Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The calcineurin inhibitor cyclosporine has been widely used as an immunosuppressant after organ transplantation (Kahan, 1989) and for the treatment of immune diseases (Faulds et al., 1993) for more than 2 decades. However, its clinical use is limited by a narrow therapeutic index and toxicity, mainly to kidney, liver, and the central nervous and cardiovascular systems (Kahan, 1989). Recent studies have indicated that cyclosporine significantly inhibits energy metabolism in brain (Serkova et al., 2001, 2002), kidney (Henke et al., 1992), and liver (Zhong et al., 2001), all primary target organs of cyclosporine toxicity. Although cyclosporine causes neurotoxicity (Kahan, 1989; Gijtenbeek et al., 1999) and was shown to significantly inhibit brain mitochondrial energy metabolism under normoxic conditions (Serkova et al., 2001), paradoxically, cyclosporine protects mitochondrial energy metabolism during ischemia and reperfusion in cell cultures and animal models (Uchino et al., 1998; Serkova et al., 2002). Although a substantial effort has been made to assess the neuroprotective effects of cyclosporine against stroke, only relatively few reports have evaluated the potential of cyclosporine to prevent ischemic/reperfusion damage in other organs, such as the heart. This is surprising since cyclosporine only poorly penetrates the blood-brain barrier, limiting its clinical application as a neuroprotective agent (Begley et al., 1990).

Recently, it has been shown that cyclosporine readily distributes into cardiac tissue, reaching concentrations greater than in blood (Serkova et al., 2000). Previous in situ and in vitro studies have suggested that cyclosporine has both detrimental and beneficial effects on the heart (Paul et al., 1991, 1992; Tatou et al., 1996; Rao et al., 1998; Park et al., 1999). For example, cyclosporine may protect the heart against toxicity of drugs such as doxorubicin (Adriamycin) and cyclophosphamide (Al-Nasser, 1998a,b), prevent cardiac hypertrophy (Force et al., 1999), and diminish ischemia/reperfusion injury (Griffiths and Halestrap, 1993; Halestrap et al., 1997; Massoudy et al., 1997; Weinbrenner et al., 1998; Squadrito et al., 1999). As of today, the changes of myocardial cell metabolism caused by cyclosporine under normoxic conditions have not yet been described in detail. If, as in the brain, cyclosporine reduces mitochondrial oxidative energy production prior to cardiac occlusion, it could have a significant impact on ischemia/reperfusion injury in the infarcted heart. It has been previously shown that depletion of energy sources prior to cardiac ischemia produces physiological preconditioning, which is beneficial for postischemic outcome (Murry et al., 1986; Garnier et al., 1996). It was our goal to study the effects of cyclosporine on cardiac metabolism in vivo using magnetic resonance spectroscopy (MRS) to determine whether there is any correlation between the infarct size and the status of cardiac metabolism, and to find the optimum cyclosporine dose for cardiac protection against ischemia/reperfusion injury.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Experimental Design. All animal protocols were reviewed and approved by the University of California, San Francisco, Committee on Animal Research, and animal care was in agreement with the National Institutes of Health guidelines for ethical animal research (National Institutes of Health publication 80-123, revised 1985). For our studies, we used young adult male Sprague-Dawley rats (6 months old). Body weight (280-320 g) was monitored daily, and no change occurred during the cyclosporine treatment. Rats were randomly assigned to two groups of cyclosporine and cyclosporine + coronary occlusion (n = 20/group). In both groups, cyclosporine (doses 0, 5, 10, 15, and 25 mg/kg/day, n = 4) was administered as its Neoral formulation (Novartis Pharma AG, Basel, Switzerland) diluted in milk by daily oral gavage for 3 days.

