Toxicology Letters 236 (2015) 34–42

Effect of fipronil on energy metabolism in the perfused rat liver

Hyllana Catarine Dias de Medeiros a, Jorgete Constantin b, Emy Luiza Ishii-Iwamoto b,Fábio Erminio Mingatto a,*a Laboratório de Bioquímica Metabólica e Toxicológica, UNESP – Univ Estadual Paulista, Campus de Dracena, 17900-000 Dracena, SP, Brazilb Laboratório de Oxidações Biológicas, Departamento de Bioquímica, Universidade Estadual de Maringá, 87020-900 Maringá, PR, Brazil

H I G H L I G H T S

� The effects of fipronil on energy metabolism were examined in perfused rat livers.� Fipronil inhibits energy-dependent processes such as gluconeogenesis and ureogenesis.� Mitochondrial respiratory chain was inhibited.� CYP-derived metabolites of fipronil had an important role in the observed effects.

A R T I C L E I N F O

Article history:Received 12 November 2014Received in revised form 26 April 2015Accepted 30 April 2015Available online 2 May 2015

Keywords:GluconeogenesisGlycogenolysisGlycolisisInsecticidesUrea cycleEnergetic metabolism

A B S T R A C T

Fipronil is an insecticide used to control pests in animals and plants that can causes hepatotoxicity inanimals and humans, and it is hepatically metabolized to fipronil sulfone by cytochrome P-450. Thepresent study aimed to characterize the effects of fipronil (10–50 mM) on energy metabolism in isolatedperfused rat livers. In fed animals, there was increased glucose and lactate release from glycogencatabolism, indicating the stimulation of glycogenolysis and glycolysis. In the livers of fasted animals,fipronil inhibited glucose and urea production from exogenous L-alanine, whereas ammonia and lactateproduction were increased. In addition, fipronil at 50 mM concentration inhibited the oxygen uptake andincreased the cytosolic NADH/NAD+ ratio under glycolytic conditions. The metabolic alterations werefound both in livers from normal or proadifen-pretreated rats revealing that fipronil and its reactivemetabolites contributed for the observed activity. The effects on oxygen uptake indicated that thepossible mechanism of toxicity of fipronil involves impairment on mitochondrial respiratory activity, andtherefore, interference with energy metabolism. The inhibitory effects on oxygen uptake observed at thehighest concentration of 50 mM was abolished by pretreatment of the rats with proadifen indicating thatthe metabolites of fipronil, including fipronil sulfone, acted predominantly as inhibitors of respiratorychain. The hepatoxicity of both the parent compound and its reactive metabolites was corroborated bythe increase in the activity of lactate dehydrogenase in the effluent perfusate in livers from normal orproadifen-pretreated rats.

ã 2015 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at ScienceDirect

Toxicology Letters

journal homepage: www.elsev ier .com/locate / tox let

1. Introduction

Fipronil, developed by Rhône-Poulenc Agro in 1987, is apesticide that belongs to the phenylpyrazole chemical group(Tingle et al., 2003). It is an insecticide with widespread use in thecontrol of many agricultural and domestic pests. Fipronil toxicity isattributed to its ability to act at the GABA receptor as anoncompetitive inhibitor of the GABA-gated chloride channels

* Corresponding author. Tel.: +55 18 3821 8200; fax: +55 18 3821 8208.E-mail address: fmingatto@dracena.unesp.br (F.E. Mingatto).

http://dx.doi.org/10.1016/j.toxlet.2015.04.0160378-4274/ã 2015 Elsevier Ireland Ltd. All rights reserved.

of neurons in the central nervous system. Impediment of the influxof the chloride ions affects the transmission of nervous impulses,causing insect death by neuronal hyperexcitation and paralysis(Rhône-Poulenc, 1995; Zhao et al., 2004).

Fipronil binding is stronger to the chloride channels of insectsthan to those of mammals, resulting in an insecticide withselective toxicity. Thus, fipronil has a greater ability to blockGABA-gated Cl� channels of insects than those of vertebrates andis therefore considered safe and is widely used in veterinarymedicine (Hainzl and Casida, 1996; Hainzl et al., 1998; Coutinhoet al., 2005; Gunasekara and Troung, 2007). However, Zhao et al.(2005) indicate that the metabolite fipronil sulfone, which results

H.C.D. de Medeiros et al. / Toxicology Letters 236 (2015) 34–42 35

from the hepatic biotransformation of fipronil via cytochromeP450, is at least 20 times more potent to block mammalian GABAA

receptors.There are several cases in the literature of animal and human

poisoning due to intentional ingestion, accidental exposure orincorrect use of fipronil (Gasmi et al., 2001; Jennings et al., 2002;Chodorowski and Anand, 2004; Mohamed et al., 2004; Lee et al.,2010; Anadon and Gupta, 2012; Gill and Dumka, 2013). Silva(2008) evaluated the effects of prolonged fipronil exposure in ratsand observed hepatic cell swelling and increased liver weights inanimals treated with a 10 mg/kg dose administered by oralgavage. This dose corresponds to one tenth of the LD50 in the ratsestablished by Hainzl and Casida (1996). The hepatotoxicmechanisms of fipronil are unknown. Vidau et al. (2011) showedthat fipronil exerts an uncoupling effect in isolated rat livermitochondria. In another study, Palma et al. (2013) demonstratedthat fipronil exerts an inhibitory effect on the electron transportchain, specifically in complex I. Both studies presented aconcentration-dependent effect of fipronil starting at a concen-tration of 5 mM.

The liver is the central organ in metabolism because it isinterposed between the digestive tract and the general circula-tion. Among the main liver functions is the uptake of amino acids,lipids, carbohydrates and vitamins, with subsequent storage,metabolic conversion and release into the blood, as well as theproduction of bile. The liver is also capable of accumulating,biotransforming and inactivating many xenobiotics, convertingthem into water-soluble substances and thus facilitating theremoval of these compounds by the organism (Guillouzo, 1998).However, this process has been considered responsible for thetoxic effects of many chemicals because the produced metabolitescan exert adverse effects on the organism (Ioannides and Lewis,2004; Mingatto et al., 2008).

