Experimental Neurology 254 (2014) 216–223

Contents lists available at ScienceDirect

Experimental Neurology

j ourna l homepage: www.e lsev ie r .com/ locate /yexnr

Taurine protects against bilirubin-induced hyperexcitation in ratanteroventral cochlear nucleus neurons

Ning-ying Song a,b, Chun-yan Li a,⁎, Xin-lu Yin a, Min Liang a, Hai-bo Shi a, Guo-ying Han a, Shan-kai Yin a,⁎a Department of Otorhinolaryngology, Affiliated Sixth People's Hospital of Shanghai Jiaotong University, 600 Yishan Road, Shanghai 200233, Chinab Department of Otorhinolaryngology, West China Hospital, Sichuan University, Chengdu, China

Abbreviations: AVCN, anteroventral cochlear nucleus;postsynaptic current; CNS, central nervous system; IGlu, gcurrent; ITau, taurine-activated current; VGCC, voltage-dep⁎ Corresponding authors. Fax: +86 21 64834143.

E-mail addresses: licycrystal@sina.com (C. Li), yinshan

0014-4886/$ – see front matter. Crown Copyright © 2014http://dx.doi.org/10.1016/j.expneurol.2013.12.014

a b s t r a c t

a r t i c l e i n f o

Article history:Received 6 June 2013Revised 24 November 2013Accepted 20 December 2013Available online 29 December 2013

Keywords:TaurineBilirubinGABAA receptorGlycine receptorVoltage-dependent calcium channel currents

No effective medication for hyperbilirubinemia yet exists. Taurine is believed to exert a neuroprotective action.The aim of the present study was to determine whether taurine protected neurons of the rat anteroventralcochlear nucleus (AVCN) against bilirubin-induced neuronal hyperexcitation. AVCN neurons were isolatedfrom 13 to 15-day-old Sprague–Dawley rats. The effects of bilirubin on the spontaneous excitatory postsynapticcurrents (sEPSCs) and action potential currents were compared with those exerted by bilirubin and taurinetogether. Bilirubin dramatically increased the frequencies of sEPSCs and action potential currents, but notsEPSC amplitude. Taurine suppressed the enhanced frequency of action potentials induced by bilirubin, in adose-dependent manner. In addition, taurine decreased the amplitude of voltage-dependent calcium channelcurrents that were enhanced upon addition of bilirubin. We explored the mechanism of the protective effectsexerted by taurine using GABAA and glycine receptor antagonists, bicuculline and strychnine, respectively.Addition of bicuculline and strychnine eliminated the protective effects of taurine. Neither bilirubin nor taurineaffected the sensitivity of the glutamate receptor. Our findings thus indicate that taurine protectedAVCNneuronsagainst bilirubin-induced neuronal hyperexcitation by activating the GABAA and glycine receptors and inhibitingcalcium flow through voltage-gated channels. Thus, taurine may be effective in treatment of neonatalhyperbilirubinemia.

Crown Copyright © 2014 Published by Elsevier Inc. All rights reserved.

Introduction

Hyperbilirubinemia is triggered by accumulation of excess bilirubinin the blood. Most infants experience physiological jaundice during thefirst 2 weeks after birth, and some develop chronic bilirubin encepha-lopathy (kernicterus) (Shapiro, 2010). Deposition of bilirubin in certainbrain regions causes temporary or permanent damage to neurons(Gourley, 1997; Ostrow et al., 2004; Watchko, 2006). Some structuresof the central nervous system (CNS), particularly the hippocampus,the auditory nuclei, the anterior ventral cochlear nucleus (AVCN), thelateral superior olive, and the inferior colliculus, are highly sensitive tobilirubin and thus vulnerable to bilirubin-induced neurotoxicity(Li et al., 2011; McDonald et al., 1998; Shi et al., 2006). For example,the cochlear nuclear volume of homozygous (jj) Gunn rats, whichexhibit congenital hyperbilirubinemia, is significantly less than that ofheterozygous (Nj) rats (Conlee and Shapiro, 1991). Bilirubin directly

sEPSC, spontaneous excitatorylutamate-evoked postsynapticendent calcium channel.

kai@china.com (S. Yin).

Published by Elsevier Inc. All rights

affects the mitochondrial and plasma membranes, causing oxidativedamage, and interrupts DNA and protein synthesis (Kashiwamataet al., 1980; Ostrow et al., 2004; Watchko, 2006; Yamada et al., 1977).Critically, bilirubin may cause excitotoxicity, triggering excessive influxof Na+, Ca2+, and water into the cell, in turn inducing cell swelling anddeath (Ostrow et al., 2004; Watchko, 2006). Glutamate, a major excit-atory neurotransmitter of the CNS, contributes to bilirubin-inducedexcitotoxicity (Johnston, 2005). Although hyperbilirubinemia has beeninvestigated intensively, no effective treatment for the condition hasyet been identified.

