N -Acetylcysteine Protects against Bupivacaine-induced Myotoxicity Caused by Oxidative and Sarcoplasmic Reticulum Stress in Human Skeletal Myotubes

• ❖ Local anesthetic injections can produce myotoxicity although the clinical significance is not entirely clear

• ❖ Mechanisms for myotoxicity by local anesthetics are not understood

• ❖ In cultured human muscle cells, local anesthetics produced myotoxicity that was reduced by the antioxidant N -acetylcysteine

• ❖ In the future, perhaps N -acetylcysteine could be used to prevent local anesthetic induced myotoxicity when this is anticipated

LOCAL anesthetics (LAs) offer the benefits of extended analgesia with greater patient satisfaction compared with intravenous morphine after orthopedic surgery.1–3Challenges remain, however, for the use of LAs to improve the comfort and postoperative pain relief in patients who receive continuous regional blocks for surgical procedures. These benefits, however, can be offset by adverse iatrogenic muscle pain caused by bupivacaine.4,5LA-induced myotoxicity seems to be a rather uncommon side effect of local and regional anesthesia. However, studies revealed that certain anesthesia techniques, especially retrobulbar and peribulbar blocks, are related to a relatively high postoperative rate of significant muscular dysfunction directly caused by these agents.6 

Therefore, skeletal muscle damage has to be considered a potentially serious complication after LA application. The frequency of these symptoms is largely unknown because they remain underreported.6Moreover, a better understanding of the mechanisms of LA-induced myotoxicity is needed to develop efficient clinical strategies to protect against LA adverse outcomes.

After LA administration, the affected muscle fibers undergo an intrinsic degenerative phase with increase in pycnotic myonuclei and pathologically condensed chromatin. However, the cellular mechanisms, particularly the apoptotic pathway(s) involved in LA myotoxicity, remain to be characterized (reviewed in7). Several studies indicate that in lidocaine-induced neurotoxicity the intrinsic mitochondrial death pathway, rather than the extrinsic death receptor pathway, might be implicated in the apoptotic response to LA administration.8,9In addition, bupivacaine could interfere with the mitochondrial energy metabolism in skeletal muscle,10,11leading to Ca²⁺deregulation. It has been proposed that increased intracellular Ca²⁺concentrations play a role in myocyte injury.12,13Indeed, in skinned fiber preparations of skeletal muscle, LAs cause Ca²⁺release from the sarcoplasmic reticulum (SR) and simultaneously inhibit Ca²⁺reuptake,14–16thus resulting in persistently increased Ca²⁺levels. On the other hand, in psoas muscle of rats anesthetized with bupivacaine and ropivacaine administered by catheter for femoral nerve block, a decrease in the mitochondrial energy metabolism was detected, but without changes in SR Ca²⁺content.5Interestingly, a microarray analysis revealed that, in HL-60 promyelocytic leukemia cells, bupivacaine treatment up-regulated the expression of several apoptosis-related genes and also of X-box-binding protein 1 (XBP-1).17XBP-1 is a transcription factor that operates as a downstream component of the unfolded protein response against sarcoplasmic or endoplasmic reticulum (SR/ER) stress caused by accumulation of unfolded/misfolded proteins.18XBP-1 also controls expression of C/EBP homologous protein (CHOP),19a key signaling component of ER stress-induced apoptosis, which, therefore, could mediate bupivacaine-induced myotoxicity.

To better characterize the mechanism(s) of LA-induced myotoxicity and, particularly, the involvement of SR/ER stress, we investigated bupivacaine effects in human primary myotubes. We have identified the caspases activated during the apoptotic response to bupivacaine and demonstrated that bupivacaine-induced apoptosis is mediated by SR stress caused by increased reactive oxygen species (ROS) production. Finally, we report that N -acetylcysteine protects against these iatrogenic effects.

