Endocrinology Vol. 138, No. 7 2856-2862
Copyright © 1997 by The Endocrine Society
Bovine Insulin-Like Growth Factor Binding Protein-3: Organization of the Chromosomal Gene and Functional Analysis of Its Promoter
N. E. Erondu,
B. Toland,
M. Boes,
B. Dake,
D. R. Moser and
R. S. Bar
Diabetes and Endocrinology Research Center, Department of Internal
Medicine, The University of Iowa and Veterans Administration Medical
Center, Iowa City, Iowa 52246
Address all correspondence and requests for reprints to: Ngozi E. Erondu, The University of Iowa Department of Internal Medicine, ENDO 3E17 VA Medical Center, Iowa City, Iowa 52246.
 |
Abstract
|
|---|
Insulin-like growth factor binding protein-3 (IGFBP-3), the major IGFBP
in the circulation, is synthesized by the vascular endothelium
in vivo and has been shown to be an important modulator
of the physiological effects of IGF. IGFBP-3 is regulated by a number
of growth factors/cytokines to which the vascular endothelium is
exposed, including IGF-I stimulation and TGF-ß1 inhibition of IGFBP-3
in cultured endothelial cells. To understand the mechanisms of
transcriptional regulation of IGFBP-3, we have cloned the bovine
IGFBP-3 gene and begun the functional analysis of its promoter.
Southern analysis indicated a single copy gene. The gene spanned
approximately 10 kb and was divided into five exons, the fifth
containing the 3' untranslated region. The transcription start site was
137 bp upstream of the initiation codon and a TATA box was located 26
bp 5' to this CAP site. No CAAT box was present but a GC rich sequence
element, containing two overlapping putative AP-2 binding elements, was
located 5' to the TATA box. Transient transfection studies with a
series of 5' truncated luciferase reporter constructs were conducted in
primary cultures of bovine aorta endothelial cells. Results of the
transfection studies indicated that 1) nearly 80% of the maximal basal
promoter activity was retained within the first 130 bp of the 5'
flanking sequence; 2) this region responded to IGF-I, despite lacking
the TTF-1/TTF-2 (thyroid specific transcription factors) binding
elements that are required for IGF-I stimulation of thyroglobulin
synthesis. These binding elements have also been suggested to be
involved in IGF-I regulation of IGFBP-3 transcription, thus, implying
the existence of novel cis-acting elements that mediate
the IGF-I stimulation of bovine endothelial cell IGFBP-3 mRNA
synthesis; 3) deletion of the GC rich sequence element resulted in a
60% reduction in basal promoter activity as well as loss of the IGF-1
stimulatory effect; 4) the TGF-ß1 mediated inhibition of IGFBP-3
transcription required sequence element(s) beyond 1.5 kb of its
promoter.
 |
Introduction
|
|---|
THE INSULIN-LIKE growth factors, IGF-I and
IGF-II, are mitogenic peptides that affect differentiation and
metabolic activity in many cell types (1, 2, 3, 4). In the circulation and in
tissues, they are noncovalently bound to one of six insulin-like growth
factor binding proteins, IGFBP-1, -2, -3, -4, -5, and -6 (3).
Considerable evidence points to the essential role of the IGFBPs in
controlling and regulating the biological activities of IGFs. In
addition, some of the IGFBPs, for example, IGFBP-3, may possess
intrinsic biological activity independent of any interaction with IGF
(5).
IGFBP-3 is the major IGFBP in the circulation, accounting for the
binding of >90% of the plasma IGFs in a trimeric 150-kDa complex
containing IGF, IGFBP-3, and an acid labile subunit (6). IGFBP-3 is
secreted by cultured endothelial cells (7) and its mRNA has been
localized in endothelial cells in a variety of human and rat tissues,
in vivo (8, 9, 10). A diverse number of substances and
conditions have been reported to modulate the production of IGFBP-3,
acting at both transcriptional and posttranscriptional levels. We have
previously shown that IGF-I and TGF-ß1 do not affect IGFBP-3 mRNA
stability in cultured bovine endothelial cells (11), suggesting that
the reported IGF-1 stimulation and TGF-ß1 inhibition of IGFBP-3 mRNA
occur at the level of gene transcription. IGF-I stimulation of IGFBP-3
mRNA has been reported in several cell types; however, the TGF-ß1
induced inhibition of IGFBP-3 mRNA synthesis has only been demonstrated
in bovine endothelial cells. To better understand the molecular basis
for these observations, we have isolated and sequenced the chromosomal
gene for bovine IGFBP-3 and have begun the characterization of its
promoter.
 |
Materials and Methods
|
|---|
Materials
Restriction enzymes were purchased from New England Biolabs
(Beverly, MA), TGF-ß1 from R & D Systems, (Minneapolis, MN),
oligonucleotides from Genosys, Biotechnologies, Inc. (Woodlands, TX),
whereas QAYL IGF-I was a kind gift from Dr. M.Cascieri (MSD, Rahway,
NJ).
cDNA library preparation and cloning of IGFBP-3 cDNA
Total RNA was isolated from bovine microvessel endothelial cells
by the CsCl gradient centrifugation method (12). Poly(A)+ RNA was
purified and oligo (dT) primed cDNA synthesized. The cDNA was ligated
into the
Uni Zap XR cloning vector (Stratagene, La Jolla, CA) and
packaged to generate a library of 5 x 106 plaque
forming units (pfu), with approximately 99% recombinants. Aliquots of
this library containing approximately 50,000 pfu each were plated on
XL1-Blue cells to generate the amplified cDNA library subsequently used
for screening. A 480-bp PCR product corresponding to nucleotides 516 to
996 was generated from primers based on the published bovine cDNA
sequence (13). This PCR product was labeled to high specific activity
by the random priming method using 32P-deoxy CTP. The
labeled probe was used to screen the microvessel cDNA library. The cDNA
clone, BP-3.511, containing the largest insert (
2.4 kb) was chosen
for further characterization.
Isolation of the bovine IGFBP-3 chromosomal gene
A commercial bovine genomic library (Stratagene) was screened
according to the manufacturers instructions using radiolabeled
IGFBP-3 probe. DNA from plaque purified positive clones was digested
with a number of restriction enzymes and subcloned into
pBluescript(Stratagene). Flourescent automated DNA sequencing was
performed in the DNA core facility at the University of Iowa utilizing
PE-ABD 373 automated DNA sequencers (Perkin-Elmer, Foster City,
CA).
Southern blotting
Total cellular DNA was isolated from bovine aorta endothelial
cells (12). DNA (20 µg) was digested with restriction enzymes
