|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 275, Issue 35, 27421-27438, September 1, 2000
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the
Received for publication, March 14, 2000, and in revised form, May 30, 2000
To identify control motifs involved in human type
II collagen gene transcription in both differentiated and
dedifferentiated rabbit articular chondrocytes, transient transfection
experiments were performed. A 715-base pair (bp) region of the
first intron (+2127/+2842), including a 153-bp sequence so far
uncharacterized (+2689/+2842), was found to mediate enhancer activity.
In dedifferentiated chondrocytes, this enhancer activity was shown to
be less effective than in primary cultures but still present. We then
demonstrated that a zinc finger protein, C-Krox, activates COL2A1 gene
transcription in differentiated chondrocytes through the enhancer
region, whereas in subcultured cells, it inhibited the gene activity
via a 266-bp promoter. Multicopies of the C-Krox binding site were
found to mediate transactivation in both primary cultures and passaged cells, whereas C-Krox overexpression inhibited transcription in dedifferentiated chondrocytes. Additionally, we showed that C-Krox binds to several cis sequences that mediate its
transcriptional effects. During chondrocyte dedifferentiation, the
protein levels and binding activity of C-Krox were reduced, whereas
those of NF- Articular cartilage contains an extracellular matrix that consists
of tissue-specific macromolecules including type II, IX, and XI
collagens and the large aggregating proteoglycan aggrecan (1). The
collagens form an insoluble fibrous network that is responsible for the
tissue tensile strength (2). These extracellular matrix components are
synthesized by chondrocytes, which are highly differentiated cells
responsible for the maintenance of a regulated balance between the
anabolism and the catabolism of these cartilage-specific macromolecules. During cartilage development, differentiation, and
repair, the highly conserved expression of the set of tissue-specific genes by chondrocytes is tightly regulated.
When chondrocytes are isolated from cartilage tissue and cultured as
monolayers, they lose their cartilage phenotype and transform into
flattened fibroblast-like upon repeated passages (3-7). They gradually
produce interstitial collagens (types I, III, and V) in addition to
type II collagen and the fibroblast-type proteoglycan (versican) at the
expense of aggrecan (3-9). The chondrocyte-specific phenotype can be
recovered when these cells are transferred to three-dimensional
matrices, such as agarose or alginate gels (7, 8). There is also
evidence that the osteoarthritic process is accompanied by a phenotypic
modulation of the chondrocytes in the altered cartilages (10-12). In
the degraded matrix, the resident cells are under the influence of
inflammatory cytokines, growth factors, and matrix molecules or
fragments that are not normally present in their microenvironment. As a
consequence, the chondrocytes respond to these signals by a progressive
loss of their specific phenotype. It is therefore crucial to
investigate the mechanisms that regulate the transcriptional activity
of the cartilage-specific type II procollagen gene (COL2A1) in adult chondrocytes. This could give clues for understanding the mechanism of
the shift of collagen type II expression to the benefit of collagens I
and III during the chondrocyte phenotype loss and provide potential
means to stabilize these cells in their functional activity despite the
stress affecting cartilage.
Although type II collagen plays a crucial role in the maintenance of
the structure and function of articular cartilage matrix, only a few
structural and functional analyses of the regulatory regions of human
COL2A1 gene have been reported. However, it has been demonstrated that
an enhancer sequence, present in the first intron of the rat, mouse,
and human genes, is responsible for the cell-specific expression of the
type II collagen in chondrocytes (13-22). More recently, a protein
complex called CSEP (chondrocyte-specific enhancer-binding proteins), in which the
transcription factor SOX9 has been identified as a regulator of
chondrocyte lineage, has been shown to be responsible for the high
level expression of COL2A1 gene in chondrocytes (21, 23). Other
transcription factors, including L-SOX5, SOX6, and proteins not yet
characterized, are likely to play a role in the function of this
complex (21). Nevertheless, SOX9 was demonstrated to be the first
transcription factor to specify the chondrocytic lineage and to be
absolutely required for complete cartilage differentiation (23). This
high mobility group protein, being expressed in different
tissues such as the central nervous system and urogenital apparatus, is
also detected during mouse embryonic development in all cartilage
primordia where it colocalizes with type II collagen expression.
SOX9 Moreover, an Sp1 binding activity has been shown to be involved in the
cartilage-specific expression of the gene, because binding appears to
be 2-3-fold greater in dedifferentiated chondrocytes than in freshly
isolated cells (24). Furthermore, a protein complex including Sp1 could
bind to both the COL2A1 promoter and enhancer, suggesting the formation
of a DNA loop structure between the COL2A1 promoter and enhancer that
is mediated by nuclear proteins (25).
However, the changes in DNA-binding proteins that may occur during the
process of phenotype modulation in chondrocytes have not been
investigated in details. Here, we studied the changes in the levels and
activity of the transcriptional factor C-Krox, a zinc finger protein
already known to take part in collagen I expression (26, 27), during
the process of chondrocyte dedifferentiation.
Cell Cultures--
RAC1 were released by
enzymatic digestion of articular cartilage slices from the shoulders
and the knees of 3-week-old rabbits, as described previously (6, 28).
Cells were seeded at 2 × 104 cells/cm2 in
either 100-mm dishes or 75- and 175-cm2 flasks and cultured
in Dulbecco's modified Eagle's medium supplemented with 10%
heat-inactivated fetal calf serum (FCS),
glutamine (2 mM), penicillin (100 IU/ml), streptomycin (100 µg/ml), and fungizone (0.25 µg/ml) in a 5% CO2
environment and fed twice a week. For dedifferentiation, RAC cultures
were passaged with trypsin (0.25%) after reaching confluency.
Human articular cartilage was obtained from two osteoarthritic patients
undergoing total knee replacement surgery. The articular cartilage was
taken from a zone without signs of alteration, comprising the uncovered
lateral tibial plateaus and the uncovered lateral femoral condyle.
Macroscopic data were validated by histologic examination of small,
full thickness specimens removed from comparative samples. They were
fixed with 10% formalin, decalcified, and embedded in paraffin.
Sections were stained with hematoxylin-eosin-safran (data not shown).
Slices of normal cartilage were cut aseptically and kept in Earle's
balanced salt solution. Chondrocytes were released by enzymatic
treatment at 37 °C: protease type XIV from Sigma (4 mg/ml) for
1.5 h and collagenase type I (1 mg/ml, Sigma) overnight. Isolated
chondrocytes were seeded at high density (3 × 106
cells/25-cm2 flask) in 10% FCS-containing Dulbecco's
modified Eagle's medium to prevent any cell dedifferentiation. They
were grown as primary monolayers at 37 °C in a 5% CO2
atmosphere, with medium change every 2-3 days. At about 80% of
confluency, they were used for nuclear extract mini-preparation.
Human dermal fibroblasts from explanted skin biopsies of healthy adults
or from infant foreskins, as well as NIH-3T3 cells, were cultured in
Dulbecco's modified Eagle's medium supplemented with antibiotics and
10% heat-inactivated FCS and used for the fourth to eighth passages.
Rabbit fibroblasts were obtained from small pieces of dermis digested
for 30 min by collagenase type I at 2 mg/ml (Sigma) in Dulbecco's
modified Eagle's medium +10% FCS, followed by an overnight incubation
in diluted collagenase (0.3 mg/ml). They were used between two to five passages.
HCS 2/8 cell line derives from a human chondrosarcoma of the proximal
part of the humeral bone of a 72-year-old Japanese male who did dot
received any surgical treatment nor chemical or radiation therapy (29).
The cells were cultured as above in 10% FCS-containing medium. They
were passaged by trypsinization, at a dilution of 1/2 to
1/4, and the medium was changed twice a week (29).
Transfection Experiments--
Chondrocytes seeded at a density
of 1 × 106 cells/100-mm culture plates were
transiently transfected by the calcium phosphate precipitation method
using HEPES at 80% of confluency (30). The reporter plasmids
(generally 9 µg each) were cotransfected with a pSV-40
For decoy experiments, primary RAC were transfected as above with 15 µg of reporter plasmids together with DNA Constructions--
A SalI-SphI DNA
fragment of ~8 kb of the human COL2A1 gene has been cloned in pUC19
(32). This clone contains approximately 1.8 kb of the promoter
sequence, exon 1, intron 1, exon 2, and part of intron 2 (
A first reporter plasmid containing 3.774 kb of COL2A1 gene sequence,
from
In the following reporter plasmids, upstream (US) and downstream (DS)
refer to the position of the intronic insert subcloned according to the
transcription initiation site, whereas (+) and (
Other constructs have been generated with the shorter pGL2-0.387 kb
plasmid. A 121-bp SmaI-BglII insert (+2127/+2248)
that was filled in has been subcloned in both orientations in the
SmaI site of pGL2-0.387 kb to give pGL2-0.387kbE121(+)US
and pGL2-0.387kbE121(
Three other types of reporter plasmids were used; one is a minimal
promoter that contains a 86-bp mouse COL1A1 promoter cloned upstream of
the luciferase gene. Upstream to this nonspecific promoter, four and
five wild-type A Generation of Recombinant C-Krox-Histidine Fusion
Proteins--
Full-length C-Krox and a deleted form of this protein
were expressed in Escherichia coli as fusion proteins with
an histidine tag (His-tagged C-Krox) (33), using the procedure
previously described (26, 27).
Nuclear Extracts--
Nuclear extracts were prepared with
buffers containing the following protease inhibitors (0.5 mM dithiothreitol, 0.5 mM
phenylmethylsulfonylfluoride, 10 µg/ml leupeptin, and 10 µg/ml
pepstatin) by the maxipreparation method of Dignam et al.
(34), or when mentioned, by the minipreparation procedure of Andrews
and Faller (35). They were resuspended in a binding buffer containing
20 mM Tris-HCl, pH 7.5, 1 mM dithiothreitol, 100 mM NaCl, 0.05 mg/ml bovine serum albumin (BSA), 0.1%
Nonidet P-40, 10% glycerol, and 25 µM
ZnCl2.
Gel Retardation Assays--
Gel retardation assays were
performed with the oligonucleotides shown in Table I, which were
synthesized on an Oligo 1000 DNA synthesizer (Beckman). They were
end-labeled with [ DNase I Footprinting--
The plasmid pGL2-3.774 kb
(
DNA binding reactions were performed in a final volume of 50 µl,
containing ~100 fmol of DNA probes in the same binding buffer used
for gel retardation assays. 4 µg of poly(dI·dC)·poly(dI·dC) were added in the reaction when nuclear extracts were used, whereas they were omitted in the case of recombinant histidine-C-Krox protein.
Two different amounts of nuclear extracts (30 and 45 µg) or
histidine-C-Krox protein (~1.5 nmol) were added in the binding
reaction and incubated for 15 min at room temperature. The DNase I
(Life Technologies, Inc.) has been diluted just before use in a cold
buffer containing 10 mM Tris-HCl, pH 7.5, 50 mM NaCl, 62.5 mM MgCl2, 1 mM
CaCl2, and 1 mg/ml BSA. 2 µl of DNase I solution (diluted
or not) have been added, and the following steps were performed as
described previously (27). The radiolabeled probes were submitted to
the A+G Maxam-Gilbert sequencing reaction (36), and the samples were
fractionated on the same gel to identify sequences protected by C-Krox
protein and other transcription factors present in the nuclear
extracts. After electrophoresis, the gel was submitted to
autoradiography at RNA Analysis--
Total RNA was isolated by the method of
Chomczynski and Sacchi (37), with an additional precipitation step in 6 M LiCl (Sigma) to remove any residual trace of genomic DNA.
Poly(A)+ RNA was extracted using the Dynabeads oligodT kit
(Dynal, France). Total and poly(A)+ RNAs were fractionated
by electrophoresis on 1% agarose-MOPS-formaldehyde gel and transferred
to a nylon membrane (Biodyne A, Pall Corp.). To facilitate the transfer
of large RNA species to the membrane, agarose gels were treated for 15 min with 50 mM NaOH prior to blotting. RNA were fixed on
the membrane by baking at 80 °C for 15 min and UV exposure.