Anti-tubulin monoclonal antibody was from Sigma-Aldrich (St. Quentin Fallavier, France). Anti-CHOP, anti-XBP-1, and anti-ATF-6 (activating transcription factor 6) polyclonal antibodies were from Abcam (Paris, France). Anti-CHOP monoclonal antibody, anti-PARP (poly(ADP-ribose)polymerase) and anti-cleaved-PARP, anti-caspases 9, 7, 6, and 3, and anti-cleaved caspases 9, 7, 6, and 3 polyclonal antibodies were from Cell Signaling Technology (Ozyme; Montigny, France). The Alexa Fluor secondary antibodies were purchased from Invitrogen (Cergy-Pontoise, Paris, France). Bupivacaine (1-butyl-N -(2,6-dimethylphenyl)-2-piperidinecarboxamide), staurosporine, thapsigargin, and N -acetylcysteine were purchased from Sigma-Aldrich, and CM-H²DCFDA derivatives (5- (and 6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate, acetyl ester) were from Invitrogen.

Human primary myoblasts were isolated as described previously20from quadriceps muscle biopsies obtained from the “AFM-BTR Banque de Tissus pour la Recherche ” (Hôpital de la Pitié-Salpêtrière, 75013 Paris). 30,000 myoblasts/ml were grown in growth medium (Dulbecco's modified Eagle's medium; Cambrex, Lonza, Verviers, Belgium) supplemented with 10% fetal calf serum (Perbio Science, Brebieres, France) and 1% Ultroser G (PALL Life Sciences, Cergy St. Christophe, France) for 5 days. At day 5, myoblasts started to fuse, and cultures were switched to differentiation medium composed of Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum and penicillin/streptomycin (Cambrex). Cultures of myotubes grown in differentiation medium were used for all experiments.

Cell extracts prepared as described previously21were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, blotted onto nitrocellulose membranes, and probed with the indicated antibodies. Bound antibodies were detected with a chemiluminescence detection kit (PerkinElmer, Courtaboeuf, France). Analysis and quantification were performed with Kodak 1D3.6 software (Kodak, Carestream Health France, Bagnolet, France).

For immunofluorescence, cells were fixed in 4% paraformaldehyde in phosphate-buffered saline and permeabilized with 0.1% Triton X-100 in phosphate-buffered saline. Cells were incubated with primary antibodies for 2 h, followed by Alexa Fluor secondary antibodies and 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) staining. Cell imaging was performed at the “Centre Regional d'Imagerie Cellulaire de Montpellier.” Fluorescent cells were viewed with a Leica Microscope (Leica DM6000, Wetzlar, Germany) using a 40×/NA = 1.30 EC Plan-Neofluar grade oil objective. Images were captured as 16 TIFF files with a MicroMax 1300 CCD cameras (RS-Princeton Instruments, Inc., Roper Scientific SAS, Evry, France) driven by the MetaMorph (version 7; Universal Imaging Corp., Roper Scientific SAS). Images were processed using ImageJ software (National Institutes of Health, Bethesda, MD; ImageJ is in the public domain).

We used CM-H²DCFDA as a cell-permeant indicator of ROS. Myotubes were cultured in 6-well plates, treated with 0.5, 1.0, and 1.5 mm bupivacaine for 8 h, washed, and incubated with 5 μm CM-H²DCFDA and DAPI in phosphate-buffered saline at 37°C in the dark for 60 min. Cells were then washed with phosphate-buffered saline three times, fixed in 4% paraformaldehyde for 5 min, and stored at 4°C for up to 16 h before image acquisition. Cells were visualized with Zeiss (Carl Zeiss SAS; Le Pecq, France) fluorescent microscope (AxioCam MRm equipped with a charged-coupled device camera), using excitation sources and filters appropriate for fluorescein isothiocyanate as described by the manufacturer. Fluorescence quantification was performed with ImageJ using for each experimental condition. Ten images from at least three independent experiments (30 images, more than 300 myotubes analyzed in total/experimental condition) were then combined to allow quantitative analysis of changes in ROS levels. In all experiments, fluorescence levels were normalized by the number of nuclei.

