Skip Navigation

Cardiovascular Research 2007 74(2):279-289; doi:10.1016/j.cardiores.2006.09.014
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Yao, Y.
Right arrow Articles by Boström, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yao, Y.
Right arrow Articles by Boström, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2006, European Society of Cardiology

Activin-like kinase receptor 1 (ALK1) in atherosclerotic lesions and vascular mesenchymal cells

Yucheng Yaoa, Amina F. Zebboudja, Alejandra Torresa, Esther Shaoa,b and Kristina Boströma,b,*

aDivision of Cardiology, David Geffen School of Medicine at UCLA, Los Angeles, CA 90095-1679, United States
bMolecular Biology Institute, UCLA, Los Angeles, CA 90095-1570, United States

* Corresponding author. Division of Cardiology, David Geffen School of Medicine at UCLA, Box 951679, Room BH-307 CHS, Los Angeles, CA 90095-1679, United States. Tel.: +1 310 794 4417; fax: +1 310 206 8553. Email address: kbostrom{at}mednet.ucla.edu

Received 25 May 2006; revised 19 September 2006; accepted 20 September 2006


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Objective: Activin-like kinase receptor 1 (ALK1) is a transforming growth factor (TGF)-β type I receptor expressed in vascular mesenchyme, yet its function in vascular mesenchymal cells (VMC) is unclear. We examined ALK1 expression in human coronary atherosclerotic lesions and bovine and human VMC undergoing cellular condensation in vitro. We also examined the effect of activated ALK1 on cell proliferation and smooth muscle cell (SMC) differentiation.

Methods and results: Our results showed that ALK1 was expressed in human coronary atherosclerotic lesions as determined by immunohistochemistry. ALK1 was also expressed in cellular condensations of bovine and human VMC as determined by real-time PCR and immunocytochemistry. Bone morphogenetic protein (BMP)-2, which is known to increase condensation size, increased ALK1 expression when induced from a BMP-2 adenoviral vector. In turn, activated ALK1 induced expression of matrix GLA protein (MGP), a BMP-2 inhibitor known to limit condensation size. Activated ALK1 enhanced proliferation of VMC as determined by 3H-thymidine incorporation, whereas MGP decreased proliferation. Activated ALK1 also enhanced expression of SMC lineage markers and ALK5, another TGF-β type I receptor, as determined by immunoblotting, real-time PCR and immunocytochemistry. Anti-TGF-β antibodies abolished expression of SMC markers in the presence of constitutively active ALK1, suggesting that ALK1 activation alone is not sufficient to promote SMC differentiation.

Conclusions: We conclude that there is a balance between the actions of BMP-2 and MGP in the initiation of vascular mesenchymal cell condensation and SMC differentiation, and that targeting ALK1, BMP2 and/or MGP may lead to novel concepts of atherosclerosis treatment.

KEYWORDS Activin-like kinase receptor 1; Bone morphogenetic protein; Matrix GLA protein; Vascular mesenchymal cells; Smooth muscle cell differentiation


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
Activin-like kinase receptor 1 (ALK1) is a TGF-β type I receptor that is important in endothelial cell biology, where it has been extensively studied in association with angiogenesis [1]. It activates the Smad1/5/8 pathway similar to BMP-receptors even though its only known ligands are TGF-β1 and -β3 [2]. ALK1 deficiency results in defective recruitment of smooth muscle cell (SMC) precursors to the vascular media and arterio-venous malformations resembling human hereditary hemorrhagic telangiectasia (HHT) [3,4]. It has been suggested that poor ALK1 signaling in the endothelium leads to insufficient amounts of TGF-β in the vicinity of SMC precursors thereby decreasing recruitment [4]. However, ALK1 expression has also been detected in vascular and pulmonary mesenchyme [5,6], yet its function in mesenchymal cells is unclear. ALK5, the first identified TGF-β type I receptor, activates the Smad2/3 pathway and has been associated with resolution of angiogenesis [1]. There is also evidence supporting a role for ALK5 in SMC differentiation [7].

Bone morphogenetic protein-2 (BMP-2) is an osteoinductive factor expressed in atherosclerotic lesions, where it may contribute to vascular calcification [8,9]. However, its role in vascular biology is incompletely understood. It has been shown to elicit chemotaxis of vascular SMC in a concentration- and time-dependent manner [10]. Furthermore, BMP-2 has been shown to inhibit SMC proliferation induced by serum, PDGF and low-density lipoproteins [11–13], and to induce SMC differentiation in multipotent progenitor cells [14].

Matrix GLA protein (MGP) is a {gamma}-carboxylated matrix protein expressed at high levels in atherosclerotic lesions [15–17]. Others and we have shown that MGP is a BMP-2 inhibitor [21,22], which is a possible explanation for its inhibition of arterial calcification in MGP null mice [18] and of arterial and valvular calcification after warfarin-intake [19,20].

Multipotent vascular mesenchymal cells (VMC, also referred to as calcifying vascular cells or CVC) are subpopulations of SMC that have been used as a model for atherosclerotic lesion cells and vascular calcification. They spontaneously form condensations and nodules and undergo differentiation along different lineages, including the SMC, osteogenic, and chondrogenic lineages [23,24]. VMC express BMP-2 and MGP at baseline, and we have previously shown that the relative levels of BMP-2 and MGP are important for condensation and nodule size as well as the overall pattern formation in VMC [25,26]. Both BMP-2 and MGP promote nodule formation but BMP-2 increases nodule size in VMC, whereas MGP limits it [25,26]. A mathematical model based on molecular morphogens interacting in a reaction-diffusion process has been shown to predict the effect of BMP-2 and MGP [26].

