Skip Navigation

Cardiovascular Research 2003 60(3):608-616; doi:10.1016/j.cardiores.2003.08.016
© 2003 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Gilles, S.
Right arrow Articles by Becker, B. F
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gilles, S.
Right arrow Articles by Becker, B. F
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2003, European Society of Cardiology

Release of TNF-{alpha} during myocardial reperfusion depends on oxidative stress and is prevented by mast cell stabilizers

Stefanie Gillesa, Stefan Zahlera, Ulrich Welschb, Christian P Sommerhoffc and Bernhard F Becker*,a

aDepartment of Physiology, University of Munich, Schillerstr. 44, 80336 Munich, Germany
bDepartment of Anatomy, University of Munich, Munich, Germany
cDepartment of Clinical Biochemistry, University of Munich, Munich, Germany

*Corresponding author. Tel.: +49-89-5996-380; fax: +49-89-5996-378. Email address: B.F.Becker{at}LRZ.UNI-MUENCHEN.DE

Received 8 May 2003; revised 13 August 2003; accepted 21 August 2003


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Objectives: Our study sought to elucidate the role of oxidative stress for shedding of tumor necrosis factor-{alpha} (TNF-{alpha}) and for activating TNF-{alpha}-converting enzyme (TACE). Background: TNF-{alpha}, a central inflammatory cytokine, is discussed as one of the mediators of reperfusion injury. Shedding of membrane-bound pro-TNF-{alpha} is thought to be largely due to TNF-{alpha}-converting enzyme (TACE). Methods: Release of TNF-{alpha} and TACE dependency were studied in isolated rat hearts and in the human mast cell line HMC-1. Results: In reperfused hearts, interstitial release of TNF-{alpha} occurred in two phases (2–10 and >45 min). It depended on the presence of oxygen during reperfusion and was attenuated by reduced glutathione. Infusion of the oxidants H2O2 or HOCl elicited release in non-ischemic hearts. TNF-{alpha} release was inhibited in hearts treated with degranulation inhibitors ketotifen or cromoglycate, suggesting mast cells as major source for myocardial TNF-{alpha}. This was confirmed by tissue staining. Post-ischemic release of histamine, however, did not parallel that of TNF-{alpha}. Heart tissue contained mainly mature TACE. HMC-1 expressed abundant pro-TACE and cleaved the pro-TNF-{alpha}-peptide Ac-SPLAQAVRSSSR-NH2. However, cleavage was nonspecific and only partly inhibited by TACE inhibitor TAPI-2 (10–100 µmol/l), while it was stimulated by H2O2 and HOCl and fully blocked by the nonspecific metalloprotease inhibitor o-phenanthroline. Conclusions: The mechanism underlying TNF-{alpha} release from post-ischemic myocardium is oxidation-dependent but largely independent of activation of TACE. Mast cell stabilizers may be useful in preventing TNF-{alpha} release during reperfusion.

KEYWORDS Cytokines; Free radicals; Ischemia; Reperfusion

Abbreviations: TNF-{alpha}, tumor necrosis factor-{alpha} • TACE, TNF-{alpha}-converting enzyme • ADAM, a disintegrin and metalloproteinase • MMP, matrix metalloproteinase • TAPI-2, TNF-{alpha} protease inhibitor: HONHCOCH2CH(CH2CH(CH3)2CO-t-Butyl-Gly-Ala-NHCH2CH2NH2 • KHB, Krebs–Henseleit buffer • GSH, reduced glutathione • HMC-1, human mast cell line-1 • CABG, coronary artery bypass grafting • DSCG, disodium cromoglycate


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
Reperfusion of ischemic myocardium leads to a number of inflammatory reactions, e. g. endothelial expression of adhesion molecules and chemokines, activation and emigration of leukocytes and degranulation of resident mast cells, with sequelae such as edema formation and tissue remodelling. Acute release of a number of proinflammatory cytokines, such as interleukines (IL-1, -6, -8) and tumor necrosis factor-{alpha} (TNF-{alpha}), has been demonstrated in reperfused myocardium of man and of various animal species [1–4]. Enhanced levels of TNF-{alpha}, for example, have been found in blood from the coronary sinus of patients undergoing CABG [1]. Although negative inotropic [5] and proinflammatory [6] as well as some cytoprotective effects [7] have been proposed for TNF-{alpha}, the role of the cytokine in reperfusion injury is controversial. Also, the cellular source and the mechanism of release of myocardial TNF-{alpha} are still rather speculative.

Proteolytic cleavage of the 26-kDa membrane-bound precursor pro-TNF-{alpha} yields the soluble, systemically active, 17-kDa cytokine. This "membrane shedding" is thought to be mainly accomplished by TNF-{alpha}-converting enzyme (TACE/ADAM17) [8], a metalloproteinase of the ADAM superfamily [9]. TACE is synthesized as proenzyme, in which a cysteine residue (Cys-184) within the prodomain interacts via its SH-group with the catalytic site Zn2+, inhibiting the proteolytic function [10]. This "cysteine switch" has been identified for many members of the ADAM and MMP families [11,12]. A postulated mechanism of TACE activation is via proteolytic removal of the entire prodomain, presumably by a furin-like activity [13]. A different mechanism might involve oxidation of the SH-group of the inhibitory cysteine, as already described for some MMPs [14]. Recent evidence suggests that, under certain circumstances, pro-TACE can be activated in a similar fashion by reactive oxygen species [15].