The OxyBlot Protein Oxidation Detection kit was purchased from Millipore (Division Bioscience Millipore SAS, St. Quentin en Yvelines, France), and analysis of the oxidation status of protein samples (dinitrophenylhydrazine derivatization) was performed following the manufacturer's instructions. One-dimensional electrophoresis was carried out using 12.5% sodium dodecyl sulfate-polyacrylamide gels after dinitrophenylhydrazine derivatization. Proteins were transferred onto nitrocellulose membranes and then stained with Red Ponceau to control for sample loading. After labeling with the rabbit anti-dinitrophenyl antibody included in the kit, blots were developed using a chemiluminescence detection system.

Distribution of the quantitative variables was fitted. Transformation to ensure normal distribution was performed when necessary. Means were compared between each experimental conditions by applying a one-way analysis of variance or a unpaired t  test when a priori  comparisons were performed versus  control. When analysis of variance was significant, post hoc  unpaired t  test were applied, and the P  value was adjusted to control the false discovery rate according to the Holm's procedure. All P  values were two-tailed and were considered significant at 0.05 or less. All data are presented as mean ± SEM, and the statistical analysis was performed using Stata Statistical Software version 6.0 (Stata Corporation, College Station, TX).

The myotoxicity of the LA bupivacaine was investigated in primary cell cultures of human myoblasts that fused in vitro  to form mature differentiated myotubes. Because the cleavage of PARP is an indicator of apoptosis, the presence of its cleaved form was determined by immunofluorescence in myotubes incubated with different concentrations (from 0.75 to 1.5 mm) of bupivacaine and with 1 μm staurosporine, a known initiator of apoptosis in vitro , for 18 h. Cleaved PARP was clearly detected in myotubes treated with staurosporine and in cells incubated with 1.25 and 1.5 mm bupivacaine (fig. 1A). In addition, the number of PARP-positive nuclei in myotubes treated with 1.25 mm bupivacaine increased proportionally to the duration of the incubation (fig. 1B). Thus, induction of apoptosis by bupivacaine is dose- and time-dependent in primary human myotubes. We then assessed whether and which caspases were implicated in bupivacaine-induced apoptosis by analyzing the expression of their cleaved forms. Immunoblot analysis showed that cleaved PARP and cleaved caspase 9 and 7 were detected in human myotubes treated with 1.5–2 mm bupivacaine but not in cells treated with 1 mm bupivacaine (fig. 2A), consistent with the results described in figure 1. Similarly, expression of cleaved caspase 9 and 7 increased with longer incubation times (figs. 2B and C). Cleavage of caspase 3 and 6 was not induced by bupivacaine differently from staurosporine, which induced strong expression of all the cleaved caspases tested (fig. 2A). These results indicate that bupivacaine induces apoptosis in human myotubes by a mechanism probably different from that of staurosporine.

Because the activation of caspase 7 is involved in ER stress-mediated apoptosis,22we next tested whether bupivacaine myotoxicity could be mediated via  induction of SR/ER stress in human myotubes. To this aim, myotubes were incubated with different concentration (from 0.75 to 1.5 mm) of bupivacaine for 18 h and with 100 nm thapsigargin, an irreversible inhibitor of the ER Ca²⁺-ATPase, for 4 h. Expression of CHOP,23–25a key downstream transcription factor in SR stress-induced apoptosis, was stimulated by thapsigargin (fig. 3A). Bupivacaine (1–1.5 mm) also strongly up-regulated CHOP expression (fig. 3A), and this effect was dose-dependent (fig. 3B). Moreover, 1 mm bupivacaine induced CHOP expression (figs. 3A, 4A, and 4B) but not activation of caspase 9 and 7 (figs. 4A and B) as observed also with thapsigargin. Thus, induction of SR stress seems to precede activation of caspases. To test this hypothesis, the kinetics of the expression of XBP-1 and ATF-6 (two transcription factors that control expression of the unfolded protein response genes like CHOP), CHOP, and cleaved caspase 9 and 7 were determined in myotubes treated with 1 mm bupivacaine. ATF-6, XBP-1, and CHOP were detected after 8 h of bupivacaine treatment, and their expression was maintained, except for XBP-1, which was transient (figs. 4C and D). Cleaved caspases were detected only at 24 h (figs. 4C and D). Together, these results show that after exposure to bupivacaine, SR stress and expression of the unfolded protein response genes are induced before apoptosis, suggesting that the apoptotic events depend on SR stress induction.