Since recruitment and proliferation of immature mesenchymal cells occur as part of atherogenesis, we hypothesized that ALK1 is induced in atherosclerotic lesions in vivo. We further hypothesized that ALK1 is induced in vascular mesenchymal cells involved in atherosclerotic lesion in vitro, and we defined a regulatory pathway that include BMP-2, ALK1 and MGP in these cells.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
2.1 Immunohistochemistry of human coronary atherosclerotic lesions
The examined arteries were autopsy specimens from nine different patients. We did not obtain any identifying or clinical information about the patients as per the regulations of the Internal Review Board of the University of California, Los Angeles, which conform to the guidelines of the National Institute of Health (NIH). We examined nine human coronary atherosclerotic lesions, 3–6 segments of each lesion, in total 36 segments. Tissues were fixed in 4% paraformaldehyde and immunohistochemistry was performed as previously described [23] using specific antibodies to ALK1 (Santa Cruz Biotechnology). Permeabilization of the tissue with 0.1 mg/ml of Proteinase K for 15 min was performed prior to immunohistochemistry. ABC reagents (Vector Laboratories) were used to detect antibody binding, and hematoxylin was used for counterstaining.

For immunocytochemistry of VMC condensations and nodules, the cells were cultured for indicated time periods in 12-well culture plates and fixed for 30 min in 4% paraformaldehyde. Immunocytochemistry was performed using the same protocol and ALK1 antibodies as for immunohistochemistry. In addition, antibodies to ALK5 (R&D Systems), SM-{alpha}-actin (BioGenex) and SM-myosin heavy chain (SM-MHC) (Biomedical Technologies) were used.

2.2 Cell culture
Bovine and human vascular mesenchymal cells (VMC) were cultured as previously described [23–25]. Bovine VMC were used between passages 8–20, and human VMC between passages 3–10. Fetal bovine serum (10% and 15% for bovine and human VMC respectively) was present at all times including cell condensation studies and cell proliferation experiments. For experiments and for adenoviral transfection, CVC were plated at approximately 70–80% confluency. The Amaxa Nucleofector® and the human AoSMC Nucleofector® kit were used for transfection by electroporation of the VMC. The number of cells per electroporation and Nucleofector settings was optimized as per manufacturer's instructions. Half a million of cells and 0– 5 µg of specific plasmid DNA were used per well in 6-well culture plates. The total amount of plasmid DNA was kept constant at 5 µg per well in 6-well plates by addition of empty plasmid (pcDNA3.1, Invitrogen). The transfection efficiency was 20–25% as determined by transfection of an expression vector for Green Fluorescent Protein (GFP). BMP-2, Noggin and neutralizing anti-TGF-β antibodies were obtained from R&D Systems. Human MGP was added in the form of conditioned medium, and the methods to prepare the media and determine the level of MGP have been described previously [25]. For preparation, we used an expression vector for human N-terminally FLAG-tagged MGP (see Vector Constructions), which also enabled us to visualize MGP in the medium by immunoblotting with anti-FLAG antibodies (compare Figs. 3 and 5Go). In experiments where conditioned medium was used, the level of conditioned medium was kept constant at 80% using sham-conditioned medium as previously described [25]. The reagents were added to the VMC at the time of plating unless otherwise stated, and the medium was renewed every 3–4 days.


Figure 3
View larger version (35K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Activated ALK1 induces expression of MGP. (A) Bovine VMC infected with inducible BMP-2 or empty control adenoviral vectors were treated with doxycycline (500 ng/ml) for 5 days. MGP expression was determined in BMP-2 expressing and control cells after the indicated amount of time by real-time PCR and normalized to GAPDH. (B) VMC were transfected with an expression vector for constitutively active (ca) ALK1 (0–5 µg/well, 6-well plates). Transfection of caALK1 was confirmed by real-time PCR and normalized to GAPDH, and visualized by immunoblotting using anti-ALK1 antibodies. (C) MGP expression was determined by real-time PCR and normalized to GAPDH. (D) VMC were transfected with an expression vector for human FLAG-tagged MGP (0–5 µg/well, 6-well plates). Transfection of FLAG-MGP was confirmed by immunoblotting using anti-FLAG antibodies. ALK1 expression was determined by real-time PCR and normalized to GAPDH. Asterisks indicate statistically significant differences compared to the first time point or control. (*p<0.05, **p<0.001, ***p<0.0001, Scheffe's test).

 

Figure 5
View larger version (48K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Activated ALK1 affects VMC proliferation. Bovine VMC were transfected with plasmids for caALK1, FLAG-MGP or empty plasmid as indicated (DNA levels corresponded to 5 µg per well in a 6-well plate). After attachment, the cells were treated with BMP-2 (100 ng/ml) and Noggin (300 ng/ml), and 3H-thymidine was added. After 4 days, the cells were harvested and the 3H-activity was determined. Increased cellular levels of ALK1 and media levels of FLAG-MGP were confirmed by immunoblotting with anti-ALK1 and anti-FLAG antibodies. Asterisks indicate statistically significant differences compared to the first sample in each triplet of samples. (*p<0.05, **p<0.001, ***p<0.0001, Scheffe's test).

 
2.3 RNA analysis
Total RNA was isolated from cultured cells using the RNeasy kit (Qiagen). Real-time PCR assays were performed as previously described [27]. The following primers and probes were used: human ALK1 (hALK1) forward (F) (5'-AGGGCAAACCAGCCATTG-3'), hALK1 reverse (R) (5'-GGTTGCTCTTGACCAGCACAT-3'), hALK1 Taqman probe (FAM-CACCGCGACTTCAAGAGCCGC-TAMRA, bovine MGP (bMGP) F (5'-GGGAAGCTTGTGATGACTTCAAA-3'), bMGP R (5'-CGCTGCCGGTCGTAGGCAGCATTGTATCCA-3'), bMGP Taqman probe (FAM-TTGCGAACGCTATGCCATGGTGT-TAMRA), bovine GAPDH (bGAPDH) F (5'-GGCGCCAAGAGGGTCAT-3'), bGAPDH R (5'-GTGGTTCACGCCCATCACA-3'), bGAPDH Taqman probe (FAM-TCTCTGCACCTTCTGCCGATGCC-TAMRA), human ALK5 (hALK5) F (5'-TGTCATTGCTGGACCAGTGTG-3'), hALK5 R (5'-CAGTGCGGTTGTGGCAGATATA-3'), hALK5 Taqman probe (FAM-TTCGTCTGC ATCTCACTCATGTTGATGGT-TAMRA). All primer pairs were tested for consistency over a wide range of concentrations (serial dilutions 100–10–7).