For this reason, based on the importance of the liver to theanimal organism and the evidence that fipronil exerts action onisolated mitochondria, the present study was planned to investi-gate the effects of fipronil in the perfused rat liver, a methodologyin which the structural and functional integrity of the organ ispreserved. Fipronil was infused into the livers at concentrationsranging from 10 to 50 mM, and several parameters related to theenergy metabolism were measured in the perfused effluent,including glycogenolysis, glycolysis and oxygen uptake in the liversof fed rats, as well as gluconeogenesis and ureagenesis in the liversof fasted rats. In order to investigate whether fipronil metabolitesare implicated in the metabolic alterations, rats were previouslytreated with proadifen, a well-known cytochrome P-450 inhibitor.

2. Materials and methods

2.1. Chemicals

The liver perfusion apparatus was built in the workshops of theUniversity of Maringá. Enzymes and coenzymes used in theenzymatic assays were purchased from Sigma–Aldrich (St. Louis,MO, USA). Fipronil was a gift from Ourofino Agribusiness,containing 96.6% purity (Cravinhos, SP, Brazil). All other reagentswere of the highest commercially available grade.

2.2. Animals

Male albino rats (Wistar) weighing 180–220 g were housed inplastic cages at a constant temperature (23 � 3 �C) and relativehumidity (55 �15%) under a regular light/dark cycle (12 h:12 h).They were fed ad libitum with a standard laboratory diet(Nuvilab1, Colombo, Brazil). The experimental protocols wereapproved by the Ethical Committee for the Use of Laboratory

Animals of the UNESP – Univ Estadual Paulista, Campus ofDracena, SP, Brazil.

All experiments were started between 7:00 AM and 8:00 AM.Rats used for studies of gluconeogenesis were fasted 24 h beforethe experiments to deplete the livers of glycogen. In someexperiments rats were pretreated with proadifen (25 mg kg bodyweight�1), a cytochrome P450 inhibitor, intraperitoneally for 3consecutive days (Somchit et al., 2009).

2.3. Liver perfusion

For the surgical procedure, fed or 24 h fasted rats wereanesthetized by intraperitoneal injection of sodium pentobarbital(50 mg kg body weight�1). Hemoglobin-free, non-recirculatingperfusion was performed. The surgical technique was the same asthat described by Scholz and Bucher (1965). After cannulation ofthe portal and cava veins, the liver was positioned in a Plexiglaschamber. The perfusion fluid was Krebs/Henseleit-bicarbonatebuffer (pH 7.4), saturated with a mixture of oxygen and carbondioxide (95:5) using a membrane oxygenator with simultaneoustemperature adjustment at 37 �C. The flow, provided by aperistaltic pump, was between 30 and 35 ml min�1, dependingon the liver weight. Samples of the effluent perfusion fluid werecollected according to the experimental protocol and analyzed fortheir metabolite contents. The oxygen concentration in theoutflowing perfusate was monitored continuously using aTeflon-shielded platinum electrode adequately positioned in aPlexiglas chamber at the point where the perfusate exits (Scholzand Bucher, 1965). Fipronil (10, 15, 25 and 50 mM) or L-alanine(2.5 mM) was dissolved in the perfusion fluid. The doses of fipronilwere selected on the basis of the previous results using liver cells(Das et al., 2006) and isolated rat liver mitochondria (Vidau et al.,2011; Palma et al., 2013). Control experiments without fiproniladdition were previously performed and showed that concen-trations of metabolites in the perfusate and oxygen uptake did notchange during the experimental time period, implying that theperfused liver was metabolically stable for the duration of theperfusion used in this study.

2.4. Analytical

The following compounds were assayed using standardenzymatic procedures: glucose (Bergmeyer and Bernt, 1974), L-lactate (Gutmann and Wahlefeld, 1974), pyruvate (Czok andLamprecht, 1974), urea (Bergmeyer, 1974) and ammonium (Kunand Kearney, 1974). Metabolic rates were calculated from input–output differences and the total flow rates, normalized to the wetweight of the liver, and expressed as mmol min�1 (gram liver wetweight)�1. The activity of the enzyme lactate dehydrogenase (LDH)was measured in the perfusate using an Assay Kit (Bioclin, Quibasa,Brazil) according to the manufacturer’s instructions.

2.5. Treatment of the data

The data in the figures are expressed as mean � standard errorof the mean (S.E.M.) of 3–5 liver perfusion experiments. Thestatistical significance of the differences between parametersamong the experimental groups was evaluated using two-wayanalysis of variance, and differences in the same experimentalgroups were tested by repeated-measures one-way analysis ofvariance (ANOVA). Significant differences among means wereidentified by Bonferroni or Newman–Keuls testing, respectively.The results are given in the text as probability values (P).P � 0.05 was adopted as the criterion of significance. Statisticalanalysis was performed using StatisticaTM or GraphPAD Softwareprograms.

36 H.C.D. de Medeiros et al. / Toxicology Letters 236 (2015) 34–42

3. Results

3.1. Effects of fipronil on glycogen catabolism and oxygen uptake

Livers from fed rats were perfused with substrate-free perfusionmedium in a non-recirculating mode. Under these conditions,energy for liver function is derived primarily from fatty acidoxidation, but it simultaneously exhibits extensive glycogenolyticand glycolytic activity (Scholz and Bucher, 1965), as indicated byglucose, L-lactate and pyruvate release. Fig. 1A shows the changes inmetabolic fluxes caused by the infusion of fipronil (50mM) for40 min. Glucose release and lactate production progressivelyincreased due to insecticide presence, becoming statisticallysignificant after 22 and 20 min of infusion, respectively, and reachinga new steady state; upon the cessation of fipronil infusion, the ratesquickly returned to baseline values. Oxygen uptake was significantlyaffected by the presence of fipronil, with significant inhibitionoccurring after 18 minof infusion; even after insecticide removal, theinhibition remained stable for another 10 min and then graduallydecreased to levels lower than baseline. Regarding pyruvate, nochange was observed during the experimental period. Assays such asthat shown in Fig. 1A were repeated with different fipronilconcentrations and are summarized in Fig. 1B. The control values(without fipronil) correspond to the basal rates obtained beforefipronil infusion (infusion for 6–10 min).

