Flumazenil (FMZ), a specific antidote for benzodiazepines, is the first benzodiazepine receptor blocker approved for clinical use [1]. It is primarily used to reverse the sedation caused by the overdose of benzodiazepines [2, 3]. The pharmacodynamics of this drug are related to its ability to block Gamma-aminobutyric acid receptor type A (GABAAR) by binding selectively at the benzodiazepine binding site [4, 5].

Given that Gamma-aminobutyric acid (GABA) is a primary inhibitory neurotransmitter in the brain [6], GABAAR is thought to have constitutive activity in reducing anxiety state and inhibiting central nervous system (CNS) convulsions. Research has indicated that GABA binds to its recognition site at the postsynaptic GABAAR, a ligand-gated chloride channel, allowing chloride ions to enter neurons, causing hyperpolarization and inhibiting action potential transmission [7]. Thus, defects in GABAAR activity constitute a risk factor for developing seizures and anxiety-related disorders, which necessitate effective medical interventions [8, 9]. The use of benzodiazepines, positive allosteric modulators for GABAAR, may help to reduce anxiety and inhibit CNS convulsion [10]. On the other hand, the use of beta-carbolines, which act as inverse agonists at benzodiazepine binding sites of GABAAR, is associated with a high risk of developing anxiety and CNS convulsions [11]. Interestingly, a line of studies found that FMZ can block the action of some benzodiazepines and beta-carbolines [12, 13]. This may suggest, at least in part, that inhibiting the constituent activity of GABAAR is an important dynamic effect for a variety of epileptogenic and anxiogenic agents, including the analgesic drug mefenamic acid (MFA).

Previous studies have indicated that MFA, a common drug in the fenamate family of non-steroidal anti-inflammatory drugs (NSAIDs), has a modulatory effect on GABAAR that depends on the specific beta subunits present in the receptor [14]. Although MFA has been used for a variety of therapeutic benefits, particularly as an anti-inflammatory [15], analgesic [16], and antipyretic properties [17], recent studies suggest that its potential risk of CNS toxicity limits its clinical applications [18]. In several case reports and small case series, MFA overdose has been shown to cause CNS convulsions and dystonic reactions [19]. In addition, acute toxicity tests in animals, particularly rats and mice, demonstrated that intraperitoneal injection of MFA (beyond the maximum tolerated dose of 20 mg/kg body weight) could induce various adverse neurophysiological effects that occurred in a dose-dependent manner [20]. Animals received 40 mg/kg showed severe muscle spasms with a prominent decrease in locomotor activity, whereas those treated with 80 mg/kg developed fatal seizures. Our earlier study [20] demonstrated that diazepam, a benzodiazepine receptor agonist, significantly reversed CNS convulsions induced by MFA in mice. However, it is unknown whether flumazenil, a benzodiazepine receptor antagonist, can alter the intensity of MFA toxicity in vivo.

The present study investigated MFA toxicity with and without pre-treatment with flumazenil in mice. It was expected from this study that pre-treatment with FMZ could, at least partially, reverse the central adverse effects caused by MFA.

Central nervous system toxicity caused by drug overdose is a common reason for admission to the emergency department. The clinical manifestation of CNS toxicity may include but is not limited to, a change in mental status, disturbance of consciousness, seizures, and movement disorders. The intensity of toxicity varies among patients and depends substantially on drug dose, genetic polymorphism, or a change in drug pharmacokinetics. The pathophysiology of the central effects of drugs is attributed, at least in part, to the ability of these drugs to affect neurotransmitters via synaptic alterations in their synthesis, release, reuptake, degradation, and dynamic interactions with the receptors. In the present study, MFA-treated mice showed an increase in an anxiety state, and catalepsy, and decrease in locomotor activity and CNS convulsions in a dose-dependent manner. These findings were consistent with previous studies findings, which showed an association between MFA overdose and the risk of developing CNS toxicity [15, 21].

Experimental mazes are a common method for evaluating anxiety and cognitive functions in mice [30]. The term “maze” is used to describe a human-made model that is designed in a way that can induce certain behavioral responses in animals. The elevated zero maze apparatus is a typical design for evaluating anxiety behaviors [31]. The apparatus comprises a couple of open and closed quadrants where mice show natural aversion toward the closed quadrants [23]. In this maze, mice display typical traits of anxiety-like behaviors, such as fast resorting to closed quadrants, tendency to remain in the closed area, unwillingness to enter open quadrants, and exhibiting escape behavior manifested by repeated head dipping [32]. Since inhibitory networks of GABAergic interneurons are thought to make up the brain circuits in the amygdala [33], this neurotransmitter is essential in regulating anxiety responses in the EZM test [34]. Through GABAergic neurotransmission, the amygdala is bidirectionally connected to the medial prefrontal cortex (mPC), crucial for controlling amygdala activity associated with processing emotions and anxiety [35].