Since different oxidants and reactive oxygen species (ROS) are generated in reperfused ischemic tissue [16], we hypothesized that an acute oxidative activation of TACE during reperfusion might be responsible for the post-ischemic release of TNF-{alpha} from myocardial tissue.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
2.1. Reagents
The peptides Ac-Ser-Pro-Leu-Ala-Gln-Ala-Val-Ser-Ser-Ser-Arg-NH2 (TNF-{alpha}-peptide) and Ac-Ser-Pro-Leu-Ala-Gln-Ala were synthesized by Biosource (Nivelles, Belgium). The TACE inhibitor N-(R)-(2-(hydroxyaminocarbonyl)methyl)-4-ethylpentanoyl-L-t-butyl-Gly-L-alanine-2-aminoethylamide (TAPI-2) was purchased from Peptides International (Louisville, USA). Ketotifen, disodium cromoglycate and ortho-phenanthroline were supplied by Sigma (Taufkirchen, Germany); +Complete® protease inhibitor cocktail by Roche Diagnostics (Mannheim, Germany). The primary anti-TACE antibody for Western blotting was a rabbit polyclonal antibody against a peptide corresponding to amino acids 807–823 of human TACE C-terminus (Chemicon International, USA). Antibodies for colocalization of TACE and TNF-{alpha} in rat ventricular tissue were (i) an affinity-purified goat polyclonal antibody (C-15) against a recombinant peptide of human TACE C-terminus (Santa Cruz Biotechnology, USA) and (ii) a rabbit monoclonal antibody against rat TNF-{alpha} (Endogen, Woburn, USA); secondary antibodies for immunofluorescence were (i) AlexaFluor-488-coupled anti-goat and (ii) AlexaFluor-546-coupled anti-rabbit antibody (Molecular Probes, Eugene, USA). All other reagents used were of analytical grade quality and supplied by Merck (Darmstadt, Germany).

2.2. Cell culture
HMC-1 cells [17] were maintained in ISCOVE's medium (Biochrom, Berlin, Germany) plus 10% iron-supplemented FBS, 1.2 mmol/l monothioglycerol and 10 µg/ml gentamycin (all from Sigma).

2.3. Animals
The investigation conforms with 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).

2.4. Perfusion of isolated rat hearts
Male Wistar rats (body weight 300–500 g) were anesthetized with ether and killed by incision into the carotid arteries. Hearts were rapidly excised and perfused via an aortic cannula at 100 cm H2O in a Langendorff-mode with a modified KHB (pH 7.4, 5.6% CO2/94.4% O2). The pulmonary, caval and azygal veins were ligated and a cannula was inserted into the pulmonary artery. After an equilibration period of 30 min, ischemic control hearts were subjected to 15 min of global ischemia (immersed in Tyrode's buffer at 37 °C) and reperfused (5.6% CO2/94.4% O2) for 120 min. A subset of hearts was treated similarly except for the presence of 100 µmol/l reduced glutathione (GSH) in the perfusate throughout. The group termed "anoxia hearts" was subjected to 15 min of global ischemia, and then reperfused in the absence of oxygen (5.6% CO2/94.4% N2) for 90 min. Longer anoxic peperfusion was not meaningful, because contractility declined markedly after 90 min. "Low-flow hearts" underwent 15 min of global ischemia and then 120 min of reperfusion (5.6% CO2/94.4% O2) at a constant, low rate of flow of 1 ml/min (normal: 4–6 ml/min). Time control hearts were not subjected to ischemia. Infusions of 100 nmol/l ketotifen or 10 µmol/l DSCG were initiated 30 or 15 min prior to ischemia, respectively, in two subset of hearts and continued throughout reperfusion. Further subsets of hearts received an infusion of oxidants (10 min 1 µM HOCl or 30 min 100 µM H2O2) instead of undergoing ischemia.

2.5. Immunodetection of TNF-{alpha} and histamine
Transudate (i.e., interstitial/lymphatic fluid appearing at the epicardial surface) was collected quantitatively from the apex of the heart over periods of 60–90 s at various points of time and snap-frozen in liquid nitrogen. Sample aliquots were analyzed by ELISA according to the manufacturer's protocols (TNF-{alpha}: rat TNF{alpha} ultrasensitive, Biosource; histamine: Beckman Coulter Bioresearch, Krefeld, Germany). Release of TNF-{alpha} and histamine from myocardial tissue was calculated for each time point by multiplication of the respective concentration and flow rate values (collected weight/sampling time) of transudate. Aliquots of coronary venous effluent were collected from the cannulated pulmonary artery and similarly analysed.

2.6. Quantitation of TNF-{alpha} mRNA by RT-PCR
About 200 mg of frozen ventriclular tissue was homogenized by pestle in liquid nitrogen, and 2 ml of TrizolTM reagent (Life Technologies) was added. Total RNA (5 µg) was reverse-transcribed using 20U of RevertAidTM M-MuLV Reverse Transcriptase (MBI Fermentas). A 5-µl aliquot of the resulting cDNA then underwent PCR using a CytoXpress quantitative PCR© Detection Kit (Biosource). The thermocycler profile was: (1 min 96°/4 min 58 °C) x 2; (1 min 94 °C/2.5 min 58 °C) x 30; 10 min 70 °C. The ethidium bromide-stained PCR products were visualized by means of a UV transilluminator (Biorad, Germany) on a 2% agarose gel.

2.7. Immunohistochemical staining of rat ventriclular tissue
Isolated hearts were perfusion-fixed for 2 min with 1% formaldehyde in PBS at a flow rate of 4 ml/min. Staining of paraffin-embedded tissue slices for the presence of mast cells and TNF-{alpha} was performed as described before [18]. Colocalization of TNF-{alpha} and TACE was investigated by immunofluorescence (for antibodies, see Section 2.1) and confocal microscopy (Zeiss LSM 410, Zeiss, Oberkochen, Germany). Samples were inspected with a 63 x oil immersion lens, and illuminated with 488 and 543 nm laser, respectively. A 510–525 nm bandpass and a 570 nm longpass served as emission filters. The size of the pinhole was 20 throughout.

