Noise exposure induces up-regulation of ecto-nucleoside triphosphate diphosphohydrolases 1 and 2 in rat cochlea

NTC

SDCa

TNb

Uc

Hd

v

AiitNciesirsi(iomtNecphiiid

*uB�EApgcgpPfrt

Neuroscience 126 (2004) 763–773

0d

OISE EXPOSURE INDUCES UP-REGULATION OF ECTO-NUCLEOSIDERIPHOSPHATE DIPHOSPHOHYDROLASES 1 AND 2 IN RAT

rcipDadL

KR

Np(cTtwo(oPi2

bct1TaotcSeai2pn(

edhB

. M. VLAJKOVIC,a,b* G. D. HOUSLEY,a

. J. B. MUNOZ,a,b S. C. ROBSON,c J. SEVIGNY,d

. J. H. WANGa AND P. R. THORNEa,b

Department of Physiology, Faculty of Medical and Health Sciences,he University of Auckland, Private Bag 92019, Auckland,ew Zealand

Discipline of Audiology, Faculty of Medical and Health Sciences, Theniversity of Auckland, Auckland, New Zealand

Department of Medicine, Beth Israel Deaconess Medical Center,arvard Medical School, Boston, MA, USA

Centre de Recherche en Rhumatologie et Immunologie, CHUQ, Uni-ersite Laval, Sainte-Foy, Quebec, Canada

bstract—Extracellular ATP acting via P2 receptors in thenner ear initiates a variety of signaling pathways that may benvolved in noise-induced cochlear injury. Nucleosideriphosphate diphosphohydrolase (NTPDase)1/CD39 andTPDase2/CD39L1 are key elements for regulation of extra-ellular nucleotide concentrations and P2 receptor signalingn the cochlea. This study characterized the effect of noisexposure on regulation of NTPDase1 and NTPDase2 expres-ion in the cochlea using a combination of real-time RT-PCR,mmunohistochemistry and functional studies. Adult Wistarats were exposed to broad band noise at 90 dB and 110 dBound pressure level (SPL) for 72 h. Exposure to 90 dB SPLnduced a small and temporary change of auditory thresholdstemporary threshold shift), while exposure to 110 dB SPLnduced a robust and permanent change of auditory thresh-lds (permanent threshold shift). NTPDase1 and NTPDase2RNA transcripts were upregulated in the cochlea exposed

o 110 dB SPL, while mild noise (90 dB SPL) altered onlyTPDase1 mRNA expression levels. Changes in NTPDasesxpression did not correlate with levels of circulating corti-osterone, implying that the up-regulation of NTPDases ex-ression was not stress-related. Semi-quantitative immuno-istochemistry in the cochlea exposed to 110 dB SPL local-

zed the increased NTPDase1 and NTPDase2 immunostainingn the stria vascularis and up-regulation of NTPDase2 in thentraganglionic spiral bundle. In contrast, NTPDase1 wasown-regulated in the cell bodies of the spiral ganglion neu-

Correspondence to: S. M. Vlajkovic, Department of Physiology, Fac-lty of Medical and Health Sciences, University of Auckland, Privateag 92019, Auckland, New Zealand. Tel: �64-9-373-7599; fax:64-9-373-7499.-mail address: [email protected] (S. M. Vlajkovic).bbreviations: ABR, auditory brainstem response; AP, artificialerilymph; CSF, cerebrospinal fluid; Ct, threshold cycle; GAPDH,lyceraldehyde-3 phosphate dehydrogenase; ICC, immuno-ytochemistry; IGSB, intraganglionic spiral bundle; MGB, minorroove binder; NTPDase, nucleoside triphosphate diphos-hohydrolase; PB, phosphate buffer; PBS, phosphate-buffered saline;TS, permanent threshold shift; RP-HPLC, reverse phase high per-

ormance liquid chromatography; RT, reverse transcription; SGN, spi-al ganglion neurones; SPL, sound pressure level; TTS, temporary

1hreshold shift.

306-4522/04$30.00�0.00 © 2004 IBRO. Published by Elsevier Ltd. All rights reseroi:10.1016/j.neuroscience.2004.04.023

763

ones. Distribution of NTPDases was not altered in the co-hlea exposed to 90 dB SPL. Functional studies revealedncreased ectonucleotidase activities in the cochlea after ex-osure to 110 dB SPL, consistent with up-regulation of NTP-ases. The changes in NTPDases expression may reflectdaptive response of cochlear tissues to limit ATP signalinguring noise exposure. © 2004 IBRO. Published by Elseviertd. All rights reserved.

ey words: hearing, ATP, NTPDases, CD39, real-timeT-PCR, gene expression.

oise exposure is a common cause of hearing loss. Ex-osure to loud sound above 85 dB sound pressure levelSPL) causes noise-induced hearing loss by damaging theochlea, particularly the organ of Corti (Spoendlin, 1971).hreshold shift (hearing loss) may be temporary (TTS) if

he organ of Corti is able to recover, or permanent (PTS)hen hair cells or neurones die (Borg, 1995). Mild damagef synapses and/or hair cell stereocilia can be repairedBorg, 1995). Severe damage inducing hair cell loss, lossf supporting cells, and neuronal apoptosis, accounts forTS. The PTS itself is a consequence of degeneration that

s secondary to the initial hair cell loss (Bohne and Harding,000).

ATP and other extracellular nucleotides are known toe important signaling and regulatory molecules in theochlea acting via ATP-gated ion channels (P2X recep-ors) and G-protein-coupled P2Y receptors (Housley,998, 2000, 2001; Marcus et al., 1998; Housley andhorne, 2000). P2 receptors in the sensory, supportingnd neural tissues of the cochlea regulate a diverse rangef physiological processes, including sound transduction,he endocochlear potential, neurotransmission, microme-hanics and cochlear blood flow (Munoz et al., 1995, 1999;kellett et al., 1996, 1997; Housley et al., 1998, 1999; Saliht al., 1998, 2002; Bobbin, 2001a). Extracellular ATP haslso been implicated in the pathophysiology of noise-

nduced cochlear injury (Munoz et al., 2001; Bobbin,001b; Thorne et al., 2002). It has been suggested that theurinergic signaling system provides a regulatory mecha-ism limiting cochlear sensitivity under stressor conditionsHousley et al., 2002).

Extracellular nucleotides can be hydrolyzed by a vari-ty of surface-located enzymes known as ectonucleoti-ases (Plesner, 1995; Zimmermann, 2001). ATP-ydrolyzing activity in the brain (Zimmermann et al., 1998;raun et al., 2000, 2003) and cochlea (Vlajkovic et al.,

ved.

nfcegaa

f(Aprm1c2

ntsNcec

niosiign

A

TwemAiwo

Qf

Tntm

cAeo(tre

N

Rslsnnobwm

H

Hec(wA4fdrsssv

M

AafaAbTb(p0tDtGaefatatTi

Q

RqcA(is

S. M. Vlajkovic et al. / Neuroscience 126 (2004) 763–773764

ucleoside triphosphate diphosphohydrolase (NTPDase)amily. Four members of the family (NTPDase1–4) areharacterized by two transmembrane domains and a largextracellular loop containing five apyrase conserved re-ions (Handa and Guidotti, 1996). In contrast, NTPDase5nd 6 lack the C-terminus and can be cleaved to producesoluble protein (Zimmermann, 2001).

NTPDase1 and NTPDase2 are likely the key elementsor termination of P2 receptor signaling in the cochleaVlajkovic et al., 1999). NTPDase1 hydrolyzes ATP andDP to a similar extent, whereas NTPDase2 has highreference for nucleoside triphosphates. In addition to theole in termination of purinergic signaling, these enzymesay prevent P2 receptor desensitization (Enjyoji et al.,999) and control the availability of ligands for either nu-leotide (P2) or adenosine (P1) receptors (Bonan et al.,001; Sevigny et al., 2002).