Nevertheless, numerous additional neurotransmitters, such as serotonin, opioid peptides, endocannabinoids, oxytocin, and corticotropin-releasing hormone, have been associated with modulating anxiety responses in the amygdala [33]. In this study, control mice, which received the vehicle without drugs, spent more than two-thirds of the time in the closed quadrants and showed few entries to the open quadrants. On the other hand, MFA, the test drug, resulted in a much greater amount of time spent in the closed quadrants and a significantly lower number of entries into the open quadrants, indicating that mice that received MFA were more anxious than the control group. Pre-treatment with FMZ significantly reversed the anxiogenic effect of MFA, suggesting the involvement of GABAAR in the development of anxiety associated with MFA treatment.

Behavioral studies have suggested that laboratory mice exhibit an innate burying behavior during the marble burying test [36]. This behavior is believed to reflect repetitive or stereotypic behavior that may have a compulsive nature [37]. Moreover, pharmacological studies have found that GABAergic drugs profoundly affect the marble-burying activity in mice [38]. In this study, MFA-treated mice significantly buried fewer marbles than control mice, suggesting an aberrant burying activity. Pre-treatment with FMZ significantly restored the burying activity, suggesting that GABAAR is involved, at least in part, in the regulation of burying behaviors.

Catalepsy is a neurological disorder characterized by sensory loss, muscle stiffness, postural rigidity and, in certain cases, loss of contact with the environment [39]. Although it is a rare condition, it often occurs as a symptom of other neurological disorders such as epilepsy and Parkinson’s disease, as a side effect of certain medications, or as a reaction to extreme emotions or trauma [36, 40]. The pathophysiology of catalepsy is not fully understood, although the loss of GABA-dopamine harmony has been identified as a causative factor [41, 42]. Previous studies have indicated that harmala alkaloids and related beta-carbolines, inverse agonists in GABAAR, can induce catalepsy in rats [43]. In this study, pre-treatment with FMZ, a selective GABA receptor antagonist, significantly inhibited the development of catalepsy in MFA-treated mice, suggesting involvement of GABA signaling in MFA-induced catalepsy. Moreover, a decrease in marble-burying activity, reduced locomotion, and manifestations of catalepsy-like behavior by MFA may be attributed to the sedative-hypnotic actions associated with GABA-modulating drugs [44].

Drug-induced seizures are a common adverse effect of drug intoxication, accounting for up to 9% of all occurrences of status epilepticus [45]. Convulsive status epilepticus is a serious neurologic condition characterized by prolonged (or recurrent without recovery) tonic-clonic seizures, with an increased risk of morbidity and mortality [46]. A line of recent studies has suggested that MFA is epileptogenic when administrated at overdoses [15, 18, 20, 47]. Although the mechanism underlying the seizures caused by MFA remains unclear, disruption of GABA signaling has been suggested as the main factor [14, 48]. In this study, pre-treatment with FMZ significantly reduced seizure intensity, indicating that GABAAR may be involved in the emergence of seizures following an overdose of MFA. Given that the CNS toxicity of mefenamic acid is related to its modulatory action on beta subunits of GABAR [14], the reverse effect of flumazenil may indicate that it interferes with the dynamic action of MFA on its binding sites on GABAR. On the other hand, diazepam can reduce MFA convulsion [20] by enhancing the binding of GABA to its receptor, which increases chloride ion influx into neurons and leads to hyperpolarized postsynaptic membranes, which exacerbates CNS depression.

Given that GABAR has constitutive activity in downregulating anxiety and CNS convulsions, the present study’s findings may suggest that MFA, at high doses, acts on GABAAR, causing anxiety-like behavior and even CNS convulsions in mice. FMZ can block MFA activity, restoring GABAAR constitutive activity and preventing MFA-induced CNS toxicity. These results suggest that FMZ is a promising agent for managing MFA overdose in clinical practice. The results also suggest that MFA-induced CNS toxicity in mice is a sensitive model for screening drugs and chemicals that act on the GABAAR system.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

QJ and HAl contributed to the experimental design. QJ, RA and HAb performed the research and conducted the behavioral tests in mice. QJ, YJ, RA, KWG and SM analyzed the data. QJ wrote the manuscript. LCM, YJ, KWG, HAl and CA provided critical feedback and reviewed the manuscript. YJ, SM, LCM and CA provided input on data interpretation. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

All animal handling, manipulation and treatment were conducted according to the international guidelines for animal care and use, and approved by Research Ethics Committee, Isra University/Jordan (2019/2018/17-174).

We would like to express our gratitude to all the peer reviewers for their opinions and suggestions.

The authors wish to thank The Deanship of Scientific Research and Postgraduate Studies, Isra University for providing financial support (Grant # 2019/2018/17-174).

The authors declare no conflict of interest.