greater than those of the SP-A2-47:hGH
fusion gene, which lacks the GT box sequence. In previous studies, we
observed that fusion genes containing the wild type GT box but a
mutated CRESP-A2 sequence also lacked cAMP inducibility
(10). These findings suggest that basal and cAMP induction of SP-A
promoter activity is mediated by the concerted actions of transcription
factors bound to the GT box and at least one upstream sequence,
CRESP-A2, within the 296-bp 5'-flanking region. Cyclic AMP
stimulation of a number of eukaryotic genes has been found to be
mediated by cooperative interactions of transcription factors bound to
their respective cis-acting elements. For example, cAMP
responsiveness of the promoter for phosphoenolpyruvate carboxykinase
appears to be dependent upon the synergistic interaction of CREB bound
to a CRE with liver-enriched transcription factors, including
CCAAT/enhancer binding protein (C/EBP) family members and AP-1, bound
to their response elements (26).
Nuclear proteins isolated from primary cultures of type II cells bound the radiolabeled GT box probe as five specific complexes. Competitive and supershift EMSA analysis of the five complexes that comprise the GT box-binding activity in type II cell nuclear extracts indicates that the transcription factor Sp1 is a component of the lowest mobility complexes 1 and 2. Binding activity was similar in the lung epithelial cell lines A549 and H441 but was dramatically reduced in lung fibroblasts, suggesting that within the lung, specific binding is restricted to pulmonary epithelial cells. Mutagenesis of the same five GT box residues that were found to dramatically reduce fusion gene expression in functional assays also abolished the ability of the mutated oligonucleotide to compete with the radiolabeled wild type GT box probe for protein binding in EMSA. Interestingly, a GT box (or a homologous CA box)-like sequence is located at -51 bp upstream of the transcription initiation site of the human SP-B gene (27) and at -61 and -272 bp within the 5'-flanking sequence of the murine SP-C gene (28). The role of the GT/CA box sequences in the SP-B and SP-C genes has not yet been studied.
Sp1 was originally identified as a ubiquitously expressed protein that
binds to the hexanucleotide consensus GC box (GGGCGG). More recently,
Sp1 has also been found to bind GT box motifs, such as those found in
the rat LH receptor (29), SV40 (30), and T cell receptor (20) genes.
Although Sp1 is known to regulate basal expression of a variety of
housekeeping genes, its role in directing tissue-specific,
developmental, and hormonal regulation of gene expression has also
recently been explored. Thus far, Sp1 has been found to regulate
expression of erythroid- (31), lymphocyte- (32), and monocyte- (33)
specific genes. Additionally, Sp1 is also believed to serve a role in
the regulation of expression of the CYP11A gene through binding to an
element required for cAMP responsiveness (34). Sp1 is also implicated
as a modulator of the retinoic acid/cAMP-dependent transcription of the
tissue plasminogen activator gene (35). The mechanism by which Sp1
regulates expression of these genes is not yet defined. However,
modulation of Sp1 activity has been found to be mediated by changes in
binding activity (36), alternative splicing of its mRNA (37, 38), and
posttranslational modification, including phosphorylation (39). Sp1 may
also regulate gene expression through interactions with transcription
factors bound to other cis-acting elements. Sp1 has been
found to interact with NF-
B (40), Ets (41), and steroid receptors
(42, 43). Our findings that cAMP induction of
SP-A2-296:hGH expression in transfected type II cells
requires the presence of both an intact GT box and CRESP-A2
sequences (Fig. 2 and Ref. 10) suggest that cAMP induction of SP-A2
gene promoter activity is dependent upon the cooperative interaction of
Sp1 and the 55-kDa protein bound to the GT box with transcription
factors bound to CRESP-A2. Studies of the leukocyte
integrin gene CD11c suggest that myeloid-specific expression and
phorbol ester induction are facilitated by the cooperative interaction
of Sp1 and AP-1 transcription factors bound to their respective
response elements (44).