NTPDase1 and 2 localization in sensory, supporting,eural, vascular and secretory tissues corresponds with

he reported distribution of some P2X and P2Y receptorubunits (Vlajkovic et al., 2002a,b). Hence, the role forTPDases in regulation of P2 receptor signaling in theochlea has been linked to control of cochlear blood flow,lectrochemical regulation of sound transduction and toochlear neurotransmission.

The aim of this study was to determine the influence ofoise exposure on NTPDase1 and NTPDase2 expression

n order to characterize their potential role in pathophysi-logy of noise-induced cochlear injury. Here, we demon-trate an overall up-regulation of these two NTPDases and

ncrease of ectonucleotidase activity in the cochlea follow-ng acoustic over-stimulation. The cochlear plasticity re-arding NTPDase expression may play a role in control ofucleotide-mediated cellular responses to noise.

EXPERIMENTAL PROCEDURES

nimals

he experiments were carried out on adult Wistar rats of either sexeighing between 250 and 300 g. Animals were allowed freexcess to food and water during noise exposures. The experi-ents described in this study were approved by the University ofuckland Animal Ethics Committee. All experiments conform to

nternational guidelines on the ethical use of animals. All effortsere made to minimize animal suffering and to reduce the numberf animals used.

uantitative studies of NTPDase expression andunction in the cochlea

he quantification of NTPDase1 and NTPDase2 in control andoise-exposed cochleas was performed using the TaqMan real-ime PCR to assess the mRNA levels and semi-quantitative im-unohistochemistry to determine the levels of protein expression.

The effect of chronic noise exposure on ATP hydrolysis in theochlea was assessed following in vivo perfusion of exogenousTP through the cochlear perilymphatic compartment. The efflu-nt was collected and analyzed for the presence of ATP metab-lites by reverse phase high performance liquid chromatographyRP-HPLC). Auditory brainstem response (ABR) to clicks andones was used to assess auditory function, and the corticoste-one levels were used as an indicator of stress response in noise-

oise exposures

ats were exposed to a broadband (16 Hz–26 kHz) noise in aound-proof booth for 72 h at either 90 or 110 dB SPL. The soundevels in the cage were measured using a calibrated Rion NL-40ound level meter. Control animals were maintained at ambientoise (ca 60 dB SPL) in the Animal Resources Unit. At the end ofoise exposure, test and control animals were euthanised by anverdose of Nembutal (90 mg/kg i.p. pentobarbitone sodium; Vir-ac Laboratories, Auckland, New Zealand) and their cochleasere removed for mRNA quantitation, immunohistochemistry, andorphological assessment.

istological preparation

istopathology was assessed in cochlear tissues of noise-xposed animals by light microscopy. Rats were perfused trans-ardially with 2.5% glutaraldehyde in a 0.08 M phosphate bufferPB). Temporal bones were removed and the round and ovalindows opened to allow perfusion of fixative through the cochlea.fter overnight post-fixation in 2.5% glutaraldehyde/0.08 PB at°C, cochleas were osmicated (2% osmium tetraoxide in ddH2O)

or 1 h and decalcified in 5% EDTA for 7 days. Cochleas wereehydrated in ethanol series and propylene oxide, embedded inesin, and sectioned with a glass knife. Sections (1 �m) weretained with Toluidine Blue, then mounted in Histomount, cover-lipped and imaged using a Zeiss Axioskop 2 mot plus micro-cope. Whole mounts of cochlear tissues were also used to pro-ide high resolution imaging.

easurement of ABR

BR to clicks and pure tones (4–20 kHz) were measured beforend immediately after noise exposure, and again at days 7 and 14ollowing noise exposures. The animals were anesthetized usingn i.p. injection of xylazine (10 mg/kg) and ketamine (40 mg/kg).BRs were recorded within a sound attenuator chamber (Shel-urg Acoustics, Pty Ltd, Croydon North, Australia) using theucker Davis Technology (TDT, Alachua, FL, USA), computer-ased digital signal processing package and associated softwareBioSig, Alachua, FL, USA). Digitally produced clicks (100 �s) orure tone stimuli (5 ms tone bursts with a rise/fall time of 1.5 ms;–110 dB SPL), were converted to an analog signal and passed

o an attenuator (TDT PA4), before delivery to a Beyer dynamicT 48 headphone speaker (Beyer, Heilbronn, Germany) posi-

ioned in a speculum placed within the external auditory canal.rass F-E2 platinum subdermal needle electrodes were insertedt the mastoid process (reference electrode), head vertex (activelectrode) and contralateral mastoid (ground electrode). Signalsrom the electrodes were passed into a TDT Bioamp head stagend amplified 100,000 times in a TDT DB4 amplifier. The ABRhreshold was defined as the lowest intensity (to the nearest 5 dB)t which a response (waves three to four) could be visually de-ected following at least two repeated traces of 1024 averages.hresholds were determined by visual analysis of traces in an

ntensity series recorded at near-threshold levels.

uantification of NTPDase mRNA

eal-time RT-PCR (Bustin, 2000, 2002) was used to provideuantification of NTPDase1 and NTPDase2 mRNA levels in co-hlear tissues using TaqMan primers and probes (Perkin-Elmerpplied Biosystems, Foster City, CA, USA). Minor groove binder

MGB) probes were used to increase the stability of probe hybrid-zation (Kutyavin et al., 2000). The quantification of mRNA tran-cripts was provided by using amplicon-specific standard curves

Cs

Tppa(1amsrea

Q

QbsTNptcs

D

RpUpU(GTrflD58ppp

psa

Cs

Ng(vPsdspdcuf(s

rA

R

Rappc2PoAa1

srnflcedtcpfcNcmd

I

Nd1ut2iB2at

EssTNpsga1atgwipvsrp

S. M. Vlajkovic et al. / Neuroscience 126 (2004) 763–773 765

ochlear mRNA isolation and first strand cDNAynthesis

he tympanic bulla was removed and placed in sterile 0.1 Mhosphate-buffered saline (PBS; pH 7.4). The cochlea was ex-osed and the otic capsule removed. The membranous labyrinthnd modiolus were dissected out and placed in lysis buffer100 mM Tris–HCl, pH 8.0; 500 mM LiCl; 10 mM EDTA, pH 8.0;% LiDS; 5 mM DTT). Cochlear tissues were homogenized using

sterile Teflon pestle and RNA extracted using DynabeadsRNA DIRECT (Dynal A.S., Oslo, Norway). First-strand cDNA

ynthesis was carried out in a 20-�l reverse transcription (RT)eaction using random primers (Invitrogen, Groningen, The Neth-rlands), dNTPs (Amersham Biosciences, Piscataway, NJ, USA)nd Superscript II reverse transcriptase (Invitrogen).

uantification of cochlear mRNA

uantification of cochlear mRNA was performed using a Ri-oGreen reagent kit (Molecular Probes, Eugene, OR, USA), aensitive fluorescent method for quantitation of RNA in solution.he labeling was performed in a standard 96-well plate (Nalgeunc Int., Naperville, IL, USA) and read in a fluorescence micro-late reader (Wallac Qy, Turku, Finland) at 535 nm. The concen-ration of mRNA in the sample was determined from the standardurve produced by serial dilutions of the control ribosomal RNAupplied with the kit.

esign of TaqMan primers and probes

at-specific primers and probes were designed using Primer Ex-ress Software v1.5 (PE Applied Biosystems, Foster City, CA,SA) and checked by BLAST search. The primers used for am-lification of NTPDase1-specific cDNA (GenBank accession no.81295) were 5�-GACAGCAACGCAGGGTGG-3� forward primer

nucleotide position 1411–1428) and 5�-GCTGGGATCATGTTG-TCAAG-3� reverse primer (nucleotide position 1469–1449).aqMan MGB probe (5�-TTTGGGCTACATGCTG) carrying a 5�

eporter FAM (6-carboxy fluorescein) and a 3� NFQ (non-uorescent quencher), was located in position 1431–1446. NTP-ase2 specific primers (GenBank accession no. Y11835) were�-TGGCAGTGCCATCGTCAG-3� upstream primer (position82–899) and 5�-GGTCACGACACAGGGTAGCA-3� downstreamrimer (position 937–918). NTPDase2-specific TaqMan MGBrobe (5�-CTGTCAGGAACCAGC-3�) was located in nucleotideosition 901–915.

