The cytotoxicity of celecoxib towards cardiac myocytes is cyclooxygenase-2 independent

Celecoxib decreases mitochondrial complex IV activity and induces oxidative stress in isolated rat heart mitochondria: An analysis for its cardiotoxic adverse effect

In spite of the cardiotoxic effect of selective cyclooxygenase-2 inhibitors, they are most widely used as anti-inflammatory and analgesic drugs. Today, valdecoxib and rofecoxib have been withdrawn in the market but celecoxib remains. In this study, we focused on an analysis of celecoxib toxic effects on isolated mitochondria. Isolated rat heart mitochondria were obtained using differential centrifugation. Using flow cytometry and biochemical assays, we searched succinate dehydrogenases, mitochondrial membrane potential (MMP), reactive oxygen species (ROS) formation, mitochondrial swelling, ATP/ADP ratio, lipid peroxidation, and mitochondrial complexes activity in rat heart isolated mitochondria. Herein, our results indicated a significant decrease in the activity of complex IV after exposure with celecoxib (16 µg/ml). This decrease in the activity of complex IV is paralleled by the MMP collapse, ROS formation, mitochondrial swelling, depletion of ATP, and lipid peroxidation. For the first time, this introductory study has shown a significant decrease in the activity of complex IV and mitochondrial dysfunction after exposure with celecoxib in rat heart isolated mitochondria.

Mechanisms of the Cardiac Myocyte-Damaging Effects of Dasatinib

The anticancer drug dasatinib (Sprycel) is a BCR-ABL1-targeted tyrosine kinase inhibitor used in treating chronic myelogenous leukemia that has been shown in clinical trials to display cardiovascular toxicities. While dasatinib potently inhibits BCR-ABL1, it is not a highly selective kinase inhibitor and may have off-target effects. A neonatal rat cardiac myocyte model was used to investigate potential mechanisms by which dasatinib damaged myocytes. The anthracycline cardioprotective drug dexrazoxane was shown to be ineffective in preventing dasatinib-induced myocyte damage. Dasatinib treatment increased doxorubicin accumulation in myocytes and doxorubicin-induced myocyte damage, likely through its ability to bind to one or more ABC-type efflux transporters. Dasatinib induced myocyte damage either after a brief treatment that mimicked the clinical situation, or more potently after continuous treatment. Dasatinib slightly induced apoptosis in myocytes as evidenced by increases in caspase-3/7 activity. Dasatinib treatment reduced pERK levels in myocytes most likely through inhibition of RAF, which dasatinib strongly inhibits. Thus, inhibition of the RAF/MEK/ERK pro-survival pathway in the heart may be, in part, a mechanism by which dasatinib induces cardiovascular toxicity.

Molecular Mechanisms of the Cardiotoxicity of the Proteasomal-Targeted Drugs Bortezomib and Carfilzomib

Bortezomib and carfilzomib are anticancer drugs that target the proteasome. However, these agents have been shown to exhibit some specific cardiac toxicities by as yet unknown mechanisms. Bortezomib and carfilzomib are also being used clinically in combination with doxorubicin, which is also cardiotoxic. A primary neonatal rat myocyte model was used to study these cardiotoxic mechanisms. Exposure to submicromolar concentrations of bortezomib and carfilzomib resulted in significant myocyte damage and induced apoptosis. Both bortezomib and carfilzomib inhibited the chymotrypsin-like proteasomal activity of myocyte lysate in the low nanomolar concentration range and exhibited time-dependent inhibition kinetics. The high sensitivity of myocytes, which were determined to contain high specific levels of chymotrypsin-like proteasomal activity, to the damaging effects of bortezomib and carfilzomib was likely due to the inhibition of proteasomal-dependent ongoing sarcomeric protein turnover. A brief preexposure of myocytes to non-toxic nanomolar concentrations of bortezomib or carfilzomib greatly increased doxorubicin-mediated damage, which suggests that the combination of doxorubicin with either bortezomib or carfilzomib may produce more than additive cardiotoxicity. The doxorubicin cardioprotective agent dexrazoxane partially protected myocytes from doxorubicin plus bortezomib or carfilzomib treatment, in spite of the fact that bortezomib and carfilzomib inhibited the dexrazoxane-induced decreases in topoisomerase IIβ protein levels in myocytes. These latter results suggest that the doxorubicin cardioprotective effects of dexrazoxane and the doxorubicin-mediated cardiotoxicity were not exclusively due to targeting of topoisomerase IIβ.

