QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 25 August 2004
doi: 10.1242/jcs.01337

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Monteiro, R. M. Articles by Mummery, C. L. Search for Related Content PubMed PubMed Citation Articles by Monteiro, R. M. Articles by Mummery, C. L.

Spatio-temporal activation of Smad1 and Smad5 in vivo: monitoring transcriptional activity of Smad proteins

¹ Hubrecht Laboratory, Netherlands Institute for Developmental Biology, Uppsalalaan 8, 3584 CT Utrecht, The Netherlands
² Division of Cellular Biochemistry, The Netherlands Cancer Institute, Plesmalaan 121, 1066 CX Amsterdam, The Netherlands
³ Department of Rheumatology, Thurston Arthritis Research Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA

View larger version (38K): [in a new window]   Fig. 1. Specificity of the BMP response element (BRE) to Smad1/5-mediated BMP signaling in vitro. (A) The BRE sequence consists of two repeats of 2xSBE (bold), 1xCGCC (underlined), 2xCAGC (bold, italic) and 1xGGCGCC (double underlined). 2C2P2C is a similar sequence that lacks 2xSBE + 1xCGCC binding sites and does not respond to BMPs in vitro (see Korchynskyi and ten Dijke, 2002). (B) In C2C12 cells, the BRE-luc reporter is specifically upregulated by BMP2, 4, 6 and 7 (100 ng/ml). 10 ng/ml of TGFß, EGF or FGF do not activate reporter transcription. (C) BRE-luc reporter activity is induced by overexpression of either Smad1 or Smad5 and to a lower extent, by Smad8 in C2C12 cells. BRE-luc reporter activity is further increased upon addition of BMP6 to control cells or cells overexpressing Smad1, 5 or 8. (D) Overexpression of Smad1+Smad4 or Smad5+Smad4 is sufficient to induce BRE-luc activity in C2C12 cells. Smad4 alone does not influence inducibility by BMP6 (100 ng/ml). (E) Binding of Smad4 to the BRE sequence is critical for the reporter response to ligand-dependent and ligand-independent Smad1 or Smad5 overexpression. In the presence of the Smad4-D4 DNA-binding mutant (harboring the K81R and R88K mutations), neither Smad1 nor Smad5 are able to drive BRE-luc expression in MDA-MB468 cells. Expression of wild-type Smad4 is sufficient to restore reporter inducibility either in response to BMP6 or to co-expressed Smad1 or Smad5. (F) In HepG2 cells, constitutively active (ca) type I receptors caALK1, caALK2, caALK3 and caALK6 specifically induce BRE-luc reporter activity, while caALK4 or caALK5, do not. (G) Specific modulation of the BRE-luc reporter in response to activation or inhibition of BMP signaling in ES cells. 10 ng/ml of BMP4 and caALK3 induce while Smad7 blocks basal and BMP induced reporter activity in transient assays. As expected, the 2C2P2C-luc reporter showed no response to modulation. **P0.005; *P0.05; #P0.1.

View larger version (53K): [in a new window]   Fig. 2. Response of stable ES reporter cell lines to BMP4. (A) Constructs used to generate transgenic ES cells. BRE-ß-galactosidase or BRE-luc reporter constructs and a PGK-hygromycin cassette were coelectroporated in wild-type ES cells. Single colonies were picked from selection medium (150 µg/ml hygromicin) and genotyped by PCR on genomic DNA. Transgenic ES cell lines were assayed for BMP-induced reporter expression. Basal reporter activity was observed in (B) unstimulated BRE-lac1 ES cells. (C) Reporter activity increased in response to 20 ng/ml BMP4. Bars, 50 µm (B,C). Activity of the reporter gene in ES lines BRE-lac1 (D), BRE-lac2 (E) and BRE-luc (F) increases with increasing concentrations of BMP4. Western blots for Id1 (anti-Id1 antibody) (G) and for activated Smad1/5/8 (PSmad1/5/8, PS1 antibody) (H) were performed on the same extracts as in D and E; expression profile of these proteins show good correlation with the reporter activity. Note that the middle band in the triplet corresponds to phosphorylated Smad1/5/8 (black arrowheads). 10 µg/ml of protein was loaded in each lane and confirmed by Ponceau S staining. Nonspecific bands are shown as loading control (l.c.). The results are represented as the mean of three independent experiments (mean±s.d.).

