Human cardiac fibroblasts express ICAM-1, E-selectin and CXC chemokines in response to proinflammatory cytokine stimulation

The International Journal of Biochemistry & Cell Biology 43 (2011) 1450–1458

Contents lists available at ScienceDirect

The International Journal of Biochemistry & Cell Biology journal homepage: www.elsevier.com/locate/biocel

Human cardiac fibroblasts express ICAM-1, E-selectin and CXC chemokines in response to proinflammatory cytokine stimulation Neil A. Turner a,b,∗ , Anupam Das a , David J. O’Regan b,c , Stephen G. Ball a,b , Karen E. Porter a,b a b c

Division of Cardiovascular Medicine, Leeds Institute of Genetics, Health and Therapeutics (LIGHT), University of Leeds, Leeds, UK Multidisciplinary Cardiovascular Research Centre (MCRC), University of Leeds, Leeds, UK Department of Cardiac Surgery, The Yorkshire Heart Centre, Leeds General Infirmary, Leeds, UK

a r t i c l e

i n f o

Article history: Received 7 April 2011 Received in revised form 13 June 2011 Accepted 14 June 2011 Available online 21 June 2011 Keywords: Cardiac fibroblasts Adhesion molecules Cytokines Chemokines Signaling pathways

a b s t r a c t Neutrophil attraction and adhesion to endothelial cells occurs via well defined mechanisms, yet the ability of other cell types to express neutrophil-binding adhesion molecules is not well studied. Cardiac fibroblasts (CF) are a key cell type involved in repair of the infarcted myocardium, a scenario in which neutrophil recruitment is perceived to be detrimental. Here we determined the effects of proinflammatory cytokines on expression of neutrophil-binding adhesion molecules and neutrophil-attracting chemokines in CF cultured from multiple patients, and explored the underlying regulatory mechanisms. An adhesion molecule focused RT-PCR array identified 5 transcripts that were increased markedly in human CF treated with the proinflammatory cytokine interleukin-1 (IL-1, 10 ng/ml, 6 h); including intercellular cell adhesion molecule (ICAM-1) and E-selectin. Real-time RT-PCR verified the array data and immunoblotting confirmed cytokine-induced ICAM-1 and E-selectin protein expression. Treatment with a panel of pharmacological inhibitors identified the NF-␬B pathway as mediating IL-1-induced ICAM1 and E-selectin mRNA and protein expression. Additionally, E-selectin expression in human CF was markedly potentiated by the JNK inhibitor SP600125, but this was not observed when a more selective inhibitor ((L)-JNKI-1) was used, or in human vascular endothelial cells. IL-1 also stimulated CF to secrete the neutrophil chemoattractant CXCL8 via a p38- and NF-␬B-dependent mechanism, as well as inducing CXCL1, CXCL2 and CXCL5 mRNA expression. In conclusion, human CF express neutrophil-binding adhesion molecules and neutrophil chemoattractants in response to proinflammatory cytokines suggesting that, in addition to EC, CF may play an important role in regulating neutrophil recruitment into the infarcted myocardium. © 2011 Elsevier Ltd. All rights reserved.

1. Introduction Pathophysiological remodeling of the infarcted myocardium is characterized by several distinct phases, including (i) cardiomyocyte injury and death; (ii) acute inflammation (elevated chemokine/proinflammatory cytokine levels and inflammatory cell infiltration); (iii) formation of granulation tissue (myofibroblast and macrophage accumulation and activation); (iv) degradation of extracellular matrix (ECM) and neovascularization; (v) deposition of new ECM protein (scar formation); and (vi) resolution of the inflammatory response (Braunwald and Kloner, 1985; Frangogiannis et al., 2002; Frangogiannis, 2008). The infarcted region is therefore dynamic and plays host to many different cell types including neutrophils, monocyte/macrophages, mast cells,

∗ Corresponding author at: Division of Cardiovascular Medicine, Worsley Building, Clarendon Way, University of Leeds, Leeds LS2 9JT, UK. Tel.: +44(0) 113 3435890; fax: +44(0) 113 3434803. E-mail address: [email protected] (N.A. Turner). 1357-2725/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocel.2011.06.008

myofibroblasts and vascular cells, that sequentially infiltrate the injured myocardium through an orchestrated series of events. Mast cells and macrophages are particularly important in initiating the post-MI inflammatory response that aids wound healing. However neutrophil infiltration, which occurs within the first few hours following coronary occlusion (Braunwald and Kloner, 1985), appears to be detrimental to the healing process by exacerbating cardiomyocyte injury and increasing infarct size (Frangogiannis, 2008; Kakkar and Lefer, 2004). Therefore therapies aimed at reducing neutrophil recruitment to the post-MI heart may be of benefit in reducing infarct size and decreasing subsequent cardiac complications (Benson et al., 2007; Frangogiannis et al., 2002; Frangogiannis, 2008; Kakkar and Lefer, 2004; Niessen et al., 2002). Neutrophil adhesion to endothelial cells (EC) occurs via welldefined mechanisms involving interactions between glycolipids and integrins expressed on the neutrophil surface and their cognate adhesion molecules expressed on the EC surface, including E-selectin and intercellular cell adhesion molecule-1 (ICAM-1) (Frangogiannis and Entman, 2005; Frangogiannis, 2008). Recruitment of neutrophils to sites of injury is stimulated by chemotaxis

