Author + information
- Received May 19, 2003
- Revision received July 25, 2003
- Accepted September 8, 2003
- Published online March 3, 2004.
- Tiziano M. Scarabelli, MD, PhD*,* (, )
- Anastasis Stephanou, PhD‡,
- Evasio Pasini, MD§,
- Gianluca Gitti, BSc§,
- Paul Townsend, PhD‡,
- Kevin Lawrence, PhD‡,
- Carol Chen-Scarabelli, MSc∥,
- Louis Saravolatz, MD†,
- David Latchman, PhD, DSc‡,
- Richard Knight, MD, PhD¶ and
- Julius Gardin, MD*
- ↵*Reprint requests and correspondence:
Dr. Tiziano M. Scarabelli, Division of Cardiology, St. John Hospital and Medical Center, 22201 Moross Road, Detroit, Michigan 48236, USA.
Objectives This study is aimed at investigating the novel use of minocycline for cardiac protection during ischemia/reperfusion (I/R) injury, as well as its mechanism of action.
Background Minocycline is a tetracycline with anti-inflammatory properties, which is used clinically for the treatment of diseases such as urethritis and rheumatoid arthritis. Experimentally, minocycline has also been shown to be neuroprotective in animal models of cerebral ischemia and to delay progression and improve survival in mouse models of neurodegenerative diseases.
Methods We studied 62 rat intact hearts exposed to I/R and cell cultures of neonatal and adult rat ventricular myocytes.
Results Minocycline significantly reduced necrotic and apoptotic cell death, both in neonatal and adult myocytes, not only when given prior to hypoxia (p < 0.001), but also at reoxygenation (p < 0.05). Moreover, in the intact heart exposed to I/R, in vivo treatment with minocycline promoted hemodynamic recovery (p < 0.001) and cell survival, with reduction of infarct size (p < 0.001), cardiac release of creatine phosphokinase (p < 0.001), and apoptotic cell death (p < 0.001). In regard to its antiapoptotic mechanism of action, minocycline significantly reduced the expression level of initiator caspases, increased the ratio of XIAP to Smac/DIABLO at both the messenger RNA and protein level, and prevented mitochondrial release of cytochrome c and Smac/DIABLO (all, p < 0.05). These synergistic actions dramatically prevent the post-ischemic induction of caspase activity associated with cardiac I/R injury.
Conclusions Because of its safety record and multiple novel mechanisms of action, minocycline may be a valuable cardioprotective agent to ameliorate cardiac dysfunction and cell loss associated with I/R injury.
Minocycline is a semisynthetic second-generation tetracycline with proven safety, which is used in humans for the treatment of acne and urethritis (1). The drug may also be considered for treatment of severe chronic inflammatory diseases, such as rheumatoid arthritis, as it exerts anti-inflammatory effects that are completely separate and distinct from its antimicrobial action (2).
It has been shown that minocycline protects the brain in rodent models of global and focal cerebral ischemia (3,4). Significant neuroprotection has been attributed to decreased expression of caspase-1 and cyclooxygenase 2 (3,4)and to inhibition of the inducible form of nitric oxide synthase (iNOS) (5). Remarkable neuroprotection has also been observed in other experimental models of neurodegeneration. In a transgenic mouse model of Huntington's disease, for instance, minocycline delayed disease progression and prolonged survival both by inhibiting caspase-1 and caspase-3 messenger RNA up-regulation, and by decreasing the activity of iNOS (6). Although minocycline-mediated neuroprotection has been extensively reported, no corresponding data are yet available about the potentially beneficial effects of minocycline as a cardioprotective agent against ischemia/reperfusion (I/R) injury. Therefore, the present study was designed to evaluate the cardioprotective effectiveness and mechanism of action of minocycline during I/R injury both in primary cultures of myocytes and in the intact heart.
Primary cultures of myocytes
Ventricular myocytes isolated from neonatal Sprague-Dawley rats were cultured as described previously (6). Briefly, after collagenase digestion, cells were pre-plated in medium consisting of Dulbecco’s Modified Eagle Medium (DMEM), supplemented with 15% (v/v) fetal calf serum (FCS) on 10 cm tissue culture dishes to allow contaminating fibroblasts to attach and myocytes remain free within the culture media. Myocyte cell suspension was then transferred onto six-well gelatin-coated plates at a density of 105 cells per well. Cells were subsequently washed in 1 × phosphate buffered saline, and the medium was replaced by one containing reduced FCS at 1% (v/v), for an additional 24 h before experimentation. Within three days, a confluent monolayer of spontaneously beating myocytes was formed.