Parameters related to pyruvate production did not change withincreasing fipronil doses. Oxygen uptake showed mild increaseswith fipronil concentrations of 10 and 15 mM (5.9% and 5.3%,respectively); with the concentrations of 25 and 50 mM, however,there was a significant decrease in oxygen uptake compared withthe control (16.1% and 19.7%, respectively). Different behavior wasobserved in glucose production; in the first two concentrations, adecrease of 18.4% and 15% occurred, respectively. With the twohighest fipronil concentrations, glucose production was stimulat-ed, reaching a significant increase (103.3%) with the infusion of a50 mM dose of fipronil. Lactate production increased in aconcentration-dependent manner; however, the increase was

Fig. 1. Effects of fipronil on metabolic fluxes in the livers of fed rats. Panel A: Time couroxygen uptake. Fipronil was infused from 10 to 50 min, as indicated in the horizontal barswas polarographically measured. Panel B: The action of different fipronil concentratiovalues � S.E.M of 3–5 experiments. Hashtags indicate significant differences comparedrepeated-measures ANOVA testing (#P � 0.05). Asterisks indicate significant differences

post hoc Newman–Keuls test (**P � 0.01 and ***P � 0.001).

only significant with fipronil concentrations of 25 mM (83%) and50 mM (156.5%).

The total glycogenolysis (Fig. 2A) was calculated by theequation glucose + 1/2 (lactate + pyruvate) released on the basisof experiments showed in Fig. 1A. According to Kimmig et al.(1983), this provides a good representation of the glycogenolyticactivity under the conditions used in this experiment. Fig. 2Aestablishes a comparison between the control series (withoutfipronil) and different fipronil concentrations at the end of infusion(46–50 min experimental period). The rate of glycogen breakdownwas found to progressively increase with increasing fipronilconcentrations. Compared with the control, the values increasedby 0.53%, 12.5%, 32.5% and 157.6% for 10, 15, 25 and 50 mM,respectively.

The rate of glycolysis, as presented in Fig. 2A, was alsocalculated from the experiments shown in Fig. 1A (lactate +pyruvate) with different fipronil concentrations. The control valuewas compared with the values found at the end of fipronil infusion(and average of 46–50 min of perfusion). The glycolysis rates alsobehaved in a concentration-dependent manner; thus, the ratesincreased with increasing fipronil concentrations and weresignificantly greater than the control at 15 mM (41%), 25 mM(56.2%) and 50 mM (181%) concentrations. The lactate to pyruvateratio, which is an indicator of the cytosolic NADH/NAD+ ratio, wassignificantly increased at 50 mM fipronil (Fig. 2B).

3.2. Effects of fipronil on gluconeogenesis, the urea cycle and oxygenuptake

In order to determine the effects of fipronil on energy-dependent biosynthetic processes, gluconeogenesis, ureogenesis,oxygen uptake, ammonium level, and L-lactate and pyruvateproduction from L-alanine were measured in perfused livers offasted rats. L-Alanine is as gluconeogenic substrate found in highconcentration in bodily fluids, which presents a plasma concen-tration of about 0.4 mM in fasted rats (Mann et al.,1988). Therefore,in the present study a supra-physiological concentration of

se of the effects of 50 mM fipronil on glucose, lactate and pyruvate production and. Perfusate samples were collected for metabolite measurements, and oxygen uptakens (10–50 mM) on the livers after 40 min of perfusion. Data represent the mean

with values obtained immediately before fipronil infusion (Panel A), according tocompared with the control period (without fipronil) by analysis of variance with the

Fig. 2. Effects of different fipronil concentrations (10–50 mM) on glycogenolysisand glycolysis (Panel A) and lactate to pyruvate ratio (Panel B) in the livers from fedrats after 40 min of infusion. Data were calculated based on the experimentspresented in Fig. 1A. Data represent the mean values � S.E.M of 3– 5 experiments.Asterisks indicate significant differences compared with the control period(without fipronil) by analysis of variance with the post hoc Newman–Keuls test(**P � 0.01 and ***P � 0.001).

H.C.D. de Medeiros et al. / Toxicology Letters 236 (2015) 34–42 37

L-alanine (2.5 mM) was used, a condition that enhances the rate ofgluconeogenesis as previously described (Mallette et al., 1969).

The results of the experiments are illustrated in Fig. 3A. In thisseries of experiments, perfusion medium alone was infused for the

Fig. 3. Effects of fipronil on metabolic fluxes in the livers of fasted rats. Panel A: Time coAlanine was infused from 10 to 90 min, and fipronil was infused from 30 to 70 min, asmeasurements, and oxygen uptake was polarographically measured. Panel B: Actions ofThe control values correspond to the parameters found in the presence of L-alanine beforeHashtags indicate significant differences compared with values obtained immediately be� 0.05). Asterisks indicate significant differences compared with the control period ((*P � 0.05and ***P � 0.001).

first 10 min. During this period, the amounts of glucose, lactate andpyruvate released were very small due to low hepatic glycogenconcentrations. After the 10-min pre-perfusion period, 2.5 mM L-alanine was infused for 80 min. After the onset of L-alanineinfusion, glucose, lactate and pyruvate production progressivelyincreased. After approximately 10 min, steady-state conditionswere attained. The infusion of 50 mM fipronil at 30 min causedglucose concentrations to progressively decrease becomingstatiscally significant after 10 min and remaining until the endof the experiment. Pyruvate concentration slightly increased beingstatistically significant only between 20 and 24 min during fipronilinfusion while lactate considerably increased, becoming statisti-cally significant 6 min after the infusion of fipronil and remainedsignificantly changed until 10 min after the insecticide removal.

Fig. 3B shows an assessment of the effect of different fipronilconcentrations on glucose, lactate and pyruvate production. Thecontrol values correspond to the average rates found in thepresence of 2.5 mM L-alanine, prior to fipronil infusion(26–30 min). The results for fipronil were presented dependingon the average values observed at the end of the infusion period(66–70 min of perfusion). Glucose production was alreadysignificantly inhibited at the lower fipronil concentration(10 mM), and the highest inhibitory effect occurred at theconcentration of 25 mM. Pyruvate production increased followingtreatment with 10, 15 and 25 mM fipronil (40.7%, 27.5% and 13.6%,respectively). However, the 50 mM concentration caused a signifi-cant reduction in pyruvate production (41.9%). In the first threeinsecticide concentrations, similar increases were observed inlactate production (19.9%, 28% and 24.2%, respectively); however,at the 50 mM concentration, there was a significant increase of 86%compared with the control.