N.A. Turner et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1450–1458

towards specific members of the ELR+ class of CXC chemokines, including CXCL1, 2, 3, 5, 7 and the archetypal CXCL8 (IL-8) (Frangogiannis and Entman, 2005; Frangogiannis, 2008). Interleukin-1␤ (IL-1␤) is one of the initiating stimuli that drive the acute inflammatory response in humans after MI (Guillen et al., 1995). Moreover, release of IL-1␣ from necrotic cells has been shown to be a key trigger for chemokine secretion and subsequent neutrophilic inflammation in the liver (Chen et al., 2007), and IL1␣ may play a similar role in the myocardium following myocyte necrosis. Both IL-1␣ and IL-1␤ exert their cellular effects via activation of the IL-1 receptor. Knockout mice lacking the IL-1 receptor exhibit reduced chemokine and cytokine expression post-MI, and decreased infiltration of neutrophils, macrophages and myofibroblasts into the infarcted myocardium (Bujak et al., 2008). Cardiac fibroblasts (CF) are the most abundant cell-type in the heart and are intricately involved in the myocardial remodeling process (Porter and Turner, 2009; Turner, 2011). In response to myocardial damage or dysfunction, CF adopt a myofibroblast phenotype and undergo increased proliferation, migration, ECM turnover and secretion of an array of bioactive molecules. CF appear to be particularly responsive to IL-1 compared with other proinflammatory cytokines (Brown et al., 2005; Turner et al., 2009). In this study we identified neutrophil-binding adhesion molecules and neutrophil chemoattractants expressed by human CF in response to proinflammatory cytokine stimulation, and delineated the intracellular signaling pathways regulating their expression. 2. Methods 2.1. Reagents Recombinant human IL-1␣ and tumor necrosis factor alpha (TNF␣) were from Invitrogen. Pharmacological signaling pathway inhibitors (PD98059, SB203580, SP600125, IMD-0354 and (L)-JNKI1) were obtained from Calbiochem, and LY294002 was from Alexis Biochemicals. 2.2. Cell culture Right atrial appendage biopsies were obtained from patients undergoing elective coronary artery bypass surgery at the Leeds General Infirmary, following local ethical committee approval and informed patient consent. Primary cultures of CF were harvested, characterized and cultured as we described previously (Mughal et al., 2009; Porter et al., 2004; Turner et al., 2003). Cells co-expressed vimentin and ␣-smooth muscle actin throughout culture (Mughal et al., 2009), indicative of a myofibroblast phenotype. Experiments were performed on passage 3–6 cells from different patients. Cells were serum-starved for 48 h before performing experiments in basal medium (Dulbecco’s modified Eagle’s medium supplemented with 0.4% FCS), unless stated otherwise. Human saphenous vein EC (SV-EC) were harvested and cultured as we described previously (Aley et al., 2005).

1451

by subtracting the mean CT value (threshold cycle number) of the 5 housekeeping (HK) genes on the array (␤2-microglobulin, hypoxanthine phosphoribosyltransferase 1, ribosomal protein L13A, glyceraldehyde-3-phosphate dehydrogenase [GAPDH] and ␤-actin) from the CT value of the target genes. Data are expressed as relative quantification (RQ) values compared with the 5 HK genes using the formula RQ = 2−CT . For undetectable transcripts, calculation of fold increase was based on the estimated detection limit of the array (RQ = 0.00012). 2.4. Quantitative RT-PCR Following specific treatments, cellular RNA was extracted and cDNA prepared as described previously (Turner et al., 2007). Real-time PCR was performed in duplicate using the Applied Biosystems 7500 Real-Time PCR System with intron-spanning human ICAM-1 (Hs99999152 m1), E-selectin (Hs00174057 m1), CXCL1 (Hs00236937 m1), CXCL2 (Hs00236966 m1), CXCL5 (Hs00171085 m1) or CXCL8 (Hs99999034 m1) primers and Taqman probes (Applied Biosystems). Data were expressed as a percentage of GAPDH endogenous control levels (Hs99999905 m1 primers) using the formula 2−CT × 100, or expressed relative to control sample using the formula 2−CT . 2.5. Immunoblotting Cells were exposed to proinflammatory cytokines before preparing whole cell homogenates and immunoblotting as we have described previously (Turner et al., 2001) using mouse monoclonal ICAM-1 antibody (sc-107) or rabbit polyclonal E-selectin antibody (sc-14011) (Santa Cruz Biotechnology). Equal sample loading was confirmed using monoclonal ␤-actin antibody (ab8226; Abcam). Immunolabelled bands were visualized by SuperSignal West Pico chemiluminescence kit (Perbio). 2.6. CXCL8 ELISA CF cultured in 6-well plates were exposed to appropriate stimuli in 1.5 ml medium for 6–24 h before collecting conditioned media. Media were centrifuged to remove cellular debris and samples stored at −40 ◦ C for subsequent analysis. CXCL8 ELISA was performed according to the manufacturer’s instructions (R&D Systems). 2.7. Statistical analysis Results are expressed as mean ± SEM with n representing the number of experiments on cells from individual patients. Data were analyzed as ratios using repeated measures one-way analysis of variance (ANOVA) and Newman–Keuls post hoc test (GraphPad Prism software). P < 0.05 was considered statistically significant. 3. Results 3.1. Effect of IL-1 on adhesion molecule mRNA levels (RT-PCR array)

2.3. RT-PCR array CF from 3 patients were treated with or without 10 ng/ml IL-1 for 6 h before extracting cellular RNA using the Aurum Total RNA kit (BioRad). RNA samples from the 3 patients were pooled before preparing cDNA and measuring expression levels of 30 adhesion molecules using a SYBR Green-based real-time PCR array (RT2 Profiler Human Extracellular Matrix and Adhesion Molecules, SABiosciences), as we described previously (Turner et al., 2010). CT values for the target genes were calculated

We investigated the effects of IL-1␣ stimulation on expression of 30 different adhesion molecules using a focused real-time RT-PCR array (Fig. 1). cDNA samples were generated following treatment of human CF (3 patients pooled) with or without 10 ng/ml IL-1 for 6 h. In control untreated cells, expression of 22 out of the 30 transcripts was detected (Fig. 1). Relatively high basal expression of several integrin subunits was evident, including ␣1, ␣2, ␣3, ␣5, ␣7, ␣8, ␣V, and ␤1 (highest expression of all transcripts) and ␤5. Integrins ␣4, ␣6, ␤2, ␤3 and ␤4 were expressed at lower levels. Other

1452

N.A. Turner et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1450–1458

Expression (relative to HK)

10

Control +IL-1

1

0.1

0.01

IT G β C 1 D IT 44 C Gα TN 3 N IT δ1 C Gα TN V N IT α1 G IT α2 G α IT 5 G IC β5 C AM TN 1 N IT β1 G IT α7 G IT α8 G α IT 1 G IT β3 G α C 4 D H IT 1 G N α6 C AM IT 1 G VC β2 AM IT 1 G E- β4 S IT EL G αM

