This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

TRPV1, NK1 receptor and substance P immunoreactivity and gene expressionin the rat lumbosacral spinal cord and urinary bladder after systemic,low dose vanilloid administration

Yujing J. Heng, Cassandra I.M. Saunders, Dale A. Kunde, Dominic P. Geraghty ⁎School of Human Life Sciences, University of Tasmania, Launceston 7250, Tasmania, Australia

a b s t r a c ta r t i c l e i n f o

Article history:Received 10 August 2010Received in revised form 13 January 2011Accepted 8 February 2011Available online 15 February 2011

Keywords:CapsaicinResiniferatoxinTachykininsImmunohistochemistrymRNAqPCR

Transient receptor potential vanilloid 1 (TRPV1), neurokinin 1 (NK1) receptor and substance P (SP)immunoreactivity (-ir) andmRNA in the rat lumbosacral spinal cord and urinary bladder were measured 24 hafter s.c. injection of the vanilloids, capsaicin (50 mg/kg) and resiniferatoxin (RTX, 100 μg/kg), or vehicle (10%ethanol/10% Tween 80/saline). In the spinal cord, capsaicin significantly reduced TRPV1 and SP-ir (40–45%) inlaminae I/II compared to controls, while RTX produced decreases of ~35%. NK1-ir in the spinal cord wasunaffected by both vanilloid treatments. In the bladder, SP-ir was reduced in urothelial cells of somecapsaicin- and RTX-treated rats, while SP-ir in the suburothelium andmuscularis was significantly reduced byRTX. A significant increase in NK1-ir was observed in the urothelium and muscularis after capsaicinadministration. Capsaicin significantly increased SP mRNA in the spinal cord, and TRPV1 and SP mRNA in thebladder, whereas RTX increased TRPV1, SP and NK1 mRNA in the spinal cord, and TRPV1 and SP mRNA in thebladder. These data suggest that stimulation of TRPV1 by low dose vanilloid administration can rapidly(within 24 h) alter both transcription and translation of TRPV1 channels, SP and NK1 receptors in the raturinary bladder and spinal cord.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Transient receptor potential vanilloid subtype 1 (TRPV1) channelsare activated by heat (N43 °C), protons (pHb5.9) and alkali (pHN8.0)[1], vanilloids, including capsaicin and resiniferatoxin (RTX), arachi-donic acid metabolites, particularly the endogenous cannabinoid/vanilloid, N-arachidonoyl-dopamine, and inflammatory mediators,such as tumour necrosis factor-α (TNF-α) and other cytokines (for areview see [2]). Originally thought to be located exclusively on thecentral and peripheral terminals of unmyelinated C-fibres and lightlymyelinated Aδ-fibres, a more ubiquitous distribution throughout theneuroaxis and on non-neuronal cells is now widely accepted [3–5].

Activation of TRPV1 channels on C- and Aδ-fibres leads to therelease of neuropeptides, including substance P (SP), from bothcentral and peripheral neuron terminals [6]. SP binds to neurokinin 1(NK1) receptors which excite ascending neurons with subsequenttransmission of nociceptive information to the brain [7] and alsotrigger a variety of autonomic reflex responses [6]. Peripherally, SPrelease mediates contraction of smooth muscle. Capsaicin-likevanilloids, but not other TRPV1 agonists, when repeatedly or

chronically applied to TRPV1 channels in high doses on capsaicin-sensitive sensory fibres, are capable of selectively exciting, desensitis-ing and eventually depleting these neurons of neuropeptides(especially SP) [8]. This phenomenon has been proposed to contributeto the long-lasting analgesic effect of capsaicin and RTX [9]. Inaddition, at sufficiently high doses, capsaicin is neurotoxic. Forexample, a single injection of capsaicin (≥50 mg/kg s.c.) into neonatalrats results in N90% ablation of Aδ- and C-fibres [10]. In adult rats,cumulative s.c. administration of≥125 mg/kg produces extensive lossof spinal afferent terminals [10]. Functional, biochemical andstructural responses to neuron ablation are usually investigatedsome time after administration of neurotoxic doses of capsaicin (e.g.,treated neonatal rats are allowed to grow to maturity whereas treatedadult rats are allowed to recover for 10–14 days after treatment), andmost likely represent long-term adaptive changes to neuron ablation.However, there are no studies that report the more acute effects ofpresumably non-neurotoxic (low) doses of capsaicin (i.e., b125 mg/kg)and RTX (b200 μg/kg) on mRNA levels and TRPV1 channel, SP and NK1receptor immunoreactivity (-ir).