Prehybridization (5 h) and hybridization (15 h) were performed at
55 °C in a buffer containing 1 M
NaH2PO4, pH 7.5, 20% SDS, and water
( Western Blotting--
Nuclear extracts (15-20 µg) and
His-C-Krox fusion protein (100 ng) were run on a polyacrylamide gel in
denaturing conditions (Tris-glycine buffer containing 1% SDS). The gel
was then equilibrated for 30 min in the transfer buffer (25 mM Tris-HCl, pH 8.3, 192 mM glycine, 20%
methanol (v/v)), and proteins were electrotransferred for 1 h at
100 V on a polyvinylidene difluoride nylon membrane. After transfer,
free protein binding sites were blocked by incubating the filters for
1 h in the TBST (Tris-buffered saline with Tween 20) buffer
containing 10% nonfat carnation milk. Then, the membrane was rinsed
six times (5 min) in TBST and incubated for 1 h with the rabbit
preimmune serum diluted in TBST containing BSA at 2 mg/ml (1:250 for
nuclear extracts and 1:2000 for the recombinant His-C-Krox protein). A
membrane containing the same samples was incubated in parallel with an
anti-C-Krox antibody at the same dilutions as the preimmune serum.
After washing with TBST, filters were incubated during 1 h in the
presence of a secondary antibody (rabbit anti-IgG) coupled to the
alkaline phosphatase (1:10000 in TBST + BSA at 2 mg/ml). The membranes
were then reacted with the substrates for the alkaline phosphatase
during 2-3 min (4-nitro blue tetrazolium, and
5-bromo-4-chloro-3-indolyl-phosphate (Roche Molecular
Biochemicals)). In one experiment, C-Krox protein expression was
revealed with a horseradish peroxidase-labeled secondary antibody (anti-rabbit IgG) (1:1000) using an ECL+Plus Western blot detection kit
(Amersham Pharmacia Biotech).
Deletion Analysis of a COL2A1 First Intron Region That Induces
Enhancer-mediated Transcription--
To delineate a putative specific
enhancer in the human COL2A1 gene, we generated different reporter
vectors containing deletions in both promoter and first intron regions.
These constructs (Fig. 1A)
were transiently transfected in primary RAC. We found that the longest
construct, spanning
To determine whether the activity of the COL2A1 regulatory sequences
was chondrocyte-specific, plasmids containing the promoter region with
or without the enhancer regions were transiently transfected in rabbit
dermal fibroblasts as control cells that do not produce type II
collagen. None of the human type II collagen gene constructs was
actually expressed in fibroblasts (Fig. 1B). The largest and most active plasmid in chondrocytes did not show significant activity (only 77 RLU) in that case. Deletion constructs containing 1.7-, 0.544-, and 0.387-kb sequences of the COL2A1 promoter region as well as
reporter plasmids presenting deletions in the first intron region
resulted in a similar weak activity. Similar data were obtained with
human dermal and mouse NIH-3T3 fibroblasts. We therefore concluded that
promoter and enhancer sequences of the COL2A1 gene are required for
high expression level in chondrocytes and that these elements do not
display any activity in cells that do not produce type II collagen.
Delineation and Functional Analysis of the Enhancer Region Present
in the First Intron of Human COL2A1 Gene--
To delineate more
precisely the region that mediates high transcription level of human
COL2A1 gene in chondrocytes, several 5' and 3' deletion fragments from
the first intron of that gene spanning the region +2127/+2842 were
isolated and subcloned upstream or downstream of two plasmids
containing 544 and 266 bp of 5'-flanking promoter sequences, localized
upstream of the luciferase reporter gene in the pGL2-basic vector.
Primary cultures of RAC were transiently transfected with each of the
human COL2A1 constructs (Fig.
2A). The transcription
activity of the 544- and 266-bp promoters represents approximately 5%
of that observed for the largest and most active construct, pGL2-3.774
kb. The 153-bp BamHI-NaeI (+2689/+2842), 458-bp
XhoI-NaeI (+2384/+2842), and 715-bp
SacI-NaeI (+2127/+2842) inserts of the first
intron, cloned upstream and/or downstream and placed in either
orientations in the 544 bp promoter, induced an increase in the
transcriptional activity of the basal promoter to levels of 412-1150%
versus the pGL2-1.167 kb control vector. However, this
overall transcriptional activation of intronic sequences remained lower
than the transcription activity observed with the longest construct
pGL2-3.774 kb.
When the shorter 266-bp promoter was used, cloning of the 153-, 458-, and 715-bp fragments of the first intron produced a dramatic
increase of expression, reaching 110-380-fold of the control
(pGL2-0.387 kb construct), and this transcription activity was even
higher when compared with that of pGL2-3.774 kb.
In addition, we found differences in transcriptional activity according
to the size of the promoter used. The activities of constructs
containing the 5' deletions of the intronic region were higher when
cloned together with the short 266-bp promoter, as compared with the
same intronic inserts cloned upstream or downstream the 544-bp
promoter. The drop of transcription level for constructs containing
544-bp promoter sequences may be due to the presence of a silencing
activity in the human COL2A1 gene, as was described in the rat COL2A1
gene (40). In fact, this 544-bp promoter includes the CIIS2 putative
silencer element present in the rat sequence (respectively
From the use of reverse-orientated constructs, we can conclude that the
increase in promoter activity induced by the first intron DNA fragments
was partially orientation-independent. The greatest variation observed
with the pGL2-0.387 kb+E715US when using both orientations was not
significant (only a 3-fold difference). However, this small difference
is relative because it should be compared with the pGL2-0.387 kb
control vector, for which we observed 113- and 382-fold increases by
the 715-bp SacI/NaeI fragment for the correct and
opposite orientations, respectively (Fig. 2A). The results
also demonstrate that a part of the chondrocyte-specific enhancer is
likely to reside in the 153-bp BamHI-NaeI
fragment, because all the intronic sequences mediating enhancer effect
included this region.
After characterization of the 3' end of the enhancer, we generated new
constructs containing 3' deletions in the first intron region spanning
+2127/+2842. These intron fragments were cloned 5' or 3' to the 266-bp
short promoter in correct or opposite orientation (Fig. 2B).
The results demonstrated that the 121-bp
SacI-BglII (+2127/+2248), the 257-bp
SacI-XhoI (+2127/+2384), and the 562-bp SacI-BamHI (+2127/+2689) intronic sequences were
also capable of increasing the transcription of a COL2A1 basal promoter
to levels observed with the largest pGL2-3.774 kb construct, the most
active in primary RAC cultures. Nevertheless, the highest transcriptional activity was observed when the 562-bp DNA fragment was
cloned upstream the short promoter, reaching a 237-fold increase compared with the pGL2-0.387 kb construct. This indicates that activating sequences are localized between +2127 and +2384 bp, although
the region spanning +2384/+2689 displays the greatest activity.
From the data shown in Fig. 2, it can be concluded that the +2127/+2842
region of the first intron of human COL2A1 gene contains activating
sequences. Moreover, two of these regions mediated a much greater
activation of COL2A1 gene expression: +2384 to +2689 bp and +2689 to
+2842 bp. Therefore, the region encompassing +2384/+2842 can be
proposed as the more potent human COL2A1-specific enhancer, but some
transactivating factors also probably bind to the +851/+2127 sequences.
Effect of Chondrocyte Subculture on the Transcription of the Human
COL2A1 Gene--
To investigate the regulation of the COL2A1 gene
expression during the dedifferentiation process of chondrocytes, we
used the subculture-induced modulation of phenotype, generally
reflected by the synthesis of type I collagen in addition to type II
(28). We found that the highest COL2A1 transcriptional expression in chondrocytes at passage 2 was observed with the pGL2-3.774 kb construct, demonstrating that part of the enhancing activity is still
present, albeit less effective, in phenotype-modulated RAC (Fig.
3). Contrasting with primary RAC, in the
dedifferentiated cells, the putative silencer sequences of the 544-bp
promoter prevented higher transcriptional activity induced by the first intron enhancer region since cloning of the 153-, 458-, and 715-bp first intron DNA fragments, downstream from this promoter, failed to
activate transcription. In addition, when these sequences were cloned
together with the 266-bp promoter in the luciferase reporter vector,
they induced transcription levels similar to those of the pGL2-3.774
kb control vector (8.5-51-fold compared with the 266-bp basal
promoter).
These data also confirm that the enhancer can be fragmented in two
segments, one spanning +2689/+2842 and another one encompassing the
+2384 to +2689 sequence, as is the case for primary RAC. It is also
obvious that dedifferentiated RAC do not behave as fibroblastic cells,
although they progressively acquire a fibroblast-like morphology, because they still possess functional regulatory sequences of COL2A1
gene and enhancer activity. The main difference observed in
dedifferentiated RAC was that the putative silencer elements present in
the 544-bp promoter were unequivocally much more potent than in fully
differentiated cells and prevented transcriptional activation induced
by the first intron. This suggests that phenotype-modulated RAC have
presumably acquired some "fibroblast-specific" silencer factors.
Detection of C-Krox Transcription Factor in Chondrocytes--
To
study the potential involvement of C-Krox in the transcriptional
regulation of human COL2A1 gene, we first searched for its presence in
nuclear extracts from human articular chondrocytes (HAC) and dermal
fibroblasts, by Western blot analysis, using a C-Krox polyclonal
antibody (Fig. 4A). Three weak
polypeptide bands were detected in both cell types, migrating in the
region of 36-64 kDa. A major polypeptide species running more slowly, was found in similar amounts in both HAC and fibroblasts (Fig. 4A, arrow). This polypeptide is of the expected
size, as revealed with the recombinant His-tagged C-Krox protein (Fig.
4B). A major band migrating to the same position was
observed when the full-length C-Krox cDNA was in vitro
transcribed and translated in a rabbit reticulocyte lysate system (data
not shown). From these results, we can conclude that C-Krox is present
in HAC and dermal fibroblast nuclear extracts to comparable
levels. Presence of the C-Krox message in chondrocytes was then
confirmed by Northern blot analysis. A transcript of the expected
3.5-kb size was identified when poly(A)+ RNA was hybridized
with a specific cDNA probe that only recognizes C-Krox among the
Krox members (Fig. 4C).
Having proved that C-Krox was expressed in chondrocytes, we performed
gel shift experiments with the A C-Krox Differentially Modulates Human COL2A1 Gene Transcription in
Differentiated and Dedifferentiated RAC--
To provide direct
evidence of the role of C-Krox in human COL2A1 gene transcription,
cotransfection experiments were performed with primary and passaged
RAC. Constructs spanning different regions of the promoter and first
intron of that gene were cotransfected with either the C-Krox
expression vector pEMSV(+) or the insertless plasmid pEMSV (Fig.
5A). Overexpression of C-Krox
induced a 3-fold increase in luciferase activity of the pGL2-3.774 kb
construct, indicating that C-Krox acts as an activator of COL2A1 gene
transcription in primary RAC. This was then confirmed in several
experiments, using primary RAC from different rabbits, with activities
ranging from 1.9- to 4.3-fold increase in transcription activity under C-Krox overexpression (Fig. 5C).
Deletion of the +2384/+2842 region of the first intron containing the
specific enhancer abolished the C-Krox effect on transcriptional activity, as demonstrated by the data obtained with the pGL2-3.316 kb.
The constructs presenting various deletions in the promoter region
remained apparently unaffected by overexpressing C-Krox. Thus, it can
be suggested that C-Krox activates transcription of type II collagen
gene through the specific +2384/+2842 enhancer region. However, cloning
of whole or part of the specific enhancer region upstream or downstream
a COL2A1 promoter of 544 or 266 bp failed to restore the C-Krox
activating effect on transcription. This strongly suggests that the
absence of effect may be due to differences in the position of the
intronic sequences in the constructs used here, which were not located
in their natural position, at the same distance of the transcription
start site. An alternative could be that the amounts of endogenous
C-Krox may be sufficient to already induce saturation of its
DNA-binding sites in COL2A1 promoter, leading to transcriptional activation.
In phenotype-altered RAC, C-Krox was acting as an inhibitor of
transcription as shown in Fig. 5A. All the constructs tested showed a reduced luciferase activity under C-Krox overexpression, suggesting that the factor exerts its transcriptional effect through the 266-bp region immediately located 5' to the start site of transcription. To check that C-Krox overexpression was actually occurring in the transfected chondrocytes and that the effects observed
were not artifactual, mini-nuclear extracts from primary and passaged
RAC that have been cotransfected with different amounts of the C-Krox
expression vector were analyzed for their C-Krox levels. As shown in
Fig. 5B, C-Krox expression was enhanced when increasing
amounts of the pEMSV(+) plasmid were transfected, being already
detectable with 10 µg of vector. When nuclear extracts from
dedifferentiated RAC (two passages) were incubated with the labeled
A
We then asked whether overexpression of C-Krox in primary and passaged
RAC did affect in the same manner the intracellular steady-state levels
of COL2A1 mRNA. C-Krox overexpression was shown to induce,
approximately, a 2-fold increase in COL2A1 mRNA levels in primary
RAC, after correction to
Thus, we concluded that C-Krox modulates the transcription of COL2A1
gene in chondrocytes as a function of their phenotype expression
because it activates COL2A1 gene transcription in differentiated RAC
through the specific enhancer region of the first intron, whereas it
inhibits transcription of that gene via the C-Krox Binding Activity Is Decreased during the RAC Phenotype
Loss--
We then asked whether C-Krox binding activity and expression
were affected by the in vitro phenotype modulation of RAC.