Primary adult rat cardiomyocytes were prepared from six-month-old female Sprague-Dawley rats, as previously described (7,8). Rats were sacrificed by cervical dislocation, hearts removed immediately, and immersed in buffer A with 5-U/ml heparin. The aorta was identified and mounted on the cannula of a Langendorff perfusion apparatus. Hearts were retrogradely perfused at 9 ml/min with buffer A containing: 750 μm CaCI2(for the first 2 min); 100 μm EGTA (for the following 4 min); and both 0.8 mg/ml collagenase type II and 200 μm CaCI2(for other 15 min). Hearts were then removed, sectioned, and placed in buffer A with 0.8 mg/ml collagenase, 10% BSA, and 200 μm CaCI2for 5 min. Cells were filtered through a 200 μm nylon gauze, resuspended in DMEM (containing 80 μm EGTA, 1% penicillin/streptomycin, and 1% fetal bovine serum), and incubated at 37°C with 5% CO2for 1 h on laminin-coated plates (10 to 15 μg/ml). After replacement of the media, cells were returned to the incubator and treated as required. All of the reagents mentioned in the Methods section, unless differently stated, were purchased from Sigma-Aldrich (United Kingdom).
Treatment of isolated myocytes
Myocytes were exposed to simulated ischemia by incubation (4 h) in an ischemic chamber and subsequently returned to a normoxic environment for 16 h to simulate reperfusion. Minocycline (0.02 μm) and tetracycline (0.04 μm) were added separately to cultured myocytes 0, 1, 2, 4, and 24 h before hypoxia or at reoxygenation.
Cell death assessment in isolated myocytes
Single-cell suspensions were incubated with annexin V and propidium iodide (PI) staining and immediately analyzed by flow cytometer as previously reported (9,10). Red fluorescent nuclei, due to PI uptake, were used as indicators of necrosis. Green annexin V staining, secondary to external exposure of phosphatidylserine on the plasma membrane, was employed as an early apoptotic marker.
Hearts from anesthetized male Sprague-Dawley rats were perfused by the non-recirculating Langendorff technique as previously described (11). Briefly, rats were anesthetized by sodium pentobarbital (6 mg/kg−1 administered intraperitoneally) and sacrificed by decapitation. The hearts were removed, immersed in an ice-cold modified Krebs-Hensleit buffer solution, and subsequently perfused by the non-recirculating Langendorff technique at a constant flow of 11 ml/min. The heart rate was continuously maintained at 300 beats/min by electrical pacing, and the left ventricular (LV) wall was kept at the steady temperature of 37°C.
Isolated hearts were randomly divided into eight groups (A, B, B1, C, D, D1, E, and E1) of at least six hearts each. Group A (control group) was aerobically perfused for 60 min. Groups B and B1 were aerobically perfused (60 min) with the addition to the perfusate of minocycline (1 μm) and tetracycline (2 μm), respectively. Group C, D, D1, E, and E1 were exposed to 30 min of regional ischemia and 2 h of reperfusion. Group D was given minocycline ex vivo, at the dose of 1 μm, 1 h before ischemia and during reperfusion. In group E, minocycline was administered in vivo, over a period of three days (45 mg/kg intraperitoneally twice daily for the first day; 22.5 mg/kg for the subsequent two days), before isolation of the hearts and their exposure to I/R. Similarly, group D1 was pretreated with tetracycline (2 μm) before ischemia (1 h) and in the course of reperfusion. Finally, in group E1, I/R was preceded by in vivo treatment with tetracycline, according to the following scheme: 90 mg/kg intraperitoneally twice a day for the first day and 45 mg/kg during the two successive days.
The left coronary artery of isolated hearts was surgically occluded for 30 min and subsequently reperfused for 60 min as previously described (11).
Infarct size measurement
Measurement of risk area and infarct size was performed by triphenyl-tetrazolium chloride staining as previously described (12). Triphenyl-tetrazolium chloride stains all living tissue brick red, leaving the infarcted area unstained (white). Left ventricular infarct zone was determined by computerized planimetry and expressed as the percentage of infarcted area within the myocardium at risk. Color enhancement was used to accentuate the differences between areas.
Left ventricular systolic pressure (LVSP) and left ventricular end-diastolic pressure (LVEDP) were recorded as previously described (13). Briefly, to obtain an isovolumetrically beating preparation, a latex balloon filled with saline, connected by a catheter to a Statham transducer (P 23 XL), was inserted into the LV 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.
