Skip Navigation


Cardiovascular Research Advance Access first published online on July 3, 2008
This version [Corrected Proof] published online on July 29, 2008
Cardiovascular Research, doi:10.1093/cvr/cvn180
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
80/1/47    most recent
cvn180v2
cvn180v1
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 Larsen, K.-O.
Right arrow Articles by Christensen, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Larsen, K.-O.
Right arrow Articles by Christensen, G.
Related Collections
Right arrowRelated Article
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

Diastolic dysfunction in alveolar hypoxia: a role for interleukin-18-mediated increase in protein phosphatase 2A

Karl-Otto Larsen1,2,3,*, Birgitte Lygren4, Ivar Sjaastad2,3,5, Kurt A. Krobert3,6, Kristin Arnkværn2,3, Geir Florholmen2,3, Ann-Kristin Ruud Larsen7, Finn Olav Levy3,6, Kjetil Taskén4, Ole Henning Skjønsberg1 and Geir Christensen2,3

1 Department of Pulmonary Medicine, Ullevål University Hospital, University of Oslo, Oslo, Norway
2 Institute for Experimental Medical Research, Surgical Building 4th floor, Ullevål University Hospital, Kirkeveien 166, N-0407 Oslo, Norway
3 Center for Heart Failure Research, University of Oslo, Oslo, Norway
4 The Biotechnology Centre of Oslo and Centre for Molecular Medicine Norway, Nordic EMBL Partnership, University of Oslo, Oslo, Norway
5 Department of Cardiology, Ullevål University Hospital, Oslo, Norway
6 Department of Pharmacology, University of Oslo, Oslo, Norway
7 Department of Immunology and Transfusion Medicine, Ullevål University Hospital, Oslo, Norway

* Corresponding author. Tel: +47 23016800; fax: +47 23016799.E-mail address: karlottl{at}medisin.uio.no

Received 16 November 2007; revised 13 June 2008; accepted 24 June 2008

Time for primary review: 17 days


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Funding
 References
 
Aims: Chronic obstructive pulmonary disease with alveolar hypoxia is associated with diastolic dysfunction in the right and left ventricle (LV). LV diastolic dysfunction is not caused by increased afterload, and we recently showed that reduced phosphorylation of phospholamban at serine (Ser) 16 may explain the reduced relaxation of the myocardium. Here, we study the mechanisms leading to the hypoxia-induced reduction in phosphorylation of phospholamban at Ser16.

Methods and results: In C57Bl/6j mice exposed to 10% oxygen, signalling molecules were measured in cardiac tissue, sarcoplasmic reticulum (SR)-enriched membrane preparations, and serum. Cardiomyocytes isolated from neonatal mice were exposed to interleukin (IL)-18 for 24 h. The β-adrenergic pathway in the myocardium was not altered by alveolar hypoxia, as assessed by measurements of β-adrenergic receptor levels, adenylyl cyclase activity, and subunits of cyclic AMP-dependent protein kinase. However, alveolar hypoxia led to a significantly higher amount (124%) and activity (234%) of protein phosphatase (PP) 2A in SR-enriched membrane preparations from LV compared with control. Serum levels of an array of cytokines were assayed, and a pronounced increase in IL-18 was observed. In isolated cardiomyocytes, treatment with IL-18 increased the amount and activity of PP2A, and reduced phosphorylation of phospholamban at Ser16 to 54% of control.

Conclusion: Our results indicate that the diastolic dysfunction observed in alveolar hypoxia might be caused by increased circulating IL-18, thereby inducing an increase in PP2A and a reduction in phosphorylation of phospholamban at Ser16.

KEYWORDS Protein phosphatases; Phospholamban; Hypoxia; Cytokines


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 Funding
 References
 
Cardiac dysfunction is frequently observed in chronic obstructive pulmonary disease (COPD), which is one of the leading causes of disability and death worldwide. Chronic hypoxia is a risk factor for COPD patients associated with an increased risk of death,1 and not only right, but also left ventricular (LV) diastolic dysfunction.2 However, little information is available regarding the molecular mechanisms for the diastolic dysfunction in chronic hypoxia. In mice subjected to chronic hypoxia, we have recently shown that LV diastolic dysfunction is associated with slower relaxation of the myocardium.3 The rate of myocardial relaxation is primarily controlled by the activity of the sarcoplasmic reticulum (SR) Ca2+ ATPase (SERCA), which pumps Ca2+ from the cytosol into the SR. Phospholamban regulates SERCA activity, and phosphorylation at the serine (Ser) 16 residue of phospholamban by cyclic AMP-dependent protein kinase (PKA) enhances relaxation rates and contractility. In our previous study, we showed that a reduction in phosphorylation of phospholamban at Ser16 in the myocardium may be an important mechanism for impaired relaxation and diastolic dysfunction in both the right ventricle (RV) and the LV in mice exposed to chronic hypoxia.3

Reduced phosphorylation of phospholamban at Ser16 may be due either to impaired phosphorylation of the protein through the β-adrenergic pathway, involving the β-adrenoceptors, adenylyl cyclases, and PKA, or to increased dephosphorylation by protein phosphatases (PPs). It is not known which of these mechanisms reduces phosphorylation of phospholamban in alveolar hypoxia. Since reduced phosphorylation of phospholamban is present in both the RV and LV, the latter not exposed to increased afterload, an increase in circulating mediators (e.g. cytokines) could contribute to the changes found in both ventricles.

The aim of the present study was to identify mechanisms reducing phospholamban phosphorylation in alveolar hypoxia. In addition, we wanted to examine if the observed mechanism can be regulated by a circulating mediator.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 Funding
 References
 
The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication No. 85-23, revised 1996) and approved by the Norwegian National Animal Research Committee. The animals were housed with a day/night cycle of 12/12 h at 21°C, and food and water were available ad libitum.

