Molecular Medicine |
From the Rudolf Virchow Center (S.M., M.P.S., M.B., R.J.A.F., S.E.), Deutsche Forschungsgemeinschaft-Research Center for Experimental Biomedicine; Department of Internal Medicine (S.F., J.B.); Department of Dermatology, Venereology and Allergology (M.P.S.); and Institute of Pharmacology and Toxicology (E.S., M.J.L.), University of Wuerzburg, Germany.
Correspondence to Stefan Engelhardt, MD PhD, Rudolf Virchow Center, DFG-Research Center for Experimental Biomedicine, University of Wuerzburg, Versbacher Strasse 9, 97078 Wuerzburg, Germany. E-mail stefan.engelhardt{at}virchow.uni-wuerzburg.de
| Abstract |
|---|
|
|
|---|
Key Words: caspase-1 transgenic mouse heart failure
| Introduction |
|---|
|
|
|---|
Caspases are cysteine proteases that cleave their specific substrates after an aspartate residue. Caspases have been divided into proapoptotic (Caspases 2, 3, 6, 7, 8, 9, 10) and proinflammatory (Caspases 1, 4, 5, 11, 12) respectively.8,9 Whereas proapoptotic caspases have been implicated in the pathogenesis of various cardiovascular disorders including heart failure and cardiac ischemia,1012 the role of the proinflammatory caspases in cardiovascular diseases is less understood.
Caspase-1 (also termed interleukin-converting enzyme), a member of the proinflammatory family of caspases, was discovered through its proteolytic activation of pro-IL-1ß and -18 into their active form.13,14 Caspase-1 is synthesized as an weakly active 45-kDa precursor that is autocatalytically processed via an intermediate into p10 and p20 subunits, forming the enzymatically full active heterodimeric or tetrameric enzyme.1518
Controversial data regarding the functional role of caspase-1 have been obtained in different species and assay systems. Whereas few studies have implicated caspase-1 in programmed cell death in certain cell types,19,20 other studies argue against a general role for caspase-1 in the regulation of apoptosis.21,22 Recent studies indicate a important role for caspase-1 in cardiac ischemia.2325 These studies implicate both the formation of interleukins25 and the induction of apoptosis23,24 in the detrimental effects of caspase-1 activation during ischemia. In contrast, the role of caspase-1 in the development of heart failure has not been studied to date.
Here we provide evidence that caspase-1 expression is increased in murine and human heart failure and that caspase-1 constitutes a critical regulator of cardiomyocyte programmed cell death in the mammalian heart.
| Materials and Methods |
|---|
|
|
|---|
isoform under the control of the murine
-myosin heavy chain (
-MHC) promoter. Experiments were performed with mice of line TG10, unless indicated otherwise. All mice were housed in a specified pathogen free facility. All animal experiments were approved by the responsible authorities (protocol no. 6212531.01-10/98, 28/01, and 31/03).
Adenovirus Generation
Recombinant adenoviral vectors were generated according to standard procedures. Caspase-1 (human caspase-1
isoform) was cloned by PCR and introduced into a adenoviral vector (pAD/CMV/V5-DEST, Invitrogen, Carlsbad, Calif) by homologous recombination. A corresponding LacZ-expressing adenoviral vector (pAd/CMV/LacZ) was used as control. Adenovirus titers were determined by plaque assays in HEK293 monolayer cultures embedded in agarose.
Human Cardiac Tissue
Failing hearts were obtained from patients undergoing heart transplantation resulting from terminal heart failure. Nonfailing donor hearts that could not be transplanted for technical reasons were used as controls. Donor patient histories and echocardiography revealed no signs of heart disease. Immediately after removal of the hearts, samples were taken, snap-frozen in liquid nitrogen and stored at 80°C. In accordance with the Declaration of Helsinki, written informed consent was obtained from all patients or the families of prospective heart donors before cardiectomy. The experiments were approved by the Ethics Committee of the Wuerzburg Medical Faculty.
Cardiomyocyte Isolation
Neonatal and adult cardiomyocytes were isolated as described previously.26,27 Cardiomyocyte length and width were determined from digitally recorded images using the MetaVue software package (Visitron, Puchheim, Germany).
Determination of IL-1ß and -18
IL-1ß ELISA (mouse IL-1ß ELISA Kit, EMIL1B, Pierce Endogen, Ill) and IL-18 ELISA (mouse IL-18 ELISA, BMS618, Bender Med Systems, Vienna, Austria) were performed with lysates from whole left ventricular myocardium and sera of 2-month-old caspase-1 transgenic and corresponding wild-type animals. For details, see the expanded Materials and Methods section in the online data supplement, available at http://circres.ahajournals.org.