Fig. 4A shows the oxygen uptake and urea and ammoniaproduction in the liver of fasted rats subjected to perfusion of50 mM fipronil. The control values (no fipronil) correspond to therates founded in the presence of L-alanine just before the onset offipronil infusion (26–30 min of perfusion). Without exogenoussubstrate (first 10 min), the basal rate of ammonia production was0.156 mmol min�1 g�1 and urea production was 0.308 mmol min�1

g�1. Upon beginning substrate infusion (L-alanine), oxygen uptakeand urea production progressively increased and then stabilizedafter 25 min, whereas ammonia production immediately increasedand remained constant until fipronil was introduced. In

urse of the effects of 50 mM fipronil on glucose, lactate and pyruvate production. L- indicated in the horizontal bars. Perfusate samples were collected for metabolite

different fipronil concentrations (10–50 mM) on the liver after 40 min of perfusion. the infusion of fipronil. Data represent the mean values � S.E.M of 3–5 experiments.fore fipronil infusion (Panel A), according to repeated-measures ANOVA testing (#Pwithout fipronil) by analysis of variance with the post hoc Newman–Keuls test

Fig. 4. Effects of fipronil on metabolic fluxes in the liver from rats fasted for 24 h. Panel A: Time course of the effects of 50 mM fipronil on ammonia and urea production andoxygen uptake. L-Alanine was infused from 10 to 90 min, and fipronil was infused from 30 to 70 min, as indicated in the horizontal bars. Perfusate samples were collected formetabolite measurements, and oxygen uptake was polarographically measured. Panel B: Actions of different fipronil concentrations (10–50 mM) on the liver after 40 min ofperfusion. The control values correspond to the parameters found in the presence of L-alanine before fipronil infusion. Data represent the mean values � S.E.M of 3–5 experiments. Hashtags indicate significant differences compared with values obtained immediately before fipronil infusion (panel A), according to repeated-measuresANOVA testing (#P � 0.05). Asterisks indicate significant differences compared with the control period (without fipronil) by analysis of variance with the post hoc Newman–Keuls test (*P � 0.05 and ***P � 0.001).

38 H.C.D. de Medeiros et al. / Toxicology Letters 236 (2015) 34–42

thepresence of the fipronil, urea production was reduced becomingstatistically significant after 12 min of infusion of the insecticide andremaining until the end of the experiment. Oxygen uptake also wasreduced after introduction of fipronil; however, oxygen uptake againincreased reaching the control values after approximately 70 min ofperfusion and reducedagain, becoming statisticallysignificant 6 minafter removal of fipronil and remaining until the end of the perfusion.Ammonia production was increased becoming statistically signifi-cant 18 min after fipronil infusion and remained significantlychanged until the end of the experiment.

Fig. 4B shows the action of different fipronil concentrations(10–50 mM) infused in experiments similar to those presented inFig. 4A. Fipronil introduction significantly inhibited urea produc-tion in a concentration-dependent manner (26.4%, 52.7%, 70.9%

Fig. 5. Concentration dependence of the actions of fipronil on gluconeogenesis from L-asimilar experiments to those illustrated in Figs. 3 and 4 but with the use of 2.5 mM L-alanrates found in the presence of L-alanine just before the onset of fipronil infusion (30 mi70 min of perfusion. Asterisks indicate significant differences compared with the controtest (*P � 0.05 and ***P � 0.001).

and 86.6%, respectively, for 10, 15, 25 and 50 mM). Ammoniaproduction also significantly increased in a concentration-depen-dent manner, increasing 147% with the lower concentration and297% with the highest concentration of insecticide. Oxygen uptakewas significantly inhibited with concentrations of 15 and 25 mMfipronil (29.4% and 25.1%, respectively).

Fig. 5 shows the average results of four assays to evaluategluconeogenesis in a similar manner to what is shown inFigs. 3 and 4A, but with different fipronil concentrations (10, 15,25 and 50 mM). The infusion of insecticide, even at lowerconcentrations, caused a significant and concentration-dependentinhibition of glucose release. Oxygen uptake was significantlyinhibited only with the intermediate concentrations.

lanine and oxygen uptake in the livers of fasted rats. The data were obtained fromine as a gluconeogenic substrate. The control values (no fipronil) correspond to then perfusion time). Rates in the presence of L-alanine + fipronil were evaluated afterl period (without fipronil) by analysis of variance with the post hoc Newman–Keuls

H.C.D. de Medeiros et al. / Toxicology Letters 236 (2015) 34–42 39

3.3. Effects of proadifen on fipronil-induced alterations on glycogencatabolism and oxygen uptake

To evaluate the influence of the biotransformation on theeffects of fipronil in the carbohydrate catabolism, fed rats werepretread with proadifen. Fig. 6 shows the changes in metabolicfluxes caused by the infusion of fipronil (50 mM) for 40 min.Glucose release and lactate production progressively increased dueto insecticide presence, becoming statistically significant after 30and 18 min of infusion, respectively, and remained elevated evenupon the cessation of fipronil infusion. No change was observed inpyruvate production during the experimental period. Oxygenuptake was reduced during the infusion of fipronil, but withoutstatistical significance. After the terminus of fipronil infusion,oxygen uptake slowly recovered to values similar to those observedbefore fipronil infusion.

To facilitate comparisons between the effects of fipronil(50 mM) in normal and proadifen-pretreated rats, the changes inthe rates of glucose, lactate, pyruvate and oxygen uptake thatfollowed the fipronil infusion in the experiments of Fig. 1 (controlrats) and Fig. 6 (proadifen-pretreated rats) are summarised inFig. 7. The values correspond to the mean values of the metabolicfluxes after the onset of fipronil infusion subtracted from the basalrates (i.e., the values measured at 10 min of perfusion time). All themetabolic fluxes changes caused by fipronil infusion were higher inlivers from proadifen-treated rats than in those of control rats. Thedifferences were progressively accentuated in the final period andafter the cessation of fipronil infusion. At the terminus ofexperimental time period (70 min of perfusion) the rates ofglucose (Panel A), lactate (Panel B) and pyruvate (Panel C)productions and the oxygen uptake (Panel D) in the proadifen-treated rats remained significantly elevated (60.5%, 41.6%, 34.3%

Fig. 6. Effects of fipronil on metabolic fluxes in the livers of fed rats pretreated withproadifen. Time course of the effects of 50 mM fipronil on glucose, lactate andpyruvate production and oxygen uptake. Fipronil was infused from 10 to 50 min, asindicated in the horizontal bars. Perfusate samples were collected for metabolitemeasurements, and oxygen uptake was polarographically measured. Data representthe mean values � S.E.M of 3–5 experiments. Hashtags indicate significantdifferences compared with values obtained immediately before fipronil infusion,according to repeated-measures ANOVA testing (#P � 0.05).

and 50.3%, respectively) in comparison with the rates found inuntreated control rats.