2.1 Animal model
A total of 250 8-week-old male C57Bl/6j mice were either placed in a tightly sealed chamber under normobaric hypoxia with 10% oxygen (hypoxia group) for 1, 2, 7, or 14 days or housed under normoxic conditions (control group).4 The number of samples in the treatment and control groups is given as n/n. Previous studies have shown that mice breathing 10% oxygen develop pulmonary hypertension3,4 and LV diastolic dysfunction.3 Blood samples were drawn from the inferior vena cava under inhalation anaesthesia with a mixture of 2% isoflurane and 98% oxygen.3 Heart and lungs were excised, and the RV and LV (free walls) were separated from the septum and weighed.

2.2 Histology
For histological examinations, hearts from mice subjected to 14 days of hypoxia and hearts from controls were fixed in 4% paraformaldehyde and embedded in paraffin. The hearts were sectioned transversely, stained with haematoxylin and eosin, acid fuchsin orange G-stain (AFOG), and Masson trichrome, and examined in a blinded manner by two pathologists.3

2.3 Preparation of subcellular fractions
Homogenates of the RV and LV were made from hearts of mice exposed to 14 days of hypoxia and controls. SR-enriched membrane preparations were obtained from the homogenates by sucrose step gradient centrifugation,5 using an Optima Max tabletop ultracentrifuge, TLS-55 rotor, 100 000 g for 1 h at 4°C (Beckman Instruments, Palo Alto, CA, USA). These SR-enriched membranes form a layer at the interface between 24% and 40% sucrose. It has in previous studies by us6 and others5 been shown that these preparations have a high content of SR as assessed by calsequestrin content. Furthermore, the content of plasma membrane in these preparations is very low.6

2.4 Western blot and R-overlay analysis
SR-enriched membrane preparations of hearts from mice exposed to 14 days of hypoxia and controls, or homogenates of isolated neonatal cardiomyocytes, were subjected to western blotting using primary antibodies to phosphorylated phospholamban at Ser16 and threonine (Thr) 17 (Badrilla, Leeds, UK), phospholamban, SERCA2 (Affinity BioReagents, Golden, CO, USA), PP1{alpha}, PP2Ac, calsequestrin (Upstate, Lake Placid, NY, USA), PP2B, PKA regulatory (R) I{alpha}, RII{alpha}, and catalytic (C) subunits (BD Transduction Laboratories, KY, USA), phosphodiesterase 4A (pan-PDE4 antibody (K116)),7 total Akt (or protein kinase B) (Cell Signaling Technology, Danvers, MA, USA), collagen I and III (Rockland Immunochemicals, Gilbertsville, PA, USA), and the Na+/Ca2+ exchanger (NCX).8 Protein concentrations were determined by the bicinchoninic acid assay (Pierce, Rockford, IL, USA) or Bradford (Bio-Rad Laboratories, Hercules, CA). PKA-R-overlays were conducted using 32P-labelled recombinant murine RII{alpha}.

2.5 Protein phosphatase activity
PP activities in SR-enriched membrane preparations from the RV and LV from mice subjected to 14 days of hypoxia and from controls, and total homogenates of isolated cardiomyocytes, were measured by the Protein Serine/Threonine Phosphatase Assay System (New England BioLabs, Ipswich, MA, USA). The PP1 activity was measured using 33P-labelled glycogen phosphorylase a as substrate in the presence of 4 nM okadaic acid (Merck, Whitehouse Station, NJ, USA), which is an inhibitor of PP2A, and 0.5 mM EDTA, which removes divalent cations that activate PP2B and PP2C.9

2.6 Radioligand binding and adenylyl cyclase activity
Membranes were prepared from RV and LV from mice exposed to 14 days of hypoxia and controls as described,3 incubated with increasing concentrations of (–)-3-[125I]iodocyanopindolol in the absence (total binding) or presence (non-specific binding) of 10 µM propranolol for 1 h at 32°C. Specific β-adrenoceptor binding was determined using the same incubation buffers and filtration method as described for other receptors.10 Adenylyl cyclase activities were measured by determining conversion of [{alpha}-32P]ATP (Amersham Biosciences) to [32P]cAMP and after stimulation with 10 µM isoproterenol (Sigma-Aldrich, St Louis, MO, USA) or 100 µM forskolin (Calbiochem, San Diego, CA, USA).10

2.7 Quantitative real-time polymerase chain reaction
A real-time quantitative polymerase chain reaction (PCR) system (ABI 7900HT Fast Real-Time PCR System, PE Biosystems, Foster City, CA, USA) was used to measure the mRNA amounts of PP2A, collagen I and III, and the normalization gene RPL32 in RV and LV from mice exposed to 14 days of hypoxia and controls. Total mRNA was isolated by using spin or vacuum total RNA isolation system (Promega, Madison, WI, USA). All RNA samples were quality assessed by Agilent Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) and RNA integrity numbers. The RNA samples were reverse transcribed by using iScript cDNA Synthesis Kit (Bio-Rad). Specific mRNA transcripts were quantified by Taqman GX assays (Applied Biosystems, Foster City, CA, USA) for PP2A (assay Mm00479816_m1, efficiency 1.88), collagen I (assay Mm00483888_m1, efficiency 1.94), collagen III (assay Mm00802331_m1, efficiency 1.87), and RPL32 (assay Mm02528467_g1, efficiency 1.94). All samples were tested in triplicate, and average values were used for quantification. Average values for PP2A and collagen I and III mRNA were normalized to RPL32 mRNA.

2.8 Luminex cytokine assays and enzyme-linked immunosorbent assay
The serum levels of 24 cytokines were assayed with Luminex cytokine assay (Bio-Rad) or ELISA (R&D Systems, MN, USA) after 1 day of hypoxia and in controls. In addition, the levels of interleukin (IL)-18 were measured with ELISA (R&D Systems) after 2, 7, and 14 days of hypoxia.