Determination of Cardiomyocyte Apoptosis and Cell Viability
Nuclear DNA fragmentation was detected by terminal deoxynucleotidyl transferasemediated dUTP nick end-labeling (TUNEL) staining using a commercial kit (Roche Diagnostics, Indianapolis, Ind). For details, see the expanded Materials and Methods section in the online data supplement.
Myocardial Infarction, Ischemia/Reperfusion, Hemodynamic and Echocardiographic Analysis
See the expanded Materials and Methods section in the online data supplement.
Histochemical and Immunohistochemical Analyses
Interstitial fibrosis, cardiomyocyte hypertrophy, and determination of necrosis and reactive oxygen species formation were determined as described previously.2831
For further details and immunohistochemical analysis, see the expanded Materials and Methods section in the online data supplement.
RNA Isolation and Real-Time RT-PCR
See the expanded Materials and Methods section in the online data supplement.
Immunoblotting
See the expanded Materials and Methods section in the online data supplement.
Statistical Analysis
Average data are presented as mean±SEM. Statistical analysis was performed using the Prism software package (GraphPad, San Diego, Calif). ANOVA followed by Bonferroni test and Students t test were used as appropriate. Differences were considered significant when P<0.05 and are indicated by an asterisk in the figures.
| Results |
|---|
|
|
|---|
|
Caspase-1 Induces Cardiomyocyte Hypertrophy Despite Unchanged Ventricular Weights
To test the hypothesis that increased expression of caspase-1 contributes to the development of heart failure, transgenic mice with overexpression of caspase-1 in the heart were generated. We used the murine
-MHC promoter to target expression of the caspase-1 cDNA to cardiac myocytes (Figure 2A, top). We identified 4 positive founder mice, 2 of which carried the transgene in their germline and gave rise to 2 independent transgenic lines (termed TG1 and TG10; Figure IA in the online data supplement). Western blotting demonstrated significant overexpression of transgenic caspase-1 protein, which was confined to the ventricular myocardium (endogenous caspase-1 is expressed in the spleen and, to a lesser extent, in the lung33,34), with slightly higher levels in TG10 as compared with TG1 (supplemental Figure IB and IC). Figure 2A, bottom, shows expression of caspase-1 (45 kDa) and its processing.
|
Despite the marked ventricular dilatation observed in caspase-1 transgenic mice (Figure 3A and 3B, bottom left), the ventricular weights of caspase-1 transgenic hearts were not increased at any age studied (Figure 2B, top). However, we detected intense remodeling of the ventricular myocardium from 4 months onward. Whereas no significant cardiomyocyte hypertrophy was present at 1 month of age, cardiomyocyte size progressively increased starting at 4 months of age (Figure 2B, bottom). In addition, we analyzed the cell dimensions of isolated adult cardiomyocytes from 5-month-old caspase-1 transgenic and wild-type mice. The average cardiomyocyte length increased from 117 to 130 µm (P<0.01), and the width increased from 26 to 37 µm (P<0.0001) through overexpression of caspase-1 (n=60 cells per genotype).
|
Messenger RNA expression of atrial natriuretic factor (ANF), ß-myosin heavy chain (ß-MHC), and sarcoplasmic reticulum ATPase (SERCA) in the myocardium of caspase-1 transgenic mice was unchanged in 1-month-old mice (Figure 2C). ANF and ß-MHC expression increased, whereas SERCA mRNA decreased at 4 months of age.
In addition, we stained left ventricular sections from wild-type and caspase-1 transgenic hearts from different age groups with Sirius red to assess the extent of interstitial collagen deposition. We detected interstitial fibrosis throughout the ventricular myocardium in caspase-1 transgenic mice from 4 months of age onward but not in 1-month-old animals (Figure 2D). Likewise, line TG1 displayed significant cardiomyocyte hypertrophy and increased interstitial collagen content as compared with wild-type control mice (data not shown).
Taking the marked cardiomyocyte hypertrophy and the prominent interstitial fibrosis into account, the absence of an increase in the left ventricular weight of these animals suggested a marked loss of cardiomyocytes. To prove loss of cardiomyocytes in this setting, one would ideally like to determine precisely the total number of myocytes as well as other cell types for an adult mouse heart, a technically very difficult undertaking. Rather, we aimed to test a direct apoptotic effect of caspase-1 in cardiomyocytes in a defined in vitro setting (see Figure 5).