2.8. Preparation of heart tissue homogenates
Hearts were snap-frozen in liquid nitrogen using pre-cooled aluminium tongs. About 175 mg of frozen material was homogenized, first by pestle and then glass homogenizer in cold homogenization buffer (50 mM Tris–HCl; 1 mM DTT; 100 µM o-phenanthroline; 0.2% Nonidet P-40; pH 7.4) supplemented with Complete® protease inhibitor cocktail. Samples were spun at 3900 x g at 4 °C for 30 min and the supernatants collected.

2.9. Lysis of HMC-1 for Western blot
About 106 cells were lysed at 4 °C in 200 µl of lysis buffer (20 mM KH2PO4; 1 mM EDTA; 1 mM PefaBlock; 1 µM leupeptin; 1 µM pepstatin; 50 mM NaF; 40 mM Na4P2O7; 1 mM Na3VO4; 1 mM o-phenanthroline; plus Complete® protease inhibitor cocktail). HMC-1 cell lysates were centrifuged at 12,000 x g for 5 min, the pellets discarded and the supernatants collected.

2.10. Determination of TACE protein by Western blotting
Samples (supernatant of heart homogenates and HMC-1 lysates) were boiled for 7 min with 1/4 volume of 4 x loading buffer (250 mM Tris–HCl, pH 6.8; 8% SDS; 40% glycerol; 0.02% bromophenol blue; 400 mM mercaptoethanol), and defined amounts of protein were loaded on 8% SDS polyacrylamide gel. For blotting after electrophoresis, primary (1:1000) and secondary antibodies were diluted in 3% dry milk powder in PBS. The immunoreactive proteins were visualized by chemiluminescence.

2.11. TNF-{alpha}-peptide cleavage assay with HMC-1 cells
Cells (107) were washed 1 x with calcium-free PBS and were then resuspended in 1 ml PBS. Incubation with 17 µM of the TNF-{alpha}-peptide Ac-SPLAQAVRSSSR-NH2 occurred at 37 °C in cylindrical cups on a rotating device. To compare specificity of cleavage, TNF-{alpha}-peptide was also incubated with recombinant human TACE catalytic domain. Disappearance of the original substrate as well as the corresponding appearance of Ac-SPLAQA, one of the two TACE-specific cleavage products, were monitored by HPLC. Ninety-five to one hundred percent inhibition of degradation by recombinant TACE was achieved with the drug TAPI at concentrations of about 1 µmol/l (IC50: 0.01 µmol/l, data not shown).

2.12. HPLC analysis
Cells were lysed by adding 4N HClO4 to the suspension (volume ratio 1:10) and vortexing immediately. Incubated rhTACE was treated similarly with HClO4. High-pressure liquid chromatography (HPLC) was used to simultaneously quantify the 12-amino-acid peptide substrate of TACE (Ac-SPLAQAVRSSSR-NH2) and the 6-amino-acid fragment Ac-SPLAQA in the acidic extracts. The conditions used were as follows: C18-Aqua® column (Phenomenex, Aschaffenburg, Germany), isocratic elution with 32% HClO4 (pH 2) in methanol (41%) and water (27%), peak detection at 200 nm (L-7400 detector, Hitachi-Merck, Darmstadt, Germany). Retention times at a flow rate of 1 ml/min were 10.1 min for Ac-SPLAQAVRSSSR-NH2 and 15.9 min for Ac-SPLAQA.

2.13. Statistical methods
The sample size of each experimental group is given in the results section. TNF-{alpha} release from hearts is expressed as mean±S.E.M., all other data as mean±S.D. Data from two independent populations were analysed for significant differences using Student t-test (normally distributed data) or Mann–Whitney rank sum test (nonnormal distribution). To test for significant differences between data of two treatment groups at a given point of time, two-way repeated measures ANOVA was performed. In the case of significant differences, the post hoc Student–Newman–Keuls multiple comparison procedure was applied. Differences were considered statistically significant if p≤0.05. All calculating procedures were carried out by the statistical software SigmaStatTM, version 1.0 (Jandel).


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
3.1. Isolated rat hearts release TNF-{alpha} into the interstitium after ischemia
Of the two myocardial fluid compartments (coronary effluent and transudate), only the interstitial transudate collected at the epicardial surface contained detectable amounts of TNF-{alpha}. This finding corresponds to previous results obtained with rat hearts. While control hearts release at most 3 pg/min TNF-{alpha} into the transudate at the end of a perfusion period of 180 min, hearts that underwent 15 min of global ischemia produce up to 50 pg/min TNF-{alpha} during 2 h of aerobic reperfusion (Fig. 1). In these hearts, TNF-{alpha} release displays an early peak 2–10 min after the onset of reperfusion, decreases and then rises again until 120 min after ischemia. Pertinently, almost no TNF-{alpha} mRNA is found in pre-ischemic heart tissue, while after 1 h of reperfusion, transcription of the cytokine gene is enhanced (Fig. 1, insert).


Figure 1
View larger version (28K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 TNF-{alpha} in transudate of rat hearts. Depicted are mean values (±S.E.M.). *Significant vs. pre-ischemic values. #Significant vs. time control hearts. Insert: RT-PCR of total RNA isolated from pre-ischemic (lane 1) and post-ischemic hearts reperfused for 1 h (lane 2), showing TNF-{alpha} and GAPDH (glyceraldehyde-phosphate dehydrogenase) mRNA.