0.001

Fig. 1. Effects of IL-1 on adhesion molecule mRNA expression in human CF. Cells from 3 patients were treated with vehicle (grey bars) or with 10 ng/ml IL-1␣ (black bars) for 6 h before RNA extraction. RNA from the 3 patients was pooled, reverse transcribed and subjected to SYBR Green-based real-time PCR array to determine expression levels of 30 adhesion molecules. Data normalized to average of 5 housekeeping genes on array (i.e., HK = 1.0) and ordered by decreasing expression in control cells. Note log scale on y axis. Arrows indicate genes modulated by IL-1. Statistical analysis was not possible due to pooling strategy. Abbreviations: CDH, cadherin; CTNN, catenin; ITG, integrin; SEL, selectin.

prevalent transcripts included CD44 and the ␣1, ␤1 and ␦1 catenins (Fig. 1). Six transcripts included on the array (integrin ␣L, catenin ␦2, contactin 1, PECAM-1, L-selectin and P-selectin) were not detectable in either unstimulated or IL-1-stimulated CF (data not shown). Five transcripts were increased markedly by IL-1 stimulation; CD44, integrin ␤3, ICAM-1, VCAM-1 and E-selectin (Fig. 1). Of these, the highest increases were observed with E-selectin (estimated 170-fold), VCAM-1 (47-fold) and ICAM-1 (35-fold). In unstimulated cells, ICAM-1 mRNA levels were 100 times higher than VCAM-1, whereas E-selectin levels were undetectable (Fig. 1). We focused our follow-up studies on ICAM-1 and E-selectin given their well established role in mediating neutrophil adhesion to EC.

3.2. Cytokine-induced ICAM-1 and E-selectin mRNA expression Array data for ICAM-1 and E-selectin were confirmed and extended using quantitative real-time RT-PCR with Taqman primer/probe sets. As a positive control, we firstly quantified the mRNA levels of ICAM-1 and E-selectin mRNA in human vascular endothelial cells (SV-EC) stimulated for 6 h with IL-1, or a second important myocardial proinflammatory cytokine, TNF␣. As expected, basal ICAM-1 and E-selectin mRNA expression was relatively low in SV-EC in the absence of a cytokine stimulus, but both IL-1 and TNF␣ markedly increased ICAM-1 and E-selectin mRNA expression (Fig. 2A). We then measured cytokine-induced ICAM1 and E-selectin mRNA in human CF (Fig. 2B). Again, basal levels were low in unstimulated CF, but both ICAM-1 and E-selectin mRNA levels increased significantly following cytokine exposure. ICAM-1 mRNA levels were twice as high following 6 h IL-1 treatment than with the same concentration of TNF␣. E-selectin mRNA levels were significantly increased following TNF␣ treatment, but only to 2% of that observed following IL-1 treatment. Thus, although both TNF␣ and IL-1 can increase ICAM-1 and E-selectin mRNA expression in human CF, IL-1 is the more potent stimulus. IL-1-induced ICAM-1 mRNA expression in CF was slightly higher (relative to GAPDH) than in SV-EC, whereas E-selectin levels were ∼20% of those observed in SV-EC (Fig. 2A vs. B). In CF, a time course of IL-1 treatment (Fig. 2C) revealed that ICAM-1 mRNA expression was increased 40-fold within 2 h of IL-1 exposure and remained steadily elevated at least 6 h after treatment. In contrast, IL-1-induced E-selectin mRNA expression was

more transient, with peak levels observed within 3 h and a marked decline thereafter. 3.3. Cytokine-induced ICAM-1 and E-selectin protein expression Cytokine-induced ICAM-1 and E-selectin protein expression in SV-EC (Fig. 3A) and CF (Fig. 3B) was then compared by Western blotting whole cell homogenates. In SV-EC, TNF␣ and IL-1 stimulated ICAM-1 and E-selectin expression to a similar extent after 6 h (Fig. 3A). ICAM-1 expression remained elevated 24 h after cytokine treatment, but E-selectin levels returned to basal. In CF, ICAM1 protein expression was evident 6 h after cytokine treatment in CF, and a greater response to IL-1 was observed compared with TNF␣ (Fig. 3B). ICAM-1 protein expression remained high 24 h after cytokine treatment, at which point levels were similar irrespective of the initiating stimulus. E-selectin protein expression in CF was observed 6 h after treatment with IL-1, but not TNF␣. No Eselectin protein expression was evident in CF homogenates 24 h after cytokine treatment. 3.4. Signaling pathways mediating IL-1-induced ICAM-1 and E-selectin expression We have previously reported that IL-1␣ can stimulate multiple signaling pathways in human CF, including ERK, p38 MAPK, JNK, PI-3-kinase/Akt and NF-␬B (Turner et al., 2009). To delineate the relative importance of these pathways in inducing adhesion molecule expression we employed a panel of pharmacological inhibitors; PD98059 (MEK inhibitor, 30 ␮M), SB203580 (p38 inhibitor, 10 ␮M), SP600125 (JNK inhibitor, 10 ␮M), LY294002 (PI3-kinase inhibitor, 10 ␮M) and IMD-0354 (IKK-2 inhibitor, 10 ␮M) respectively. After pre-treatment with inhibitors, cells were stimulated with IL-1 and changes in mRNA (2 h) and protein (6 h) levels studied (Fig. 4). Only the NF-␬B pathway inhibitor IMD-0354 had a significant effect on IL-1-induced ICAM-1 expression, reducing mRNA levels by 50% (Fig. 4A, upper panel) and fully attenuating protein expression (Fig. 4A, lower panel). A very different profile was observed for E-selectin expression (Fig. 4B). IMD-0354 reduced IL-1-induced E-selectin mRNA expression by 40%, whereas both PD98059 and SP600125 markedly increased IL-1-induced E-selectin mRNA expression by 2.4-fold and 7.8-fold respectively (Fig. 4B, upper panel). These changes were reflected at the protein level (Fig. 4B, lower panel).