2.9 Isolation of neonatal mouse cardiomyocytes and exposure to hypoxia or cytokines
Cardiomyocytes were isolated from 1- to 3-day-old C57Bl/6j mice using a modified version of a previously described protocol.11 The purity of the cardiomyocyte cultures was >98%, assessed by DAPI nuclear staining and myosin light chain-2V (Synaptic Systems GmbH, Goettingen, Germany). Neonatal cardiomyocytes respond with an increase in Ser16 phosphorylation similar to adult cardiomyocytes.12,13 The neonatal cardiomyocytes were treated with recombinant IL-18 dissolved in distilled water (1000 ng/mL; R&D Systems) or vehicle (distilled water) for 24 h, or exposed to 1% oxygen for 24 h. The cells were kept in Dulbecco’s modified Eagle’s medium (Gibco-BRL) supplemented with medium 199, 1 M HEPES (Gibco-BRL), and penicillin/streptomycin/glutamine (Sigma). An IL-18 concentration of 1000 ng/mL was used based on information from previous cell studies with IL-18.14 Hypoxic cultures were incubated in an Invivo2 400 hypoxic workstation (Ruskinn, Cincinnatti, OH, USA) in a gas mixture containing 4% CO2, 1% O2, and balanced with N2. An oxygen level of 1% has been shown to induce cellular hypoxia with an increase in the hypoxia inducible transcription factor-1{alpha},15 without causing increased cardiomyocyte cell death (97% viable cells), as assessed by exclusion of trypan blue. Cardiomyocytes were treated for 24 h to allow synthesis of phosphatases as shown in previous studies on myoblasts.16 For measurements of phosphorylation of phospholamban at Ser16 in IL-18 treated cells, cardiomyocytes were stimulated with 1 nM isoproterenol for 5 min before harvesting. Based on our own preliminary experiments, and previous studies,12,13 a concentration of 1 nM was used as it induced a marked increase in Ser16 phosphorylation of phospholamban. In each of the above-mentioned conditions, samples from the experiments were analysed in duplicate or triplicate. Results from western blotting were normalized to calsequestrin. Cardiomyocytes were obtained in separate cell isolations on several days to avoid possible errors arising from imperfect cells on 1 day. The n given in the Results section denotes the number of experiments performed. A possible limitation of the cell experiments can be that isolated neonatal cardiomyocytes respond differently than adult cardiomyocytes in a beating heart. However, both protein kinases and phosphatases are active in isolated neonatal cardiomyocytes, and phosphorylation of phospholamban occurs in response to β-adrenergic stimulation as in adult cells.12


Figure 5
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 5 (A) Protein phosphatase 2A (PP2A) abundance (n = 6/6) and (B) relative PP2A activity (n = 6/6) in homogenates of isolated cardiomyocytes exposed to IL-18 for 24 h (open bar) or vehicle (filled bar). (C) Phosphorylation of phospholamban (PLB) at serine 16 (PS16) in isolated cardiomyocytes stimulated with isoproterenol (1 nM) after exposure to IL-18 (open bar) or vehicle (filled bar) for 24 h (n = 6/6), and (D) representative western blot showing PS16-PLB. *P < 0.05 vs. control.

 
2.10 Data analysis and statistics
Data are presented as means ± SEM. Comparisons between groups were made using unpaired Student’s t-test or the non-parametric Mann–Whitney rank sum test in SigmaStat 3.1.1 (Systat Software, Richmond, CA, USA), or ANOVA followed by Student–Newman–Keul’s post hoc test. Differences were considered significant for P < 0.05.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 Funding
 References
 
3.1 Body, heart, and lung weights of mice exposed to alveolar hypoxia
The ratio RV weight/tibial length (TL) increased by 35% in mice exposed to 14 days of hypoxia compared with controls (P < 0.05, Table 1). There were no significant differences between the hypoxia and the control group with regard to the LV weight normalized to TL. The lung weight normalized to TL increased by 36% in the hypoxia group compared with controls (P < 0.05, Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1 Organ weights at 14 days of hypoxia

 
3.2 Levels of Ca2+-handling proteins
The levels of proteins involved in the regulation of Ca2+ extrusion from the cytosol were measured in SR-enriched membrane preparations from RV and LV. Mice exposed to hypoxia had reduced phosphorylation of phospholamban at Ser16 in both the RV and LV amounting to 84 ± 6% (P < 0.05, n = 7/7) and 59 ± 9% (P < 0.05, n = 7/7) of controls, respectively (Figure 1A and B). Neither in the RV nor in the LV, significant changes in total phospholamban (RV 102 ± 5%, n = 5/5; LV 106 ± 12%, n = 5/5) or its phosphorylation at Thr17 (RV 98 ± 10%, n = 7/7; LV 121 ± 16%, n = 7/7) were observed compared with controls. The amounts of SERCA2 were not significantly changed in the RV or LV compared with controls (Figure 1C, n = 7/7 for RV and 5/5 for LV). The protein levels of NCX were increased to 152 ± 17% of control in the RV (P < 0.05, n = 7/7) and to 147 ± 10% in the LV (P < 0.05, n = 7/7, Figure 1D).


Figure 1
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1 (A) Abundance of phospholamban (PLB) phosphorylated at serine 16 (PS16) in isolated sarcoplasmic reticulum-enriched membrane preparations from the right ventricle (RV) (n = 7/7) and left ventricle (LV) (n = 7/7), and (B) representative western blot showing PS16-PLB. (C) Sarcoplasmic reticulum Ca2+ ATPase (SERCA2) (n = 7/7 for RV and n = 5/5 for LV) and (D) Na+/Ca2+-exchanger (NCX) protein abundance (n = 7/7 for RV and LV). Immunolabelling density for alveolar hypoxia (open bars) was compared with the control groups (filled bars), which was set to 100%. *P < 0.05 vs. control.