View larger version (28K): [in a new window]   Fig. 3. ES cells express active BMP4 and 7 but not BMP2. (A) RT-PCR on cDNA from wild type and transgenic ES cell lines BRE-luc, BRE-lac1 and BRE-lac2. All the lines tested express BMP4 and BMP7, but not BMP2. Using the same primer sets, transcripts for BMP2, 4 and 7 were detected by RT-PCR on cDNA from E9.5 mouse hearts, as expected (results not shown). ß-actin was a positive control for the reverse transcription reaction; expression of Oct4 indicates the undifferentiated state of the ES cells. (B) Transactivation assay in HepG2 cells. Medium conditioned by ES cells induces BRE-luc reporter activity in HepG2 cells. Non-conditioned medium had no effect. As a positive control, 10 ng/ml BMP4 was added to transfected cells to activate the BRE-luc reporter. As a negative control, HepG2 cells were transfected with the BRE-luc reporter and no BMP was added. Results are presented as mean fold induction (±s.d.) of triplicates in relation to unstimulated cells, set to 1 (*P0.05).

View larger version (23K): [in a new window]   Fig. 4. Autocrine BMP pathway in ES cells. Noggin and BMP were added to reporter ES cell lines (A) BRE-luc and (B) BRE-lac2 at different concentrations to assess the reporter activity in response to activation/inhibition in vitro. The same results were obtained in BRE-lac1 ES cells (results not shown). (C) Id1 protein levels in a western blot correlated with the reporter activity in BRE-lac2 protein extracts. Equal amounts of protein (10 µg) were loaded in each lane and confirmed with Ponceau S staining; nonspecific bands are shown as loading control (l.c.). Results are presented as the mean of triplicate experiments (mean±s.d.).

View larger version (191K): [in a new window]   Fig. 5. Teratomas derived from reporter ES cells in adult male mice. At 4 weeks, ES cells injected in the testis capsule of syngenic mice (129 OLA) give rise to different tissues. Compare the normal testis tissue in A with a representative sample of ES-derived tissues in B, C and D. Bone (b), cartilage (ca), striated muscle (s), adipose tissue (a) and epithelial tissue (ep, red arrow) were found (B,C). ß-Galactosidase staining of BRE-lac1-derived teratomas showed staining in areas of cartilage and perichondrium (D, black arrowhead). Note that not all the chondrocytes show ß-galactosidase staining, suggesting regional differences in activation of Smad1/5 signaling in chondrocytes. Sections (7 µm) were stained either with Eosin/Hematoxilin (A-C) or Neutral Red and ß-galactosidase (D). Abbreviations: it, interstitial tissue; sp, spermatidia. Bar, 100 µm.

View larger version (103K): [in a new window]   Fig. 6. Analysis of reporter expression in transgenic E9.5 embryos. (A) Chimeric embryos were generated with the BRE-lac1 ES cell line by injection into blastocysts of ß-actin-GFP mice and analyzed at E9.5. ß-galactosidase staining was present in the heart, the optic vesicle (op), the ventral forelimb (l), neural crest (nc, blue arrowhead) and in the branchial arches (ba). Adult male chimeras generated with BRE-lac1 and BRE-lac2 ES lines were crossed with wild-type females and the progeny analyzed at E9.5. BRE-lac1 (B) and BRE-lac2 (C) transgenic embryos showed reporter expression pattern similar to that in transient chimeras. ß-Galactosidase staining was also detectable in the roof of the midbrain neuroepithelium (green arrowheads), in the hindbrain, where the region around rhombomere 5 showed very low ß-Galactosidase staining, in the posterior mesenchyme and in the aorta (white arrows). (D-E) Transversal sections through the heart of the embryo shown in A. ß-Galactosidase staining is found in the outflow tract (ot), the bulbus cordis (bc) and the ventricle (v), as well as in the atrioventricular canal (avc, red arrowhead). (F) In the optic vesicle, reporter expression is restricted to the dorsal side, while low expression was observed in the overlying ectoderm. (G,H) Transverse sections through the heart of BRE-lac2 transgenic embryos showed reporter expression in endocardium, pericardium and myocardium, with stronger expression in the outflow tract and the atrioventricular canal. Frontal (I) and lateral (J) views of a transgenic embryo showing ß-galactosidase expression in the cardiac crescent and in the amnion at E7.5. (K-N) Co-staining for phosphorylated Smad1/5/8 (PS1 antibody) and ß-galactosidase in E9.5 transgenic embryos. ß-Galactosidase expression coincides with PS1 staining in the heart and in the foregut (K,L), while in the neural tube and neural crest some cells are stained for PS1 and not ß-galactosidase (M, black arrowheads). High correlation between phosphorylated Smad1/5/8 and ß-galactosidase was observed in the branchyal arches (N). (O) Quantification of the incidence of ß-galactosidase and PS1 staining. Sections containing posterior body wall (pbw), heart and foregut (h+f), neurectoderm (nt) and branchyal arches (ba) were scored for double or single ß-galactosidase and PS1 staining. Results are presented as the mean of the percentage of single or double stained versus total of stained cells within three sections per count. Bars, 500 µm (A-C,I-J) and 100 µm (D-H,K-N). Abbreviations: a, atrium; am, amnion; as, aortic sac; cc, cardiac crescent; f, foregut; l, forelimb bud; p, pericardium; r5, rhombomere 5.