2.4 Construction and infection with the doxycycline-inducible BMP-2 adenoviral vector
The recombinant BMP-2 adenoviral vector was constructed using the Adeno-XTM Expression System (Clontech). Full-length human BMP-2 cDNA was inserted in the pTRE-Shuttle2 vector downstream of the Tet-responsive element (TRE) and the CMV promoter. The resulting expression cassette was excised and ligated into the pre-digested Adeno-X viral DNA as per the manufacturer's protocol. The viral DNA was linearized, transfected into HEK293 cells using Superfect (Qiagen), and recombinant BMP-2 adenovirus crude lysates were harvested and analyzed as per manufacturer's instructions. Amplification and purification of viral particles were performed by Viraquest Inc.

The optimal infection efficiency was determined to be 100 multiplicity of infection (moi) with cell densities of 3x105 and 5x105 for VMC using a GFP adenovirus and the manufacturer's protocol. Adenoviral BMP-2 expression required co-infection with a Tet-On containing virus (Clontech). We determined that a 2:3 (Tet-On:BMP-2 adenovirus) ratio in combination with 500 ng/ml of doxycycline yielded efficient induction of BMP-2 with cell toxicity of <5% as determined by the trypan blue exclusion test [28]. Cells floating in the medium were included in the test.

2.5 Immunoblotting
Immunoblotting was performed as previously described [27]. Equal amounts of cellular protein or equal volumes of culture medium were used. Blots were incubated with specific antibodies to BMP-2 (0.4 µg/ml; N-14), P-Smad1 (0.4 µg/ml), Smad1 (0.4 µg/ml), ALK1 (0.4 µg/ml; D-20) (all from Santa Cruz Biotechnology), FLAG (2.5 µg/ml; Sigma), SM-{alpha}-actin (1 µg/ml; BioGenex), calponin (0.9 µg/ml; DakoCytomation), SM-MHC (6.7 µg/ml; Biomedical Technologies), ALK5 (0.3 µg/ml; R&D Systems), and β-actin (0.6 µg/ml; Sigma). Quantification of BMP-2 in culture medium was performed using an ELISA from R&D Systems.

2.6 3H-thymidine incorporation
Cells were seeded in 24-well plates at a density of 100,000 cells per well, and allowed to attach for 4–6 h. 3H-thymidine was added at 1 µCi/ml for 4 days, and 3H-thymidine incorporation was determined as previously described [29].

2.7 Vector constructions
The expression vector pcDNA-ALK1(Q201D)/caALK1 contains the constitutively active ALK1 in the expression vector pcDNA 3.1 (Invitrogen) and was provided by Karen Lyons, University of California, Los Angeles. To construct the expression vector pcDNA-N-FLAG-hMGP, which contains FLAG-tagged human MGP in the expression vector pcDNA3.1, a FLAG-tag was placed in the N-terminus of the secreted, mature protein by subcloning of a synthesized FLAG-coding DNA fragment between the coding regions for the signal peptide and the mature protein. The FLAG-containing hMGP DNA fragment was amplified by PCR and subcloned into pcDNA3.1(+) (Invitrogen) using restriction sites in the primers. The FLAG-tagged MGP was similar to non-tagged MGP in inhibiting BMP-2 activity as determined by BMP-2 responsive luciferase reporter gene assays (data not shown). The expression vector pcDNA3.1 without inserted cDNA was used as empty plasmid in transfection experiments.

2.8 Statistics
Data was analyzed for statistical significance by ANOVA with post-hoc Scheffe's analysis, unless otherwise stated. The analyses were performed using StatView, version 4.51 (Abacus Concepts). Experiments that involved adenoviral vectors were repeated three times. Experiments that involved no treatment, or treatment or transfection of cells were repeated four or more times. Immunocytochemistry was repeated on three different batches of bovine VMC and two different batches of human VMC. Immunohistochemistry of the atherosclerotic lesions was performed at least twice on serial sections.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
3.1 Expression of ALK1 is increased in atherosclerotic lesions
Since proliferation and recruitment of immature mesenchymal cells are part of atherogenesis, we examined ALK1 expression in human coronary atherosclerotic lesions using immunohistochemistry. We were unable to obtain a completely normal coronary artery. However, minimal ALK1 expression was observed in apparently disease-free segments of diseased coronaries except in the adventitia (Fig. 1A and B). In lesions, ALK1 staining was detected in the neointima and in scattered cells in the media (Fig. 1C and D). It was also detected in the coronary endothelium (Fig. 1E) and in areas of the shoulder region that appeared to be sites of neoangiogenesis (Fig. 1F). In advanced lesions, diffuse ALK1 staining was detected in the core of the lesions, in the vicinity of lipid and in parts that appeared to undergo cellular organization (Fig. 1G and H).


Figure 1
View larger version (156K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 ALK1 is expressed in human coronary atherosclerotic lesions. Atherosclerotic lesions were stained by immunohistochemistry using anti-ALK1 antibodies (A, C, E, F, G) or non-immune control antibodies (B, D, H) (original magnification 10x). Minimal ALK1 expression was observed in disease-free segments except in the adventitia (Fig. 1A and B). ALK1 staining was detected in the neointima and in scattered cells in the media (Fig 1C and D). It was also detected in the coronary endothelium (Fig. 1E) and in areas of the shoulder region (Fig. 1F). Diffuse ALK1 staining was detected in the core of advanced lesions and in areas that appeared to undergo cellular organization (Fig. 1G and H). Lu, lumen; m, media; neo, neointima.