Analysis of the GT box-binding activity by competition and supershift
EMSA revealed that whereas Sp1 is a component of complexes 1 and 2, a
protein(s) distinct from Sp1 is a component of complexes 3, 4, and 5.
The finding, that similar complexes were not detected by EMSA using the
radiolabeled GC box as probe, supports our hypothesis that this
factor(s) differs from Sp1 in its binding activity. By UV cross-linking
analysis using a body-labeled GT box oligonucleotide as probe, we
consistently detected a single binding protein that migrated at 55
kDa. The finding that binding of the 55-kDa factor was reduced in lung
fibroblasts as compared with type II cells supports those of EMSA,
which indicated that the GT box-binding activity is enriched in lung
epithelial cells. Our inability to detect a binding protein of the
apparent size of Sp1 by UV cross-linking may be due to the relatively
poor ability of Sp1 to cross-link to DNA. To further characterize the
GT box factors, we resolved type II cell nuclear proteins by SDS-PAGE
and analyzed binding of the eluted proteins by EMSA. Our findings
confirmed that a factor of a molecular mass between 50–60 kDa does
indeed interact with the GT box and that this factor manifested
significantly higher binding activity for the GT box of SP-A2 as
compared with the GC box. By contrast, a higher molecular mass protein,
corresponding in size to Sp1, manifested increased binding activity
toward the consensus GC box as compared with the GT box. Our finding
that the 50–60 kDa factor can bind the GT box in the absence of Sp1
suggests that the binding of Sp1 and the 55-kDa factor to the GT box
can occur independently in vitro. This is further supported
by the findings of competitive EMSA, in which a nonradiolabeled GC box
specifically competed for complexes 1 and 2 without altering the
intensity or pattern of migration of complexes 3, 4, and 5 (Fig. 4A).
We considered the possibility that this 55-kDa band could represent an Sp1 degradation product. However, this is unlikely because the same size protein was consistently observed in all type II cell nuclear extracts as well as in nuclear extracts from A549 and H441 cells (data not shown). On the other hand, the 55-kDa protein could be an alternatively spliced form of Sp1. Each of the previously characterized alternatively spliced Sp1 mRNAs encode proteins that exhibit varying degrees of amino-terminal truncation and contain an intact Sp1 DNA-binding domain. Consequently, these isoforms exhibit very similar DNA- binding activity to that of full-length Sp1 (38). However, we observed that a 1000-fold excess of consensus Sp1 oligonucleotide competed only weakly with the radiolabeled GT box for binding to the 55-kDa protein in UV cross-linking assays, whereas a 1000-fold excess of nonradiolabeled GT box oligonucleotide effectively competed for binding to this protein. This finding indicates that the 55-kDa protein binds the GT box preferentially over the GC box sequence and suggests that the 55-kDa protein is distinct from Sp1. Furthermore, a 55- kDa alternatively spliced Sp1 isoform has not as yet been reported (37, 38).
Utilizing EMSA, supershift analysis, UV cross-linking, and SDS-PAGE combined with EMSA, we have demonstrated clearly that both Sp1 and a factor(s) distinct from Sp1 in molecular mass, binding activity, and antigenicity interact with the GT box of SP-A2. Although Sp1 has long been thought to be a unique GC box- and GT/CA box-binding protein, several groups have recently cloned cDNAs for novel GT/CA box-binding proteins. Like Sp1, these proteins contain three zinc fingers in their DNA-binding domains, but exhibit varying degrees of similarity within their transactivation domains (20, 45, 46, 47). On the basis of homology of Sp1 and the GT box-binding proteins to that of a Drosophila body pattern-determining gap gene, Krüppel, these factors constitute a new family of Krüppel-like proteins. Sp1, basic transcription element binding protein-1 (BTEB1), basic Krüppel-like factor (BKLF/TEF-2), and Krox20 represent members of this family that are widely expressed (21, 45, 48, 49). By contrast, basic transcription element binding protein-2 (BTEB2) appears to be restricted to the testis and placenta (46), while expression of erythroid Krüppel-like factor (EKLF) is limited to erythroid cells, suggesting that these proteins may be involved in the regulation of tissue-specific gene expression (50). Analysis of the DNA-binding activity of EKLF and BKLF indicate that they bind to CACCC sequence more avidly than to the GC consensus sequence (48).
