Author + information
- Received September 11, 2001
- Revision received March 27, 2002
- Accepted April 5, 2002
- Published online July 3, 2002.
- Tiziano M Scarabelli, MD*,* (, )
- Evasio Pasini, MD†,
- Anastasis Stephanou, PhD*,
- Laura Comini, MSc†,
- Salvatore Curello, MD‡,
- Riccardo Raddino, MD‡,
- Roberto Ferrari, MD, PhD†,
- Richard Knight, MD, PhD§ and
- David S Latchman, PhD, DSc*
- ↵*Reprint requests and correspondence:
Dr. Tiziano M. Scarabelli, Medical Molecular Biology Unit, Institute of Child Health and Great Ormond Street Hospital, University College London, London, WC1H 9DQ, United Kingdom.
Objectives This study evaluates the hemodynamic, bioenergetic and cytoprotective effects of urocortin (Ucn) in the isolated rat heart exposed to ischemia (I)/reperfusion (R).
Background We have previously demonstrated that administration of exogenous Ucn reduces infarct size in ischemic-reperfused rat hearts.
Methods Urocortin 10−8M was added to the perfusate before I, before I and during R, and during R alone in the isolated pulsed rat heart exposed to 35 min I followed by 60 min R.
Results Partial to complete recovery of diastolic pressure and developed pressure was seen irrespective of when Ucn was perfused. In particular, beneficial effects are observed when Ucn is only given during R. Urocortin given only before I, and before I and over R, although not during R alone, also produces significant recovery of high-energy phosphate pools. In each group, improvement in ventricular function is associated with reduction both in myocardial damage, assessed by creatine phosphokinase release, and in endothelial cell and cardiomyocyte apoptosis, assessed by caspase 3 activity and fluorescent-based terminal deoxynucleotidyl transferase mediated nick end labelling enhanced with counterstains. These improvements in ventricular performance, bioenergetics and cell survival are not secondary to any inotropic effects of Ucn.
Conclusions This is the first report to show enhanced cardiac function induced by Ucn during I/R. Because the cytoprotective and functional benefits are still produced when Ucn is given only at R, these data suggest that Ucn may be useful clinically in the management of myocardial infarction.
Myocardial ischemia/reperfusion (I/R) triggers a cascade of key pathogenetic cellular events such as massive influx of calcium, mitochondrial damage and rapid depletion of high energy stores, leading to both necrotic and apoptotic cell death. In addition, the mechanical efficiency of the heart is compromised, with a fall in developed pressure (DP) and progressive increases of diastolic pressure (dP) throughout I and R. The degree of myocyte death and the severity of the hemodynamic consequences are major determinants of the clinical outcome of I/R, and therapies for myocardial infarction must both improve cell survival and promote mechanical recovery.
Urocortin (Ucn) is a 40 amino acid member of the corticotropin-releasing hormone family, initially characterized in rat brain (1). Subsequently, Ucn expression has also been described in a number of other sites including the placenta (2), the immune system (3), the gastrointestinal tract (4)and the cardiovascular system (5). Urocortin and other members of the family exert their physiologic effects through binding to two G-protein coupled receptors (corticotropin-releasing hormone-R1 and -R2), both of which can be expressed in alternatively spliced forms. Corticotropin-releasing hormone-R2 has at least 10-fold higher affinity for Ucn than for corticotropin-releasing hormone and is the only such receptor found in the heart (6,7). This coexpression in the heart of corticotropin-releasing hormone-R2 together with its preferred Ucn ligand suggests that cardiac Ucn may exert autocrine/paracrine physiologic effects on the heart.
Several studies have, indeed, shown effects of exogenously administered corticotropin-releasing hormone-like peptides on cardiac function, although these may depend on the route of administration. For example, heart rate, cardiac output and mean arterial pressure are increased when corticotropin-releasing hormone is given into the cerebral ventricles (8–10), although mean arterial pressure is decreased by intravenous injection (11,12). In contrast, intravenous injection of Ucn into conscious sheep increases mean arterial pressure, together with heart rate, cardiac output and coronary blood flow (13). However, because both corticotropin-releasing hormone (14)and Ucn (15)have been shown to stimulate atrial natriuretic peptide release from cultured neonatal rat cardiomyocyte (CM), some of their physiologic effects may be indirect. In contrast, Ucn perfused through the isolated rat heart has a coronary vasodilator effect and enhances left ventricular pressure (16), and this effect must be atrial natriuretic peptide-independent. In agreement with the presumed role of corticotropin-releasing hormone-R2 in mediating the effects of Ucn on the heart, intravenous Ucn does not affect cardiac function in corticotropin-releasing hormone-R2 knockout mice (17).