Assay of creatine phosphokinase (CPK) in the coronary effluent
Creatine phosphokinase activity was evaluated by spectrophotometry, as previously reported (14).
Caspase activity assay
Cardiac activation of caspase-3, -7, -8, and -9 was evaluated in tissue extracts, as previously described (15).
TUNEL enhanced with additional staining
After terminal deoxynucleotidyl transferase mediated nick end labeling (Boehringer Mannheim, Lewes, Sussex, United Kingdom) (TUNEL) staining, a previously described multiple-step immunocytochemical procedure was used (16). Myocardial sections were labeled with either anti-desmin (Insight Biotechnology, Wembley, Middlesex, United Kingdom) or anti–von Willebrand (Boehringer Mannheim Biochemica, Lewes, Sussex, United Kingdom) factor antibodies, in order to selectively identify myocytes and endothelial cells, respectively. After incubation with specific secondary antibodies, the slides were counterstained with PI and finally examined by confocal fluorescent microscopy. Data are expressed as the means of 12 to 15 high-power fields ±SD.
Reverse transcription-polymerase chain reaction (RT-PCR)
The following specific primers were used to semiquantitate transcript levels: caspase-1, forward primer (FP), 5′-GACCTCAGAGAAATGAAGTTG-3′, reverse primer (RP), 5′-CACGGCATGCCTGAATAATG-3′; caspase-3, FP, 5′-GGTATTGAGACAGACAGTGG-3′, RP, 5′-CATGGGATCTGTTTCTTTGC-3′; caspase-7, FP, 5′-GATAAAGGATCTGACAGCTC-3′; RP, 5′-CATGGACACCATACATGGAATC-3′; caspase-8, FP, 5′-GAGCTGACATCTTACTTCAC-3′, RP, 5′-GAAGATGGGCTGTGGCATC-3′; caspase-9, FP, 5′-GTACATCGAGACCTTGGATG-3′, RP, 5′-GACAGGATTACACAACCTCATG-3′; caspase-12, FP, 5′-CAGAAGTACAGGATTCACTG-3′, RP, 5′-CATTCCTCATCTGTATCAGC-3′; DIABLO, FP, 5′-GCATGACACTGTGTGCGGTTC-3′, RP, 5′-CCAACTGGATGTGATTCCTG-3′; XIAP, FP, 5′-CTACCTCTGGAACAAGGTGG-3′, RP, 5′-CAAGCTGCTCAGGCTGAAC-3′; and rat cyclophilin, FP, 5′-CGAGCTGTTTGCAGACAAAG-3′, RP, 5′-TTCTTGCTGGTCTTGCCATT-3′.
Western blot analysis was performed following standard protocols. Anti-procaspase-1, -3, -7, -8, -9, -12, anti-actin, anti-XIAP, and anti-Smac/DIABLO antibodies were obtained from Santa Cruz Biotechnology, Inc.
Mitochondrial and cytoplasmic protein fractionation, characterization, and assessment
Mitochondrial and cytoplasmic fractions from cardiac ventricular tissue were isolated as previously described (17). Anti-cytochrome C and anti-HSP60 antibodies were purchased from Abcam (Cambridge-Science-Park, United Kingdom).
Data are expressed as means ± SD, and the significance level overall was set at p < 0.05. Experiments were repeated at least three times. Single-factor one way analysis of variance was performed for each group of treatments. Differences between treatment groups were evaluated using Student ttest, and the Bonferroni correction was applied for multiple comparisons. The resulting significance level was p < 0.008 (0.05/6) for cell death comparisons, and p < 0.0125 (0.05/4) for infarct risk ratio, TUNEL, and active caspase-3%. Analysis of covariance, with time as covariate and post hoc contrast testing, was used to evaluate the hemodynamic variables.
