Author + information
- Received May 6, 2008
- Revision received June 19, 2008
- Accepted June 20, 2008
- Published online September 23, 2008.
- Diego Romero-Perez, BS,
- Eduardo Fricovsky, PharmD,
- Katrina Go Yamasaki, BS,
- Michael Griffin, PhD,
- Maraliz Barraza-Hidalgo, BS,
- Wolfgang Dillmann, MD and
- Francisco Villarreal, MD, PhD⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Francisco Villarreal, UCSD Cardiology, 9500 Gilman Drive, 0613J, BSB 4028, La Jolla, California 92093
Objectives The ability of minocycline to be transported into cardiac cells, concentrate in normal and ischemic myocardium, and act as a cardioprotector in vivo was examined. We also determined minocycline's capacity to act as a reducer of myocardial oxidative stress and matrix metalloproteinase (MMP) activity.
Background The identification of compounds with the potential to reduce myocardial ischemic injury is of great interest. Tetracyclines are antibiotics with pleiotropic cytoprotective properties that accumulate in normal and diseased tissues. Minocycline is highly lipophilic and has shown promise as a possible cardioprotector. However, minocycline's potential as an in vivo cardioprotector as well as the means by which this action is attained are not well understood.
Methods Rats were subjected to 45 min of ischemia and 48 h of reperfusion. Animals were treated 48 h before and 48 h after thoracotomy with either vehicle or 50 mg/kg/day minocycline. Tissue samples were used for biochemical assays and cultured cardiac cells for minocycline uptake experiments.
Results Minocycline significantly reduced infarct size (∼33%), tissue MMP-9 activity, and oxidative stress. Minocycline was concentrated ∼24-fold in normal (0.5 mmol/l) and ∼50-fold in ischemic regions (1.1 mmol/l) versus blood. Neonatal rat cardiac fibroblasts, myocytes, and adult fibroblasts demonstrated a time- and temperature-dependent uptake of minocycline to levels that approximate those of normal myocardium.
Conclusions Given the high intracellular levels observed and results from the assessment of in vitro antioxidant and MMP inhibitor capacities, it is likely that minocycline acts to limit myocardial ischemic injury via mass action effects.
Tetracyclines (TTCs) are broad-spectrum antibiotics and exert antimicrobial effects by inhibiting protein synthesis. Tetracyclines are classified as natural, semisynthetic, and chemically modified (1). Two of the more common semisynthetic TTCs used clinically as antibiotics are doxycycline and minocycline. Tetracyclines are known to chelate calcium and as a consequence incorporate into teeth (which during dentition stains them), cartilage, and bone (2). They also are found at high levels in gingival tissue/saliva, and their accumulation has been linked with an active cellular uptake process (3). In the 1970s, it was reported that radiolabeled TTCs accumulate in infarcted myocardium in proportion to the degree of tissue damage (4,5). However, the extent and means by which TTCs may accumulate in normal and ischemic myocardium has not been examined.
Doxycycline and minocycline possess cytoprotective properties (1,6–8). Effects are likely secondary to their capacity to act as anti-inflammatory, antiapoptotic, reactive oxygen species (ROS) scavenger, and matrix metalloproteinase (MMP) inhibitors (1,6–8). However, controversy exists in this area, given that, for example, doxycycline is not a particularly potent MMP inhibitor, because IC50values are in the micromolar range (9). Thus, it is of great importance to determine compound tissue concentrations because they would ultimately define how effective TTCs may be in exerting cytoprotective actions. Nonetheless, the promise of this class of drugs in preventing and/or limiting organ injury is currently being explored in more than 30 ongoing clinical trials, several of which cover cardiovascular pathologies such as stroke, vascular aneurisms, coronary bypass surgery, post-infarct ventricular remodeling, and diseases such as Parkinson's.