Taurine is one of the most abundant free amino acids in the brainand exertsmany physiological functions, including promotion of neuro-nal proliferation and differentiation, scavenging of free radicals, andregulation of membrane excitability (Chen et al., 1998; Godfrey et al.,2000; Hanna et al., 2004). Neuroprotection is one of themost importantfunctions of taurine; the amino acid protects neurons from glutamate-induced excitotoxicity (Tang et al., 1996) and inhibits accumulation ofexcess intracellular Ca2+, thus preventing neuronal damage (Kontroand Oja, 1988; Leon et al., 2009; Oja and Saransaari, 2007). Zhanget al. investigated cultured fetal neurons, and found that taurinereduced bilirubin-induced apoptotic cell death,maintained intracellularCa2+ homeostasis, and aided cells to recover from bilirubin-induceddamage (Zhang et al., 2010). Although the effects of taurine on neurons

reserved.

217N. Song et al. / Experimental Neurology 254 (2014) 216–223

have been investigated extensively, the mechanism underlying theneuroprotective function remains poorly understood.

The AVCN, the first synaptic station along the central auditory path-way, plays important roles in the processing of peripheral auditorysignals. In the present study, we recorded spontaneous excitatory post-synaptic currents (sEPSCs) and action potentials from AVCN neuronsusing the gramicidin-perforated patch-clamp technique. We found thatbilirubin markedly increased the frequencies of sEPSCs, and firing, cre-ating an excitotoxic effect. Further, we showed that taurine protectedthe neurons against bilirubin-induced hyperexcitation, and that addi-tion of antagonists of the GABAA and glycine receptors eliminated thisprotective action. Taurine also inhibited ion flow through bilirubin-activated voltage-gated calcium channels. Together, the findings sug-gest that taurine may possibly be used to treat hyperbilirubinemia-induced neuronal excitotoxicity.

Materials and methods

All experimental protocols complied with institutional principles forthe care and use of animals, and were approved by the Ethics ReviewCommittee for Animal Experimentation at Shanghai JiaotongUniversity.All efforts were made to minimize animal suffering and the number ofanimals used.

Preparation of AVCN neurons

AVCN neurons were prepared as described previously (Rhee et al.,1994). Briefly, Sprague–Dawley rats (of either gender), 13–15 daysold, were anesthetized with sodium pentobarbital (55 mg/kg, i.p.) anddecapitated. Each brain was removed quickly and placed in ice-coldartificial cerebrospinal fluid (ACSF) containing the following (in mM):124 NaCl, 5 KCl, 1.2 KH2PO4, 1.3 MgSO4, 2.4 CaCl2, 24 NaHCO3, and 10glucose. The brain region containing the AVCN was sectioned intocoronal slices (300 μm thick) using a vibratome (VT-1000s, Leica).Brain slices were pre-incubated in ACSF saturated with 95% O2/5% CO2

(both v/v) at room temperature (24–26 °C) for 20–30 min. Neuronswere next mechanically isolated using fire-polished glass pipettes andplaced on 35-mm-diameter dishes (Corning) containing a solutioncomprising (in mM) 150 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 10 glucose, and10 HEPES. The pH of the solutionwas adjusted to pH 7.4with Tris-base.

Reagents

Reagents used included free bilirubin, taurine, bicuculline, strychnine,and glutamate. All were purchased from Sigma (St. Louis, MO). Bilirubinwas first dissolved in 0.1 M NaOH to a concentration of 1 mM, stored insingle-use aliquots in the dark at −20 °C (for b24 h), and diluted to afinal concentration of 3 μM immediately before use (Li et al., 2012).Bilirubin is photosensitive and all bilirubin solutions were protectedfrom light at all times. Isolated neuronswere exposed to variousmaterialsusing a Y-tube system which completely replaced the existing solutionwith the material-containing solution within 10–20 ms (Nabekuraet al., 1996).

Electrophysiological recordings

Dissociated neuronswere bathed in a standard external solution andvisualized using the phase-contrast mode of an inverted microscope(TE-2000U; Nikon, Tokyo, Japan). Gramicidin-perforated patch-clamprecordings were made with the aid of an amplifier (EPC 10; HEKA,Lambrecht/Pfalz, Germany). All sEPSCs were recorded in voltageclamp mode at a holding potential of −60 mV. Patch electrodes werepulled from borosilicate capillary glass using a vertical pipette puller(P-9; Narishige, Tokyo, Japan) and had a resistance of 4–7 MΩ. Thepatch pipette solution used to record sEPSCs and action potentialscontained (in mM): 150 KCl and 10 HEPES (adjusted to pH 7.2 with

Tris-base). Gramicidin was dissolved in methanol to 10 mg/ml, dilutedto a final concentration of 50–80 μg/ml prior to application, and usedwithin 2 h. Action potentials were recorded in the current clamp mode,and the injected current was zero. Voltage-dependent calcium channel(VGCC) currents were recorded in whole cells. The external solutioncontained 10 mM CaCl2, 130 mM tetraethylammonium chloride, 5 mM4-AP, 5 mM HEPES, 25 mM D-glucose, and 1 μM tetrodotoxin (pH 7.4);and the internal solution 105 mM CsCl, 40 mMHEPES, 5 mM D-glucose,2.5 mMMgCl2, 10 mMEGTA, 2 mM Mg-ATP, and 0.5 mMGTP (pH 7.2).Each VGCC current was activated by ramp depolarization from −80 to60 mV in 10 mV increments, using test pulses 150 ms in duration.Capacitance and series resistance were corrected; the series resistancecompensation was 80–94%. The peak current developing during stepdepolarization was considered to be the maximum current. Signalswere filtered at 1–3 kHz, sampled at 3–10 kHz using a Dell computerequipped with Pulse 6.0 software (HEKA), and collected with the aidof the patch-clamp amplifier.