 
3.2. Oxidative stress is required for the post-ischemic release of TNF-{alpha}
A group of hearts was equilibrated in the presence of the antioxidant GSH (100 µmol/l), then underwent global ischemia and was finally reperfused again in the presence of GSH. In this group, both the early and the late release of TNF-{alpha} were reduced, although the late (60–120 min) post-ischemic TNF-{alpha} release still differed significantly from the mean pre-ischemic value (Fig. 2A). Two further groups of ischemic hearts were subjected to modified reperfusion protocols: one was reperfused with KHB gassed with 95% N2, instead of with 95% O2 as in the ischemic control group, the other was reperfused with oxygen-gassed buffer but at a constant, reduced flow rate (1 ml/min compared to reperfusion at constant pressure in ischemic control hearts with a mean coronary flow of 4–6 ml/min). Post-ischemic transudates of low-flow and anoxically reperfused hearts both contained significantly lower levels of TNF-{alpha} than transudates of ischemic control hearts (Fig. 2A).


Figure 2
View larger version (25K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 (A) Mean TNF-{alpha} release (±S.E.M.) of ischemic rat hearts reperfused with different protocols. GSH=perfusate augmented with 100 µmol/l reduced glutathione. *Significant vs. pre-ischemic value. #Significant vs. ischemic control hearts. (B) Mean TNF-{alpha} release (±S.E.M.) of perfused rat hearts treated with oxidants. *Significant vs. pre-ischemic value. #Significant vs. time control hearts.

 
3.3. Infusion of oxidants mimics the effect of ischemia and reperfusion
Two groups of non-ischemic hearts, after an equilibration period of 30 min, were infused for another 30 min with 100 µmol/l H2O2 or for 10 min with 1 µmol/l HOCl, respectively. Both treatment groups showed elevated TNF-{alpha} levels as compared to the untreated control group (Fig. 2B), the TNF-{alpha} production of hearts treated with H2O2 already being significantly enhanced during the early phase of infusion. In contrast, the TNF-{alpha} release of HOCl-treated hearts was not significantly enhanced until 90 min after onset. Interestingly, release continued to rise even after cessation of infusion of either oxidant.

3.4. Ketotifen (100 nmol/l) or DSCG (10 µmol/l) inhibit release of TNF-{alpha} but not of histamine
A quantitatively identical suppression of post-ischemic release of myocardial TNF-{alpha} as seen in the absence of oxygen was observed in the presence of the mast cell stabilizers ketotifen or DSCG (Fig. 3A). Surprisingly, they inhibited both early and late release of the cytokine.


Figure 3
View larger version (22K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 (A) Mean TNF-{alpha} release (±S.E.M.) of ischemic rat hearts perfused in the absence or presence of ketotifen (100 nmol/l) or DSCG (10 µmol/l). *Significant vs. pre-ischemic values. #Significant vs. ischemic control hearts. (B) Mean histamine release (±S.E.M.) of ischemic rat hearts perfused in the absence or presence of 100 nmol/l ketotifen or DSCG (10 µmol/l).

 
Histamine, like TNF-{alpha}, was restricted to the transudate of rat hearts, with no detectable release into the coronary effluent (data not shown). While some hearts released minor amounts of histamine even before ischemia, maximal levels were reached immediately after ischemia. Histamine was then rapidly washed out (Fig. 3B). In transudates of hearts treated with 100 nmol/l ketotifen, the post-ischemic histamine levels as well as the kinetics of histamine release remained unaltered (Fig. 3B). The same result was obtained with the degranulation inhibitor DSCG (Fig. 3B).

3.5. Tissue staining of rat ventricle
Mast cells, as detectable by Alcian blue staining, were frequent in rat, mouse and human myocardium (data not shown). In all three species, they exhibited a preferential perivascular localization. In the heart, mast cells are the major cell type showing positive staining for TNF-{alpha} (Fig. 4). Moreover, they showed a clear colocalization of TACE and TNF-{alpha} (Fig. 4, panels A–D). Especially bright TACE staining appeared perinuclearly and in some granules near the plasma membrane (Fig. 4A). The same peripheral granules also stained positive for TNF-{alpha} (Fig. 4B). Like TACE, most of the TNF-{alpha} was concentrated in the perinuclear space, possibly representing the trans-Golgi network.


Figure 4
View larger version (88K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Immunostaining of tissue sections through rat ventricle. (A) Immunoflourescent staining of TACE in a tissue mast cell. (B) Immunofluorescent staining of TNF{alpha} in the same cell. (C) Colocalization of TACE and TNF-{alpha}. (D) Phase-contrast image of the tissue containing the mast cell (arrow).

 
3.6. Forms of TACE expressed in HMC-1 and rat heart tissue
As previously reported, TACE is located at the plasma membrane of HMC-1 [18]. In lysates of the same cells, TACE protein is abundant (Fig. 5A, lane 1), and two immunoreactive bands can be distinguished by Western blotting: a larger, 100-kDa and a shorter, 80-kDa form, likely representing the zymogen and the mature enzyme, respectively. Interestingly, HMC-1 predominantly express the unprocessed 100-kDa version of TACE.


