Skip Navigation


Cardiovascular Research Advance Access first published online on June 16, 2008
This version [Corrected Proof] published online on July 2, 2008
Cardiovascular Research, doi:10.1093/cvr/cvn159
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
80/1/96    most recent
cvn159v2
cvn159v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
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
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Heller, R.
Right arrow Articles by Wetzker, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Heller, R.
Right arrow Articles by Wetzker, R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2008. For permissions please email: journals.permissions@oxfordjournals.org

Overlapping and distinct roles for PI3Kβ and {gamma} isoforms in S1P-induced migration of human and mouse endothelial cells

Regine Heller1,*, Qing Chang1, Gunter Ehrlich1, Sherry N. Hsieh1, Simone M. Schoenwaelder2, Peter J. Kuhlencordt3, Klaus T. Preissner4, Emilio Hirsch5 and Reinhard Wetzker1

1 Center for Molecular Biomedicine, Institute of Molecular Cell Biology, Friedrich Schiller University, Jena, Am Leutragraben 3, 07743 Jena, Germany
2 Australian Centre for Blood Diseases, Monash University, Alfred Medical Research Centre and Education Precinct (AMREP), Melbourne, Victoria, Australia
3 Medizinische Klinik I/Herz-Kreislaufzentrum, Universitätsklinikum, Julius Maximilian University, Würzburg, Germany
4 Institute for Biochemistry, Medical Faculty, Justus Liebig University, Giessen, Germany
5 Department of Genetics, Biology and Biochemistry, Center for Molecular Biotechnology, University of Torino, Torino, Italy

* Corresponding author. Tel: +49 3641 938 750; fax: +49 3641 938 752. E-mail address: regine.heller{at}mti.uni-jena.de

Received 19 October 2007; revised 3 June 2008; accepted 6 June 2008

Time for primary review: 22 days


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Supplementary material
 Funding
 References
 
Aims: Sphingosine-1-phosphate (S1P), a key regulator of vascular homeostasis and angiogenesis, promotes endothelial cell migration via stimulation of phosphoinositide 3-kinase (PI3K). The aim of this study was to identify the role of PI3Kβ and {gamma} isoforms and their downstream effector pathways in mediating endothelial cell migration induced by S1P.

Methods and results: Experiments were performed in human umbilical vein endothelial cells (HUVEC) and murine lung endothelial cells (MLEC). A combination of specific inhibitors, RNA interference, and PI3K{gamma}–/– mice were used to investigate the role of PI3Kβ and {gamma} isoforms in endothelial cell migration. Both PI3Kβ and {gamma} isoforms are required for full migration induced by S1P, with Rac1 being a major mediator downstream of both isoforms. In addition, PI3Kβ but not PI3K{gamma} mediates migration via Akt but independent of Rac1 and endothelial NO synthase (eNOS). Further, a S1P-mediated activation of extracellular signal-regulated kinases (Erk) 1/2 contributes to the chemotactic response independent of PI3Kβ or PI3K{gamma}.

Conclusions: Our data demonstrate that both PI3Kβ and PI3K{gamma} isoforms are required for S1P-induced endothelial cell migration through activation of Rac1. In addition, PI3Kβ initiates an Akt-sensitive chemotactic response which is independent of Rac1 and eNOS. Thus, PI3Kβ and PI3K{gamma} have both overlapping and distinct roles in regulating endothelial cell migration, which may underlie S1P-triggered angiogenic differentiation.

KEYWORDS Angiogenesis; Endothelial function; Lipid signalling; Protein kinases; Signal transduction


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 Supplementary material
 Funding
 References
 
Sphingosine-1-phosphate (S1P), a biologically active sphingolipid metabolite, plays a key role in the regulation of vascular homeostasis.1,2 In endothelial cells, S1P participates in processes that are important in angiogenesis such as proliferation, migration, and barrier integrity.1,2 S1P stimulates endothelial migration and subsequent angiogenic differentiation through the Gi-coupled receptor, S1P1, and the Gi/Gq/G12-coupled receptor, S1P3, leading to activation of the Rho GTPases Cdc42, Rac, and Rho.3–10

S1P-induced migration is also mediated via activation of class I phosphoinositide 3-kinases (PI3Ks).7–9 These enzymes produce phosphatidylinositol 3,4,5-trisphosphate (PIP3), which recruits pleckstrin homology domain-containing molecules such as Akt and guanine nucleotide exchange factors (GEFs) to the membrane. Importantly, several lines of evidence support a role for Akt and GEFs in mediating migration.11

Class I PI3Ks are classified into two subclasses: class IA, which has three members, PI3K{alpha}, β, and {delta}; and class IB, which has one member, PI3K{gamma}. PI3K{alpha} and {delta} isoforms are activated by receptor tyrosine kinases, while PI3K{gamma} is activated by G protein β{gamma} subunits (Gβ{gamma}), which are usually derived from Gi-coupled receptors. In contrast, the PI3Kβ isoform is regulated by receptor tyrosine kinases and by Gβ{gamma}-subunits.11–13

Although a role for PI3K in S1P-induced endothelial cell migration has been proposed, the responsible isoforms have not been characterized. Further, the contribution and interrelationship of the downstream effectors, Rac1 and Akt and its substrate endothelial NO synthase (eNOS), in mediating migration is not fully understood.

In the present study, we have focused on the role of the two Gi-sensitive PI3K isoforms, β and {gamma}, in mediating migration of endothelial cells to S1P. Importantly, both isoforms have been shown to control chemotaxis in leukocytes.11,14 The present observations demonstrate differential functions for the two isoforms in regulating endothelial cell migration upstream of Rac1 and Akt following activation by S1P. The results may have important consequences for the role of the two PI3K isoforms in regulation endothelial cell migration in vivo.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 Supplementary material
 Funding
 References
 
2.1 Materials
Endothelial mitogen was obtained from Biomedical Technologies (Stoughton, MA, USA). Rabbit polyclonal antibodies reacting with Akt1 or with the PI3Kβ catalytic subunit p110β were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The monoclonal PI3K{gamma} antibody (clone H1) recognizing the sequence aa 97–335 of human and mouse PI3K{gamma} (catalytic subunit p110{gamma}) was generated as described15 and is distributed by Jena Bioscience GmbH (Jena, Germany). Monoclonal antibodies against human eNOS (clone 2) and Rac1 (clone 102) or against mouse CD102 were from BD Transduction or BD Pharmingen, respectively (Heidelberg, Germany). The phosphospecific antibodies recognizing phosphorylated Akt (serine 473), eNOS (serine 1177), or Erk1/2 (threonine 202/tyrosine 204 of Erk1, threonine 185/tyrosine 187 of Erk2) and the antibodies against Erk1/2 and Akt were from Cell Signaling Technologies (Frankfurt, Germany).