 
3.2 BMP-2 induces expression of ALK1 in VMC
The VMC are multipotent cells that form cellular condensations and nodules and have been used as a model for atherosclerotic lesion cells [23,24,26]. To examine if bovine VMC express ALK1, we allowed the cells to spontaneously form condensations over 12 days. Cellular RNA was prepared at different stages and ALK1 expression was assessed by real-time PCR and normalized to GAPDH. The results showed that expression of the ALK1 increased about 3-fold (2.8±0.6) during condensation (Fig. 2A). To determine if ALK1 was specifically expressed in the condensations, we allowed the cells to form condensations for 7 days and then visualized expression of ALK1 by immunocytochemistry. Non-immune IgG was used for comparison. The results showed that ALK1 was specifically expressed in the condensations (Fig. 2B).


Figure 2
View larger version (65K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 BMP-2 induces expression of ALK1. (A) Bovine VMC were allowed to form condensations for 12 days. ALK1 expression was determined after 1, 4, 8 and 12 days by real-time PCR and normalized to GAPDH. Photos were taken after 4 and 12 days (original magnification 10x). (B) VMC were allowed to form condensations for 7 days. ALK1 expression was visualized by immunocytochemistry and compared to non-immune IgG control (original magnification 10x). (C) VMC infected with inducible BMP-2 or empty control adenoviral vector were treated with doxycycline (500 ng/ml) for 70 h. BMP-2 in the culture media of BMP-2 expressing and control cells was visualized by immunoblotting after the indicated number of hours. (D) ALK1 expression was determined in BMP-2 expressing and control cells after the indicated number of hours by real-time PCR and normalized to GAPDH. (E) Activation of the Smad1 pathway was determined by visualizing phosphorylated Smad1 (P-Smad1) and comparing it to total Smad1 after the indicated number of hours. It correlated with the induction of ALK1 expression in panel D. Asterisks indicate statistically significant differences compared to the first time point. (*p<0.05, **p<0.001, ***p<0.0001, Scheffe's test).

 
BMP-2 has been found to accelerate and enlarge the VMC condensations [25,26]. To determine the effect of BMP-2 on ALK1 expression, we used a doxycycline-inducible adenoviral vector for BMP-2 to ensure high expression of BMP-2 and to allow us to study both dose-dependency and the time course of induced ALK1 expression. We infected the VMC with the BMP-2 adenoviral vector or an empty control vector. We used doxycycline at 500 ng/ml for induction, which resulted in a BMP-2 secretion of 70–80 ng per ml and 24 h (8 ml medium per 100 mm culture dish) as determined by BMP-2 ELISA. The BMP-2 was also visualized by immunoblotting (Fig. 2C), which has a lower sensitivity than the ELISA. Cellular RNA was prepared at different time points after doxycycline induction and ALK1 expression was assessed by real-time PCR and normalized to GAPDH. The results showed that BMP-2 induced ALK1 expression in a transient pattern, ALK1 expression increased up to 20-fold (19.6±0.7) compared to 1.5-fold for controls (1.4±0.1), and then rapidly decreased over the course of 70 h after induction (Fig. 2D) despite the continuous increase of BMP-2 secretion. There was an increase in Smad1 activation (phosphorylation) between 24 and 54 h after BMP-2 induction, correlating with ALK1 induction (Fig. 2E). High BMP-2 secretion accelerated the formation and increased the size of the VMC condensations consistent with previous reports [25,26] (data not shown).

3.3 BMP-2 and activated ALK1 induces expression of MGP in VMC
MGP, a BMP-2 inhibitor, is expressed in VMC and restricts condensation size [25,28]. We determined the MGP expression at different time points after BMP-2 induction in bovine VMC by real-time PCR and normalization to GAPDH. The results showed that MGP expression increased up to 11-fold (11.1±1.0) and then decreased (Fig. 3A). Peak expression of MGP was delayed compared to that of ALK1.

To determine if activation of ALK1 induced MGP, we transfected the VMC with an expression vector for constitutively active (ca) ALK1 (0–5 µg/well, 6-well plates). To avoid influence from endogenous ALK5 signaling, the cells were treated with neutralizing anti-TGF-β antibodies. Expression of TGF-β1 did not change significantly during these experiments (data not shown). ALK1 expression was confirmed by real-time PCR with normalization to GAPDH and immunoblotting (Fig. 3B). MGP expression was determined after 24 h by real-time PCR and normalized to GAPDH. The results showed that MGP expression increased 5-fold (5.1±0.2) after transfection of caALK1 (Fig. 3C). The induction is less than when using the BMP-2 adenoviral vector, which is most likely because only 20–25% of the cells express caALK1 after transfection, whereas nearly 100% of the cells express high levels of BMP-2 that induces ALK1 after adenoviral infection. Furthermore, a plateau in signaling would be expected in cells with excess ALK1 due to limited availability of other signaling factors. The results suggest that ALK1 is a mediator between BMP-2 and its inhibitor MGP. Attempts to knock-down ALK1 expression using human ALK1 siRNA in bovine cells were unsuccessful, likely due to differences in the bovine ALK1 sequence, which is not available.

MGP would be expected to decrease ALK1 expression through BMP-2 inhibition. To test this, we transfected VMC with an expression vector for human FLAG-tagged MGP (0–5 µg/well, 6-well plates). The transfection was confirmed by immunoblotting of the medium using anti-FLAG antibodies (Fig. 3D). ALK1 expression was determined after 24 h by real-time PCR and normalized to GAPDH. The results showed that ALK1 expression decreased by approximately 30% (0.3±0.1) (Fig. 3D).

Transfection of caALK1 (0–5 µg/well, 6-well plates) increased condensation formation after 4 days compared empty control plasmid (Fig. 4A, 3 µg/well). Since BMP-2 increased ALK1 and MGP decreased ALK1, but both promoted condensations in VMC, we predicted that there would be a difference in ALK1 expression in the condensations formed after addition of BMP-2 versus MGP. We treated VMC with BMP-2 (100 ng/ml) or FLAG-MGP (60 ng/ml) for 4 days. Since MGP was added in the form of conditioned medium (see Methods), control cells were treated with sham-conditioned medium generated in parallel with FLAG-MGP. ALK1 expression was visualized by immunocytochemistry; non-immune IgG was used for comparison. The results showed that ALK1 expression was induced after addition of BMP-2 but not FLAG-MGP (Fig. 4B). The cells treated with control conditioned medium was indistinguishable from the spontaneously formed condensations with ALK1 expression shown in Fig. 2B.