The urinary bladder is richly innervated by capsaicin-sensitive(TRPV1-expressing) Aδ- and C-fibres that detect bladder distensionand/or chemical irritants and which terminate in the lumbosacralspinal cord. In addition to neuron terminals, TRPV1-ir and/or vanilloidbinding sites have also been described on smooth muscle cells andnon-neuronal cells of the bladder, in particular, the urothelium [11–

Regulatory Peptides 167 (2011) 250–258

⁎ Corresponding author at: School of Human Life Sciences, University of Tasmania,Locked Bag 1320, Launceston 7250, Tasmania, Australia. Tel.: +61 3 63245488;fax: +61 3 63243658.

E-mail address: D.Geraghty@utas.edu.au (D.P. Geraghty).

0167-0115/$ – see front matter © 2011 Elsevier B.V. All rights reserved.doi:10.1016/j.regpep.2011.02.004

Contents lists available at ScienceDirect

Regulatory Peptides

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

Author's personal copy

14]. Hence, the rat urinary bladder and lumbosacral spinal cord areideal tissues to study the regulation of central and peripheralneurotransmitter/neurotransmitter receptor changes following vanil-loid-induced disruption of TRPV1 function. The aim of the presentstudy was to investigate TRPV1, SP and NK1 receptor protein and geneexpression in the adult rat lumbosacral spinal cord and urinarybladder, 24 h after stimulation of TRPV1 by systemic administration oflow doses of capsaicin and RTX using double labelling immunofluo-rescence and quantitative real time (qRT)-PCR.

2. Materials and methods

2.1. Experimental animals

Animal experiments were approved by the University of TasmaniaAnimal Ethics Committee (approval number A7193) and all proce-dures adhered to the guidelines of the Australian Code of Practice forthe Care and Use of Animals for Scientific Purposes [15]. Twenty eightadult male Hooded Wistar rats 190–210 g (Central Animal House,University of Tasmania, Hobart, TAS, Australia) were kept in an air-conditioned facility maintained at 21 °C±1 °C with a 12 h light/darkcycle. Commercial food and water were replenished daily ad libitum.

Vehicle (10% ethanol and 10% Tween 80 in saline), 50 mg/kgE-capsaicin (Tocris, Bristol, UK) or 100 μg/kg RTX (Tocris), dissolvedin vehicle, were administered as a single s.c. injection in the looseskin on the back of the neck under sodium pentobarbitoneanaesthesia (60 mg/kg, i.p.; Boehringer Ingelheim, Artarman, NSW,Australia). These doses were chosen based on the premise that125 mg/kg capsaicin is required to produce permanent ablation ofcapsaicin-sensitive neurons in adult rats [10]. As RTX is known to beabout 100- to 1000-foldmore potent than capsaicin, a 500-fold lowerdose of RTX was used. The rats were kept warm with a heating padand radiator for 3 h after vehicle or vanilloid administration.

2.2. Immunohistochemistry

Twenty-four hours after systemic administration of vehicle,capsaicin or RTX, the rats were sacrificed. Rats (n=4 per treatment)were administered 500 IU of heparin (Weddel Pharmaceuticals,Thornleigh, NSW, Australia) before being deeply anaesthetised withsodium pentobarbitone (60 mg/kg, i.p.). The absence of withdrawaland head-shake reflexes indicated that the degree of anaesthesia wasadequate for dissection.

An abdominal midline incision was made. The urinary bladder wasexposed and drained of urine using a syringe and a 28 ga needle. Allrats were perfused via the left ventricle (ascending aorta) with freshlyprepared 0.1 M phosphate buffered saline (PBS) for 2 min followed by4% paraformaldehyde in 0.1 M PBS for 7 min. Bladders were excisedfollowed by the intact spinal cord using a modified method of deSousa and Horrocks [16]. Briefly, the vertebral column was cut at C3and the cauda equina. A 16 ga needle attached to a 10 ml syringe filledwith 0.1 M PBS was inserted into the caudal central canal and, usinghydraulic pressure, the spinal cord was rapidly ejected from thecervical opening. The lumbosacral spinal cord (L1-S2) and bladderwere post-fixed in 4% paraformaldehyde in 0.1 M PBS at 4 °C for 6 h,washed 3×20 min in 0.1 M PBS, before being stored in 30% sucrosesolution in 0.1 M PBS at 4 °C for at least 24 h for cryoprotection. Serialfrozen sections (30 μm) were obtained using a CO2-cooled freezingmicrotome (Leica CM1325, Wetzlar, Germany). Sections werecollected in 0.01 M PBS as free-floating sections and stored at 4 °C.