Mitochondrial toxicity of selective COX-2 inhibitors via inhibition of oxidative phosphorylation (ATP synthesis) in rat liver mitochondria

Cyclooxygenase-2 (COX-2) inhibitors (coxibs) are non-steroidal anti-inflammatory drugs (NSAIDs) designed to selectively inhibit COX-2. However, drugs of this therapeutic class are associated with drug induced liver injury (DILI) and mitochondrial injury is likely to play a role. The effects of selective COX-2 inhibitors on inhibition of oxidative phosphorylation (ATP synthesis) in rat liver mitochondria were investigated. The order of potency of inhibition of ATP synthesis was: lumiracoxib (IC50: 6.48 ± 2.74 μM)>celecoxib (IC50: 14.92 ± 6.40 μM)>valdecoxib (IC50: 161.4 ± 28.6 μM)>rofecoxib (IC50: 238.4 ± 79.2 μM)>etoricoxib (IC50: 405.1 ± 116.3 μM). Mechanism based inhibition of ATP synthesis (Kinact 0.078 min(-1) and KI 21.46 μM and Kinact/KI ratio 0.0036 min(-1)μM(-1)) was shown by lumiracoxib and data suggest that the opening of the MPT pore may not be the mechanism of toxicity. A positive correlation (with r(2)=0.921) was observed between the potency of inhibition of ATP synthesis and the log P values. The in vitro metabolism of coxibs in rat liver mitochondria yielded for each drug substance a major single metabolite and identified a hydroxy metabolite with each of the coxibs and these metabolites did not alter the inhibition profile of ATP synthesis of the parent compound. The results suggest that coxibs themselves could be involved in the hepatotoxic action through inhibition of ATP synthesis.

Selective and non-selective non-steroidal anti-inflammatory drugs differentially regulate pulmonary vein and atrial arrhythmogenesis

BACKGROUND

Non-steroidal anti-inflammatory drugs (NSAIDs) increase the risk of atrial fibrillation (AF). This study investigated whether selective and non-selective NSAIDs differentially regulate the arrhythmogenesis of pulmonary veins and atria.

METHODS

Conventional microelectrodes were used to record action potentials (APs) in isolated rabbit PVs, sinoatrial node (SAN), left atrium (LA), and right atrium (RA) preparations before and after celecoxib or indomethacin administration. A whole-cell patch clamp was used to record the sodium-calcium exchanger (NCX) current, L-type calcium current (ICa-L), and late sodium current (INa-late) before and after celecoxib administration in isolated PV cardiomyocytes.

RESULTS

Celecoxib (0.3, 1, and 3 μM) reduced PV spontaneous beating rates, and induced delayed afterdepolarizations and burst firings in four of eight PV preparations (50%, p<0.05). Celecoxib also reduced SAN beating rates and decreased AP durations (APDs) in RA and LA, but did not change the resting membrane potential. Indomethacin (0.3, 1, 3, and 10 μM) changed neither the PV or SAN beating rates nor RA APDs, but it reduced LA APDs. Celecoxib (3 μM) significantly increased the NCX current and decreased the ICa-L, but did not change the INa-late. Ranolazine (10 μM) suppressed celecoxib (3 μM)-induced PV burst firings in 6 (86%, p<0.05) of 7 PVs. KB-R7943 (10 μM) suppressed celecoxib (3 μM)-induced PV burst firings in 5 (71%, p<0.05) of 7 PVs.

CONCLUSIONS

Selective and non-selective NSAIDs differentially modulate PV and atrial electrophysiological characteristics. Celecoxib increased PV triggered activity through enhancement of the NCX current, which contributed to its arrhythmogenesis.