Figure 4
View larger version (145K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 ALK1 and condensation formation. (A) Bovine VMC were transfected with an expression vector for constitutively active (ca) ALK1 or empty vector (3 µg/well, 6-well plates). The cells were allowed to form condensations for 4 days (original magnification 10x). (B) VMC were treated with BMP-2 (100 ng/ml) or FLAG-MGP (60 ng/ml) for 4 days. ALK1 expression was visualized by immunocytochemistry and compared to non-immune IgG control (original magnification 10x).

 
3.4 Activated ALK1 affects proliferation and SMC differentiation in VMC
To determine if activated ALK1 affects bovine VMC proliferation, we transfected caALK1 and compared the effect to that of BMP-2 and transfected MGP using 3H-thymidine incorporation. All cells underwent transfection with plasmid DNA containing either caALK1, FLAG-tagged MGP or empty plasmid. The total amount of DNA was kept constant at a level corresponding to 5 µg/well in 6-well plates with empty plasmid. After attachment, the cells were treated with BMP-2 (100 ng/ml) and Noggin (300 ng/ml) as per Fig. 4, and 3H-thymidine was added. After 4 days, the cells were harvested and the 3H-activity was determined. No serum deprivation was used; FBS was present at the usual concentration in all experiments due to the long days of incubation period. The results showed that BMP-2 treatment increased proliferation significantly, an effect that was neutralized by addition of Noggin, a known BMP-inhibitor (Fig. 5). Transfection of caALK1 similarly increased proliferation. However, MGP, which is induced by activated ALK1, inhibited proliferation (Fig. 5). Thus, ALK1 may mediate the proliferative effect of BMP-2, which is subsequently inhibited by MGP.

To determine if activated ALK1 affects SMC differentiation in the VMC, we transfected VMC with caALK1 vector (0–4 µg/well, 6-well plates). After 4 days, cell lysates were prepared and SM-{alpha}-actin (early SMC lineage marker), calponin (intermediate marker), and SM-MHC (late marker) were visualized by immmunoblotting and compared to ALK1 and β-actin. The results showed that caALK1 increased the expression of SMC markers (Fig. 6A), suggesting an enhancement of SMC differentiation.


Figure 6
View larger version (99K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 Activated ALK1 enhances SMC differentiation in VMC. (A) Bovine VMC were transfected with caALK1 vector (0–5 µg/well, 6-well plate). After 4 days, cell lysates were prepared and {alpha}-actin (early SMC lineage marker), calponin (intermediate marker) and SM-MHC (late marker) were visualized by immmunoblotting and compared to ALK1 and β-actin. (B) VMC were transfected with caALK1 (0–5 µg/well, 6-well plate), and incubated with or without neutralizing anti-TGF-β antibodies (300 ng/ml) to inhibit endogenous ALK5 signaling. After 4 days, cell lysates were prepared and {alpha}-actin was visualized by immmunoblotting and compared to β-actin. (C) VMC were allowed to form spontaneous condensations for 7 days, or were treated with BMP-2 (100 ng/ml) or FLAG-MGP (60 ng/ml) for 4 days. {alpha}-Actin and SM-MHC were visualized by immunocytochemistry and compared to non-immune IgG control (original magnification 10x).

 
To determine if activated ALK1 was sufficient to induce SMC differentiation, we transfected the cells with caALK1 vector (0–4 µg/well, 6-well plates), and incubated them without or with neutralizing anti-TGF-β antibodies to inhibit endogenous signaling through ALK1 and ALK5. After 4 days, cell lysates were prepared and {alpha}-actin was visualized by immmunoblotting and compared to β-actin. The results showed that inhibition of endogenous TGF-β signaling abolished the appearance of the SMC marker (Fig. 6B). Thus, activated ALK1 alone is not sufficient to induce SMC-differentiation.

We then determined if the SMC markers were specifically expressed in the condensations and if they were associated with ALK1 expression. We performed immunocytochemistry for {alpha}-actin and SM-MHC in condensations formed spontaneously for 7 days or with 4 days of treatment with BMP-2 (100 ng/ml) or FLAG-MGP (60 ng/ml). The results showed that {alpha}-actin and SM-MHC were specifically induced in condensations, but only in the condensations that formed spontaneously or after BMP-2 treatment (Fig. 6B), and that expressed ALK1 (Figs. 2B and 4BGo). No SMC markers were detected in the condensations that formed after FLAG-MGP treatment (Fig. 6B) and were without ALK1 expression (Fig. 4B).

3.5 Activated ALK1 induces ALK5 expression
We then examined the expression levels of ALK5, the TGF-β type I receptor that may be connected with SMC differentiation. First, bovine VMC were allowed to spontaneously form condensations over 12 days. Cellular RNA was prepared at different stages and ALK5 expression was assessed by real-time PCR and normalized to GAPDH. The results showed that expression of the ALK5 increased 3-fold (2.8±0.2) in condensations (Fig. 7A). Second, we examined if activated ALK1 affected the expression of ALK5. VMC were transfected with caALK1 vector (0–5 µg/well, 6-well plates). Anti-TGF-β antibodies were included to inhibit endogenous ALK5 signaling. RNA was prepared after 24 h and ALK5 expression was determined by real-time PCR with normalization to GAPDH and immunoblotting. The results showed a significant increase (3.4±0.4) in ALK5 expression (Fig. 7B), suggesting that activated ALK1 promotes ALK5 expression. To determine if ALK5 expression was induced in the condensations, we allowed condensations to form spontaneously for 7 days and visualized ALK5 expression by immunocytochemistry. The results showed that ALK5 expression was increased in the condensations (Fig. 7C). However, there was a low level of ALK5 staining in all cells consistent with ALK5 being a more ubiquitous TGF-β receptor than ALK1.