We have previously reported that exposure of cultured neonatal CM to I/R increased Ucn messenger ribonucleic acid abundance and release of Ucn peptide (18). The addition of exogenous Ucn reduced necrotic and apoptotic myocyte death after simulated I/R in cultured CM, and also reduced infarct size in the Langendorff perfused heart exposed to I/R, even when addition of the peptide was delayed until the onset of R (19). In the present study, we have examined these cytoprotective effects in more detail and asked whether the beneficial effects of Ucn on myocyte survival are associated with an equally marked improvement in functional recovery. The isolated heart has been preferred as it allows assessment of the direct cardioprotective effects of Ucn in the absence of interfering peripheral hemodynamic and neurohumoral alterations. We demonstrate that Ucn exerts a cytoprotective action that is independent from a negative inotropic effect and is associated with both recovery of cardiac performance and reduced depletion of endogenous high-energy phosphates.
Male Sprague-Dawley rats weighing 300 g to 350 g were anesthetized by sodium pentobarbital (6 mg kg−1administered intraperitoneally) and sacrificed by decapitation. The heart were removed, immersed in an ice-cold modified Krebs-Hensleit buffer solution (BS) and subsequently perfused by the nonrecirculating Langendorff technique at a constant flow of 11 ml/min as previously described (20). The heart rate was continuously maintained at 300 beats/min by electrical pacing, and the left ventricular wall was kept at the steady temperature of 37°C. After a period of stabilization of at least 30 min, the isolated hearts were randomly divided into six groups of six hearts each. Group A (I/R control) and three treated groups (B, C and D) were continuously perfused for 60 min, made globally ischemic for 35 min and reperfused for 60 min. Treated hearts were then perfused with Ucn dissolved in the perfusate at the dose of 10−8M either 60 min before I alone (group B), 60 min before I and during R (group C) or during R alone (group D). Finally, two other groups (E and F) were simply perfused for 2 h with either BS or Ucn (10−8M).
Left ventricular pressure
To obtain an isovolumetrically beating preparation, a latex balloon filled with saline, connected by a catheter to a Statham transducer (P 23 XL, VWR International, Milan, Italy), was inserted into the left ventricle through an atriotomy and secured by a suture around the atrioventricular groove. The balloon was inflated to provide an end-diastolic pressure <1.0 mm Hg (20).
Assay of creatine phosphokinase in the coronary effluent
During each perfusion, the coronary effluent was collected at different time points in chilled glass vials and promptly assayed for creatine phosphokinase (CPK) activity by spectrophotometry, as previously reported (21).
Assay of high-energy phosphates
After each perfusion, the hearts were freeze-clamped with aluminum tongues precooled in liquid nitrogen. Separation and quantification of adenosine triphosphate (ATP) and creatine phosphate were performed in ventricular tissue extracts by using a reversed-phase 3-μm C18 column, as previously described (22).
Caspase 3 (C3) activity measurement (aspartyl glutamylvalylaspartic acid cleavage enzyme assay)
Cardiac activation of C3 was evaluated in tissue extracts using a commercial kit (caspase-3/CPP32 Fluorometric Assay Kit, Biovision, Mountain View, California) with the following changes to the recommended protocol. Cardiac ventricular tissue from each group was placed in ice-cold lysis buffer and subsequently homogenized. The homogenates were centrifuged at 750 × gfor 5 min at 4°C. Supernatants were then centrifuged at 10,000 × gfor 15 min at 4°C. Enzyme reactions were performed with ∼300 μg of cytosolic proteins per assay and a final concentration of 50 μM aspartyl glutamylvalylaspartic acid–7-amino-4-trifluoromethyl courmarin. Samples were read in a fluorimeter equipped with a 400-nm excitation and a 505-nm emission filter. Fold-increase in C3 activity was determined by comparing fluorescence of 7-amino-4-trifluoromethyl courmarin in control and treated hearts with BS perfused control.