Gel retardation experiments, with nuclear extracts from primary RAC and
fibroblasts showed that binding was seen on two cis elements (Fig. 6) that were identified in
footprint analysis (see Fig. 8). These cis-binding regions
cover the sequences
The data raised the possibility that the dedifferentiation process of
RAC could be associated with a decreased DNA binding of transcription
factors in general. To test this hypothesis, we performed gel shift
experiments, using a NF- C-Krox Protein Expression Is Reduced in Phenotype-modulated
RAC--
Then we tested the hypothesis that decrease of C-Krox binding
activity could be associated with a reduced C-Krox gene transcription during the phenotype alteration of RAC. As seen in Fig. 6D,
C-Krox mRNA amounts were not significantly altered during phenotype
modulation of RAC because only 10-20% reduction of this mRNA
level was observed after three and five culture passages. Such
variations of C-Krox mRNA levels could not account for the
significant effects observed on C-Krox DNA binding activity during
phenotype alteration of RAC. The decrease of C-Krox binding to its
cis element accompanying the dedifferentiation of RAC is
likely to involve other mechanisms than transcription rate of the
factor. However, Western blot analysis of primary and dedifferentiated
RAC nuclear extracts with an anti-C-Krox antibody demonstrated that
C-Krox amount was decreased after one and two culture passages (Fig.
6E). So, a part of the decrease in C-Krox binding activity
during RAC dedifferentiation may result from reduced C-Krox protein synthesis.
C-Krox Specifically Binds to Several cis-acting Elements in Both
Promoter and First Intron Regions of COL2A1 Gene--
To further
investigate the role of C-Krox in human COL2A1 gene transcriptional
regulation, we carried out DNase I footprint experiments with both the
promoter and first intron regions. We first looked at the specific
enhancer region. Nuclear extracts from primary and dedifferentiated
RAC, HCS 2/8 cells, and NIH-3T3 fibroblasts were used, as well as
His-tagged C-Krox protein. A strong footprint was induced by C-Krox
when we used the +2689/+2842 fragment of the COL2A1 gene (Fig.
7A, lane 11). The
protected area is located between +2817 and +2845 and is centered by a
GC-rich sequence, as indicated on Table I (+2817+2845
Using the +2384/+2689
DNase I footprints were also performed on the promoter region mediating
the C-Krox inhibition of COL2A1 gene transcription in dedifferentiated
RAC. When we used the
To test the specificity of C-Krox binding, we performed gel retardation
assays, using the oligonucleotides identified in footprint analyses as
probes. Point mutations were introduced in these cis elements (Table I), as described previously (27). In direct binding
experiments, recombinant C-Krox bound to all of the wild-type oligonucleotides (Fig. 8) (data not shown
for the
Specificity of C-Krox binding was also studied by competition
experiments in gel retardation assay. As shown in Fig.
9, C-Krox bound to the labeled
C-Krox protein also bound to the +2440+2485
As shown in Fig. 11, His-C-Krox binds
also to the wild-type oligonucleotides C-Krox Consensus Binding Site Activates Transcription of a
Nonspecific Promoter in RAC, whereas C-Krox Overexpression Reduced
Transcription of This Promoter in Dedifferentiated RAC--
The
function of a consensus C-Krox binding site was studied by transient
transfection experiments. Reporter vectors containing four (AP35) or
five (AR5) copies of the wild-type A
In dedifferentiated RAC, where C-Krox binding activity and amounts are
greatly reduced, the transcription of AP35 and AR5 reporter constructs
was always increased approximately to the same extent when compared
either with the mutated A'N°3 plasmid or with the effects observed in
primary RAC (Fig. 12). This indicates that a putative residual
C-Krox-like activity is still present in dedifferentiated RAC and that
this activity was undetectable in gel retardation assays, certainly
because of the sensitivity of the method. In these phenotype-modulated
RAC, cotransfection with an expression vector containing C-Krox
cDNA resulted in a decreased transcriptional activity (2-3 times)
of the two reporter vectors AP35 and AR5 (Fig. 12, P3 and
P4). Therefore, C-Krox was acting as an inhibitor of
transcription of a nonspecific reporter gene in dedifferentiated RAC.
This effect is correlated with the C-Krox inhibitory action on the
266-bp short COL2A1 promoter transcription in phenotype-altered RAC,
where C-Krox DNA-binding sites have been localized.
Finally, we examined whether double-stranded oligonucleotide could
modulate COL2A1 gene transcriptional activity through C-Krox-specific DNA-binding site in decoy experiments performed on primary RAC with the objective of confirming that an endogenous C-Krox-like activity was present in these cells. In this study, we chose to transfect the In this study, we found an enhancer in the human COL2A1 gene
between +2384 and +2842 bp, responsible for its specific expression in
chondrocytes, as reported for the rat and mouse genes (13, 14, 16, 20,
21). Within this enhancer, two regions (+2384/+2689 and +2689/+2842 bp)
mediate separately the enhancing effect. Nevertheless, sequences
encompassing +2127/+2384 contain also binding sites for transactivating factors.
The COL2A1 enhancer has a strong homology among different species. In
rat (14), mouse (20, 21, 42), and human genes (43), there is at least a
binding site for a transcription factor of the high mobility group
family (L-SOX5, SOX6, and SOX9), and another high mobility group-like
element (+2398/+2404 and +2388/+2394 in the human gene, respectively)
(21, 43). The sequences could be responsible, at least in part, for the
high transcriptional activity of the +2384/+2689 region observed in our
study, as already reported for the human COL2A1 gene (44).
SOX9 binds to a 48-bp region of the mouse COL2A1 enhancer and directs
the specific expression of this gene in transgenic mice as well as in
chondrocyte cultures (20, 42, 43). It is the first transcription factor
that was shown to determine the chondrocyte lineage and cartilage
differentiation (23). However, the control of the specific regulation
of the mouse COL2A1 gene rather involves a protein complex called CSEP
(chondrocyte-specific enhancer-binding proteins), including SOX9,
L-SOX5, SOX6, and other proteins so far unknown (20-22). Our study is
consistent with a recent report (43) showing that the human COL2A1
enhancer is more extended than that of the rat and mouse genes, in
which minimum size intronic fragments can direct COL2A1 gene-specific
expression. We also demonstrate that other proteins, different from the
SOX family, namely C-Krox, Sp1, and Sp3, bind to three
cis-acting elements in the two enhancer subregions. These
zinc finger proteins might be involved in the CSEP complex because we
demonstrated that C-Krox increased transcription of human COL2A1 gene
via the enhancer region in differentiated RAC. Sp1 and Sp3 were also
shown to bind C-Krox elements presenting the GC or CT characteristic sequence.
A 309-bp human-specific enhancer, linked to 5'-flanking COL2A1
sequences and lacZ gene, has also been shown to
direct transgene expression similar to that of the endogenous COL2A1
gene in the para-axial mesoderm and in somites of mouse embryos 8.5 days post-coitum (22). Thus, the enhancer could be regulated
by a multiproteic complex including at least SOX5, SOX6, SOX9, C-Krox,
and Sp1. This is highly probable because COL2A1, SOX9, C-Krox, and Sp1 have a similar gene expression pattern, being expressed almost at the
same time during mouse embryonic development (8.5-9.5 days post-coitum) (21, 22, 27, 44-46). The functional
significance of the enhancer C-Krox binding sites can be also supported
by the fact that two sequences (Y and Z), which correspond to the 458-bp enhancer described here, are implicated in the regulation of
limb cartilage formation (22).
We demonstrated that the enhancer effect is partly maintained during
the dedifferentiation process of chondrocytes. This suggests that type
II collagen synthesis, which is reduced in favor of types I and III
collagen production in this process (6, 47), could be potentially
decreased through changes of level (or binding properties) of
transactivating factors.
Despite the presence of C-Krox DNA-binding sites in the COL2A1
promoter, overexpression of this factor does not significantly alter
the transcription activity. This may be due to the fact that C-Krox
basal levels are already high in primary RAC cultures and may saturate
the binding sites of COL2A1 promoter. Alternatively, saturation of the
proximal promoter may provide a basal transcription activity, whereas
C-Krox overexpression would be rather effective on the first intron
enhancer region. This is the case for Sp1, a C-Krox homologue, which
stimulates the transcription of a reporter gene when it binds at 1.7 kb
upstream or downstream from the initiation site of transcription.
However, this stimulation required intracellular concentrations 5-10
times higher than those necessary for the enhancing effect induced by
Sp1 sites at the vicinity of the initiation site of transcription (48).
This could apply to C-Krox and more generally to zinc finger proteins.
It could be emphasized also that both the enhancer and promoter are
taking part in the type II collagen expression in chondrocytes.
Our results demonstrate the functional importance of the C-Krox
DNA-binding sites in both the promoter and enhancer regions and provide
evidence of the relative contribution of each region to the
transcriptional activity of COL2A1 gene. We showed that C-Krox
activates COL2A1 gene transcription in differentiated RAC via the
458-bp enhancer region. Nevertheless, subcloning of the 458-bp enhancer
upstream of a 266-bp short COL2A1 promoter failed to restore the
transcriptional activation induced by C-Krox overexpression on the
longest 3.774-kb construct. This absence of effect may result from a
difference in the C-Krox-sensitive element positioning, because they
are not located at their natural position, in respect to the
transcription initiation site. Interaction between the promoter and
enhancer regions is required to produce an optimal activation of the
transcription, probably via a transcription factor. This is supported
by previous data showing that the promoter and enhancer of the rat
COL2A1 gene are interacting with each other through a heteromeric
complex containing Sp1 and CIIZFP (CII zinc finger protein) (25).
C-Krox could be another candidate implicated in that complex,
facilitating the interaction between distant gene regions. Moreover,
several reports mentioned that 5'-flanking DNA sequences and first
intron must work synergistically to direct the expression of COL2A1
transgenes in tissues that normally express COL2A1 gene. Thus, the
expression of type II collagen transgenes requires the 309-bp enhancer
intronic fragment (+2388 to +2696: region Y) (22, 43). Moreover, the
presence of a 90-bp promoter region is necessary for tissue-specific
expression in different components of the developing cartilaginous
skeleton (22). Interestingly, we found C-Krox DNA-binding sites in both regions and in the 153-bp enhancer sequence, this latter being responsible for the inhibition of type II collagen expression in the
developing midbrain neuroepithelium of transgenic mice (22). Here, we
found three C-Krox binding sites in the COL2A1 enhancer, two being
located in the human 309-bp COL2A1 enhancer (43). So far, we do not
know whether the stimulation of the enhancer upon overexpression of
C-Krox is due to a direct or indirect mechanism, such as an increased
expression of a factor participating to the CSEP complex. In that
latter case, a member of the SOX family, like SOX9, would be a good
candidate (20, 21).
DNA binding activities of C-Krox, Sp1, and Sp3 were specifically
reduced during the RAC dedifferentiation process. This is apparently
not consistent with a previous report showing a 2-3-fold increase of
Sp1 binding in dedifferentiated human chondrocytes compared with
primary cultures (24). There is no clear explanation for this
discrepancy, because we confirmed this reduction of both Sp1 amount and
binding in passaged cultures with different methods (Western and
Southwestern analyses; data not shown). Nevertheless, it was of
interest to determine whether the overexpression of a transcriptional
factor shown to activate COL2A1 gene expression in differentiated RAC
could reactivate the type II collagen synthesis in already
dedifferentiated chondrocytes. However, overexpression of C-Krox in
dedifferentiated RAC was found to rather inhibit the transcription of
the COL2A1 gene, via the promoter region. C-Krox should therefore be
considered as a bifunctional transcription factor, the effect of which
depends on the chondrocyte differentiation state. In this regard, it is
worth noting that other C-Krox homologues, such as Krüppel and
Yin-Yang1 (YY1), have been shown to either activate or inhibit gene
transcription depending on whether they form homo- or heterodimers and
in the context where the gene is found (49-56). This may be the case
for C-Krox because this factor can form homodimers (27). Furthermore,
we have also shown that C-Krox, Sp1, and Sp3 bind to several common
cis elements, suggesting that they could interact together.