Celecoxib and ion channels: a story of unexpected discoveries

Celecoxib (Celebrex), a highly popular selective inhibitor of cyclooxygenase-2, can modulate ion channels and alter functioning of neurons and myocytes at clinically relevant concentrations independently of cyclooxygenase inhibition. In experimental systems varying from Drosophila to primary mammalian and human cell lines, celecoxib inhibits many voltage-activated Na(+), Ca(2+), and K(+) channels, including NaV1.5, L- and T-type Ca(2+) channels, KV1.5, KV2.1, KV4.3, KV7.1, KV11.1 (hERG), while stimulating other K(+) channels-KV7.2-5 and, possibly, KV11.1 (hERG) channels under certain conditions. In this review, we summarize the information currently available on the effects of celecoxib on ion channels, examine mechanistic aspects of drug action and the concomitant changes at the cellular and organ levels, and discuss these findings in the therapeutic context.

Inhibition of ion channels and heart beat in Drosophila by selective COX-2 inhibitor SC-791

Recent findings suggest that modulation of ion channels might be implicated in some of the clinical effects of coxibs, selective inhibitors of cyclooxygenase-2 (COX-2). Celecoxib and its inactive analog 2,5-dimethyl-celecoxib, but not rofecoxib, can suppress or augment ionic currents and alter functioning of neurons and myocytes. To better understand these unexpected effects, we have recently investigated the mechanism of inhibition of human K_v2.1 channels by a highly selective COX-2 inhibitor SC-791. In this study we have further explored the SC-791 action on ion channels and heartbeat in Drosophila, which lacks cyclooxygenases and thus can serve as a convenient model to study COX-2-independent mechanisms of coxibs. Using intracellular recordings in combination with a pharmacological approach and utilizing available Drosophila mutants, we found that SC-791 inhibited voltage-activated K⁺ and L-type Ca²⁺ channels in larval body-wall muscles and reduced heart rate in a concentration-dependent manner. Unlike celecoxib and several other K⁺ channel blockers, SC-791 did not induce arrhythmia. Instead, application of SC-791 resulted in a dramatic slowing of contractions and, at higher concentrations, in progressively weaker contractions with gradual cessation of heartbeat. Isradipine, a selective blocker of L-type Ca²⁺ channels, showed a similar pattern of heart arrest, though no prolongation of contractions was observed. Ryanodine was the only channel modulating compound of those tested additionally that was capable of slowing contractions. Like SC-791, ryanodine reduced heart rate without arrhythmia. However, it could not stop heartbeat completely even at 500 µM, the highest concentration used. The magnitude of heart rate reduction, when SC-791 and ryanodine were applied together, was smaller than expected for independent mechanisms, raising the possibility that SC-791 might be interfering with excitation-contraction coupling in Drosophila heart.

A multifaceted evaluation of imatinib-induced cardiotoxicity in the rat

Cardiotoxicity was an unanticipated side effect elicited by the clinical use of imatinib (Imb). This toxicity has been examined in only a limited number of experimental studies. The present study sought, by a variety of approaches, to identify important characteristics of Imb-induced cardiac alterations. Male spontaneously hypertensive rats (SHRs) received oral doses of 10, 30, or 50 mg/kg Imb or water daily for 10 d. Cardiac lesions, detected at all doses, were characterized by cytoplasmic vacuolization and myofibrillar loss. In a second experiment, cardiac lesions were found in Sprague Dawley (SD) and SHR rats given 50 or 100 mg/kg Imb for 14 d. Mean cardiac lesion scores and serum levels of cardiac troponin I were higher in SHRs than in SD rats. Imb induced myocyte death by necrosis, autophagy, and apoptosis. Dose-related increases in cardiac expression were observed for several genes associated with endoplasmic reticulum stress response, protein folding, and vascular development and remodeling. Imb caused alterations in isolated myocytes (myofibrillar loss, highly disrupted and disorganized sarcomeric α-actinin, apoptosis, and increased lactate dehydrogenase release) at low concentrations (5 mM). The authors conclude that Imb exerts cardiotoxic effects that are manifest through a complex pattern of cellular alterations, the severity of which can be influenced by arterial blood pressure.