Minocycline is cardioprotective even when given at reoxygenation
To investigate whether minocycline plays a role in cardioprotection, we first assessed cell death in primary cultures of neonatal and adult cardiomyocytes treated with minocycline. Minocycline significantly reduced necrotic and apoptotic cell death, both in neonatal and adult myocytes, not only when given before hypoxia, but also, more importantly, when given at reoxygenation. In neonatal myocytes, the cardioprotective effect peaked when minocycline was administered 2 h before hypoxia (Fig. 1A). Necrotic cell death decreased from 35.2 ± 2.8% to 24.5 ± 1.7% (p < 0.001) and apoptotic death from 28.3 ± 2% to 18.1 ± 1.7% (p < 0.001). In adult myocytes, the greatest cardioprotection was achieved when minocycline was given 4 h before hypoxia (Fig. 1B). The magnitude of necrosis (38.1 ± 2.7%) and apoptosis (29.5 ± 2.3%) in untreated myocytes diminished to 22.7 ± 1.9% and 20.4 ± 1.6%, respectively (p < 0.001). Post-hypoxic treatment with minocycline prevented to a lesser, though still statistically significant, extent, the degree of total cell death both in neonatal and adult myocytes (p < 0.05) (Figs. 1A and 1B). In contrast, no cardioprotection was observed after addition of tetracycline at any time point (data not shown).
Minocycline promotes post-ischemic recovery of cardiac function independently of any direct hemodynamic effects
To further validate these in vitro findings, we investigated a potential cardioprotective effect of minocycline in the isolated rat heart. First, we assessed whether minocycline produced any hemodynamic changes in the isolated heart in the absence of any ischemic insult. Addition to the perfusate of minocycline (concentration range: 10−5to 10−7M) had no effect on either LVSP (Fig. 2A) or LVEDP (Fig. 2B). Similarly, ex vivo infusion of increasing doses of tetracycline had no effect on both LVSP and LVEDP (data not shown).
We then evaluated the hemodynamic effects induced by minocycline in rat hearts exposed to 30 min of regional ischemia followed by 2 h of reperfusion. Minocycline was either infused ex vivo (dose: 1 μm) 1 h before ischemia and during reperfusion, or administered in vivo, over a period of three days (45 mg/kg intraperitoneally twice a day for the first day; 22.5 mg/kg for the subsequent two days), before isolation of the hearts and their exposure to I/R. Consistent with previous reports using the same experimental model of I/R injury (13), ischemic/reperfused control hearts showed a progressive rise in LVEDP and a rapid decline in LVSP, which recovered poorly during reperfusion (Fig. 2C). The pre- and post-ischemic infusion of minocycline significantly reduced the progressive rise of LVEDP observed during I/R. This recovery in LVEDP started after 10 min of ischemia (29.3 ± 3.2 mm Hg after 20 min ischemia; p < 0.025 vs. I/R control) and progressively improved throughout reperfusion. Because no concurrent enhancement in LVEDP was observed, the post-ischemic recovery of function in the hearts infused pre- and post-ischemia with minocycline was only modest, though statistically significant (p < 0.025 vs. control) (Fig. 2D). In contrast, in vivo treatment with minocycline, extending over three days, induced not only a greater and earlier recovery of LVEDP during I/R, but also a rapid recovery of LVSP, which became significant after 5 min of reperfusion (72 ± 5.2 mm Hg; p < 0.025 vs. I/R control), and which progressively improved with partial normalization (86.0 ± 5.9 mm Hg; p < 0.001 vs. I/R control) at the end of reperfusion (Fig. 2E). Hence, the in vivo pretreatment with minocycline consistently prevented the functional impairment in cardiac function associated with I/R injury (p < 0.001 vs. I/R control). In contrast, both ex vivo and in vivo administration of tetracycline had no effect on the hemodynamics of the isolated rat heart (data not shown).
Minocycline minimizes infarct size and cardiac release of CPK after I/R
To address whether this amelioration in cardiac performance is due to enhanced cell survival, we evaluated the extent of myocardial infarction, cardiac release of CPK, and apoptotic cell death in the isolated rat heart pretreated with minocycline and subsequently exposed to regional ischemia. In control hearts exposed to I/R, the percentage of infarction within the risk zone was 35 ± 1.9% (Fig. 3A). Following pre- and post-ischemic infusion of minocycline, the cardiac release of CPK (905 ± 44 mU/min/GWW; p < 0.05 vs. I/R control), though not the percentage of infarction (p > 0.0125), was significantly reduced (Figs. 3A and 3B). Conversely, the in vivo administration of minocycline over a three-day period resulted in a marked limitation of infarct size (p < 0.001 vs. I/R control) (Figs. 3A, 3C to 3F), which was also associated with a strong attenuation of CPK release in the coronary effluent (737 ± 36 mU/min/GWW after 30 min reperfusion; p < 0.001 vs. I/R control) (Fig. 3B). In contrast, both ex vivo and in vivo treatment with tetracycline did not result in any significant reduction in infarct size and CPK release (Figs. 3A and 3B).