Effect of Cyclosporine on Cardiac Energy Metabolism. To assess cyclosporine-induced metabolic changes, we quantified key cell metabolites in perchloric acid (PCA) and lipid extracts using MRS. On day 4, 4 h after the last cyclosporine dose, rats were anesthetized, and the beating hearts were excised and immediately frozen in liquid nitrogen. Frozen hearts were weighed, homogenized in a mortar grinder in the presence of liquid nitrogen, and extracted with 4 ml of ice-cold PCA (12%) as described in detail by Serkova et al. (2001). For each extract, 1 g of heart tissue was used. The samples were centrifuged, the aqueous phase was removed, and the samples were neutralized using KOH and centrifuged once again. The lipid fraction was extracted from the pellets remaining after the aqueous extraction step. The pellets were dissolved in 4 ml of ice-cold water, and the solution was neutralized. Both the aqueous extract and the redissolved pellet were lyophilized overnight. The lyophilisates of the aqueous extracts were redissolved in 0.45 ml of deuterium oxide (D₂O) and adjusted to pH 7 using DCl and NaOD. The lipid fraction was extracted from the lyophilisates of the redissolved pellet by the addition of 1 ml of deuterated chloroform/methanol mixture (CDCl₃/CD₃OD; 2:1 v/v). After centrifugation, the supernatants were analyzed by MRS.

Effect of Cyclosporine on Infarct Size in Rats after Coronary Occlusion and Reperfusion. On day 4, 4 h after the last cyclosporine dose, rats in the cyclosporine + coronary occlusion group were subjected to 30 min of coronary artery occlusion followed by 24 h of reperfusion. The rats were anesthetized with a single dose of 50 mg/kg ketamine and 1.4 mg/kg xylazine injected intraperitoneally. The trachea was exposed, and a tracheotomy was performed. The rats were ventilated using a constant-volume ventilator with a volume of 1 to 2 ml/100 g of body weight at a rate of 65 to 70 strokes/min. The chest was opened through the fourth intercostal space. The pericardium was opened, and the main branch of the left coronary artery was occluded for 30 min 1 to 2 mm below the left atrial appendage by an intramural stitch with a 6.0 polypropylene suture. Subsequently, the suture was removed and the chest was closed; the animals were allowed to recover from anesthesia. Animals were monitored for 2 h after completion of surgery.

Measurement of Cyclosporine Concentrations in Blood and Heart Tissues. At the time of sacrifice of the cyclosporine + coronary occlusion group, blood samples (0.25 ml, anticoagulated with EDTA) and cardiac tissue samples from the normally perfused heart, the area at risk, and the infarcted area were collected. Samples were frozen in liquid nitrogen and stored at 80°C until HPLC/MS analysis. All samples were analyzed within 7 days after collection. For quantification of cyclosporine concentrations, an HPLC/electrospray-MS assay with automated on-line sample preparation by column-switching (LC/LC-MS) based on methods described previously (Christians et al., 2000; Serkova et al., 2000) was used. In brief, heart tissue samples were weighed and homogenized in 0.5 ml of KH₂PO₄ buffer, pH 7.4 (1 M). For protein precipitation, 1 ml of methanol/1 M ZnSO₄ (80:20 v/v) containing 100 µg/l internal standard cyclosporine D was added to 0.25 ml of blood/tissue homogenate. Samples were vortexed for 30 s and centrifuged at 10,000g for 10 min. One hundred microliters of the supernatant were injected onto the 10.2-mm extraction column filled with Hypersil ODS-1 of 10-µm particle size (Shandon Scientific, Chadwick, UK). Automated sample preparation and analysis were carried out using two HPLC systems (all series HP1100 components; Hewlett-Packard, Palo Alto, CA) connected by a 7240 Rheodyne 6-port switching valve mounted on a step motor (Rheodyne, Rohnert Park, CA). Samples were washed with a 4:6 (v/v) mobile phase of methanol and 0.1% formic acid (flow 5 ml/min). After 0.75 min, the switching valve was activated, and the analytes were eluted in the backflush mode from the extraction column onto the 50 × 4.6 mm C8, 3.5-µm analytical column (Zorbax; Agilent Technologies, Palo Alto, CA). The mobile phase was methanol and 0.1% formic acid supplemented with 1 µmol/l sodium formate. The gradient was run time 0 min, 35% methanol; 7 min, 75% methanol; and 9 min, 90% methanol (flow rate, 0.5 ml/min). Both columns were kept at 65°C. Single ions [M + Na]⁺ were recorded. The mass spectrometer (G1946A mass selective detector; Hewlett-Packard) was focused on m/z = 1224 (cyclosporine), m/z = 1238 (internal standard cyclosporin D) (Christians et al., 2000; Serkova et al., 2000).