Figure 5
View larger version (41K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 TACE in Western blot of cell lysates and rat heart homogenates. (A) Lane 1=HMC-1 lysate, lane 2=tissue homogenate of a heart perfused for 180 min (time control), lane 3=tissue homogenate of a heart equilibrated for 45 min, subjected to 15 min of ischemia and reperfused for 120 min. Amounts of total protein loaded per lane: 50 µg for cell lysate, 150 µg for heart homogenates. (B) Lanes 1 and 2=lysates of untreated HMC-1 cells, lanes 3 and 4=lysates of HMC-1 cells stimulated for 1 h with 50 µmol/l HOCl, lanes 5 and 6=lysates of HMC-1 cells stimulated for 1 h with 1 mmol/l H2O2. Amounts of total protein loaded: 50 µg (lanes 1, 3 and 5) or 25 µg (lanes 2, 4 and 6). 100 kDa=pro-TACE, 80 kDa=mature TACE. All blots are representative of at least three independent experiments.

 
In homogenates of rat hearts, the same two protein species can be found, even though the overall expression is lower than in HMC-1 lysates if normalized to the total protein amount loaded (Fig. 5A, lanes 2 and 3). In contrast to HMC-1, the 80-kDa form is the prevalent one in heart tissue. Although the intensity of the 100-kDa "immature TACE" band seems to decrease with ischemia/reperfusion in Fig. 5A (cf. lanes 2 and 3), the ratio of the two bands did not differ significantly beween homogenates of time controls and hearts reperfused for 120 min (ratio 80 kDa/100 kDa TACE=2.99±0.57 and 4.30±2.33 for time control and ischemia/120 min reperfusion, respectively; mean±S.D., n = 4 each). Similarly, the ratio of the two TACE bands in untreated HMC-1 cells (about 0.6) remained the same in lysates of cells treated for 1 h with 50 µmol/l HOCl or 1 mmol/l H2O2. A representative Western blot is shown in Fig. 5B.

3.7. TNF-{alpha}-peptide cleaving activity of HMC-1 cells
Degradation of the peptide Ac-SPLAQA*VRSSSR-NH2 by HMC-1 cells was dependent on incubation time and cell number (Fig. 6A). About 100% degradation was seen after 1 h in the presence of 107 cells/ml. Although the 12-mer was readily degraded by HMC-1, there was never a trace of the TACE-specific cleavage product Ac-SPLAQA in supernatants or lysates. Incubation of HMC-1 cells with Ac-SPLAQA for up to 3 h revealed that the peptide is stable in supernatants and is not taken up by the cells (data not shown).


Figure 6
View larger version (17K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 TNF-{alpha}-peptide cleavage assay with HMC-1 cells. (A) Substrate degradation depending on cell number and incubation time in PBS (representative of two to three similar determinations per condition). (B) Substrate degradation in the absence or presence of different protease/metalloproteinase inhibitors. All incubations were carried out for 1 h at 1 x 107 cells/ml. CON=control; Complete®=protease-inhibitor cocktail; phen=1 mmol/l o-phenanthroline. (C) Stimulation of cells with oxidants. Incubations were carried out for 1 h with 107 cells/ml PBS supplemented with Complete® protease inhibitor cocktail. Concentrations were 100 µmol/l for TAPI-2, 300–500 µmol/l for H2O2 and 50 µmol/l for HOCl. Depicted are mean values (±S.D.); #significant vs. control; *significant vs. value without TAPI-2.

 
Incubation of TNF-{alpha}-peptide for 1 h with 107 HMC-1 cells/ml PBS together with Complete® inhibitor cocktail led to an approximately 40% reduction in nonspecific peptide degradation (Fig. 6B). A further reduction was achieved by adding 100 µmol/l of the TACE inhibitor TAPI-2, while 10 µmol/l TAPI-2 was less effective (Fig. 6B). An almost complete inhibition was accomplished by adding 1 mmol/l o-phenanthroline, a chelator of divalent metal ions and universal metalloproteinase inhibitor (Fig. 6B).

Exposure of cells to oxidants (300–500 µmol/l H2O2 or 50 µmol/l HOCl) stimulated TNF-{alpha}-peptide degradation significantly (Fig. 6C). In both cases, this increase was fully inhibitable by addition of 100 µmol/l TAPI-2 (Fig. 6C).


   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
4.1. Post-ischemic release of TNF-{alpha} depends on oxidative stress
The acute release of TNF-{alpha} from the heart has been observed previously in studies of patients after CABG and in work on reperfused isolated hearts [1,6,19]. However, neither the initiator of release nor the cellular source of the cytokine in the heart was identified. Our initial finding now was that ischemic rat hearts release TNF-{alpha} into the interstitial fluid during early as well as late reperfusion, the early release peaking at 2–10 min after ischemia. These rapid kinetics of TNF-{alpha} release make this cytokine an interesting candidate as a mediator of ischemic preconditioning [20]. The early peak could represent either a release of preformed TNF-{alpha} from granular stores or a rapid shedding of membrane-bound pro-TNF-{alpha} due to enhanced proteolytic cleavage. The late rise in TNF-{alpha} release, in contrast, involves de novo synthesis of the cytokine, as seen by the up-regulation of TNF-{alpha} m-RNA 1 h after ischemia.

The observations that ischemic rat hearts reperfused either anoxically, under low-flow conditions or in the presence of 100 µmol/l GSH produce less TNF-{alpha} than ischemic control hearts are strong indications of the importance of oxidative stress. The absence of the initial TNF-{alpha}-peak in transudate of the GSH-treated hearts suggests that rapid TNF-{alpha} release is particularly dependent on the extracellular redox status and that ischemia itself is not the direct trigger.