Expression of Caspase-1 Is Sufficient to Induce Cardiac Failure, Whereas Deletion of Endogenous Caspase-1 Ameliorates Postinfarction Heart Failure
Although caspase-1 transgenic mice displayed normal cardiac morphology at young age (2 months), we observed thinning of the left ventricular wall and cardiac dilatation with increasing age. In addition, left atrial dilatation and thrombus formation were frequently observed in caspase-1 transgenic mice, consistent with the development of cardiac failure (Figure 3A). When the left ventricular function of these mice was assessed in vivo by left ventricular catheterization using a 1.4 F micromanometer tip catheter, a significant impairment of left ventricular contractility in caspase-1 transgenic mice (1 year of age) became apparent (P<0.05) (Figure 3B, top left). This was not caused by alterations in heart rate, as the beating frequency was the same in wild-type and caspase-1 transgenic hearts (471±18 bpm, wild-type, versus 468±22 bpm, caspase-1 transgenic) (data not shown). In addition, there was a modest but significant decline of the maximal left ventricular pressure (105±2 mm Hg, wild-type, versus 88±2 mm Hg, caspase-1 transgenic, P<0.0001) and a marked increase of left ventricular end-diastolic pressure, indicating impaired cardiac function (Figure 3B, top right). Furthermore, echocardiographic analysis demonstrated cardiac dilatation and a decrease in fractional shortening in caspase-1 transgenic mice (Figure 3B, bottom left and right).
We then assessed the functional role of endogenous caspase-1 in a model of postinfarction cardiac failure. Homozygous caspase-1deficient mice were subjected to myocardial infarction through permanent left anterior descending ligation and followed for 8 weeks thereafter. Although we did not observe a significant influence of endogenous caspase-1 on infarct size (Figure 3C, top left), caspase-1deficient mice displayed reduced cardiomyocyte hypertrophy in comparison with wild-type control mice (Figure 3C, top middle and right). Survival was doubled in caspase-1deficient mice compared with wild-type mice (Figure 3C, bottom left). Furthermore, deletion of caspase-1 rescued in part the decline of left ventricular function observed in wild-type animals after myocardial infarction (Figure 3C, bottom right). Interestingly, this occurs independent of infarct size. These data provide evidence that endogenous caspase-1 activity indeed contributes to the deterioration of left ventricular function after myocardial infarction.
Caspase-1 Induces Cardiomyocyte Apoptosis Independent of Interleukin Formation and Inflammation
Given that caspase-1 and its downstream regulated cytokines have been implicated in the proinflammatory response to a variety of stimuli in various cell types,35,36 we determined sera and cardiac IL-1ß and IL-18 levels of caspase-1 transgenic and wild-type mice. Despite robust overexpression of caspase-1, IL-1ß, and -18 levels in the myocardium and sera of caspase-1 transgenic mice were essentially indistinguishable from those of control hearts (Figure 4A). The absence of IL-1ß formation was confirmed by immunohistochemistry (Figure 4B, left). Furthermore, we studied several proinflammatory mediators that have been implicated in the downstream response of IL-1ß formation. IL-1
, IL-4, IL-6, tumor necrosis factor-
, interferon-
, and granulocyte/macrophage colonystimulating factor were essentially unchanged within the myocardium of caspase-1 transgenic compared with wild-type mice (data not shown). In parallel, we assessed the presence of inflammatory cells such as neutrophilic granulocytes and lymphocyte subsets, which represent the prototypical inflammatory response to necrotic cell death. All of these markers yielded identical results for wild-type and caspase-1 transgenic mice. However, staining for macrophages (CD18 positive and CAE negative) as typical postapoptotic response showed increased numbers of macrophages in the myocardium of caspase-1 transgenic mice already at 2 months of age (supplemental Figure II).
|
To directly determine whether cardiomyocyte necrosis occurs to a significant extent, we performed immunohistochemistry and stained for complement component C9, a marker for necrotic cell death. We did not observe a significant increase of complement 9 staining in caspase-1 transgenic mice as compared with wild-type mice (Figure 4B, right). In accordance with these findings, we found reactive oxygen species formation as assessed by 8-hydroxy-2'-deoxyguanosine and dihydroethidium staining to be essentially unaltered between wild-type and caspase-1 transgenic mice (data not shown).
Given that an alternative mechanism of action of caspase-1 is the induction of programmed cell death, we determined the extent of apoptosis in the myocardium of caspase-1 transgenic mice by TUNEL. In contrast to the structural alterations observed during the remodeling process, which appeared only with increasing age of the animals (see Figure 2), apoptosis was markedly increased already at the youngest age group studied (Figure 4C, top and bottom left). In addition, we found a massive increase of apoptosis in adult cardiomyocytes isolated from caspase-1 transgenic mice (Figure 4C, right), confirming that caspase-1 is sufficient to induce cardiomyocyte apoptosis.