Preparation and staining of sections
Serial 5 μm sections were cut from paraffin blocks and, after dewaxing and heat-mediated antigen retrieval, stained with terminal deoxynucleotidyl transferase mediated nick end labeling (TUNEL) reagents and propidium iodide. In other sections, TUNEL staining was combined with anti-desmin or von Willebrand factor antibodies to selectively identify CM and endothelial cells (EC), respectively (23,24). After washing, slides were mounted and examined by confocal fluorescent microscopy as described before.
Data are expressed as the means of 12 to 15 high power fields ± SD.
Analysis of covariance, with time as the covariate and post-hoc analyses were used to test the principal component with contrast. The Bonferroni correction was then applied and p values <0.05 were considered significant.
Developed pressure and dP in control and treated hearts exposed to I/R are shown in Figure 1. Untreated hearts showed a constant DP (103.3 ± 9.8 mm Hg) with no change in dP during stabilization and aerobic perfusion (Fig. 1a). Abolition of the coronary flow (CF) caused a rapid decline in DP. Diastolic pressure started to rise 10 min after I (27.7 ± 8.2 mm Hg) and progressively increased during the ischemic phase, reaching 30.7 ± 5.3 mm Hg at the end of I. With restoration of CF, dP was further augmented. After 5 min R, it reached a peak of 55.7 ± 12.1 mm Hg and then slowly decreased to 51.0 ± 11.3 mm Hg at the completion of R. In addition, the I/R control group showed only limited recovery of DP, which increased to 64.5 ± 8.3 mm Hg by the end of R.
Administration of Ucn during aerobic perfusion had no effect on the hemodynamics of isolated hearts, apart from a transient, nonsignificant increase in DP during the first phase of infusion (p > 0.05 vs. I/R control, Fig. 1b). However, in the hearts treated with Ucn before I alone (Fig. 1b) as well as before I and during R (Fig. 1c), Ucn significantly reduced the progressive rise of dP observed during I (14.3 ± 3.2 mm Hg and 13.5 ± 3.2 mm Hg in groups B and C, respectively, after 30 min I, p < 0.05 vs. I/R control), and, in both groups, recovery of dP after R was complete. In the same groups (B and C,Figs. 1b and 1c), Ucn produced a rapid recovery of DP, which began after only 1 min of R (44.0 ± 6.9 mm Hg and 52.8 ± 7.2 mm Hg, respectively; p < 0.05 vs. I/R control), and which progressively improved with partial (group B) and complete (group C) normalization (86.0 ± 7.9 mm Hg and 110.4 ± 12.8 mm Hg, respectively; p < 0.001 vs. I/R control) at the end of R. Isolated hearts from group D, treated with Ucn only during R (Fig. 1d), also showed a significantly improved post ischemic DP recovery (82.3 ± 8.3 mm Hg; p < 0.001 at the end of R) with progressively diminishing dP values over the length of R (29.1 ± 5.3 at 10 min; p < 0.05 vs. I/R control and 25.3 ± 4.1 at 60 min: p < 0.001 vs. I/R control).
Hence, Ucn has no effect on DP and dP in control perfused hearts, but ameliorates the fall in DP and the rise in dP observed during I/R, when given before the onset of the ischemic insult. Pre- and postischemic treatment with Ucn produces complete recovery of dP and DP, while preischemic treatment only results in normalization of dP and partial restoration of DP, during R. Progressive but partial recovery of both dP and DP is also seen when Ucn is only given after restoration of the flow.
Figure 2ashows the effects of Ucn on R-induced release of CPK, a typical index of cell damage. During the aerobic preischemic period, only small amounts of CPK were released. Urocortin did not affect the release curve of the enzyme. In control hearts exposed to I/R, R resulted in marked and sustained release of CPK, which peaked at 30 min from the restoration of flow (1,785 ± 88 at 30 min R). Urocortin administered both 60 min before I alone, and before I and during R significantly attenuated the cardiac CPK release (620 ± 46 and 510 ± 43 after 30 min R, respectively; p < 0.001 vs. I/R control). Consistent with the hemodynamic findings, the release of CPK into the CF was reduced even when Ucn treatment was performed only during R (1,005 ± 56; p < 0.05 vs. I/R control). Therefore, Ucn given before I alone, before I and during R, and during R alone significantly reduces CPK release during the reperfusion phase.