Statistical Analysis. Values are expressed as mean ± standard deviation, if not otherwise designated. In the MRS study, concentrations between the controls and cyclosporine-treated animals were compared using unpaired Student's t test. The effects of cyclosporine on infarct size were compared between controls and cyclosporine dose groups using analysis of variance in combination with Duncan grouping as a post hoc test. The relationship among doses, metabolite concentrations, and blood and tissue concentrations as well as cyclosporine concentrations and infarction size were assessed using correlation analysis. The SPSS software package (version 10.07; SPSS Inc., Chicago IL) was used to calculate distribution statistics and for all statistical analyses. A P value of <0.05 was considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of Cyclosporine on Myocardial Energy Metabolism. ATP, ADP, phosphocreatine, NAD⁺/NADH, and water-soluble phospholipids were quantified based on ³¹P MR spectra (Fig. 1). The following cell metabolites could be detected and were quantified using ¹H MRS of heart PCA extracts: 1) glycolysis intermediates lactate and alanine; 2) Krebs' cycle products glutamate, glutamine, and succinate; 3) high-energy phosphates creatine/phosphocreatine; 4) osmolytes taurine; and 5) glutathione. Three days of cyclosporine oral treatment significantly reduced cardiac high-energy phosphates as indicated by a dose-dependent decrease of ATP and phosphocreatine concentrations as well as by reduction of the [ATP/ADP] and [phosphocreatine/creatine] ratios surrogate markers for (Table 1). The concentration of NAD⁺/NADH, surrogate markers for oxidative phosphorylation, was decreased after treatment (Table 1) and correlated positively (r = 0.89, P < 0.002) with the ATP levels. Furthermore, cyclosporine significantly reduced Krebs' cycle intermediates by as much as 35% (glutamate at a dose of 25 mg/kg/day) (Table 1). A positive correlation with the ATP concentrations was found for both glutamate (r = 0.95, P < 0.001) and glutamine (r = 0.84, P < 0.002). In parallel, the concentrations of lactate, the end product of anaerobic glycolysis, were significantly increased in all treatment groups (Table 1) and inversely correlated with ATP concentrations (r = 0.93, P < 0.001). The increase in lactate production was associated with a decrease in glucose concentrations that correlated with the reduction in high-energy phosphate concentrations (ATP: r = 0.92, P < 0.001).

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Representative 31P MRS spectra of heart PCA extracts from rats in the control (A), 5 mg/kg/day (B), 10 mg/kg/day (C), 15 mg/kg/day (D), and 25 mg/kg/day (E) cyclosporine-treated groups . Peak assignment: PCr, phosphocreatine; PME, phosphomonoesters.

Effect of Cyclosporine on Infarct Size in Rats. Cyclosporine at doses of 10 and 15 mg/kg/day significantly reduced the size of infarct relative to the area at risk (P < 0.03). At a dose of 15 mg/kg, the percentage of left ventricular infarct was 4-fold smaller than in the control rats (15 mg/kg/day, 7.3 ± 6.7% versus control, 29.4 ± 5.7%). Cyclosporine dose of 25 mg/kg/day resulted in slightly less protection than the dose of 15 mg/kg/day (area of infarct/area at risk at 25 mg/kg/day, 17.0 ± 14.6% versus 15 mg/kg/day, 13.9 ± 12.9%; Fig. 2). The difference between the two doses, however, was not statistically significant. The relative sizes of the areas at risk/area of the left ventricle were similar in the different dose groups as evaluated by analysis of variance (Fig. 2). Concentrations of the high-energy phosphates ATP and NAD⁺/NADH in the control group significantly correlated with the ratios [area of infarction/area at risk] and [area of infarction/area of the left ventricle] (r = 0.97, P < 0.001 and r = 0.82, P < 0.01, respectively).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Assessment of the dose-dependent cardioprotective effect of orally administered cyclosporine on the area at risk/area of the left ventricle [%], the area of infarct/area at risk [%], and the area of infarct/left ventricle [%] in the rat. Data are presented as mean ± standard error (n = 4).