M-450 sheep anti-rat beads were acquired from Dynal Biotech (Hamburg, Germany). The PI3K inhibitors TGX-221 (selective for PI3Kβ) and AS-252424 (selective for PI3K{gamma}), both synthesized as described,16,17 were kind gifts from the Baker Heart Institute. S1P, PD-98059, Akt inhibitor VIII, and the Rac1 inhibitor NSC-23766 were purchased from Merck Chemicals Ltd. (Darmstadt, Germany). The protease inhibitor cocktail Complete was from Roche Diagnostics (Mannheim, Germany) and prepared as described by the manufacturer. Wortmannin, pertussis toxin, L-nitroarginine methylester (L-NAME), and other reagents were acquired from Sigma-Aldrich (Deisenhofen, Germany).

2.2 Animals
PI3K{gamma}–/– mice were obtained as previously reported.18 Experiments were performed with sex-matched 3-month-old wild-type (PI3K{gamma}+/+) and PI3K{gamma}–/– mice. Both mutants and control animals were littermates from heterozygous crosses. The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health.

2.3 Cell culture
Human umbilical cord vein endothelial cells (HUVEC) were isolated from anonymously acquired umbilical cords according to the Declaration of Helsinki, cultured as described earlier,19 characterized by flow cytometric staining for platelet endothelial cell adhesion molecule-1 (PECAM-1, >98% positive) and used at the first or second passage. Murine lung endothelial cells (MLEC) were isolated from mouse lungs with collagenase A and selected twice in a magnetic field after incubating the cultures with magnetic beads coated with anti-mouse CD102 antibody.20 Typically, >90% of cells were PECAM-1 positive as measured by flow cytometry. Cultures from wild-type and PI3K{gamma}–/– mice were started simultaneously.

Experiments were carried out after 5 h incubation in serum-free M199 containing 0.25% human serum albumin (M199/HSA). Pre-treatment with inhibitors and stimulation was performed in HEPES/HSA buffer [10 mM HEPES (pH 7.4), 145 mM NaCl, 5 mM KCl, 1 mM MgSO4, 10 mM glucose, 1.5 mM CaCl2, 0.25% HSA] or M199/HSA (migration assays, pertussis toxin pre-treatment). Wortmannin, TGX-221, AS-252424, genistein, and Akt inhibitor VIII were dissolved in dimethyl sulfoxide. Control cells received the same volume of solvent and the final concentration did not exceed 0.1%. The S1P stock solution was prepared in methanol and, after evaporation of the solvent, reconstituted with HEPES/albumin buffer [20 mM HEPES (pH 7.4), 150 mM NaCl, 0.25 mM bovine serum albumin].

2.4 Small interfering RNA (siRNA)-mediated knockdown of PI3K isoforms, Akt1, and eNOS
siRNA duplex oligonucleotides used in this study are based on the human cDNA sequences encoding PI3Kβ and {gamma} (catalytic subunits), Akt1, and eNOS. PI3K{gamma}-specific siRNA duplexes were custom-made by Sigma-Proligo (Hamburg, Germany). Sense and antisense strands of the siRNA targeting PI3K{gamma} had the following sequences: 5'-GCAUAUCCUAAGCUAUUUA-dTdT-3' and 5'-UAAAUAGCUUAGGAUAUGCdTdT-3'. Validated PI3Kβ-siRNA (Catalog number SI02622214), Akt1-siRNA (Catalog number SI00299145), and non-silencing siRNA [5'-UUCUCCGAACGUGUCACGUdTdT-3' (sense) and 5'-ACGUGACACGUUCGGAGAAdTdT-3' (antisense)] were obtained from Qiagen (Hilden, Germany). ON-TARGETplus SMARTpool siRNA for downregulation of eNOS (L-006490-00-0005) and ON-TARGETplus Non-targeting Pool (D-001810-10-0005) as control were obtained from Thermo Fisher Scientific (Lafayette, CO, USA).

Transfection was performed using the amphiphilic delivery system SAINT-RED (Synvolux Therapeutics B.V., Groningen, The Netherlands) as described earlier.19 The transfection efficiency of fluorescence-labelled siRNA was ≥99%.

2.5 Transwell migration assay
Migration assays were performed in the absence of serum or growth factors. Cells were seeded on membrane inserts (8 µm pore size, BD FalconTM) coated with 0.2% gelatin and placed into 12-well plates (2.5 x 105 cells/insert). Inhibitors were added to insert and bottom well, and following pre-incubation, S1P was applied to the lower chamber. After 4 h, cells were fixed with 4% paraformaldehyde and stained with haematoxylin. Non-migrated cells were wiped away and cells migrated to the lower surface of the membrane were scored under a microscope in 10 randomly selected fields. Basal migration was very low (<5 cells per field) and not significantly changed by the used inhibitors or siRNAs. S1P-induced migration is typically shown as the difference from stimulated and the corresponding unstimulated cells and expressed as average cell number per field.

2.6 Western blot analysis
Cells were lysed in ice-cold Tris buffer [50 mM Tris (pH 7.4), 2 mM EDTA, 1 mM EGTA] containing 1% Triton X-100, 0.1% SDS, 50 mM NaF, 10 mM Na4P2O7, 1 mM Na3VO4, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, and 10 µL/mL protease inhibitor cocktail. For the detection of PI3Kβ, cell lysates were subjected to immunoprecipitation with a polyclonal anti-PI3Kβ antibody and protein A sepharose. Lysate proteins or immune complexes were solubilized in Laemmli buffer and separated by SDS–polyacrylamide gel electrophoresis. Blots were immunostained with primary antibodies overnight and peroxidase-conjugated secondary antibodies for 1 h. Protein bands were visualized by chemiluminescence and evaluated by densitometry. Phosphospecific signals were normalized against the amount of total protein. Changes in phosphorylation were evaluated by comparing the differences between basal and stimulated values of specifically treated samples and their respective controls.