(EcoRI, HindIII and XhoI) according to
the suppliers instructions. The fragments were separated on a 0.8%
agarose gel and transferred to a nylon membrane. The IGFBP-3 PCR
product generated as described above was labeled with 32P
by the random priming method and used to probe the filter under
standard conditions (14).
Primer extension analysis
A synthetic 23 base oligonucleotide (5'-CTGGGC GGC AGC GAG CTG
AGC GA-3') complementary to nucleotides -50 to -73 relative to the
translation initiation site of the bovine cDNA was end-labeled with
[32P]ATP using T4 polynucleotide kinase (Promega,
Madison, WI). The labeled oligonucleotide was hybridized to either
bovine endothelial cell poly(A)+ RNA or yeast transfer RNA
at 65 C for 30 min and slowly cooled to room temperature. After
hybridization, the primer was extended at 42 C for 30 min. The
hybridization buffer, extension buffer and reverse transcriptase were
supplied in the Promega primer extension kit. The reaction products
were separated on a 6% polyacrylamide-7 M urea sequencing
gel in parallel with a sequencing reaction to determine the sizes of
the reaction products.
Construction of recombinant plasmids
A 3.0-kb EcoRI genomic fragment containing
1.5 kb
of IGFBP-3 5' flanking region was subcloned into pBluescript
(pBS:3000R). This plasmid was double digested with EcoRV and
Cel II to release a 1480 bp fragment corresponding to
nucleotides -1409 to +70 relative to the transcription start site.
This fragment was subcloned into the SmaI site of the basic
luciferase reporter vector, pGL3 in sense (pGL3:1480S) and antisense
(pGL3:1480AS) orientation. To generate a larger construct, a 6-kb
EcoRI genomic fragment, immediately upstream of the 3-kb
EcoRI fragment, was subcloned into the EcoRI site
of pGL3:1480S to generate a construct that had 7.4 kb of 5' flanking
region (pGL3:7400S). Plasmid pGL3:7400S was digested with
HindIII and the 3.0 kb HindIII fragment subcloned
into the HindIII site of pGL3 in two
orientationspGL3:3000S and pGL3:3000AS. pGL3:7400S was also double
digested with HincII and XhoI and an
approximately 4 kb 5' flanking fragment subcloned into pGL3 to yield
pGL3:4000S. Plasmid pBS:3000R was digested with RsaI to
yield a 520-bp fragment that was subcloned into the SmaI
site of pBluescript (pBS:520S). An
300 bp SacII fragment
was excised from pBS:520S and the remaining recombinant containing 220
bp of 5' flanking region religated (pBS:220S). Plasmids pBS:520S and
pBS:220S were double digested with SacI and
HindIII and the released fragments subcloned into pGL3
vector that had been predigested with SacI and
HindIII, to generate pGL3:520S and pGL3:220S. Plasmid
pGL3:220S was double digested with KpnI (3' overhang) and
Ecl 136 I (blunt end) followed by sequential incubation with
Exonuclease III and Mung bean nuclease under conditions recommended by
the manufacturer (Stratagene). A number of nested deletions were
created and one of them, pGL3:180S, contained 188 bp of 5' flanking
region. Primers were designed for specific amplication of both a 140 bp
and a 110 bp segment of the 5' flanking region of the IGFBP-3 gene. The
upstream primer for the 140 bp segment (5'-GAG AGA TAG GAG
CTC AGC CGG CGC GCC GCT-3') is a composite primer
which contains 16 bases at the 3' end (bold print) that
correspond to nt -76 to -61 relative to the CAP site. The upstream
primer for the 110 bp segment (5'-GAG AGA TAG TGA GCT
CGG CCG CCC GGC TTC-3') is also a composite primer which
contains 17 bases at the 3' end (bold print) which correspond to nt -48
to -32 relative to the CAP site. The 5' end of each primer was designed
to contain a SacI site, GAG CTC (italicized). The
downstream primer was T7 primer (5'-GTA ATACG ACTC ACTA TAGG GC-3') and
the template used in the PCR was pBS:220S. The polymerase chain
reaction was carried out under standard conditions (12). Following
purification, the product was double digested with SacI and
HindIII and the released fragment subcloned into pGL3 basic
vector to generate pGL3:140S and pGL3:110S.
Cell culture and DNA transfection
Bovine aorta endothelial cells were prepared and characterized
as previously described (11). Cells were grown to confluence in 35-mm
well plates, washed and incubated in serum free media before
transfection. Plasmids were purified using the QAIGEN (Chatsworth, CA)
midi plasmid prep kit. Cells in each well were transfected by
incubation at 37 C for 4 h in a reaction mixture containing 2 µg
test plasmid, 1 µg pCMV SEAP (Tropix, Bedford, MA) and 10 µl
Transfectam, a cationic lipid purchased from Promega. After
transfection, cells were incubated for an additional 4 h in serum
free media at 37 C. Secreted alkaline phosphatase (from pCMV SEAP) in
each well was assayed (Tropix) to check for any differences in
transfection efficiency. Subsequently, fresh serum free media ±
the indicated hormones was added to the wells and incubation continued
for 18 h. The cells were washed with
Mg2+/Ca2+ free PBS and lysed with reporter
lysis buffer (Promega). Luciferase activity was determined in duplicate
according to the manufacturers recommendations. The protein
concentration in the lysate was determined by the Bradford method
(15).
 |
Results
|
|---|
Complete sequence of bovine IGFBP-3 cDNA
IGFBP-3 mRNA, across species and tissues, is approximately 2.5 kb
in size (7). By comparing published human and bovine IGFBP-3 cDNAs, it
became apparent that the published bovine sequence is incomplete,
lacking nearly 900 bp of the 3' untranslated region (13). In view of
the potential role of this region of the mRNA in determining mRNA
stability (16) we considered it necessary to isolate a full length
bovine IGFBP-3 cDNA clone. Sequence analysis of the clone with the
largest insert (BP-3.511) revealed that it contained the full coding
sequence, 120 bp of 5' untranslated region and 1420 bp of 3'
untranslated region. The partial sequence of BP-3.511, shown in Fig. 1
, begins with the last nucleotide residue (G) of the
published sequence (13). It should be noted that a stretch of 13 A
residues follows the last G in the previously published sequence and
this poly A stretch probably served as a site for oligo dT primed cDNA
synthesis to generate the 1.6 kb cDNA isolated from cow liver (13).
Underlined and in bold print are an AAATAAA
polyadenylation signal motif and an ATTTA motif which may promote mRNA
degradation when located in the 3' untranslated region of an mRNA
transcript (16).