Quantitation of a house-keeping gene, glyceraldehyde-3hosphate dehydrogenase (GAPDH) cDNA, was performed for allamples as an endogenous reference. GAPDH-specific primersnd VIC-labeled probes are proprietary to Applied Biosystems.

onstruction of plasmids and preparation oftandard curves

TPDase1-specific cDNA (GenBank accession no. U81295) li-ated in pCR2.1 vector (Invitrogen) and NTPDase2-specific cDNAGenBank accession no. Y11835) ligated in pcDNA3 vector (In-itrogen) were used to generate standard curves for real-timeCR. The concentrations of the plasmids were determined bypectrophotometry using GeneQuant (Pharmacia, Uppsala, Swe-en) and converted to a copy number based on a linear relation-hip between the size of DNA and its weight in Daltons (one baseair equals 660 Da). For the standard curve, a series of 10-foldilutions of both plasmids were prepared, representing 102–107

opies of DNA/�l. Standard curves for GAPDH were producedsing GAPDH cDNA (GenBank accession no. X02231) amplified

rom cochlear tissues (RT-PCR) using specific 838–859sense5�-TCAAGAAGGTGGTGAAGCAGGC-3�) and 1201–1180anti-

ified using a QIAquick PCR purification kit (Qiagen, Clifton Hill,ustralia).

eal-time PCR

eal-time PCR was carried out in MicroAmp Optical 96-well re-ction plates using TaqMan Universal PCR Master Mix (PE Ap-lied Biosystems), Taqman primers (100 nM each) and MGBrobes (200 nM). As a template, 2 �l of sample cDNA or positiveontrol plasmid DNA was added to a total reaction volume of5 �l. A negative control without a template was included in everyCR run. Amplification and fluorescence detection were carriedut using the ABI Prism 7700 Sequence Detection System (PEpplied Biosystems). The thermal cycling protocol included 2 mint 50 °C, 10 min at 95 °C, and a 40 cycle profile: 15 s at 95 °C andmin at 60 °C.

The data were analyzed using the Sequence Detector v1.7oftware (Perkin Elmer). The threshold cycle (Ct) was set foreporter dyes (FAM and VIC) so that it crossed through theon-limiting linear phase of amplification above the backgrounduorescence. Ct was defined as the cycle at which the fluores-ence exceeded 10 times the mean baseline emission for thearly cycles (3–15). The samples were tested in duplicate andata were expressed as a mean of two replicates. Amplification inhe absence of reverse transcriptase (control reaction showingontamination with genomic DNA) was subtracted from the am-lification in the presence of reverse transcriptase. Copy numbersor NTPDase1 and NTPDase2 in normal and noise-exposed co-hleas were calculated from the standard curves. The number ofTPDase1 and NTPDase2 copies was normalized to GAPDHopy number for each cochlea and expressed per nanogram ofRNA. Comparison between the groups was performed by Stu-ent’s unpaired t-test or Mann-Whitney U test.

mmunohistochemistry

TPDase1 and NTPDase2 antibodies were raised in rabbits byirect injection of the encoding cDNA in pcDNA3 (Enjyoji et al.,999). The specificity of these antibodies was previously verifiedsing cytochemistry and Western blots on transfected cells andheir protein extracts, respectively (Braun et al., 2000; Heine et al.,001). An additional control for specificity was immunohistochem-

stry of tissues from NTPDase1-null mice (Enjyoji et al., 1999;raun et al., 2000) and differential distribution of NTPDase1 andin the rat and mouse cochlear and vascular tissues (Vlajkovic etl., 2002a,b; Sevigny et al., 2002). These studies demonstratedhe lack of cross-reactivity between the two NTPDase antibodies.

Rat cochleas fixed in 4% PFA were decalcified in 5%DTA/PB solution for 7 days and cryoprotected overnight in 30%ucrose/PB solution. The tissues were then rinsed in 0.1 M PB,nap-frozen in isopentane at �80 °C and cryosectioned at 20 �m.he sections were placed in 48-well plates (Nalge Nunc Int.,aperville, USA) containing the sterile 0.1 M PBS (pH 7.4) andermeabilized with 1% Triton X-100 for 1 h. Non-specific bindingites were blocked with 2% non-fat milk powder and 1.5% normaloat serum (Vector Laboratories, Burlingame, CA, USA). Primaryntibodies to NTPDase1 or NTPDase2 were applied at a titer of:5000 in 0.1 M PBS containing 1.5% normal goat serum (Vector)nd 0.1% Triton X-100 and incubated overnight at 4 °C. Vec-astain Elite ABC kit (Vector) containing a secondary biotinylatedoat anti-rabbit antibody and avidin–biotin peroxidase complexas used for detection of NTPDase immunoreactivity. Sequential

ncubations with the secondary antibody and avidin–biotin com-lex were for 40 min at room temperature. The reactions wereisualized using diaminobenzidine (Vector) chromogen. In controlections, the primary antibody was replaced with the preimmuneabbit serum. The tissues were photographed on positive trans-arency color film (Fujichrome Provia 400F) using a microscope

iSCsm“

Q

EPtavistb(sawtc

Mn

Tv7p(mtmtctirtmpbism47Svomvmo(2

twwasAnRA

cwBp51np

C

Aafieckrct

Fp1ciTne

S. M. Vlajkovic et al. / Neuroscience 126 (2004) 763–773766

oskop, Thornwood, NY, USA) and the images were scanned on acanMaker 35t plus (Microtek Laboratory Inc., Redondo Beach,A, USA). The contrast enhancement was used to matchcanned images with the original microscopic appearance of im-unolabelled tissues. The images were also sharpened using the

Unsharp Mask” tool.

uantification of NTPDase translation

xpression of enzyme proteins was semi-quantified usinghotoshop-based image analysis (Lehr et al., 1997, 1999). Al-

hough this method cannot provide information for the absolutemount of chromogen present (Matkowskyj et al., 2000), thealidity of Photoshop-based image analysis is comparable to othermage analysis programs, which use similar features for coloreparation as Photoshop (Lehr et al., 1999). Immunostaining in-ensity was calculated as the difference between specific andackground staining, and expressed as the immunocytochemicalICC) index using arbitrary units (Lehr et al., 1997). Betweeneven and 11 images were analyzed for each experimental groupnd data compared using Student’s unpaired t-test. All specimensere processed in a double-blind manner; the individuals doing

he staining and the analyses were unaware of the experimentalonditions.

easurement of ecto-nucleotidase activity in theoise-exposed cochlea

he hydrolysis of ATP (0.5 mM) in the cochlea was studied inivo and compared between noise-exposed (110 dB SPL for2 h) and control animals. The rats were anesthetized withentobarbitone sodium (60 mg/kg, i.p.), placed in a heated37 °C) stereotaxic head-holder with the ventral side upper-ost, tracheostomized and left to breathe on their own. Body

emperature, monitored through a rectal thermocouple, wasaintained at 37 °C by a heating blanket. The temperature of

he cochlea, measured by a thermistor probe attached to theochlear basal turn, was constant (36.8 –37.5 °C) throughouthe experiment in all animals tested (n�17). Prior to the open-ng of the left tympanic bulla, the cisterna magna was opened toeduce cerebrospinal fluid (CSF) pressure and contamination ofhe effluent with CSF. The cochlea was approached via ventro-edial surface in the neck region. The entire cochlear perilym-hatic space was perfused with ATP through a hole made in theasal turn scala tympani. The effluent was collected by a cap-

llary tube placed against an outlet hole in scala vestibuli of theame cochlear turn. The cochlea was perfused for the first 20in with artificial perilymph (AP) solution (133.5 mM NaCl,.5 mM KCl, 0.4 mM MgCl2, 1.8 mM CaCl2, 20 mM NaHCO3; pH.4, 295 mOsm), and then with 0.5 mM ATP (disodium salt;igma, St Louis, MO, USA) dissolved in AP. Our previous inivo studies in the cochlea (Vlajkovic et al., 1996) have dem-nstrated that at this concentration of ATP, adenine nucleotideetabolites are accurately measured by HPLC, and tissue

iability is fully preserved. Effluent collection started after 15in of ATP perfusion, due to dead space constraints. The ratef perfusion (2.5 �l/min) was controlled by a syringe pumpSage 355; Orion Research Inc, Boston, MA, USA) using a50 �l syringe.