2.7 Rac activity assay
The activation of Rac was measured in pull-down assays, in which the glutathione S-transferase-tagged Cdc42- and Rac-interacting binding domain of p21-activated kinase (GST-PAK) was used to isolate the active GTP-bound form of Rac. Cells were lysed on ice in a buffer consisting of 50 mM Tris (pH 7.4), 50 mM NaCl, 5 mM MgCl2, 1 mM EGTA, 1% NP-40, 10% glycerol, 100 µM GDP, 10 µL/mL proteinase inhibitor cocktail, and 20 µg/mL GST-PAK. Lysates (200 µg protein) were incubated with glutathione sepharose beads (40 µL of 50% slurry, 30 min, 4°C). The beads were washed three times with lysis buffer. The bead bound protein was eluted with Laemmli buffer and analysed by immunoblotting using Rac1 antibody.

2.8 Statistical analysis
Data represent means ± SEM of three to five independent experiments. For evaluation of statistical significance, analyses of variance with Bonferroni's correction for multiple comparisons, or t-tests were performed. A P-value of <0.05 was accepted as statistically significant.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 Supplementary material
 Funding
 References
 
3.1 S1P induces endothelial cell migration via a Gi protein and PI3K-dependent pathway
S1P stimulated migration of both HUVEC and MLEC as assayed in transwell filter systems (Figure 1A). The effect was dose-dependent with a threshold at 0.01 µM and a maximal response at 1 µM (Figure 1B). S1P-induced cell migration was blocked by pertussis toxin (0.1 µg/mL) and decreased by wortmannin (0.1 µM), a general inhibitor of PI3K, by almost 50% demonstrating a role for heterotrimeric Gi proteins and PI3K activity, respectively (Figure 1C). In contrast, general tyrosine kinase inhibition by genistein did not affect S1P-induced migration suggesting that transactivation of receptor tyrosine kinases was not involved (data not shown).


Figure 1
View larger version (27K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1 Effect of Gi protein inhibition, PI3K inhibition, or PI3K{gamma} deficiency on S1P-induced migration. HUVEC (A–D) or MLEC from wild-type or PI3K{gamma}-deficient mice (A and E) were seeded on the top of migration chambers. Pertussis toxin (0.1 µg/mL, 3 h) or PI3K inhibitors (0.1 µM wortmannin, 0.1 µM TGX-221, 1 µM AS-252424, 30 min pre-incubation) were added to both compartments and S1P was added to the bottom. After 4 h, migrated cells were counted. (A and B) Means ± SEM, n = 7 (A) or n = 4 (B), *P < 0.05 vs. unstimulated cells. (C) Separate series of experiments for each inhibitor are shown. (C–E) Means ± SEM, n = 4 (C) or 5 (E and D). Differences between S1P-stimulated and unstimulated cells are shown and cells with and without inhibitor pre-treatment or PI3K{gamma}-deficient and wild-type cells were compared. *P < 0.05.

 
3.2 PI3Kβ and PI3K{gamma} mediate S1P-induced endothelial cell transwell migration
PI3Kβ and PI3K{gamma}, which are both candidates for the wortmannin-sensitive component of S1P-induced migration, are expressed in HUVEC and MLEC (Supplementary material online, Figure S1). Further, the level of PI3Kβ is not altered in MLEC isolated from PI3K{gamma}–/– mice, which as expected lack detectable PI3K{gamma} (Supplementary material online, Figure S1B).

To investigate the role of the two PI3K isoforms in S1P-induced endothelial migration, we employed TGX-221 and AS-252424, which have been shown to inhibit PI3Kβ and PI3K{gamma}, respectively, with high specificity.16,17 Both TGX-221 (0.1 µM) and AS-252424 (1 µM) decreased S1P-induced migration in HUVEC significantly by more than 30% (Figure 1D). Further, the combined effect of TGX-221 and AS-252424 (52% inhibition) was comparable to that of wortmannin indicating that PI3Kβ and {gamma} were the major isoforms in mediating migration. In line with the results in HUVEC, MLEC derived from PI3K{gamma}–/– mice exhibited a reduced migratory response towards S1P when compared with wild-type cells (48% inhibition) (Figure 1E). Furthermore, the PI3Kβ inhibitor TGX-221 reduced S1P-induced migration in wild-type MLEC by approximately 56% and in PI3K{gamma}–/– MLEC by about 52% (Figure 1E). In contrast, AS-252424 inhibited S1P-induced migration in wild-type MLEC to a similar degree to that seen in HUVEC while it had no effect in the absence of PI3K{gamma} (data not shown). Taken together, these results reveal that both PI3Kβ and PI3K{gamma} are required for the full migratory response of endothelial cells to S1P.

3.3 PI3Kβ but not PI3K{gamma} mediates S1P-induced Akt phosphorylation
One of the downstream effectors of PI3K implicated in cell migration is the serine/threonine kinase Akt.7–9 Stimulation of HUVEC with 1 µM S1P resulted in a time-dependent phosphorylation of Akt (Supplementary material online, Figure S2A). As expected, the response was completely blocked by pertussis toxin (0.1 µg/mL) and wortmannin (0.1 µM) (Figure 2A). Interestingly, the PI3Kβ inhibitor TGX-221 (0.01–0.1 µM) decreased S1P-stimulated Akt phosphorylation in a concentration-dependent manner with 90% reduction at 0.1 µM (Figure 2A and Supplementary material online, Figure S3A), whereas AS-252424 (1 µM) had no effect (Figure 2A). Pre-treatment of HUVEC with inhibitors of the vascular endothelial cell growth factor (VEGF) receptor-2 (SU5614, 1 µM), of the epidermal growth factor (EGF) receptor (AG1478, 10 µM), or of tyrosine kinases in general (genistein, 10 µM) did not affect the S1P response indicating that a transactivation of receptor tyrosine kinases by S1P21 was not involved (Supplementary material online, Figure S3B).