Of the TTCs, minocycline is unique in that it not only has excellent bioavailability, absorption, and a long half life, but it is also lipophilic (7). Thus, this agent may reach very high tissue levels (secondary to active transport systems and lipophilicity) and, as a consequence, have the ability to protect organs from ischemic injury. Indeed, reports from studies in which the authors used ex vivo heart systems and cultured cells indicate that minocycline can protect myocardium from ischemic injury (10). However, its potential for in vivo cardioprotective actions has not been determined. Thus, the purpose of our study was to determine the extent to which minocycline renders significant cardioprotection from ischemic injury in vivo, to examine myocardial levels attained, and to determine whether a cellular uptake process is involved. For this purpose, we used a rat model of ischemia/reperfusion (I/R) injury. We also wished to establish how myocardial minocycline levels may determine the effectiveness of the compound in exerting cytoprotective actions secondary to its antioxidant and MMP inhibitor properties.
Treatment with minocycline
Male Sprague-Dawley rats (Harlan) of 250 to 300 g were used. Two groups of rats were generated: control and minocycline-treated rats. Subgroups were used for the assessment of I/R injury (n = 12 control, n = 10 minocycline), hemodynamics (n = 5/each), and biochemical determinations (n = 4 control, n = 5 minocycline). A separate subgroup was generated for measurements of tissue minocycline (n = 5). Minocycline hydrochloride was administered via intraperitoneal injection at 25 mg/kg every 12 h, a dose within a range known to attain effective cardioprotection ex vivo under regional ischemia (10). In control rats, water was used. Treatment began 48 h before thoracotomy and continued until 48 h of reperfusion. Procedures were performed according to guidelines by the American Association for Accreditation of Laboratory Animal Care. The protocols were approved by the University of California, San Diego Institutional Animal Care and Use Committee and conform to published National Institutes of Health guidelines for animal research.
Animals were anesthetized by intraperitoneal injection of ketamine (100 mg/kg) and xylazine (10 mg/kg), intubated, and positive-pressure ventilated. Hearts were exposed through a left thoracotomy. The left anterior descending coronary artery was occluded and released after 45 min and the suture left in place.
In subgroups of anesthetized animals before sacrifice, a micromanometer was inserted into the right carotid artery to measure heart rate and carotid and left ventricular (LV) pressures.
Tissue sampling and staining
At 48 h of reperfusion, heparinized/anesthetized animals were euthanized and underwent cardiectomy. To delineate the area at risk, the left anterior descending coronary artery suture was retied, hearts were perfused with 0.4% trypan blue using a syringe, and then they were sliced into 2-mm thick sections. A single equatorial slice was photographed and the extent of the area at risk and infarct areas was quantified by the use of computer planimetry. The remaining slices were used for biochemical assays. The LV freewall was separated and divided into ischemic region and border zone. The right ventricle was used as nonischemic tissue. To distinguish between viable and infarcted myocardium, the equatorial slice was stained in 1% w/v 2,3,5-triphenyltetrazolium chloride (i.e., TTC) in phosphate buffer, pH 7.4, for 20 min at 37°C. Infarct area was quantified by digital planimetry and expressed as a percentage of area at risk.
Tissue (50 mg) was homogenized in 10 mmol/l HEPES, pH 7.5, 150 mmol/l NaCl, 0.2 mmol/l ethylene diamine tetraacetic acid, 25% glycerol, 100 μg/ml phenylmethyl sulfonylfluoride, and 0.2 kallikrein inhibitory U/ml aprotinin. Samples (10 μg) were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis as described (11). An internal control (human MMP-2/-9) was loaded to normalize gels. Gelatinolytic activity was digitally quantified.
Glutathione (glutathione disulfide/glutathione [GSSG/GSH]) assay
Tissue was homogenized in buffer (154 mmol/l KCl, 5 mmol/l diethylenetriaminepentaacetic acid, and 0.1 mol/l potassium phosphate, pH 6.8). After centrifugation, an aliquot was removed for protein determination. One volume of cold acid buffer (40 mmol/l HCl, 10 mmol/l diethylenetriaminepentaacetic acid, 20 mmol/l ascorbic acid, and 10% trichloroacetic acid) was added to 1 volume of homogenate. The suspension was centrifuged and the supernatant solution filtered. Reduced and oxidized glutathione (GSH) and GSSG levels, respectively, were determined by the use of fluorophore o-phthalaldehyde (12).