Reperfusion of ischemic tissue leads to the generation of a variety of oxygen free radicals and oxidants [16,21]. Zhang et al. [15] have recently demonstrated that H2O2 can activate recombinant TACE by oxidizing the sulfhydryl group of the inhibitory cysteine switch, and Fu et al. [14] have reported activation of an MMP by HOCl, proposing a similar mechanism of interaction. In our experiments, infusion of either HOCl or H2O2 into non-ischemic hearts led to enhanced TNF-{alpha} production as compared to untreated non-ischemic hearts. These results demonstrate that oxidants, even without ischemia, suffice to mediate release of TNF-{alpha} from myocardial tissue. Protracted up-regulation may occur via NF{kappa}B (intracellular redox stress); for early release, enhanced mast cell degranulation or acute activation of TACE and/or related metalloproteinases seem likely mechanisms (extracellular redox stress).

4.2. Role of mast cells
Immunohistochemical staining of rat hearts revealed TNF-{alpha} in high concentration localizing to perivascular mast cells. Similar localizations were found in tissue sections of human and mouse hearts (unpublished data). The proposed role of cardiac mast cells for post-ischemic release of TNF-{alpha} [4,7] could largely be confirmed in our model for ischemia/reperfusion. Notwithstanding, additional sources for TNF-{alpha} in the heart cannot be ruled out strictly by tissue staining. Besides CD68+ tissue macrophages bearing TNF-{alpha} (results not shown), the cytokine might also be shed from cardiac myocytes, especially in hearts subjected to prolonged inflammatory stimulation [22,23].

Immunostaining of rat ventricle revealed that TNF-{alpha} of mast cells is lost within 10 min of reperfusion (unpublished data). Accordingly, at least in cardiac mast cells, a mechanism must exist by which TNF-{alpha} is rapidly released following reperfusion. The observed inhibition of early release by ketotifen or DSCG, at first sight, supports a role of degranulation. In contrast to TNF-{alpha} release, however, histamine release was not influenced by treatment with ketotifen or DSCG at the chosen concentrations.

There is evidence in the literature that the concentration of 100 nmol/l ketotifen, while fully sufficient to prevent release of TNF-{alpha} in our experiments, might simply not be high enough to inhibit discharge of the histamine granules [24]. Interestingly, ketotifen has been reported as an inhibitor of TNF-{alpha} release from peripheral blood mononuclear cells (PBMCs) and was successfully used on HIV-infected patients to treat malnutrition [25]. Furthermore, it has been shown to be beneficial in the treatment of some forms of physical urticaria (PU), where ketotifen reduced both cutaneous and serum expression of TNF-{alpha} in patients [26]. Additionally, a protective effect has been assigned to ketotifen in a model of experimental inflammation in the small intestine held to mimic Crohn's disease and which may involve TNF-{alpha} [27].

The fact that myocardial release of histamine in our model was also not suppressed by 10 µmol/l DSCG further supports the hypothesis that release of TNF-{alpha} and histamine from mast cells during early reperfusion are unlinked processes. While the post-ischemic TNF-{alpha} release is clearly triggered by reperfusion, histamine is already released in substantial amounts by mast cells during ischemia, leading to elevated histamine levels in the interstitium [28] and allowing rapid washout during reperfusion. Interestingly, even in the presence of 10 µmol/l DSCG, tissue histamine levels rose during the first 15 min of global ischemia [28]. The obvious difficulty in inhibiting histamine release from mast cells during myocardial ischemia might be due to ATP-depletion, leading to Ca2+ influx and piecemeal degranulation.

The protracted post-ischemic release of myocardial TNF-{alpha} observed in our experiments is presumably a result of enhanced TNF-{alpha} expression (Fig. 1, insert). The inhibition of the late phase obtained with ketotifen or DSCG indicates that mast cell stimulation seems to be vital for initiating this process. An autocrine stimulation of TNF-{alpha} expression has been described before [29]. Alternatively, as yet unidientified products of mast-cell degranulation might serve as trigger for late TNF-{alpha} release. Histamine alone obviously does not suffice.

4.3. Role of TACE
Both forms of TACE, the zymogen (100 kDa) and the mature enzyme (80 kDa), were detected in homogenized rat heart tissue, although the processed form (lacking the cysteine switch containing prodomain) was clearly the predominant one. Thus, there seems to be rather little scope for oxidative activation of pro-TACE. Furthermore, the ratio of the 100-kDa to the 80-kDa band did not shift significantly towards the smaller species after ischemia, which argues against acutely enhanced proteolytic processing of TACE during reperfusion. However, should oxidative or proteolytic activation of TACE occur only in the small subset of cardiac mast cells, it would be overlooked if whole heart homogenates are analysed.

HMC-1 served as model system for cardiac mast cells in order to better assess the role of oxidative stress in TACE-activation. Pro-TACE was not only abundant in lysates of HMC-1; it even predominated over the 80-kDa form. However, exposure to oxidants did not shift the relationship. Thus, one can imagine a rapid rise of enzyme activity via oxidation of the "cysteine switch" as likely alternative to its activation via conversion by a furin-like protease. Unfortunately, HMC-1 expressed hardly any TNF-{alpha} in the nonstimulated state. Hence, use was made of the TNF-{alpha}-peptide Ac-SLPAQAVRSSSR-NH2 as substrate for cleavage. HMC-1 rapidly degraded the peptide, but not to the TACE-specific hexamer Ac-SPLAQA, one of two products obtained after incubation of TNF-{alpha}-peptide with recombinant human TACE catalytic domain [8]. On the other hand, the rate of substrate degradation was increased by addition of 500 µmol/l H2O2 or 50 µmol/l HOCl to the cells and proved susceptible to inhibition by TAPI-2, but only at 100 µmol/l, at which concentration TAPI-2 is no longer specific for TACE. One ready explanation for these observations is the existence of (an)other substrate-degrading activity(ies), similar to but distinct from TACE. As degradation of the TNF-{alpha}-peptide was almost completely blocked by 1 mmol/l o-phenanthroline, the enzyme(s) in question must be (a) metalloproteinase(s). TNF-{alpha}-converting activity has been demonstrated previously for at least ADAM-10 and MMP-7 and -17 [30–32].