Data analysis

The numbers of sEPSCs (≥6 pA) and action potentials (≥40 mV)were automatically counted and analyzed using MiniAnalysis software(Synaptosoft, Leonia, NJ). The frequencies and amplitudes of all synapticevents (including action potentials) that developed during and afterapplication of test materials were normalized and compared to controlvalues using the Wilcoxon signed-rank test. The amplitudes ofglutamate-evoked postsynaptic currents and peak VGCC currentswere normalized to those of the control, averaged, and subjected toanalysis using Student's paired t-test. All statistical analyses wereperformedwith the aid of SPSS version 17.0 software (SPSS Inc., Chicago,IL). All values are expressed as means ± standard errors. A value ofP b 0.05 was taken to indicate statistical significance.

Results

Bilirubin induced AVCN hyperexcitation

Ablood bilirubin level of 3 μMcauses encephalopathy in rats (DaoodandWatchko, 2006).We tested the effects of bilirubin at this concentra-tion on spontaneous firing of isolated AVCN neurons at the restingpotential. Fig. 1 shows that the firing rate increased dramatically duringaddition of bilirubin. The firing rate was 694.0 ± 256.2% that of thecontrol 10 min after bilirubin addition (F = 34.543, P b 0.01, n = 6)and remained higher than the control value (527.7 ± 311.3% of con-trol) even 3 min after bilirubin washout. These values were not signifi-cantly different (F = 7.569, P N 0.05, n = 6) (Fig. 1C). Thus, bilirubintriggered hyperexcitation of AVCN neurons.

Spontaneous postsynaptic currents in dissociated AVCN neurons

Spontaneous postsynaptic currents (sPSCs) featuring two compo-nents were recorded in AVCN neurons clamped at −60 mV (Fig. 2A,top trace). The outward current (Fig. 2B, top trace) had a muchlonger time course and a slower decay time than did the inward sPSC(Fig. 2B, bottom trace). The former current was partially blockedby 10 μM bicuculline (a GABAA receptor antagonist) and completelyblocked upon co-addition of 0.3 μM strychnine (a glycine receptor an-tagonist) (Fig. 2A, bottom two traces), suggesting that the outwardPSCs were inhibitory in nature and were mediated by both the GABAA

and glycine receptors. The inward PSCs were partially suppressed by50 μM APV (an NMDA receptor antagonist) and completely blockedupon co-addition of 10 μM NBQX (an AMPA receptor antagonist)(Fig. 2A, top; second and third traces), indicating that the inward PSCswere excitatory and mediated by both the NMDA and AMPA receptors.

Fig. 1. Bilirubin triggered hyperexcitation of neurons of the anteroventral cochlear nucleus (AVCN). (A) Action potentials were recorded from an AVCN neuron before, during, and afteraddition of 3 μM bilirubin. (B) The time course of changes in firing rate of the same AVCN neuron before, during, and after addition of 3 μM bilirubin. (C) The average firing rates of sixAVCN neurons before, during (10 min addition time), and after (to 3 min after washout) bilirubin addition were compared. NS: no significant difference; **: P b 0.01.

Fig. 2. Spontaneous outward GABA/glycinergic inhibitory postsynaptic currents (sIPSCs)and inward glutamatergic excitatory postsynaptic currents (sEPSCs) recorded from anAVCN neuron. (A) The sIPSCs were partially inhibited by 10 μMbicuculline and completelyblocked by co-addition of 0.3 μMstrychnine (bottom two traces). The sEPSCswerepartiallysuppressed by 50 μM APV and completely blocked by co-addition of 10 μM NBQX (top:second and third traces). (B) The average waveforms of sIPSCs and sEPSCs were recordedover 4 min from a neuron in control solution.

218 N. Song et al. / Experimental Neurology 254 (2014) 216–223

Bilirubin increased the frequency of sEPSCs

To explore whether glutamatergic synaptic transmission con-tributed to bilirubin-induced excitotoxicity of AVCN neurons, weadded bilirubin to 3 μM. The average sEPSC frequency increased to562.2 ± 131.7% that of the control during bilirubin addition, and actionpotentials could still be evoked even 4–7 min after bilirubin addition(Figs. 3A and B, F = 19.116, P b 0.01, n = 6). The sEPSC frequencyrecovered to 176.2 ± 25.7% that of the control 3 min after bilirubinwashout (Figs. 3A and B, F = 9.855, P N 0.05, n = 6). However, thesEPSC amplitude was not affected by bilirubin (being 112.7 ± 29.7%that of the control during bilirubin application, and 103.7 ± 24.7% ofcontrol after washout; F = 7.136 and 24.162, P N 0.05, respectively;n = 6; Figs. 3A and C). Together, the data suggest that bilirubinincreases the extent of presynaptic glutamate release, triggeringhyperexcitation of AVCN neurons.