Indirect immunofluorescence was performed to visualise TRPV1,SP and NK1 receptors in the rat lumbosacral spinal cord and bladder.Sections were immersed in blocking solution [5% bovine serumalbumin (Sigma-Aldrich, St Louis, MO, USA) in 0.01 M PBS with 0.3%Triton X] for 1 h at room temperature. Double-staining experimentswere performed to assess co-localisation and for detailed comparison

of the distribution of TRPV1-, SP- and NK1 receptor-ir. Sections wereplaced in blocking solution containing a mixture of 2 primaryantibodies at 1:500 dilution of either rabbit anti-SP (BioGenex, SanRamon, CA, USA) and guinea pig anti-NK1 (Chemicon, Temecula, CA,USA), or rabbit anti-SP and guinea pig anti-TRPV1 (Chemicon)antibodies. The sections were incubated overnight at 4 °C.

After incubating with primary antibodies, sections were washed3×15 min in 0.01 M PBS with 0.3% Triton X, followed by 1.5 h darkincubation at room temperature with secondary goat IgG at 1:1000dilution, conjugated with either anti-rabbit Alexa Fluor 488 (green;Molecular Probes, Eugene, OR, USA) for SP or anti-guinea pig AlexaFluor 594 (red; Molecular Probes) for NK1 receptor and TRPV1.Washes were repeated. Sections were then collected onto slides, airdried in the dark and mounted with Permafluor (Beckman Coulter,Marseille, France). To test for non-specific immunofluorescenceproduced by the secondary antibodies, primary antibody wasexcluded in negative controls.

All slides were viewed using anOlympus IX71 inverted fluorescencemicroscope (Melville, NY, USA). Images were captured with anOptronics 60800 digital camera (Goleta, CA, USA). Exposure times andlight intensity were optimised using vehicle sections for quantitativepurposes. All images were captured and colourmerged using OptronicsMagnaFire software (version 2.1). Images for visualisation wereimported into Adobe Photoshop (version 8.0) (San Jose, CA, USA) tooptimise brightness, contrast and sharpness. Where images fromvehicle, capsaicin and RTX administration were visually compared, allimages were enhanced (brightness and contrast) to the same extent.

The merged images (40× objective lens) were split, converted togreyscale and inverted so that fluorescence appeared as black on awhite background. Immunoreactivity was quantified by measuringrelative optical densities (ROD) of the tissues using Analytical ImagingStation (software version 3.0; Imaging Research, St Catherines,Ontario, Canada). Three images per tissue from each rat wereanalysed. For the lumbosacral spinal cord (L4-S1), four RODmeasurements of each lamina (laminae I to X) were obtained usinga 4000 μm2 template. The atlas of Paxinos and Watson [17] andhaematoxylin and eosin-stained sections were used as a guide to thelaminae. In the bladder, four ROD measurements were obtained fromthe urothelium, lamina propria (subepithelium) and muscularis usinga 400 μm2 template. To provide specific fluorescence, the average offour ROD measurements of the background (glass slide) wassubtracted from the raw ROD values of each image.

2.3. Quantitative real-time RT-PCR (qRT-PCR)

Twenty-four hours after a single s.c. injection of vehicle (n=6),capsaicin (n=5) or RTX (n=5), rats were sacrificed using a lethaldose of sodium pentobarbitone (200 mg/kg, i.p.). The lumbosacralspinal cord and bladder were quickly removed (as before) and placedin RNAlater (Ambion, Austin, TX, USA).

Lumbosacral spinal cords and bladders were removed fromRNAlater and weighed. To extract RNA, the tissue was cut up finelywith a scalpel blade and placed into 1 ml of QIAzol Lysis Reagent(QIAGEN, Valencia, CA, USA). The tissue was homogenised at highspeed using the Omni Tissue Homogeniser (Omni International,Marietta, GA, USA)with a 10 mmprobe for 3×20 s. The tissuewas leftto stand on the bench for 5 min and processed according to theRNeasy Lipid Tissue mini kit protocol (QIAGEN). The final volume ofextracted total RNA was 60 μl.