N.A. Turner et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1450–1458

1453

150

***

100

E-selectinmRNA (% GAPDH)

ICAM-1 mRNA (% GAPDH)

A. Endothelial cells

***

50

0

Ctrl

TNF

***

40

30

***

20

10 0

Ctrl

IL-1

IL-1

TNF

***

100

E-selectin mRNA (% GAPDH)

ICAM-1 mRNA (% GAPDH)

B. Cardiac fibroblasts

75

***

50 25 0

Ctrl

TNF

15

***

10

5

***

0

Ctrl

IL-1

TNF

IL-1

200

*** *** ***

150

***

***

100 50 0 0

1

2

3

4

5

IL-1 incubation time (h)

6

E-Selectin mRNA (% GAPDH)

ICAM-1 mRNA (% GAPDH)

C. Cardiac fibroblasts 25

***

20

*** ***

15

***

10

*** 5 0

0

1

2

3

4

5

6

IL-1 incubation time (h)

Fig. 2. Proinflammatory cytokines induce ICAM-1 and E-selectin mRNA expression in human CF. Messenger RNA levels quantified by real-time RT-PCR using ICAM-1, Eselectin and GAPDH Taqman primer/probe sets. Data expressed as percentage of GAPDH mRNA levels. ***P < 0.001, NS, not significant vs. control. (A) SV-EC from 3 patients treated with 10 ng/ml TNF␣ or IL-1␣ for 6 h (n = 3). (B) CF from 3 different patients treated with 10 ng/ml TNF␣ or IL-1␣ for 6 h (n = 3). (C) CF from 3 further patients treated with 10 ng/ml IL-1␣ for 2–6 h (n = 3).

Fig. 3. Proinflammatory cytokines induce ICAM-1 and E-selectin protein expression in human CF. Western blots demonstrating ICAM-1 and E-selectin protein expression in whole cell homogenates from human SV-EC (A) and CF (B) following vehicle treatment (control, C) or stimulation with 10 ng/ml TNF␣ or IL-1␣ for 6 h or 24 h. Blots reprobed with ␤-actin antibody as a loading control. Representative of n = 3. Position of molecular weight markers (in kDa) is indicated.

1454

N.A. Turner et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1450–1458

Fig. 4. Effects of signaling inhibitors on IL-1-induced adhesion molecule expression in human CF. Cells from 5 patients pre-treated with signaling inhibitors (30 ␮M PD98059, 10 ␮M SB203580, 10 ␮M SP600125, 10 ␮M LY294002 or 10 ␮M IMD-0354) before stimulation with 10 ng/ml IL-1␣ for 2 h (mRNA) or 6 h (protein) and measurement of ICAM-1 (A) or E-selectin (B) expression. Upper bar chart: mRNA levels. Lower bar chart and blot: protein levels. ***P < 0.001, **P < 0.01, *P < 0.05 vs. IL-1 alone (n = 5). Position of molecular weight markers (in kDa) is indicated.

The effect of SP600125 on E-selectin expression in CF was different from that in SV-EC, which was unaffected by SP600125 (Fig. 5). Because SP600125 is not a particularly specific JNK inhibitor (Bain et al., 2007), we also investigated the effects of a highly specific cell-permeable peptide inhibitor of JNK, (L)-JNKI-1 (Bonny et al., 2001). In marked contrast to the results with SP600125, (L)-JNKI-1 had no effect on E-selectin expression in CF (Fig. 5). (L)-JNKI-1 also had no effect on E-selectin expression in SV-EC, in agreement with the SP600125 results in this cell type.

3.5. Cytokine-induced CXC chemokine mRNA expression and protein secretion To investigate whether human CF expressed neutrophilic ELR+ CXC chemokines in response to cytokine stimulation, we employed real-time RT-PCR to measure CXCL1, 2, 5 and 8 mRNA levels (Fig. 6). Both IL-1 and TNF␣ increased mRNA levels of all four chemokines, with IL-1 being the more potent stimulus. Cytokinestimulated CXCL1 and CXCL8 mRNA levels were particularly high;

Cardiac fibroblasts NS

12

2.0

***

10

E-selectin mRNA (relative)

E-selectin mRNA (relative)

Endothelial cells

8 6 4

***

2

NS NS

1.5

*** 1.0

0.5

0.0

0 Ctrl

IL-1

IL-1 +SP

IL-1 +JNKI-1

Ctrl

IL-1

IL-1

IL-1

+SP

+JNKI-1

Fig. 5. Effects of JNK inhibitors on IL-1-induced E-selectin mRNA expression in human CF and SV-EC. CF (n = 5) and SV-EC (n = 3) from different patients were pre-treated with 10 ␮M SP600125 or 10 ␮M (L)-JNKI-1 before stimulation with 10 ng/ml IL-1␣ for 2 h and measurement of E-selectin mRNA expression by RT-PCR. ***P < 0.001, NS not significant vs. IL-1 alone.

N.A. Turner et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1450–1458

200

100

*** 0 Ctrl

5

** 0 Ctrl

TNF

10

5

***

IL-1

CXCL8

300

***

***

10

IL-1

CXCL5

15

CXCL5 mRNA (% GAPDH)

TNF

CXCL8 mRNA (% GAPDH)

CXCL1 mRNA (% GAPDH)

300

CXCL2

15

*** CXCL2 mRNA (% GAPDH)

CXCL1

400

1455

***

200

100

*** 0

0 Ctrl

TNF

Ctrl

IL-1

TNF

IL-1

Fig. 6. IL-1 stimulates CXCL1, 2, 5 and 8 mRNA expression in human CF. CF from 3 patients were stimulated with 10 ng/ml TNF␣ or IL-1␣ for 6 h before performing real-time RT-PCR with Taqman primer/probes for CXCL1, 2, 5 and 8. Data expressed as percentage GAPDH levels. ***P < 0.001 **P < 0.01 vs. control (n = 3).