In a series of complementary experiments, we demonstrate the contribution of endogenous caspase-1 to the induction of cardiomyocyte apoptosis in vivo. For the rapid induction of a high extent of cardiomyocyte apoptosis, we used a 30-minute ischemia/reperfusion protocol in wild-type and caspase-1deficient mice. In line with the equal infarct sizes after myocardial infarction (see Figure 3C, top left), the planimetric analysis of the infarct area yielded identical results for wild-type and caspase-1deficient mice (Figure 4D, left). This protocol resulted in a marked induction of cardiomyocyte apoptosis in wild-type mice. In contrast, caspase-1deficient mice were significantly protected from cardiomyocyte apoptosis, as evidenced by a 3-fold reduction of TUNEL-positive cardiomyocytes (Figure 4D, top and bottom right). The reduction of cardiomyocyte apoptosis in caspase-1deficient mice may eventually result in a reduction of infarct size, possibly becoming evident after longer periods of follow-up.
These data indicate that caspase-1 causes its effects primarily through the induction of apoptosis rather than through a proinflammatory response.
Caspase-1 Induces Cardiomyocyte Apoptosis via Activation of Caspases-3 and -9
To assess the proapoptotic effect of caspase-1 in a defined experimental setting in vitro, we generated recombinant adenoviral vectors (Adv-Casp-1) to express caspase-1 in neonatal rat cardiomyocytes. Western blot analysis confirmed activation of adenoviral-expressed caspase-1 (Figure 5A). Adv-Casp-1 led to a dose-dependent induction of cardiomyocyte apoptosis 48 hours after infection. Already at low multiplicity of infection (MOI) (MOL), cardiomyocyte apoptosis increased 4.6±0.6-fold relative to neonatal rat cardiomyocytes infected with a adenovirus expressing LacZ (Adv-lacZ) (Figure 5B, left and top right). Analysis of cell viability confirmed the dose-dependent cardiomyocyte death induced by expression of caspase-1 (Figure 5B, bottom right).
|
We further assessed the activation of caspase-3 as a hallmark of cardiomyocyte apoptosis. Expression of caspase-1 in neonatal rat cardiomyocytes led to a significant increase of caspase-3 activation (Figure 5C, left). To investigate whether the intrinsic and/or extrinsic pathways are activated through caspase-1 in neonatal rat cardiomyocytes, we determined the activation of caspase-9 (intrinsic pathway) and caspase-8 (extrinsic pathway). Whereas we found caspase-8 activity essentially unchanged (data not shown), cleaved caspase-9 was significantly increased after incubation of neonatal rat cardiomyocytes with Adv-Casp-1 (P<0.0001; Figure 5C, right). The apoptotic response induced by caspase-1 was blocked both by the broad-spectrum caspase inhibitor zVAD-fmk and the caspase-1specific inhibitor RU 36384 but not by the caspase-8specific inhibitor zIETD-fmk (Figure 5D).
| Discussion |
|---|
|
|
|---|
We confirmed that caspase-1 protein is upregulated in murine heart failure, which corroborates earlier data obtained on the mRNA level.3739 The early and parallel occurrence of caspase-1 expression and cardiomyocyte apoptosis supports the notion that these 2 events are causally related. This is followed (but not paralleled) by myocardial remodeling, the induction of a fetal gene expression program, and, finally, the development of overt heart failure. Deletion of endogenous caspase-1 was without effect under basal conditions but ameliorated the development of heart failure after myocardial infarction. In the setting of ischemia/reperfusion (known to be a very strong stimulus for cardiomyocyte apoptosis), deletion of caspase-1 effectively protected against cardiomyocyte apoptosis. Comparably modest expression of caspase-1 in isolated cardiomyocytes at low multiplicities of infection revealed the potent proapoptotic activity of caspase-1. Thus increased caspase-1 expression appears to primarily induce cardiomyocyte apoptosis which, consecutively, leads to the development and progression of heart failure.
In contrast to heart failure, several studies have investigated the role of caspase-1 in the pathogenesis of ischemic heart disease. Most importantly, mice deficient in caspase-1 have been subjected to myocardial infarction and followed for up to 9 days.24 These mice displayed a significant reduction in mortality within 24 hours after myocardial infarction, correlating with reduced levels of cardiomyocyte apoptosis. We also saw a marked improvement of survival resulting from deletion of endogenous caspase-1. Syed et al reported that cardiomyocyte caspase-1 induces apoptosis during cardiac ischemia,23 whereas a study on isolated human atrial tissue found ischemic myocardial dysfunction to be a result of caspase-1dependent IL-1 formation.25
Our data argue for activation of caspase-1 under basal conditions in the heart in vivo and thus are in line with several other studies that have defined an autocatalytic mechanism of caspase-1 activation.17,18 However, there is 1 recent study in which caspase-1 activation was detected only under conditions of cardiac ischemia.23 At present, we cannot identify the relevant factors, which might contribute to the observed differences to what Syed et al have reported. Potential differences are the gender of the animals studied (whereas we have studied exclusively male mice, Syed et al did not indicate the sex of the mice used for their study) and factors such as the animal chow40 and the level of transgene expression.