3.4. Hepatotoxicity of fipronil

The activity of LDH in the perfusate released from livers of fedrats was measured to find if fipronil caused hepatotoxic effect bothin normal versus proadifen pre-treated rats. In the perfusate ofliver of normal rats, only small activity was found in the first 10 minof perfusion (Fig. 8). With the addition of 50 mM fipronil,concentration that presented the most significant effects in thepresent study, the activity of LDH significantly increased, remain-ing elevated even after the removal of the insecticide. Thepretreatment of the rats with proadifen did not result in differenceof LDH release compared to livers from normal rats.

4. Discussion

The results of the present study showed that fipronil affectsenergy-linked hepatic metabolism. The insecticide was able toactivate glycogenolysis (glucose production) and glycolysis (theproduction of lactate over pyruvate) in the livers of fed animals andinhibit gluconeogenesis (glucose production from L-alanine) andureagenesis (urea production) in the livers of fasted animals.Additionally it changed the redox state of the cytosolic NAD+–

NADH couple, since increased NADH/NAD+ ratios were observedunder glycolytic conditions (Scholz and Bucher, 1965). The reducedoxygen uptake observed during fipronil infusion is a strongindicator that the speed of electron flow in the mitochondrialrespiratory chain was reduced, thereby compromising thephosphorylation of ADP and all energy-dependent cellularprocesses such as gluconeogenesis and ureagenesis (Colturatoet al., 2012). On the other hand, significant increases in lactateproduction from glucose (glycolysis) and glycogenolysis in thelivers of fed animals are features of the so-called Pasteur effect,which involves the increase in anaerobic glycolysis to generate ATPand to increase the energy supply as a compensatory mechanismfollowing the reduction of mitochondrial oxidative phosphoryla-tion (Dickman and Mandel, 1990; Williams et al., 2007).

Our results indicate that the effect of fipronil in isolatedmitochondria reported by Palma et al. (2013), in which fipronilinhibits the mitochondrial respiratory chain on complex 1, alsooccurs in perfused rat livers. When infused into perfused rat livers,classical respiratory chain inhibitors, such as KCN or antimycin,inhibit respiration, gluconeogenesis and ureagenesis and stimulateglycogenolysis and glycolysis (Younes and Strubelt, 1988; Con-stantin et al., 1995; Pagadigorria et al., 1996). Therefore, it seemslikely that the main mechanism of the cytotoxic effect of fipronil inthe rat liver is the inhibition of mitochondrial energy metabolism.

There are numerous consequences of the use of inhibitors inanimals. It is well known that liver glycogenolysis and gluconeo-genesis are the major sources of circulating glucose and that theimpairment of these processes can have negative consequences onother organs (Ruderman, 1975). Hepatic glycogen is postprandiallyreplenished during the absorptive period. As fasting progresses,when blood glucose is lower, a decrease occurs in the insulin:glucagon ratio, thus stimulating glycogenolysis to maintain theblood glucose levels (Park and Exton, 1972).

Gluconeogenesis is triggered by decreased liver glycogenreserves to maintain circulating blood glucose levels, made fromnon-carbohydrate compounds with an ATP cost (Ruderman, 1975).Fipronil significantly inhibited gluconeogenesis from L-alanine in aconcentration-dependent manner, likely due to its inhibitory effecton oxidative phosphorylation. Thus, it was expected thatgluconeogenesis would be compromised, independent of thepresence of gluconeogenic substrate. The inhibition of this process

Fig. 7. Changes on metabolic fluxes caused by 50 mM fipronil infusion in the livers from fed rats pretreated with proadifen (&) or untreated-control rats (&). Fipronil wasinfused from 10 to 50 min, as indicated in the horizontal bars. Data were calculated based on the experiments presented in Figs. 1 and 6, and represent the mean values � SEMof the values of metabolite productions subtracted from the rates before fipronil infusion. Data represent the mean values � S.E.M of 3–5 experiments. Significant differencesin the time course of glucose (Panel A), lactate (Panel B) and pyruvate (Panel C) productions and on oxygen uptake (Panel D) of proadifen-pretreated rats as compared to thecontrol (no treated) rats are indicated by horizontal lines (two-way analysis of variance, with post hoc Bonferroni test, P � 0.05).

40 H.C.D. de Medeiros et al. / Toxicology Letters 236 (2015) 34–42

is extremely important, especially for cattle; these animals obtainmost of their required glucose through gluconeogenesis, whichoccurs mainly during lactation, when the demand for this hexose isgreatly increased. Moreover, just a small fraction of all glucoserequired by ruminants is absorbed in the gastrointestinal tract(Lindsay, 1970). Ureagenesis inhibition in mammals is alsoextremely important because the urea cycle is well known asthe main pathway for removing potentially toxic ammonium from

0 10 20 30 40 50 60 70

0

10

20

30

Proad ifenNo Proad ifen

*

****

****

**

*** ** ** ** **

Fipronil infusion (50 µM)

Perfusion time (minu tes)

LD

H A

ctiv

ity (

U/L

)

Fig. 8. Time courses of lactate dehydrogenase (LDH) release into the outflowingperfusate elicited by the infusion of 50 mM fipronil. Livers from fed rats pretreatedor no with proadifen were perfused with substrate-free perfusion medium asdescribed under Section 2. Fipronil was infused from 10 to 50 min, as indicated inthe horizontal bars. Samples of the effluent perfusion fluid were collected forenzyme assay. Data � mean standard errors are from 3–5 liver perfusionexperiments. Asterisks indicate significant differences compared with the controlperiod (without fipronil) by analysis of variance with the post hoc Newman–Keulstest (*P � 0.05 and **P � 0.01).

the blood of ruminants, as well as of other mammals (Huntingtonand Archibeque, 1999).

The liver is able to metabolize fipronil (Tang et al., 2004) andtherefore the question arises whether the parent compound or itsmetabolites were the causative agents for the alteration seen in theenergy metabolism in the perfused liver. The biotransformation offipronil occurs via cytochrome P450; there is no information in theliterature about which rat CYP isozymes metabolize fipronil,however, it is known that CYP3A4 is the major isoform responsiblefor fipronil oxidation in humans while CYP2C19 is considerably lessactive. From the S-oxidation of fipronil, fipronil sulfone arises (Tanget al., 2004). Das et al. (2006) showed that the biotransformation offipronil does not eliminate its toxicity. Fipronil and fipronil sulfoneat micromolar concentrations induced cell death in HepG2 cellsand human hepatocytes. Ferreira et al. (2012) demonstrated theoccurrence of hypertrophy in the liver tissue of rats treated withfipronil using histological techniques. According to the authors,this hypertrophy was likely due to the increased number oforganelles involved in the degradation of toxic products, mainlythe endoplasmic reticulum, which is able to synthesize isoforms ofcytochrome P450.