This strongly suggests that the resulting transcriptional effect
could be different, depending on the type of association between these
trans factors. C-Krox may also facilitate interactions
between proteins bound to nonadjacent DNA sites, promoting the
formation of multiprotein-enhancer and multiprotein-promoter complexes
and bringing them close to the transcriptional machinery, as it has
been postulated for SOX9 (20, 21).
NF- In summary, the present work documents the crucial influence of
chondrocyte differentiation state on the response of COL2A1 gene to
transcription factors. In contrast to SOX9, a key factor in the
cartilage-specific COL2A1 gene expression during development (23),
C-Krox can be considered rather as a bifunctional regulator of that
gene, its role depending on cooperating factors. However, its function
in the maintenance of the chondrocyte differentiated phenotype seemed
to be crucial because both high levels and binding activity of this
factor are associated with a high transcription of the human COL2A1
gene. Our study strongly suggests that several nuclear factors can act
together in combination to promote the loss of chondrocyte phenotype.
It is also clear that C-Krox overexpression is not sufficient to
restore the transcription of the human COL2A1 gene in dedifferentiated
articular chondrocytes and make them recover their phenotype.
Nevertheless, the present data contribute to our knowledge of the
transactivating mechanisms involved in the type II collagen expression
in chondrocytes and provide tools for further investigating their
modulation by cytokines and growth factors in the articular cartilage.
The transcription factor machinery may ultimately provide potential
targets to maintain or restore the chondrocyte phenotype in cartilage
and to design new therapeutic treatment of joint diseases in the future.
*
This work was supported by INSERM (CJF 91-06), the
University of Caen, the Regional Council of Normandy, and the Ligue
Nationale Contre le Cancer.The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
§
These authors contributed equally to this work.
¶
Fellow of the Association pour la Recherche sur le Cancer.
Published, JBC Papers in Press, June 15, 2000, DOI 10.1074/jbc.M002139200
2
C. Ghayor, C. Chadjichristos, M. Demoor-Fossard,
J.-F. Herrouin, F. Rédini, L. Ala-Kokko, G. Suske, J.-P. Pujol,
and P. Galéra, manuscript in preparation.
The abbreviations used are:
RAC, rabbit
articular chondrocyte(s);
FCS, fetal calf serum;
RLU, relative
luciferase unit(s);
kb, kilobase(s);
bp, base pair(s);
BSA, bovine
serum albumin;
MOPS, 4-morpholinepropanesulfonic acid;
GAPDH, glyceraldehyde-3-phosphate dehydrogenase;
HAC, human articular chondrocyte(s).
Regulation of Human COL2A1 Gene Expression in Chondrocytes
IDENTIFICATION OF C-Krox-RESPONSIVE ELEMENTS AND MODULATION BY
PHENOTYPE ALTERATION*
§,§¶,,
,, and
Laboratoire de Biochimie du Tissu
Conjonctif, Centre Hospitalier Universitaire de Caen,
Faculté de Médecine, Avenue de la Côte de Nacre,
14032, Caen Cedex, France, the Collagen Research Unit, Biocenter
and Department of Medical Biochemistry, University of Oulu, 90220 Oulu,
Finland, and the ** Department of Biochemistry and Molecular
Dentistry, Okayama University Dental School, Shikata-cho,
Okayama 700, Japan
ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
B were increased. This was not associated with
variations of mRNA levels, suggesting that post-transcriptional
regulatory mechanisms could be involved in C-Krox expression. These
results suggest that C-Krox plays a major role in type II collagen
expression and the chondrocyte phenotype modulation.
INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
/ cells from mouse chimaeras were shown to be excluded from the chondrocyte lineage during chondrogenesis at the condensation mesenchyme step and, as a consequence, display a rather mesenchymal phenotype instead of a cartilaginous phenotype (23).
EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
-galactosidase expression vector (2 µg) as an internal control of
transfection efficiency and, in some experiments, with increasing
amounts of a C-Krox expression vector (10-30 µg) (26). 10-15 h
after transfection, the medium was replaced by a fresh one, and cells
were harvested 24 h later. Luciferase activity was assayed on
total cell extracts (kit Promega) in a luminometer (Berthold Lumat LB
9501). -Galactosidase activity was assayed by a colorimetric assay
(31), whereas the protein amount was determined by the Bradford's
colorimetric procedure (Bio-Rad). Luciferase activities were normalized
to transfection efficiency and protein amount and expressed in relative
luciferase units (RLU) as the means ± S.D. of three independent samples.
327-305
1(II) wild-type
and mutant oligodeoxynucleotides (2.6 µM). After
overnight transfection, the culture medium was changed, and cell
cultures were incubated for a further 24-h period, and the relative
luciferase activity was determined.
1840 bp to
+5800 bp).
932 to +2842 bp (including a part of the promoter, exon 1, and
approximately 60% of the first intron), has been generated using the
pGL2 basic luciferase reporter plasmid (Promega). The 3.774-kb
NaeI insert of the COL2A1 gene has been subcloned in the
SmaI site of pGL2 basic to give pGL2-3.774 kb. pGL2-3.316 kb clone has been obtained by digestion of pGL2-3.774 kb with XhoI, excision of the 458-bp DNA insert and religation of
the vector. The pGL2-3.059 kb construct was obtained by digestion of
pGL2-3.774 kb by SacI, excision of the 715-bp DNA fragment, and ligation of the vector. The pGL2-1.783 kb clone (932 to +851) derived from the pGL2-3.774 kb construct in which the 2-kb
KpnI insert has been removed and the vector has been
religated. The pGL2-1.167 kb clone was obtained by subcloning a
1.167-kb BglII DNA insert of COL2A1 gene (544 to +623) in
the BglII site of pGL2 basic. The pGL2-2.367 kb clone was
prepared by cutting the SalI-SphI initial clone
by XbaI (present in the molecular cloning site of pUC19) and
NheI (position 531 in the COL2A1 gene), the resulting
1.213-kb insert being subcloned in pGL2-1.167 kb digested by
NheI. The pGL2-0.387 kb reporter vector was prepared by
insertion of a 387-bp SmaI insert (266 to +121) in the
NheI site of pGL2 basic whose extremities have been rendered
blunt ends.
) correspond
respectively to the correct or reverse orientations of the inserts. To
obtain the different clones, we have inserted: (a) a 153-bp
BamHI-HindIII DNA fragment (+2689/+2842) treated with klenow enzyme in both correct (+) or reverse () orientations in
the XhoI site of pGL2-1.167 kb that was filled in; these
subclones were called pGL2-1.167kbE153(+)US and
pGL2-1.167kbE153()US; (b) the same 153-bp
BamHI-HindIII insert in both orientations in the HindIII site of pGL2-1.167 kb that was blunted, to give
pGL2-1.167kbE153(+)DS and pGL2-1.167kbE153()DS; (c)
a blunt ended 458-bp XhoI insert (+2383/+2842), in opposite
orientation in the HindIII site of pGL2-1.167 kb whose
extremities were filled in. This construct was called
pGL2-1.167kbE458()DS; and (d) a 715-bp
SacI DNA fragment (+2127/+2842) in both orientations in the
SacI site of the pGL2-1.167 kb subclone, respectively
called pGL2-1.167kbE715(+)US and pGL2-1.167kbE715()US.
)US. The same insert was cloned in the
BglII site of pGL2-0.387 kb rendered blunt ended, giving
pGL2-0.387kbE121(+)DS and pGL2-0.387kbE121()DS. A 257-bp
SmaI-XhoI DNA fragment (+2127/+2384), subjected
to blunt ending by klenow polymerase, was subcloned in both
orientations in the filled in BglII site of pGL2-0.387 kb.
The resulting constructs were called pGL2-0.387kbE257(+)DS and
pGL2-0.387kbE257()DS. A blunt ended 562-bp
SmaI-BamHI insert (+2127/+2689) was subcloned in
the correct orientation in the SmaI site of pGL2-0.387 kb, and this construct was designated pGL2-0.387kbE562(+)US. A 715-bp SacI insert (+2127/+2842) was subcloned in either correct or
opposite orientation in the SacI site of pGL2-0.387 kb. The
resulting subclones were pGL2-0.387kbE715(+)US and
pGL2-0.387kbE715()US. A filled in 458-bp XhoI DNA
fragment (+2384/+2842) was inserted in the correct orientation in the
SmaI site of pGL2-0.387 kb: pGL2-0.387kbE458(+)US. A blunt
ended 153-bp BamHI-HindIII insert (+2689/+2842)
was ligated in the two orientations in the SmaI site of
pGL2-0.387 kb to give pGL2-0.387kbE153(+)US and
pGL2-0.387kbE153()US.
1(I) C-Krox consensus binding sites were subcloned.
These plasmids were designated, respectively, AP35 and AR5. To
determine the specificity of the effects observed, four mutant C-Krox
binding sites were also subcloned upstream to the minimal nonspecific
promoter (A'N°3) (26). Finally, a C-Krox expression vector,
pEMSV(+), obtained by subcloning the C-Krox cDNA in the pEMSV
scribe (pEMSV) expression vector was also used (26).
-32P]dATP (NEN Life Science
Products) using T4 polynucleotide kinase (Life Technologies, Inc.). 1 µl of RAC nuclear extracts (3-5 µg of total protein) or
full-length His-C-Krox protein (2-3 fmol) was incubated for 15 min at
room temperature with 5 fmol of the probe in a 10-µl reaction mixture
containing 20 mM Tris-HCl, pH 7.5, 1 mM
dithiothreitol, 100 mM NaCl, 0.05 mg/ml BSA, 0.1% Nonidet P-40, 10% glycerol, 25 µM ZnCl2, and 4 µg
of poly(dI·dC)·poly(dI·dC) (Amersham Pharmacia Biotech)
(except when recombinant C-Krox was used). Samples were fractionated by
electrophoresis for 1.5 h at 150 V on a 5% polyacrylamide gel
(acrylamide/bis at 30:1) in 0.5X TBE (45 mM Tris borate, 1 mM Na2EDTA) and visualized by autoradiography.
932/+2842) was digested by BamHI, dephosphorylated with
the calf alkaline phosphatase, and radiolabeled at its 5'
BamHI end (+2689) by the T4 polynucleotide kinase in the
presence of [-32P]dATP. Then, the DNA was cut by
XhoI. The 183-bp insert has been later fractionated on a
native polyacrylamide gel. After overnight elution at 37 °C in 10 mM Tris-HCl, pH 8, 1 mM EDTA, 100 mM NaCl, the probe was concentrated by ethanol
precipitation and dissolved in 10 mM Tris, pH 7.5, 1 mM EDTA. The level of [-32P]dATP
incorporated into the probe was determined in a Packard 1600 TR
scintillation counter. The 305-bp XhoI-BamHI
fragment (+2384/+2689) of the plasmid pGL2-3.774 kb has been
radiolabeled by [-32P]dATP at its 5' XhoI
end. The StyI-NarI DNA insert (391/21) of the
plasmid pGL2-1.167 kb has been radiolabeled at its 5' StyI end and finally the SmaI-NarI insert (266/21)
of the plasmid pGL2-0.387 kb at its 5' NarI end. These last
three DNA fragments were then gel purified, eluted, concentrated, and
counted as above.
80 °C.
::, v/v/v). Blots were washed
several times in 2× SSC (standard saline citrate) and 0.1% SDS at
room temperature and at 55 °C. Final washes were in 0.1× SSC plus
0.1% SDS at 55 °C. The 32P-labeled cDNA-mRNA
hybrids were visualized by autoradiography. To assess for variations in
RNA loading, filters were stripped by boiling in a 0.1% SDS solution
for 5 min and reprobed using a labeled -actin probe generated by
reverse transcription-polymerase chain reaction (38). The relative
density of detected signals was measured on a PhosphorImager (Molecular
Dynamics) and treated by the ImageQuant program (Molecular Dynamics).
The pG1 clone, containing approximately 1600 bp of C-Krox cDNA
sequence, was digested by XhoI and HindIII to
give a 322-bp cDNA probe covering the 5'-untranslated region and
part of the coding sequence (26). A 1.2-kb EcoRI fragment of
human COL2A1 cDNA cloned in pBluescript KS (HC 21 plasmid),
covering parts of the N-propeptide and the triple helical regions
(exons 2-19) coding sequence, was used as a probe (kindly provided by
Dr. Francesco Ramirez (Mt. Sinaï School of Medicine, New York).