Neonatal and adult rat ventricular fibroblasts (NRVFs and ARVFs, respectively) and neonatal rat ventricular myocytes (NRVMs) were prepared, as previously described, with a collagenase-based method and gradient separations (13). We seeded NRVFs and ARVFs into 12-well cell culture plates (200,000 cells/well) and allowed them to proliferate to confluence (∼350,000 cells/well). We seeded NRVMs into 6-well cell culture plates (500,000 cells/well). Cells were cultured in standard serum-containing media until used (13). Cells were serum deprived for 24 h before use.
Minocycline transport assay in cardiac cells
Minocycline transport in cardiac cells was measured fluorometrically in each seeded well as described (3,14) with modifications. In brief, culture plates containing cell monolayers were washed 4× with Hank's balanced salt solution (HBSS), overlaid with HBSS, and warmed to 37°C before the assay. Warm HBSS (0.2 ml) containing twice the desired final minocycline concentration (40 μM) was simultaneously added to each well with multichannel pipettes. After incubation at 37°C and 4°C for the indicated times, the minocycline solution was quickly removed. Each well was then rapidly washed 4× according to incubation temperature with either warm (37°C) or cold (4°C) HBSS to eliminate extracellular minocycline. Cell monolayers were lysed with 1 ml of ddH2O and then sonicated on ice 3× for 5-s bursts (resting 10 s in between to prevent overheating). The lysate was centrifuged at 13,000g for 10 min. The supernatant was recovered and mixed with 1 ml of ethylene glycol containing 200 mmol/l citric acid and 200 mmol/l magnesium acetate before fluorescence measurement. Calibration plots were constructed to relate fluorescence to well minocycline content. To estimate the intracellular concentration of minocycline, NRVF and ARVF cell volume was measured using a cell and particle counter (Z2 Coulter Counter, Beckman Coulter, Fullerton, California) yielding an average diameter of 12 μm. The diameter of the NRVMs was obtained from previously published reports (19.5 μm) (15).
Minocycline determination in heart tissue and plasma
Fluorometric assay of minocycline in heart homogenates was done as previously described (16) with modifications. In brief, 150 mg of heart tissue was homogenized with a Polytron, in 2 ml of 20 mmol/l HEPES and 10 mmol/l MgCl2solution pH 7.4. Samples were centrifuged at 13,000g for 10 min. Supernatants were recovered and deproteinized with 0.6 ml of 1.5 mol/l trichloracetic acid. Samples were then vortexed and allowed to sit at room temperature for 10 min. Samples were centrifuged at 13,000g for 10 min. Supernatants were recovered and transferred to 15-ml conical tubes. We added 0.5 ml of 0.1 mol/l HCl containing 1 mg/ml thiopropionic acid along with 1.0 ml of Sorense's buffer (pH 6.0). After mixing, 0.5 ml of 0.75 mol/l aluminum chloride was added, and the tubes were shaken vigorously. After 15 min incubation at room temperature, fluorescence was measured (λex= 384, λem= 450 nm). A similar procedure was used for minocycline determinations in plasma. Calibration plots were constructed to relate fluorescence to tissue minocycline content.