A Secreted IL-8 (ng/ml)

approximately 30 times greater than those of CXCL2 and CXCL5 (Fig. 6). We selected one of the more prevalent chemokines, CXCL8 (IL8), for further investigation of the role of IL-1-activated intracellular signaling pathways. Pre-treatment of CF with a panel of pharmacological inhibitors revealed that the IL-1-induced increase in CXCL8 mRNA expression observed after 2 h treatment was relatively insensitive to inhibition of the ERK, p38 MAPK, JNK, PI-3-kinase or NF-␬B pathways (Fig. 7). A CXCL8 ELISA was used to confirm that IL-1- and TNF␣-induced changes in CXCL8 mRNA levels led to increased protein secretion (Fig. 8A). In agreement with the mRNA data, IL-1 was the more potent stimulus for CXCL8 secretion. Assessment of the role of sig-

***

40

30

*** 20

10

0 Ctrl

B Secreted IL-8 (ng/ml)

IL-8 mRNA (relative)

1.5

+++ 1.0

0.5

TNF

IL-1

10

8

6

**

4

2

***

0 IL-1

0.0 Ctrl

IL-1

+PD

+SB

+SP

+LY

+PD

+SB

+SP

+LY

+IMD

+IMD

Fig. 7. Signaling pathways mediating IL-1-induced CXCL8 mRNA expression in human CF. CF were pre-treated with signaling inhibitors (30 ␮M PD98059, 10 ␮M SB203580, 10 ␮M SP600125, 10 ␮M LY294002 or 10 ␮M IMD-0354) before stimulation with 10 ng/ml IL-1␣. Levels of CXCL8 mRNA were determined by RT-PCR after 2 h. +++ P < 0.001 for effect of IL-1 vs. control (n = 5). No statistical difference for effects of inhibitors on IL-1-induced CXCL8 expression.

Fig. 8. Signaling pathways mediating IL-1-induced CXCL8 protein secretion in human CF. (A) ELISA measurement of CXCL8 protein levels in conditioned medium from CF treated with 10 ng/ml TNF␣ or IL-1␣ for 24 h. ***P < 0.001 vs. control (n = 5). (B) CF were pre-treated with signaling inhibitors (30 ␮M PD98059, 10 ␮M SB203580, 10 ␮M SP600125, 10 ␮M LY294002 or 10 ␮M IMD-0354) before stimulation with 10 ng/ml IL-1␣ for 6 h. CXCL8 secretion measured by ELISA. ***P < 0.001, **P < 0.01 vs. IL-1 alone (n = 3).

1456

N.A. Turner et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1450–1458

naling pathways in mediating IL-1-induced CXCL8 secretion in CF revealed that the p38 MAPK inhibitor SB203580 and the NF-␬B pathway inhibitor IMD-0354 reduced CXCL8 secretion by 58% and 98%, respectively (Fig. 8B).

4. Discussion ICAM-1 is expressed by many different cell types including EC, epithelial cells, cardiomyocytes, fibroblasts, smooth muscle cells, lymphocytes and monocytes (Benson et al., 2007; Niessen et al., 1999). In the heart, ICAM-1 has been shown to be expressed by cardiomyocytes following MI (Devaux et al., 1997; Hawkins et al., 1996; Kukielka et al., 1995b; Niessen et al., 1999) and there is good evidence that myocardial expression of ICAM-1 is detrimental in the MI setting, due to increased neutrophilic inflammation (Barnes, 2007; Graciano et al., 2001; Ioculano et al., 1994; Jones et al., 2000; Kakkar and Lefer, 2004; Lefer et al., 1996; Palazzo et al., 1998; Sato et al., 2006). The interaction between fibroblasts and neutrophils has been overlooked until recently, when it was demonstrated that neutrophils could adhere to CF or dermal fibroblasts via interactions with ICAM-1 (Couture et al., 2009). CF were also shown to engulf neutrophils, a phenomenon required for transcellular migration (Couture et al., 2009). This particular study used rat neutrophils in combination with rat CF or human dermal fibroblasts, and ICAM1 expression was induced by TNF␣ or phorbol ester. Our study is the first to demonstrate this phenomenon in human CF. Previous studies have also shown that cultured neonatal rat CF express ICAM-1 in response to proinflammatory cytokines, but not hypoxia (Kacimi et al., 1998), whereas hypoxia induced soluble ICAM-1 secretion in adult rat CF (Sapna and Shivakumar, 2007). It is possible that neutrophil binding to CF via ICAM-1 could have direct modulatory effects on CF function. For example, cell–cell contact between leukocytes and renal fibroblasts has been shown to stimulate monocyte chemoattractant protein-1 secretion from fibroblasts, an effect that could be inhibited by an ICAM-1 neutralizing antibody (Hao et al., 2003). Thus, one could postulate that neutrophil-fibroblast interactions in vivo would exacerbate the inflammatory response by increasing local production of chemokines, thereby increasing leukocyte influx and further augmenting the inflammatory response. ICAM-1 gene expression can be regulated by the NF-␬B pathway, as well as by MAPK pathways (ERK, p38, JNK) acting on the AP-1 transcription factor (Roebuck and Finnegan, 1999). We have previously demonstrated that IL-1 activates the ERK, p38 MAPK, JNK, PI-3-kinase/Akt and NF-␬B pathways in human CF (Turner et al., 2009), but the role of these pathways in regulating ICAM-1 expression in human CF was unknown. Our inhibitor studies revealed that IL-1-induced ICAM-1 mRNA and protein expression were regulated exclusively by the NF-␬B pathway in these cells. A previous report using neonatal rat CF indicated a role for p38 MAPK (but not ERK) in mediating cytokine-induced ICAM-1 expression (Kacimi et al., 1998), but we saw no effect with the p38 inhibitor SB203580. The discrepancies between these two studies may relate to species (rat vs. human) or developmental (neonatal vs. adult) differences of the CF studied. In contrast to ICAM-1, E-selectin expression is highly restricted to EC and a limited number of other cell types, due to methylation of the E-selectin promoter (Smith et al., 1993). Our study is the first to report E-selectin expression in CF. Tail vein injection of IL1␤ in mice induced E-selectin expression specifically in the heart, whereas ICAM-1 expression was observed in many other tissues as well (Tamaru et al., 1998). Interestingly, IL-1-induced E-selectin expression was localized not only to microvascular EC in the heart, but also with other stromal cells, but not cardiomyocytes (Tamaru et al., 1998). Myocardial E-selectin expression was elevated in mice