A GAPDH probe was also used as a reference (39). The probes were
labeled by random priming (type II collagen, C-Krox, -actin) using
the Rad prime kit or by nick translation (GAPDH) using Life
Technologies, Inc. kits.
RESULTS
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
932 to +2842 bp, yielded the highest
transcriptional activity. Further 3' deletion in the first intron of
COL2A1 gene to +2127 decreased the transcription level by about 60%.
Deletion of the first intron to the KpnI site (+851) further
reduced the transcription which represents ~2% of the pGL2-3.774 kb
expression, suggesting that the +851/+2127 region mediates ~32% of
transcription activity of the COL2A1 gene and probably binds
transactivating factors. We also studied the effect of deletions in the
5'-flanking promoter region of COL2A1 gene. Transfection of recombinant
plasmids containing various lengths of the promoter region showed that
the 1.7-kb segment had a weak activity and induced less than 7% of the
transcriptional activity observed with the pGL2-3.774 kb construct.
Transfection of shorter promoter fragments bearing 5' deletions to
544 and 266 bp further reduced the transcriptional activity. The
data of Fig. 1A indicated that several enhancer regions are
present in the first intron of human COL2A1 gene and that the promoter
alone, when not coupled to the first intron enhancer regions, is not
involved in the high transcription activity of that gene in
differentiated RAC, albeit the transcription activity of the promoter
was slightly detectable. Two large enhancer regions of the first intron
were shown to drive high transcription activity, one comprising the
+851/+2127 sequence (~32% of transcription) and the other covering
+2127/+2842 sequence (~65% of transcription).
View larger version (37K):
[in a new window]
Fig. 1.
Evidence of a specific enhancer in the first
intron of human COL2A1 gene. Different reporter plasmids were
generated as described under "Experimental Procedures." The
DNA constructs were transiently transfected into fully differentiated
RAC (A) and rabbit dermal fibroblasts (B;
bracket F). 20 µg of reporter plasmid pGL2-3.774 kb were
transfected, and equal molar amount of COL2A1 sequences were used for
the other constructs, with the insertless pGL2 basic vector as a
complement to 20 µg. RAC and fibroblasts were cotransfected with a -galactosidase expression vector, pSV-gal (2 µg), used as
an internal control for transfection efficiency. Relative luciferase
activities were measured in cell extracts and normalized for
transfection efficiency. Values are expressed as percentages relative
to that of the pGL2-3.774 kb plasmid (A) and as RLU
(B). Note in the later case that the RLU scale has been
adjusted for each cell type, because of the differences in the extent
of transcriptional COL2A1 constructs activities, and that the RLU are
indicated at the top of each respective histogram. RLU
represents the mean ± S.D. of three independent samples of a
representative experiment.
View larger version (38K):
[in a new window]
Fig. 2.
Delineation of an enhancer region in the
first intron of the human COL2A1 gene by transient transfection
experiments. Different DNA fragments of the first intron of human
COL2A1 gene were cloned in correct and opposite orientation, upstream
(US) or downstream (DS) of two different promoter
regions. A, the first intron DNA inserts (called E153, E458,
and E715) were obtained by restriction enzyme digestion
(scheme on the left side of the figure). The
position of the first and last nucleotides relative to the COL2A1
transcription start site is indicated at the bottom left and
right sides of the promoter and intron fragments. These
constructs were transfected into primary RAC cultures, and RLU values
were determined in cell extracts as described under "Experimental
Procedures." Values are expressed as the average of RLU × 103 ± S.D. of three independent samples of a
representative experiment. The numbers indicated on the top
of the histograms represented the RLU percentages of constructs
containing intron fragments, expressed versus respective
control vectors, including only 544 or 266 bp of promoter sequences.
B, experimental design and presentation are similar to those
described in for A. The differences reside in the first
intron regions cloned beside the 266 bp short promoter as indicated
on the scheme of the left side of the figure. The
numbers on top of the histograms represent the
RLU percentages of constructs containing intronic inserts, expressed
versus the control vector with 266 bp of promoter sequences
(pGL2-0.387 kb).
458/440
bp in the human and 438/420 bp in the rat promoters) (40, 41). We
therefore suggest that the enhancing transcriptional activity of first
intron fragments is reduced by the putative silencer present in the
544-bp promoter; when this element was deleted in the 266-bp promoter, only the enhancing effect of first intron region remained effective.
View larger version (39K):
[in a new window]
Fig. 3.
The enhancer is still functional in
phenotypically altered chondrocytes. Part of the significative
plasmids described in the legend to Fig. 2 were transiently transfected
into chondrocytes dedifferentiated by two passages. Experimental
conditions were the same as described in the legend to Fig. 2.
Luciferase activities measured after transfection efficiency correction
are expressed as percentages (100% = 16206 RLU). The promoter-induced
transcriptional activation by intron fragments is given relative to the
activity obtained with the pGL2-3.774 kb control vector and is
presented as the average with standard deviation of three independent
samples. Numbers on top of the histograms
represented the RLU fold induction of constructs containing intronic
sequences, expressed versus the control vector including 266 bp of promoter sequences (pGL2-0.387 kb).
View larger version (50K):
[in a new window]
Fig. 4.
The transcription factor C-Krox is present in
chondrocytes. A and B, Western analysis.
Cells were harvested at 80-90% of confluency, and nuclear extracts
were prepared by the micropreparation technique. Protein concentration
of the extracts was determined, and 15 µg of nuclear extracts from
two sources of human dermal fibroblasts (lanes 2,
3, 6, and 7) and two different samples
of HAC (lanes 1, 4, 5, and
8) were fractionated on a 10% polyacrylamide gel and
transferred to polyvinylidene difluoride membrane (A).
Similarly, 100 ng of recombinant C-Krox were also separated on a 10%
SDS-polyacrylamide gel and processed as described for nuclear extracts
samples (B). The Western blot was performed as described
under "Experimental Procedures." Numbers on the
left correspond to the molecular mass of prestained marker
proteins. A, lanes 1-4, nuclear extracts were
incubated with the preimmune serum of the rabbit used to generate
C-Krox polyclonal antibodies; lanes 5-8, nuclear extracts
were incubated with rabbit anti-C-Krox. B, polyclonal
antibodies against C-Krox (aC-Krox, lane 2) or
preimmune serum (P, lane 1) were incubated
together with C-Krox recombinant protein. Sera were used at 1:150
dilution for nuclear extracts samples and 1:2000 dilution for
recombinant C-Krox protein. C-Krox protein was localized by a secondary
antibody (anti-rabbit IgG conjugated with alkaline phosphatase) used at
a 1:10000 dilution. The arrows indicate the position of the
C-Krox-antibody complex. Numbers on the left correspond to
the molecular masses of prestained marker proteins. C,
Northern analysis. 0.75 µg of poly(A)+ RNA from confluent
primary RAC cultures were fractionated on a MOPS-formaldehyde gel,
transferred to nylon membranes, and hybridized with a 330-bp labeled
probe located 5' of the zinc finger region of the C-Krox cDNA and a
specific GAPDH probe. The arrows denote the positions of
C-Krox and GAPDH mRNAs and 28 and 18 S ribosomal RNAs.
D, DNA binding assay. DNA binding was analyzed by
electrophoretic mobility shift assay. A double-stranded labeled
consensus C-Krox DNA-binding site called A 1(I) oligonucleotide was
incubated 15 min at room temperature with 5 µg of nuclear extracts
from primary RAC (RAC P0, lane 1), NIH-3T3
fibroblasts (lane 2), human chondrosarcoma chondrocytes
(HCS, lane 3), and HAC (lane 4). The
resulting DNA-proteins complexes were resolved in nondenaturing
conditions by SDS-polyacrylamide gel electrophoresis (5% gel) and
exposed to x-ray film overnight for autoradiography. The
arrows indicate the position of the different nuclear
proteins-DNA complexes (a corresponds to A1(I)-Sp1 and
Sp3 complexes, b represents Sp3-DNA complex, c
indicates C-Krox-probe complex, and NS indicates nonspecific
DNA binding).
1(I) DNA cis-acting element of the mouse COL1A1 promoter, previously used to clone C-Krox
by Southwestern (26). We found that the radiolabeled A1(I)
oligonucleotide bound nuclear proteins from RAC primary cultures (P0),
HCS 2/8 chondrosarcoma cells, HAC, and NIH-3T3 fibroblasts (Fig.
4D). The mobility shift assays showed three specific
protein-DNA complexes. The complex a involves Sp1 and Sp3
transcription factors, the complex b contains
Sp3,2 and the complex
c represents the C-Krox-A1(I) complex, because NIH-3T3
nuclear extract protein-DNA complex was shown to migrate with a
mobility similar to that of the His-tagged C-Krox-DNA complex, and it
was shown that C-Krox binding to the probe could be prevented by
high EDTA concentrations and restored by zinc addition (26). However,
C-Krox binding activity is detected at greater levels in primary RAC,
HAC, and NIH-3T3 than in HCS 2/8 nuclear extracts. This suggests that
HCS 2/8 chondrosarcoma cells probably contain lower amounts of C-Krox.
Binding of other nuclear proteins to the A1(I) oligonucleotide was
nonspecific because addition of increasing excess of cold COL1A1 or
COL2A1 wild-type or mutant oligonucleotides did not prevent their
binding (data not shown).
View larger version (21K):
[in a new window]
Fig. 5.
C-Krox stimulates human COL2A1 gene
transcription through part of the enhancer region in fully
differentiated RAC and inhibits expression of COL2A1 gene by the
promoter region in dedifferentiated RAC. A, summary of
expression of regulatory sequences of human COL2A1 gene in RAC and
responses to C-Krox overexpression. The top
diagram represents the entire 3.774-kb regulatory region
spanning 932 to +2842 bp. The position of the specific restriction
enzyme sites used for cloning are indicated below or above (for
intronic DNA inserts indicated in smaller character size) the fragments
inserted upstream of the luciferase reporter gene. Primary RAC and RAC
at passage 2 were cotransfected with 10 µg of different COL2A1
reporter plasmids along with 10 µg of pEMSV expression vector that
either did contain (pEMSV (+)) or did not contain (pEMSV) the C-Krox
cDNA. A -galactosidase expression vector, pSVgal (2 µg) was
used as an internal control for transfection. Luciferase activities of
a representative experiment are presented as the average with standard
deviation of three independent samples. RLU values upon C-Krox
expression vector cotransfection are expressed as percentages relative
to that of the respective reporter construct cotransfected by the
C-Krox insertless expression vector pEMSV. US and
DS in the plasmid names refer, respectively, to the upstream
and downstream cloning positions of the first intron regions, 5' or 3'
of the COL2A1 promoter used. B, overexpression of C-Krox in
RAC. DNA binding was analyzed by electrophoretic mobility shift assay.
A labeled wild-type A1(I) oligonucleotide was incubated with 5 µg
of mini-nuclear extracts from transfected primary RAC (lanes
1-5) and dedifferentiated RAC (two passages) (lanes
6-9). Mini-nuclear extracts were prepared after transient
transfection with 30 µg of pEMSV (lanes 1 and
6), 20 µg of pEMSV and 10 µg of pEMSV(+) (lanes
2 and 7), with 10 µg of pEMSV and 20 µg of pEMSV(+)
(lanes 3 and 8), and 30 µg of pEMSV(+)
(lanes 4 and 9). The sample of lane 5 corresponds to nuclear extracts from untransfected primary RAC. The
C-Krox-DNA complex is indicated by an arrow. C,
C-Krox activates transcription of the human COL2A1 gene through a
region spanning +2384 to +2842 bp. The RLU data of 10 separate
experiments performed on primary chondrocytes from different rabbits
are presented. The percentages represented the mean of relative
luciferase activity of three samples cotransfected with 10 µg of
expression vector pEMSV(+) and 10 µg of pGL2-3.774 kb plasmid
versus the transcription activity of three samples of RAC P0
cotransfected with the same quantity of pGL2-3.774 kb reporter vector
and the insertless expression vector pEMSV. D, C-Krox
effects on COL2A1 gene transcription are correlated with endogenous
COL2A1 mRNA amounts in RAC. Cultures of primary RAC (RAC
P0, lanes 1 and 2) and chondrocytes
dedifferentiated by two passages (RAC P2, lanes 3 and 4) were transfected with an expression vector containing
C-Krox cDNA (pEMSV+, 20 µg; lanes 2 and 4)
or not (pEMSV, 20 µg; lanes 1 and 3) at 80%
confluency. Total RNAs were extracted, and 15 µg were fractionated by
electrophoresis and processed for Northern blot analysis. Filters were
hybridized with COL2A1 (HC21) and -actin cDNA probes.