 
3.3 β-Adrenergic receptor levels and adenylyl cyclase activity
Possible mechanisms leading to reduced phosphorylation of phospholamban at Ser16 were explored by examining molecules in the β-adrenergic signalling cascade. Exposure to hypoxia did not alter the levels of β-adrenoceptors in the RV or LV (Table 2). Moreover, hypoxia did not induce any changes in the basal, isoproterenol-stimulated or forskolin-stimulated adenylyl cyclase activities in the RV or LV (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2 β-Adrenergic signaling at 14 days of hypoxia

 
3.4 PDE4, PKA-subunits and Akt
The β-adrenergic signalling cascade downstream of adenylyl cyclase was also examined. PDE4, as well as PDE3, represents the major cAMP hydrolytic activities in cardiomyocytes.17 However, in SR-enriched membrane preparations, hypoxia did not induce any significant changes in PDE4 in the RV (112 ± 5%, n = 7/7) or LV (99 ± 5%, n = 7/7) compared with controls (Table 2). In addition, alveolar hypoxia did not change the amount of PKA-C in the RV (110 ± 10%, n = 7/7) or LV (102 ± 13%, n = 7/7, Table 2). Neither were the levels of PKA-RI{alpha} and PKA-RII{alpha} in the RV (125 ± 10%, n = 7/7 and 110 ± 9%, n = 7/7) nor LV significantly altered (109 ± 6%, n = 7/7 and 109 ± 12%, n = 7/7, Table 2). To investigate whether PKA has altered binding to A-kinase anchoring proteins in the hypoxia group, PKA-RII{alpha}-overlay was performed. No significant changes in A-kinase anchoring protein levels were found. Akt is an important signal transduction molecule in IL-18-mediated gene transcription,18 and hypoxia induced an increase in cytosolic Akt in the RV to 116 ± 5% (P < 0.05, n = 7/7) and in the LV to 110 ± 5% (P < 0.05, n = 7/7) of respective controls.

3.5 Protein phosphatase levels
To explore if changes in the levels of PPs could explain reduced phosphorylation of phospholamban at Ser16, PP2A and PP1 levels were measured in SR-enriched membrane preparations. Exposure to hypoxia induced an increase in the amount of PP2A in the RV to 122 ± 6% (P < 0.05, n = 7/6) and in the LV to 124 ± 12% (P < 0.05, n = 7/6) of respective controls (Figure 2A and B). Moreover, the levels of PP1 in the hypoxia group were increased to 143 ± 16% (P < 0.05, n = 7/7) and 177 ± 27% (P < 0.05, n = 7/7) of control in the RV and LV, respectively (Figure 2C). In mice subjected to hypoxia, the amount of PP2B (calcineurin), an important mediator of cardiac hypertrophy, increased to 141 ± 19% of control in the hypertrophic RV (P < 0.05, n = 7/7, Figure 2D). No significant alteration in the level of PP2B was found in the non-hypertrophic LV (n = 5/7, Figure 2D).


Figure 2
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2 (A) Protein phosphatase (PP) 2A abundance in sarcoplasmic reticulum-enriched membrane preparations from the right ventricle (RV) (n = 7/6) and left ventricle (LV) (n = 7/6), and (B) representative western blot showing PP2A. (C) PP1 (n = 7/7 for RV and LV) and (D) PP2B (calcineurin) protein abundance (n = 7/7 for RV and n = 5/7 for LV). Immunolabelling density for alveolar hypoxia (open bars) was compared with the control groups (filled bars), which was set to 100%. *P < 0.05 vs. control.

 
3.6 Protein phosphatase activities
PP activities were measured in SR-enriched membrane preparations. The PP2A activity was increased in the RV to 202 ± 32% (P < 0.05, n = 6/5) and in the LV to 234 ± 74% (P < 0.05, n = 6/6). In absolute values, the PP2A activity increased in both the RV (9 ± 1 pmol/min/mg in the hypoxia group vs. 4 ± 1 pmol/min/mg in the control group, P < 0.05, n = 6/5, Figure 3A) and LV (14 ± 4 pmol/min/mg in the hypoxia group vs. 6 ± 1 pmol/min/mg in the control group, P < 0.05, n = 6/6, Figure 3A). The increase in PP1 activity did not reach statistical significance (Figure 3B). The absolute PP1 activity in the RV was 32 ± 2 pmol/min/mg in the hypoxia group vs. 29 ± 6 pmol/min/mg in the control group (n = 6/5, Figure 3B) and in the LV 36 ± 6 pmol/min/mg in the hypoxia group vs. 26 ± 5 pmol/min/mg in the control group (n = 6/6, Figure 3B).


Figure 3
View larger version (17K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3 (A) Protein phosphatase (PP) 2A activity in sarcoplasmic reticulum-enriched membrane preparations from the right ventricle (RV) (n = 6/5) and left ventricle (LV) (n = 6/6). (B) PP1 activity (n = 6/5 in RV and n = 6/6 in LV). PP activities in alveolar hypoxia (open bars) were compared with the control groups (filled bars). *P < 0.05 vs. control.

 
3.7 Protein phosphatase mRNA
In the LV, hypoxia induced an increase in the PP2A mRNA (144 ± 14% of control, P < 0.05, n = 6/6), whereas the PP2A mRNA was not significantly increased in the RV (115 ± 8% of control, n = 6/6).

3.8 Cytokine and chemokine levels induced by alveolar hypoxia
To investigate whether alveolar hypoxia altered the levels of circulating cytokines and chemokines, of whom several are known to impair cardiac function, serum concentrations of 24 mediators were measured. One day of hypoxia increased the serum concentration of IL-18, IL-12p40, regulated on activation normally T-cell expressed and secreted, IL-5, keratinocyte chemoattractant, and granulocyte-macrophage colony-stimulating factor (P < 0.05 for all, n = 7/7, Figure 4A). The increase in IL-18 concentration was pronounced, and the IL-18 level was also increased at 2, 7, and 14 days of hypoxia (all P < 0.05, n = 6/7, 6/6, and 9/10, respectively, Figure 4B). No significant changes in the serum concentrations of IL-1{alpha}, IL-1β, IL-2, IL-3, IL-4, IL-6, IL-9, IL-10, IL-12p70, IL-13, IL-17, eotaxin, granulocyte colony-stimulating factor, interferon-{gamma}, monocyte chemoattractant protein-1, macrophage inflammatory protein (MIP)-1{alpha}, MIP-1β, or tumour necrosis factor-{alpha} (TNF-{alpha}) were found after 1 day of hypoxia (data not shown).