A novel COX-2 inhibitor pyrazole derivative proven effective as an anti-inflammatory and analgesic drug

The introduction of new COX-2 inhibitors with high efficacy and enhanced safety profile would be a great achievement in the development of anti-inflammatory drugs. This study was designed to screen and assess the anti-inflammatory and analgesic activities as well as some of the expected side effects of some pyrazole derivatives, newly synthesized as potential COX-2 inhibitors at the Faculty of Pharmacy, Alexandria University and compared to indomethacin and celecoxib. Twelve compounds were screened for their anti-inflammatory activity using carrageenan-induced paw oedema and cotton pellet granuloma tests. On the basis of their apparent anti-inflammatory activity, four compounds with different substitutions were selected for the evaluation of their analgesic activity using the formalin-induced hyperalgesia and hot-plate tests. Compound AD 532, ((4-(3-(4-Methylphenyl)-4-cyano-1H-pyrazol-1-yl)benzenesulfonamide)), showed very promising results. In the single-dose and subchronic toxicity studies, compound AD 532 showed no ulcerogenic effect and produced minimal effects on renal function. Furthermore, compound AD 532 was a less potent inhibitor of COX-2 in vitro than celecoxib, which may indicate lower potential cardiovascular toxicity. It is concluded that compound AD 532 appears to be a promising and safe option for the management of chronic inflammatory conditions. This study recommends more in-depth investigation into the therapeutic effects and toxicity profile of this compound including its cardiovascular toxicity.

Mechanisms of myocyte cytotoxicity induced by the multikinase inhibitor sorafenib

The use of the anticancer multikinase inhibitor sorafenib is associated with cardiac ischemia or infarction and an increase in hypertension. We investigated various mechanisms that might be responsible for its cardiotoxicity in a neonatal rat myocyte model. As measured by lactate dehydrogenase release, sorafenib treatment of myocytes caused dose-dependent damage at therapeutically relevant concentrations. It had been hypothesized that inhibition of RAF1 and BRAF kinases may be responsible for sorafenib induced cardiotoxicity. However, because sorafenib treatment did not inhibit phosphorylation of ERK (extracellular signal-regulated kinase), it was concluded that sorafenib did not exert its damaging effects through RAF inhibition of the RAF/MEK/ERK kinase cascade. The clinically approved doxorubicin cardioprotective agent dexrazoxane did not protect myocytes from damage. At lower sorafenib concentrations, at least, these results are consistent with sorafenib not being able to induce significant oxidative damage. In conclusion, given the extreme lack of kinase selectivity that sorafenib exhibits, it is likely that inhibition of kinases other than RAF, or combinations of kinases, contributes to the cardiotoxic effects of sorafenib.

The use of the Integrated Discrete Multiple Organ Co-culture (IdMOC) system for the evaluation of multiple organ toxicity

The application of the Integrated Discrete Multiple Organ Co-culture (IdMOC) system in the evaluation of organ-specific toxicity is reviewed. In vitro approaches to predict in vivo toxicity have met with limited success, mainly because of the complexity of in vivo toxic responses. In vivo properties that are not well-represented in vitro include organ-specific responses, multiple organ metabolism, and multiple organ interactions. The IdMOC system has been developed to address these deficiencies. The system uses a 'wells-within-a-well' concept for the co-culturing of cells or tissue slices from different organs as physically separated (discrete) entities in the small inner wells. These inner wells are nevertheless interconnected (integrated) by overlying culture medium in the large outer containing well. The IdMOC system thereby models the in vivo situation, in which multiple organs are physically separated but interconnected by the systemic circulation, permitting multiple organ interactions. The IdMOC system, with either cells or tissue slices from multiple organs, can be used to evaluate cell type-specific or organ-specific toxicity.

Mechanisms of myocyte cytotoxicity induced by the multiple receptor tyrosine kinase inhibitor sunitinib