1–24 h after ischemia–reperfusion injury and E-selectin knockout mice exhibited smaller resultant infarct size than non-transgenic controls due to reduced neutrophil infiltration (Jones et al., 2000). Further studies are warranted to determine the functional relevance of E-selectin expression in CF, compared with its welldefined role in EC. Neutrophil binding to endothelial E-selectin results in phosphorylation of the cytoplasmic domain of E-selectin (Yoshida et al., 1998) and activation of MAP kinase signaling pathways (Hu et al., 2000). Hence, the engagement of E-selectin on CF may initiate signaling responses in CF that modulate cellular function. Identifying the differences in molecular mechanisms regulating E-selectin expression in CF compared with EC may facilitate cell type-specific therapies aimed at reducing myocardial inflammation. Our signaling inhibitor experiments indicated that E-selectin expression was regulated at the transcriptional level via the NF-␬B pathway, in agreement with the classical model of E-selectin transcription in EC (Collins et al., 1995). However, we also observed a striking increase in IL-1-induced E-selectin mRNA and protein expression in CF treated with the JNK inhibitor SP600125; effects that were not evident in human SV-EC under the same conditions. Although SP600125 is widely used as a JNK inhibitor, it also inhibits several other protein kinases with similar or greater potency and therefore its effects are not necessarily attributable to JNK (Bain et al., 2007). Despite this, the majority of evidence for a role of the JNK pathway in CF, as well as many other cell types, has come from the use of SP600125 (Turner, in press). To clarify the role of JNK in regulating E-selectin expression, we employed a more specific JNK inhibitor, (L)-JNKI-1 (Bonny et al., 2001). This cell-permeable peptide inhibitor mimics the JNK-binding domain and thereby inhibits JNK through a different mechanism from SP600125. Our finding that (L)-JNKI-1 did not increase IL-1-induced E-selectin expression in CF indicates that the stimulatory effect of SP600125 on E-selectin expression is independent of JNK inhibition and may relate to “off-target” effects. Nevertheless, these data highlight potential differences in regulation of E-selectin expression in CF and EC. Upregulation of ELR+ chemokines is observed in animal models of ischemia–reperfusion injury (Chandrasekar et al., 2001; Ivey et al., 1995; Kukielka et al., 1995a) and reduced neutrophil accumulation and infarct size is apparent after ischemia–reperfusion injury in mice lacking CXCR2, the receptor responsible for the effects of most ELR+ chemokines (Tarzami et al., 2003). CXCL1 (KC/GRO␣) is one of the most prevalent cytokines secreted by neonatal rat CF in culture (LaFramboise et al., 2007). CXCL1 and CXCL5 (LIX/ENA-78) secretion can be induced by oncostatin M, but not other IL-6 family ligands, in cultured mouse (Lafontant et al., 2006) and human (Hohensinner et al., 2009) CF. Although unaffected by oncostatin M, CXCL2 (MIP2/GRO␤) expression was induced by TNF␣ (Lafontant et al., 2006). Our study is the first to report induction of CXCL1, 2 and 5 in CF following stimulation with IL-1. We also established that IL-1 stimulates human CF to secrete CXCL8 (IL-8), the archetypal ELR+ chemokine that is not expressed in mice. CXCL8 gene expression is reported to be maximally activated by a combination of de-repression of the gene promoter, transcriptional activation via NF-␬B and JNK pathways, and mRNA stabilization through p38 MAPK activation (Hoffmann et al., 2002). Surprisingly, we observed no significant effect of kinase inhibitors on IL-1-induced CXCL8 mRNA expression in CF. We deliberately studied the effects of pathway inhibition after only 2 h IL-1 stimulation in order to focus on early responses to IL-1 receptor activation. It is possible that later time points may have revealed more pronounced effects on mRNA levels, when mRNA half-life may have been more relevant. Indeed, our observation that CXCL8 protein secretion was prevented by NF-␬B pathway inhibition and partially reduced by p38 MAPK inhibition suggests the possibility of post-transcriptional regulation.

N.A. Turner et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1450–1458

Aside from our focus on ICAM-1 and E-selectin, our adhesion molecule array analysis also revealed that human CF express mRNA encoding a range of integrin subunits at relatively high levels, including ␣1, ␣2, ␣3, ␣5, ␣7, ␣8, ␣V, ␤1 and ␤5. Integrins are a large family of ␣/␤ dimeric cell surface receptors that mediate cell–ECM interactions (Manso et al., 2009). Expression of particular integrin subunits varies between cell types and our results extend those of previous studies that reported expression of ␣1, ␣2, ␣5, ␣V, ␤1, ␤3 and ␤5 in rat CF (Burgess et al., 2002; Kawano et al., 2000; Manso et al., 2009). The most prevalent ␣ integrin subunit mRNA we detected in human CF was ␣3, which has been shown to bind type III and type VI collagen and is upregulated in CF after experimental MI (Bryant et al., 2009). The ␣3 integrin data therefore match well with our recent ECM-focused array study in which we identified type VI collagen as being the most highly expressed collagen mRNA in unstimulated cultured human CF (Turner et al., 2010). In addition to the integrins, CF also expressed high levels of CD44, a transmembrane receptor that is important for resolution of the inflammatory response post-MI as well as in regulating CF function (Huebener et al., 2008). CD44 mRNA levels were increased 3.4-fold by IL-1 treatment; an observation that may explain some of these in vivo effects. In summary, we have demonstrated that human CF express ICAM-1, E-selectin and ELR+ CXC chemokines in response to proinflammatory cytokines that are elevated in the infarcted heart. We have also identified the underlying signaling pathways mediating these effects, some of which differ from those previously reported in EC. Our data support the concept that, in addition to EC, CF may play an important role in regulating neutrophil recruitment into the infarcted myocardium. Further studies will be required to assess the extent of the functional contribution played by CF in inflammatory cardiac dysfunction.

Acknowledgments N.A.T. is the recipient of a Research Councils UK Academic Fellowship. We are grateful to Stacey Galloway, Philip Warburton, John Sinfield and Kirsten Riches for laboratory assistance.