Minocycline prevents both endothelial and myocyte apoptosis in the intact heart exposed to ischemia/reperfusion
In agreement with our previous studies, TUNEL-positive staining was detected only during reperfusion (11,13,15,16)and always colocalized with caspase-3-positive labeling (13,16). Additionally, the magnitude of apoptotic cell death was much higher in endothelial cells than in cardiac myocytes (15,16). The three-day in vivo treatment with minocycline produced a highly significant reduction in these apoptotic markers in both cell types (p < 0.001 vs. I/R control) (Figs. 3G, 3H, and 4). ⇓Although less profound, the lessening in apoptosis remained significant when minocycline was given by infusion before and after ischemia (p < 0.001 vs. I/R control). Reduction of TUNEL- and caspase-3-positive staining was not observed after either ex vivo nor in vivo treatment with tetracycline (Figs. 3G and 3H). Hence, the modest, though significant, post-ischemic recovery of function observed after ex vivo infusion of minocycline was associated with a significant decrease in cardiac release of CPK and a non-significant limitation of infarct size. In contrast, after in vivo treatment with minocycline, the post-ischemic recovery of cardiac function was greatly enhanced and paralleled a marked reduction of both infarct size and myocardial release of CPK.
Minocycline induces cardiac downregulation of both initiator and effector caspases
By RT-PCR and Western blot analysis, ischemic/reperfused control hearts showed a threefold or more transcriptional and translational up-regulation of caspase-1, -3, -7, -8, -9, and -12 (Figs. 5A and 5B). In hearts given three-day treatment with minocycline before exposure to I/R, the messenger RNA and protein induction of all the above caspases was dramatically reduced. Similarly, in control hearts unexposed to I/R receiving minocycline intraperitoneally for three days, the basal messenger RNA and protein expression level of caspases-1, -3, -7, -8, -9, and -12 was greatly downregulated compared with control hearts unexposed to I/R.
In agreement with these findings, in control hearts exposed to I/R, the activity level of initiator (caspase-8 and -9) and effector caspases (caspase-3 and -7) was increased compared with control hearts. Hearts subjected to three-day in vivo treatment with minocycline showed a greatly reduced level of functional activation of caspase-3, -7, -8, and -9 after I/R compared with ischemic/reperfused control hearts (p < 0.001 vs. I/R control) (Fig. 5C). Therefore, in addition to the downregulation of effector caspase expression previously reported in the brain (3,4,6), this study shows for the first time that minocycline downregulates not only the expression, but also the activity, of both initiator and effector caspases in a different target organ.
Effect of minocycline on XIAP and the mitochondrial proteins, cytochrome C and Smac/DIABLO
One stimulus by which cells are induced to undergo apoptosis is mitochondrial damage with release of cytochrome c and activation of caspase-9. The cytoplasmic relocation of cytochrome c seen in ischemic control hearts was remarkably reduced in rat hearts exposed to I/R after in vivo treatment with minocycline, although the total mitochondrial levels of cytochrome c were unaffected (Fig. 5D). Caspase activation is also dependent on the ratio between the inhibitors of apoptosis and another mitochondrial protein, Smac/DIABLO. In untreated perfused hearts exposed to I/R, expression of XIAP was reduced, together with a marked induction of Smac/DIABLO at both transcript and protein levels (Fig. 5D). In contrast, administration of minocycline for three days before I/R induced enhanced expression of XIAP with concurrent downregulation of Smac/DIABLO expression. Strong induction of XIAP, with reduced Smac/DIABLO expression, was also seen in the hearts of minocycline-treated animals that were not subjected to I/R (Fig. 5D). Consistent with these findings, the levels of Smac/DIABLO in the mitochondria of hearts from minocycline-treated rats was also reduced. Moreover, minocycline treatment reduced translocation of Smac/DIABLO from mitochondria into the cytosol, similar to its effects on cytochrome c translocation. Therefore, the protective effects of minocycline are mediated by a number of processes. This antibiotic diminishes the level and activity of caspases; it reduces mitochondrial leakage of cytochrome c after I/R, thereby reducing activation of the apical caspase, caspase-9; and, finally, it increases the cytoplasmic ratio of XIAP to Smac/DIABLO, both by altering the expression of these proteins and by minimizing mitochondrial leakage of Smac/DIABLO.