ROS alter cellular redox reactions, interfere with protein disulfide bonding, and result in protein misfolding leading to SR stress and the unfolded protein response activation (reviewed in26). Therefore, primary myotubes were incubated with different concentrations (from 0.5 to 1.5 mm) of bupivacaine for 8 h, and ROS production was evaluated indirectly by measuring the oxidation status of CM-H²DCFDA derivatives. Dichlorofluorescein oxidation was significantly increased in myotubes treated with 1 mm and 1.5 mm bupivacaine compared with control, untreated cells (fig. 5A). Moreover, ROS production in bupivacaine-treated myotubes was inhibited by the antioxidant drugs N -acetylcysteine, trolox (a water-soluble derivative of vitamin E), and resveratrol (antioxidant polyphenol) in a dose-dependent manner (fig. 5Band data not shown). To confirm that bupivacaine-induced ROS caused oxidative stress in myotubes, the content in carbonylated proteins of samples treated with 1.5 mm bupivacaine for 8 h was assessed. As shown in figure 5C, carbonylated proteins increased in response to bupivacaine treatment, and this effect was prevented by antioxidant treatment.

To test whether inhibition of ROS production could protect against bupivacaine-induced SR stress and apoptosis, primary myotubes were incubated with 1 or 1.5 mm bupivacaine and increasing concentrations of the antioxidant N -acetylcysteine for 18 h. Addition of N -acetylcysteine decreased CHOP expression and inhibited expression of activated caspases in myotubes (figs. 6A and B). CHOP expression was also monitored by immunofluorescence, and the percentage of nuclei expressing CHOP decreased from 48% in bupivacaine-treated myotubes to 26%, 16%, and 9% in cells coincubated with 2, 5, and 10 mm N -acetylcysteine, respectively (fig. 6C).

These results show that, in primary human myotubes, blocking ROS production with the antioxidant N -acetylcysteine has a protective effect against bupivacaine-induced myotoxicity. These findings could have an important clinical impact, which will have to be evaluated in practice.

A variety of experimental findings indicate that LA myotoxicity has a complex pathogenesis. The experiments described indicate that treatment of human primary myotubes with bupivacaine induces ROS production and SR stress leading to apoptosis characterized by expression of cleaved caspase 9 and 7. Coincubation of myotubes with bupivacaine and antioxidant drugs, especially N -acetylcysteine, inhibits ROS production and SR stress and consequently caspase activation (fig. 7).