Recently, a new member of this family of transcription factors, termed
lung Krüppel-like factor (LKLF), has been identified. Expression
of LKLF occurs predominantly in lung and spleen. Reduced levels of LKLF
expression can also be detected in heart, skeletal muscle, and testis
(51). Within the zinc finger DNA-binding domain, LKLF shares a high
degree of amino acid identity with the other GT box-binding proteins,
EKLF and BTEB2, and somewhat lower similarity with Sp1. In a
cotransfection assay of mouse NIH 3T3 cells, LKLF transactivated the
human ß-globin promoter through a GT (CA) box. This suggests that a
GT box may be the in vivo binding site for LKLF (51). The
calculated size of LKLF deduced from the amino acid sequence is 38
kDa; however, the apparent molecular mass estimated from SDS-PAGE of
LKLF has not been reported. Intriguingly, LKLF contains a highly
proline-rich transactivation domain (23% proline-rich excluding zinc
finger region); high proline content has been found to contribute to
slower migration in SDS-PAGE than expected based on the amino acid
sequence (52). Thus, we must consider the possibility that the apparent
molecular mass of LKLF on SDS/PAGE may be somewhat higher than the
calculated value, and that LKLF could be the factor that transactivates
SP-A2 promoter activity through the GT box at -61 bp. Currently, we
are investigating whether the 55-kDa factor represents LKLF or is a new
member of the Krüppel family. Isolation and characterization of
this 55-kDa factor and further study of its interaction with Sp1 will
provide insight into the regulatory mechanisms involved in SP-A gene
transcription in pulmonary type II cells.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Preparation and Maintenance of Cell Lines
The lung adenocarcinoma cell line A549 (ATCC CCL 185) of
presumed type II cell origin was maintained in Waymouth’s MB752/1
medium (GIBCO, Grand Island, NY) supplemented with FCS (10%, vol/vol),
100 U/ml penicillin, and 100 µg/ml streptomycin. The human lung
adenocarcinoma-derived cell line of presumed Clara cell origin,
NCI-H441 (54), was maintained in RPMI 1640 medium (GIBCO) containing
FCS (10%, vol/vol). Cells were grown to approximately 75% confluence
on 60-mm dishes and infected with 1 x 106 recombinant
infectious adenoviral particles containing various SP-A2:hGH fusion
genes. Alternatively, the cultured cells were used to prepare nuclear
extracts.
Construction of SP-A2:Human GH (hGH) Fusion Genes and Preparation
of Recombinant Adenoviruses
Fusion genes containing various amounts of the human SP-A2
5'-flanking DNA linked to the human GH (hGH) structural gene, as
reporter, were constructed as described previously (10). Briefly, SP-A2
genomic sequences were subcloned into the BamHI site of the
plasmid pACOGH, which contains the promoterless hGH structural gene
subcloned into the BamHI and EcoRI sites in the
polylinker of plasmid pAC1RR.5 (11). In this manner, the first 20-bp
segment of exon I of the SP-A2 gene was fused to the first exon of the
hGH structural gene. pAC1RR.5 contains sequences corresponding to the
left end of the adenovirus 5 genome from 0 to 1.4 and 9.1–17 map
units.