Determination of IC50
The inhibition of human recombinant MMP-7 and -9 (Calbiochem, San Diego, California) activity by minocycline was measured with a fluorescent assay kit from BIOMOL International L.P. (Plymouth Meeting, Pennsylvania). Assays contained 0.6 μM MMP-7 or -9 in buffer [50 mmol/l N-(2-hydroxyethyl)piperazine-N′-ethanesulfonic acid, 10 mmol/l CaCl2, 0.05% Brij-35, pH 7.5] in the presence of increasing concentrations of minocycline (0, 0.01, 0.1, 0.25, 0.5, 1, 5, 50, 100, 250, 500, 1,000, 1,500, and 2,000 μM). Protein was incubated with drug for 1 h at 37°C in the dark. Cleavage reactions were started by addition of the fluorescent substrate. Fluorescence (λex= 335, λem= 405 nm) was measured at 30-s intervals for 1 h. Cleavage rates were determined from the linear regions of kinetic curves. We determined IC50from graphs of MMP activity versus log [minocycline].
The activity of MMP heart homogenates
Frozen tissue samples were homogenized with the use of a hand-held homogenizer on ice-cold lysis buffer (50 mmol/l Tris, pH 7.4; 150 mmol/l NaCl; 5 mmol/l CaCl2; 0.2 mmol/l NaN3; 0.1% Triton X-100). Homogenates were centrifuged and supernatants isolated. We mixed 100 μg of total protein and 20 μM OmniMMP fluorogenic substrate (BIOMOL International L.P.) in buffer. Given that the initial tissue concentration was diluted during homogenization, heart homogenates (in 100 μl of buffer) derived from minocycline-treated rats were supplemented with minocycline to reach a final concentration of 0.5 mmol/l (nonischemic regions) and 1.1 mmol/l (ischemic regions), as determined by results from the minocycline tissue uptake assays described in this study. Kinetic fluorescence measurements were performed (λex= 335, λem= 405 nm), and substrate cleavage rates were determined from the linear regions of kinetic curves.
Determination of antioxidant activity
Antioxidant activity for minocycline and Trolox (internal control) was measured in triplicates using an in vitro Antioxidant Assay Kit purchased from Cayman Chemical Company (Ann Arbor, Michigan).
Results are expressed as mean ± SEM. Comparisons between means were analyzed, as appropriate, by Student ttest or 1-way analysis of variance followed by Bonferroni t test. A value of p < 0.05 was considered statistically significant.
Minocycline reduced infarct size
Body and heart weights (data not shown) were similar for both groups after I/R. In the control group the mean area at risk (47 ± 3.7% of LV) was not significantly different from that in the minocycline-treated group (41 ± 2.7% of LV) (Fig. 1A).The infarct area in the minocycline-treated group (30 ± 3.5% of the area at risk) was significantly reduced compared with that of the control group (45 ± 4.2% of the area at risk), representing an overall infarct reduction of ∼33% by minocycline (p = 0.01) (Fig. 1B). The assessment of hemodynamics yielded nonsignificant differences between control versus minocycline-treated rats in heart rate (385 ± 25 beats/min vs. 363 ± 35 beats/min), LV end-diastolic pressure (6.5 ± 4 mm Hg vs. 4.1 ± 3.3 mm Hg), LV systolic pressure (101 ± 5 mm Hg vs. 104 ± 7 mm Hg), or mean arterial pressure (86 ± 7 mm Hg vs. 86 ± 5 mm Hg). Thus, changes in hemodynamics fail to explain the observed differences in infarct size.
Minocycline attenuated MMP-9 up-regulation in the ischemic region
Gelatin zymography of tissue homogenates from the ischemic and nonischemic regions revealed visible bands corresponding to 92-kDa MMP-9 (Fig. 2A).Densitometric analysis indicated that MMP-9 levels were reduced by ∼50% in the ischemic region of the minocycline-treated group versus controls (p < 0.01, Bonferroni, 6 comparisons) (Fig. 2B). We found that MMP-9 levels in the nonischemic regions remained similar between groups. No differences in MMP-2 levels were observed in any group (data not shown). The 86-kDa MMP-9 band was not quantified because it was not visible.