View larger version (46K):
[in this window]
[in a new window]
|
Figure 1. 3' untranslated sequence of bovine IGFBP-3
cDNA. Sequence from the bovine cDNA clone BP-3.511 beginning with the
last nucleotide residue (G) in the previously published sequence (13 ).
Underlined and in bold print are the
polyadenylation signal motif (AAATAAA) and an ATTTA motif that plays a
role in eukaryotic mRNA degradation.
|
|
Isolation and characterization of the bovine IGFBP-3 chromosomal
gene
A commercial bovine genomic library (Stratagene), was screened
using radiolabeled IGFBP-3 probe. DNA from plaque purified positive
clones was digested with a number of restriction enzymes and subcloned
into pBluescript. The two phage clones (bov BP31 and bov BP32, Fig. 2A
) that were isolated initially, lacked Exon 5 and the
library was rescreened with a 3' fragment of IGFBP-3 cDNA (a 550-bp
SacI-XhoI fragment). The third phage clone (bov
BP33) contained all the exons and both 5' and 3' flanking
regions(Fig. 2A
). A schematic representation of the bovine IGFBP-3
gene, which depicts the relative size and location of its exons and
introns is shown in Fig. 2A
. Like human IGFBP-3, the bovine gene
consists of five exons, the fifth containing the 3' untranslated region
(Fig. 2B
). These exons are separated by four introns of 2954, 474,
1049, and 1544 bp; the corresponding introns in the human IGFBP-3 gene
are 3285, 544, 1595, and 941 bp, respectively (17). We have sequenced
the gene including
4 kb of the 5' flanking region and
1 kb
beyond the polyadenylation site. The sequence information is a
composite generated from the three overlapping phage clones, bov
BP31, -2, and -3 (Fig. 2A
). A TATA box was located 162 bp 5' to the
translation initiation codon. No CAAT box was present but a GC rich
sequence element was located 5' to the TATA box.

