First published online 2 December 2004
doi: 10.1242/dev.01573
Development 132, 215-225 (2005)
Published by The Company of Biologists 2005
Smad1, ß-catenin and Tcf4 associate in a molecular complex with the Myc promoter in dysplastic renal tissue and cooperate to control Myc transcription
Ming Chang Hu1 and
Norman D. Rosenblum2,*
1 Program in Developmental Biology, Research Institute, The Hospital for Sick
Children, 555 University Avenue, Toronto, Ontario M5G 1X8, Canada
2 Program in Developmental Biology, Research Institute, The Hospital for Sick
Children and Division of Nephrology, Department of Paediatrics, University of
Toronto, 555 University Avenue, Toronto, Ontario M5G 1X8, Canada
View larger version (40K):
[in a new window]
|
Fig. 3. Associations of Smad1, ß-catenin and Tcf4 with the Myc promoter are
enhanced in TgAlk3QD kidney tissue. (A) Schematic
representation of mouse Myc promoter. The 1409 nucleotide region upstream of
the transcription start site is organized into Tcf-binding element (TBE),
TBE-A, consisting of three Tcf-binding consensus sequences, an adjacent
Smad-binding element (SBE), SBE-A, consisting of five Smad-binding consensus
sequences, and a second SBE, SBE-B, consisting of four Smad-binding consensus
sequences. (B) Results of ChIP. DNA amplified using region-specific primers is
shown above graphs demonstrating quantitation of corresponding amplified DNA
controlled for the amount of input DNA. Antibody controls using non-immune
sera are shown for each ChIP. Left panel: ChIP of TBE-A using anti-Tcf4
antibody showing increased association of Tcf4 with TBE-A in
TgAlk3QD kidney tissues versus wild type. Middle and right
panels: ChIP of SBE-A and SBE-B using anti-Smad1 antibody showing increased
association of Smad1 with SBE-A and SBE-B in TgAlk3QD
kidney tissue. (C, left panels) ChIP of SBE-A using anti-Tcf4 antibody showing
association of Tcf4 with SBE-A in TgAlk3QD kidney tissue
but not in wild-type tissue. (C, right panels) ChIP of SBE-B using anti-Tcf4
antibody. No association of Tcf4 with SBE-B was detected in either wild-type
or TgAlk3QD kidney tissue.
|
|
View larger version (64K):
[in a new window]
|
Fig. 1. TgAlk3QD mice exhibit medullary cystic renal dysplasia
and aberrant Myc expression. (A) Histological phenotype and Myc expression.
Histological analysis of 4 µm Hematoxylin and Eosin-stained kidney tissue
sections. Upper and middle panels: the renal medulla of wild-type mice is
characterized by a high density of closely apposed tubules without intervening
extracellular matrix. By contrast, the renal medulla of
TgAlk3QD mice is characterized by a heterogeneous
population of tubules of irregular shape and variable diameter. Cysts (C) with
greatly increased diameter and flattened epithelium are contrasted with normal
tubules (T). Lower panels: Myc was detected using an anti-Myc antibody.
Although Myc was barely detected in wild-type kidney, it was widely expressed
in a nuclear pattern in dysplastic renal tubules. (B) Association of acetyl
histone 4 (H4) with the RNA polymerase II core promoter using ChIP. The
location and characteristics of the RNA polymerase II core promoter including
the TATA box and TFII recognition elements (BRE) in the murine Myc
gene are shown in schematic form. Left lower panel: ChIP using an
anti-acetyl-histone 4 (H4) and kidney tissue isolated from wild-type or
TgAlk3QD kidney tissue demonstrated increased
amplification of the 103 nucleotide core promoter region in dysplastic (QD)
tissue. Lower right panel: quantitation of DNA amplified after ChIP,
demonstrating a 1.6-fold increase in acetyl-H4 association with the core
promoter amplified from TgAlk3QD tissue. The amount of DNA
amplified was controlled for the amount of input DNA. P<0.05,
n=3 independent experiments.
|
|
View larger version (47K):
[in a new window]
|
Fig. 2. Nuclear association of Smad1/ß-catenin/Tcf4 is increased in
TgAlk3QD kidney tissue. (A) Cellular localization in
wild-type and cystic TgAlk3QD kidney tissue.