Figure 4
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4 (A) Concentrations of upregulated cytokines and chemokines in serum induced by 1 day of alveolar hypoxia (open bars) and in respective controls (filled bars) (n = 10/8). (B) Serum concentrations of IL-18 after 2, 7, and 14 days of alveolar hypoxia and in respective controls (n = 6/7, 6/6, and 9/10, respectively). *P < 0.05 vs. control.

 
3.9 Protein phosphatases and phosphorylation of phospholamban at Ser16 in cardiomyocytes following IL-18 stimulation and hypoxia
Based on the observed increase in circulating IL-18 at 1, 2, 7, and 14 days of hypoxia together with previous observations showing that this cytokine can induce cardiac dysfunction,19 we hypothesized that IL-18 might, at least partly, mediate the reduction in phosphorylation of phospholamban at Ser16. IL-18 has also been shown to induce signalling through Akt,18 which is assumed to be involved in the regulation of PP2A.20 Stimulation of isolated neonatal cardiomyocytes with IL-18 for 24 h increased the amount of PP2A to 133 ± 8% of control (P < 0.05, n = 6/6, Figure 5A), but the level of PP1 was not significantly changed (102 ± 7% of control, n = 6/6). IL-18 also increased PP2A activity to 114 ± 5% of control (P < 0.05, n = 6/6, Figure 5B), whereas PP1 activity was not significantly altered (102 ± 6% of control, n = 6/6). In addition, IL-18 treatment before isoproterenol stimulation reduced the amount of Ser16 phosphorylated phospholamban to 54 ± 8% of control (P < 0.05, n = 6/6, Figure 5C and D). The amounts of PP2A and PP1 in isolated cardiomyocytes exposed to 1% oxygen for 24 h were not significantly changed compared with control (100 ± 5%, n = 6/6 and 104 ± 6%, n = 6/6). Nor were changes in PP2A or PP1 activity observed in isolated cardiomyocytes exposed to 1% oxygen (97 ± 13%, n = 6/6 and 112 ± 14%, n = 6/6).

3.10 Histology and collagen content in hearts exposed to alveolar hypoxia
In the LV, hypoxia induced an increase in collagen I and III mRNA to 153 ± 16% (P < 0.05, n = 6/6) and 150 ± 13%, respectively (P < 0.05, n = 6/6). A corresponding higher amount of collagen I protein was also observed (143 ± 7%, P < 0.05, n = 7/7), whereas the amount of collagen III was not significantly changed (110 ± 8%, n = 7/7). The increase in collagen I in the LV was not associated with any visible fibrosis, as assessed by the examination of haematoxylin and eosin, AFOG, and Masson trichrome-stained sections (n = 3/3). The collagen III mRNA was significantly increased in the RV (206 ± 40%, P < 0.05, n = 6/6) in the hypoxia group, but not the collagen I mRNA (138 ± 17%, n = 6/6). Collagen I and III protein in the RV was not significantly altered (100 ± 12%, n = 7/7 and 99 ± 11%, n = 7/7, respectively), and no visible fibrosis was observed.


   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
Funding
 References
 
The reduction in Ser16 phosphorylation of phospholamban observed in the free wall of the LV may be a molecular mechanism contributing to the reduced relaxation rate and diastolic dysfunction observed in mice subjected to alveolar hypoxia,3 as well as in patients with COPD and chronic hypoxia.21 For the development of novel therapies of diastolic dysfunction, identification of the signalling pathways causing reduced phospholamban phosphorylation may be of importance. By examining SR-enriched membrane preparations from mice subjected to hypoxia, we found that the amounts of PP1 and PP2A and the activity of PP2A in the LV increased considerably compared with normoxic controls. Moreover, IL-18 levels were significantly increased in hypoxic mice, and IL-18 upregulated both the amount and activity of PP2A in isolated cardiomyocytes. Finally, IL-18 suppressed phospholamban phosphorylation in cardiomyocytes.

In heart failure caused by myocardial infarction or systemic hypertension,22,23 the reduced phosphorylation of phospholamban has, at least partly, been ascribed to diminished β-adrenergic signalling. In alveolar hypoxia, however, we did not observe any alterations in either the number of β-adrenergic receptors or adenylyl cyclase activity. In addition, we did not find an increase in circulating catecholamines in this hypoxia model.3 Thus, it would appear that signals responsible for the reduced phosphorylation of phospholamban at Ser16 in alveolar hypoxia differ from those observed in heart failure following myocardial infarction or systemic hypertension, where also disturbances in anchored pools of PKA have been found.24 In our study, however, the amounts of SR-associated PKA-R and -C subunits and A-kinase anchoring proteins were not altered. Additionally, the amount of PDE4, which regulates PKA activity in the heart,24 was not changed in alveolar hypoxia. Thus, our findings indicate that reduced β-adrenergic signalling is not an important mechanism for the reduction in phosphorylation of phospholamban at Ser16 in alveolar hypoxia.

Phosphorylation of phospholamban is a dynamic process, in which the action of protein kinases is counterbalanced by PPs. PP1 and PP2A are the main phosphatases in the heart that regulate the phosphorylation of phospholamban at Ser16.25 In our study, we showed that the PP activity in the SR-enriched membrane preparations from the LV free wall was substantially increased. The high PP activity was mainly caused by elevated PP2A activity, which was increased more than two-fold, accompanied by an increased amount of PP2A. This increase in PP2A activity was found in SR-enriched membrane preparations consistent with an increase in phosphatases localized to the SR. The importance of an increase in PP2A in vivo has been studied in transgenic mice overexpressing the catalytic subunit of PP2A. Those mice have increased PP2A activity, reduced phosphorylation of phospholamban at Ser16 and diastolic dysfunction,26 indicating that an isolated increase in PP2A may contribute to the reduced LV function observed in alveolar hypoxia. In cardiac myocytes from PP2A-transgenic mice, the diastolic Ca2+ concentration in the cytosol was increased and the time course of the decay of the Ca2+ transients was not hastened by β-adrenergic stimulation.26 We also found increased amounts of NCX in both ventricles, which can increase calcium extrusion and serve as a compensatory mechanism for impaired SR uptake due to reduced phospholamban phosphorylation. In addition to being involved in the regulation of cardiac relaxation and contractility, PP2A activity and level of expression are of importance for growth and development of the lung.27–29 Thus, it cannot be excluded that alterations in pulmonary PP2A during hypoxia may also play a role for hypoxia-related changes in the lung, in addition to the alterations observed in the heart.