1(I) oligonucleotide, an important decrease in C-Krox binding
activity was observed, compared with primary RAC nuclear extracts,
suggesting that C-Krox levels or binding activity might be reduced
during the phenotype alteration of chondrocytes (Fig. 5B,
lanes 5 and 6). Cotransfection of the C-Krox
expression vector led to the restoration of C-Krox DNA binding activity
(Fig. 5B, lanes 7-9).
-actin mRNA amounts (Fig.
5D, lanes 2 versus lanes
1). In RAC dedifferentiated by two passages, C-Krox overexpression
led to a decrease of both COL2A1 and COL1A1 mRNA levels (Fig.
5D, lanes 4 compared with lanes 3). It
must be noted that in these conditions, the COL2A1 cDNA probe
(HC21) hybridized with the two mRNA species, because of its
conserved sequence coding for the collagen triple helix. It is also
clear that COL1A1 expression is reflected by the presence of the 5.8-kb
mRNA transcript, but the other mRNA species observed is
probably due to the overlapping of 5-kb COL2A1 and 4.8-kb COL1A1 mRNAs, hardly separated in these electrophoretic conditions.
266-bp region in
dedifferentiated RAC.
73/98 bp and 337/360 bp of COL2A1 promoter
(Fig. 6, A, lanes 1 and 2, and B, lanes 1 and 2). Binding was also
observed with the positive control represented by the A1(I)
oligonucleotide (Fig. 6A, lanes 8 and
9). Complex c corresponds to C-Krox interacting
with the probe. Using nuclear extracts from dedifferentiated RAC, we
observed a decreased binding of C-Krox, Sp1, and Sp3 to the different
labeled oligonucleotides. This effect was already detected after one
culture passage (Fig. 6, A, lanes 3 and
10 compared with lanes 2 and 9, and
B, lane 3 compared with lane 2). After
three passages, no more binding of the three transcription factors was
observed.
View larger version (47K):
[in a new window]
Fig. 6.
C-Krox levels are specifically reduced during
the dedifferentiation process of RAC. A-C, DNA binding
was analyzed by gel retardation assay. A, labeled wild-type
7398 1(II) (lanes 1-7) and wild-type A1(I)
(lanes 8-14) double-stranded oligonucleotides were
incubated for 15 min with 5 µg of nuclear extracts from NIH-3T3
fibroblasts (lanes 1 and 8), primary RAC
(lanes 2 and 9), RAC P1 (one passage; lanes
3 and 10), RAC P2 (two passages; lanes 4 and
11), RAC P3 (three passages; lanes 5 and
12), RAC P4 (four passages; lanes 6 and
13), and RAC P5 (five passages; lanes 7 and
14). B, a labeled wild-type 337360 1(II)
double-stranded oligonucleotide was incubated with 5 µg of nuclear
extracts from NIH-3T3 fibroblasts (lane 1), primary RAC
(lane 2), and dedifferentiated chondrocytes by one, two,
three, four, and five passages (respectively, lanes 3,
4, 5, 6, and 7).
C, a labeled wild-type NF-B binding site was incubated
with 5 µg of nuclear extracts from RAC (lanes 1-6).
Lane 1, primary RAC; lanes 2-5, nuclear extracts
from chondrocytes dedifferentiated by one, two, three, and five
passages, respectively; lane 6, competition performed with a
100-fold molar excess of unlabeled wild-type NF-B binding site. The
arrows in A-C indicate the protein-DNA
complexes. Arrow a, Sp1 + Sp3; arrow b, Sp3; c:
C-Krox; arrows d-g, the NF-B complexes involving
homodimers or heterodimers of different subunits such as p50, p65, etc.
D, analysis of C-Krox mRNA expression during the
in vitro dedifferentiation process of RAC.
Poly(A)+ RNA (approximately 2.5 µg/lane) from primary RAC
(RAC P0), RAC P3 (three passages), and RAC
P5 (five passages) were analyzed by Northern blot hybridization
using as a probe a 330-bp XhoI-HindIII fragment
located 5' of the zinc finger-coding region of the mouse C-Krox
cDNA. The same filter was hybridized with a human -actin
cDNA probe as a control. E, C-Krox protein expression is
reduced during phenotypic alteration of RAC. 20 µg of primary
(lane 1) and dedifferentiated by one or two passages
(lanes 2 and 3, respectively) nuclear extracts
prepared by the maxipreparation method were separated on a 6%
acrylamide gel in denaturing conditions. Then the proteins were
transferred to a polyvinylidene difluoride membrane and reacted with a
polyclonal antibody against C-Krox (1:100 dilution). C-Krox protein was
revealed with a horseradish peroxidase-labeled secondary antibody
(anti-rabbit IgG, 1:1000 dilution) using an ECL+Plus Western blot
detection kit.
B binding site as a probe. Incubation of
primary RAC nuclear extracts with the NF-B probe revealed the
presence of four specific nuclear protein-DNA complexes, as
demonstrated by addition of a 100-fold molar excess of cold NF-B
wild-type binding site that competed away the binding of nuclear
proteins to the probe (Fig. 6). As the dedifferentiation proceeded, we
noticed an increase of the DNA binding activity of the transcription
factor(s) involved in the complex e, whereas the binding
activity of complexes d, f, and g was
unaffected by subculturing the RAC. The decrease in C-Krox, Sp1, and
Sp3 binding activities during RAC dedifferentiation could therefore be
considered as rather specific because a complex involving one or two
NF-B subunits showed an increased DNA binding activity in the same conditions.
1(II)
oligonucleotide). Part of this region was equally protected when the
same amounts of primary RAC and NIH-3T3 nuclear extracts were used,
indicating that the transcription factors binding to this region are
present in similar quantity in both cell types (Fig. 7A,
lanes 4-6). A weaker footprinted area was detected with HCS
2/8 nuclear extracts, probably because of lower levels of the factors
binding to this region (Fig. 7A, lanes 9 and
10). When nuclear extracts of dedifferentiated RAC (three
and five passages) were used, we observed a reduced protection on the
+2817/+2845 region, suggesting that the amounts or/and the binding
activity of transcription factors recognizing this sequence were
reduced during RAC dedifferentiation (Fig. 7A, compare
lanes 7 and 8 versus lane
6).
View larger version (59K):
[in a new window]
Fig. 7.
C-Krox binding sites are present in both
promoter and first intron regions of the human COL2A1 gene. DNA
binding was analyzed by DNase I footprint assay. A, the
153-bp BamHI-NaeI fragment of plasmid pGL2-3.774
kb (+2689+2842) was labeled at the NaeI 5' end, incubated
with or without nuclear extracts or His-tagged C-Krox, and treated with
DNase I. Lane 1, A+G sequencing reaction of the
end-labeled fragment; lanes 2 and 3, DNase
I digestion pattern of the naked DNA; lanes 4 and
5, DNase I digestion pattern of the DNA incubated,
respectively, with 30 and 45 µg of NIH-3T3 extracts; lane
6, with 45 µg of primary RAC nuclear extracts; lanes
7 and 8, with 45 µg of RAC P3 and RAC P5 nuclear
extracts, respectively; lanes 9 and 10, with 30 and 45 µg of HCS 2/8 nuclear extracts; lane 11, with 1.5 nmol of His-tagged C-Krox. B, the 305-bp
XhoI-BamHI fragment (+2384 to +2689) of plasmid
pGL2-3.774 kb was labeled at the XhoI 5' end, incubated
with or without nuclear extracts or recombinant C-Krox, and treated
with DNase I. Lane 1, A+G sequencing reaction; lanes
2 and 3, DNase I digestion pattern of the naked DNA;
lane 4, DNase I digestion pattern of the DNA incubated with
45 µg of NIH-3T3 extracts; lane 5, probe incubated with 45 µg of primary RAC nuclear extracts; lanes 6 and
7, end-labeled DNA with 45 µg of RAC P3 or RAC P5 nuclear
extracts respectively; lanes 8, with 45 µg of HCS 2/8
extracts; lane 9, with 1.5 nmol of His-tagged C-Krox.
C, the 370-bp StyI-NarI fragment
( 391 to 21) of plasmid pGL2-1.167 kb was labeled at the
StyI 5' end, incubated with or without nuclear extracts or
His-tagged C-Krox, and treated with DNase I. Lane 1, A+G
sequencing reaction; lanes 2 and 3, naked DNA;
lane 4, DNase I digestion of the DNA incubated with 45 µg
of primary RAC nuclear extracts; lanes 5 and 6,
with 45 µg of RAC P3 and RAC P5 extracts respectively; lane
7, with 45 µg of HCS 2/8 nuclear extracts; lane 8,
with 1.5 nmol of recombinant C-Krox. D, the 245-bp
SmaI-NarI fragment (266 to 21) of plasmid
pGL2-1.167 kb was labeled at the NarI 5' end and incubated
with or without nuclear extracts or recombinant C-Krox. Lane
1, A+G sequencing reaction; lanes 2 and 3,
DNase I digestion pattern of the naked DNA; lanes 4 and
5, DNase I digestion of the DNA incubated respectively with
30 and 45 µg of NIH-3T3 nuclear extracts; lane 6, probe + 45 µg of primary RAC nuclear extracts; lane 7, probe + 45 µg of RAC P3 nuclear extracts; lane 8, probe + 45 µg of
HCS 2/8 nuclear extracts; lane 9, DNA probe with 1.5 nmol of
recombinant C-Krox. The different protected areas on A-D
are marked by brackets. The numbers on the
right correspond to base pairs upstream or downstream the
start site of transcription.
1(II) collagen enhancer fragment as a template,
we observed that C-Krox bound to another site located between +2440 and
+2485 bp, which is also a GC-rich region (Table I, +2440+24851(II) wt oligonucleotide)
(Fig. 7B, lane 9). This region was protected by
nuclear proteins from both primary RAC and NIH-3T3 fibroblasts (Fig.
7B, lanes 5 and 4, respectively) and a
fainter footprint is observed with extracts of HCS 2/8 chondrosarcoma cells (Fig. 7B, lane 8). Amounts or binding
activity of the transcription factors interacting with this region are
decreased during the RAC dedifferentiation, because no further
protection was seen when the same quantity of nuclear extracts from
phenotype-altered RAC was used (Fig. 7B, lanes 6 and 7 compared with lane 5).
Oligonucleotides used in this study
391/21-bp promoter fragment as a template, on
both strands, we observed that C-Krox bound to three sites covering the
222 to 360-bp region (Fig. 7C, lane 8) and
three other sequences: 66/103, 107/135, and 188/219 bp
(Fig. 7D, lane 9). Transcription factors present in nuclear extracts of primary RAC, HCS 2/8, and NIH-3T3 fibroblasts protected these regions (Fig. 7, C, lanes 4 and
7, and D, lanes 4-6 and
8), except for the 188/219 bp region, which is not
protected by any of the nuclear extracts. When nuclear extracts of
dedifferentiated RAC were used, we confirmed that the amounts or
binding activity of transcription factors interacting with the
different GC-rich or CTC sequences were reduced. The sequences of the
protected areas are shown in Table I.
3042741(II) oligonucleotide). By contrast, C-Krox did
not bind, or bound only very slightly, to the mutant oligonucleotides,
indicating that the fixation of this transcription factor on the
cis sequence of COL2A1 gene is specific (Fig. 8). The
differences in binding activity according to the oligonucleotides might
be explained by variations in DNA affinity or differences in
interacting sequences surrounding the cis element.
View larger version (34K):
[in a new window]
Fig. 8.
C-Krox binds specifically to the
corresponding cis-acting elements identified by
footprinting analyses. DNA binding was analyzed by gel mobility
shift assay. 32P-Labeled double-stranded wild-type
(Wt.) or mutant (Mut.) oligonucleotides indicated
on the figure that have been identified in footprint experiments were
incubated for 15 min with approximately 2 fmol of His-tagged C-Krox
protein (A, lanes 1-15, and B,
lanes 1-9). Lane 6 in B corresponds
to C-Krox protein incubated with a consensus A 1(I) oligonucleotide
from COL1A1 mouse gene. C-Krox protein-DNA complex is indicated by an
arrow.
3373601(II) oligonucleotide (Fig. 9, A, lanes
1 and 10, and B, lane 1). This
binding was specific because it could be inhibited by molar excesses of
the following wild-type nucleotides: 2252651(II),
1071351(II), 1882251(II), 3373601(II), 3053271(II), and 73981(II). C-Krox binding was not
competed away by the counterpart mutant oligonucleotides,
respectively: 225- 2651(II), 1071351(II),
1882251(II), 337360 1(II), 3053271(II), and
73981(II).