Immunofluorescence imaging of 4 µm tissue sections was performed after
incubation with primary antibodies and fluorescein- or rhodamine-conjugated
secondary antibodies. Left panels: co-localization of Tcf4 and Smad1. In
tissue isolated from wild-type mice, Tcf4 and PSmad1 were expressed in a
cytoplasmic pattern and showed little co-localization (merge image). By
contract, in tissue isolated from TgAlk3QD mouse kidneys,
Tcf4 and P-Smad1 were expressed in a nuclear pattern with co-localization in
the majority of imaged cells (yellow). Right panels: co-localization of
ß-catenin and Tcf4. In tissue isolated from wild-type mice,
ß-catenin is expressed in a cytoplasmic pattern and demonstrates little
co-localization with Tcf4. By contrast, in tissue isolated from
TgAlk3QD mouse kidney, ß-catenin is expressed in a
nuclear pattern and co-localizes with Tcf4 in many cells (yellow). (B)
Detection of molecular complexes in cytoplasmic and nuclear fractions.
Proteins derived from cytoplasmic (actin-positive) and nuclear (acetyl
H4-positive) fractions were submitted to immunoprecipitation and
immunoblotting with specific antibodies. Quantitation of immunoblotted
proteins controlled for the quantity of immunoprecipitated protein is shown.
Molecular complexes consisting of ß-catenin and P-Smad1 were detected in
increased amounts in both cytoplasmic and nuclear fractions of
TgAlk3QD kidney tissue. Molecular complexes consisting of
ß-catenin and Tcf4 were increased in the nuclear fraction of
TgAlk3QD tissue. (C) Association of Smad1, ß-catenin
and Tcf4 with chromatin isolated from kidney tissue. Protein associations with
nuclear DNA were assessed using cisplatin crosslinking and
immunoprecipitation/immunoblotting. Molecular complexes consisting of
ß-catenin and Smad1 and of Tcf4 and Smad1 were detected in increased
quantities in TgAlk3QD versus wild-type kidney
tissues.
|
|
View larger version (46K):
[in a new window]
|
Fig. 4. Association of Smad1/Tcf4/ß-catenin molecular complexes with
oligo-duplexes encoding either TBE-A or SBE-A. EMSA showing inhibition of
migration generated by incubating nuclear extracts isolated from wild-type or
TgAlk3QD (QD) kidney tissue with 32P-labeled
oligo-duplexes corresponding to wild-type and mutant consensus binding regions
within TBE-A and SBE-A (broken arrow). Unbroken arrows mark enhanced
inhibition of migration associated with addition of specific antibodies. Free
probe (FP) is observed at the lower end of each autoradiogram. (A) Left panel:
migration of 32P-labeled oligo-duplex, TBE-A, was retarded by
addition of nuclear extract (broken arrow). The amount of oligo-duplex bound
by extract from QD was greater than that from wild type. Addition of
specific antibody caused a further migratory retardation (unbroken arrow).
Association of Tcf4 was detected in both wild type and QD but at much
higher levels in QD. Association of ß-catenin was detected at
low levels in wild type and at much higher levels in QD. Association
of Smad1 with TBE-A sequences was observed only in QD. The mobility
of the supershifted band was greater in QD lysates compared with
wild-type lysates. Right panel: specificity of TBE-A is demonstrated by
failure to detect any specific retardation using a mutant version of TBE-A.
(B) The specificity of DNA-protein interactions (broken arrow) is shown by
progressive diminution of binding in the presence of tenfold and 100-fold
excess of unlabeled oligo-duplex. (C) The specificity of antibody-mediated
supershifts (unbroken arrow) is shown by the absence of a supershift in the
presence of non-immune mouse or rat antisera. (D) Left panel: migration of
32P-labeled oligo-duplex, SBE-A, was retarded by addition of
nuclear extract (broken arrow). The amount of oligo-duplex bound by extract
from QD was greater than that from wild type. Addition of specific
antibody caused a further migratory retardation (unbroken arrow). Association
of Smad1 with SBE-A was detected in both wild type and QD with
greater amounts detected in QD. Association of ß-catenin was
detected in wild type and at much higher levels in QD. Association of
Tcf4 was detected only in QD. The mobility of the supershifted band
was greater in QD lysates compared with wild-type lysates. Right
panel: specificity of SBE-A is demonstrated by failure to detect any specific
retardation using a mutant version of SBE-A. (E) The specificity of
DNA-protein interactions (broken arrow) is shown by progressive diminution of
binding in the presence of tenfold and 100-fold excess of unlabeled
oligo-duplex. (F) The specificity of antibody-mediated supershifts (unbroken
arrow) is shown by the absence of a supershift in the presence of non-immune
mouse or rat antisera.