The degradation of ATP was assessed by RP-HPLC usinghe procedure modified from Heine et al. (1999). The samplesere centrifuged at 300�g for 10 min and the supernatantsere precipitated with an equal volume of 6.5% trichloroaceticcid for 30 min at 4 °C and spun at 14,000�g for 45 min. Theamples were then filtered through a 0.2 �m filter (Alltechssociates Inc., Deerfield, IL, USA) and stored at �80 °C. Theucleotides and nucleosides were separated and quantified byP-HPLC (Agilent 1100 Series; Agilent Technologies, Palo

leoside column (Alltech; 7 �m; 250�4.6 mm). The samplesere eluted at 1 ml/min with the mobile phase consisting ofuffer A (60 mM NH4H2PO4, 5 mM tetrabutylammonium phos-hate, pH 5.0) and Buffer B (80% HPLC-grade methanol withmM TBAP); gradient: 10 –50% Buffer B in Buffer A in the first

0 min. The absorbance was detected at 254 nm, and theucleotide/nucleoside concentrations determined from theeak area in the chromatogram.

orticosterone assay

basic assessment of animal stress levels was undertaken usingcorticosterone assay. Blood samples (0.5 ml) were obtained

rom control and noise-exposed rats by heart puncture at timentervals (0.5, 24, 48 and 72 h) after the commencement of noisexposures. The serum corticosterone levels were determined by aompetitive immunoassay using a Correlate-EIA Corticosteroneit (Assay Designs Inc., Ann Arbor, MI, USA). The reactions wereead in a microplate reader (Wallac Qy) at 405 nm and theoncentrations of corticosterone in the samples calculated fromhe standard curve.

ig. 1. ABR thresholds in response to acoustic clicks (A) and 4 kHzure tone (B) before and immediately after noise exposure, and afterand 2 weeks. Exposure to 90 dB SPL induced a small and temporary

hange of auditory thresholds (TTS), while exposure to 110 dB SPLnduced a robust and permanent change of auditory thresholds (PTS).he animals were unresponsive to auditory clicks immediately afteroise exposure; hence, 100 dB is an arbitrary threshold ceiling. Dataxpressed as mean�S.D. (n�4–11 animals per group). ** P�0.01;

Ns

CS(tglttS

saa(sicafs

ai

R

Tva((tFp(D

C

Rqct(l

FcceaNa ney test.

S. M. Vlajkovic et al. / Neuroscience 126 (2004) 763–773 767

RESULTS

oise-induced cellular injury and auditory thresholdhifts

ellular pathology in the rat cochleas exposed to 110 dBPL included swelling of the organ of Corti supporting cells

inner sulcus, Hensen and Claudius cells) and fibrocytes ofhe spiral ligament, and occasionally damage to the mar-inal cells of stria vascularis (data not shown). High reso-

ution light microscopy showed lesions in the organ of Cortihat included scattered loss of outer hair cells and distor-ion of stereocilia. In contrast, mild noise exposure (90 dBPL) did not produce apparent damage to the cochlea.

Cochlear function in noise-exposed animals was as-essed by determining the threshold of the ABR to clicksnd pure tones. ABR thresholds were similar in all testednimals before the commencement of noise exposuresFig. 1). In the 90 dB group, ABR thresholds were notignificantly affected by noise exposure. A small increase

n threshold recovered within the first week (Fig. 1). Inontrast, exposure to 110 dB SPL induced an increase inuditory thresholds for auditory clicks and a loss of 31 dBor 4 kHz pure tone (Fig. 1), as for 10–20 kHz (data not

ig. 2. NTPDase1 and NTPDase2 mRNA expression in the noise-eurves for NTPDase1, NTPDase2 and GAPDH using amplicon-specopies) has been plotted against the log10 of the initial template concefficiency and linear detection range of the real-time PCR assay. (B) Cssay. (C) Expression of NTPDase1 and NTPDase2 mRNA in conTPDase2 copy numbers normalized to mRNA concentration and GAs mean�S.E.M. (n�7–8 cochleas per group); * P�0.05; Mann-Whit

hown). There was a minimal recovery of hearing 2 weeks a

fter noise exposure (Fig. 1), indicating permanent noise-nduced functional loss.

eal-time PCR standard curves

he plasmids used for generation of standard curves pro-ided a highly reproducible measure of real-time cDNAmplification. The dynamic range of DNA concentrations102–107 copies) demonstrated a strong linear relationshipr20.995) between Ct values and initial template concen-rations (Fig. 2a). The coefficients of regression shown inig. 2a were representative for all experiments. The am-lification efficiency, calculated as (10(�1/slope)�1)�100Dehee et al., 2002), was 94%, 94% and 88%, for NTP-ase1, NTPDase2 and GAPDH respectively.

t values for NTPDase1 and NTPDase2

eal-time PCR assay provided accurate and reproducibleuantitation of NTPDase mRNA levels in the cochlea. Theoefficient of variation for Ct data was less than 3% be-ween duplicates in the same run or between different runsdata not shown). Contamination with genomic DNA wasess than 1%, estimated by the Ct difference between �RT

at cochlea. (A) Representative examples of TaqMan PCR standardplasmids as templates. The Ct for each plasmid dilution (102–107

The slope and the coefficient of regression indicate high amplificationRNA concentrations in pre-reaction aliquots measured by RiboGreennoise-exposed rat cochlea. Arbitrary units denote NTPDase1 andmolecules) to correct for RNA quantity and integrity. Data expressed

xposed rific cDNAntration.ochlear mtrol and

PDH (103

nd �RT reactions (5.7–9.6; Table 1). There was no am-

pvuciNbfpmw

Nl

CsagattllaNeN(

S

Nc3osccclvcdc

t

(1coNa

cc3c1v(c(Tc

iincti9

sacbr

A

Ite(tacdth

TR

C91

a

fb

* udent’s t

S. M. Vlajkovic et al. / Neuroscience 126 (2004) 763–773768

lification in the absence of template (data not shown). Ctalues for NTPDase1 and NTPDase2 were lower (P�0.05,npaired t-test) in noise-exposed cochlea (110 dB SPL)ompared with control exposed to ambient noise, suggest-

ng up-regulation of transcription (Table 1). Ct values forTPDase1 were lower than controls in the 90 dB group,ut this failed to reach significance prior to normalizationor mRNA and GAPDH levels. Ct values for GAPDH ex-ression were similar in noise-exposed and control ani-als, indicating that GAPDH mRNA levels in the cochleaere unaffected by the noise exposures.

ormalization to mRNA concentration and GAPDHevels

oncentrations of mRNA measured by the RiboGreen as-ay were similar across the groups (Fig. 2B). NTPDase1nd NTPDase2 copy numbers were expressed per nano-ram of mRNA and normalized to GAPDH expression tollow for comparison under different experimental condi-ions. NTPDase1 mRNA expression in control cochlea waswo orders of magnitude lower compared with GAPDHevels, while NTPDase2 was three orders of magnitudeower (data not shown). NTPDase1 mRNA expression waspproximately 20-fold higher than NTPDase2 (Fig. 2C).TPDase1 mRNA expression increased in both noise-xposed groups (P�0.05, Mann-Whitney test), whileTPDase2 up-regulation only reached significance

P�0.05) in the 110 dB group (Fig. 2C).

emi-quantitative immunohistochemistry

TPDase1 immunostaining was observed in diverse co-hlear tissues (Fig. 3). Staining in the spiral ligament (Fig.A–C) was confined to blood vessels as reported previ-usly (Vlajkovic et al., 2002a). Weak immunostaining waseen in cells and blood vessels of the stria vascularis in theontrol and 90 dB noise-exposed cochlea (Fig. 3A, B). Inontrast, blood vessels of the stria vascularis in the co-hlea exposed to 110 dB SPL showed strong NTPDase1-ike immunolabelling (Fig. 3C). The ICC index for the striaascularis in the 110 dB group increased to 157% of theontrol group (P�0.05, unpaired t-test) and 185% of the 90B group (P�0.01; Fig. 4A). The ICC index was similar inontrol and 90 dB groups.