References Aley PK, Porter KE, Boyle JP, Kemp PJ, Peers C. Hypoxic modulation of Ca2+ signaling in human venous endothelial cells. Multiple roles for reactive oxygen species. J Biol Chem 2005;280:13349–54. Bain J, Plater L, Elliott M, Shpiro N, Hastie CJ, McLauchlan H, et al. The selectivity of protein kinase inhibitors: a further update. Biochem J 2007;408:297–315. Barnes PJ. New molecular targets for the treatment of neutrophilic diseases. J Allergy Clin Immunol 2007;119:1055–62. Benson V, McMahon AC, Lowe HC. ICAM-1 in acute myocardial infarction: a potential therapeutic target. Curr Mol Med 2007;7:219–27. Bonny C, Oberson A, Negri S, Sauser C, Schorderet DF. Cell-permeable peptide inhibitors of JNK: novel blockers of beta-cell death. Diabetes 2001;50:77–82. Braunwald E, Kloner RA. Myocardial reperfusion: a double-edged sword? J Clin Invest 1985;76:1713–9. Brown RD, Ambler SK, Mitchell MD, Long CS. The cardiac fibroblast: therapeutic target in myocardial remodeling and failure. Annu Rev Pharmacol Toxicol 2005;45:657–87. Bryant JE, Shamhart PE, Luther DJ, Olson ER, Koshy JC, Costic DJ, et al. Cardiac myofibroblast differentiation is attenuated by alpha(3) integrin blockade: potential role in post-MI remodeling. J Mol Cell Cardiol 2009;46:186–92. Bujak M, Dobaczewski M, Chatila K, Mendoza LH, Li N, Reddy A, et al. Interleukin-1 receptor type I signaling critically regulates infarct healing and cardiac remodeling. Am J Pathol 2008;173:57–67. Burgess ML, Terracio L, Hirozane T, Borg TK. Differential integrin expression by cardiac fibroblasts from hypertensive and exercise-trained rat hearts. Cardiovasc Pathol 2002;11:78–87. Chandrasekar B, Smith JB, Freeman GL. Ischemia–reperfusion of rat myocardium activates nuclear factor-␬B and induces neutrophil infiltration via lipopolysaccharide-induced CXC chemokine. Circulation 2001;103:2296–302. Chen CJ, Kono H, Golenbock D, Reed G, Akira S, Rock KL. Identification of a key pathway required for the sterile inflammatory response triggered by dying cells. Nat Med 2007;13:851–6.

1457

Collins T, Read MA, Neish AS, Whitley MZ, Thanos D, Maniatis T. Transcriptional regulation of endothelial cell adhesion molecules: NF-␬B and cytokine-inducible enhancers. FASEB J 1995;9:899–909. Couture P, Paradis-Massie J, Oualha N, Thibault G. Adhesion and transcellular migration of neutrophils and B lymphocytes on fibroblasts. Exp Cell Res 2009;315:2192–206. Devaux B, Scholz D, Hirche A, Klovekorn WP, Schaper J. Upregulation of cell adhesion molecules and the presence of low grade inflammation in human chronic heart failure. Eur Heart J 1997;18:470–9. Frangogiannis NG. The immune system and cardiac repair. Pharmacol Res 2008;58:88–111. Frangogiannis NG, Entman ML. Chemokines in myocardial ischemia. Trends Cardiovasc Med 2005;15:163–9. Frangogiannis NG, Smith CW, Entman ML. The inflammatory response in myocardial infarction. Cardiovasc Res 2002;53:31–47. Graciano AL, Bryant DD, White DJ, Horton J, Bowles NE, Giroir BP. Targeted disruption of ICAM-1, P-selectin genes improves cardiac function and survival in TNF-alpha transgenic mice. Am J Physiol Heart Circ Physiol 2001;280:H1464–71. Guillen I, Blanes M, Gomez-Lechon MJ, Castell JV. Cytokine signaling during myocardial infarction: sequential appearance of IL-1 beta and IL-6. Am J Physiol 1995;269:R229–35. Hao L, Okada H, Kanno Y, Inoue T, Kobayashi T, Watanabe Y, et al. Direct contact between human peripheral blood mononuclear cells and renal fibroblasts facilitates the expression of monocyte chemoattractant protein-1. Am J Nephrol 2003;23:208–13. Hawkins HK, Entman ML, Zhu JY, Youker KA, Berens K, Dore M, et al. Acute inflammatory reaction after myocardial ischemic injury and reperfusion. Development and use of a neutrophil-specific antibody. Am J Pathol 1996;148:1957–69. Hoffmann E, ttrich-Breiholz O, Holtmann H, Kracht M. Multiple control of interleukin-8 gene expression. J Leukoc Biol 2002;72:847–55. Hohensinner PJ, Kaun C, Rychli K, Niessner A, Pfaffenberger S, Rega G, et al. The inflammatory mediator oncostatin M induces stromal derived factor-1 in human adult cardiac cells. FASEB J 2009;23:774–82. Hu Y, Kiely JM, Szente BE, Rosenzweig A, Gimbrone Jr MA. E-selectin-dependent signaling via the mitogen-activated protein kinase pathway in vascular endothelial cells. J Immunol 2000;165:2142–8. Huebener P, bou-Khamis T, Zymek P, Bujak M, Ying X, Chatila K, et al. CD44 is critically involved in infarct healing by regulating the inflammatory and fibrotic response. J Immunol 2008;180:2625–33. Ioculano M, Squadrito F, Altavilla D, Canale P, Squadrito G, Campo GM, et al. Antibodies against intercellular adhesion molecule 1 protect against myocardial ischaemia–reperfusion injury in rat. Eur J Pharmacol 1994;264:143–9. Ivey CL, Williams FM, Collins PD, Jose PJ, Williams TJ. Neutrophil chemoattractants generated in two phases during reperfusion of ischemic myocardium in the rabbit. Evidence for a role for C5a and interleukin-8. J Clin Invest 1995;95:2720–8. Jones SP, Trocha SD, Strange MB, Granger DN, Kevil CG, Bullard DC, et al. Leukocyte and endothelial cell adhesion molecules in a chronic murine model of myocardial reperfusion injury. Am J Physiol Heart Circ Physiol 2000;279:H2196– 201. Kacimi R, Karliner JS, Koudssi F, Long CS. Expression and regulation of adhesion molecules in cardiac cells by cytokines: response to acute hypoxia. Circ Res 1998;82:576–86. Kakkar AK, Lefer DJ. Leukocyte and endothelial adhesion molecule studies in knockout mice. Curr Opin Pharmacol 2004;4:154–8. Kawano H, Cody RJ, Graf K, Goetze S, Kawano Y, Schnee J, et al. Angiotensin II enhances integrin and alpha-actinin expression in adult rat cardiac fibroblasts. Hypertension 2000;35:273–9. Kukielka GL, Smith CW, LaRosa GJ, Manning AM, Mendoza LH, Daly TJ, et al. Interleukin-8 gene induction in the myocardium after ischemia and reperfusion in vivo. J Clin Invest 1995a;95:89–103. Kukielka GL, Smith CW, Manning AM, Youker KA, Michael LH, Entman ML. Induction of interleukin-6 synthesis in the myocardium. Potential role in postreperfusion inflammatory injury. Circulation 1995b;92:1866–75. Lafontant PJ, Burns AR, Donnachie E, Haudek SB, Smith CW, Entman ML. Oncostatin M differentially regulates CXC chemokines in mouse cardiac fibroblasts. Am J Physiol Cell Physiol 2006;291:C18–26. LaFramboise WA, Scalise D, Stoodley P, Graner SR, Guthrie RD, Magovern JA, et al. Cardiac fibroblasts influence cardiomyocyte phenotype in vitro. Am J Physiol Cell Physiol 2007;292:C1799–808. Lefer DJ, Flynn DM, Anderson DC, Buda AJ. Combined inhibition of P-selectin and ICAM-1 reduces myocardial injury following ischemia and reperfusion. Am J Physiol 1996;271:H2421–9. Manso AM, Kang SM, Ross RS. Integrins, focal adhesions, and cardiac fibroblasts. J Investig Med 2009;57:856–60. Mughal RS, Warburton P, O’Regan DJ, Ball SG, Turner NA, Porter KE. Peroxisome proliferator-activated receptor ␥-independent effects of thiazolidinediones on human cardiac myofibroblast function. Clin Exp Pharmacol Physiol 2009;36:478–86. Niessen HW, Krijnen PA, Visser CA, Meijer CJ, Hack CE. Intercellular adhesion molecule-1 in the heart. Ann N Y Acad Sci 2002;973:573–85. Niessen HW, Lagrand WK, Visser CA, Meijer CJ, Hack CE. Upregulation of ICAM-1 on cardiomyocytes in jeopardized human myocardium during infarction. Cardiovasc Res 1999;41:603–10. Palazzo AJ, Jones SP, Girod WG, Anderson DC, Granger DN, Lefer DJ. Myocardial ischemia–reperfusion injury in CD18- and ICAM-1-deficient mice. Am J Physiol 1998;275:H2300–7.