Here, we show for the first time that minocycline induces protection against I/R injury in the heart and substantiates our findings by providing new insights into its mechanism of action. Our study demonstrates that minocycline effectively protects cardiac myocytes against I/R injury, inducing a marked reduction in both necrotic and apoptotic cell death. This cardioprotective effect has been validated at three levels: in vitro, using primary cultures of neonatal and adult cardiomyocytes; ex vivo, infusing minocycline to the isolated rat heart; and in vivo, injecting the animals with minocycline over a period of three days. Importantly, reduction of infarct size and apoptotic cell death observed after in vivo treatment with minocycline was associated with a remarkable post-ischemic recovery of cardiac function. However, because our findings refer solely to the use of in vitro models of ischemia/reperfusion, data supporting similar effects of minocycline on chronic infarct size in the blood-perfused heart subjected to regional ischemia are also required.
There are many determinants of cellular fate after stressors such as I/R injury, including the balance between intrinsic death and survival factors within the apoptotic program. Although we have identified some mechanisms of minocycline-mediated cardioprotection, these are not necessarily the only, or even the most important, effects. For example, modulation of the production of free radicals, and/or attenuation of their damaging effects, may be a more fundamental property of the drug, and this possibility is currently under investigation.
The cardioprotective properties of minocycline are not shared by its ancestral antibiotic, tetracycline. The newer tetracyclines have a variety of substituted groups at various positions in the basic four-ring structure, which have the effect of broadening their antibiotic spectrum, at least partially as a result of increasing their solubility in the lipid membranes of bacteria. Minocycline, uniquely, has an N(CH3)2 group at position 7, and this may confer an ability to protect cardiomyocytes and neurons and enhance its anti-bacterial repertoire.
Besides the known downregulation of caspase-1 and -3, we show that minocycline reduces the cardiac expression of caspase-7, -8, -9, and -12 under basal conditions, and prevents the post-ischemic up-regulation of expression of all these caspases. Minocycline also interferes with upstream and downstream mechanisms leading to secondary caspase activation and reactivation. After in vivo treatment with minocycline, we report reduced mitochondrial leakage of cytochrome c and Smac/DIABLO, together with an increased ratio of XIAP to Smac/DIABLO. Therefore, the effects achieved with in vivo administration of minocycline effectively cooperate to restrict the level of caspase activity in the heart. Although rats given minocycline at doses as high as 100 to 125 mg/kg/day showed no side effects (18), the in vivo doses of minocycline we used to achieve cardioprotection (90 mg/kg for the first day, and 45 mg/kg during the following two days) are undoubtedly high. These dosages, which in humans would correspond to approximately 1.5 g/day instead of the conventional 200 mg/day, may compromise long-term utilization of the antibiotic as a cardioprotective agent. Although this issue will have to be formally investigated, previous reports have shown, however, that daily tetracycline doses even higher than 2 g are still relatively safe in healthy nonpregnant women (19). In conclusion, we believe that minocycline could be valuable in both acute and possibly chronic clinical settings, where it may augment conventional cardioprotective agents in counteracting the occurrence and progression of myocyte cell loss. Additionally, provided that long-term treatment and safety issues can be successfully resolved, minocycline, by virtue of its oral bioavailability, may also be useful in delaying the progression of chronic cardiovascular conditions, such as heart failure and cardiomyopathies, where progressive cardiac cell loss is the ultimate mechanism accounting for cardiac dysfunction.
The authors are deeply grateful to Kathleen Steiner for her valuable support in the preparation of this manuscript.
- creatine phosphokinase
- forward primer
- inducible form of nitric oxide synthase
- left ventricle/ventricular
- left ventricular end-diastolic pressure
- left ventricular systolic pressure
- propidium iodide
- reverse primer
- reverse transcription-polymerase chain reaction
- Received May 19, 2003.
- Revision received July 25, 2003.
- Accepted September 8, 2003.
- American College of Cardiology Foundation
- Aronson A.L.
- Yrjanheikki J.,
- Tikka T.,
- Goldsteins G.,
- et al.
- Yrjanheikki J.,
- Keinanen R.,
- Pellikka M.,
- et al.
- Simpson P.,
- Savion S.
- Koopman G.,
- Reutlingsperger C.,
- Kuijten G.,
- et al.
- Walker D.,
- Pasini E.,
- Kucukoglu S.,
- et al.
- Scarabelli T.M.,
- Pasini E.,
- Stephanou A.,
- et al.
- Oliver T.A.
- Scarabelli T.M.,
- Stephanou A.,
- Pasini E.,
- et al.
- Scarabelli T.,
- Stephanou A.,
- Rayment N.,
- et al.
- Watmough N.J.,
- Bhuiyan A.K.,
- Bartlett K.,
- et al.