Neither bilirubin nor taurine affected postsynaptic glutamatereceptor sensitivity

To explore whether bilirubin and/or taurine influenced glutamatereceptor sensitivity, we recorded the currents evoked by exogenousglutamate (20 μM) and next determined how bilirubin and taurineaffected these currents. The glutamate-activated current (the IGluparam-eter) was the same during (104.6 ± 4.6% of control; P N 0.05, n = 6)and 10 min after (96.7 ± 5.6% of control; P N 0.05, n = 6) addition ofbilirubin to 3 μM (Fig. 4A). The IGlu amplitude during (92.8 ± 6.5% ofcontrol; P N 0.05, n = 5) and 10 min after (84.6 ± 6.0% of control;P N 0.05, n = 5) addition of taurine (to 0.1 mM) also did not change

Fig. 3. Bilirubin increased the frequency but not the amplitude of spontaneous excitatory postsynaptic currents (sEPSCs). (A) Addition of bilirubin (3 μM) increased sEPSC frequency(left panel) and evoked action potential currents (right panel) in an AVCN neuron. (B) Histogram showing the average sEPSC frequencies of six neurons before, during (10 min), andafter (to 3 min) bilirubin addition. The sEPSC frequency increased significantly after bilirubin treatment. (C) The sEPSC amplitude was not affected by bilirubin. Vertical bars: standarderrors; NS: no significant difference; **: P b 0.01.

219N. Song et al. / Experimental Neurology 254 (2014) 216–223

(Fig. 4B). Thus, neither bilirubin nor taurine affected glutamate receptoractivity in AVCN neurons.

Taurine protected ANCN neurons against bilirubin-induced excitotoxicity

To determine whether bilirubin-induced excitotoxicity could besuppressed by taurine, we added bilirubin (to 3 μM), together with dif-ferent concentrations of taurine, to AVCN neurons. The frequency ofbilirubin-evoked action potentials decreased dramatically during tau-rine (0.1 mM) addition and recovered rapidly after taurine washout

Fig. 4. The effects of bilirubin (3 μM) and taurine (0.1 mM) on the glutamate-evoked postsynbilirubin perfusion. Bottom: Normalized IGlu amplitudes from six neurons before, during, andafter taurine perfusion. Bottom: Normalized IGlu amplitudes from five neurons before, during, a

(Fig. 5A). We examined the dose-dependency of this effect. As shownin Fig. 5B, taurine at 0.01 mM did not change the firing frequency(126.4 ± 23.2% of control; F = 13.91, P N 0.05, n = 4). However, a sig-nificant decrease in firing rate was observed as the taurine concentra-tion increased, commencing at 0.03 mM (Fig. 5B). The frequencies fellto 46.5 ± 13.2% (F = 10.063, P b 0.05, n = 4), 25.4 ± 12.3%(F = 5.836, P b 0.05, n = 4), and 6.3 ± 2.0% (F = 19.50, P b 0.05,n = 4) that of the control when taurine was present at 0.03 mM,0.1 mM, and 0.3 mM, respectively. Thus, taurine significantly decreasedbilirubin-induced hyperexcitation in a dose-dependent manner.

aptic current (IGlu). (A) Top: IGlu recorded from an AVCN neuron before, during, and afterafter bilirubin perfusion. (B) Top: IGlu recorded from an AVCN neuron before, during, andnd after taurine perfusion. NS: no significant difference.

Fig. 5. Taurine antagonized bilirubin-induced hyperexcitation. (A) The enhanced firing elicited by 3 μM bilirubin was suppressed by taurine (0.1 mM). (B) The combined data show theaverage neuronal firing frequencies during application of bilirubin, and co-addition of bilirubin with different levels of taurine. The results are representative of those of four independentexperiments, each on four neurons. NS: no significant difference; *: P b 0.05.

220 N. Song et al. / Experimental Neurology 254 (2014) 216–223

Protection by taurine was inhibited by bicuculline and strychnine

We investigated whether antagonists of the GABAA/glycine recep-tors (the GABAARs/GlyRs) affected the taurine-induced protection ofneurons from bilirubin-induced hyperexcitation. First, we evaluatedthe effects of GABAAR and GlyR antagonists, bicuculline and strychnine,on the taurine-activated current (the ITau parameter). This current fell to44.0 ± 4.7% of the control value (F = 25.64, P b 0.01, n = 6; Fig. 6A)upon addition of strychnine to 0.3 μM, and to 60.5 ± 7.9% of the controlvalue (F = 18.189, P b 0.01, n = 6; Fig. 6A) upon addition of bicucullineto 10 μM. Co-addition of strychnine (0.3 μM) and bicuculline (10 μM)almost completely eliminated the ITau (9.8 ± 1.6% of the control;F = 8.723, P b 0.01, n = 6; Fig. 6A). These results suggested that ITauwas mediated by the GABAA and glycine receptors.

We next examined how taurine exerted its protective effect byactivating the GABAA and glycine receptors. Taurine, bicuculline, andstrychnine were added to the bilirubin perfusion solution. The firingrate after co-application of bilirubin, taurine, and strychnine (0.3 μM),with bicuculline (10 μM), did not significantly differ from that notedwhen bilirubin alone was applied (122.0 ± 11.2% of the control;F = 16.00, P N 0.05, n = 5; Figs. 6B and C), indicating that taurine-mediated protection was inhibited by strychnine and bicuculline.These results further suggested that taurine protected AVCN neuronsfrom excitotoxicity by activating the GABAA and glycine receptors.