This biotransformation process is relatively slow; in this study,the effects of fipronil observed occurred within the first 40 min ofinfusion, and metabolic changes were observed in first minutes ofinfusion of fipronil, suggesting that the acute effects are likelycaused by fipronil itself. However, the metabolites generated mighthave contributed to the observed effects, especially at the end ofthe infusion period. This hypothesis is supported by the slowrecovery of metabolic fluxes after fipronil infusion. Someparameters were irreversible, including glucose production fromL-alanine in fasted animals and oxygen uptake in fed rats. This

H.C.D. de Medeiros et al. / Toxicology Letters 236 (2015) 34–42 41

phenomenon may represent the accumulation of fipronil or itsactive metabolites in cellular compartments.

The inhibitory effect on oxygen uptake observed in livers of fedanimals that was gradually accentuated during the period of50 mM fipronil infusion and remained reduced even after theterminus of fipronil infusion, may be caused by an action of fipronilmetabolites. This interpretation is corroborated by observationthat the pretreatment of rats with proadifen, an inhibitor ofcytochrome P450 isoforms (Somchit et al., 2009; Shi et al., 2011)eliminated the inhibitory effect on oxygen uptake. Along theselines, in experiments using isolated mitochondria, Palma et al.(2013) showed that fipronil inhibits the electron transport chain,specifically in complex I, and Vidal et al. (2011) demonstrated thatfipronil also acts as an uncoupler. Both studies used fipronilconcentrations between 5 and 25 mM. According to these studies,the fipronil cytotoxicity observed in Caco-2 and SHSY5Y cells couldbe related to the ability of fipronil to affect the commitment toenergy production of cells (Vidau et al., 2009, 2011). In studiesconducted simultaneously with isolated mitochondria, we foundthat the metabolite fipronil sulfone at concentrations between0.5 and 5 mM was able to uncouple mitochondria, and acted as arespiratory chain inhibitor in concentrations ranging from 5 to25 mM (unpublished data).

The predominat effect of fipronil on oxygen uptake in liversfrom fed rats was of activation at low concentrations and ofinhibition at higher concentration. This probably reflected the dualeffects of fipronil and its metabolites as uncoupler or as inhibitor ofthe respiratory chain. That the products of fipronil metabolismacted predominantly as inhibitors of the respiratory chain wasfurther supported by the increased in the ratio between lactate andpyruvate only at a 50 mM concentration when the level of fipronilmetabolites was possibly higher. This is an expected consequenceof an inhibition of NADH reoxidation by the respiration chain(Scholz and Bucher, 1965).

It seems clear, however, that both the parent compound and itsmetabolites contributed to impairment of energy metabolism inthe perfused livers since the stimulation of glucose and lactateproduction was further accentuated by treatment in proadifen-treated rats. In accordance to this, it was found a similar increase inLDH release into the perfusion medium under 50 mM fipronilinfusion in both livers from normal and proadifen-treated rats,confirming that the parent compound and its reactive metabolitesulfone can cause cellular damage, as previously demonstrated(Das et al., 2006).

The effects of fipronil found in literature indicate that it inducesapoptotic cell death. Liver cells exposed to low concentrations offipronil (0.1 mM), in addition to having decreased ATP levels,displayed the activation of enzymes involved in the apoptoticprocess, caspases 3 and 7 (Das et al., 2006). Histological studiesalso showed that the nuclei of hepatocytes were disorganized andmodified, with chromatin condensation and marginalization,indicating that the cells were undergoing apoptotic cell death(Ferreira et al., 2012). In intact livers, fipronil exerted significanteffects at higher concentrations than those reported in isolatedmitochondria or hepatocytes. This is a common phenomenon,considering that a higher arterial concentration of insecticidewould be necessary to reach the desired intracellular concentra-tion in the target organelles or enzymes. Accordingly, Birckel et al.,1997 detemined that fipronil and its sulfone metabolite werehighly bound to plasma protein, with a percentage fixation ofapproximately 98% with predominant interaction with serumalbumin, so only a free portion of the compounds will be availableto cells and tissues.

Cid et al. (2012)subcutaneously administered a single dose of1 mg/kg in cattle; they found a peak concentration of approxi-mately 1 mM fipronil in the blood plasma 10 h after

administration. In another study, Leghait et al. (2009) used adose of 3 mg/kg/day in rats for 14 or 28 days and found plasmafipronil levels of less than 0.2 mM; however, concentrations of thesulfone metabolite of approximately 20 times higher were found.In addition, Guo-Xin et al. (2006) found plasma concentrations ofapproximately 8 mM fipronil and 2.3 mM fipronil sulfone afterintravenous injection of a single dose of 3 mg/kg fipronil to rabbitsand also concluded that the half life of fipronil sulfone is longerthan that of fipronil. Although these plasma concentrations arelower than the concentrations used in the present study we haveto consider that due its portal location within the circulation andits central role in xenobiotic metabolism, the liver is exposed tohigh concentrations of xenobiotic/metabolite(s), making theorgan particularly susceptible to damage. Therefore, despite theuse of the recommended dose of fipronil seems to be safe, theapplication of excessive or prolonged fipronil doses has beenobserved to cause damage to animals and humans exposed to theinsecticide. Because the inhibition of energy metabolism mayaffect several processes important for cell maintenance, it isreasonable that the energy-linked liver metabolic changesdescribed in the present work might contribute to fipronilhepatotoxicity.

Conflict of interest

The authors declare that there are no conflicts of interest.

Acknowledgement

The authors thank Fundação de Amparo à Pesquisa do Estado deSão Paulo (FAPESP) by grants (Process number 2012/15135-6) andfor providing the M.Sc. scholarship (Process number 2012/21869-2). The authors also thank Aparecida Pinto Munhoz Hermoso forthe technical assistance. The results were presented by H.C.D.Medeiros to the Faculdade de Medicina Veterinária de Araçatuba,Universidade Estadual Paulista “Júlio de Mesquita Filho” in partialfulfillment of the requirements for a Master degree in CiênciaAnimal.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.toxlet.2015.04.016.

References

Anadon, A.N., Gupta, R.C., 2012. Fipronil. In: Gupta, R.C. (Ed.), Veterinary Toxicology:Basic and Clinical Principles. Academic Press/Elsevier, Amsterdam, pp. 604–608.