To construct the fusion gene containing mutations in the GT box of SP-A2, -1500 bp of SP-A2 5'-flanking sequence was subcloned into pBluescript KS. CJ236 Escherichia coli strain deficient in dUTPase and uracyl-N-glycosylase was transformed with pBluescript:SP-A2-1500 and M13K07 helper phage; single-stranded uracil containing phage DNA was isolated. This DNA template was used for site-directed mutagenesis that was performed according to the method outlined in Bio-Rad Mutagene Kit (Bio-Rad Laboratories, Richmond, CA) using the following oligonucleotide as primer (5'... GTAGAGCTCTCAGAATTCAGGAA GAAGCCTG... 3' [mutated nucleotides in italics and underlined]). The resulting plasmid, pBluescript:SP-A2-1500GTmut, was used to derive the SP-A2-296GTmut:hGH fusion gene. The accuracy of each construct was verified by double-stranded sequencing using the dideoxy chain termination method and a Sequenase kit (US Biochemical, Cleveland, OH).
To obtain recombinant viruses, 293 cells, a permissive human embryonic kidney cell line that has the capacity to produce E1a, were cotransfected with the recombinant pAC1RR.5 plasmids containing the fusion genes and with pJM17. The pJM17 plasmid, which contains the entire adenovirus genome plus insertion of a 4.3-kb pBR322 plasmid, is too large to be packaged into viral particles. Homologous recombination in vivo of the plasmids results in the formation of a recombinant viral genome that lacks the inserted pBR322 sequence and thus can be packaged into infectious viral particles (55, 56). Viral DNA was analyzed for the presence of the fusion genes by restriction endonuclease digestion followed by Southern analysis; the sequence was further verified by PCR sequencing (New England Biolabs, Beverly, MA). The recombinant viruses were titered to determine the concentration of infectious particles.
Expression of SP-A Fusion Genes in Transfected Cells
Type II cells in primary monolayer culture and lung cell lines
were incubated for 1 h with 1 x 106 recombinant
infectious viral particles; these were limiting with respect to the
number of plated cells, to achieve a multiplicity of infection of 0.2.
In this manner, the same number of cells (1 x 106)
were infected in each experiment, and fusion gene expression remained
consistent from experiment to experiment. After incubation with
recombinant adenoviruses, the medium was aspirated and replaced with
fresh medium with or without (Bu)2cAMP (1 mM).
Media from transfected cells were collected at 24-h intervals and
assayed for hGH by RIA using an hGH kit (Nichols Institute, San Juan
Capistrano, CA). The levels of hGH in the culture medium have been
shown to be proportional to the levels of hGH mRNA transcripts in the
cultured cells as determined by Northern blotting (11).
Synthetic Oligonucleotides for EMSAs and for UV Cross-Linking
Oligonucleotides were purchased from the custom primer synthesis
laboratory of GIBCO BRL (Gaithersburg, MD). Complementary strands were
annealed and chromatographed on Biospin 6 columns (Bio-Rad). The
following sequences were synthesized: the region from -66 to -48 bp
upstream of the SP-A2 gene transcription initiation site that contains
the GT box (5'..TCTCAGGGGTGGGGAAGAA..3') and a GT box in which the
nucleotides in italics have been mutated
(5'..TCTCAGAATTCAGGAAGAA. 3'). Oligonucleotides
were also synthesized containing the consensus binding sites of: Sp1,
(5'..ATTCGATCGGGGCGGGGCGAG..3'), AP2
(5'..GATCGAACTGACCGCCCGGCCCGT..3'); the CRE (in italics) of
the somatostatin gene (5'..AGCTCTCTCTGACGTCAGCCAAGG..3').
Purified double stranded oligonucleotides were used both as
radiolabeled probe and nonradiolabeled competitor.