Our data suggest that the marked increase of caspase-1 expression in human failing myocardium is pathophysiologically relevant, however, not through the formation of proinflammatory cytokines, as might be assumed. Increased IL-1ß serum levels have been described in patients with heart failure and have been suggested to contribute to the progression of the disease.41,42 Although we initially speculated that this might be caused, at least partly, by the marked increase of caspase-1 expression in the failing heart, our data do not support a role for cardiac caspase-1 in this regard. We have determined serum and tissue levels of IL-1ß in caspase-1 transgenic mice and found that IL-1ß levels were low and unaltered. This is unlikely attributable to the expression of the human caspase-1 isoform in our transgenic model, as human caspase-1 has been found to be highly similar to murine caspase-1 with respect to its substrate requirements and its inhibition by specific inhibitors.43 Specifically, the cleavage of the murine IL-1ß precursor is nearly identical to that of the human IL-1ß precursor.44
As an additional test for potential subthreshold quantities of IL-1ß formation, we assessed typical IL-1ßmediated effects on cardiomyocytes that have been previously reported. In this respect, several studies have shown cardiomyocyte hypertrophy and activation of the fetal gene expression program through IL-1ß.4548 In sharp contrast, we found unchanged cardiomyocyte dimensions and a completely normal fetal gene expression profile in 1-month-old animals. Furthermore, IL-1ß, as well as various proinflammatory mediators known to be upregulated through IL-1ß were not detectably elevated in caspase-1 transgenic mice. However, even at this early age, cardiomyocyte apoptosis was very prominent throughout the left ventricular myocardium, thus providing evidence for caspase-1 exerting a proapoptotic function specifically in cardiomyocytes.
In contrast to previous findings obtained in other cell types implying a proinflammatory role of caspase-1, our data suggest a primary proapoptotic role for caspase-1 in the heart. Caspase-1 expression results in cleavage of caspases-9 and -3, but not caspase-8, suggesting activation of the intrinsic pathway of apoptosis. Taken together, our data reveal a novel functional role for caspase-1 in the pathogenesis of experimental and human heart failure. Contrary to previous assumptions, cardiac caspase-1 appears to primarily induce apoptotic cell death, as opposed to the formation of inflammatory cytokines. Given the availability of highly specific caspase-1 inhibitors,49 caspase-1 may represent a therapeutic target molecule to prevent the progression of cardiac remodeling during the course of heart failure.
| Acknowledgments |
|---|
Sources of Funding
These studies were supported by grants from the Deutsche Forschungsgemeinschaft (SFB 598, SFB 688), the German Federal Ministry of Education and Research (Competence Network of Heart Failure [TP7 to J.B. and S.E.]), and the Rudolf Virchow Center/Deutsche Forschungsgemeinschaft-Research Center for Experimental Biomedicine, supported by ProCorde, Sanofi-Aventis, and the Bavarian Ministry of Economics.
Disclosures
None.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. J. Birks, N. Latif, K. Enesa, T. Folkvang, L. A. Luong, P. Sarathchandra, M. Khan, H. Ovaa, C. M. Terracciano, P. J.R. Barton, et al. Elevated p53 expression is associated with dysregulation of the ubiquitin-proteasome system in dilated cardiomyopathy Cardiovasc Res, August 1, 2008; 79(3): 472 - 480. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Abbate, F. N. Salloum, E. Vecile, A. Das, N. N. Hoke, S. Straino, G. G.L. Biondi-Zoccai, J.-E. Houser, I. Z. Qureshi, E. D. Ownby, et al. Anakinra, a Recombinant Human Interleukin-1 Receptor Antagonist, Inhibits Apoptosis in Experimental Acute Myocardial Infarction Circulation, May 20, 2008; 117(20): 2670 - 2683. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. J.A. Frost and S. Engelhardt A Secretion Trap Screen in Yeast Identifies Protease Inhibitor 16 as a Novel Antihypertrophic Protein Secreted From the Heart Circulation, October 16, 2007; 116(16): 1768 - 1775. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||