1458

N.A. Turner et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1450–1458

Porter KE, Turner NA. Cardiac fibroblasts: at the heart of myocardial remodeling. Pharmacol Ther 2009;123:255–78. Porter KE, Turner NA, O’Regan DJ, Balmforth AJ, Ball SG. Simvastatin reduces human atrial myofibroblast proliferation independently of cholesterol lowering via inhibition of RhoA. Cardiovasc Res 2004;61:745–55. Roebuck KA, Finnegan A. Regulation of intercellular adhesion molecule-1 (CD54) gene expression. J Leukoc Biol 1999;66:876–88. Sapna S, Shivakumar K. Hypoxia and antioxidants enhance soluble ICAM-1 release from cardiac fibroblasts. Mol Cell Biochem 2007;303:259–62. Sato T, Suzuki H, Shibata M, Kusuyama T, Omori Y, Soda T, et al. Tumornecrosis-factor-␣-gene-deficient mice have improved cardiac function through reduction of intercellular adhesion molecule-1 in myocardial infarction. Circ J 2006;70:1635–42. Smith GM, Whelan J, Pescini R, Ghersa P, DeLamarter JF, Hooft van HR. DNAmethylation of the E-selectin promoter represses NF-␬B transactivation. Biochem Biophys Res Commun 1993;194:215–21. Tamaru M, Tomura K, Sakamoto S, Tezuka K, Tamatani T, Narumi S. Interleukin-1␤ induces tissue- and cell type-specific expression of adhesion molecules in vivo. Arterioscler Thromb Vasc Biol 1998;18:1292–303. Tarzami ST, Miao W, Mani K, Lopez L, Factor SM, Berman JW, et al. Opposing effects mediated by the chemokine receptor CXCR2 on myocardial ischemiareperfusion injury: recruitment of potentially damaging neutrophils and direct myocardial protection. Circulation 2003;108:2387–92.

Turner NA. The cardiac fibroblast. Kerala, India: Research Signpost; 2011. Turner NA. Therapeutic regulation of cardiac fibroblast function: targeting stressactivated protein kinase pathways. Fut Cardiol; in press. Turner NA, Ball SG, Balmforth AJ. The mechanism of angiotensin II-induced extracellular signal-regulated kinase-1/2 activation is independent of angiotensin AT1A receptor internalisation. Cell Signal 2001;13:269–77. Turner NA, Das A, Warburton P, O’Regan DJ, Ball SG, Porter KE. Interleukin-1␣ stimulates pro-inflammatory cytokine expression in human cardiac myofibroblasts. Am J Physiol Heart Circ Physiol 2009;297:H1117–27. Turner NA, Mughal RS, Warburton P, O’Regan DJ, Ball SG, Porter KE. Mechanism of TNF␣-induced IL-1␣, IL-1␤ and IL-6 expression in human cardiac fibroblasts: effects of statins and thiazolidinediones. Cardiovasc Res 2007;76:81–90. Turner NA, Porter KE, Smith WH, White HL, Ball SG, Balmforth AJ. Chronic ␤2 -adrenergic receptor stimulation increases proliferation of human cardiac fibroblasts via an autocrine mechanism. Cardiovasc Res 2003;57: 784–92. Turner NA, Warburton P, O’Regan DJ, Ball SG, Porter KE. Modulatory effect of interleukin-1␣ on expression of structural matrix proteins, MMPs and TIMPs in human cardiac myofibroblasts: role of p38 MAP kinase. Matrix Biol 2010;29:613–20. Yoshida M, Szente BE, Kiely JM, Rosenzweig A, Gimbrone Jr MA. Phosphorylation of the cytoplasmic domain of E-selectin is regulated during leukocyte-endothelial adhesion. J Immunol 1998;161:933–41.