EMSA
Nuclear extracts were prepared using a procedure described by
Dignam et al. (57) except that protease inhibitors (0.5
mM phenylmethylsulfonyl fluoride, 50 µg/ml aprotinin, 200
µg/ml leupeptin, and 10 µM pepstatin) were added to all
extraction solutions. Recombinant Sp1 and AP2 protein were purchased
from Promega (Madison, WI). Synthetic oligonucleotides were end-labeled
with T4 polynucleotide kinase and 
[32P]ATP, incubated
with nuclear proteins (5 µg) at room temperature for 30 min in
binding buffer (20 mM HEPES, pH 7.6, 12% glycerol, 70
mM KCl, 1 mM EDTA, 1 mM
dithiothreitol) and 0.5 µg of
poly(deoxyinosinic-deoxycytidylic)-poly(deoxyinosinic-deoxycytidylic)
acid as nonspecific competitor and resolved on a 6% polyacrylamide gel
(10, 58). The DNA-protein complexes were visualized by
autoradiography.
UV Cross-Linking
Double-stranded oligonucleotides containing the GT box of SP-A2
were phosphorylated with T4 kinase and subcloned into the
SmaI site of pUC 19 followed by body labeling using strand
synthesis as follows. The recombinant plasmids were denatured with 0.2
M NaOH and hybridized to a 5-fold molar excess of pUC/M13
universal primer (-40) and used as template for Sequenase (US
Biochemical Corp.) in the presence of dATP (60 µM), dTTP
(60 µM), dGTP (5 µM), dCTP (5
µM), [
-32P]dCTP (0.5 µM),
and [-32P]dGTP (0.5 µM). The labeled
DNAs were digested with BamHI and EcoRI to
produce a 52-bp fragment containing a portion of pUC19 polylinker plus
the sequences between -65 and -47 bp of the SP-A2 gene 5'-flanking
region. The resulting fragments were purified by microcon 100 (Amicon,
Beverly, MA) and bio-spin 6 (Bio-Rad), incubated with nuclear proteins
isolated from human fetal lung type II cells before and after culture
in the presence of (Bu)2cAMP using conditions described
above for EMSA, and subjected to UV irradiation for 60 min.
After digestion with DNase I and micrococcal nuclease, the DNA protein
complexes were resolved on an 11% SDS-polyacrylamide gel and
visualized by autoradiography.
Renaturation of Gel-Purified Type II Cell Nuclear Proteins
Followed by EMSA
Renaturation of lung nuclear proteins was performed as described
by Ossipow et al. (59) with minor modifications. Briefly,
type II cell nuclear proteins were boiled for 5 min in SDS-PAGE loading
buffer and separated in a SDS 11% polyacrylamide mini gel in parallel
with molecular mass standards (Amersham). The gel lane containing type
II cell nuclear proteins that had been separated by molecular mass was
subsequently cut into 13 slices of 30 mg, and each piece was
homogenized in 3 volumes of elution-renaturation buffer (1% Triton
X-100, 20 mM HEPES, pH 7.6, 1 mM EDTA, 100
mM NaCl, 5 mg/ml BSA, 2 mM dithiothreitol, 1
mM phenylmethylsulfonyl fluoride, 50 µg/ml aprotinin, 200
µg/ml leupeptin, and 10 µM pepstatin) to remove SDS
from protein-SDS complexes. This process sequesters SDS into micelles
so that SDS no longer interferes with DNA binding. Microcentrifuge
tubes containing the various homogenized gel fragments were incubated
for 3 h at 37 C and the polyacrylamide gel pieces were removed by
spinning at 56,000 rpm for 20 min using TLA 100.3 rotor (Beckman,
Fullerton, CA). Aliquots (14 ml) of supernatant from each fraction were
used in an EMSA using the radiolabeled GT box or GC box as probe.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
This research was supported in part by Basic Research Grant 1-FY94–0879 from the March of Dimes Birth Defects Foundation and by NIH Grant HL-50022. Pampee P. Young was supported in part by NIH Training Grant 5-T32-GM08014, a grant from the Perot Family Foundation, and by a predoctoral fellowship from The Chilton Foundation, Dallas, Texas.