Taurine suppressed the bilirubin-induced elevation of the VGCC current

Calciumplays an important role inmaintaining neuronal excitabilityand modulating hyperexcitation-induced neuronal death in vivo. Intra-cellular calcium overloading is the major molecular mechanism bywhich bilirubin exerts an excitotoxic action (Watchko, 2006; Zhanget al., 2010). Taurine inhibited calcium re-entry into neurons andchanged VGCC activity, mediating neuroprotection (Chen et al., 2001;Wu and Prentice, 2010;Wu et al., 1992).We next investigated the effectof bilirubin on the VGCC current and determined whether taurineaffected the action of bilirubin on that current in AVCN neurons. AVGCC current was evident at all potentials from −80 mV to 60 mV(Figs. 7A and B, n = 4) and bilirubin (3 μM) increased the current

amplitude. The maximum VGCC current was 171.4 ± 15.0% that ofthe control (P b 0.05, n = 4; Fig. 7C, left). However, the maximumcurrent amplitude significantly decreased (to 34.7 ± 13.1% that of thecontrol; P b 0.05, n = 4; Figs. 7B and C, right) during 5 min of exposureto taurine (0.1 mM). Moreover, bilirubin (3 μM) had little effect on themaximum current amplitude in the presence of taurine (0.1 mM) (themaximum VGCC current was 38.9 ± 12.2% that of the control,P b 0.05, n = 4; Figs. 7B and C, right). These results suggested thattaurine suppressed the bilirubin-induced elevation of the VGCC currentin AVCN neurons.

Discussion

Phototherapy and exchange transfusion are two popular treatmentsfor newborns with hyperbilirubinemia. Although phototherapy is non-invasive and easily applied, the therapy can induce photodynamicstress and lipid peroxidation (Ostrea et al., 1985; Tozzi-Ciancarelliet al., 1985). About 30% of newborns with hyperbilirubinemia causedby ABO hemolytic disease require exchange transfusions (Miqdadet al., 2004), which are invasive and can cause serious complications,including thrombocytopenia, necrotizing enterocolitis, or even death,particularly in preterm and low-birth-weight infants (Jackson, 1997;Maisels and Watchko, 2003). Minocycline has been proposed as apotential treatment for bilirubin-induced neurotoxicity (Geiger et al.,2007; Lin et al., 2005; Rice et al., 2011). However, we earlier showedthat minocycline did notmodulate the excitatory synaptic transmissionor hyperexcitation induced by bilirubin (Li et al., 2012). Moreover,minocycline causes the teeth of young adults to become pigmented(Poliak et al., 1985), and negatively affects other body tissues, such asthe skull (White and Besanceney, 1983) and thyroid gland (Gordonet al., 1984). Thus, minocycline is not appropriate for treatment ofneonatal hyperbilirubinemia. An ideal treatment should be noninvasiveand very safe.

Current treatment of neonatal hyperbilirubinemia is dictated by thetotal serum bilirubin level. However, the free bilirubin concentration isthe main determinant of the severity of neonatal hyperbilirubinemia,because free bilirubin can easily pass the blood–brain barrier to exerttoxic effects on the brain (Bratlid et al., 1983; Levine, 1979; Wennberg,

Fig. 6. Taurine-mediated protection was inhibited by bicuculline and strychnine. (A) Left panel: The AVCN neuronal current evoked by 0.1 mM taurine (ITau) was suppressed by strychnine(0.3 μM), bicuculline (10 μM), or strychnine plus bicuculline. Right panel: The combined data show that ITau was reduced significantly by strychnine, bicuculline, or strychnine plusbicuculline. (B) Co-addition of taurine, strychnine, and bicuculline did not affect the enhanced frequency offiring induced by 3 μMbilirubin. (C) The combined data show that the frequenciesof firing elicited by bilirubin; co-addition of bilirubin, taurine, strychnine and bicuculline; and after washout of strychnine, bicuculline, taurine, and bilirubin, were similar. NS: no significantdifference. **: P b 0.01.

221N. Song et al. / Experimental Neurology 254 (2014) 216–223

2000). Taurine protects neurons from excitotoxicity (Chen et al., 2001;Hanna et al., 2004; Oja and Saransaari, 2007; Schaffer et al., 2003).Taurine also passes freely through the blood–brain barrier (Angeliniet al., 1998), inhibiting bilirubin-induced neurotoxicity. The neuropro-tective functions of taurine in the context of bilirubin-induced apoptosis(Zhang et al., 2010) and oxidative stress (Chen et al., 2001; Schafferet al., 2003) are exerted via inhibition of NMDA receptors and VGCCcurrents (Leon et al., 2009; Wu et al., 2005; Wu et al., 2009). Taurineconjugation with bile acids plays an important role in acceleratingbilirubin excretion (Yamao et al., 1996). Taurine reduced the durationof jaundice in patients with acute hepatitis (Matsuyama et al., 1983).However, any electrophysiological effect of taurine on neuronal excit-ability and hyperexcitation remains unclear. The results of our presentstudy add electrophysiological evidence to support the idea that taurineexerts a powerful protective action when neurons are exposed tobilirubin-induced excitotoxicity.