Cyclosporine Blood and Cardiac Tissue Concentrations. Cyclosporine doses were significantly correlated with cyclosporine concentrations in blood (r = 0.81, P < 0.001), in normal myocardial tissue (r = 0.86, P < 0.001), and in the infarcted tissue (r = 0.87, P < 0.001) but not in the area of risk (Fig. 3). A significant inverse correlation was evident between the cyclosporine dose (blood concentration) and the percentage of area of infarct/area of risk as well as the percentage of area of infarct/area of the left ventricle with r = 0.81, P < 0.01 and r = 0.86, P < 0.01, respectively. If the 25 mg/kg/day was excluded from the analysis, the correlation coefficients improved to 0.86 and 0.91, respectively. In addition, there was a significant negative correlation between the cyclosporine dose and ATP (r = 0.89, P < 0.01), NAD⁺/NADH (r = 0.94, P < 0.001), glutamate concentrations (r = 0.83, P < 0.02), and a significant positive correlation with the lactate concentrations (r = 0.87, P < 0.002).

View larger version (38K): [in this window] [in a new window]   Fig. 3.   Cyclosporine concentrations in blood, noninfarcted normal myocardial tissue, myocardial area at risk, and infarcted myocardial tissue 24 h after onset of hypoxia (30 min) and reperfusion and 28 h after the last cyclosporine dose. Cyclosporine concentrations are presented as mean ± standard error (n = 4) and were measured by a specific LC/LC-MS assay.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The main result of the present study is that cyclosporine inhibits cardiac high-energy phosphate and Krebs' cycle metabolism in a dose-dependent fashion. Comparable data have not been previously reported in the literature. Our MRS data indicate that cyclosporine inhibits the Krebs' cycle (decreased glutamate/glutamine concentrations) and mitochondrial oxidative phosphorylation (decreased concentrations of NAD⁺), resulting in a significant cellular reduction in ATP and phosphocreatine with a concomitant increase of ADP concentration. The inhibition of mitochondrial energy production caused lactate accumulation followed by elevated fatty acid oxidation at the high cyclosporine doses. The increased lactate concentrations in the cyclosporine-treated rats can be explained by compensatory stimulation of anaerobic ATP synthesis via glycolysis (Schlant, 1978). The decrease of fatty acids at the highest treatment dose (25 mg/kg/day cyclosporine) can be attributed to the increased stimulation of fatty acid oxidation, which is another additional alternative pathway for energy production (Schlant, 1978), when oxidative mitochondrial pathway remains largely inhibited by cyclosporine. This may explain the increased lipid peroxidation in the 25 mg/kg/day cyclosporine-treated group. Malondialdehyde, one of the main products of lipid peroxidation, indirectly measured in our TBA test, is generated during hydrolysis by oxidation of PUFA (Draper and Hadley, 1990; Suttnar et al., 1997). PUFA concentrations, as well as the total fatty acid pool, were decreased in 25 mg/kg/day treated rats and correlated with the increased LPO. The biochemical depression of mitochondrial glucose and high-energy phosphate metabolism found in our study are very similar to those induced by cyclosporine in the brain (Serkova et al., 2001, 2002). The ability of cyclosporine to decrease the energy state by inhibiting mitochondrial metabolic pathways has also been reported for the kidney (Henke et al., 1992) and, more recently, it has been shown that cyclosporine produces hypoxia-like conditions in the normoxic liver as well (Zhong et al., 2001). The mechanism of cardiac energy depression is not clear, and the present study design does not allow one to answer this specific question. However, our MRS results suggest that the effects of cyclosporine on myocardial cell metabolism mimic to a certain extent the effects of hypoxia. Hypoxic-like conditions, which result in the depletion of mitochondrial energy homeostasis, can produce the protective effect on myocardium against prolonged ischemia/reperfusion damages (Murry et al., 1986; Garnier et al., 1996). It can be speculated that by inducing metabolic changes similar to hypoxia, cyclosporine may also have ischemic preconditioning effects. Besides direct depression of mitochondrial respiration, another possible mechanism, such as induction of heat shock protein expression as shown for the kidney (Yang et al., 2001), may be involved in cyclosporine-induced pharmacological preconditioning. However, the potential contribution and the exact mechanism of cyclosporine-induced preconditioning will require further assessment.