Received for publication January 8, 1997. Revision received March 17, 1997. Accepted for publication April 2, 1997.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
globin gene binds Sp1 and is a
functional promoter element in vitro and in vivo.
J Biol Chem 266:8907–8915 globin gene promoter may regulate developmental
specificity of expression. Mol Cell Biol 13:3272–3281This article has been cited by other articles:
 |
|
 |
|
  D. Liu, H. Benlhabib, and C. R. Mendelson cAMP Enhances Estrogen-Related Receptor {alpha} (ERR{alpha}) Transcriptional Activity at the SP-A Promoter by Increasing Its Interaction with Protein Kinase A and Steroid Receptor Coactivator 2 (SRC-2) Mol. Endocrinol., June 1, 2009; 23(6): 772 - 783. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Liu, M. Yi, M. Smith, and C. R. Mendelson TTF-1 response element is critical for temporal and spatial regulation and necessary for hormonal regulation of human surfactant protein-A2 promoter activity Am J Physiol Lung Cell Mol Physiol, August 1, 2008; 295(2): L264 - L271. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Liu, M. M. Hinshelwood, V. Giguere, and C. R. Mendelson Estrogen Related Receptor-{alpha} Enhances Surfactant Protein-A Gene Expression in Fetal Lung Type II Cells Endocrinology, November 1, 2006; 147(11): 5187 - 5195. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Nakano, K. Fukuhara-Takaki, T. Jono, K. Nakajou, N. Eto, S. Horiuchi, M. Takeya, and R. Nagai Association of Advanced Glycation End Products with A549 Cells, a Human Pulmonary Epithelial Cell Line, Is Mediated by a Receptor Distinct from the Scavenger Receptor Family and RAGE. J. Biochem., May 1, 2006; 139(5): 821 - 829. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. N. Islam and C. R. Mendelson Permissive Effects of Oxygen on Cyclic AMP and Interleukin-1 Stimulation of Surfactant Protein A Gene Expression Are Mediated by Epigenetic Mechanisms Mol. Cell. Biol., April 15, 2006; 26(8): 2901 - 2912. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-Y. Park, H. D. Jang, S. Y. Lee, K.-J. Lee, and E. Kim Fas-associated Factor-1 Inhibits Nuclear Factor-{kappa}B (NF-{kappa}B) Activity by Interfering with Nuclear Translocation of the RelA (p65) Subunit of NF-{kappa}B J. Biol. Chem., January 23, 2004; 279(4): 2544 - 2549. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Gao, Y. Wang, J. L. Alcorn, and C. R. Mendelson Transcription factor USF2 is developmentally regulated in fetal lung and acts together with USF1 to induce SP-A gene expression Am J Physiol Lung Cell Mol Physiol, June 1, 2003; 284(6): L1027 - L1036. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Phillips, J. Bailey, S. C. Robson, and G. N. Europe-Finner Differential Expression of the Adenylyl Cyclase-Stimulatory Guanosine Triphosphate-Binding Protein Gs{alpha} in the Human Myometrium during Pregnancy and Labor Involves Transcriptional Regulation by Cyclic Adenosine 3',5'-Monophosphate and Binding of Phosphorylated Nuclear Proteins to Multiple GC Boxes within the Promoter J. Clin. Endocrinol. Metab., December 1, 2002; 87(12): 5675 - 5685. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. N. Islam and C. R. Mendelson Potential Role of Nuclear Factor {kappa}B and Reactive Oxygen Species in cAMP and Cytokine Regulation of Surfactant Protein-A Gene Expression in Lung Type II Cells Mol. Endocrinol., June 1, 2002; 16(6): 1428 - 1440. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-S. Yang, M.-C. W. Yang, B. Wang, and J. C. Weissler BR22, a Novel Protein, Interacts with Thyroid Transcription Factor-1 and Activates the Human Surfactant Protein B Promoter Am. J. Respir. Cell Mol. Biol., January 1, 2001; 24(1): 30 - 37. [Abstract] [Full Text]