Taken together, our results clearly show that an oxidation-sensitive mechanism initiates post-ischemic myocardial release of TNF-{alpha}. Data obtained with HMC-1 cells and myocardial tissue make it unlikely that this is accomplished via enhanced processing of pro-TACE. In fact, shedding of TNF-{alpha} even seems to be largely independent of TACE, relying on other oxidatively activated metalloproteinases. In our heart model, early TNF-{alpha} release seems to stem from mast cell activation. Preliminary data using KitW/KitW/W-v mice [33], a knockout line with very few myocardial mast cells, show that these animals lack the early post-ischemic peak of cytokine release (unpublished data). The later release, in contrast, requires de novo synthesis of TNF-{alpha}, possibly induced by TNF-{alpha} itself and/or intracellular oxidative stress. These conclusions have direct bearing on the situation in the human heart: After CABG, an absolutely identical time-course of coronary TNF-{alpha} release has been observed, with an early and a later phase, and both were mitigated by generating the radical scavenger nitric oxide [34]. Interestingly, a negative correlation exists between the amount of TNF-{alpha} released during the first 5 min of reperfusion and the cardiac index determined several hours after the intervention [19]. Owing to the early release of TNF-{alpha}, as well as the inflammatory potential of the cytokine, antioxidative interventions and mast cell stabilization in general will surely influence post-ischemic reperfusion damage. While TACE inhibitors may turn out to be ineffective, mast cell stabilizers could possess an unexpected potential in this regard.


   Acknowledgements
 
The authors thank Veronika Frei, Dora Kiesl, Pia Unterberger and Dorothee Goessel for expert technical assistance.


   Notes
 
* Supported by SFB 469 of the German Research Council (DFG), Bonn, Germany.