Most of the spiral ganglion neurones (SGN) in control

able 1. Ct values (mean�S.D.) for NTPDase1, NTPDase2 and GAT-PCR assaya

N NTPDase1

�RT �RTb

ontrol 8 27.43�1.05 34.62�1.110 dB 8 26.93�1.54 34.11�0.5310 dB 7 26.02�1.27* 34.63�1.27

Ct value is the cycle number at which the amplification plot rises logarrom the controls indicates increased transcription.Control reaction in the absence of reverse transcriptase representinStatistically significant difference versus control group: * P�0.05 (St

issues were immunostained with NTPDase1 antibody p

Fig. 3D–F). This was significantly reduced following the10 dB noise exposure. The SGN ICC index in the 110 dBochlea decreased to 61% of the control cochlea and 62%f the 90dB cochlea (P�0.01; Fig. 4B). The levels ofTPDase1 immunostaining were comparable in controlnd 90 dB groups.

NTPDase2 immunostaining in the stria vascularis ofontrol cochlea was confined to basal and intermediateells (Fig. 3G). The cochlea exposed to 90 dB SPL (Fig.H) showed a similar pattern of immunostaining as theontrol cochlea, while NTPDase2 immunoreactivity in the10 dB cochlea extended to the entire surface of the striaascularis, including the marginal cells facing scala mediaFig. 3I). The ICC index of the stria vascularis in the 110 dBochlea increased to 135% of the control cochleaP�0.01) and 124% (P�0.05) of the 90 dB group (Fig. 4C).he ICC index was similar in the control and 90 dBochlea.

The cell bodies of the SGN were devoid of NTPDase2mmunoreactivity, while the efferent nerve fibers of thentraganglionic spiral bundle (IGSB) were strongly immu-olabelled (Fig. 3J–L). The ICC index of the IGSB wasomparable between the groups (results not shown), buthe area of IGSB immunoreactivity in the 110 dB cochleancreased to 171% of the control cochlea and 212% of the0 dB cochlea (P�0.01; Fig. 4D).

The intensity of NTPDase1 and NTPDase2 immuno-taining in the organ of Corti (Vlajkovic et al., 2002a)ppeared to be similar in normal and noise-exposed co-hleas (data not shown). Apart from the tectorial mem-rane, control sections incubated with the preimmune se-um were unstained (Fig. 3).

TP hydrolysis in the cochlea

n control cochlea exposed to ambient noise ATP perfusedhrough the perilymphatic fluid space was hydrolyzed byctonucleotidases (Fig. 5). The initial ATP concentration0.5 mM) was reduced by 53.5% during passage throughhe perilymphatic compartment, with associated appear-nce of metabolites (ADP, AMP and adenosine). The ac-umulation of ADP in the perfusate is consistent with pre-ominant expression of NTPDase2 in the tissues exposedo perilymph. The rate of ATP hydrolysis was significantlyigher in cochlea exposed to 110 dB SPL (P�0.05, un-

ontrol and noise-exposed cochleas obtained by a TaqMan real-time

e2 GAPDH

�RT �RT �RT

.14 35.28�1.04 20.50�1.38 29.66�1.69

.05 34.64�0.90 21.26�1.03 29.67�0.98

.55* 35.14�0.85 20.30�1.40 29.97�1.96

above the background level of fluorescence. A decrease in Ct values

plification of genomic DNA.-test).

PDH in c

NTPDas

�RT

29.09�128.92�227.53�1

ithmically

g the am

aired t-test). The initial ATP concentration was reduced

bmisogi

E

Svmaashai

csr

TpUthnadpds

FwcS(eN9o

S. M. Vlajkovic et al. / Neuroscience 126 (2004) 763–773 769

y 71.9% (Fig. 5), accompanied by increased production ofetabolites (ADP and AMP). These data provide biochem-

cal evidence of NTPDase up-regulation in cochlear tis-ues following 110 dB noise exposure. The concentrationf adenosine was similar in control and noise-exposedroups, indicating no change in ecto-5�nucleotidase activ-

ty (Fig. 5).

ffect of noise exposure on corticosteroid levels

erum levels of corticosterone were assayed at time inter-als after the onset of noise exposures (Fig. 6). After 30in, there was a 10-fold increase of corticosterone innimals exposed to 90 dB SPL (P�0.01, Student’s t-test),nd this level was sustained throughout the noise expo-ure (72 h). Loud sound at 110 dB SPL induced evenigher elevation of corticosterone levels than 90 dB SPLfter 30 min. However, after 24 h circulating corticosterone

ig. 3. NTPDase immunohistochemistry. (A–F) NTPDase1 immunolaith the preimmune serum (NTPDase1 and NTPDase2, left and rigochleas. The arrows indicate NTPDase-like immunostaining of bloodGN of the control and noise-exposed cochleas. (G–I) NTPDase2 imm

sl). In the normal and 90 dB cochlea, staining in the stria is confined toxtends to the marginal cells (mc). (J–L) Perikarya of the SGN areTPDase2 immunoreactivity. In the cochlea exposed to 110 dB SPL,0 dB cochleas. Control sections stained with the preimmune sera shrgan of Corti; rm, Reissner’s membrane; sm, scala media; tm, tector

n the 110 dB group dropped to the level similar to the i

ontrol group, and remained low for the rest of the expo-ure. The latter obviates against likely corticosterone-elated effect on NTPDase expression.

DISCUSSION

his study demonstrates noise-induced changes in ex-ression of NTPDase1 and NTPDase2 in the cochlea.sing quantitative real-time RT-PCR to assess mRNA

ranscription levels, overall expression of both nucleotideydrolyzing enzymes was upregulated in response tooise exposure. This was functionally confirmed by in vivonalysis of nucleotide hydrolysis. Chronic exposure to 110B SPL increased NTPDase1 and NTPDase2 mRNA ex-ression in the cochlea, while exposure to mild noise (90B SPL) increased only NTPDase1 mRNA levels withinpecific tissues. Immunohistochemistry showed changes

–L, NTPDase2 immunolabelling; Control, cochlear tissues incubatedtively). (A–C) Lateral wall tissues of the control and noise-exposedin the stria vascularis (sv). (D–F) NTPDase1 immunoreactivity in theivity is present in the stria vascularis (sv), but not in the spiral ligamentc) and intermediate (ic) cells, whereas staining in the 110 dB cochleaed, while the efferent nerve fibers of the IGSB (IGSB) show strong2 immunoreactivity in the IGSB is more extensive than in control andnon-specific staining of the tectorial membrane (tm). lim, limbus; oc,rane. Scale bars�50 �m.

belling; Ght respec

vesselsunoreactbasal (b

not labelNTPDaseow onlyial memb

n the distribution of these two ectonucleotidases. NTP-

DlSscNdeit

sw

tennamgiT

Fva9cr

FPIicm dent’s t-t

FEAt(tse

S. M. Vlajkovic et al. / Neuroscience 126 (2004) 763–773770

ase1 was upregulated in the secretory tissues of theateral wall, and down-regulated in the cell bodies of theGN, while NTPDase2 immunostaining increased in thetria vascularis and efferent nerve fibers of the IGSB. Inontrast, changes in the distribution of NTPDase1 andTPDase2 were not detected in cochleas exposed to 90B SPL, implying a differential regulation of NTPDasexpression in response to varied noise levels. These data

ndicate that the regulation of NTPDase1 and NTPDase2ranslation in the cochlea is responsive to noise as a

ig. 4. Semi-quantitative NTPDase immunohistochemistry in the cochotoshop analysis of the luminosity of selected pixels. ICC index (im

GSB. The cochlea exposed to 110 dB SPL showed increased NTPmmunostaining of the SGN (B) compared with control cochlear tissueochlea (C, D). Control and 90 dB cochlea showed similar levelsean�S.E.M. (n�7–11 images per group); * P�0.05; ** P�0.01; Stu

ig. 5. ATP degradation in control and noise-exposed rat cochlea.xposure to 110 dB SPL for 72 h increased the rate of extracellularTP hydrolysis resulting in accumulation of metabolites (ADP, AMP) in

he perfusate. The concentrations of ATP, ADP, AMP and adenosineADO) were expressed as a percentage of the initial ATP concentra-ion (0.5 mM) introduced into the cochlear perilymphatic space. Datahown as mean�S.E.M. Number of experiments: control, n�8; noise-

(xposed, n�9; * P�0.05; Student’s t-test.

timulus that also upregulates P2 receptor signaling path-ays (Wang et al., 2003).