|
|
|
|
|
|
P. R. Provost, C. H. Blomquist, C. Godin, X.-F. Huang, N. Flamand, V. Luu-The, D. Nadeau, and Y. Tremblay Androgen Formation and Metabolism in the Pulmonary Epithelial Cell Line A549: Expression of 17{beta}-Hydroxysteroid Dehydrogenase Type 5 and 3{alpha}-Hydroxysteroid Dehydrogenase Type 3 Endocrinology, August 1, 2000; 141(8): 2786 - 2794. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Suga, M. Kurabayashi, Y. Sando, Y. Ohyama, T. Maeno, Y. Maeno, H. Aizawa, Y. Matsumura, T. Kuwaki, M. Kuro-o, et al. Disruption of the klotho Gene Causes Pulmonary Emphysema in Mice . Defect in Maintenance of Pulmonary Integrity during Postnatal Life Am. J. Respir. Cell Mol. Biol., January 1, 2000; 22(1): 26 - 33. [Abstract] [Full Text]
|
|
|
|
|
|
M. D. Bruno, J. A. Whitsett, G. F. Ross, and T. R. Korfhagen Transcriptional Regulation of the Murine Surfactant Protein-A Gene by B-Myb J. Biol. Chem., September 24, 1999; 274(39): 27523 - 27528. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Alcorn, R. E. Hammer, K. R. Graves, M. E. Smith, S. D. Maika, L. F. Michael, E. Gao, Y. Wang, and C. R. Mendelson Analysis of genomic regions involved in regulation of the rabbit surfactant protein A gene in transgenic mice Am J Physiol Lung Cell Mol Physiol, August 1, 1999; 277(2): L349 - L361. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Ahlgren, G. Suske, M. R. Waterman, and J. Lund Role of Sp1 in cAMP-dependent Transcriptional Regulation of the Bovine CYP11A Gene J. Biol. Chem., July 2, 1999; 274(27): 19422 - 19428. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R. Farina, A. Tacconelli, A. Vacca, M. Maroder, A. Gulino, and A. R. Mackay Transcriptional Up-Regulation of Matrix Metalloproteinase-9 Expression during Spontaneous Epithelial to Neuroblast Phenotype Conversion by SK-N-SH Neuroblastoma Cells, Involved in Enhanced Invasivity, Depends upon GT-Box and Nuclear Factor {{kappa}}B Elements Cell Growth Differ., May 1, 1999; 10(5): 353 - 367. [Abstract] [Full Text]
|
|
|
|
|
|
D. R. Breed, L. R. Margraf, J. L. Alcorn, and C. R. Mendelson Transcription Factor C/EBP{delta} in Fetal Lung: Developmental Regulation and Effects of Cyclic Adenosine 3',5'-Monophosphate and Glucocorticoids Endocrinology, December 1, 1997; 138(12): 5527 - 5534. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Gao, Y. Wang, J. L. Alcorn, and C. R. Mendelson The Basic Helix-Loop-Helix-Zipper Transcription Factor USF1 Regulates Expression of the Surfactant Protein-A Gene J. Biol. Chem., September 12, 1997; 272(37): 23398 - 23406. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. He, E. C. Crouch, K. Rust, E. Spaite, and S. L. Brody Proximal Promoter of the Surfactant Protein D Gene. REGULATORY ROLES OF AP-1, FORKHEAD BOX, AND GT BOX BINDING PROTEINS J. Biol. Chem., September 29, 2000; 275(40): 31051 - 31060. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Yi, G.-X. Tong, B. Murry, and C. R. Mendelson Role of CBP/p300 and SRC-1 in Transcriptional Regulation of the Pulmonary Surfactant Protein-A (SP-A) Gene by Thyroid Transcription Factor-1 (TTF-1) J. Biol. Chem., January 18, 2002; 277(4): 2997 - 3005. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