View larger version (17K):
[in this window]
[in a new window]
|
Figure 2. A, Structure of the bovine IGFBP-3 chromosomal
gene. The three overlapping genomic clones (bov BP31, -2 and -3) are
shown on top. Below is the map of the IGFBP-3 gene: the solid
boxes are exons 15 (E1-E5), whereas the introns are shown as
lines between exons. The number at the very bottom depict the distance,
in nucleotides, 5' (negative) or 3' (positive) to the TATA box
(position 1). Restriction sites used for the Southern blot analysis are
also shown: H (HindIII), R (EcoRI), and X
(XhoI). (B) Comparison of the organization of the bovine
and human (18 ) IGFBP-3 gene. The exons are represented by boxes
(protein coding regions being solid) and their sizes in bp are shown on
top. The introns are shown as lines between the exons with their sizes
(bp) shown below.
|
|
Genomic Southern analysis
Southern blot analysis of bovine genomic DNA was performed with a
480 bp PCR product corresponding to nucleotides 516 to 996 of the
previously published bovine cDNA sequence (13). This PCR product
contains exons 2, 3, and 4 of the IGFBP-3 gene and, as shown in Fig. 3
, hybridized to two restriction fragments in each lane.
The genomic DNA in this region has single sites for the enzymes used
(Fig. 2A
). This result is consistent with the presence of a single copy
of the IGFBP-3 gene in the bovine genome.