Bergmeyer, H.U., 1974. Determination of urea with glutamate dehydrogenase asindicator enzyme. In: Bergmeyer, H.U. (Ed.), Methods of Enzymatic Analysis.Academic Press, New York, pp. 1794–1801.

Bergmeyer, H.U., Bernt, E., 1974. Determination of glucose with glucose oxidase andperoxidase. In: Bergmeyer, H.U. (Ed.), Methods of Enzymatic Analysis. AcademicPress, New York, pp. 1205–1215.

Birckel, P., Weil, A., Galtier, P., Alvinerie, M., 1997. Plasma protein binding of fiproniland its sulfone metabolite in the dog. J. Vet. Pharmacol. Ther. 20 (Suppl. 1),79–80.

Chodorowski, Z., Anand, J.S., 2004. Accidental dermal and inhalation exposure withfipronil – a case report. J. Toxicol. 42 (2), 189–190.

Cid, Y.P., Ferreira, T.P., Medeiros, D.M.V.C., Oliveira, R.M., Silva, N.C.C., 2012.Determination of fipronil in bovine plasma by solid-phase extraction and liquidchromatography with ultraviolet detection. Quim. Nova 35, 2063–2066.

Colturato, C.P., Constantin, R.P., Maeda Junior, A.S., Constantin, R.P., Yamamoto, N.S.,Bracht, A., Ishii-Iwamoto, E.L., Constantin, J., 2012. Metabolic effects of silibininin the rat liver. Chem. Biol. Interact. 195, 119–132.

Constantin, J., Iwamoto, E.L.I., Kemmelmeier, F.S., Yamamoto, N.S., Bracht, A., 1995.The action of glucagon infused via the hepatic artery in anterograde andretrograde bivascular perfused of rat liver is not a function of the accessiblecellular spaces. Biochim. Biophys. Acta 1244, 169–178.

42 H.C.D. de Medeiros et al. / Toxicology Letters 236 (2015) 34–42

Coutinho, C.F.B., Tanimoto, S.T., Galli, A., Garbellini, G., Takayama, M., Amaral, R.B.,Mazo, L.H., Avaca, L.A., Machado, S.A.S., 2005. Pesticidas: mecanismo de ação:degradação e toxidez. Pesticidas: Revista de Ecotoxicologia e Meio Ambiente 15,65–72.

Czok, R., Lamprecht, W., 1974. Pyruvate, phosphoenolpyruvate and D-glycerate2-phosphate. In: Bergmeyer, H.U. (Ed.), Methods of Enzymatic Analysis.Academic Press, New York, pp. 1446–1451.

Das, P.C., Cao, Y., Cherrington, N., Hodgson, E., Rose, R.L., 2006. Fipronil induces CYPisoforms and cytotoxicity in human hepatocytes. Chem. Biol. Interact. 164,200–214.

Dickman, K.G., Mandel, L.J., 1990. Differential effects of respiratory inhibitors onglycolysis in proximal tubules. Am. J. Physiol. 258, 1608–1615.

Ferreira, M., Oliveira, P.R., Denardi, S.E., Bechara, G.H., Mathias, M.I.C., 2012. Action ofthe chemical agent fipronil (active ingredient of acaricide Frontline) on the liverof mice: an ultrastructural analysis. Microsc. Res. Techniq. 75, 197–205.

Gasmi, A., Chataigner, D., Gamier, T., Lagier, G., 2001. Toxicity of fipronil-containinginsecticides. Report of 81 cases from the Paris Poison Center. Vet. Hum. Toxicol.43, 247.

Gill, K.K., Dumka, V.K., 2013. Biochemical alterations induced by oral subchronicexposure to fipronil, fluoride and their combination in buffalo calves. Environ.Toxicol. Pharm. 36, 1113–1119.

Guillouzo, A., 1998. Liver cell models in in vitro toxicology. Environ. Health Persp.106 (2), 511–532.

Gunasekara, A.S., Troung, T., 2007. Environmental fate of fipronil. California Environ.Prot. Agency 28, 1–28.

Guo-Xin, H., Xiao-Yu, C., Hong-Yu, Z., Xiang-Jun, Q., Bing-Bing, C., Zhong-Qiu, L.,2006. Toxicokinetics of fipronil and fipronil sulfone in rabbits. Chin. J.Pharmacol. Toxicol. 20 (4), 356–360.

Gutmann, I., Wahlefeld, A.W.L., 1974. L-(+)-Lactate determination with lactatedehydrogenase and NAD+. In: Bergmeyer, H.U. (Ed.), Methods of EnzymaticAnalysis. Academic Press, New York, pp. 1464–1466.

Hainzl, D., Casida, J.E., 1996. Fipronil insecticide: novel photochemicaldesulfinylation with retention of neurotoxicity. P. Natl. Acad. Sci. 93,12746–12767.

Hainzl, D., Cole, L.M., Casida, J.E., 1998. Mechanisms for selective toxicity of fipronilinsecticide and its sulfone metabolite and desulfinyl photoproduct. Chem. Res.Toxicol. 11, 1529–1535.

Huntington, G.B., Archibeque, S.L., 1999. Practical aspects of urea and ammoniametabolism in ruminant. P. Natl. Acad. Sci. 1, 1–11.

Ioannides, C., Lewis, D.F.V., 2004. Cytochromes P450 in the bioactivation ofchemicals. Curr. Top. Med. Chem. 4, 1767–1788.

Jennings, K.A., Canerdy, T.D., Keller, R.J., Atieh, B.H., Doss, R.B., Gupta, R.C., 2002.Human exposure to fipronil from dogs treated with Frontline. Vet. Hum. Toxicol.44, 301–303.

Kimmig, R., Mauch, T.J., Kerzl, W., Schwabe, U., Scholz, R., 1983. Actions of glucagonon flux rates in perfused rat liver. Kinetics of the inhibitory effect on glycolysisand the stimulatory effect on glycogenolysis. Eur. J. Biochem. 136, 609–616.

Kun, F., Kearney, E.B., 1974. Ammonia. In: Bergmeyer, H.U. (Ed.), Methods ofEnzymatic Analysis. Academic Press, New York, pp. 1802–1806.

Lee, S.J., Mulay, P., Diebolt-Brown, B., Lackovic, M.J., Mehler, L.N., Beckman, J., Waltz,J., Prado, J.B., Mitchell, Y.A., Higgins, S.A., Schwartz, A., Calvert, G.M., 2010. Acuteillnesses associated with exposure to fipronil – surveillance data from 11 statesin the United States, 2001–2007. Clin. Toxicol. 48 (7), 737–744.