To obtain cDNA, 5 μl of RNA extracted from each tissue was addedto 45 μl of Master Mix (Omniscript Reverse Transcriptase, QIAGEN)containing 10× Reverse Transcriptase Buffer, dNTP mix, ReverseTranscriptase, Oligo dT (Promega, Madison, WI, USA), RNase Inhibitor(Promega) and water. The tubes were briefly centrifuged andprocessed in a Peltier Thermal Cycler 150 (MJ Research, Waltham,

251Y.J. Heng et al. / Regulatory Peptides 167 (2011) 250–258

Author's personal copy

MA, USA) for 1 h at 44 °C, followed by 92 °C for 15 min to terminatethe process.

qRT-PCRwas performed by theMYiQ™ single colour real-time PCRdetection system i-Cycler (Bio-Rad Laboratories, Austin, TX, USA)using iQ™ SYBR Green Supermix (Bio-Rad) with the following specificprimers: glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (for-ward: 5′-GGC ATG GAC TGT GGT CAT GA-3′, reverse: 5′-TGC ACC ACCAAC TGC TTA GC-3′), SP (forward: 5′-ACC AAA TCA AGG AGG CAA TG-3′, reverse: 5′-AGC CTT TAA CAG GGC CAC TT-3′), NK1 receptor(forward: 5′-CGT CCT TCC TAT GGA CTC TGA-3′, reverse: 5′-TCA TAAGTC CTG TTG GGA TGC-3′) and TRPV1 (forward: 5′-CCT TAA GCC AGAGGA TGC TG-3′, reverse: 5′-TGG TTC CCT AAG CAG ACC AC-3′). TheqRT-PCR protocol for all was: 95 °C for 3 min; 40 cycles of 95 °C for20 s, 55 °C for 20 s and 72 °C for 20 s; 95 °C for 1 min; 65 °C for 1 min;30 cycles of 65 °C for 10 s. Specific cultured plasmids containing thedesired mRNA sequences were used to obtain the standard curve.cDNA and plasmids were performed in triplicate. Raw copy numbersof mRNA obtained for each rat were normalised against GAPDH andexpressed as copies per 106 copies GAPDH.

2.4. Statistical analysis

Comparison of data for immunohistochemistry (ROD measure-ments) and qRT-PCR (mRNA copies) for vehicle-, capsaicin- and RTX-treated rats was performed using (multivariate) general linearmodelling with correction for repeated measures and calculation ofrobust standard errors using STATA (version 8.2; Intercooled Stata,College Station, TX, USA). Pb0.05 was considered statisticallysignificant. All replicate results were included in the statisticalanalysis and data are presented as mean±S.E.M. of 4 to 6 rats pergroup.

3. Results

3.1. General distribution of TRPV1, SP and NK1 receptorimmunoreactivity in the rat lumbosacral spinal cord

Strongly immunoreactive TRPV1 fibres were found in lamina I andlight TRPV1-ir was observed in lamina II (Fig. 1A). The area aroundthe central canal was intensely TRPV1 positive. Large motor neuroncell bodies in the ventral horn were lightly TRPV1 positive (notshown). SP-ir was intense in laminae I/II of the dorsal horn (Fig. 1B)and in lamina X (ventral to the central canal) (not shown). Laminae Vand VI had moderate SP-ir with some thick strands of SP fibres alsoobserved within these two laminae (Fig. 1E, arrow). SP co-localisedwith TRPV1 immunoreactive fibres in laminae I/II (Fig. 1C). NK1fibres were located mainly in laminae I/II and X, co-localising withdense SP-ir in laminae I/II (Fig. 1F). NK1-ir was most intense inlaminae VII. Some medium sized, strongly NK1 immunoreactiveneurons were found at the lateral border of laminae VI/VII extendingto the central canal, with web-like sprouting of fibres into the lateralspinal nucleus. These neuron cell bodies were also intensely SPpositive (not shown).

3.2. Effect of capsaicin and RTX on TRPV1, SP and NK1 receptorimmunoreactivity in the rat lumbosacral spinal cord

Relative optical densities of TRPV1-, SP- and NK1-ir werecompared in the rat lumbosacral spinal cord in laminae I to X, 24 hafter s.c. administration of vehicle, capsaicin or RTX (Fig. 2). There wasa statistically significant decrease (45%) in TRPV1-ir in laminae I/II incapsaicin-treated rats when compared to vehicle-treated rats(p=0.005) (Fig. 3D). There was also a 35% decrease in TRPV1-ir inlaminae I/II in RTX-treated rats compared to vehicles (p=0.051)(Fig. 3G). In the remaining laminae, there were no significant

differences in TRPV1-ir in either capsaicin or RTX-treated rats whencompared to vehicles.