View larger version (48K):
[in this window]
[in a new window]
|
Figure 3. Genomic Southern blot analysis. Genomic DNA,
isolated from bovine aorta endothelial cells, was digested with three
restriction enzymes, HindIII (H), EcoRI
(R) and XhoI (X). The DNA was electrophoresed on an 8%
agarose gel and transferred to a nylon membrane, then probed with a
32P-labeled PCR product that was generated from bovine
IGFBP-3 cDNA. Hybridizing fragments were visualized by autoradiography.
Each lane contains 20 µg DNA.
|
|
Determination of the transcription start site by primer
extension
A synthetic 23-bp oligonucleotide (5'-CTGGGC GGC AGC GAG CTG AGC
GA-3'), complementary to nucleotides -50 to -73 relative to the
translation initiation site of the bovine cDNA, was end-labeled with
[32P]ATP using T4 polynucleotide kinase (Promega). The
labeled oligonucleotide was hybridized to either poly (A+)
RNA from bovine endothelial cells or yeast transfer RNA at 65 C for 30
min, then slowly cooled to room temperature. After hybridization, the
primer was extended at 42 C for 30 min. The reaction products were
separated on a 6% polyacrylamide-7 M urea sequencing gel
in parallel with a sequencing reaction to determine the size of the
reaction product. As shown in Fig. 4
, an 87 base product
was generated indicating that the mRNA cap site is 137 bp 5' to the
translation initiation code. A similar result was obtained when total
RNA from cow liver was substituted for the endothelial cell poly
(A+) RNA (data not shown).

View larger version (92K):
[in this window]
[in a new window]
|
Figure 4. Primer extension. A 23 bp oligonucleotide
complementary to nucleotides -50 to -73 relative to the translation
start site was end labeled with [32P] ATP, hybridized to
20 µg yeast tRNA (lane 1) and 3 µg of poly (A)+ RNA
from bovine endothelial cells (lane 2) at 65 C for 30 min; and then
extended with AMV reverse transcriptase. The same oligonucleotide was
used to sequence pBS:3000R, a recombinant plasmid containing 1.5 kb
of 5' flanking IGFBP-3 genomic fragments. The products of the
sequencing reaction (lanes G, A, T, and C) and of the primer extension
were run on the same gel. The sequence surrounding the 5' end of the
IGFBP-3 mRNA appears on the right, the circled A being the adenosine
that serves as the cap site.
|
|
Functional characterization of the bovine IGFBP-3 promoter
To test whether the 5' flanking region of the bovine IGFBP-3 gene
exhibits promoter activity, chimeric IGFBP-3: luciferase plasmids were
used in transient transfection experiments. Upon transfection into
bovine aorta endothelial cells, pGL3:1480S, consisting of 1410 bp of
bovine IGFBP-3 5' flanking region, directed the expression of 400 times
more luciferase than did a promoterless plasmid. Experiments with
deletion constructs indicated that 50 bp of 5' flanking region
(PGL3:110S) retained some promoter activity being eight times more
potent than a promoterless vector, but only 2% of the maximal basal
promoter activity shown with pGL3:1480S (Fig. 5
). In
addition, the first 130 bp of the 5' flanking region (PGL3:180S)
retained nearly 80% of the maximal promoter activity. By contrast,
none of these constructs exhibited promoter activity when the 5'
flanking regions were inserted in the antisense orientation (only data
for PGL3:1480AS is shown in Fig. 5
).