This study showed the salutary effect of cyclosporine on infarct size in rats subjected to 30 min of coronary occlusion. Pretreatment with cyclosporine for 4 days was based on the rationale that protective and/or adaptive mechanisms, such as pharmacological preconditioning, induced by cyclosporine may depend on changes of gene expression that may take hours to days for full effect. We found that within 24 h of 30 min of coronary artery occlusion, cyclosporine pretreatment produced a dose-dependent reduction in infarct size relative to the area at risk and to the area of the left ventricle. Our evaluation of different doses suggested maximum protective effect at doses between 10 and 15 mg/kg/day, with a fall-off at higher doses (25 mg/kg). This finding is in good agreement with those of Griffiths and Halestrap (1993) who observed that in isolated perfused hearts the cardioprotective effect was dose-dependent with an apparent reversal of the protective effect at higher doses. Furthermore, the ratio [area of infarct/area at risk] in the occlusion group strongly correlated with the decrease of cellular ATP and NAD⁺ concentrations in control rats. Doses of 10 and 15 mg/kg/day cyclosporine produced inhibition of mitochondrial energy production prior to occlusion, resulting in a decrease in infarct size after 24 h of reperfusion. At 25 mg/kg/day cyclosporine, the positive effect of cyclosporine appears to be antagonized, possibly due to an increased lipid peroxidation.

Other effects of cyclosporine, such as inhibition of the immune response following ischemia/perfusion injury, have also been described, which may contribute to the beneficial effect of cyclosporine on ischemia/reperfusion injury (Squadrito et al., 1999). Griffiths and Halestrap (1993) found that cyclosporine keeps the mitochondrial nonspecific pores closed during reperfusion. This effect was not dependent on calcineurin inhibition but most likely due to inhibition of cyclophilin D, since a nonimmunosuppressive cyclosporine derivative was equally as effective as cyclosporine. Weinbrenner et al. (1998) found evidence that cyclosporine protects the ischemic rabbit heart by inhibiting the calcium-calmodulin-dependent protein phosphate 2B, calcineurin. Massoudy et al. (1997) reported that cyclosporine acts as a cardioprotective agent by an endothelin-dependent mechanism. None of these studies pretreated animals with cyclosporine or were based on isolated hearts; in the only in vivo study reported (Squadrito et al., 1999), cyclosporine was administered intravenously 15 min after coronary occlusion. We also measured cyclosporine concentrations in the remote nonischemic myocardium, the area at risk, and the infarcted area by using a highly specific and sensitive, validated HPLC/MS assay. Since cyclosporine oral bioavailability is known to be erratic and variable (Oellerich et al., 1995), we also measured cyclosporine blood concentrations. Our study suggested a good correlation between the oral cyclosporine dose and cardiac tissue concentrations as well as a good correlation between blood and cardiac tissue concentrations. The cyclosporine doses and blood concentrations that produced significant reduction of the area of infarction are comparable with those obtained in/desired in patients after heart transplantation (Oellerich et al., 1995). In conclusion, cyclosporine, in a dose-dependent fashion, induced depression of the cardiac energy metabolism before occlusion and decreased infarct size in a rat coronary occlusion model. The depression of cardiac energy metabolism induced by cyclosporine correlated with the reduction in infarct size and suggests that cyclosporine pretreatment produces an energetic milieu similar to that achieved by ischemic preconditioning. Further studies are needed to test the effect of cyclosporine on cardiac function and the severity of myocardial injury.