The anticancer tyrosine kinase inhibitor sunitinib has been shown recently to be cardiotoxic. Using a neonatal rat myocyte model, we investigated various mechanisms that might be responsible for its cardiotoxicity. Sunitinib potently inhibited the enzyme activity of both AMP-activated protein kinase (AMPK) and the ribosomal S6 kinase RSK1 at therapeutically relevant concentrations. Heart tissue with its high energy needs might be particularly sensitive to inhibition of AMPK because of its role as an energy sensor regulating ATP levels. As measured by lactate dehydrogenase release, sunitinib treatment of myocytes caused dose-dependent damage at therapeutic levels. Sunitinib treatment also caused a dose-dependent reduction in myocyte protein levels of the phosphorylated alpha and beta isoforms of the AMPK phosphorylation target acetyl-Coenzyme A carboxylase. However, myocytes were not protected from sunitinib treatment by pretreating them with the AMPK-activating antidiabetic drug metformin. Sunitinib treatment of myocytes also did not affect cellular ATP levels. Together, these last two results do not suggest a major role for inhibition of AMPK in sunitinib-induced myocyte damage. Dexrazoxane, which is a clinically approved doxorubicin cardioprotective agent, also did not protect myocytes from damage, which suggests that sunitinib did not induce oxidative damage. In conclusion, even though sunitinib potently inhibits AMPK and RSK1, given the extreme lack of kinase selectivity that sunitinib exhibits, it is likely that inhibition of other kinases or combinations of kinases are responsible for the cardiotoxic effects of sunitinib.

Celecoxib but not the combination of celecoxib+atorvastatin prevents the development of monocrotaline-induced pulmonary hypertension in the rat

The present study aimed to assess the effects of a COX-2 inhibitor, celecoxib, a HMG-CoA reductase inhibitor, atorvastatin, and the association of both on monocrotaline (MC)-induced pulmonary hypertension in rats. Celecoxib (Cib, 25 mg kg(-1) day(-1)), atorvastatin (AS, 10 mg kg(-1) day(-1)) or vehicle, were given orally, separately or in combination, for 26 days to Wistar male rats injected or not with MC (60 mg/kg intraperitoneally). At 4 weeks, MC-injected rats developed a severe pulmonary hypertension, with an increase in lung to body weight ratio (L/BW), right ventricular pressure (RVP in mmHg, 31 +/- 3 and 14 +/- 1 for MC and control groups, respectively, P < 0.05) and right ventricle/left ventricle + septum weight ratio (RV/LV+S) associated with a decrease in acetylcholine- and sodium-nitroprusside-induced pulmonary artery vasodilation in vitro. Hypertensive pulmonary arteries exhibited an increase in wall thickness (wall thickness to external diameter ratio, 0.42 +/- 0.01 vs 0.24 +/- 0.01 for MC and control groups, respectively, P < 0.001). Whole lung eNOS expression was decreased, and an increase in apoptosis, evaluated by cleaved caspase-3 expression, was evidenced by Western blotting. Cib (RVP in mmHg, 19 +/- 3 and 31 +/- 3 for MC+Cib and MC groups, respectively, P < 0.05), but neither AS nor AS+Cib significantly limited the development of pulmonary hypertension (P < 0.05), although the three treatments exhibited protective effects against MC-induced lung and right ventricle hypertrophy evaluated by L/BW and RV/(LV+S) ratios, respectively (P < 0.05). AS, Cib and AS+Cib treatments reduced MC-induced thickening of small intrapulmonary artery wall (0.42 +/- 0.01, 0.24 +/- 0.01, 0.26 +/- 0.01 and 0.28 +/- 0.01 for MC, MC+AS, MC+Cib and MC+AS+Cib groups, respectively, P < 0.001). In control rats, Cib reduced acetylcholine-induced pulmonary artery vasorelaxation. Treatment of MC rats by either Cib or AS did not modify acetylcholine-induced pulmonary artery relaxation, whereas combination of both drugs significantly worsened it (P < 0.05). AS, but neither Cib nor the combination of both, prevented apoptosis (AS, P < 0.05) and partially restored eNOS expression (AS, P < 0.05) in whole lung of MC rats. In conclusion, celecoxib exhibited beneficial effects against the development of monocrotaline-induced pulmonary artery hypertension and right ventricular hypertrophy. These beneficial effects of celecoxib might be, at least partly, explained by its effects on pulmonary artery thickening and pulmonary hypertrophy, even if it did not show any effect on pulmonary artery vasorelaxation and whole lung eNOS expression or apoptosis. The combination of celecoxib and atorvastatin was unable to prevent MC-induced pulmonary hypertension, decreased endothelium-dependent vasorelaxation and showed a trend toward an increased in RVP that deserves further studies.