As excitotoxicity is the main cause of bilirubin-induced neurologicaldisorders, we believed that activation of inhibitory neurotransmitterreceptors would suppress the observed increase in excitability. Taurinehas been suggested to be an inhibitor of neurotransmitter action, viaactivation of GABARs and/or GlyRs (Okamoto et al., 1983). The type ofreceptor activated depends on the brain region in question and thetaurine concentration (Hussy et al., 1997; Wu et al., 2008; Xu et al.,2006). We found that taurine activated both the GABAARs and GlyRsof AVCN neurons (Fig. 6). Also, taurine reduced bilirubin-evoked

hyperexcitation (Fig. 5), and this suppression was inhibited by GABAARand GlyR antagonists (Fig. 6B). Thus, taurine-mediated protectionof AVCN neurons against excitotoxicity involves GABAAR and GlyRactivation.

In addition, intracellular calcium overloading is a molecular featureof bilirubin excitotoxicity (Watchko, 2006; Zhang et al., 2010). Severalprevious studies have shown that taurine protects neurons from deathby inhibiting extracellular calcium influx, of outflow of calcium fromintracellular pools (Chen et al., 2001; El Idrissi and Trenkner, 1999;Zhang et al., 2010). Also, taurine-mediated neuroprotection is associat-ed with inhibition of VGCC currents (Wu and Prentice, 2010; Wu et al.,1992; Wu et al., 2005; Zhang et al., 2010). We found, in the presentstudy, that bilirubin-induced enhancement of VGCC current was signif-icantly reduced in the presence of taurine, which is consistent with thefindings of many other works.

Glutamate is a major excitatory neurotransmitter of the CNS, andplays an important role in bilirubin-generated neuronal excitotoxicity.We reported previously that bilirubin facilitated presynaptic glutamaterelease, activated postsynaptic AMPA and NMDA receptors, and trig-gered neuronal hyperexcitation (Li et al., 2011). In addition, Oermannet al. have shown that knockout of the mouse taurine transporter geneincreased AMPA receptor density (Oermann et al., 2005). Therefore,we considered whether bilirubin and taurine might affect the sensitivi-ties of postsynaptic glutamate receptors. However, this was not the casein AVCN neurons (Fig. 4).

Fig. 7. Taurine inhibited the bilirubin-induced elevation of VGCC current. (A) I–V curves of peak neuronal calcium currents before and after exposure to bilirubin (left), or taurine andtaurine + bilirubin (right). (B) Representative traces of maximum peak calcium currents before and after exposure to bilirubin, taurine, and taurine + bilirubin. (C) The combineddata show that bilirubin increased peak VGCC currents, and taurine suppressed this bilirubin-induced effect. NS: no significant difference; *: P b 0.05.

222 N. Song et al. / Experimental Neurology 254 (2014) 216–223

Although the electrophysiological and neuroprotective actions oftaurine have been explored (Chen et al., 2001; Wu and Prentice, 2010;Wu et al., 2009; Xu et al., 2006), themechanisms underlying the protec-tive effect are not fully understood. We showed that taurine activatedboth GABAARs and GlyRs, and suppressed calcium currents, in AVCNneurons, eliminating bilirubin-induced excitotoxicity. Further clinicalresearch on specific taurine therapies for bilirubin encephalopathyshould be conducted.

In conclusion, we found that taurine protected AVCN neurons frombilirubin-induced excitotoxicity by activating both GABAARs and GlyRsand inhibiting the increase in VGCC current mediated by bilirubin.Taurine may thus be used to effectively treat hyperbilirubinemia inneonates.

Acknowledgments

Thisworkwas sponsored by the State KeyDevelopment Program forBasic Research of China (grant no. 2011CB504503), and the NationalNatural Science Foundation of China (grant no. 81300822 and81170918). We are grateful to Professor Shu Hui Wu for her many sug-gestions and supportive work on the manuscript.

References

Angelini, E., Teixeira, M., Aran, J.M., Ferrary, E., 1998. Taurine entry into perilymph of theguinea pig. Eur. Arch. Otorhinolaryngol. 255, 331–333.

Bratlid, D., Cashore, W.J., Oh, W., 1983. Effect of serum hyperosmolality on opening ofblood–brain barrier for bilirubin in rat brain. Pediatrics 71, 909–912.

Chen, W.Q., Jin, H., Nguyen, M., Carr, J., Lee, Y.J., Hsu, C.C., Faiman, M.D., Schloss, J.V., Wu,J.Y., 2001. Role of taurine in regulation of intracellular calcium level and neuroprotec-tive function in cultured neurons. J. Neurosci. Res. 66, 612–619.

Chen, X.C., Pan, Z.L., Liu, D.S., Han, X., 1998. Effect of taurine on human fetal neuron cells:proliferation and differentiation. Adv. Exp. Med. Biol. 442, 397–403.

Conlee, J.W., Shapiro, S.M., 1991. Morphological changes in the cochlear nucleus andnucleus of the trapezoid body in Gunn rat pups. Hear. Res. 57, 23–30.