Figure 7
View larger version (39K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 Acitvated ALK1 induces expression of ALK5. (A) Bovine VMC were allowed to spontaneously form condensations over 12 days. Cellular RNA was prepared at different stages and ALK5 expression was assessed by real-time PCR with normalization to GAPDH. (B) VMC were transfected with caALK1 (0–5 µg/well, 6-well plate). RNA was prepared after 24 h and ALK5 expression was determined by real-time PCR with normalization to GAPDH and immunoblotting. (C) VMC were allowed to form spontaneous condensations for 7 days. ALK5 expression was visualized by immunocytochemistry and compared to non-immune IgG control (original magnification 4x). Asterisks indicate statistically significant differences compared to control. (*p<0.05, **p<0.001, ***p<0.0001, Scheffe's test).

 
3.6 ALK1 is expressed in condensations of human VMC
We have previously shown that human VMC react similarly to bovine VMC in their response to BMP-2 and MGP [25]. To determine if ALK1 was expressed in condensations of human VMC, we allowed them to form condensations for 7 days, and then visualized ALK1 expression by immunocytochemistry. The results showed that the human VMC expressed ALK1 specifically in the condensations similarly to the bovine VMC (Fig. 8A).


Figure 8
View larger version (67K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 8 ALK1 is expressed in human VMC. (A) Human VMC were allowed to form condensations for 7 days. ALK1 expression was visualized by immunocytochemistry and compared to non-immune IgG control (original magnification 10x). (B) Schematic representation of the current working model. BMP-2 induces expression of ALK1, which in turn induces expression of MGP when activated by TGF-β. BMP-2 and MGP interact to generate the pattern and size of the cell condensations. ALK1 is also critical in SMC differentiation, which can be separated from the process of cell aggregation. ALK1 induces expression of SMC markers possibly through ALK5.

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
In this study, we show that ALK1 is induced in several areas of atherosclerotic lesions including endothelium, neointima and the core of the lesion. Focusing our studies on vascular mesenchymal cells, we found that these cells express ALK1 when they undergo condensation. We also show that BMP-2 induces expression of ALK1 in a transient fashion. Activation of ALK1 results in increased expression of MGP, a BMP-2 inhibitor that may restrict BMP-2 activity through negative feedback regulation. In addition, ALK1 activation enhances cell proliferation, SMC differentiation and expression of ALK5, which has been associated with SMC differentiation [7].

BMP-2 and MGP have previously been shown to act as an activator and an inhibitor respectively in a diffusion–reaction model that determines the pattern formation in VMC [26]. BMP-2 and MGP has also been shown to increase and decrease nodule size respectively [25]. The current study shows that ALK1 may function as a link between BMP-2 and MGP allowing them to regulate each other. Thus, appropriate function of ALK1 may be critical in maintaining the balance between BMP-2 and MGP, a balance that also determines the expression of ALK1 itself. The finding that condensations formed without ALK1 expression when the cells were treated with MGP supports the diffusion-reaction model and that it is the balance between BMP-2 and MGP that determines the pattern of cell aggregation. This also implies that one type of cells would be able to affect the aggregation of a second type of cells by secreting these factors.

Homozygosity for ALK1 deficiency is embryonic lethal in mice, whereas heterozygosity leads to a phenotype resembling HHT in humans [3,4]. It has been proposed that defective endothelial ALK1 signaling and decreased TGF-β levels in the vicinity of the SMC precursor cells causes poor recruitment [4]. The ALK1 signaling in the endothelium may be important in the initiation of the lesion. However, ALK1 expression has been reported in vascular and pulmonary mesenchyme [5,6], and our results suggest that expression of ALK1 in the SMC precursors themselves may contribute to regulation of proliferation and promotion of SMC differentiation both during development and atherosclerosis progression.

An atherosclerotic lesion could be considered the opposite of the deficient SMC recruitment in HHT, in that immature mesenchymal cells are stimulated to migrate and proliferate within the lesion. BMP-2 and MGP are both expressed in atherosclerotic lesions [8,9,15–17], where they might be linked to ALK1. In normal vascular wall, MGP would be expected to limit BMP-2 activity. However, functional MGP requires intact {gamma}-carboxylation, which is vitamin K-dependent. Insufficient {gamma}-carboxylation results in undercarboxylated MGP, which has been detected in atherosclerotic plaques [30,31] and is incapable of binding BMP-2 [21]. Uninhibited BMP-2 may explain the calcification and the osteogenic differentiation detected in plaques as well as the high levels of ALK1. This may also be the reason why MGP gene deletion results in arterial calcification with replacement of SMC with chondrocyte-like cells [18].

In endothelial cells, signaling by TGF-β/ALK1 through Smad1/5/8 directly antagonizes signaling through ALK5 and Smad2/3 [32]. Nevertheless, ALK5 is important for optimal TGF-β/ALK1 signaling in that ALK5 mediates a TGF-β dependent recruitment of ALK1 into a TGF-β receptor complex, and ALK5 kinase activity is required for efficient ALK1 activation. However, it has not been shown previously that activated ALK1 induces expression of ALK5, which supports a later role for ALK5 in e.g. angiogenesis [1].

TGF-β1 signaling through Smad2/3 has been shown to contribute to development of SMC from embryonic stem cells [7], suggesting that ALK5 is the responsible TGF-β type I receptor. Furthermore, SMC differentiation was diminished by a truncated form of the TGF-β type II receptor (TβR-II), which is required for ALK5 signaling. A significant down-regulation of TβR-II has been reported in atherosclerotic lesions and lesions cells whereas little change was reported for ALK5 [33,34]. The absence of TβR-II in lesions cells was connected to absence of growth inhibitory response to TGF-β1, which was restored after transfection of TβR-II [34]. The ALK1 receptor was not studied in these experiments. Our results indicate that ALK1 is critical in SMC induction but further studies are needed to determine if this is mediated by ALK5. It is possible that ALK1 and ALK5 are both induced by BMP-2. Even though TβR-II is known to signal with ALK1, it cannot be excluded that other type II receptors are used by ALK1. If other type II receptors are involved, it may fundamentally alter TGF-β signaling in lesions with low expression of the TGF-β type II receptor.