Figure 2
Figure 2
View larger version (60K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2 Role of PI3Kβ isoform in S1P-stimulated Akt phosphorylation and the function of Akt1 in migration and Rac1 activation. (A–C) HUVEC or MLEC from wild-type or PI3K{gamma}-deficient mice were stimulated with S1P (1 µM, 2 min). Akt was analysed using antibodies against phosphorylated Akt (serine 473) and total Akt. (A) HUVEC were pre-incubated with pertussis toxin (0.1 µg/mL, 3 h) or with PI3K inhibitors (wortmannin and TGX-221: 0.1 µM, AS-252424: 1 µM, 30 min). (B) HUVEC were pre-treated with PI3Kβ- or PI3K{gamma}-specific siRNA or non-silencing control siRNA (1.0 µg/30 mm dish, 72 h). (D) HUVEC were seeded on the top of migration chambers, pre-treated with Akt1/2 inhibitor (20 µM, 30 min), and stimulated with S1P applied to the bottom (1 µM, 4 h). Alternatively, cells were transfected with Akt1-specific or control siRNA (1 µg/30 mm dish, 72 h) before migration experiments. S1P-induced response was calculated as difference of non-stimulated and stimulated migrated cells (means ± SEM). Cells with and without inhibitor or siRNA pre-treatment (n = 5 or n = 3, respectively) were compared. *P < 0.05. (E and F) HUVEC were pre-treated with Akt1/2 inhibitor or with Akt1 siRNA (D), stimulated with S1P (1 µM, 1 min), and lysed in the presence of GST-PAK (see Methods). GTP-bound Rac was precipitated with glutathione-sepharose beads and immunostained using Rac1 antibody. Total Rac1, phospho-Akt, total Akt and Akt1 were stained in cell lysates. (A–C, E, and F) Immunoblot data are shown as representative experiments and densitometry analyses [means ± SEM, n = 3 (C, E, and F) or n = 4 (A and B)]. Phospho-Akt or GTP-Rac1 signals in S1P-stimulated cells with or without inhibitor pre-treatment, RNA interference, or PI3K{gamma}-knockdown were compared. *P < 0.05.

 
To confirm these results, we used a specific siRNA treatment leading to a downregulation of PI3Kβ or PI3K{gamma} expression in HUVEC by 74 ± 2% (n = 3) or 88 ± 7% (n = 3), respectively (Supplementary material online, Figure S1A). S1P-induced Akt phosphorylation was significantly reduced in cells deficient in PI3Kβ (60% inhibition) but was not altered in cells treated with PI3K{gamma} siRNA (Figure 2B). Similarly, no difference was found between S1P-induced Akt phosphorylation in MLEC from wild-type and PI3K{gamma}–/– mice (Figure 2C). Thus, these data confirm that S1P-induced phosphorylation of Akt is mediated via PI3Kβ but not via PI3K{gamma}.

3.4 S1P-induced Akt activation mediates endothelial cell migration
To investigate whether PI3Kβ-mediated Akt activation contributes to the chemotactic response of HUVEC towards S1P, we employed Akt inhibitor VIII, an Akt1/2 inhibitor, or a specific Akt1 siRNA in our study. Akt inhibitor VIII (20 µM) decreased S1P-induced Akt-phosphorylation (1 µM, 2 min) by 98 ± 1.1% (n = 4). Since Akt2 is not expressed in endothelial cells,22 these data strongly suggest that Akt1, the primary Akt isoform in endothelial cells, is stimulated by S1P. As a second approach, we reduced the level of expression of Akt1 by 77.9 ± 1.9% (n = 5) using siRNA (Supplementary material online, Figure S4A).

S1P-induced migration (1 µM) measured in the presence of the Akt1/2 inhibitor or following pre-treatment of HUVEC with Akt1 siRNA was decreased by 54 or 55%, respectively (Figure 2D), pointing to a role of Akt1 in mediating the migratory response. Interestingly, Akt seems to affect S1P-induced migration independent of the small GTPase Rac1 since neither inhibition nor downregulation of Akt1 decreased S1P-triggered activation of Rac1 (Figure 2E and F).

3.5 S1P-stimulated eNOS phosphorylation does not contribute to S1P-induced migration
Akt has been reported to phosphorylate eNOS at serine 1177 and thereby promote NO formation and support growth factor-stimulated migration.8,23 eNOS phosphorylation at serine 1177 was time-dependently stimulated by 1 µM S1P (Supplementary material online, Figure S2B). The effect of S1P was blocked by pertussis toxin, inhibited by wortmannin and TGX-221 (42 and 38%, respectively), but not affected by AS-252424 (Figure 3A). Accordingly, pre-treatment of HUVEC with PI3Kβ-specific siRNA significantly decreased S1P-induced eNOS phosphorylation at serine 1177 (60%), whereas PI3K{gamma} downregulation in HUVEC or the absence of PI3K{gamma} in MLEC was without effect (Figure 3B and C). These data indicate that S1P stimulates eNOS phosphorylation partially via a PI3Kβ/Akt-dependent pathway whereas PI3K{gamma} is not involved.