Little information is available regarding extracellular stimulators that may regulate the amount and activity of PP2A in the heart. Since LV wall stress is not increased during alveolar hypoxia, circulating mediators might cause the alterations in PP2A amount and function. A link between chronic hypoxemia and cardiovascular disease through inflammatory pathways has been suggested.30 Moreover, accumulating evidence indicates that pro-inflammatory cytokines such as TNF-{alpha} are involved in the development of cardiac failure.31 To investigate whether circulating mediators could cause the observed alterations in PPs in alveolar hypoxia, we examined an array of cytokines and chemokines in serum from mice exposed to alveolar hypoxia. We found increased concentrations of six cytokines, and the increase in the concentration of IL-18 was particularly pronounced. Patients suffering from severe COPD, who are prone to having hypoxia and cardiac dysfunction, have increased expression of IL-18 protein in pulmonary alveolar macrophages.32 In addition to alveolar macrophages, IL-18 is produced by airway and alveolar epithelial cells.33 All these cell types can be exposed to air with decreased partial pressure of oxygen, and thereby be stimulated to generate IL-18, since hypoxia has been shown to promote synthesis of IL-18 in cultures of other types of cells.34

To our knowledge, the relationship between circulating IL-18 and alveolar hypoxia or pulmonary hypertension has not yet been studied. We found increased serum concentrations of IL-18 throughout the investigation period of 2 weeks of hypoxia, showing that the heart in alveolar hypoxia is chronically exposed to increased concentrations of IL-18. Increased circulating IL-18 has also been found in a post-infarction model of heart failure and in patients with congestive heart failure.35,36 Moreover, daily administration of IL-18 to healthy mice caused LV myocardial dysfunction,19 but little is known about the mechanism causing this dysfunction. In the present study, IL-18 increased both the amount and activity of PP2A in isolated cardiomyocytes. Importantly, we also showed that in cardiomyocytes treated with IL-18 for 24 h, there was an ~50% decrease in phosphorylation of phospholamban at Ser16 in response to submaximal β-adrenergic stimulation. The intracellular pathway by which IL-18 exerts its effect on cardiac hypertrophy has been shown to be via Akt.18 We found elevated amounts of cytosolic Akt, which may be induced by IL-18 and possibly lead to increased synthesis of PP2A. However, although our findings indicate a role for IL-18, we cannot exclude that also other circulating mediators may influence the amount and activity of PP2A acting on phospholamban. Another possible regulator of PP2A or PP1 could be hypoxia per se. However, in isolated cardiomyocytes exposed to 1% oxygen, neither the amount nor the activity of PP1 and PP2A was altered.

Another possible mechanism for the LV diastolic dysfunction in alveolar hypoxia is myocardial fibrosis. We did not observe myocardial fibrosis by histological examination, although increased content of collagen I mRNA and protein was observed in the LV. The increase in collagen content might contribute to increased passive chamber stiffness and altered late-diastolic filling. We have previously measured slower relaxation in the early phase of diastole in alveolar hypoxia,3 and the relaxation in this phase of the diastole is predominantly regulated by SERCA activity.37

PP2B (calcineurin) has been shown to mediate cardiac hypertrophy.38 In our study, calcineurin was significantly increased in the RV, consistent with a role as a mediator of hypertrophy in the RV, which is exposed to increased afterload. No significant increase in calcineurin was found in the LV, which has a normal afterload and no hypertrophy. In contrast to the PP2A activity that was increased in both ventricles, possibly due to circulating mediators, the expression of PP2B seems to be stronger related to mechanical stress, such as increased afterload. In a rat model with chronic hypoxia, increased mRNA of the modulatory calcineurin-interacting protein-1, an indicator of calcineurin activity, was found in the RV.39 However, to our knowledge, the amount of PP2B has not previously been measured in alveolar hypoxia or in any types of pulmonary hypertension.

In summary, alveolar hypoxia leads to decreased phosphorylation of phospholamban at Ser16. Changes in the β-adrenergic signalling cascade could not account for this hypophosphorylation. However, alveolar hypoxia induced an increase in the amount and activity of PP2A, which may lead to reduced phospholamban phosphorylation in the myocardium. Serum concentrations of IL-18 were substantially increased in alveolar hypoxia, and IL-18 stimulation of isolated cardiomyocytes reduced phosphorylation of phospholamban at Ser16 and increased the amount and activity of PP2A. Thus, the chronically elevated serum levels of IL-18 in alveolar hypoxia may contribute to diastolic dysfunction via increased PP2A activity causing reduced phosphorylation of phospholamban.


   Funding
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Funding
References
 
This study was supported by Anders Jahre’s Fund for Promotion of Science, The Research Council of Norway, The Norwegian Council for Cardiovascular Diseases and The Norwegian Cancer Society.


   Acknowledgements
 
We are grateful to Bjørg Austbø, Hilde Dishington, Liv Marit Skaug, Dina Behmen, and Almira Karahasan for skilful laboratory work, to Else Marit Løberg for heading the histological examinations, to Siv Leng Tran, Siv Rong Tran, and Carsten Lund for animal care, and to Roy Trondsen for expert technical help.

Conflict of interest: none declared.


