Caspofungin Modulates Ryanodine Receptor-Mediated Calcium Release in Human Cardiac Myocytes

TEXT

Caspofungin (CAS) is known as the first echinocandin antifungal agent recommended for the empirical and specific treatment of invasive Candida infections and aspergillosis (1, 2). According to several clinical case reports describing adverse cardiac events during echinocandin therapy in critically ill patients, animal studies revealed an altered contractility of isolated rat cardiomyocytes after CAS administration (3–6). According to these results, studies in isolated heart models treated with CAS also found impaired cardiac function (7). In line with this, after central venous administration of CAS, rats were found to exhibit impaired left ventricular function associated with reduced survival rates (8, 9). Furthermore, hemodynamic measurement in endotoxemic rats, mimicking the clinical setting, revealed that CAS application induced cardiac impairment even at subclinical concentrations (10). Therefore, we suggested that mechanisms similar to septic cardiomyopathy might be causative (4–6, 10–13). The aim of our study was to investigate the impact of CAS on intracellular calcium homeostasis in rat and human cardiac myocytes.

First, we identified CAS-induced human cardiac myocyte (HCM) cytotoxicity using a cell viability assay of cultured HCMs (C-12810; PromoCell GmbH, Heidelberg, Germany). Cytotoxicity increased dose-dependently after the administration of dosages of 50 μg/ml CAS and higher (13.50% ± 3.42%; n = 4). Dosages of >100 μg/ml were associated with severe cytotoxicity (100 μg/ml, 76.15% ± 7.03%; 140 μg/ml, 80.93% ± 10.79%; 150 μg/ml, 85.63% ± 8.89%; 200 μg/ml, 82.52% ± 11.84%; all n = 4).

Second, we showed a dose-dependent increase in intracellular Ca²⁺ concentration ([Ca²⁺]_i) oscillation frequency in freshly isolated rat cardiomyocytes, with significant effects at CAS concentrations of ≥50 μg/ml, by performing Ca²⁺ imaging with 2.5 μM Fura-2-AM-microscopy (methodology as described earlier [10]) (Fig. 1; n_CTRL = 24; n_CAS = 10 to 22 per group; 12.5 μg/ml, fold change [FC] = 1.13, P = 0.2; 25 μg/ml, FC = 1.11, P = 0.5; 50 μg/ml, FC = 1.60, P < 0.001; 100 μg/ml, FC = 2.28, P < 0.01; 200 μg/ml, FC = 2.74, P < 0.001). All experiments were approved by the local committee for animal care (JLU-no. 540_M; Regierungspraesidium, Giessen, Germany). CAS treatment (75 to 200 μg/ml) in HCMs caused a dose-dependent increase in [Ca²⁺]_i in physiologic calcium-containing buffer medium. Dosages of >130 μg/ml were associated with a significant elevation in the Fura-2-AM ratio attributed to [Ca²⁺]_i (n_CTRL = 20; n_CAS = 29 to 40 per group; P < 0.01) of [Ca²⁺]_i with CAS >130 μg/ml and increased [Ca²⁺]_i between 1.6- and 3.3-fold (control <1-fold; Fig. 2). Furthermore, the CAS-dependent increase in [Ca²⁺]_i was also found in experiments involving Ca²⁺-free buffer medium (n_CTRL = 20; n_CAS = 30 to 50 per group; P < 0.01) of [Ca²⁺]_i with CAS at >100 μg/ml and increased [Ca²⁺]_i between 2.2- and 2.4-fold (control <1-fold)) (Fig. 3).

• Open in new tab
• Download powerpoint

FIG 1

Boxplot diagram showing maximum (Max.) of % oscillation frequency in 1/s⁻¹ under the influence of different dosages of CAS in rat cardiomyocytes. Boxplot whiskers represent confidence intervals of median. ***, P < 0.001; n.s., not significant.

• Open in new tab
• Download powerpoint

FIG 2

Boxplot diagram demonstrating the influence of physiological and Ca²⁺-free buffer media, CAF, RYN, and CAS on changes in intracellular Ca²⁺ levels in individual rat cardiomyocytes determined by fluorescence microscopy of Fura-2-AM signal. Administration of substances is characterized with + (CAS +, 140 μg/ml CAS; Ca²⁺ +, 2.5 mM Ca²⁺; RYN +, 40 μM RYN) and −. ***, P < 0.001.

• Open in new tab
• Download powerpoint

FIG 3

Dose-response curve of intracellular calcium levels measured with fluorescence microscopy in relation to different dosages of CAS. Ca²⁺ + and Ca²⁺ − symbolize Ca²⁺-containing and Ca²⁺-free buffer media, respectively. ED₅₀, 50% effective dose.

Third, we were able to identify potential mechanisms of [Ca²⁺]_i release by measuring the ratio of Fura-2-AM fluorescence in HCM. While CAS-induced elevation in [Ca²⁺]_i was found in physiological and Ca²⁺-free buffer media (Fig. 2A), CAS-induced (140 μg/ml) elevation of [Ca²⁺]_i was significantly reduced in the presence of caffeine (CAF; 30 mM) (n_CTRL = 20, n_Effect = 40; CAF+ versus CAF−, FC = 1.04; P < 0.001; Fig. 2B). Application of ryanodine (RYN; 40 μM) to inhibit ryanodine receptors prior to 140 μg/ml CAS administration resulted in a significant suppression of [Ca²⁺]_i release (RYN n_CTRL = 20, n_Effect = 40; RYN+ versus RYN−, FC = 1.15; P < 0.01; Fig. 2C).