View larger version (87K):
[in a new window]
Fig. 9.
C-Krox binds specifically to the
3373601(II) cis-acting
element. DNA binding was analyzed by electrophoretic mobility
shift assay. A labeled 3373601(II) double-stranded
oligonucleotide was incubated for 15 min with the recombinant C-Krox
protein. DNA competition experiments were performed with the indicated
molar excess of wild type (Wt.) 2252651(II)
oligonucleotide (A, lanes 2 and 3),
2252651(II) mutant (Mut.) 1 (A,
lanes 4 and 5), mutant 2 (A,
lanes 6 and 7), and mutant 3 oligonucleotides
(A, lanes 8 and 9), wild type
1071351(II) oligonucleotide (A, lanes 11 and 12), mutant 1071351(II) oligonucleotide
(A, lanes 13 and 14), wild type
1882251(II) oligonucleotide (A, lanes 15 and 16), mutant 1882251(II) oligonucleotide
(A, lanes 17 and 18), wild type
(B, lanes 2 and 3) and mutant
3373601(II) oligonucleotides (B, lanes 4 and 5), wild type (B, lanes 6-8) and
mutant 305-3271(II) oligonucleotides (B, lanes
9-11), wild type (B, lanes 12-14) and
mutant 73981(II) oligonucleotides (B, lanes
15-17). In A, lane 1 represents the control
of lanes 2-9, and lane 10 represents the control
of lanes 11-18.
1(II) (Fig.
10). The binding was competed away by
addition of molar excesses of the following wild-type oligonucleotides:
+2440+24851(II), +2817+28451(II), 2252651(II),
1071351(II), and 2743041(II). The respective mutant
counterparts were not able to abolish C-Krox binding to the
+2440+24851(II) probe (Fig. 10).
View larger version (84K):
[in a new window]
Fig. 10.
C-Krox binds specifically to the
+2440+2485 1(II) cis
element. Experimental conditions are the same as described
in the legend of Fig. 9. A labeled +2440+24851(II) double-stranded
oligonucleotide was incubated with C-Krox protein. DNA competition
experiments were performed with the indicated cold molar excesses of
wild type (Wt.) and mutant (Mut.)
oligonucleotides. In A, lane 1 represents the
control of lanes 2-7, and lane 12 represents the
control of lanes 8-11. In B, lane 1 is the control of lanes 2-19, and lane 20 is
that of lanes 21-24.
1071351(II) (Fig.
11A, lane 1), 1882251(II) (Fig.
11A, lane 6), and 3053271(II) (Fig.
11B, lane 1). The binding was competed away by
addition of the respective wild-type oligonucleotides nor by their
mutant counterparts (Fig. 11). Additional gel retardation experiments
performed with a number of combinations, varying the promoter and first
intron cis COL2A1 gene elements used as a probe and also the
wild-type and mutant competitors, demonstrated that C-Krox binding to
these newly characterized sequences was specific (data not shown). In
conclusion, we identified several cis elements in human
COL2A1 gene that specifically bind C-Krox and could be putative targets
whereby C-Krox mediates its transcriptional effects on type II collagen
expression in both differentiated and dedifferentiated RAC.
View larger version (42K):
[in a new window]
Fig. 11.
C-Krox binds specifically to other promoter
cis-acting elements. Experimental conditions are
the same as in Fig. 9. Labeled 1071351(II) (A,
lanes 1-5), 1882251(II) (A, lanes
6-8), and 3273051(II) (B, lanes
1-7) double-stranded oligonucleotides were incubated with
recombinant C-Krox protein. DNA competition experiments were performed
with the indicated cold molar excesses of wild type (Wt.)
and mutant (Mut.) oligonucleotides.
1(I) cis element cloned upstream of a minimum COL1A1 promoter fused to the luciferase gene were transfected in primary RAC cultures. They induced an increase
in relative luciferase activity of 10- and 55-fold, respectively (Fig.
12, top left panel),
compared with the transcriptional activity observed with the reporter
plasmid containing four mutated C-Krox binding sites (A'N°3). This
suggested that an endogenous C-Krox-like activity enhances
transcription. An expression vector including C-Krox cDNA was
unable to modify the observed effects on transcription of the reporter
vector (Fig. 12, top left panel). The mechanism whereby
C-Krox activates transcription of the AP35 and AR5 constructs in
primary RAC seemed to be different, as previously observed in NIH-3T3
fibroblasts where C-Krox overexpression was still able to mediate a
further increase in transcriptional activity (26). It is therefore
possible that cofactors involved in C-Krox-induced transactivation are
present in fibroblasts and not in chondrocytes or that their nature
differs according to the cell type. We may also suggest that high
levels of endogenous C-Krox are present in RAC and could be sufficient
to saturate the four or five copies of C-Krox binding site present in
the reporter vector.
View larger version (28K):
[in a new window]
Fig. 12.
C-Krox binding site mediates
transcriptional activation of a nonspecific promoter whatever the
differentiation state of RAC, and overexpression of C-Krox inhibits
transcription in dedifferentiated RAC. Primary RAC (RAC
P0) and dedifferentiated RAC (by three and four passages,
respectively indicated by RAC P3 and RAC P4) were
transiently transfected. The transfections were performed with 9 µg
of reporter plasmid (A'N°3, AP35, or AR5), along with 9 µg of pEMSV
expression vector that either contained (pEMSV(+)) or did not contain
(pEMSV) the C-Krox cDNA. pSV gal was used as an internal control
for transfection efficiency. For each passage, the relative luciferase
activities were expressed as percentages of the transcriptional
activity determined with the A'N°3 plasmid cotransfected with the
insertless expression vector pEMSV. RLU represented the averages ± S.D. of three independent samples of a representative
experiment.
3273051(II) consensus C-Krox binding site
identified in the COL2A1 gene promoter in footprint experiments. For
the reporter plasmids, we deliberately used the pGL2-3.774 kb and pGL2-0.387 kb constructs which were previously shown to mediate the
transcriptional function of C-Krox on COL2A1 gene in differentiated (i.e. activation of pGL2-3.774 kb expression) and
dedifferentiated RAC (i.e. inhibition of pGL2-0.387 kb
transcription). The data presented in Fig.
13 demonstrated that expression of the
reporter plasmids pGL2-3.774 kb and pGL2-0.387 kb was decreased by
52.6 and 47.9%, respectively, when the chondrocytes are transfected with the wild-type 3273051(II) oligonucleotide, when compared with the cells transfected with the mutant counterpart. These results
demonstrate that an endogenous C-Krox-like activity is present in
primary RAC cultures and is responsible for transactivation of the
COL2A1 gene.
View larger version (20K):
[in a new window]
Fig. 13.
Inhibition of COL2A1 gene expression by the
double-stranded C-Krox consensus DNA-binding site. Primary RAC
were transiently transfected with 15 µg of pGL2-3.774 kb and
pGL2-0.387 kb together with 2.6 µM of wild-type or
mutant double-stranded 3273051(II) C-Krox cis binding
site and 2 µg of pSV-gal. RLU values are expressed as percentages
relative to that of the pGL2-3.774 kb or pGL2-0.387 kb
transfected with the mutant 3273051(II)
oligonucleotide. RLU represent the means ± S.D. of three
independent samples of a representative experiment.
DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
B also contributes to the control of human COL2A1 gene
transcription because increased binding of a particular NF-B complex was shown during the RAC dedifferentiation. NF-B could take part in
the loss of chondrocyte phenotype because inflammatory cytokines such
as IL-1 and TNF-, which are responsible for alteration of articular
cartilage in osteoarthritis, are known to induce activation of
metalloprotease genes through AP1- and NF-B-responsive elements. These cytokines also inhibit the synthesis of type II collagen (57-59).
FOOTNOTES
Supported by fellowships of the Fondation pour la Recherche
Médicale and Association pour la Recherche sur le Cancer. To whom
correspondence should be addressed: Laboratoire de Biochimie du Tissu
Conjonctif, CHU de Caen, Faculté de Médecine, Avenue de la
Côte de Nacre 14032, Caen Cedex, France. E-mail:
Galera.Philippe@wanadoo.fr.
ABBREVIATIONS
REFERENCES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
1.
Eyre, D. R.,
Wu, J. J.,
and Woods, P.
(1992)
in
Articular Cartilage and Osteoarthritis
(Kuettner, K. E.
, Schleyerbach, R.
, Peyron, J. G.
, and Hascall, V. C., eds)
, pp. 119-131, Raven Press, New York
2.
Petit, B.,
Freyria, A. M.,
van der Rest, M.,
and Herbage, D.
(1992)
in
Biological Regulation of the Chondrocytes
(Adolphe, M., ed)
, pp. 33-84, CRC Press, Boca Raton, FL
3.
Benya, P. D.,
Padilla, S. R.,
and Nimni, M. E.
(1978)
Cell
15,
1313-1321
4.
Kuettner, K. E.,
Memoli, V. A.,
Pauli, B. U.,
Wzobel, N. C.,
Thonar, E. J. M.,
and Daniel, J. C.
(1982)
J. Cell Biol.
93,
751-757
5.
Archer, C. W.,
McDowell, J.,
Bayliss, M. T.,
Stephens, M. D.,
and Bentley, G.
(1990)
J. Cell Sci.
97,
361-371
6.
Galéra, P.,
Rédini, F,
Vivien, D.,
Bonaventure, J.,
Penfornis, H.,
Loyau, G.,
and Pujol, J.-P.
(1992)
Exp. Cell Res.
200,
379-392
7.
Bonaventure, J.,
Khadom, N.,
Cohen-Solal, L.,
Ng, K. H.,
Bourguignon, J.,
Lasselin, C.,
and Freissinger, P.
(1994)
Exp. Cell Res.
212,
97-104
8.
Hauselman, H. J.,
Fernandes, R. J.,
Mok, S. S.,
Schmid, T. M.,
Block, J. A.,
Aydelotte, M. B.,
Kuettner, K. E.,
and Thonar, E. J.-M. A.
(1994)
J. Cell Sci.
107,
17-27
9.
Reginato, A. M.,
Iozzo, R. V.,
and Jimenez, S. A.
(1994)
Arthritis Rheum.
37,
1338-1349
10.
Nimni, M.,
and Deshmukh, K.
(1973)
Science
181,
751-752
11.
Golwasser, M.,
Astley, T.,
van der Rest, M.,
and Glorieux, F. H.
(1982)
Clin. Orthop.
167,
296-302
12.
Von der Mark, K.,
Kirsch, T.,
Nerlich, A.,
Kuss, G.,
Weseloh, G.,
Gluckert, K.,
and Stoss, H.
(1992)
Arthritis Rheum.
35,
806-811
13.
Horton, W.,
Miyashita, T.,
Kohno, K.,
Hassall, J. R.,
and Yamada, Y.
(1987)
Biochemistry
84,
8864-8868
14.
Wang, L.,
Balakir, R.,
and Horton, W. E., Jr.
(1991)
J. Biol. Chem.
266,
19878-19881
15.
Seghatoleslami, M. R.,
Lichtler, A. C.,
Upholt, W. B.,
Kosher, R. A.,
Clark, S. H.,
Mack, K.,
and Rowe, D. W.
(1994)
Matrix Biol.
14,
753-764
16.
Mukhopadhyay, K.,
Lefebvre, V.,
Zhou, G.,
Garofalo, S.,
Kimura, J. H.,
and de Crombrugghe, B.
(1995)
J. Biol. Chem.
270,
27711-27719
17.
Zhou, G.,
Garofalo, S.,
Mukhopadhyay, K.,
Lefebvre, V.,
Smith, C. N.,
Eberspaecher, H.,
and de Crombrugghe, B.
(1995)
J. Cell Sci.
108,
3677-3684
18.
Krebsbach, P. H.,
Nakata, K.,
Bernier, S. M.,
Hatano, O.,
Miyashita, T.,
Rhodes, C. S.,
and Yamada, Y.
(1996)
J. Biol. Chem.
271,
4298-4303
19.
Lefebvre, V.,
Zhou, G.,
Mukhopadhyay, K.,
Smith, C. N.,
Zhang, Z.,
Eberspaecher, H.,
Zhou, X.,
Sinha, S.,
Maity, S. N.,
and de Crombrugghe, B.
(1996)
Mol. Cell. Biol.
16,
4512-4523
20.