Figure 3
View larger version (35K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3 Role of the PI3Kβ isoform in S1P-stimulated eNOS phosphorylation and the function of eNOS in migration. (A–C) HUVEC or MLEC from wild-type or PI3K{gamma}-deficient mice were stimulated with S1P (1 µM, 2 min). eNOS was analysed using antibodies against phosphorylated eNOS (serine 1177) and total eNOS. (A) HUVEC were pre-incubated with pertussis toxin (0.1 µg/mL, 3 h) or with PI3K inhibitors (wortmannin and TGX-221: 0.1 µM, AS-252424: 1 µM, 30 min). (B) HUVEC were pre-treated with PI3Kβ- or PI3K{gamma}-specific or control siRNA (1.0 µg/30 mm dish, 72 h). (A–C) Typical experiments and densitometry analyses are shown [mean ± SEM, n = 4 (A and B) or n = 3 (C)]. Phospho-eNOS signals in S1P-stimulated cells with or without inhibitor pre-treatment, RNA interference, or PI3K{gamma}-knockdown were compared. *P < 0.05. (D) HUVEC were seeded on upper wells of migration chambers, pre-treated with L-NAME (1 mM, 30 min), and stimulated by adding 1 µM S1P to the bottom. Alternatively, cells were transfected with eNOS-specific or control siRNA (1 µg/30 mm dish, 72 h) before measuring migration. Four hours after seeding, migrated cells were counted and S1P-induced migration was calculated by subtracting the respective control values (means ± SEM). Cells with and without L-NAME or siRNA pre-treatment (n = 6 or n = 4, respectively) were compared. *P < 0.05.

 
To investigate the possible role of NO in mediating migration of endothelial cells, we inhibited eNOS using L-NAME (1 mM). L-NAME had no effect on the migration of HUVEC induced by S1P (1 µM) (Figure 3D). Moreover, downregulation of eNOS in HUVEC using RNA interference (85 ± 2.7% reduction, n = 5, Supplementary material online, Figure S4B) did not decrease but even increase S1P-triggered chemotaxis suggesting an inhibitory function of the protein (Figure 3D).

3.6 PI3Kβ and PI3K{gamma} mediate S1P-induced activation of the small GTPase Rac
Rac has been proposed to play a major role in mediating S1P-induced cell motility.24,25 Indeed, a time-dependent Rac1 activation by S1P (1 µM) was observed in HUVEC in our study (Supplementary material online, Figure S2C). Moreover, the Rac1 inhibitor NSC-23766 (100 µM) led to an inhibition of endothelial migration towards S1P by 72% (Figure 4A).


Figure 4
View larger version (27K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4 Role of PI3Kβ and {gamma} isoforms in S1P-stimulated Rac activation and the effect of Rac inhibition on Akt activation and migration. (A) HUVEC were pre-treated with NSC-23766 (100 µM, 4 h) and seeded on upper wells of migration chambers. NSC-23766 was re-added to both compartments and S1P (1 µM, 4 h) was applied to the bottom. Migrated cells were counted and S1P-induced migration was calculated by subtracting the respective control values (means ± SEM, n = 4). Cells with and without NSC-23766 pre-treatment were compared. *P < 0.05. The insert shows the inhibitory effect of NSC-23766 on S1P-induced GTP-loading of Rac measured as described below. (B and C) HUVEC or MLEC from wild-type or PI3K{gamma}-deficient mice were pre-treated with 0.1 µM TGX-221 or 1 µM AS-252424 for 30 min, stimulated with S1P (1 µM, 1 min), and lysed in the presence of GST-PAK (see Methods). GTP-bound Rac was precipitated with glutathione-sepharose beads and immunostained using Rac1 antibody. Total Rac1 was stained in cell lysates. (D) HUVEC were pre-treated with NSC-23766 (100 µM, 4 h) and stimulated with 1 µM S1P at the indicated times. Akt was analysed using antibodies against phosphorylated Akt (serine 473) and total Akt. (B–D) Typical experiment and densitometry analyses (means ± SEM, n = 4) are shown. Rac1-GTP or phospho-Akt signals in S1P-stimulated cells with or without inhibitor pre-treatment or PI3K{gamma}-knockdown were compared. *P < 0.05.

 
To understand the role of PI3Kβ or PI3K{gamma} in Rac activation, we performed pull-down assays in which GTP-bound Rac was isolated from HUVEC pre-treated with isoform-specific inhibitors and stimulated with S1P. Inhibition of PI3Kβ by TGX-221 (0.1 µM) and of PI3K{gamma} by AS-252424 (1 µM) significantly impaired Rac1 activation induced by S1P (1 µM, 1 min) by 45 and 41%, respectively (Figure 4B). Further, Rac activation was reduced by 56% in MLEC derived from PI3K{gamma}–/– mice relative to controls (Figure 4C) confirming an essential role of PI3K{gamma}. In addition, pre-treatment with TGX-221 decreased S1P-induced Rac1 activation in MLEC from both PI3K{gamma}+/+ and PI3K{gamma}–/– mice emphasizing the importance of PI3Kβ (Figure 4C). Together, these data identify Rac1 as a common downstream effector for both PI3K isoforms. In contrast, Rac1 does not seem to act upstream of Akt, since Rac1 inhibition by NSC-23766 (100 µM) did not affect S1P-induced Akt phosphorylation in HUVEC (Figure 4D).

3.7 S1P-induced activation of Erk1/2 is independent of PI3Kβ or {gamma} and contributes to endothelial cell migration
S1P stimulation of HUVEC with 1 µM S1P leads to time-dependent stimulation of Erk1/2 (Supplementary material online, Figure S2D). The effect was blocked by pertussis toxin (0.1 µg/mL), but not affected by TGX-221 (0.1 µM) or AS-252424 (1 µM) (Figure 5A) and not altered in MLEC from PI3K{gamma}–/– mice (Figure 5B). Interestingly, inhibition of Erk1/2 activation by PD-98059 (10 µM), an inhibitor of the upstream kinase mitogen activated kinase kinase (MEK), reduced S1P-triggered endothelial migration by 49% indicating a role of Erk1/2 activation in this process (Figure 5C).


Figure 5
View larger version (45K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5 Role of PI3Kβ and {gamma} isoforms in S1P-stimulated Erk1/2 activation and the effect of Erk1/2 inhibition on migration. (A and B) HUVEC pre-treated with inhibitors (0.1 µM TGX-221, 1 µM AS-252424, 30 min) or MLEC from wild-type or PI3K{gamma}-deficient mice were stimulated with S1P (1 µM, 2 min). Erk1/2 was analysed using antibodies against phosphorylated Erk1/2 (threonine 202/tyrosine 204 of Erk1, threonine 185/tyrosine 187 of Erk2) and total Erk1/2. Typical experiments and densitometry analyses (means ± SEM, n = 4) are shown. Phospho-Erk1/2 signals in S1P-stimulated cells with or without inhibitor pre-treatment or PI3K{gamma}-knockdown were compared. *P < 0.05. (C) HUVEC were seeded on upper wells of migration chambers, pre-treated with PD-98059 (10 µM, 30 min), and stimulated by 1 µM S1P added to the lower well. After 4 h, migrated cells were counted and S1P-induced migration was calculated by subtracting the respective control values (means ± SEM, n = 4). Cells with and without PD-98059 pre-treatment were compared. *P < 0.05.