El Idrissi, A., Trenkner, E., 1999. Growth factors and taurine protect against excitotoxicity bystabilizing calcium homeostasis and energy metabolism. J. Neurosci. 19, 9459–9468.

Geiger, A.S., Rice, A.C., Shapiro, S.M., 2007. Minocycline blocks acute bilirubin-inducedneurological dysfunction in jaundiced Gunn rats. Neonatol. 92, 219–226.

Godfrey, D.A., Farms,W.B., Godfrey, T.G., Mikesell, N.L., Liu, J., 2000. Amino acid concentra-tions in rat cochlear nucleus and superior olive. Hear. Res. 150, 189–205.

Gordon, G., Sparano, B.M., Kramer, A.W., Kelly, R.G., Iatropoulos, M.J., 1984. Thyroid glandpigmentation and minocycline therapy. Am. J. Pathol. 117, 98–109.

Gourley, G.R., 1997. Bilirubin metabolism and kernicterus. Adv. Pediatr. 44, 173–229.Hanna, J., Chahine, R., Aftimos, G., Nader, M., Mounayar, A., Esseily, F., Chamat, S., 2004.

Protective effect of taurine against free radicals damage in the rat myocardium.Exp. Toxicol. Pathol. 56, 189–194.

Hussy, N., Deleuze, C., Pantaloni, A., Desarmenien, M.G., Moos, F., 1997. Agonist action oftaurine on glycine receptors in rat supraoptic magnocellular neurones: possible rolein osmoregulation. J. Physiol. 502 (Pt 3), 609–621.

Jackson, J.C., 1997. Adverse events associated with exchange transfusion in healthy and illnewborns. Pediatrics 99, E7.

Johnston, M.V., 2005. Excitotoxicity in perinatal brain injury. Brain Pathol. 15, 234–240.Kashiwamata, S., Aono, S., Semba, R.K., 1980. Characteristic changes of cerebellar proteins

associated with cerebellar hypoplasia in jaundiced Gunn rats and the prevention ofthese by phototherapy. Experientia 36, 1143–1144.

Kontro, P., Oja, S.S., 1988. Effects of taurine on the influx and efflux of calcium in brainslices of adult and developing mice. Int. J. Neurosci. 38, 103–109.

Leon, R., Wu, H., Jin, Y.,Wei, J., Buddhala, C., Prentice, H.,Wu, J.Y., 2009. Protective functionof taurine in glutamate-induced apoptosis in cultured neurons. J. Neurosci. Res. 87,1185–1194.

Levine, R.L., 1979. Bilirubin: worked out years ago? Pediatrics 64, 380–385.Li, C.Y., Shi, H.B., Song, N.Y., Yin, S.K., 2011. Bilirubin enhances neuronal excitability by

increasing glutamatergic transmission in the rat lateral superior olive. Toxicology284, 19–25.

Li, C.Y., Shi, H.B., Ye, H.B., Song, N.Y., Yin, S.K., 2012. Minocycline cannot protect neuronsagainst bilirubin-induced hyperexcitation in the ventral cochlear nucleus. Exp.Neurol. 237, 96–102.

Lin, S., Wei, X., Bales, K.R., Paul, A.B., Ma, Z., Yan, G., Paul, S.M., Du, Y., 2005. Minocyclineblocks bilirubin neurotoxicity and prevents hyperbilirubinemia-induced cerebellarhypoplasia in the Gunn rat. Eur. J. Neurosci. 22, 21–27.

Maisels, M.J., Watchko, J.F., 2003. Treatment of jaundice in low birthweight infants. Arch.Dis. Child. Fetal Neonatal Ed. 88, F459–F463.

Matsuyama, Y., Morita, T., Higuchi, M., Tsujii, T., 1983. The effect of taurine administrationon patients with acute hepatitis. Prog. Clin. Biol. Res. 125, 461–468.

McDonald, J.W., Shapiro, S.M., Silverstein, F.S., Johnston, M.V., 1998. Role of glutamatereceptor-mediated excitotoxicity in bilirubin-induced brain injury in the Gunn ratmodel. Exp. Neurol. 150, 21–29.

223N. Song et al. / Experimental Neurology 254 (2014) 216–223

Miqdad, A.M., Abdelbasit, O.B., Shaheed, M.M., Seidahmed, M.Z., Abomelha, A.M., Arcala, O.P.,2004. Intravenous immunoglobulin G (IVIG) therapy for significant hyperbilirubinemiain ABO hemolytic disease of the newborn. J. Matern. Fetal Neonatal Med. 16, 163–166.

Nabekura, J., Omura, T., Akaike, N., 1996. Alpha 2 adrenoceptor potentiates glycinereceptor-mediated taurine response through protein kinase A in rat substantianigra neurons. J. Neurophysiol. 76, 2447–2454.

Oermann, E., Warskulat, U., Heller-Stilb, B., Haussinger, D., Zilles, K., 2005. Taurine-transporter gene knockout-induced changes in GABA(A), kainate and AMPA butnot NMDA receptor binding in mouse brain. Anat. Embryol. (Berl) 210, 363–372.