ATP and creatine phosphate tissue reserves
The cardiac levels of ATP and creatine phosphate assessed after 2 h of aerobic BS perfusion are reported in Figure 2b. The infusion of Ucn over the aerobic phase did not modify the reservoir of both high energy molecules. In ischemic-reperfused control hearts, the endogenous stores of ATP and creatine phosphate evaluated at the end of I showed a dramatic decline (5.4 ± 0.9 and 4.95 ± 0.7, respectively), which R failed to normalize (6.6 ± 1.3 and 6.3 ± 1.1, respectively). Preischemic infusion of Ucn reduced the tissue drop of ATP and creatine phosphate levels measured both at the end of I (12.5 ± 2.2 and 11.6 ± 1.8, respectively; p < 0.05 vs. I/R control) and R (13.4 ± 2.4 and 18.7 ± 3.5, respectively; p < 0.001 vs. I/R control). Adenosine triphosphate and creatine phosphate contents were maximally preserved in hearts receiving Ucn both before I and after R (16.2 ± 2.8 and 25.5 ± 4.3, respectively; p < 0.0001 vs. I/R control). The administration of Ucn only during R resulted in a small recovery of tissue ATP and creatine phosphate contents, which failed to reach reach statistical significance (7.9 ± 1.8 and 7.1 ± 1.5, respectively; p < 0.05). Hence, Ucn administered before I alone and before I and during R partially restore ATP and creatine phosphate tissue levels both after I and R. Urocortin only given during R has no significant effect on tissue reserves of both high energy molecules measured after R.
Evaluation of C3 activity in tissue extracts
Apoptosis is one form of cell death after I/R injury and is mediated through sequential activation of the caspase cascade. Caspase 3 is one of the terminal effector caspases, which cleaves substrates important for cell survival, and the level of C3 activity is, therefore, an important marker of the level of apoptosis. We, therefore, used an enzymatic assay to measure the activity of C3 at the end of R in each of the treatment groups (Fig. 2c). We also employed an antibody specific for active, cleaved C3 to evaluate the total number of EC and CM, which contained active C3 (Fig. 2c). The number of positive cells is expressed as the mean of 15 neighboring high power fields.
Caspase 3 activity, and the numbers of active C3 positive cells, were maximally reduced when Ucn was given pre-I and before I and during R (p < 0.001 vs. BS perfused control). A significant (p < 0.05) reduction in both enzyme activity and active C3-positive cells was still observed even when Ucn was administered only during R.
Assessment of apoptosis by TUNEL
Terminal deoxynucleotidyl transferase mediated nick end labelling is a commonly used method for detecting the oligonucleosomal deoxyribonucleic acid fragmentation characteristic of end-stage apoptotic cells. We have previously shown that cells positive for active C3 colocalize with TUNEL-positive cells (24).Figure 2dshows the numbers of active C3 and TUNEL-positive EC and CM in the different treatment groups.
In agreement with our previous study (24), more EC than CM show evidence of apoptosis after I/R as assessed by both techniques. Preischemic treatment with Ucn produced a highly significant reduction in these markers of apoptotic cell death in both cell types (groups D and E: p < 0.001 vs. I/R control). Though less profound, the reduction in apoptosis when Ucn was only given during R remained significant (group E: p < 0.001 vs. I/R control).
This study shows that administration of the endogenous cardiac peptide, Ucn, at a dose of 10−8M, reduces necrotic and apoptotic cell death in the isolated rat heart exposed to I/R and, in parallel, partially prevents depletion of cellular energy stores with enhancement of ventricular function. These beneficial effects, apart from the preservation of myocardial high energy reservoirs, are still seen when Ucn perfusion is performed only during R, suggesting that Ucn may have therapeutic potential in the management of myocardial infarction.
In our study, we have intentionally used the isolated paced rat heart exposed to zero flow ischemia in order to assess the direct effects of Ucn on the heart without the influence of secondary peripheral effects of the peptide. Developed pressure and dP are commonly used methods for assessing cardiac dysfunction after I/R injury (25)and for evaluating the efficacy of candidate cardioprotective agents (26). At the 10−8M concentration used, Ucn did not show a significant inotropic effect when given before I during the aerobic period. However, when given pre-I, pre-I and during R and during R alone, Ucn produced significant functional recovery.