Minocycline as an antioxidant
We explored whether minocycline ameliorated oxidative stress in our in vivo model of cardiac I/R. We used GSSG/GSH as a measure of tissue oxidative stress. Ischemic tissue from minocycline-treated animals displayed a significant decrease (p < 0.05, Bonferroni, 6 comparisons) of GSSG/GSH by 97% versus control ischemic tissue (Fig. 3).In vitro assays indicate that minocycline bears similar antioxidant potential as the vitamin E analog Trolox when compared at the same concentrations (44, 88, 135, 180, 225, and 330 μM) (Fig. 4).
Accumulation of minocycline in the heart
Fluorometric analysis indicated that animals treated with minocycline displayed significant (p < 0.05, Bonferroni, 4 comparisons) accumulation of the antibiotic in the heart by revealing ∼24-fold increase relative to plasma levels. Moreover, minocycline accumulation was enhanced upon induction of I/R injury, where antibiotic concentrations reached levels ∼2-fold greater versus those of nonischemic regions (∼50-fold vs. plasma) (Fig. 5).Previous studies have shown that human gingival fibroblasts display an active transport mechanism for minocycline uptake (3,14). In cultured ARVFs (Fig. 6A),as well as in NRVF (Fig. 6B), minocycline accumulation also takes place in a time- and temperature-dependent manner. At 37°C, minocycline uptake in cardiac fibroblasts follows a kinetic pattern of active transport reaching a saturation stage at ∼6 min, whereas at 4°C, no significant accumulation occurred over time. Myocytes display a similar minocycline uptake process as in fibroblasts yielding a 6-fold concentration of the compound versus extracellular levels (Fig. 7).
Minocycline as MMP inhibitor
On the basis of the IC50values determined, minocycline revealed an IC50for MMP-7 (125 μM) and MMP-9 (180 μM) (Fig. 8).On the basis of the results obtained regarding minocycline uptake in heart tissue, MMP activity assays were supplemented with 0.5 mmol/l for nonischemic region and 1.1 mmol/l for ischemic region homogenates to approximate in vivo concentrations. Minocycline at 0.5 mmol/l in the nonischemic tissue reduced global MMP activity by 74% versus controls (ischemic and nonischemic). Minocycline at 1.1 mmol/l in the ischemic-tissue reduced global MMP activity by 91% versus controls (Fig. 9).Decreases were significant (p < 0.0001, Bonferroni, 6 comparisons). In addition, ischemic tissue treated with minocycline displayed a further reduction of 66% in global MMP activity when compared with that of nonischemic tissue treated with minocycline.
Results demonstrate the treatment of rats with minocycline before and after I/R yields a significant ∼33% reduction in infarct size. This is the first report for in vivo cardioprotection by minocycline. Scarabelli et al. (10) reported minocycline cardioprotection in cultured cardiac myocytes and ex vivo rat hearts. The cardioprotective effects were attributed to minocycline actions on apoptotic cell pathways. We have also previously reported on the cardioprotective actions of doxycycline (17,18). The ability of minocycline to limit tissue damage in the setting of ischemia has been documented in kidney and lung (19,20). Minocycline also demonstrates neuroprotection in vitro and in vivo (7,21,22). Neuroprotective mechanisms include reduction of microglial activation, down-regulation of proinflammatory/cell death mediators, and control of mitochondrial permeability (21–23). Minocycline, by potentially acting as an MMP inhibitor, also can prevent blood-brain barrier disruption (7,22,23).
Given this evidence, clinical studies in stroke patients have been performed where minocycline was administered orally for 5 days (200 mg) with a therapeutic window of time of 6 to 24 h after onset of stroke (24). Patients had significantly better outcome with minocycline versus placebo. Currently, clinical trials are being pursued to examine for doxycycline cardioprotection in coronary artery bypass patients and for amelioration of adverse post-infarction remodeling.