Significant decreases in SP-ir in laminae I/II of 40% and 37% wereobserved in rats treated with capsaicin (p=0.001) and RTX(p=0.012), respectively, when compared to vehicles (Fig. 3E andH). Capsaicin-treated rats demonstrated a significant decrease (19%)in SP-ir in laminae III compared to vehicles (p=0.023; Fig. 2B). Ratstreated with either vanilloid demonstrated insignificant decreases inSP-ir compared to vehicles in laminae IV–X.

Neither capsaicin nor RTX treatment had a substantial effect onNK1-ir in the lumbosacral spinal cord. Laminae I/II–VII displayed smallincreases of 0.1% to 17% in NK1-ir in capsaicin-treated rats comparedto vehicles, but these differences were not statistically significant(Fig. 3F). RTX-treated rats demonstrated small but not statisticallysignificant increases of 3% to 15% in NK1-ir in laminae V, VI, VII and X(Fig. 3I).

3.3. General distribution of TRPV1, SP and NK1 receptorimmunoreactivity in the rat urinary bladder

TRPV1-, SP- and NK1-ir were most intense in the urothelium,followed by the muscularis (Fig. 1G, H and J). TRPV1-ir was alsoobserved immediately beneath the urothelium, co-localising with SP(Fig. 1M). Strong TRPV1-ir was found on or in urothelial cells (Fig. 1G).Numerous, punctate SP fibres were present in themuscularis, runningparallel to the smooth muscles (Fig. 1O). There was some co-localisation of TRPV1-ir with SP immunoreactive varicosities in themuscularis. Strong SP-ir was also observed in urothelial cells (Fig. 1N).NK1-ir co-localised with SP in the smooth muscle cells of themuscularis (Fig. 1L).

3.4. Effect of capsaicin and RTX on TRPV1, SP and NK1 receptorimmunoreactivity in the rat urinary bladder

Relative optical densities of TRPV1-, SP- and NK1-ir in the ratbladder were compared in vehicle-, capsaicin- and RTX-treated rats(Fig. 4). Significant (42% and 55%) decreases in TRPV1-ir wereobserved in the subepithelium of the bladder after capsaicin(p=0.048) and RTX administration (p=0.005), respectively, whencompared to vehicles (Fig. 3M and P). TRPV1-ir in the muscularis ofRTX-treated rats was significantly lower (44%, p=0.032) thancontrols, whereas the decrease following capsaicin was not statisti-cally significant (p=0.246).

Whilst SP-ir decreased by 33% and 35% in the urothelium ofcapsaicin- and RTX-treated rats, respectively, when compared tovehicles (Fig. 4B), and also decreased by 30% in both the subepithe-lium and muscularis in capsaicin-treated rats (Fig. 3N), thesedifferences were not statistically significant (p=0.067 top=0.118). RTX administration, however, produced a significant(40%) decrease in the SP-ir in the subepithelium (p=0.007;Fig. 3Q) and in the muscularis (38%, p=0.047).

NK1-ir was significantly increased (76%, p=0.012) in theurothelium of capsaicin-treated rats (Fig. 4C). However, no apparentchange was detected in RTX-treated rats. NK1-ir was similar invehicle-, capsaicin- and RTX-treated rats in the subepithelium. Therewas a significant (75%) increase in NK1-ir in the muscularis followingcapsaicin administration (p=0.015) (Fig. 3O). In contrast, RTXadministration had minimal effects on NK1-ir in the muscularis(Fig. 3R).

3.5. Effect of capsaicin and RTX on TRPV1, SP and NK1 receptor mRNA inthe rat lumbosacral spinal cord

TRPV1, SP and NK1 receptor mRNA copies in the ratlumbosacral spinal cord were compared 24 h after s.c. adminis-tration of vehicle, capsaicin or RTX (Fig. 5). There was no

252 Y.J. Heng et al. / Regulatory Peptides 167 (2011) 250–258

Author's personal copy

difference in the quantity of TRPV1 mRNA obtained from thespinal cords of vehicle- and capsaicin-treated rats (p=0.296).However, RTX-treated rats displayed a significant (50%) increasein TRPV1 mRNA expression (p=0.015). Capsaicin administrationincreased SP mRNA 2.4-fold (p=0.002), whereas RTX adminis-

tration caused a marked, 16.2-fold increase in the quantity of SPmRNA (p=0.010). Vehicle- and capsaicin-treated rats had similarquantities of NK1 receptor mRNA, whereas there was a significant(2-fold) increase of NK1 receptor mRNA in RTX-treated rats(p=0.001).