Effect of Ucn on cell death
The degree of cardiac dysfunction after I/R injury reflects the level of myocyte injury and death. We have previously shown, in isolated rat hearts exposed to 35 min I, that TUNEL-positive EC appear after 5 min and TUNEL-positive CM within 1 h of R. Consistent with our previous findings that Ucn given both before and after a simulated ischemic insult in the intact rat heart reduced infarct size (19), in the present study Ucn reduced both CPK release and the numbers of apoptotic cells, even when only administered during R. Urocortin administration significantly reduces activation of C3, as well as oligonucleosomal deoxyribonucleic acid fragmentation, in both EC and CM.
Mechanisms of Ucn-mediated cardioprotection
Cell death after injury, and the form that it takes, is largely dependent on levels of intracellular ATP and other high energy phosphates (27). For example, viable cells have an ADP:ATP ratio of <0.11, apoptotic cells a ratio of between 0.11 and 1.0 and necrotic cells a ratio of up to 15 (28). The Ucn-induced recovery of ATP stores, with reduction of the intracellular ADP:ATP ratio, might allow damaged myocytes that would otherwise die by necrosis to die by the alternative apoptotic pathway. Because necrosis, unlike apoptosis, is associated with release of intracellular contents and subsequent inflammatory reaction, reduction in the proportion of necrotic death will result in a smaller final lesion, with functional benefit. The reduction in CPK release produced by Ucn treatment (Fig. 2a) would be consistent with this interpretation. Similarly, restoration of ATP in damaged cells predisposed to apoptosis may allow them to remain viable, again consistent with the data in Figures 2c and 2d.
One critical role for ATP is in the cytochrome C-mediated formation of Apaf-1/procaspase-9 complexes, which results in the activation of the initiator caspase-9 after mitochondrial injury (29), and we have previously shown that caspase-9 is the only initiator caspase activated in hypoxic neonatal cardiac myocytes (30). The partial restoration of ATP and creatine phosphate levels at the end of both I and R suggests that Ucn acts on mitochondria to maintain the respiratory transport chain, thus preventing mitochondrial injury. One mechanism for these mitochondrial effects of Ucn may be its ability to increase expression and function of the mitochondrial KATPchannel (Lawrence et al., unpublished data, 2002), which has been implicated in cardioprotection (31).
The data reported here have several important implications. First, because Ucn is an endogenous cardiac peptide, basal levels of Ucn may determine the degree of cell loss and functional compromise in individual patients suffering from myocardial infarction. Second, because endogenous levels of Ucn are increased by I/R injury (5), raised basal levels of Ucn may play a role in the preconditioning effect. Finally, the fact that Ucn improves both cell survival and ventricular performance even when only given during R suggests that Ucn may be clinically useful in the management of established infarction.
☆ Supported by British Heart Foundation. Dr. Scarabelli is supported by a University College London Graduate School Research Studentship. Dr. Stephanou is a British Heart Foundation Intermediate Fellow.
- adenosine triphosphate
- buffer solution
- coronary flow
- creatine phosphokinase
- caspase 3
- diastolic pressure
- developed pressure
- endothelial cell
- terminal deoxynucleotidyl transferase mediated nick end labelling
- Received September 11, 2001.
- Revision received March 27, 2002.
- Accepted April 5, 2002.
- American College of Cardiology Foundation
- Chen R.,
- Lewis K.A.,
- Perrin M.H.,
- et al.
- Lovenberg T.W.,
- Oltersdorf T.
- Grosskreutz C.,
- Brody M.
- Parkes D.,
- Vaughan J.,
- Rivier J.,
- et al.
- Ikeda K.,
- Tojo K.,
- Sato S.,
- et al.
- Terui K.,
- Higashiyama A.,
- Horiba N.,
- et al.
- Brar B.K.,
- Jonassen A.K.,
- Stephanou A.,
- et al.
- Ferrari R.,
- Cargnoni A.,
- Bernocchi P.,
- et al.
- Oliver T.A.
- Bernocchi P.,
- Ceconi C.,
- Cargnoni A.,
- et al.
- Scarabelli T.M.,
- Stephanou A.,
- Rayment N.B.,
- et al.
- Inagaki K.,
- Kihara Y.,
- Hayashida W.,
- et al.
- Leist M.,
- Single B.,
- Castoldi A.F.,
- et al.
- Grover G.J.