Investigators using radiolabeled TTC noted its capacity to accumulate in damaged myocardium and serve to diagnose infarcts (5,25). Results demonstrated a correlation between infarct size, as determined by radiolabeled TTC, and serum creatine kinase. The ability of TTCs to concentrate in other tissues is well known (26). Dentists take advantage of the high concentration of doxycycline in saliva as a means to treat periodontal (gum) disease, which compromises dental ligaments via MMPs. Periostat (a form of doxycycline available from CollaGenex Pharmaceuticals, Inc., Fort Worth, Texas) is commercialized for this purpose (6). To investigate TTCs' accumulation in gingival tissue, Yang et al. (3) examined the capacity of gingival fibroblasts to uptake the compounds. Gingival fibroblasts transport minocycline in a concentration- and temperature-dependent manner. At steady state, the cellular/extracellular concentration ratio was >60 for minocycline. It was concluded that gingival fibroblasts possess active transporters for TTCs that may explain high saliva levels of these agents. The uptake of TTCs also has been observed in neutrophils and may partly explain high levels observed in injured tissues (27).
We explored the capacity of minocycline to accumulate in myocardial tissue and cells. Minocycline accumulates in myocardium several-fold greater than plasma levels. Accumulation was more pronounced in ischemic versus normal myocardium or plasma. To identify the mechanism that may account for myocardial accumulation, we performed experiments using neonatal and adult rat cardiac fibroblasts and neonatal rat cardiac myocytes. Cardiac fibroblasts possess a comparable uptake system to that reported for gingival cells (3). Uptake was saturable in addition to time and temperature dependent. Cardiac myocytes also demonstrated uptake but with reduced magnitude. The estimation of intracellular minocycline levels indicates a capacity for isolated cardiac cells to accumulate the compound in a magnitude compatible with our in vivo normal tissue results. This observation allows us to conclude that most of the TTCs accumulated in myocardium are found in physical association with cellular structures. However, as observed in gingiva, high intracellular concentrations likely lead to high interstitial levels. We speculate that minocycline accumulation in ischemic tissues may also arise from its chelation properties binding to divalent cations such as calcium in cells whose permeability is compromised (28).
We previously reported a significant down-regulation of MMP-9 activity in ischemic myocardium with doxycycline treatment 48 h after I/R (18). Minocycline treatment also led to a significant decrease in 92-kDa MMP-9 levels in the ischemic region. No changes were observed in the 86-kDa form of MMP-9 or in MMP-2 levels. We also previously published the IC50of doxycycline for MMP-7 (28 μM) (9). We performed similar in vitro determinations by using minocycline as a means to determine the capacity of this agent to act as an MMP inhibitor. Results yielded an IC50for MMP-7 of 125 μM and for MMP-9 of 180 μM. Thus, minocycline is a weaker MMP inhibitor versus doxycycline in vitro, confirming previous reports (29,30).
However, because minocycline can attain millimolar concentrations, mass action effects may be possible on intracellular or immediate extracellular MMP targets (such as MMP-2 or -9), which may activate with I/R (31). To investigate this possibility, we performed global MMP activity assays. Samples were homogenized in assay buffer. Because homogenization requires the dilution of the tissue sample, we supplemented minocycline to the assay buffer to generate concentrations comparable to those derived from in vivo measurements. At the minocycline concentrations tested (0.5 mmol/l for nonischemic and 1.1 mmol/l for ischemic), a significant, dose-dependent reduction in MMP activity was observed. A recent report shows a significant correlation between the degree of MMP inhibition and brain levels of TTCs (32). Thus, it is possible that part of the cardioprotective effects of minocycline may be derived from high tissue levels of the compound.