View larger version (21K):
[in this window]
[in a new window]
|
Figure 5. Identification of promoter activity in the 5'
flanking region of the bovine IGFBP-3 gene. Different fragments of the
5' flanking region from -1410 to -50 relative to the transcription
start site were cloned 5' to a luciferase reporter gene, pGL3 Basic.
Bovine endothelial cells were transfected with the various constructs
and luciferase activity determined as described in Materials and
Methods. The data, which have been normalized for transfection
efficiency against the secreted alkaline phosphatase activity of a
cotransfected pCMV SEAP vector (Tropix), are expressed as a percentage
of the activity of the pGL3 control vector. The data are mean ±
SD of at least two experiments. A schematic of the 5'
portion of the IGFBP-3 gene and relevant restriction enzyme sites are
indicated.
|
|
A comparison of the 5' flanking region of the bovine IGFBP-3 gene with
the corresponding region of the human IGFBP-3 gene reveals the presence
of several consensus hormone response elements in corresponding
positions in both genes (Fig. 6
). These putative
response elements include TGF-ß activating sequence (TAE), GH and
steroid (ERE) response elements as well as binding sites for activating
protein-2 (AP-2), thyroid-specific transcription factors (TTF-1 and
-2), ATF and nuclear factor-1 (NF-1).

View larger version (42K):
[in this window]
[in a new window]
|
Figure 6. Nucleotide sequences of the 5'-flanking regions of
bovine (top line) and human (bottom line)
IGFBP-3 genes. The sequence is numbered relative to the transcription
start site (+1). Human sequences that are homologous to the bovine 5'
flanking region are indicated by an asterisk. Gaps have
been introduced to maximize the alignment. The TATA box and putative
hormone response elements(18 22 ) are underlined; the
transcription start site is in bold print. These
putative elements include TGF-ß1 activating sequence (TAE), GH, and
steriod (ERE) response elements as well as binding sites for activating
protein -2 (AP-2), thyroid-specific transcription factors (TTF-2), ATF,
and nuclear factor-1 (NF-1).
|
|
To determine which segments of the 5' flanking region mediated
the IGF-1 stimulation and TGF-ß1 inhibition of IGFBP-3 transcription,
bovine aorta endothelial cells were transfected with the various
constructs, followed by exposure to the growth factors. The data shown
in Fig. 7
have been normalized for transfection
efficiency against the secreted alkaline phosphatase and expressed as
percentage of the activity of the PGL3:1480S. QAYL IGF-I, an analog of
IGF-I that retains affinity for the type I IGF-receptor but has lower
affinity for IGFBPs, stimulated the promoter activity of the construct
2- to 3-fold, which is similar to the fold increase in IGFBP-3 mRNA
levels previously reported in bovine endothelial cells exposed to QAYL
IGF-I (11). It was of particular interest that pGL3:220S and pGL3:180S
were also stimulated by QAYL IGF-I because these constructs lack
TTF-I/TTF-2 (thyroid specific transcription factors) binding elements
that are required for IGF-I stimulation of thyroglobulin synthesis and
have been suggested to be involved in IGF-I regulation of IGFBP-3
transcription (18). However, QAYL IGF-1 had no significant effect on
the promoter activity of pGL3:140S, a construct which lacks the GC rich
sequence element that contains two overlapping putative AP-2 binding
sites. Finally, TGF-ß1 caused a 2-fold increase in the promoter
activity of pGL3:1480S but had no significant effect on the activity of
pGL3:7400S, pGL3:4000S, and pGL3:3000S (constructs containing 3kb or
more of the 5' flanking region). In these transient transfection
studies, we have been unable to demonstrate significant TGF-ß1
induced decrease in promoter activity, which indicates that the
TGF-ß1 mediated inhibition of IGFBP-3 transcription requires sequence
element(s) that are not present in any of the constructs tested.