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

  1. Celli BR, Cote CG, Marin JM, Casanova C, Montes de Oca M, Mendez RA, et al. The body-mass index, airflow obstruction, dyspnea, and exercise capacity index in chronic obstructive pulmonary disease. N Engl J Med (2004) 350:1005–1012.[Abstract/Free Full Text]
  2. Baum GL, Schwartz A, Llamas R, Castillo C. Left ventricular function in chronic obstructive lung disease. N Engl J Med (1971) 285:361–365.[Web of Science][Medline]
  3. Larsen KO, Sjaastad I, Svindland A, Krobert KA, Skjønsberg OH, Christensen G. Alveolar hypoxia induces left ventricular diastolic dysfunction and reduces phosphorylation of phospholamban in mice. Am J Physiol Heart Circ Physiol (2006) 291:H507–H516.[Abstract/Free Full Text]
  4. Levi M, Moons L, Bouche A, Shapiro SD, Collen D, Carmeliet P. Deficiency of urokinase-type plasminogen activator-mediated plasmin generation impairs vascular remodeling during hypoxia-induced pulmonary hypertension in mice. Circulation (2001) 103:2014–2020.[Abstract/Free Full Text]
  5. Kapiloff MS, Jackson N, Airhart N. mAKAP and the ryanodine receptor are part of a multi-component signaling complex on the cardiomyocyte nuclear envelope. J Cell Sci (2001) 114:3167–3176.[Abstract/Free Full Text]
  6. Lygren B, Carlson CR, Santamaria K, Lissandron V, McSorley T, Litzenberg J, et al. AKAP complex regulates Ca2+ re-uptake into heart sarcoplasmic reticulum. EMBO Rep (2007) 8:1061–1067.[CrossRef][Web of Science][Medline]
  7. Iona S, Cuomo M, Bushnik T, Naro F, Sette C, Hess M, et al. Characterization of the rolipram-sensitive, cyclic AMP-specific phosphodiesterases: identification and differential expression of immunologically distinct forms in the rat brain. Mol Pharmacol (1998) 53:23–32.[Abstract/Free Full Text]
  8. Thomas MJ, Sjaastad I, Andersen K, Helm PJ, Wasserstrom JA, Sejersted OM, et al. Localization and function of the Na+/Ca2+-exchanger in normal and detubulated rat cardiomyocytes. J Mol Cell Cardiol (2003) 35:1325–1337.[CrossRef][Web of Science][Medline]
  9. Braz JC, Gregory K, Pathak A, Zhao W, Sahin B, Klevitsky R, et al. PKC-{alpha} regulates cardiac contractility and propensity toward heart failure. Nat Med (2004) 10:248–254.[CrossRef][Web of Science][Medline]
  10. Krobert KA, Bach T, Syversveen T, Kvingedal AM, Levy FO. The cloned human 5-HT7 receptor splice variants: a comparative characterization of their pharmacology, function and distribution. Naunyn Schmiedebergs Arch Pharmacol (2001) 363:620–632.[CrossRef][Web of Science][Medline]
  11. Florholmen G, Andersson KB, Yndestad A, Austbø B, Henriksen UL, Christensen G. Leukaemia inhibitory factor alters expression of genes involved in rat cardiomyocyte energy metabolism. Acta Physiol Scand (2004) 180:133–142.[CrossRef][Web of Science][Medline]
  12. El Armouche A, Rau T, Zol kO, Ditz D, Pamminger T, Zimmermann WH, et al. Evidence for protein phosphatase inhibitor-1 playing an amplifier role in β-adrenergic signaling in cardiac myocytes. FASEB J (2003) 17:437–439.[Abstract/Free Full Text]
  13. El Armouche A, Bednorz A, Pamminger T, Ditz D, Didie M, Dobrev D, et al. Role of calcineurin and protein phosphatase-2A in the regulation of phosphatase inhibitor-1 in cardiac myocytes. Biochem Biophys Res Commun (2006) 346:700–706.[CrossRef][Web of Science][Medline]
  14. Yamada N, Niwa S, Tsujimura T, Iwasaki T, Sugihara A, Futani H, et al. Interleukin-18 and interleukin-12 synergistically inhibit osteoclastic bone-resorbing activity. Bone (2002) 30:901–908.[CrossRef][Web of Science][Medline]
  15. Nguyen SV, Claycomb WC. Hypoxia regulates the expression of the adrenomedullin and HIF-1 genes in cultured HL-1 cardiomyocytes. Biochem Biophys Res Commun (1999) 265:382–386.[CrossRef][Web of Science][Medline]
  16. Krankel N, Adams V, Gielen S, Linke A, Erbs S, Schuler G, et al. Differential gene expression in skeletal muscle after induction of heart failure: impact of cytokines on protein phosphatase 2A expression. Mol Genet Metab (2003) 80:262–271.[CrossRef][Web of Science][Medline]
  17. Rochais F, Abi-Gerges A, Horner K, Lefebvre F, Cooper DM, Conti M, et al. A specific pattern of phosphodiesterases controls the cAMP signals generated by different Gs-coupled receptors in adult rat ventricular myocytes. Circ Res (2006) 98:1081–1088.[Abstract/Free Full Text]
  18. Colston JT, Boylston WH, Feldman MD, Jenkinson CP, de la Rosa SD, Barton A, et al. Interleukin-18 knockout mice display maladaptive cardiac hypertrophy in response to pressure overload. Biochem Biophys Res Commun (2007) 354:552–558.[CrossRef][Web of Science][Medline]
  19. Woldbaek PR, Sande JB, Strømme TA, Lunde PK, Djurovic S, Lyberg T, et al. Daily administration of interleukin-18 causes myocardial dysfunction in healthy mice. Am J Physiol Heart Circ Physiol (2005) 289:H708–H714.[Abstract/Free Full Text]
  20. Beaulieu JM, Sotnikova TD, Marion S, Lefkowitz RJ, Gainetdinov RR, Caron MG. An Akt/β-arrestin 2/PP2A signaling complex mediates dopaminergic neurotransmission and behavior. Cell (2005) 122:261–273.[CrossRef][Web of Science][Medline]