Fig. 1. Double immunostaining of transient receptor potential vanilloid 1 (TRPV1; red) or neurokinin 1 (NK1; red) with substance P (SP; green) in rat lumbosacral spinal cord (A–F)or rat urinary bladder (G–O). (A) Intense TRPV1 immunoreactivity (-ir) observed in lamina I of the dorsal horn of a vehicle rat (arrow). (B) Abundant SP-ir in laminae I/II, showingdefinitive co-localisation with TRPV1-ir in (C). (D) NK1-ir is present in laminae I/II, co-localising with SP-ir in lamina I (F). (G) Strong TRPV1-ir (red) was found on the urothelium.(J) Dense NK1-ir was observed in the urothelium andmuscularis. (M) TRPV1-ir co-localised with SP immediately beneath the urothelium. (N) Strong SP-ir was observed in urothelialcells and (O) running parallel to the smooth muscle cells. (P) Negative control was performed by omittance of primary antibody. Scale bar=500 μm in A–L, 50 μm in M and O and100 μm in N.

253Y.J. Heng et al. / Regulatory Peptides 167 (2011) 250–258

Author's personal copy

3.6. Effect of capsaicin and RTX on TRPV1, SP and NK1 receptor mRNA inthe rat urinary bladder

TRPV1, SP and NK1 receptor mRNA in the rat bladder werecompared 24 h after s.c. administration of vehicle, capsaicin or RTX(Fig. 6). There were significant increases in the amount of TRPV1mRNA in capsaicin- (75%, pb0.001) and RTX-treated rats (32%,p=0.010). SP mRNA was increased 1.6-fold in capsaicin-treated rats(p=0.024) and 5-fold in RTX-treated rats (p=0.001). There were nosignificant differences in NK1 receptor mRNA expression in theurinary bladders of vehicle- and vanilloid-treated rats.

4. Discussion

The present study shows for the first time that stimulation ofTRPV1 by systemic administration of low doses of capsaicin and RTXproduces significant changes in protein and mRNA expression ofTRPV1, SP and NK1 receptors in the rat lumbosacral spinal cord andbladder within 24 h of administration. TRPV1-ir detected in the dorsalhorn of the lumbosacral spinal cord undoubtedly represents theneuronal terminals of capsaicin-sensitive Aδ- and C-fibres. Lessintense and more uniform TRPV1-ir was also observed in laminaeIII–X. A similar pattern has been reported recently by others [18,19],although Cruz et al. [20] observed minimal TRPV1-ir outside oflaminae I/III, suggesting that in the present study, at least somebackground staining may be present in laminae III–X. There was co-localisation of SP and TRPV1-ir and dense plexuses of NK1immunoreactive fibres in laminae I/II. These observations are similarto the elegant studies previously reported by Mantyh et al. [21] andothers [19]. The mismatches in SP- and NK1-ir in laminae I/II that areobserved in this study suggests that SP may not bind exclusively toNK1 receptors in the spinal cord.

TRPV1-ir was reduced in laminae I/II after capsaicin and RTXadministration, suggesting selective destruction of TRPV1 channelsand/or TRPV1 expressing neuron terminals in the lumbosacral spinalcord. In conjunction with the reported 40% reduction in SP-ir aftercapsaicin and RTX administration in laminae I/II, this (apparent)partial ablation of TRPV1 expressing terminals in laminae I/II appearsto be associated with a loss of SP. These results are consistent withprevious studies in which capsaicin and RTX abolished or reducedTRPV1-ir in rat dorsal horns [18,19,22–24]. In addition, our study hasdemonstrated that TRPV1-ir is reduced by much lower doses of eithervanilloid and over a much shorter timeframe (24 h) than previouslyreported. For example, Jeffry et al. [19] assessed TRPV1-ir 10 days and5 months after 20 days RTX administration (1.9 μg/kg per day i.t.),whereas Pan and colleagues' [18,24] studies were performed 4–5 weeks after a single injection of RTX (200 μg/kg i.p.). As desensitisa-tion of capsaicin-sensitive sensory neurons is generally attributed tothe ‘over’ release and depletion of SP from their neuronal terminalsafter vanilloid application [8,25-28], systemic administration of lowdoses of capsaicin and RTX may be adequate to deplete SP since SP-irwas observed in the present study to be significantly reduced after24 h.