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
Supplementary material
 Funding
 References
 
Our study in human and mouse endothelial cells demonstrates that both Gi protein sensitive PI3K species β and {gamma} contribute to S1P-stimulated migration. Specific inhibition of either isoform or genetic ablation of PI3K{gamma} resulted in a significantly reduced activation of Rac1 and an impaired capacity of endothelial cells to migrate in response to S1P. In addition, our data reveal that PI3Kβ but not PI3K{gamma} stimulates Akt and that the PI3Kβ/Akt pathway contributes to S1P-induced migration independent from Rac1 and eNOS activation. S1P-induced Erk1/2 activation, on the contrary, is not mediated by PI3Kβ or {gamma} but participates in the migratory response.

Both PI3Kβ and PI3K{gamma} isoforms have been identified in endothelial cells,25–28 but, so far, their functions are poorly characterized. PI3Kβ has been thought to stimulate the Akt/eNOS pathway,26 whereas PI3K{gamma} has been shown to be required for selectin-mediated neutrophil adhesion27 and for TNF{alpha}-induced NADPH oxidase activation and ICAM-1 expression.28 To our knowledge, our study is the first to describe a coordinated role of PI3Kβ and PI3K{gamma} in endothelial cell migration towards a chemotactic gradient. The requirement of both isoforms points to differential cellular functions of these enzymes as described in other cells types. PI3K{gamma}, for example, has been shown to define the direction of migration in leukocytes and epithelial cells,29,30 whereas PI3Kβ has been proposed to regulate cell adhesion via integrins in platelets.17 Our study has used pharmacological and siRNA approaches to demonstrate that PI3Kβ has a distinct function in mediating S1P-triggered Akt phosphorylation and Akt-mediated migration. Previous investigations have already shown that PI3Kβ is able to transmit signals from G protein-coupled receptors to Akt31,32 and that S1P stimulates PI3Kβ activity and Gβ{gamma}-dependent Akt activation in bovine aortic endothelial cells (which lack PI3K{gamma}).26 Our data extend the latter observations and show a direct link between PI3Kβ and Akt phosphorylation using PI3Kβ-specific tools. In addition, our results indicate that PI3Kβ is activated via a Gβ{gamma}-dependent pathway rather than via transactivation of receptor tyrosine kinases21 since neither specific inhibition of the EGF receptor or of the VEGF receptor-2 nor general tyrosine kinase inhibition affected S1P-induced Akt phosphorylation.

Prevention of Akt phosphorylation by an Akt1/2 inhibitor or downregulation of Akt1 by specific siRNA led to a significant decrease of S1P-induced migration underlining the importance of the PI3Kβ/Akt pathway and pointing to Akt1 as the responsible isoform. Interestingly, although eNOS was partially phosphorylated via the PI3Kβ/Akt pathway in our study, it did not mediate the effect of Akt on S1P-induced migration. Neither eNOS inhibition by L-NAME nor downregulation of eNOS by specific siRNA impaired the chemotactic effect of S1P. This is in agreement with a previous study and supports the view that in contrast to its role in VEGF-induced migration the eNOS/NO pathway is of minor importance in promoting S1P-induced chemotaxis.8 Akt effects in S1P-induced migration were also independent of Rac1. In contrast, a previous report in CHO cells transfected with the receptor S1P1 suggested that Akt mediates Rac activation via phosphorylation of S1P1.7 Our data demonstrate that neither pharmacological inhibition nor downregulation of Akt1 via RNA interference inhibited Rac1 activation in endothelial cells. Thus, other Akt targets involved in cytoskeleton remodelling (Girdin, PAK1, glycogen synthase kinase 3) or integrin transport may be involved in the chemotactic response towards S1P.33

Our study confirms that Rac1, activated via Akt-independent pathways, is a major effector molecule in S1P-triggered endothelial migration. Inhibition of Rac1 by NSC-23766 resulted in a significant inhibition of the chemotactic response towards S1P. Interestingly, Rac1 appears to be a downstream target of both PI3Kβ and PI3K{gamma} since specific inhibition of either isoform as well as knockout of PI3K{gamma} suppresses S1P-induced Rac1 activation. These findings are in line with previous studies showing that Rac1 activation was dependent on PI3K stimulation.24,25 Conversely, inhibition of Rac1 by NSC-23766 did not affect S1P-induced Akt phosphorylation indicating that Rac activation is not a proximal event in S1P-mediated PI3K/Akt signalling as suggested recently.34 PI3Kβ and PI3K{gamma} may stimulate Rac1 via the activation of PIP3-sensitive GEFs35 including common and/or distinct pathways. Notably, Rac activation by S1P has been shown to be mediated via Tiam1 or P-Rex2b24,25,34 and the latter was suggested to involve PI3K{gamma}.24

S1P-induced migration was only partially blocked by PI3K inhibition in our study suggesting the involvement of PI3K-independent pathways. One of these mechanisms might include Erk1/2 activation which was not dependent on PI3Kβ or {gamma} activities but contributed to the chemotactic effect of S1P. Inhibition of MEK, the upstream kinase of Erk1/2, led to a significant inhibition of S1P-stimulated cell migration in our experiments. A role of Erk1/2 in promoting S1P effects on cell movement has not been observed in an earlier study,36 but likely involves the substrates myosin light chain kinase, calpain, and/or focal adhesion kinase as suggested recently.37

In summary, our study shows that both PI3Kβ and PI3K{gamma} are required for endothelial migration towards S1P but their downstream signalling pathways and functions differ. PI3Kβ mediates the phosphorylation of Akt and initiates an Akt-sensitive part of cell migration which is independent from eNOS and Rac1 activation. In addition, both PI3Kβ and PI3K{gamma} mediate endothelial migration via the activation of Rac1. The characterization of these pathways may reveal novel approaches for the modulation of the S1P response and therapies affecting angiogenesis.