The present study attempted to discriminate betweenhe local effect of noise and systemic stress response onxpression levels of NTPDase1 and 2 in the cochlea. Aumber of studies have demonstrated a correlation ofoise exposures with elevation of glucocorticoids (Rarey etl., 1995; Yoshida and Liberman, 1999; Wang and Liber-an, 2002). Glucocorticoid receptors, when bound to li-and, act as transcription factors, affecting gene regulation

n a tissue-specific manner (Wang and Liberman, 2002).he measurement of corticosterone levels in our study has

ig. 6. The levels of circulating corticosterone were significantly ele-ated in both noise-exposed groups (90 dB and 110 dB SPL) 30 minfter the onset of noise exposure. This elevation was sustained in the0 dB group throughout the exposure, while in the 110 dB grouporticosterone levels were reduced to the baseline level after 24 h, andemained low after 48 and 72 h. Data presented as mean�S.E.M.

lculation of the ICC index expressed in arbitrary units was based onining intensity) was calculated for the stria vascularis (SV), SGN andmunoreactivity of the stria vascularis (A), and reduced NTPDase1

ase2 was upregulated in the stria vascularis and IGSB of the 110 dBase1 and NTPDase2 immunostaining (A–D). Data expressed as

est.

hlea. CamunostaDase1 ims. NTPDof NTPD

n�5 for each data point); * P�0.05; ** P�0.01; Student’s t-test.

tsmaqciNoltN

Fo

IciAtcaenAcnsaaeicaar

pcttdo1INsetatrut

iiTl

ameMmiric

tri2Drfs

IcbmpCrrPiasaucv

tcltiScdcasatefmi

AGDD

S. M. Vlajkovic et al. / Neuroscience 126 (2004) 763–773 771

wo important implications. Firstly, 110 dB SPL induced aignificant increase of corticosterone levels in the first 30in of exposure and subsided afterward, suggesting thatdaptation to noise during exposure to 110 dB SPL occursuickly, as hearing loss develops. Secondly, the lack oforrelation between corticosterone levels and the changesn enzyme expression implied that the up-regulation ofTPDase expression was not stress-related. Local releasef ATP from noise-damaged cochlear tissues is thus more

ikely to induce increased NTPDase expression, althoughhe signaling mechanisms associated with the regulation ofTPDase gene expression are unknown.

unctional significance of noise-induced modulationf NTPDase expression

n addition to the role as a regulatory and signaling mole-ule, extracellular ATP has a significant role in pathophys-ology (Abbrachio and Burnstock, 1998; Burnstock, 2001).TP can be released from the vesicular stores by exocy-

osis or through membrane channels such as ATP-bindingassette transporters, volume regulated anion channelsnd connexin hemichannels (Zimmermann, 1996; Josepht al., 2003). Changes in membrane permeability duringoise exposure would also enable considerable release ofTP into the extracellular space (Munoz et al., 2001). Highoncentrations of ATP can irreversibly damage afferenteural output, inner hair cell receptor potential, and causewelling of nerve terminals at the inner hair cells, possiblycting via P2X7 receptor subunit expressed in the SGNnd their terminals at hair cells (Thorne et al., 2002; Nikolict al., 2003). ATP becomes cytotoxic by turning ATP gated

on channels into large pores allowing Ca2� entry into theells (Ferrari et al., 1999; Virginio et al., 1999; Chessell etl., 2001). Along with glutamate excitotoxicity (Jagger etl., 2000), this might be an underlying mechanism of neu-onal cell death by necrosis or apoptosis in the cochlea.

Recent studies indicate that ectonucleotidases maylay a major regulatory role in pathophysiological pro-esses mediated by extracellular nucleotides. The induc-ion of status epilepticus and ischemia-reperfusion injury inhe brain leads to up-regulation of the enzyme chain hy-rolyzing extracellular ATP (NTPDase1 and ecto-5�nucle-tidase; Schetinger et al., 1997, 1998; Braun et al., 1997,998; Braun and Zimmermann, 2001; Bonan et al., 2001).n a similar fashion, we show here that NTPDase1 andTPDase2 in cochlear tissues are upregulated in re-ponse to noise. The overexpression of these ecto-nzymes appears to be a response to increased nucleo-ide release and clearly represents a protective adaptationgainst the deleterious effect of high nucleotide concen-rations (Braun et al., 1998). Effective regulation of P2eceptor mediated signaling may therefore be dependentpon adaptive changes in NTPDase expression in theseissues.

NTPDase 1 and NTPDase2 distribution in the cochleas well positioned to protect the hearing organ from the cellnjury caused by excessive stimulation of P2 receptors.he up-regulation of NTPDase1 and 2 in the stria vascu-

aris may serve to protect these tissues by regulating the p

ctivation of G protein-coupled P2Y4 receptor involved inaintenance of endolymphatic potassium levels and thendocochlear potential (Marcus et al., 1998; Sage andarcus, 2002). In addition, NTPDase2 overexpression inarginal cells bordering scala media is strategically local-

zed to provide neutralization of high ATP concentrationseleased into cochlear fluids during noise exposures. ATPs concentrated within the marginal cells of the stria vas-ularis at this site (Munoz et al., 2001).

Noise-induced reduction of NTPDase1 distribution inhe SGN coincides with increased P2X2 expression in thategion (Wang et al., 2003). It is likely that ATP signalingnfluences the level of excitability of SGN (Housley et al.,002). We speculate that the down-regulation of NTP-ase1 in SGN in 110 dB noise-exposed cochlea may be a

esponse to the loss of excitability by these neuronesollowing onset of PTS. Thus, purinergic neurotransmis-ion may determine NTPDase1 expression levels in SGN.

Loud sound induced NTPDase2 overexpression in theGSB that provides efferent innervation to the outer hairells. This medial efferent system, originating from theilateral medial superior olivary complex, is involved inodulation of cochlear sound transduction through hyper-olarisation of the outer hair cells (Zheng et al., 1997;anlon et al., 1999). Overexpression of NTPDase2 in this

egion may provide a protection of outer hair cells byeducing ATP signal levels required for the activation of the2X7 receptor localized pre-synaptically on the efferent

nnervation of the outer hair cells (Nikolic et al., 2003). Welso acknowledge the possibility that NTPDase2 immuno-taining in the interganglionic spiral bundle may be partlyssociated with Schwann cells. In either case, NTPDase2p-regulation would provide reduction of ATP levels thatould be damaging to the axons providing efferent inner-ation to the outer hair cells.