Tetracyclines also are known as antioxidants. Reductions in ROS levels can prevent pro-MMP activation by the cysteine switch mechanism, as observed in stroke (33). Studies have demonstrated that the structural features of phenolic compounds and their hydroxyl radicals (as those of TTCs) confer ROS scavenger potential (28). Treatment with minocycline led to a significant blunting of ischemic tissue oxidative stress. These data are compatible with previous reports in which TTCs have been shown to reduce tissue oxidative stress (34,35). Leon et al. (35) recently reported the capacity of doxycycline to protect cardiac myocytes from peroxynitrite-induced contractile failure independent of MMP inhibitor actions. We compared the antioxidant potential of minocycline versus a known ROS scavenger, Trolox. Results indicated that minocycline has antioxidant potential that is comparable with Trolox. On the basis of the antioxidant results observed and the concentrations measured in samples of myocardium, we can assume that minocycline acts as an effective ROS scavenger. Again, this effect would result not only as a consequence of its unique antioxidant structural feature but also because of mass action effects. Minocycline has also been shown to be cytoprotective by attenuating apoptotic events/pathways (7). It is thus possible that high concentrations of minocycline found in normal and ischemic tissue also allow for notable effects on these other targets.
It is important to note the limitations associated with our study. We used a pretreatment scheme to examine the cardioprotective effects of minocycline. As such, the effects apply to ischemic injury and not necessarily to reperfusion (because we did not only give the drug either just before or after it). Thus, the extrapolation of these results to the clinical scenario awaits further studies. Furthermore, the use of an animal model of I/R injury has inherent limitations when compared with patients with coronary heart disease.
Results from our study indicate that minocycline can act in the in vivo setting to reduce infarct size after ischemic injury. Minocycline cardioprotection may not only be secondary to the capacity of the compound to modulate specific signaling pathways but also from mass action effects on end points such as antioxidant and MMP inhibitor actions. The levels of minocycline observed in tissue samples and cells justify a re-evaluation by investigators to consider how this class of drugs acts to limit tissue injury. Given their unusual capacity to accumulate in tissues and areas of injury, one would even be tempted to qualify TTCs as “smart” drugs.
This work was supported by National Institutes of Health (NIH) grants HL-43617 and HL-67922. Dr. Romero-Perez received support from a CONACYT-UC MEXUS doctoral fellowship and Dr. Fricovsky received support from a post-doctoral NIH fellowship (T32-DK007044). The first two authors contributed equally to this work. Deepak L. Bhatt, MD, FACC, FSCAI, FESC, FAHA, served as Guest Editor for this article.
- Abbreviations and Acronyms
- adult rat ventricular fibroblast
- glutathione disulfide
- Hank's balanced salt solution
- ischemia/ reperfusion
- left ventricular
- matrix metalloproteinase
- neonatal rat ventricular fibroblasts
- neonatal rat ventricular myocyte
- reactive oxygen species
- Received May 6, 2008.
- Revision received June 19, 2008.
- Accepted June 20, 2008.
- American College of Cardiology Foundation
- Yang Q.,
- Nakkula R.J.,
- Walters J.D.
- Holman B.L.,
- Idoine J.,
- Fliegel C.P.,
- et al.
- Holman B.L.,
- Zweiman F.G.
- Garcia R.A.,
- Pantazatos D.P.,
- Gessner C.R.,
- Go K.V.,
- Woods V.L. Jr..,
- Villarreal F.J.
- Scarabelli T.M.,
- Stephanou A.,
- Pasini E.,
- et al.
- Garcia R.A.,
- Brown K.L.,
- Pavelec R.S.,
- Go K.V.,
- Covell J.W.,
- Villarreal F.J.
- Villarreal F.J.,
- Kim N.N.,
- Ungab G.D.,
- Printz M.P.,
- Dillmann W.H.
- Walters J.D.,
- Nakkula R.J.,
- Maney P.
- Villarreal F.J.,
- Griffin M.,
- Omens J.,
- Dillmann W.,
- Nguyen J.,
- Covell J.
- Stirling D.P.,
- Koochesfahani K.M.,
- Steeves J.D.,
- Tetzlaff W.
- Nelson M.L.
- Wang W.,
- Schulze C.J.,
- Suarez-Pinzon W.L.,
- Dyck J.R.,
- Sawicki G.,
- Schulz R.
- Lee C.Z.,
- Yao J.S.,
- Huang Y.,
- et al.
- Gu Z.,
- Kaul M.,
- Yan B.,
- et al.