Unlike TRPV1- and SP-ir, NK1-ir was not significantly affected bysystemic vanilloid administration. As NK1 receptors are undoubtedlypostsynaptic, changes in NK1-ir are unlikely to be similar to thoseobserved for SP- and TRPV1-ir (presynaptic). NK1 receptors are rapidlyinternalised within minutes of exposure to SP and return to baseline(pre-exposure) levels on the cell surface after 30 to 60 min [29–31].Hence, whereas capsaicin and RTX administration might initiallystimulate SP release leading to NK1 receptor internalisation in laminaeI/II, one would assume that NK1 receptors would have returned to thecell surface by 24 h. More importantly, the NK1 receptor antibody usedin the present study is capable of picking up cell surface as well asinternalised receptors [29,32]. Thus, although there was no obviouschange in NK1 receptor complement, there may have been aredistribution of NK1 receptors (i.e. increased internalised receptorsthat were non-functional or increased cell surface receptors makingneurons more responsive). Nevertheless, no ‘bead-like’ cytoplasmicstructures, indicative of internalised receptors, were observed [30].

Fig. 2. (A) Transient receptor potential vanilloid 1 (TRPV1), (B) substance P (SP) and(C) neurokinin 1 (NK1) receptor immunoreactivity in rat lumbosacral spinal cord 24 hafter s.c. administration of vehicle, 50 mg/kg capsaicin or 100 μg/kg resiniferatoxin. Barsrepresent the mean (±SEM) relative optical density (ROD) of 4 rats per group.* Indicates significantly different from vehicle.

Fig. 3. Transient receptor potential vanilloid 1 (TRPV1), substance P (SP) and neurokinin 1 (NK1) receptor immunoreactivity (-ir) in rat lumbosacral spinal cord (A–I) and urinarybladder (J–R) 24 h after s.c. administration of vehicle, 50 mg/kg capsaicin or 100 μg/kg resiniferatoxin (RTX). TRPV1-ir in laminae I/II was reduced in both capsaicin- and RTX-treatedrats (D, G). There was a decrease in SP immunoreactive fibres in laminae I/II in rats treated with capsaicin (E). Small increases in NK1-ir were observed in both capsaicin- and RTX-treated rats in all laminae (F, I). A reduction in TRPV1-ir in the subepithelium was observed in both capsaicin- and RTX-treated rats (M, P), whereas a reduction in TRPV1-ir in themuscularis was only seen in RTX-treated animals. A decrease in SP-ir was observed in RTX- (Q), but not capsaicin-treated rats (N) in the muscularis. NK1-ir was increased in theurothelium and muscularis of capsaicin-treated rats (O), while in RTX-treated animals (R), no apparent change was detected. Scale bar=500 μm.

254 Y.J. Heng et al. / Regulatory Peptides 167 (2011) 250–258

Author's personal copy

255Y.J. Heng et al. / Regulatory Peptides 167 (2011) 250–258

Author's personal copy

To our knowledge, this is the first study to use qRT-PCR to measureTRPV1, SP andNK1 receptormRNA levels in the lumbosacral spinal cordand bladder after acute systemic administration of vanilloids. TRPV1, SPandNK1 receptormRNAexpression in the lumbosacral spinal cordwereall significantly increased 24 h after RTX administration, whereascapsaicin induced a significant increase in SP mRNA only. Althoughthe dose of RTX administered in this study was 500-fold lower than thedose of capsaicin, there was a 7-fold greater increase in SP mRNA and amarked increase in TRPV1 mRNA in RTX-treated animals compared tocapsaicin-treated rats. It appears that the superior potency of RTXcompared to capsaicin at producing desensitisation [9] also holds truefor inducing SP and TRPV1mRNA expression changes at the lower dosesused in the present study. There appears to be no difference in NK1receptor mRNA 24 h after s.c. administration of capsaicin whilst NK1mRNA doubled after RTX treatment compared to vehicle. A previousstudy by Szallasi et al. [33] reported that NK1 receptor mRNA in the ratlumbar dorsal horn remained unchanged at 2 and 8 weeks after RTX

administration. SinceRTX is known tohave a faster action thancapsaicin[34], NK1 receptor mRNAmay be initially increased by RTXwithin 24 hof administration and multiple factors may return NK1 receptor mRNAback to pre-vanilloid treatment levels [33].