   Supplementary material
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Supplementary material
Funding
 References
 
Supplementary material is available at Cardiovascular Research online.


   Funding
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Supplementary material
 Funding
References
 
Deutsche Forschungsgemeinschaft (SFB 604 to R.W., KU-1206 (1-2) to P.K.). Interdisziplinäres Zentrum für Klinische Forschung (IZKF Jena, TP 4.6 to R.H., IZKF Würzburg to P.K.).


   Acknowledgements
 
We thank Gunda Guhr, Elke Teuscher, and Gabriele Riehl for their excellent technical assistance. We are grateful to Michael Gruen and Christian Koenig for supporting mice experiments and to Nadine Stahmann for flow cytometric analyses.

Conflict of interest: none declared.


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

  1. Yatomi Y. Sphingosine 1-phosphate in vascular biology: possible therapeutic strategies to control vascular diseases. Curr Pharm Des (2006) 12:575–587.[CrossRef][Web of Science][Medline]
  2. Peters SL, Alewijnse AE. Sphingosine-1-phosphate signaling in the cardiovascular system. Curr Opin Pharmacol (2007) 7:186–192.[CrossRef][Web of Science][Medline]
  3. Wang F, Van Brocklyn JR, Hobson JP, Movafagh S, Zukowska-Grojec Z, Milstien S, et al. Sphingosine 1-phosphate stimulates cell migration through a G(i)-coupled cell surface receptor. Potential involvement in angiogenesis. J Biol Chem (1999) 274:35343–35350.[Abstract/Free Full Text]
  4. Paik JH, Chae SS, Lee MJ, Thangada S, Hla T. Sphingosine 1-phosphate-induced endothelial cell migration requires the expression of EDG-1 and EDG-3 receptors and Rho-dependent activation of alphavbeta3- and beta1-containing integrins. J Biol Chem (2001) 276:11830–11837.[Abstract/Free Full Text]
  5. Vouret-Craviari V, Bourcier C, Boulter E, van Obberghen-Schilling E. Distinct signals via Rho GTPases and Src drive shape changes by thrombin and sphingosine-1-phosphate in endothelial cells. J Cell Sci (2002) 115:2475–2484.[Abstract/Free Full Text]
  6. Lee JF, Ozaki H, Zhan X, Wang E, Hla T, Lee MJ. Sphingosine-1-phosphate signaling regulates lamellipodia localization of cortactin complexes in endothelial cells. Histochem Cell Biol (2006) 126:297–304.[CrossRef][Web of Science][Medline]
  7. Lee MJ, Thangada S, Paik JH, Sapkota GP, Ancellin N, Chae SS, et al. Akt-mediated phosphorylation of the G protein-coupled receptor EDG-1 is required for endothelial cell chemotaxis. Mol Cell (2001) 8:693–704.[CrossRef][Web of Science][Medline]
  8. Morales-Ruiz M, Lee MJ, Zollner S, Gratton JP, Scotland R, Shiojima I, et al. Sphingosine 1-phosphate activates Akt, nitric oxide production, and chemotaxis through a Gi protein/phosphoinositide 3-kinase pathway in endothelial cells. J Biol Chem (2001) 276:19672–19677.[Abstract/Free Full Text]
  9. Rikitake Y, Hirata K, Kawashima S, Ozaki M, Takahashi T, Ogawa W, et al. Involvement of endothelial nitric oxide in sphingosine-1-phosphate-induced angiogenesis. Arterioscler Thromb Vasc Biol (2002) 22:108–114.[Abstract/Free Full Text]
  10. English D, Welch Z, Kovala AT, Harvey K, Volpert OV, Brindley DN, et al. Sphingosine 1-phosphate released from platelets during clotting accounts for the potent endothelial cell chemotactic activity of blood serum and provides a novel link between hemostasis and angiogenesis. FASEB J (2000) 14:2255–2265.[Abstract/Free Full Text]
  11. Hawkins PT, Anderson KE, Davidson K, Stephens LR. Signalling through Class I PI3Ks in mammalian cells. Biochem Soc Trans (2006) 34:647–662.[CrossRef][Web of Science][Medline]
  12. Hazeki O, Okada T, Kurosu H, Takasuga S, Suzuki T, Katada T. Activation of PI 3-kinase by G protein betagamma subunits. Life Sci (1998) 62:1555–1559.[CrossRef][Web of Science][Medline]
  13. Maier U, Babich A, Nurnberg B. Roles of non-catalytic subunits in gbetagamma-induced activation of class I phosphoinositide 3-kinase isoforms beta and gamma. J Biol Chem (1999) 274:29311–29317.[Abstract/Free Full Text]
  14. Ferguson GJ, Milne L, Kulkarni S, Sasaki T, Walker S, Andrews S, et al. PI(3)Kgamma has an important context-dependent role in neutrophil chemokinesis. Nat Cell Biol (2007) 9:86–91.[CrossRef][Web of Science][Medline]
  15. Leopoldt D, Hanck T, Exner T, Maier U, Wetzker R, Nürnberg B. Gbetagamma stimulates phosphoinositide 3-kinase-gamma by direct interaction with two domains of the catalytic p110 subunit. J Biol Chem (1998) 273:7024–7029.[Abstract/Free Full Text]
  16. Condliffe AM, Davidson K, Anderson KE, Ellson CD, Crabbe T, Okkenhaug K, et al. Sequential activation of class IB and class IA PI3K is important for the primed respiratory burst of human but not murine neutrophils. Blood (2005) 106:1432–1440.[Abstract/Free Full Text]
  17. Jackson SP, Schoenwaelder SM, Goncalves I, Nesbitt WS, Yap CL, Wright CE, et al. PI 3-kinase p110beta: a new target for antithrombotic therapy. Nat Med (2005) 11:507–514.[CrossRef][Web of Science][Medline]
  18. Hirsch E, Katanaev VL, Garlanda C, Azzolino O, Pirola L, Silengo L, et al. Central role for G protein-coupled phosphoinositide 3-kinase gamma in inflammation. Science (2000) 287:1049–1053.[Abstract/Free Full Text]