In conclusion, we demonstrated tissue-specific regula-ion of NTPDase1 and NTPDase2 expression in rat co-hlea following exposure to loud sound. Differential regu-

ation of NTPDase1 and 2 gene transcription and transla-ion in the cochlea was dependent on noise level, butndependent of glucocorticoid-mediated stress pathways.ustained nucleotide release, particularly from injured co-hlear tissues, is likely a potent stimulus for ectonucleoti-ase up-regulation associated with the lateral wall of theochlea. The changes in NTPDase expression may reflectn adaptive response of cochlear tissues to limit ATPignaling during noise exposure. Adenosine, the final met-bolic product of ectonucleotidase pathway, may also con-ribute to the cochlear response to noise stress (Rudolphit al., 1992; Sweeney, 1997). Further studies involvingunctional and molecular studies on gene knockout ani-als, will likely clarify the role of NTPDases in noise-

nduced cochlear injury.

cknowledgements—Supported by the New Zealand Lotteryrants Board, the Health Research Council of New Zealand, theeafness Research Foundation of New Zealand and the Lodgeiscovery 501 and Freemasons of New Zealand. J.S. was sup-

taDDU

A

B

B

B

B

B

B

B

B

B

B

B

B

B

C

C

D

E

F

H

H

H

H

H

H

H

H

H

H

J

J

K

L

L

M

M

M

S. M. Vlajkovic et al. / Neuroscience 126 (2004) 763–773772

utes of Health Research and S.C.R. by NIH, USA. We gratefullycknowledge that NTPDase2 cDNA used for generation of NTP-ase2 antibody was provided by Prof. H. Zimmermmann andr. N. Braun, AK Neurochemie, Biozentrum der J. W. Goethe-niversitat, Frankfurt am Main, Germany.

REFERENCES

bbrachio MP, Burnstock G (1998) Purinergic signalling-pathophysiological roles. Jpn J Pharmacol 78:113–145.

obbin RP (2001a) ATP-induced movement of the stalks of isolatedcochlear Deiters’ cells. Neuroreport 12:2923–2926.

obbin RP (2001b) PPADS, an ATP antagonist, attenuates the effectsof a moderately intense sound on cochlear mechanics. Hear Res156:10–16.

ohne B, Harding GW (2000) Degeneration in the cochlea after noisedamage: primary versus secondary events. Am J Otolaryngol 21:505–509.

onan CD, Schetinger MRC, Battastini AMO, Sarkis JJF (2001) Ec-tonucleotidases and synaptic plasticity: implications in physiologicaland pathological conditions. Drug Dev Res 52:57–65.

org E (1995) Noise-induced hearing loss-literature review and exper-iments in rabbits-morphological and electrophysiological features,exposure parameters and temporal factors, variability and interac-tions. Scand Audiol 24 (Supp 40):6–147.

raun N, Zimmermann H (2001) Microglial ectonucleotidases: identi-fication and functional roles. Drug Dev Res 53:208–217.

raun N, Lenz C, Gillardon F, Zimmermann M, Zimmermann H (1997)Focal ischemia enhances glial expression of ecto-5�-nucleotidase.Brain Res 766:213–226.

raun N, Zhu Y, Krieglstein J, Culmsee C, Zimmermann H (1998)Up-regulation of the enzyme chain hydrolyzing extracellular ATPafter transient forebrain ischemia in the rat. J Neurosci 18:4891–4900.

raun N, Sevigny J, Robson SC, Enjyoji K, Guckelberger O, HammerK, Di Virgilio F, Zimmermann H (2000) Assignment of ecto-nucleoside triphosphate diphosphohydrolase-1/cd39 expression tomicroglia and vasculature of the brain. Eur J Neurosci 12:4357–4366.

raun N, Sevigny J, Mishra SK, Robson SC, Brath SW, Gerstberger R,Hammer K, Zimmermann H (2003) Expression of the ecto-ATPaseNTPDase2 in the germinal zones of the developing and adult ratbrain. Eur J Neurosci 17:1355–1364.

urnstock G (2001) Expanding field of purinergic signaling. Drug DevRes 52:1–10.

ustin S (2000) Absolute quantification of mRNA using real-time re-verse transcription polymerase chain reaction assays. J Mol Endo-crinol 25:169–193.

ustin SA (2002) Quantification of mRNA using real-time reversetranscription PCR (RT- PCR): trends and problems. J Mol Endocri-nol 29:23–29.

anlon B, Fransson A, Viberg A (1999) Medial olivocochlear efferentterminals are protected by sound conditioning. Brain Res 850:253–260.

hessell IP, Grahames CBA, Michel AD, Humphrey PPA (2001) Dy-namics of P2X7 receptor pore dilation: pharmacological and func-tional consequences. Drug Dev Res 53:60–65.

ehee A, Cesaire R, Desire N, Lezin A, Bourdonne O, Bera O,Plumelle Y, Smadja D, Nicolas J-C (2002) Quantitation of HTLV-Iproviral load by a TaqMan real-time PCR assay. J Virol Methods102:37–51.

njyoji K, Sevigny J, Lin Y, Frenette PS, Christie PD, Esch JSA, ImaiM, Edelberg JM, Rayburn H, Lech M, Beeler DL, Csizmadia E,Wagner DD, Robson SC, Rosenberg RD (1999) Targeted disrup-tion of cd39/ATP diphosphohydrolase results in disordered hemo-stasis and thromboregulation. Nat Med 5:1010–1017.

errari D, Los M, Bauer MKA, Vandenabeele P, Wesselborg S,

tivation of caspases with distinct roles in apoptotic and necroticalterations of cell death. FEBS Lett 447:71–75.

anda M, Guidotti G (1996) Purification and cloning of a solubleATP-diphosphohydrolase (apyrase) from potato tubers (Solanumtuberosum). Biochem Biophys Res Comm 218:916–923.

eine P, Braun N, Heilbronn A, Zimmermann H (1999) Functionalcharacterization of rat ecto-ATPase and ecto-ATP diphosphohydro-lase after heterologous expression in CHO cells. Eur J Biochem262:102–107.

eine P, Braun N, Sevigny J, Robson SC, Servos J, Zimmermann H(2001) The C-terminal cysteine-rich region dictates specific cata-lytic properties in chimeras of the ectonucleotidases NTPDase1and NTPDase2. Eur J Biochem 268:364–373.

ousley GD (1998) Extracellular nucleotide signaling in the inner ear.Mol Neurobiol 16:21–48.

ousley GD (2000) Physiological effects of extracellular nucleotides inthe inner ear. Clin Exp Pharmacol Physiol 27:575–580.

ousley GD (2001) Nucleoside and nucleotide transmission in sen-sory systems. In: Handbook of experimental pharmacology: puri-nergic and pyrimidinergic signalling, vol. 151 (Abbracchio MP, Wil-liams M, eds), pp 339–369. Berlin: Springer.

ousley GD, Thorne PR (2000) Purinergic signalling: an experimentalperspective. J Auton Nerv Syst 81:139–145.

ousley GD, Luo L, Ryan AF (1998) Localization of mRNA encodingthe P2X(2) receptor subunit of the adenosine 5�-triphosphate-gatedion channel in the adult and developing rat inner ear by in situhybridization. J Comp Neurol 393:403–414.

ousley GD, Kanjhan R, Raybould NP, Greenwood D, Salih SG,Jarlebark L, Burton LD, Setz VCM, Cannell MB, Soeller C, ChristieDL, Usami S, Matsubara A, Yoshie H, Ryan AF, Thorne PR (1999)Expression of the P2X2 receptor subunit of the ATP-gated ionchannel in the cochlea: implications for sound transduction andauditory neurotransmission. J Neurosci 19:8377–8388.

ousley GD, Jagger DJ, Greenwood D, Raybould NP, Salih SG,Jarlebark LE, Vlajkovic SM, Kanjhan R, Nikolic P, Munoz DJB,Thorne PR (2002) Purinergic regulation of sound transduction andauditory neurotransmission. Audiol Neurootol 7:55–61.

agger W, Goiny M, Herrera-Marschitz M, Brundin L, Fransson A,Canlon B (2000) Noise-induced aspartate and glutamate efflux inthe guinea pig cochlea and hearing loss. Exp Brain Res 134:426–434.

oseph SM, Buchakjian MR, Dubyak GR (2003) Colocalization of ATPrelease sites and ecto-ATPase activity at the extracellular surfaceof human astrocytes. J Biol Chem 278:23331–23342.