  21. Boussuges A, Pinet C, Molenat F, Burnet H, Ambrosi P, Badier M, et al. Left atrial and ventricular filling in chronic obstructive pulmonary disease. An echocardiographic and Doppler study. Am J Respir Crit Care Med (2000) 162:670–675.[Abstract/Free Full Text]
  22. Sande JB, Sjaastad I, Hoen IB, Bøkenes J, Tønnessen T, Holt E, et al. Reduced level of serine16 phosphorylated phospholamban in the failing rat myocardium: a major contributor to reduced SERCA2 activity. Cardiovasc Res (2002) 53:382–391.[Abstract/Free Full Text]
  23. Satoh S, Ueda Y, Suematsu N, Oyama J, Kadokami T, Sugano M, et al. Beneficial effects of angiotensin-converting enzyme inhibition on sarcoplasmic reticulum function in the failing heart of the Dahl rat. Circ J (2003) 67:705–711.[CrossRef][Web of Science][Medline]
  24. Tasken K, Aandahl EM. Localized effects of cAMP mediated by distinct routes of protein kinase A. Physiol Rev (2004) 84:137–167.[Abstract/Free Full Text]
  25. Kirchhefer U, Baba HA, Boknik P, Breeden KM, Mavila N, Bruchert N, et al. Enhanced cardiac function in mice overexpressing protein phosphatase inhibitor-2. Cardiovasc Res (2005) 68:98–108.[Abstract/Free Full Text]
  26. Gergs U, Boknik P, Buchwalow I, Fabritz L, Matus M, Justus I, et al. Overexpression of the catalytic subunit of protein phosphatase 2A impairs cardiac function. J Biol Chem (2004) 279:40827–40834.[Abstract/Free Full Text]
  27. Taylor BK, Stoops TD, Everett AD. Protein phosphatase inhibitors arrest cell cycle and reduce branching morphogenesis in fetal rat lung cultures. Am J Physiol Lung Cell Mol Physiol (2000) 278:L1062–L1070.[Abstract/Free Full Text]
  28. Xue C, Heller F, Johns RA, Everett AD. Developmental expression and localization of the catalytic subunit of protein phosphatase 2A in rat lung. Dev Dyn (1998) 211:1–10.[CrossRef][Web of Science][Medline]
  29. Everett AD, Kamibayashi C, Brautigan DL. Transgenic expression of protein phosphatase 2A regulatory subunit B56{gamma} disrupts distal lung differentiation. Am J Physiol Lung Cell Mol Physiol (2002) 282:L1266–L1271.[Abstract/Free Full Text]
  30. Sin DD, Anthonisen NR, Soriano JB, Agusti AG. Mortality in COPD: role of comorbidities. Eur Respir J (2006) 28:1245–1257.[Abstract/Free Full Text]
  31. Prabhu SD. Cytokine-induced modulation of cardiac function. Circ Res (2004) 95:1140–1153.[Abstract/Free Full Text]
  32. Imaoka H, Hoshino T, Takei S, Kinoshita T, Okamoto M, Kawayama T, et al. Interleukin-18 production and pulmonary function in COPD. Eur Respir J (2008) 31:287–297.[Abstract/Free Full Text]
  33. Cameron LA, Taha RA, Tsicopoulos A, Kurimoto M, Olivenstein R, Wallaert B, et al. Airway epithelium expresses interleukin-18. Eur Respir J (1999) 14:553–559.[Abstract]
  34. Kim J, Shao Y, Kim SY, Kim S, Song HK, Jeon JH, et al. Hypoxia-induced IL-18 increases hypoxia-inducible factor-1{alpha} expression through a Rac1-dependent NF-{kappa}B pathway. Mol Biol Cell (2008) 19:433–444.[Abstract/Free Full Text]
  35. Woldbaek PR, Tønnessen T, Henriksen UL, Florholmen G, Lunde PK, Lyberg T, et al. Increased cardiac IL-18 mRNA, pro-IL-18 and plasma IL-18 after myocardial infarction in the mouse; a potential role in cardiac dysfunction. Cardiovasc Res (2003) 59:122–131.[CrossRef][Web of Science][Medline]
  36. Seta Y, Kanda T, Tanaka T, Arai M, Sekiguchi K, Yokoyama T, et al. Interleukin-18 in patients with congestive heart failure: induction of atrial natriuretic peptide gene expression. Res Commun Mol Pathol Pharmacol (2000) 108:87–95.[Web of Science][Medline]
  37. Kass DA, Bronzwaer JG, Paulus WJ. What mechanisms underlie diastolic dysfunction in heart failure? Circ Res (2004) 94:1533–1542.[Abstract/Free Full Text]
  38. Molkentin JD, Lu JR, Antos CL, Markham B, Richardson J, Robbins J, et al. A calcineurin-dependent transcriptional pathway for cardiac hypertrophy. Cell (1998) 93:215–228.[CrossRef][Web of Science][Medline]
  39. Koulmann N, Novel-Chate V, Peinnequin A, Chapot R, Serrurier B, Simler N, et al. Cyclosporin A inhibits hypoxia-induced pulmonary hypertension and right ventricle hypertrophy. Am J Respir Crit Care Med (2006) 174:699–705.[Abstract/Free Full Text]

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

Related Article

New pathophysiological function of protein phosphatase 2A?
Joachim Neumann
Cardiovasc Res 2008 80: 7-8. [Extract] [Full Text] [PDF]



This article has been cited by other articles:


Home page
 Cardiovasc ResHome page
J. Neumann
New pathophysiological function of protein phosphatase 2A?
Cardiovasc Res, October 1, 2008; 80(1): 7 - 8.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
80/1/47    most recent
cvn180v2
cvn180v1
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 Larsen, K.-O.
Right arrow Articles by Christensen, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Larsen, K.-O.
Right arrow Articles by Christensen, G.
Related Collections
Right arrowRelated Article
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?