  19. Stahmann N, Woods A, Carling D, Heller R. Thrombin activates AMP-activated protein kinase in endothelial cells via a pathway involving Ca2+/calmodulin-dependent protein kinase kinase beta. Mol Cell Biol (2006) 26:5933–5945.[Abstract/Free Full Text]
  20. Kuhlencordt PJ, Rosel E, Gerszten RE, Morales-Ruiz M, Dombkowski D, Atkinson WJ, et al. Role of endothelial nitric oxide synthase in endothelial activation: insights from eNOS knockout endothelial cells. Am J Physiol Cell Physiol (2004) 286:C1195–C1202.[Abstract/Free Full Text]
  21. Tanimoto T, Jin ZG, Berk BC. Transactivation of vascular endothelial growth factor (VEGF) receptor Flk-1/KDR is involved in sphingosine 1-phosphate-stimulated phosphorylation of Akt and endothelial nitric-oxide synthase (eNOS). J Biol Chem (2002) 277:42997–43001.[Abstract/Free Full Text]
  22. Fieber CB, Eldridge J, Taha TA, Obeid LM, Muise-Helmericks RC. Modulation of total Akt kinase by increased expression of a single isoform: requirement of the sphingosine-1-phosphate receptor, Edg3/S1P3, for the VEGF-dependent expression of Akt3 in primary endothelial cells. Exp Cell Res (2006) 312:1164–1173.[CrossRef][Web of Science][Medline]
  23. Dimmeler S, Dernbach E, Zeiher AM. Phosphorylation of the endothelial nitric oxide synthase at ser-1177 is required for VEGF-induced endothelial cell migration. FEBS Lett (2000) 477:258–262.[CrossRef][Web of Science][Medline]
  24. Li Z, Paik JH, Wang Z, Hla T, Wu D. Role of guanine nucleotide exchange factor P-Rex-2b in sphingosine 1-phosphate-induced Rac1 activation and cell migration in endothelial cells. Prostaglandins Other Lipid Mediat (2005) 76:95–104.[Web of Science][Medline]
  25. Singleton PA, Dudek SM, Chiang ET, Garcia JG. Regulation of sphingosine 1-phosphate-induced endothelial cytoskeletal rearrangement and barrier enhancement by S1P1 receptor, PI3 kinase, Tiam1/Rac1, and alpha-actinin. FASEB J (2005) 19:1646–1656.[Abstract/Free Full Text]
  26. Igarashi J, Michel T. Sphingosine 1-phosphate and isoform-specific activation of phosphoinositide 3-kinase beta. Evidence for divergence and convergence of receptor-regulated endothelial nitric-oxide synthase signaling pathways. J Biol Chem (2001) 276:36281–36288.[Abstract/Free Full Text]
  27. Puri KD, Doggett TA, Huang CY, Douangpanya J, Hayflick JS, Turner M, et al. The role of endothelial PI3Kgamma activity in neutrophil trafficking. Blood (2005) 106:150–157.[Abstract/Free Full Text]
  28. Frey RS, Gao X, Javaid K, Siddiqui SS, Rahman A, Malik AB. Phosphatidylinositol 3-kinase gamma signaling through protein kinase Czeta induces NADPH oxidase-mediated oxidant generation and NF-kappaB activation in endothelial cells. J Biol Chem (2006) 281:16128–16138.[Abstract/Free Full Text]
  29. Hannigan M, Zhan L, Li Z, Ai Y, Wu D, Huang CK. Neutrophils lacking phosphoinositide 3-kinase gamma show loss of directionality during N-formyl-Met-Leu-Phe-induced chemotaxis. Proc Natl Acad Sci USA (2002) 99:3603–3608.[Abstract/Free Full Text]
  30. Zhao M, Song B, Pu J, Wada T, Reid B, Tai G, et al. Electrical signals control wound healing through phosphatidylinositol-3-OH kinase-gamma and PTEN. Nature (2006) 442:457–460.[CrossRef][Medline]
  31. Murga C, Fukuhara S, Gutkind JS. A novel role for phosphatidylinositol 3-kinase beta in signaling from G protein-coupled receptors to Akt. J Biol Chem (2000) 275:12069–12073.[Abstract/Free Full Text]
  32. Kubo H, Hazeki K, Takasuga S, Hazeki O. Specific role for p85/p110beta in GTP-binding-protein-mediated activation of Akt. Biochem J (2005) 392:607–614.[CrossRef][Web of Science][Medline]
  33. Stambolic V, Woodgett JR. Functional distinctions of protein kinase B/Akt isoforms defined by their influence on cell migration. Trends Cell Biol (2006) 16:461–466.[CrossRef][Web of Science][Medline]
  34. Gonzalez E, Kou R, Michel T. Rac1 modulates sphingosine 1-phosphate-mediated activation of phosphoinositide 3-kinase/Akt signaling pathways in vascular endothelial cells. J Biol Chem (2006) 281:3210–3216.[Abstract/Free Full Text]
  35. Welch HC, Coadwell WJ, Stephens LR, Hawkins PT. Phosphoinositide 3-kinase-dependent activation of Rac. FEBS Lett (2003) 546:93–97.[CrossRef][Web of Science][Medline]
  36. Kimura T, Watanabe T, Sato K, Kon J, Tomura H, Tamama K, et al. Sphingosine 1-phosphate stimulates proliferation and migration of human endothelial cells possibly through the lipid receptors, Edg-1 and Edg-3. Biochem J (2000) 348:71–76.[CrossRef][Web of Science][Medline]
  37. Huang C, Jacobson K, Schaller MD. MAP kinases and cell migration. J Cell Sci (2004) 117:4619–4628.[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
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
80/1/96    most recent
cvn159v2
cvn159v1
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
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
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Heller, R.
Right arrow Articles by Wetzker, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Heller, R.
Right arrow Articles by Wetzker, R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?