utyavin IV, Afonina IA, Mills A, Gorn VV, Lukhtanov EA, BelousovES, Singer MJ, Walburger DK, Lokhov SG, Gall AA, Dempcy R,Reed MW, Meyer RB, Hedgpeth J (2000) 3�-Minor groove binder-DNA probes increase sequence specificity at PCR extension tem-peratures. Nucleic Acids Res 28:655–661.

ehr HA, Mankoff DA, Corwin D, Santeusanio G, Gown AM (1997)Application of Photoshop-based image analysis to quantification ofhormone receptor expression in breast cancer. J Histochem Cyto-chem 45:1559–1565.

ehr HA, van der Loos CM, Teeling P, Gown AM (1999) Completechromogen separation and analysis in double immunohistochemi-cal stains using Photoshop-based image analysis. J HistochemCytochem 47:119–125.

arcus D, Sunose H, Liu J, Shen Z, Scofield M, Ryan A (1998) Proteinkinase C mediates P2U purinergic receptor inhibition of K� channelin apical membrane of strial marginal cells. Hear Res 115:82–92.

atkowskyj KA, Schonfeld D, Benya RV (2000) Quantitative immuno-histochemistry by measuring cumulative signal strength using com-mercially available software Photoshop and Matlab. J HistochemCytochem 48:303–311.

unoz DJB, Thorne PR, Housley GD, Billett TE, Battersby JM(1995) Extracellular adenosine 5�-triphosphate (ATP) in the en-dolymphatic compartment influences cochlear function. Hear

M

M

N

P

R

R

S

S

S

S

S

S

S

S

S

S

T

V

V

V

V

V

V

V

W

W

Y

Z

Z

Z

Z

S. M. Vlajkovic et al. / Neuroscience 126 (2004) 763–773 773

unoz DJB, McFie C, Thorne PR (1999) Modulation of cochlear bloodflow by extracellular purines. Hear Res 127:55–61.

unoz DJB, Kendrick IS, Rassam M, Thorne PR (2001) Vesicularstorage of adenosine triphosphate in the guinea-pig cochlear lateralwall and concentrations of ATP in the endolymph during soundexposure and hypoxia. Acta Otolaryngol 121:10–15.

ikolic P, Housley GD, Thorne PR (2003) Expression of the P2X7

receptor subunit of the adenosine 5�-triphosphate (ATP)-gated ionchannel in the developing and adult rat cochlea. Audiol Neuro-Otol8:28–37.

lesner L (1995) Ecto-ATPases: identities and functions. Intern RevCytol 158:141–214.

arey KE, Gerhardt KJ, Curtis LM, ten Cate WJF (1995) Effect ofstress on cochlear glucocorticoid protein: acoustic stress. Hear Res82:135–138.

udolphi KA, Schubert P, Parkinson FE, Fredholm BB (1992) Neuro-protective role of adenosine in cerebral ischemia. Trends Pharma-col Sci 13:439–445.

age CL, Marcus DC (2002) Immunolocalisation of P2Y4 and P2Y2

purinergic receptors in strial marginal cells and vestibular dark cells.J Memb Biol 185:103–115.

alih SG, Housley GD, Burton LD, Greenwood D (1998) P2X2 receptorsubunit expression in a subpopulation of cochlear type I spiralganglion neurones. Neuroreport 9:279–282.

alih SG, Jagger DJ, Housley GD (2002) ATP-gated currents in ratprimary auditory neurones in situ arise from a heteromultimeric P2Xreceptor subunit assembly. Neuropharmacology 42:386–395.

chetinger MRC, Bonan CD, Schierholt RC, Webber A, Sarkis JJF,Dias RD, Netto CA (1997) ATP diphosphohydrolase and 5�-nucleotidase activities from hippocampal synaptosomes after brainischemia. In: Ecto-ATPases: recent progress on structure and func-tion (Plesner L, Kirley TL, Knowles AF, eds), pp 213–219. NewYork: Plenum Press.

chetinger MRC, Bonan CD, Schierholt RC, Webber A, Arteni N,Emanuelli T, Dias RD, Sarkis JJF, Netto CA (1998) Nucleotidehydrolysis in rats submitted to global cerebral ischemia: a possiblelink between preconditioning and adenosine production. J StrokeCerebrovasc Dis 7:281–286.

evigny J, Sundberg C, Braun N, Guckelberger O, Csizmadia E, QawiI, Imai M, Zimmermann H, Robson SC (2002) Differential catalyticproperties and vascular topography of murine nucleoside triphos-phate diphosphohydrolase (NTPDase1) and NTPDase2 have im-plications for thromboregulation. Blood 99:2801–2809.

kellett RA, Crist JR, Fallon M, Bobbin RP (1996) Chronic low-levelnoise exposure alters distortion product otoacoustic emissions.Hear Res 98:68–76.

kellett RA, Chen C, Fallon M, Nenov AP, Bobbin RP (1997) Phar-macological evidence that endogenous ATP modulates cochlear

poendlin H (1971) Primary structural changes in the organ of Cortiafter acoustic overstimulation. Acta Otolaryngol 71:166–173.

weeney MI (1997) Neuroprotective effects of adenosine in cerebralischemia, window of opportunity. Neurosci Biobehav Rev 21:207–217.

horne PR, Munoz DJB, Nikolic P, Mander L, Jagger D, Greenwood D,Vlajkovic SM, Housley GD (2002) Potential role of purinergic sig-nalling in cochlear pathology. Audiol Neuro-Otol 7:180–184.

irginio C, MacKenzie A, Rassendren FA, North RA, Surprenant A(1999) Pore dilation of neuronal P2X receptor channels. Nat Neu-rosci 2:315–321.

lajkovic SM, Thorne PR, Munoz DJB, Housley GD (1996) Ectonucle-otidase activity in the perilymphatic compartment of the guinea pigcochlea. Hear Res 99:31–37.

lajkovic SM, Thorne PR, Housley GD, Munoz DJB, Kendrick IS(1998a) Ecto-nucleotidases terminate purinergic signalling in thecochlear endolymphatic compartment. Neuroreport 9:1559–1565.

lajkovic SM, Thorne PR, Housley GD, Munoz DJB, Kendrick IS(1998b) The pharmacology and kinetics of ecto-nucleotidases inthe perilymphatic compartment of the guinea-pig cochlea. Hear Res117:71–80.

lajkovic SM, Housley GD, Greenwood D, Thorne PR (1999) Evi-dence for alternative splicing of ecto-ATPase associated with ter-mination of purinergic transmission. Mol Brain Res 73:85–92.

lajkovic SM, Thorne PR, Sevigny J, Robson SC, Housley GD (2002a)Distribution of ectonucleoside triphosphate diphosphohydrolases 1and 2 in rat cochlea. Hear Res 170:48–59.

lajkovic SM, Thorne PR, Sevigny J, Robson SC, Housley GD (2002b)NTPDase1 and NTPDase2 immunolocalisation in mouse cochlea:implications for regulation of P2 receptor signalling. J HistochemCytochem 50:1435–1441.

ang JCC, Raybould NP, Luo L, Ryan AF, Cannell MB, Thorne PR,Housley GD (2003) Noise induces up-regulation of P2X2 receptorsubunit of ATP-gated ion channels in the rat cochlea. Neuroreport14:817–823.

ang Y, Liberman MC (2002) Restraint stress and protection fromacoustic injury in mice. Hear Res 165:96–102.

oshida N, Liberman MC (1999) Heat stress and protection fromacoustic injury in mice. J Neurosci 19:10116–10124.

heng X, Henderson D, McFadden SL, Hu B (1997) The role of thecochlear efferent system in acquired resistance to noise-inducedhearing loss. Hear Res 104:44–56.

immermann H (1996) Biochemistry, localization and functional rolesof ecto-nucleotidases in the nervous system. Prog Neurobiol 49:589–618.

immermann H (2001) Ectonucleotidases: some recent developmentsand a note on nomenclature. Drug Dev Res 52:44–56.

immermann H, Braun N, Kegel B, Heine P (1998) New insights intomolecular structure and function of ecto-nucleotidases in the ner-

(Accepted 7 April 2004)(Available online 18 May 2004)