Other investigators have shown that BMP-2 inhibits proliferation of SMC induced by serum, PDGF and low-density lipoproteins [10–13]. In our experiments, BMP-2 enhanced proliferation, although this may be progressively limited by increased MGP expression induced by activated ALK1. The differences between our and previous studies may be due to serum deprivation in other studies [10–13], the use of VMC instead of SMC, difference in time periods or BMP-2 concentrations, or availability of TGF-β. Previous studies showed that proliferation of VMC was affected by various treatments despite the presence of 5 or 15% of serum in the culture medium. Proliferation decreased to about 15–20% of baseline when treated with cAMP or forskolin [29] and increased 2–3-fold when treated with insulin-like growth factor-I [35]. Therefore we performed the proliferation experiments in the presence of the usual concentration of serum in order to avoid potential adverse effects on cell viability.

Altogether, our studies suggest a working model (Fig. 8B) where BMP-2 connects with TGF-β signaling through induction of ALK1 expression. ALK1 in turn induces expression of MGP when activated by TGF-β. BMP-2 and MGP interact to generate the pattern and size of condensations. ALK1 is also critical in SMC differentiation, which can be separated from the process of cell aggregation. ALK1 induces expression of SMC markers possibly through ALK5. Imbalance in this system may lead to the excessive ALK1 expression in atherosclerotic lesions, possibly due to poor feedback regulation of BMP-2 by insufficient or undercarboxylated MGP.


   Acknowledgement
 
This work was funded in part by NIH grants HL30568 and HL81397, the American Heart Association, and the International HDL Awards program (Pfizer). We thank the Department of Pathology and Laboratory Medicine at UCLA for assistance with autopsy specimens.