Time for primary review 25 days


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

  1. Zahler S., Massoudy P., Hartl H., Hahnel C., Meisner H., Becker B.F. Acute cardiac inflammatory responses to postischemic reperfusion during cardiopulmonary bypass. Cardiovasc. Res. (1999) 41:722–730.[Abstract/Free Full Text]
  2. Suzuki K., Murtuza B., Smolenski R.T., et al. Overexpression of interleukin-1 receptor antagonist provides cardioprotection against ischemia–reperfusion injury associated with reduction in apoptosis. Circulation (2001) 104:I308–I313.[ISI][Medline]
  3. Kawamura T., Wakusawa R., Okada K., Inada S. Elevation of cytokines during open heart surgery with cardiopulmonary bypass: participation of interleukin 8 and 6 in reperfusion injury. Can. J. Anaesth. (1993) 40:1016–1021.[Abstract/Free Full Text]
  4. Frangogiannis N.G., Lindsey M.L., Michael L.H., et al. Resident cardiac mast cells degranulate and release preformed TNF-alpha, initiating the cytokine cascade in experimental canine myocardial ischemia/reperfusion. Circulation (1998) 98:699–710.[Abstract/Free Full Text]
  5. Yokoyama T., Vaca L., Rossen R.D., Durante W., Hazarika P., Mann D.L. Cellular basis for the negative inotropic effects of tumor necrosis factor-alpha in the adult mammalian heart. J. Clin. Invest. (1993) 92:2303–2312.[ISI][Medline]
  6. Kupatt C., Habazettl H., Goedecke A., et al. Tumor necrosis factor-alpha contributes to ischemia and reperfusion-induced endothelial activation in isolated hearts. Circ. Res. (1999) 84:392–400.[Abstract/Free Full Text]
  7. Entman M.L., Youker K.A., Frangogiannis N., et al. Is inflammation good for the ischemic heart—perspectives beyond the ordinary. Z. Kardiol. (2000) 89(Suppl. 9):IX/82–IX/87.[CrossRef]
  8. Black R.A., Rauch C.T., Kozlosky C.J., et al. A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells. Nature (1997) 385:729–733.[CrossRef][Medline]
  9. Primakoff P., Myles D.G. The ADAM gene family: surface proteins with adhesion and protease activity. Trends Genet. (2000) 16:83–87.[CrossRef][ISI][Medline]
  10. Black R.A. Tumor necrosis factor-alpha converting enzyme. Int. J. Biochem. Cell Biol. (2002) 34:1–5.[CrossRef][ISI][Medline]
  11. Loechel F., Overgaard M.T., Oxvig C., Albrechtsen R., Wewer U.M. Regulation of human ADAM 12 protease by the prodomain. Evidence for a functional cysteine switch. J. Biol. Chem. (1999) 274:13427–13433.[Abstract/Free Full Text]
  12. Van Wart H.E., Birkedal-Hansen H. The cysteine switch: a principle of regulation of metalloproteinase activity with potential applicability to the entire matrix metalloproteinase gene family. Proc. Natl. Acad. Sci. U. S. A. (1990) 87:5578–5582.[Abstract/Free Full Text]
  13. Clarke H.R., Wolfson M.F., Rauch C.T., et al. Expression and purification of correctly processed, active human TACE catalytic domain in Saccharomyces cerevisiae. Protein Expr. Purif. (1998) 13:104–110.[CrossRef][ISI][Medline]
  14. Fu X., Kassim S.Y., Parks W.C., Heinecke J.W. Hypochlorous acid oxygenates the cysteine switch domain of pro-matrilysin (MMP-7). A mechanism for matrix metalloproteinase activation and atherosclerotic plaque rupture by myeloperoxidase. J. Biol. Chem. (2001) 276:41279–41287.[Abstract/Free Full Text]
  15. Zhang Z., Oliver P., Lancaster J.R. Jr., et al. Reactive oxygen species mediate tumor necrosis factor alpha-converting, enzyme-dependent ectodomain shedding induced by phorbol myristate acetate. FASEB J. (2001) 15:303–305.[Free Full Text]
  16. Becker B.F., Kupatt C., Massoudy P., Zahler S. Reactive oxygen species and nitric oxide in myocardial ischemia and reperfusion. Z. Kardiol. (2000) 89(Suppl. 9):IX/88–IX/91.[CrossRef]
  17. Butterfield J.H., Weiler D., Dewald G., Gleich G.J. Establishment of an immature mast cell line from a patient with mast cell leukemia. Leuk. Res. (1988) 12:345–355.[CrossRef][ISI][Medline]
  18. Becker B.F., Gilles S., Gonscherowski V., et al. Assessing experimental models in myocardial injury: lack of activation of proteases TACE and calpain in brief ischemia and reperfusion. Heart Lung Circ. (2003) 12:51–59.[CrossRef][Medline]
  19. Freyholdt T., Massoudy P., Zahler S., et al. Beneficial effect of sodium nitroprusside after coronary artery bypass surgery. J. Cardiovasc. Pharmacol. (2003) 42:372–378.[CrossRef][ISI][Medline]
  20. Belosjorow S., Bolle I., Duschin A., Heusch G., Schulz R. TNF-{alpha} antibodies are as effective as ischemic preconditioning in reducing infarct size in rabbits. Am. J. Physiol. Heart Circ. Physiol. (2002) 284:H927–H930.[ISI][Medline]
  21. Lefer D.J., Granger D.N. Oxidative stress and cardiac disease. Am. J. Med. (2000) 109:315–323.[CrossRef][ISI][Medline]
  22. Meldrum D.R. Tumor necrosis factor in the heart. Am. J. Physiol. (1998) 274:R577–R595.[ISI][Medline]
  23. Kapadia S., Lee J., Torre-Amione G., Birdsall H.H., Ma T.S., Mann D.L. Tumor necrosis factor-alpha gene and protein expression in adult feline myocardium after endotoxin administration. J. Clin. Invest. (1995) 96:1042–1052.[ISI][Medline]
  24. Vleeming W., van Rooij H.H., Wemer J., Porsius A.J. Characterization and modulation of antigen-induced effects in isolated rat heart. J. Cardiovasc. Pharmacol. (1991) 18:556–565.[ISI][Medline]
  25. Ockenga J., Rohde F., Suttmann U., Herbarth L., Ballmaier M., Schedel I. Ketotifen in HIV-infected patients: effects on body weight and release of TNF-alpha. Eur. J. Clin. Pharmacol. (1996) 50:167–170.[CrossRef][ISI][Medline]
  26. Frezzolini A., De Pita O., Cassano N., et al. Evaluation of inflammatory parameters in physical urticarias and effects of an anti-inflammatory/antiallergic treatment. Int. J. Dermatol. (2002) 41:431–438.[CrossRef][ISI][Medline]
  27. Rachmilewitz D., Okon E., Karmeli F. Sulphydryl blocker induced small intestinal inflammation in rats: a new model mimicking Crohn's disease. Gut (1997) 41:358–365.[Abstract/Free Full Text]
  28. Davani S., Muret P., Royer B., et al. Ischaemic preconditioning and mast cell histamine release: microdialysis of isolated rat hearts. Pharmacol. Res. (2002) 45:383–390.[CrossRef][ISI][Medline]
  29. Wu S., Boyer C.M., Whitaker R.S., et al. Tumor necrosis factor alpha as an autocrine and paracrine growth factor for ovarian cancer: monokine induction of tumor cell proliferation and tumor necrosis factor alpha expression. Cancer Res. (1993) 53:1939–1944.[Abstract/Free Full Text]
  30. Lunn C.A., Fan X., Dalie B., et al. Purification of ADAM 10 from bovine spleen as a TNFalpha convertase. FEBS Lett. (1997) 400:333–335.[CrossRef][ISI][Medline]
  31. Haro H., Crawford H.C., Fingleton B., Shinomiya K., Spengler D.M., Matrisian L.M. Matrix metalloproteinase-7-dependent release of tumor necrosis factor-alpha in a model of herniated disc resorption. J. Clin. Invest. (2000) 105:143–150.[ISI][Medline]
  32. English W.R., Puente X.S., Freije J.M., et al. Membrane type 4 matrix metalloproteinase (MMP17) has tumor necrosis factor-alpha convertase activity but does not activate pro-MMP2. J. Biol. Chem. (2000) 275:14046–14055.[Abstract/Free Full Text]
  33. Maurer M., Echtenacher B., Hültner L., et al. The c-kit ligand, stem cell factor, can enhance innate immunity through effects on mast cells. J. Exp. Med. (1998) 188:2343–2348.[Abstract/Free Full Text]
  34. Massoudy P., Zahler S., Freyholdt T., et al. Sodium nitroprusside in patients with compromised left ventricular function undergoing coronary bypass: reduction of cardiac proinflammatory substances. J. Thorac. Cardiovasc. Surg. (2000) 119:566–574.[Abstract/Free Full Text]

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



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