Lefebvre, V.,
Huang, W.,
Harley, V. R.,
Goodfellow, P. N.,
and de Crombrugghe, B.
(1997)
Mol. Cell. Biol.
17,
2336-2346
21.
Lefebvre, V.,
Li, P.,
and de Crombrugghe, B.
(1998)
EMBO J.
17,
5718-5715
22.
Leung, K. K. H.,
Ng, L. J.,
Ho, K. K. Y.,
Tam, P. P. L.,
and Cheah, K. S. E.
(1998)
J. Cell Biol.
141,
1291-1300
23.
Bi, W.,
Deng, J. M.,
Zang, Z.,
Behringer, R. R.,
and de Crombrugghe, B.
(1999)
Nat. Genet.
22,
85-89
24.
Dharmavaram, R. M.,
Liu, G.,
Mowers, S. D.,
and Jimenez, S. A.
(1997)
J. Biol. Chem.
272,
26918-26925
25.
Savagner, P.,
Krebsbach, P. H.,
Miyashita, T.,
Liebman, J.,
and Yamada, Y.
(1995)
DNA Cell Biol.
14,
501-510
26.
Galéra, P.,
Musso, M.,
Ducy, P.,
and Karsenty, G.
(1994)
Proc. Natl. Acad. Sci. U. S. A.
91,
9372-9376
27.
Galéra, P.,
Park, R. W.,
Ducy, P.,
Mattéi, M.-G.,
and Karsenty, G.
(1996)
J. Biol. Chem.
271,
21331-21339
28.
Benya, P. D.,
Padilla, S. R.,
and Nimni, M. E.
(1977)
Biochemistry
16,
865-872
29.
Takigawa, M.,
Koji, T.,
Hai-Ou, P.,
Enomoto, M.,
Kinoshita, A.,
Suzuki, F.,
Takano, Y.,
and Mori, Y.
(1989)
Cancer Res.
49,
3996-4002
30.
Chen, C.,
and Okayama, H.
(1987)
Mol. Cell. Biol.
7,
2745-2752
31.
Goldberg, H.,
Helaakoski, T.,
Garrett, L.-A.,
Karsenty, G.,
Pellegrino, A.,
Lozano, G.,
Maity, S.,
and de Crombrugghe, B.
(1992)
J. Biol. Chem.
267,
19622-19630
32.
Ala Kokko, L.,
Hyland, J.,
Smith, C.,
Kivirikko, K. I.,
Jimenez, S. A.,
and Prockop, D. J.
(1991)
J. Biol. Chem.
266,
14175-14178
33.
Van Dyke, W. M.,
Sirito, M.,
and Sawadago, M.
(1992)
Gene (Amst.)
111,
99-104
34.
Dignam, J. D.,
Lebowitz, R. M.,
and Roeder, R. G.
(1983)
Nucleic Acids Res.
11,
1475-1489
35.
Andrews, N. C.,
and Faller, D. V.
(1991)
Nucleic Acids Res.
19,
2499
36.
Maxam, A. M.,
and Gilbert, W.
(1980)
Methods Enzymol.
65,
499-559
37.
Chomczynski, P.,
and Sacchi, N.
(1987)
Anal. Biochem.
162,
153-159
38.
Boumediene, K.,
Félisaz, N.,
and Pujol, J.-P.
(1998)
Exp. Cell Res.
243,
173-184
39.
Fort, P.,
Marty, L.,
Piachekzyk, M.,
El Sabrouty, S.,
Dani, C.,
Jeanteur, P.,
and Blanchard, J.-M.
(1985)
Nucleic Acids Res.
13,
1431-1442
40.
Savagner, P.,
Miyashita, T.,
and Yamada, Y.
(1990)
J. Biol. Chem.
265,
6669-6674
41.
Vikkula, M.,
Metsäranta, M.,
Syvänen, A. C.,
Ala-Kokko, L.,
Vuorio, E.,
and Peltonen, L.
(1992)
Biochem. J.
285,
287-294
42.
Zhou, G.,
Lefebvre, V.,
Zhang, Z. P.,
Eberspaecher, H.,
and de Crombrugghe, B.
(1998)
J. Biol. Chem.
273,
14989-14997
43.
Bell, D. M.,
Leung, K. K. H.,
Wheatley, S. C.,
Ng, L. J.,
Zhou, S.,
Ling, K. W.,
Sham, M. H.,
Koopman, P.,
Tam, P. P. L.,
and Cheah, K. S. E.
(1997)
Nat. Genet.
16,
174-178
44.
Ng, L. J.,
Wheatley, S.,
Muscat, G. E. O.,
Conway-Campbell, J.,
Bowles, J.,
Wright, E.,
Bell, D. M.,
Tam, P. L. M.,
Cheah, K. S. E.,
and Koopman, P.
(1997)
Dev. Biol.
183,
108-121
45.
Marin, M.,
Karis, A.,
Visser, P.,
Grosveld, F.,
and Philipsen, S.
(1997)
Cell
89,
619-628
46.
Zhao, Q. I.,
Eberspaecher, H.,
Lefebvre, V.,
and de Crombrugghe, B.
(1997)
Dev. Dynam.
209,
377-386
47.
Benya, P. D.,
and Padilla, S. R.
(1986)
Dev. Biol.
118,
296-305
48.
Courey, A. J.,
Holtzman, D. A.,
Jackson, S. P.,
and Tjian, R.
(1989)
Cell
59,
827-838
49.
Sauer, F.,
and Jäckle, H.
(1991)
Nature
353,
563-566
50.
Sauer, F.,
and Jäckle, H.
(1993)
Nature
364,
454-457
51.
Seto, E.,
Shi, Y.,
and Shenk, T.
(1991)
Nature
354,
241-245
52.
Seto, E.,
Lewis, B.,
and Shenk, T.
(1993)
Nature
365,
462-464
53.
Shi, Y.,
Seto, E.,
Chang, L. C.,
and Shenk, T.
(1991)
Cell
67,
377-388
54.
Majello, B.,
De Luca, P.,
Suske, G.,
and Lania, L.
(1995)
Oncogene
10,
1841-1848
55.
Majello, B.,
De Luca, P.,
Hagen, G.,
Suske, G.,
and Lania, L.
(1994)
Nucleic Acids Res.
22,
4914-4581
56.
Majello, B.,
de Luca, P.,
and Lania, L.
(1997)
J. Biol. Chem.
272,
4021-4026
57.
Pujol, J.-P.,
and Loyau, G.
(1987)
Life Sci.
41,
1187-1198
58.
Chandrasekhar, S.,
Harvey, A. K.,
Higginbotham, J. D.,
and Horton, W. E.
(1990)
Exp. Cell Res.
191,
104-114
59.
Goldring, M. B.,
Fukuo, K.,
Birkhead, J. R.,
Dudek, E.,
and Sandell, L. J.
(1994)
J. Cell. Biochem.
54,
85-99
Copyright © 2000 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:
|
|
 |
|
  L. Maneix, G. Beauchef, A. Servent, Y. Wegrowski, F. X. Maquart, N. Boujrad, G. Flouriot, J. P. Pujol, K. Boumediene, P. Galera, et al. 17{beta}-Oestradiol up-regulates the expression of a functional UDP-glucose dehydrogenase in articular chondrocytes: comparison with effects of cytokines and growth factors Rheumatology, March 1, 2008; 47(3): 281 - 288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Poree, M. Kypriotou, C. Chadjichristos, G. Beauchef, E. Renard, F. Legendre, M. Melin, S. Gueret, D.-J. Hartmann, F. Mallein-Gerin, et al. Interleukin-6 (IL-6) and/or Soluble IL-6 Receptor Down-regulation of Human Type II Collagen Gene Expression in Articular Chondrocytes Requires a Decrease of Sp1{middle dot}Sp3 Ratio and of the Binding Activity of Both Factors to the COL2A1 Promoter J. Biol. Chem., February 22, 2008; 283(8): 4850 - 4865. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kypriotou, G. Beauchef, C. Chadjichristos, R. Widom, E. Renard, S. A. Jimenez, J. Korn, F.-X. Maquart, T. Oddos, O. Von Stetten, et al. Human Collagen Krox Up-regulates Type I Collagen Expression in Normal and Scleroderma Fibroblasts through Interaction with Sp1 and Sp3 Transcription Factors J. Biol. Chem., November 2, 2007; 282(44): 32000 - 32014. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Badariotti, M. Kypriotou, C. Lelong, M.-P. Dubos, E. Renard, P. Galera, and P. Favrel The Phylogenetically Conserved Molluscan Chitinase-like Protein 1 (Cg-Clp1), Homologue of Human HC-gp39, Stimulates Proliferation and Regulates Synthesis of Extracellular Matrix Components of Mammalian Chondrocytes J. Biol. Chem., October 6, 2006; 281(40): 29583 - 29596. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sironi, G. Menozzi, G. P. Comi, R. Cagliani, N. Bresolin, and U. Pozzoli Analysis of intronic conserved elements indicates that functional complexity might represent a major source of negative selection on non-coding sequences Hum. Mol. Genet., September 1, 2005; 14(17): 2533 - 2546. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Meech, D. B. Edelman, F. S. Jones, and H. P. Makarenkova The homeobox transcription factor Barx2 regulates chondrogenesis during limb development Development, May 1, 2005; 132(9): 2135 - 2146. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Chadjichristos, C. Ghayor, M. Kypriotou, G. Martin, E. Renard, L. Ala-Kokko, G. Suske, B. de Crombrugghe, J.-P. Pujol, and P. Galera Sp1 and Sp3 Transcription Factors Mediate Interleukin-1{beta} Down-regulation of Human Type II Collagen Gene Expression in Articular Chondrocytes J. Biol. Chem., October 10, 2003; 278(41): 39762 - 39772. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Matsuo, W. Yu-Hua, H. Sumiyoshi, K. Sakata-Takatani, H. Nagato, K. Sakai, M. Sakurai, and H. Yoshioka The Transcription Factor CCAAT-binding Factor CBF/NF-Y Regulates the Proximal Promoter Activity in the Human {alpha}1(XI) Collagen Gene (COL11A1) J. Biol. Chem., August 29, 2003; 278(35): 32763 - 32770. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Tan, H. Peng, M. Osaki, B. K. Choy, P. E. Auron, L. J. Sandell, and M. B. Goldring Egr-1 Mediates Transcriptional Repression of COL2A1 Promoter Activity by Interleukin-1beta J. Biol. Chem., May 9, 2003; 278(20): 17688 - 17700. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Luchetti, P. Paroncini, P. Majlingova, J. Frampton, M. Mucenski, S. S. Baroni, P. Sambo, J. Golay, M. Introna, and A. Gabrielli Characterization of the c-Myb-responsive Region and Regulation of the Human Type I Collagen alpha 2 Chain Gene by c-Myb J. Biol. Chem., January 10, 2003; 278(3): 1533 - 1541. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Zhang, S. A. Jimenez, and D. G. Stokes Regulation of Human COL9A1 Gene Expression. ACTIVATION OF THE PROXIMAL PROMOTER REGION BY SOX9 J. Biol. Chem., January 3, 2003; 278(1): 117 - 123. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Chadjichristos, C. Ghayor, J.-F. Herrouin, L. Ala-Kokko, G. Suske, J.-P. Pujol, and P. Galera Down-regulation of Human Type II Collagen Gene Expression by Transforming Growth Factor-beta 1 (TGF-beta 1) in Articular Chondrocytes Involves SP3/SP1 Ratio J. Biol. Chem., November 8, 2002; 277(46): 43903 - 43917. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Garcia-Ruiz, P. de la Torre, T. Diaz, E. Esteban, I. Fernandez, T. Munoz-Yague, and J. A. Solis-Herruzo Sp1 and Sp3 Transcription Factors Mediate Malondialdehyde-induced Collagen alpha 1(I) Gene Expression in Cultured Hepatic Stellate Cells J. Biol. Chem., August 16, 2002; 277(34): 30551 - 30558. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. R. Stenmark Cell-, age-, and phenotype-dependent differences in the control of gene expression Am J Physiol Lung Cell Mol Physiol, October 1, 2001; 281(4): L762 - L765. [Full Text] [PDF]
|
|
|
|
|
|
C. Ghayor, C. Chadjichristos, J.-F. Herrouin, L. Ala-Kokko, G. Suske, J.-P. Pujol, and P. Galera SP3 Represses the SP1-mediated Transactivation of the Human COL2A1 Gene in Primary and De-differentiated Chondrocytes J. Biol. Chem., September 28, 2001; 276(40): 36881 - 36895. [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 |