|
|
View larger version (24K):
[in a new window]
|
Fig. 5. Treatment of mIMCD-3 cells with Bmp2 increases ß-catenin/P-Smad1
molecular complexes and Myc expression. Immunoblots and quantitation are shown
on the left and right panels, respectively. Treatment with recombinant 10 nM
Bmp2 increased intracellular levels of ß-catenin 1.9-fold (A),
ß-catenin/P-Smad1 molecular complexes 2.3-fold (B) and Myc 2.4-fold (C).
(D) Bmp2 treatment of mIMCD-3 cells transfected with a plasmid encoding the
Myc promoter (-1490 to -1) upstream of luciferase increased luciferase
activity 10-fold by 1 hour and 26-fold by 2 hours after addition of Bmp2.
|
|
View larger version (29K):
[in a new window]
|
Fig. 6. Tcf4, ß-catenin and Smad1 positively regulate Myc. (A,B) Effect of
RNAi-mediated decrease in Tcf4, ß-catenin or Smad1 on endogenous Myc mRNA
and protein in mIMCD-3 cells. In control cells transfected with an empty
plasmid, Bmp2 significantly increased levels of endogenous Myc mRNA and
protein. (A,B) Knock-down of Tcf4 or ß-catenin decreased basal levels of
Myc mRNA and protein, and totally blocked Bmp2-mediated increases in these
species. Knock-down of Tcf4 exerted a larger inhibitory effect on basal
levels. Knock-down of Smad1 did not exert a significant effect on basal mRNA
and protein levels but did partially block the Bmp2 mediated increase observed
in controls. (C) Effect of RNAi-mediated decrease in Tcf4, ß-catenin or
Smad1 on luciferase activity downstream of the -1490 to -1 segment of the Myc
promoter. In controls transfected with an empty RNAi-inducing plasmid, Bmp2
markedly induced luciferase activity. Knock-down of Tcf4 significantly
decreased basal luciferase activity and totally abrogated the Bmp2-dependent
increase in luciferase activity. Knock-down of either ß-catenin or Smad1
exerted no significant effect on luciferase activity under basal conditions.
Knock-down of ß-catenin blocked the Bmp2-dependent increase. By contrast,
the response to Bmp2 was limited to a partial but significant response in
cells with a quantitatively similar reduction in Smad1 levels. (D) Role of
TBE-A and SBE-A in Bmp2-dependent induction of Myc luciferase. mIMCD-3 cells
were transfected with plasmid DNA encoding luciferase downstream of either
wild-type or mutant forms of the Myc promoter (-1490 to -1). Under basal
conditions (no Bmp2 treatment), luciferase activity was significantly
decreased under the control of the mutant TBE-A and significantly increased
downstream of the mutant SBE-A. Mutant TBE-A significantly decreased
Bmp2-dependent induction of luciferase activity. Mutant SBE-A also decreased
Bmp2-dependent induction of luciferase activity but not to the same extent as
mutant TBE-A.
|
|
View larger version (22K):
[in a new window]
|
Fig. 7. Model of Smad/ß-catenin signaling in normal and dysplastic renal
tissues. In the absence of Bmp-stimulated Smad1/ß-catenin interactions,
Myc expression is inhibited by Smad1 bound to Smad-binding regions. By
contrast, in dysplastic tissue, formation of Smad1/ß-catenin/Tcf4
molecular complexes stimulates Myc transcription via mechanisms to be
defined.
|
|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
© The Company of Biologists Ltd 2005