The dihydroorotase inhibitor 5-aminoorotic acid inhibits the metabolism in the rat of the cardioprotective drug dexrazoxane and its one-ring open metabolites

Dexrazoxane (ICRF-187) is clinically used as a doxorubicin cardioprotective agent and to prevent anthracycline extravasation injury. It may act by preventing iron-based oxygen free radical damage through the iron-chelating ability of its metabolite N,N'-[(1S)-1-methyl-1,2-ethanediyl]bis[(N-(2-amino-2-oxoethyl)]glycine (ADR-925). Dexrazoxane undergoes an initial metabolism to its two one-ring open intermediates [N-(2-amino-2-oxoethyl)-N-[(1S)-2-(3,5-dioxo-1-piperazinyl)-1-methylethyl]glycine (B) and N-(2-amino-2-oxoethyl)-N-[(2S)-2-(3,5-dioxo-1-piperazinyl)propyl]glycine (C)] and is then further metabolized to its presumably active metal-chelating form ADR-925. We previously showed that the first ring opening reaction is catalyzed by dihydropyrimidinase and the second by dihydroorotase (DHOase), but not vice versa. To determine whether DHOase was important in the metabolism of dexrazoxane, its metabolism and that of B and C to ADR-925 were measured in rats that were pretreated with the DHOase inhibitor 5-aminoorotic acid. In rats pretreated with 5-aminoorotic acid the area-under-the-curve concentration of ADR-925 was reduced 5.3-fold. In rats treated with a mixture of B and C, the maximum concentration of ADR-925 in the plasma was significantly decreased in rats pretreated with 5-aminoorotic acid, which indicates that DHOase directly metabolized B and C. Both heart and liver tissue levels of ADR-925 in rats were also greatly reduced by pretreatment with 5-aminoorotic acid. Together these results indicate that the metabolism of dexrazoxane and of B and C is mediated by DHOase. These results provide a mechanistic basis for the antioxidant cardioprotective activity of dexrazoxane.

Effects of celecoxib on ionic currents and spontaneous firing in rat retinal neurons

Accumulating evidence suggests that the side effects of celecoxib, widely used to treat muscle and joint pain, may be mediated in part through cyclooxygenase-2 (COX-2) independent mechanisms, such as inhibition of ion channels. In this study we report effects of celecoxib on ionic currents and neuronal activity in isolated rat retinal neurons. We found that celecoxib suppressed voltage-gated potassium currents in retinal bipolar cells with an effective concentration to inhibit 50% of function (EC(50)) of 5.5 microM. In retinal amacrine and ganglion cells, celecoxib inhibited voltage-dependent sodium channels with an EC(50) of 5.2 microM, and voltage-dependent transient and sustained potassium currents with EC(50)s of 16.3 and 9.1 microM, respectively. Notably, the rate of spontaneous spike activity was dramatically suppressed in ganglion and amacrine cells with an EC(50) of 0.76 microM. All actions of celecoxib on ionic currents and action potentials occurred from the extracellular side and were completely reversible. These findings indicate that inhibition of ion channels by celecoxib in the CNS may affect neuronal function at clinically relevant concentrations.

Apoptosis is not the major death mechanism induced by celecoxib on rheumatoid arthritis synovial fibroblasts

Synovial hyperplasia in rheumatoid arthritis (RA) has been associated with apoptosis deficiency of RA fibroblast-like synoviocytes (FLSs). Celecoxib is a non-steroidal anti-inflammatory drug that has been demonstrated to induce apoptosis in some cellular systems. We have therefore examined the dose- and time-dependent effects of celecoxib on RA FLS viability. Treatment of RA FLSs with celecoxib for 24 hours reduced their viability in a dose-dependent manner. Analysis of celecoxib-treated RA FLSs for their content of apoptotic and necrotic cells by Annexin V staining and TO-PRO-3 uptake displayed only few apoptotic cells. Caspase 3, a key mediator of apoptosis, was not activated in celecoxib-treated RA FLSs, and the presence of specific caspase 3 or pan-caspase inhibitors did not affect celecoxib-induced cell death. Moreover, we could not detect other signs of apoptosis, such as cleavage of poly(ADP-ribose) polymerase, caspase 8 or 9, or DNA fragmentation. We therefore conclude that apoptosis is not the major death pathway in celecoxib-treated RA FLSs.