   Notes
 
Time for primary review 39 days


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. Lebrin F., Deckers M., Bertolino P., Ten Dijke P. TGF-beta receptor function in the endothelium. Cardiovasc Res (2005) 65:599–608.[Abstract/Free Full Text]
  2. Lux A., Attisano L., Marchuk D.A. Assignment of transforming growth factor β1 and β3 and a third new ligand to the type I receptor ALK-1. J Biol Chem (1999) 274:9984–9992.[Abstract/Free Full Text]
  3. Urness L.D., Sorensen L.K., Li D.Y. Arteriovenous malformations in mice lacking activin receptor-like kinase-1. Nature (2000) 26:328–331.
  4. Carvalho R.L.C., Jonker L., Goumans M.-.J., Larsson J., Bouwman P., Karlsson S., et al. Defective paracrine signalling by TGFβ in yolk sac vasculature of endoglin mutant mice: a paradigm for hereditary haemorrhagic telangectasia. Development (2004) 131:6237–6247.[Abstract/Free Full Text]
  5. Panchenko M.P., Williams M.C., Brody J.S., Yu Q. Type I receptor serine-threonine kinase preferentially expressed in pulmonary blood vessels. Am J Physiol (1996) 270:L547–L548.[ISI][Medline]
  6. Roelen B.A.J., Van Rooijen M.A.V., Mummery C.L. Expression of ALK-1, a type I serine/threonine kinase receptor, coincides with sites of vasculogenesis and angiogenesis in early mouse development. Dev Dyn (1997) 209:418–430.[CrossRef][ISI][Medline]
  7. Sinha S., Hoofnagle M.H., Kinston P.A., McCanna M.E., Owens G.K. Transforming growth factor-beta1 signaling contributes to development of smooth muscle cells from embryonic stem cells. Am J Physiol Cell Physiol (2004) 287:C1560–C1568.[Abstract/Free Full Text]
  8. Boström K., Watson K.E., Horn S., Wortham C., Herman I.M., Demer L.L. Bone morphogenetic protein-2A expression in human atherosclerotic lesions. J Clin Invest (1993) 91:1800–1809.[ISI][Medline]
  9. Dhore C.R., Cleutjens J.P.M., Lutgens E., Cleutjens K.B.J.M., Geusens P.P.M., Kitslaar P.J.E.H.M., et al. Differential expression of bone matrix regulatory proteins in human atherosclerotic plaques. Arterioscler Thromb Vasc Biol (2001) 21:1998–2003.[Abstract/Free Full Text]
  10. Willette R.N., Gu J.L., Lysko P.G., Anderson K.M., Minehart H., Yue T. BMP-2 gene expression and effect on human vascular smooth muscle cells. J Vasc Res (1999) 289:48–53.
  11. Nakaoka T., Gonda K., Ogita T., Otawara-Hamamoto Y., Okabe F., Kira Y., et al. Inhibition of rat vascular smooth muscle proliferation in vitro and in vivo by bone morphogenetic protein-2. J Clin Invest (1997) 100:2824–2832.[ISI][Medline]
  12. Wong G.A., Tang V., El-Sabeawy F., Weiss R.H. BMP-2 inhibits proliferation of human aortic smooth muscle cells via p21Cip1/Waf1. Am J Physiol Endocrinol Metab (2003) 284:E972–E979.[Abstract/Free Full Text]
  13. Emmanuele L., Ortmann J., Doerflinger T., Traupe T., Barton M. Lovastatin stimulates human vascular smooth muscle cell expression of bone morphogenetic protein-2, a potent inhibitor of low-density lipoprotein-stimulated cell growth. Biochem Biophys Res Commun (2003) 28:67–72.
  14. Shah N.M., Groves A.K., Anderson D.J. Alternative neural crest cell fates are instructively promoted by TGFβ superfamily members. Cell (1996) 85:331–343.[CrossRef][ISI][Medline]
  15. Shanahan C.M., Cary N.R.B., Metcalfe J.C., Weissberg P.L. High expression of genes for calcification-regulating proteins in human atherosclerotic plaques. J Clin Invest (1994) 93:2393–2402.[ISI][Medline]
  16. Spronk H.M.H., Soute B.E.M., Schurgers L.J., Cleutjens J.P.M., Thijssen H.H.W., De Mey J.G.R., et al. Matrix Gla protein accumulate at the border of regions of calcification and normal tissue in the media of the arterial vessel wall. Biochem Biophys Res Commun (2001) 289:485–490.[CrossRef][ISI][Medline]
  17. Engelese M.A., Neele J.M., Bronckers A.L.J.J., Pannekoek H., de Vries C.J.M. Vascular calcification: expression patterns of the osteoblast-specific gene core binding factor alpha-1 and the protective factor matrix gla protein in human atherogenesis. Cardiovasc Res (2001) 52:281–289.[Abstract/Free Full Text]
  18. Luo G., Ducy P., McKee M.D., Pinero G.J., Loyer E., Behringer R.R., et al. Spontaneous calcification of arteries and cartilage in mice lacking matrix GLA protein. Nature (1997) 386:78–81.[CrossRef][Medline]
  19. Price P.A., Faus S.A., Williamson M.K. Warfarin causes rapid calcification of the elastic lamellae in rat arteries and heart valves. Arterioscler Thromb Vasc Biol (1998) 18:1400–1407.[Abstract/Free Full Text]
  20. Schurgers L.J., Aebert H., Vermeer C., Bultmann B., Janzen J. Oral anticoagulant treatment: friend or foe in cardiovascular disease? Blood (2004) 104:3231–3232.[Abstract/Free Full Text]
  21. Wallin R., Cain D., Hutson S.M., Sane D.C., Loeser R. Modulation of the binding of matrix Gla protein (MGP) to bone morphogenetic protein-2 (BMP-2). Thromb Haemost (2000) 84:1039–1044.[ISI][Medline]
  22. Zebboudj A.F., Imura M., Boström K. Matrix GLA protein, a regulatory protein for bone morphogenetic protein-2. J Biol Chem (2002) 277:4388–4394.[Abstract/Free Full Text]
  23. Watson K.E., Bostrom K., Ravindranath R., Lam T., Norton B., Demer L.L. TGF-beta 1 and 25-hydroxycholesterol stimulate osteoblast-like vascular cells to calcify. J Clin Invest (1994) 93:2106–2113.[ISI][Medline]
  24. Tintut Y., Alfonso Z., Saini T., Radcliff K., Watson K., Boström K., et al. Multilineage potential of cells from the artery wall. Circulation (2003) 108:2505–2510.[Abstract/Free Full Text]
  25. Zebboudj A.F., Shin V., Boström K. Matrix GLA protein and BMP-2 regulate osteoinduction in calcifying vascular cells. J Cell Biochem (2003) 90:756–765.[CrossRef][ISI][Medline]
  26. Garfinkel A., Tintut Y., Petrasek D., Boström K., Demer L.L. Spatial pattern formation in a tissue culture model of vascular calcification. Proc Natl Acad Sci U S A (2004) 101:9247–9250.[Abstract/Free Full Text]
  27. Boström K., Zebboudj A.F., Yao Y., Lin T.S., Torres A. Matrix GLA protein stimulates VEGF expression through increased TGF-β1 activity in endothelial cells. J Biol Chem (2004) 279:52904–52913.[Abstract/Free Full Text]
  28. Sugiyama S., Kugiyama K., Matsumura T., Suga S.-I., Itoh H., Nakao K., et al. Lipoproteins regulate C-type natriuretic peptide secretion from cultured vascular endothelial cells. Arterioscler Thromb Vasc Biol (1995) 15:1968–1974.[Abstract/Free Full Text]
  29. Tintut Y., Parhami F., Boström K., Jackson S.M., Demer L.L. cAMP stimulates osteoblast-like differentiation of calcifying vascular cells. J Biol Chem (1998) 273:7547–7553.[Abstract/Free Full Text]
  30. Sweatt A., Sane D.C., Hutson S.M., Wallin R. Matrix Gla protein (MGP) and bone morphogenetic protein-2 in aortic calcified lesions of aging rats. J Thromb Haemost (2003) 1:178–185.[CrossRef][ISI][Medline]
  31. Schurgers L.J., Teunissen K.J., Knapen M.H., Kwaijtaal M., van Diest R., Appels A., et al. Novel conformation-specific antibodies against matrix gamma-carboxyglutamic acid (Gla) protein: undercarboxylated matrix Gla protein as marker for vascular calcification. Arterioscler Thromb Vasc Biol (2005) 25:1629–1633.[Abstract/Free Full Text]
  32. Goumans M.-J., Valdimarsdottir G., Itoh S., Lebrin F., Larsson J., Mummery C., et al. Activin receptor-like kinase (ALK)1 is an antagonistic mediator of lateral TGFβ/ALK5 signaling. Mol Cell (2003) 12:817–828.[CrossRef][ISI][Medline]
  33. McCaffrey T.A. TGF-betas and TGF-beta receptors in atherosclerosis. Cytokine Growth Factor Rev (2000) 11:103–114.[CrossRef][ISI][Medline]
  34. Bobik A., Agrotis A., Kanellakis P., Dilley R., Krushinsky A., Smirnov V., et al. Distinct patterns of transforming growth factor-β isoform and receptor expression in human atherosclerotic lesions. Colocalization implicates TGF-β in fibrofatty lesion development. Circulation (1999) 99:2883–2891.[Abstract/Free Full Text]
  35. Radcliff K., Tang T.-B., Lim J., Zhang Z., Abedin M., Demer L.L., et al. Insulin-like growth factor-I regulates proliferation and osteoblastic differentiation of calcifying vascular cells via extracellular signal-regulated protein kinase and phosphatidylinositol 3-kinase pathways. Circ Res (2005) 96:398–400.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Circ. Res.Home page
Y. Yao, A. Shahbazian, and K. I. Bostrom
Proline and {gamma}-Carboxylated Glutamate Residues in Matrix Gla Protein Are Critical for Binding of Bone Morphogenetic Protein-4
Circ. Res., May 9, 2008; 102(9): 1065 - 1074.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Yao, Y.
Right arrow Articles by Boström, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yao, Y.
Right arrow Articles by Boström, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?