Oja, S.S., Saransaari, P., 2007. Pharmacology of taurine. Proc. West. Pharmacol. Soc. 50, 8–15.Okamoto, K., Kimura, H., Sakai, Y., 1983. Evidence for taurine as an inhibitory neurotransmit-

ter in cerebellar stellate interneurons: selective antagonism by TAG (6-aminomethyl-3-methyl-4H,1,2,4-benzothiadiazine-1,1-dioxide). Brain Res. 265, 163–168.

Ostrea Jr., E.M., Cepeda, E.E., Fleury, C.A., Balun, J.E., 1985. Red cell membrane lipidperoxidation and hemolysis secondary to phototherapy. Acta Paediatr. Scand. 74,378–381.

Ostrow, J.D., Pascolo, L., Brites, D., Tiribelli, C., 2004. Molecular basis of bilirubin-inducedneurotoxicity. Trends Mol. Med. 10, 65–70.

Poliak, S.C., DiGiovanna, J.J., Gross, E.G., Gantt, G., Peck, G.L., 1985. Minocycline-associatedtooth discoloration in young adults. JAMA 254, 2930–2932.

Rhee, J.S., Ebihara, S., Akaike, N., 1994. Gramicidin perforated patch-clamp techniquereveals glycine-gated outward chloride current in dissociated nucleus solitarii neu-rons of the rat. J. Neurophysiol. 72, 1103–1108.

Rice, A.C., Chiou, V.L., Zuckoff, S.B., Shapiro, S.M., 2011. Profile of minocycline neuroprotec-tion in bilirubin-induced auditory system dysfunction. Brain Res. 1368, 290–298.

Schaffer, S., Azuma, J., Takahashi, K., Mozaffari, M., 2003. Why is taurine cytoprotective?Adv. Exp. Med. Biol. 526, 307–321.

Shapiro, S.M., 2010. Chronic bilirubin encephalopathy: diagnosis and outcome. Semin.Fetal Neonatal Med. 15, 157–163.

Shi, H.B., Kakazu, Y., Shibata, S., Matsumoto, N., Nakagawa, T., Komune, S., 2006. Bilirubinpotentiates inhibitory synaptic transmission in lateral superior olive neurons of therat. Neurosci. Res. 55, 161–170.

Tang, X.W., Deupree, D.L., Sun, Y., Wu, J.Y., 1996. Biphasic effect of taurine on excitatoryamino acid-induced neurotoxicity. Adv. Exp. Med. Biol. 403, 499–505.

Tozzi-Ciancarelli, M.G., Amicosante, G., Menichelli, A., Di Giulio, S., Del Principe, D., 1985.Photodynamic damage induced by bilirubin on human platelets: possible relevanceto newborn pathology. Biol. Neonate 48, 336–340.

Watchko, J.F., 2006. Kernicterus and the molecular mechanisms of bilirubin-induced CNSinjury in newborns. Neruomol. Med. 8, 513–529.

Wennberg, R.P., 2000. The blood–brain barrier and bilirubin encephalopathy. Cell. Mol.Neurobiol. 20, 97–109.

White, S.W., Besanceney, C., 1983. Systemic pigmentation from tetracycline andminocycline therapy. Arch. Dermatol. 119, 1–2.

Wu, H., Jin, Y., Wei, J., Jin, H., Sha, D., Wu, J.Y., 2005. Mode of action of taurine as aneuroprotector. Brain Res. 1038, 123–131.

Wu, J., Kohno, T., Georgiev, S.K., Ikoma, M., Ishii, H., Petrenko, A.B., Baba, H., 2008. Taurineactivates glycine and gamma-aminobutyric acid A receptors in rat substantia gelati-nosa neurons. Neuroreport 19, 333–337.

Wu, J.Y., Prentice, H., 2010. Role of taurine in the central nervous system. J. Biomed. Sci. 17(Suppl. 1), S1.

Wu, J.Y., Tang, X.W., Tsai, W.H., 1992. Taurine receptor: kinetic analysis and pharmacolog-ical studies. Adv. Exp. Med. Biol. 315, 263–268.

Wu, J.Y., Wu, H., Jin, Y., Wei, J., Sha, D., Prentice, H., Lee, H.H., Lin, C.H., Lee, Y.H., Yang, L.L.,2009. Mechanism of neuroprotective function of taurine. Adv. Exp. Med. Biol. 643,169–179.

Xu, H., Wang, W., Tang, Z.Q., Xu, T.L., Chen, L., 2006. Taurine acts as a glycine receptor ag-onist in slices of rat inferior colliculus. Hear. Res. 220, 95–105.

Yamada, N., Sawasaki, Y., Nakajima, H., 1977. Impairment of DNA synthesis in Gunn ratcerebellum. Brain Res. 126, 295–307.

Yamao, J., Matsumura, Y., Hokaze, Y., Yoshikawa, M., Umemoto, N., Matsui, Y., Fukui, H.,Tsujii, T., 1996. Significance of taurine conjugation of bile acid in the biliary excretionof bilirubin. Adv. Exp. Med. Biol. 403, 99–106.

Zhang, B., Yang, X., Gao, X., 2010. Taurine protects against bilirubin-induced neurotoxicityin vitro. Brain Res. 1320, 159–167.