View larger version (20K):
[in this window]
[in a new window]
|
Figure 7. Hormonal responsiveness of bovine IGFBP-3
luciferase constructs in bovine aorta endothelial cells. After
transfection, the cells were incubated for 18 h in serum free
media containing no growth factor (C), or containing 100 ng/ml QAYL
IGF-I (Q) or 1 nM TGFß1 (T). The cell lysates were then
assayed for luciferase activity. The data, which have been normalized
for transfection efficiency against the secreted alkaline phosphatase
activity of a cotransfected pCMV SEAP vector, are expressed as a
percentage of the activity of the pGL3:1480S vector. The data are
mean ± SD of at least two separate experiments.
|
|
 |
Discussion
|
|---|
The present study represents the first report of the cloning and
characterization of the complete bovine IGFBP-3 gene. The bovine
IGFBP-3 chromosomal gene and its promoter, like human IGFBP-3, exhibit
a simple organization, consisting of four protein coding exons and a
fifth exon containing the 3' untranslated region. AII the intron/exon
splice junctions of the bovine IGFBP-3 gene conformed to the GT/AG rule
(19) and are located at identical positions as those of the human
IGFBP-3 gene (17). Southern blot analysis indicated the presence of a
single copy gene in the bovine genome and the transcription start site
was mapped 137 bp upstream of the initiation codon by primer extension
to an adenosine residue that has been shown to be the CAP site for the
human and rat genes (17, 18). Sequence analysis of the 5' flanking
region revealed a TATA box 26 bp upstream of the CAP site. No CAAT box
was present, but a GC rich sequence element was located 5' to the TATA
box. To determine whether the putative promoter element was functional,
a 1480 bp 5' flanking segment was inserted upstream of a luciferase
reporter gene and transfected into bovine aorta endothelial cells. The
results indicated that this segment has significant orientation
dependent promoter activity as is expected for eukaryotic promoter
elements. Transient transfection experiments with deletion constructs
revealed that a promoter segment containing 130 bp of 5' flanking DNA
mediated nearly 80% of maximal promoter activity (pGL3:180S), and a
segment extending only 18 nt 5' to the TATA box (pGL3:110S) retained
some promoter function, albeit only 2% of maximal activity. The
minimal sequence element required for promoter function in the bovine
IGFBP-3 gene is similar to results obtained for human IGFBP-3 gene (17)
and the mouse IGFBP-5 promoter (20).
A number of cytokines/growth factors to which the vascular endothelium
is exposed, have been shown to play a major role in the synthesis of
IGFBP-3 by the endothelium and in the pathogenesis of diabetic
complications, such as poor wound healing, retinopathy, and nephropathy
(21). We have previously shown that two such factors, IGF-1 and
TGF-ß1 regulate the transcription of IGFBP-3 in bovine endothelial
cells (11). To understand the molecular basis for this transcriptional
regulation it was necessary to clone and sequence the bovine IGFBP-3
gene. The 5' flanking region of the bovine IGFBP-3 chromosomal gene
contains several consensus hormone response elements (Fig. 6
), similar
to results reported for the human and rat homologs(17, 18). Of
particular interest are the NF-1 and TAE elements known to mediate
TGF-ß activation of transcription (22, 23); TTF-1 and 2 binding
elements that have been suggested to be involved in IGF-1 regulation of
IGFBP-3 mRNA synthesis (18) as well as AP-2 binding elements shown to
contribute to the constitutively high expression of IGFBP-5 and its
cAMP responsiveness in human fibroblasts (24). Despite the cloning of
the human and rat IGFBP-3 promoters, very limited information is
available regarding IGFBP-3 mRNA regulation. In fact, attempts to
demonstrate hormonal responsiveness of the human IGFBP-3 promoter have
been unsuccessful (25).
As a first step towards identifying the cis-acting elements
that mediate the IGF-1 stimulation and TGF-ß1 inhibition of bovine
endothelial IGFBP-3 mRNA synthesis, we generated a number of luciferase
constructs containing 50 to 7400 bp of bovine IGFBP-3 5' flanking
region. Following transfection, bovine aorta endothelial cells were
exposed to serum free media in the absence or presence of IGF-I or
TGF-ß1. The promoter activity of almost all the constructs tested was
stimulated by QAYL IGF-I, an analog of IGF-I that has high affinity for
the type 1 IGF receptor, but lower affinity for IGFBPs. Interestingly,
plasmids pGL3:220S and pGL3:180S were also stimulated by QAYL IGF-I
despite the fact that they lack TTF-1 and -2 binding elements that have
been recently suggested to be involved in the regulation of IGFBP-3
transcription by IGF-I (18). This implies that IGF-I stimulation of
bovine endothelial IGFBP-3 mRNA synthesis requires novel
cis-acting sequence elements. Equally fascinating is the
fact that the IGF-1 stimulation of IGFBP-3 promoter activity was lost
with pGL3:140S, a construct lacking the GC rich sequence element. It
should be noted that the GC rich box has two overlapping putative AP-2
binding sites (Fig. 6
), raising the possibility that AP-2 may
contribute to the IGF-1 responsiveness of the IGFBP-3 promoter in
bovine endothelial cells. This is interesting as AP-2 has been shown to
contribute to the consitutively high expression of IGFBP-5 and its cAMP
responsiveness in human fibroblasts (24). TGF-ß1 stimulated the
promoter activity of constructs containing 1.5 kb or less of IGFBP-3 5'
flanking region, an effect that may be mediated by the TGF-ß1
activating sequences present in this region (TAE, NF-1 Fig. 6
). On the
other hand, TGF-ß1 had no effect on the promoter activity of
constructs containing 3kb or more of 5' flanking region. Although the
ability of TGF-ß1 to stimulate promoter activity was lost in
constructs containing 3 kb or more of 5' flanking region, no
significant decrease below basal promoter activity was seen in any of
the constructs. Thus, the TGF-ß mediated inhibition of IGFBP-3
transcription previously reported in cultured endothelial cells(11),
likely requires additional sequence elements(s) that are not present in
any of the constructs tested thus far.
In summary, we have cloned the bovine IGFBP-3 chromosomal gene and
demonstrated both basal and hormonally responsive promoter activity in
its 5' flanking region. Further studies are required to identify and
characterize endothelial specific cis and
trans-acting element(s)/factor(s) involved in the
regulation of IGFBP-3 transcription. The data presented in this report
will provide a basis for such future experiments.
Received November 1, 1996.
 |
References
|
|---|
-
Bach LA, Rechler MM 1995 Insulin-like growth
factor binding proteins. Diabetes Rev 3:3861
-
Baxter RC, Martin JL 1989 Binding proteins for the
insulin-like growth factors: structure, regulation and function. Prog
Growth Factor Res 1:4968[CrossRef][Medline]
-
Jones JI, Clemmons DR 1995 Insulin-like growth
factors and their binding proteins: biological actions. Endocr Rev 16:334[CrossRef][Medline]
-
Shimasaki S, Ling N 1991 Identification and
molecular characterization of insulin-like growth factor binding
proteins (IGFBP-1, -2, -3, -4, -5 and -6). Prog Growth Factor Res 3:253266
-
Oh Y, Muller H, Lamson G, Rosenfeld R 1993 Insulin-like growth factor (IGF)- independent action of IGF-binding
protein-3 in Hs578T human breast cancer cells. J Biol Chem 268:1496414971[Abstract/Free Full Text]
-
Baxter RC 1993 Circulating binding proteins for
the insulin-like growth factors. Trends Endocrinol Metab 4:9196[Medline]
-
Moser DR, Lowe Jr WL, Dake BL, Booth BA, Boes M,
Clemmons DR, Bar RS 1992 Endothelial cells express insulin-like
growth factor binding proteins 2 to 6. Mol Endocrinol 6:18051814[Abstract]
-
Zhou J, Bondy CB 1992 Insulin-like growth
factor-II and its binding proteins in placental development.
Endocrinology 131:12301240[Abstract]
-
Zhou J, Bondy CB 1993 Anatomy of the insulin-like
growth factor system in the human testis. Fertil Steril 60:897904[Medline]
-
Chin E, Zhou J, Dai J, Baxter RC, Bondy CA 1994 Cellular localization and regulation of gene expression for components
of the insulin-like growth factor ternary binding protein complex.
Endocrinology 134:24982504[Abstract]
-
Erondu NE, Dake BL, Moser DR, Lin M, Boes M, Bar RS 1996 Regulation of endothelial IGFBP-3 synthesis and secretion by IGF-I
and TGF-ß. Growth Regul 6:19[Medline]
-
Ausubel FM, Brent R, Kingston RE, Moore DD, Smith JA,
Seidman JG, Struhl K 1988 Current Protocols in Molecular Biology.
Greene/Wiley Interscience, New York
-
Spratt SK, Tatsuno GP, Sommer A 1991 Cloning and
characterization of bovine insulin-like growth factor binding
protein-3. Biochem Biophys Res Commun 177:10251032[CrossRef][Medline]
-
Erondu NE, Donelson JE 1992 Differential expression
of two mRNAs from a single gene encoding an HMG1-like DNA binding
protein of African trypanosomes. Mol Biochem Parasitol 51:111118[CrossRef][Medline]
-
Bradford MM 1976 A rapid and sensitive method for
the quantification of microgram quantities of protein using the
principle of protein-dye binding. Anal Biochem 72:248254[CrossRef][Medline]
-
Sachs A, Wahle E 1993 Poly(A) tail metabolism and
function in eukaryotes. J Biol Chem 268:2295522958[Free Full Text]
-
Cubbage ML, Suwanichkul A, Powell DR 1990 Insulin-like growth factor binding protein-3. Organization of the human
chromosomal gene and demonstration of promoter activity. J Biol
Chem 265:1264212649[Abstract/Free Full Text]
-
Albiston AL, Saffery R, Herington AC 1995 Cloning
and characterization of the promoter for the rat insulin-like growth
factor-binding protein-3 gene. Endocrinology 136:696704[Abstract]
-
Padgett RA, Grabowski PJ, Maria MM, Seiler S, Sharp
PA 1986 Splicing of messenger RNA precursors. Annu Rev Biochem 55:11191150[CrossRef][Medline]
-
Kou K, Mittanck DW, Fu C, Rotwein P 1995 Structure
and function of the mouse insulin-like growth factor binding protein 5
gene promoter and DNA. Cell Biol 14:241249
-
Pankewycz OG, Guan J-X, Benedict JF 1995 Cytokines
as mediators of autoimmune diabetes and diabetic complications. Endocr
Rev 16:164176[CrossRef][Medline]
-
Faisst S, Meyer S 1992 Compilation of
vertebrate-encoded transcription factors. Nucleic Acids Res 20:326[Free Full Text]
-
Ritzenthaler JD, Goldstein RH, Fine A, Smith BD 1993 Regulation of the
1(I) collagen promoter via a transforming
growth factor-ß activation element. J Biol Chem 268:1263513631
-
Duan C, Clemmons R 1995 Transcription factor AP-2
regulates human insulin-like growth factor binding protein-5 gene
expression. J Biol Chem 42:2484424851
-
Powell DR, Suwanichkul A 1994 Role of IGFBP-1 and
IGFBP-3 promoter regions in regulating IGFBP expression. Growth Regul
[Suppl 1] 4:63
This article has been cited by other articles:

|
 |

|
 |
 
S Llewellyn, R Fitzpatrick, D A Kenny, J J Murphy, R J Scaramuzzi, and D C Wathes
Effect of negative energy balance on the insulin-like growth factor system in pre-recruitment ovarian follicles of post partum dairy cows
Reproduction,
March 1, 2007;
133(3):
627 - 639.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
E. M. Ongeri, M. F. Verderame, and J. M. Hammond
Follicle-Stimulating Hormone Induction of Ovarian Insulin-Like Growth Factor-Binding Protein-3 Transcription Requires a TATA Box-Binding Protein and the Protein Kinase A and Phosphatidylinositol-3 Kinase Pathways
Mol. Endocrinol.,
July 1, 2005;
19(7):
1837 - 1848.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
E. M. Ongeri, Q. Zhu, M. F. Verderame, and J. M. Hammond
Insulin-Like Growth Factor-Binding Protein-3 in Porcine Ovarian Granulosa Cells: Gene Cloning, Promoter Mapping, and Follicle-Stimulating Hormone Regulation
Endocrinology,
April 1, 2004;
145(4):
1776 - 1785.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
J. Tsubaki, V. Hwa, S. M. Twigg, and R. G. Rosenfeld
Differential Activation of the IGF Binding Protein-3 Promoter by Butyrate in Prostate Cancer Cells
Endocrinology,
May 1, 2002;
143(5):
1778 - 1788.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
C. Deal, J. Ma, F. Wilkin, J. Paquette, F. Rozen, B. Ge, T. Hudson, M. Stampfer, and M. Pollak
Novel Promoter Polymorphism in Insulin-Like Growth Factor-Binding Protein-3: Correlation with Serum Levels and Interaction with Known Regulators
J. Clin. Endocrinol. Metab.,
March 1, 2001;
86(3):
1274 - 1280.
[Abstract]
[Full Text]
|
 |
|

|
 |

|
 |
 
W. S. Cohick, B. Wang, P. Verma, and Y. R. Boisclair
Insulin-Like Growth Factor I (IGF-I) and Cyclic Adenosine 3',5'-Monophosphate Regulate IGF-Binding Protein-3 Gene Expression by Transcriptional and Posttranscriptional Mechanisms in Mammary Epithelial Cells
Endocrinology,
December 1, 2000;
141(12):
4583 - 4591.
[Abstract]
[Full Text]
[PDF]
|
 |
|

|
 |

|
 |
 
N. E. Erondu, J. Nwankwo, Y. Zhong, M. Boes, B. Dake, and R. S. Bar
Transcriptional and Posttranscriptional Regulation of Insulin-Like Growth Factor Binding Protein-3 by Cyclic Adenosine 3',5'-Monophosphate: Messenger RNA Stabilization Is Accompanied by Decreased Binding of a 42-kDa Protein to a Uridine-Rich Domain in the 3'-Untranslated Region
Mol. Endocrinol.,
March 1, 1999;
13(3):
495 - 504.
[Abstract]
[Full Text]
|
 |
|