Leghait, J., Gayrard, V., Picard-Hagen, N., Camp, M., Perdu, E., Toutain, P.L., Viguié, C.,2009. Fipronil-induced disruption of thyroid function in rats is mediated byincreased total and free thyroxine clearances concomitantly to increasedactivity of hepatic enzymes. Toxicology 255, 38–44.

Lindsay, D.B., 1970. Carbohydrate metabolism in ruminants. In: Phillipson, A.T. (Ed.),Physiology of Digestion and Metabolism in the Ruminant. Oriel Press, Newcastleupon Tyne, England, pp. 438–541.

Mallette, L.E., Exton, J.H., Park, C.R., 1969. Control of gluconeogenesis from aminoacids in the perfused rat liver. J. Biol. Chem. 244, 5713–5723.

Mann, G.E., Smith, S.A., Norman, P.S., Emery, P.W., 1988. Fasting, refeeding anddiabetes modulate free amino acid concentrations in the rat exocrine pancreas:role of transstimulation in amino acid efflux. Pancreas 3, 67–76.

Mingatto, F.E., Maioli, M.A., Bracht, A., Ishii-Iwamoto, E.L., 2008. Effects ofmonocrotaline on energy metabolism in the rat liver. Toxicol. Lett. 182,115–120.

Mohamed, F., Senarathna, L., Percy, A., Abeyewardene, M., Eagleshmam, G., Cheng,R., Azher, S., Hittarage, A., Dissanayake, W., Sheriff, M.H.R., Davies, W., Buckley,N., Eddleston, M., 2004. Acute human poisoning with the N-phenylpyrazoleinsecticide fipronil – a GABAA – gated chloride channel blocker. J. Toxicol. Clin.Toxic. 42 (7), 955–963.

Pagadigorria, C.L.S., Bracht, A., Constantin, J., Nascimento, E.A., 1996. Effects of thenon-steroidal anti-inflammatory drug piroxicam on energy metabolism in theperfused rat liver. Comp. Biochem. Phys. C 113, 93–98.

Palma, I.D.F., Medeiros, H.C.D., Santana, A.T., Guelf, M., Mingatto, F.E., 2013. Efeito dofipronil na bioenergética de mitocôndrias isoladas de fígado de ratos. In: IXSimpósio de Ciências da UNESP-Dracena. http://www.dracena.unesp.br/Home/Eventos/sicud2013/ivo-dias-ferreira-da-palma-efeitos.pdf.

Park, C.R., Exton, J.H., 1972. Glucagon and the metabolism of glucose. In: Lefebvre, P.J., Unger, R.H. (Eds.), Glucagon: Molecular Physiology, Clinical and TherapeuticImplications. Pergamon, Oxford, pp. 77–108.

Rhône-Poulenc, 1995. Atelier International Fipronil/lutte antiacridienne. Rhône-Poulenc Agrochimie, Lyon.

Ruderman, N.B., 1975. Muscle amino acid metabolism and gluconeogenesis. Annu.Rev. Med. 26, 245–258.

Scholz, R., Bucher, T., 1965. Hemoglobin-free perfusion of rat liver. In: Chance, B.,Estabrook, R.W., Willianson, J.R. (Eds.), Control of Energy Metabolism. AcademicPress, New York, pp. 393–414.

Shi, Q., Yang, X., Greenhaw, J., Salminen, W.F., 2011. Hepatic cytochrome P450sattenuate the cytotoxicity induced by leflunomide and its active metaboliteA771726 in primary cultured rat hepatocytes. Toxicol. Sci. 122, 579–586.

Silva, A.S. 2008. Efeitos neurocomportamentais da exposição prolongada de ratos aofipronil. Dissertação (Mestrado em Toxicologia Experimental), Departamentode Patologia da Faculdade de Medicina Veterinária e Zootecnia, USP. http://www.teses.usp.br/teses/disponiveis/9/9141/tde-02042009-160216/pt-br.php.

Somchit, N., Ngee, C.S., Yaakob, A., Ahmad, Z., Zakaria, Z.A., 2009. Effects ofcytochrome P450 inhibitors on itraconazole and fluconazole inducedcytotoxicity in hepatocytes. J. Toxicol. 2009, 1–7.

Tang, J., Amin Usmani, K., Hodgson, E., Rose, R.L., 2004. In vitro metabolism offipronil by human and rat cytochrome P450 and its interactions withtestosterone and diazepam. Chem. Biol. Interact. 147, 319–329.

Tingle, C.C., Rother, J.A., Dewhurst, C.F., Lauer, S., King, W.J., 2003. Fipronil:environmental fate, ecotoxicology, and human health concerns. Rev. Environ.Contam. T. 176, 1–66.

Vidau, C., Brunet, J.L., Badiou, A., Belzunces, L.P., 2009. Phenylpyrazole insecticidesinduce cytotoxicity by altering mechanisms involved in cellular energysupply in the human epithelial cell model Caco-2. Toxicol. In Vitro 23,589–597.

Vidau, C., González-Polo, R.A., Niso Santano, M., Goméz Sánchez, R., Bravo-SanPedro, J.M., Pizarro-Estrella, E., Blasco, R., Brunet, J.L., Belzunces, L.P., Fuentes, J.M., 2011. Fipronil is a powerful uncoupler of oxidative phosphorylation thattriggers apoptosis in human neuronal cell line SHSY5Y. Neurotoxicology 32,935–943.

Williams, Z.R., Goodman, C.B., Soliman, K.F., 2007. Anaerobic glycolysis protectionagainst 1-methyl-4-phenylpyridinium (MPP+) toxicity in C6 glioma cells.Neurochem. Res 32, 1071–1080.

Younes, M., Strubelt, O., 1988. Cyanide-induced injury to the isolated perfused ratliver. Basic Clin. Pharmacol. Toxicol. 63, 382–385.

Zhao, X., Yeh, J.Z., Salgado, V.L., Narahashi, T., 2004. Fipronil is a potent open channelblocker of glutamate-activated chloride channels in cockroach neurons. J.Pharmacol. Exp. Ther. 310 (1), 192–201.

Zhao, X., Yeh, J.Z., Salgado, V.L., Narahashi, T., 2005. Sulfone metabolite of fipronilblocks aminobutyric acid and glutamate-activated chloride channels inmammalian and insect neurons. J. Pharmacol. Exp. Ther. 314, 363–373.