Skeletal muscle fiber degeneration induced by bupivacaine has been confirmed in multiple studies, but the pathway of cell death remained to be clarified.7In neuroblastoma cells, the LA lidocaine induces a cell death8,9that is prevented by the lack of caspase 9,9indicating an apoptosis mediated by the intrinsic mitochondrial death pathway. Consistent with these observations, we detected the cleaved form of PARP, which is an indicator of apoptosis, and expression of activated caspase 9 and 7 in human myotubes treated with bupivacaine, suggesting the involvement of the intrinsic pathway. LA also can decrease the mitochondrial membrane potential,10,11leading to progressive fragmentation of the mitochondrial network in human myoblasts.27However, mitochondria are unlikely to be the only targets of LAs. The increase of the intracellular Ca²⁺concentrations by LAs could play a role in myocyte injury by inducing pathways of cell death.12,13In isolated myofibers, bupivacaine induces Ca²⁺release from the SR by acting on the Ca²⁺release channel ryanodine receptors at SR membranes and inhibits Ca²⁺reuptake into the SR, which is mainly regulated by SR Ca²⁺-ATPase activity.16,28Bupivacaine could inhibit these receptors by direct binding, but this has never been demonstrated. Therefore, and based on the results of this work, we propose an alternative mechanism linked to induction of oxidative stress. In human myotubes, bupivacaine treatment increases production of ROS, which may then modify oxidizable residues (cysteine, tyrosine) of SR-associated calcium channels, such as ryanodine receptors (S -nitrosylation) and SR/ER Ca²⁺-ATPases (by tyrosine nitration), causing their dysfunction and SR calcium depletion.29–31However, disturbance of intracellular Ca²⁺homeostasis should not be considered the only mechanism by which bupivacaine induces apoptosis in muscle. Indeed, in vivo , bupivacaine injected via  femoral nerve block catheters has deleterious effects on mitochondrial energy without affecting calcium homeostasis,5and in our experiments, bupivacaine induced apoptosis at concentrations that do not influence Ca²⁺release from SR. Another mechanism, associated with SR/ER stress, should therefore be considered. Here, we show that, in myotubes, bupivacaine induces activation of caspase 7, which is involved in ER stress-mediated apoptosis22as well as expression of the ER stress markers XBP-1, ATF-6, and CHOP. Oxidative stress can cause SR/ER stress. Production of ROS, which alter cellular redox reactions, and production of nitric oxide, a mediator of protein nitrosylation, interfere with protein disulfide bonding and result in protein misfolding.26The excessive presence of misfolded proteins disrupts SR/ER function, resulting in accumulation of unfolded or misfolded proteins in the reticulum lumen, which leads to SR/ER stress and the unfolded protein response (reviewed in19). In bupivacaine-treated myotubes, addition of the antioxidant N -acetylcysteine protects against ROS production, as expected, but also inhibits expression of CHOP and of activated caspases. These results strongly suggest that bupivacaine causes muscle injury by increasing ROS production, which leads to SR/ER stress and finally to apoptosis. The implication of ROS production in bupivacaine-mediated toxicity was also shown in a Schwann cell line,32but how bupivacaine induces ROS production remains a major question for future studies.

Furthermore, the dose of 1–2 mm bupivacaine used for the experiments described in this paper was chosen to mimic the effects of direct exposure of myotubes to this agent at clinically relevant concentrations (7.7 mm = 0.25% bupivacaine solution used in peripheral nerve blocks). These concentrations are in the same range as those that altered mitochondrial energy metabolism in human myoblasts (1–5 mm bupivacaine)27and induced apoptosis in a Schwann cell line (0.5 mm bupivacaine).32,In vivo , lipophilic LAs accumulate in tissues, but the real concentrations in the cells remain unknown. Bupivacaine concentration in the muscle 1 h after the last injection of a clinically relevant protocol performed in rats was approximately 30 nmol/mg of tissue or 30 μm.5Nevertheless, this concentration is underestimated because the authors could not measure precisely the local concentration of bupivacaine in the strict diffusion space within the muscle because of the limited sensitivity of the high-performance liquid chromatography system that would have required a larger biopsy. Thus, 1 mm bupivacaine is probably not so far away from real concentrations in the vicinity of the nerve catheter. Anyhow, in clinical practice, the relevance of concentration depends mainly on the dose effect, which play a key role in bupivacaine toxicity. This suggests that the current clinical protocols for LA administration should be reevaluated and adapted to use dose close to the minimal LA concentration needed.

The observation that an antioxidant protects against LA-induced myotoxicity constitutes an important result for clinical practice. Recently, we showed that recombinant human erythropoietin (hrEPO), often used for its hematopoietic effects, can partially protect against LA-induced myotoxicity in rats and cultured human myoblasts.27Nevertheless, the recombinant human erythropoietin doses used in these experiments were much higher than those typically used clinically to treat anemia.33Therefore, the risk of side effects combined with its extremely high cost does not make of recombinant human erythropoietin an ideal treatment against LA-induced myotoxicity. Conversely, N -acetylcysteine might be an interesting alternative, although the clinical impact of our results remains to be evaluated in practice.