Echinocandins represent a well-established group of antifungal agents which are widely used for intensive care treatment in critically ill patients (1, 2). However, recent data raise doubts about their safe use in critically ill patients with regard to hemodynamic stability (14–16). Several cases of severe hemodynamic instability following echinocandin infusion were reported, which led to the definition of an antifungal-associated drug-induced cardiac disease by Cleary et al. (4–6, 11). Cardiac impairment following echinocandin administration was reported in isolated rat cardiomyocytes (3). Especially, high-dose central-venous application of anidulafungin or caspofungin in adult rats resulted in a significant reduction in cardiac output and was also associated with reduced survival rate compared to control animals. These effects might be explained by a rapid onset of high peak levels of echinocandins (8). Since adverse effects of echinocandins have been observed especially in septic patients, endotoxemic rats were examined subsequently. While control and micafungin-treated animals did not show hemodynamic alterations, administration of anidulafungin or CAS in clinically used dosages led to a significant and dose-dependent decrease in cardiac output (10). Mitochondrial damage was suspected as a result of analysis of isolated rat hearts in transmission electron microscopy following echinocandin infusion (7). In contrast to these findings, we were not able to identify mitochondrial damage in spectrophotometric measurements of rat left ventricular cardiac tissue or altered mitochondrial enzyme activity after echinocandin treatment (8, 10). Therefore, we hypothesized that the mechanism behind CAS-induced cardiac alterations was not yet fully identified. With respect to the hemodynamic impairment of endotoxemic rat hearts following echinocandin infusion and studies supporting an endotoxin-induced suppression of the l-channel-dependent calcium flow, we hypothesized that alterations in Ca²⁺ homeostasis induced by echinocandins might be one potential harmful mechanism (10, 17–19).

CAS is the oldest and a well-established echinocandin and was therefore chosen as the substrate in this study. Beyond, CAS offers lipophilic features, in contrast to micafungin, and is therefore able to penetrate cell membranes. Our results might explain some of the observed detrimental effects of echinocandins in former studies. Long-term incubation of CAS in freshly isolated rat cardiomyocytes leads to severe cell toxicity. Therefore, all further results were performed by short-time application of CAS. In HCMs, the dose-dependent increase in [Ca²⁺]_i and maximum oscillation frequency prove a CAS-induced intracellular Ca²⁺ increase. Ca²⁺-free buffer medium led to a reduced maximum but still dose-dependent effect of CAS-induced increase of [Ca²⁺]_i, supporting the hypothesis of Ca²⁺ release from intracellular Ca²⁺ stores. The computed 50% effective dose (ED₅₀) of CAS in Ca²⁺-containing buffer medium amounted to 148.1 μg/ml, which is approximately 7.5-fold higher than plasma levels of healthy probands (20). These findings are in line with cardiac impairment in high-dose-caspofungin-treated rats (8). However, critically ill patients suffering from multiple organ failure might be at high risk for excessive plasma levels, especially if CAS is infused rapidly via a central venous catheter. This might explain previous observations of sustained cardiac impairment following standard-dose administration of echinocandins in endotoxemic rats (10).

Subsequently, we addressed the mechanism of Ca²⁺ release. Caffeine (CAF) is well known to release Ca²⁺ most probably from sarcoplasmic reticulum (SR) by activation of RYN receptors. Therefore, depleting caffeine-sensitive SR Ca²⁺ stores was found to inhibit CAS (140 μg/ml)-induced Ca²⁺ release (21, 22). Also, RYN used in doses that inhibit ryanodine receptors was able to inhibit CAS-induced increase in [Ca²⁺]_i. These results lead to the conclusion that CAS interferes with intracellular caffeine-sensitive stores, most probably via the activation of ryanodine receptors. Interestingly, these effects were found in physiological and Ca²⁺-free buffer media and therefore are independent of the influence of extracellular Ca²⁺, further supporting the hypothesis of Ca²⁺ release from the SR.

However, our study exhibits some limitations. Variations in pharmacokinetic and pharmacodynamic parameters might vary between HCMs and rat cardiac myocytes (RCMs). HCM-derived cell lines might feature different reactions to substrates than native cells. Furthermore, isolated cells are not able to represent physiologic in vivo reactions because of their lack of cell-to-cell or other physiologic interactions.

In conclusion, first, we were able to prove a severe grade of toxicity of long-term application of CAS in HCMs, which is in line with the findings of earlier studies (7, 8). Second, measurements in the oscillation frequency of cytosolic Ca²⁺ showed a dose-dependent impact of CAS with an increase in oscillation frequency. Third, we addressed these results and asked for the reasons of elevation in oscillation frequency and found a dose-dependent increase in [Ca²⁺]_i following CAS treatment in a 340/380 nm fluorescence intensity ratio. Fourth, Ca²⁺ ions were found to be released from intracellular caffeine-sensitive stores, most probably via the activation of ryanodine receptors.