The observation in this study that TRPV1-ir is located in urothelialcells further confirms that this ion channel is not exclusively neuronalin its location [12–14]. In addition, some individual urothelial cellswere positive for SP. One might speculate that when noxiouschemicals are present in the bladder, TRPV1 channels are activatedon the urothelium, subsequently resulting in the release of SP by SPpositive urothelial cells. SP may, in turn, act on NK1 receptors presenton 1) urothelial cells resulting in the release of NO (nitric oxide) [12];2) sensory nerve endings leading to the perception of pain and itch[35]; 3) smooth muscles causing the detrusor muscle to contract,expelling urine as a form of defence [35,36]; and/or 4) theendothelium of blood vessels leading to vasodilation, swelling andinflammation [6,37]. Diffuse TRPV1-ir has also been reported inmouse [13,14] and rat bladder [13,14]. Although the specificity ofTRPV1 staining in both mouse and rat bladder has been questioned[38], it is interesting to note that both bladder mucosa and detrusor inhumans express significant quantities of TRPV1 mRNA [39]. Theapparent co-localisation of bead-like SP-ir in capsaicin-sensitivesensory fibres expressing TRPV1-ir in the smooth muscles and thedemonstration of TRPV1 channels and SP in/on urothelial cellssuggests that urothelial cells and capsaicin-sensitive sensory fibresmay work together to detect noxious stimuli in the bladder.

SP-ir appeared to be co-localised with NK1-ir in the urotheliumand muscularis, as reported by previous [125I]-SP and [125I]-SPanalogue autoradiographic studies [21,40–42]. The localisation ofNK1-ir corresponds well with the presence of SP and the contractilefunction of SP in the bladder.

Fig. 6.mRNA copies of (A) transient receptor potential vanilloid 1 (TRPV1), (B) substanceP (SP) and (C) neurokinin 1 (NK1) receptor in the rat urinary bladder 24 h after s.c.administration of vehicle (n=6), 50 mg/kg capsaicin (n=5) or 100 μg/kg resiniferatoxin(n=5). Bars represent the mean (±S.E.M.) copies per 106 copies of glyceraldehyde-3-phosphate dehydrogenase of each group. * Indicates significantly different from vehicle.

Fig. 5. mRNA copies of (A) transient receptor potential vanilloid 1 (TRPV1),(B) substance P (SP) and (C) neurokinin 1 (NK1) receptor in the rat lumbosacralspinal cord 24 h after s.c. administration of vehicle (n=6), 50 mg/kg capsaicin (n=5)or 100 μg/kg resiniferatoxin (n=5). Bars represent the mean (±S.E.M.) copies per 106

copies of glyceraldehyde-3-phosphate dehydrogenase of each group. * Indicatessignificantly different from vehicle.

Fig. 4. (A) Transient receptor potential vanilloid 1 (TRPV1), (B) substance P (SP) and(C) neurokinin 1 (NK1) receptor immunoreactivity in rat urinary bladder 24 h after s.c.administration of vehicle, 50 mg/kg capsaicin or 100 μg/kg resiniferatoxin. UR, urothelium;SE, subepithelium; MU, muscularis. Bars represent the mean (±S.E.M.) relative opticaldensity (ROD) of 4 rats per group. * Indicates significantly different from vehicle.

256 Y.J. Heng et al. / Regulatory Peptides 167 (2011) 250–258

Author's personal copy

Acute systemic administration of capsaicin and RTX appears todramatically reduce the number of TRPV1 channels (and/or neuronterminals) in the bladder. As expected, SP-ir in the bladder was alsogenerally decreased after vanilloid administration, although not alldecreases were statistically significant. Interestingly, NK1-ir in thebladder increased 24 h after capsaicin administration, while RTXadministration did not produce any effect. Morrison et al. [43]reported a 20% increase in [125I]Bolton-Hunter [Sar9,Met(O2)11]SPbinding sites in the rat bladder smooth muscles and epithelium inneonatal rats treated with 50 mg/kg of capsaicin, somewhat validat-ing the NK1 receptor results of this study. The explanation for theapparent increase in NK1 receptors after capsaicin administrationmay be to compensate for the lack or loss of SP. Nevertheless, there areconflicting reports regarding the effects of systemic capsaicin onNK1 receptors (assessed using [125I]-SP or SP analogues) in the ratbladder. Using autoradiography, Banasiak and Burcher [40] reportedno change in the density of binding sites in the rat bladder after s.c.administration of 150 mg/kg of capsaicin over 3 days. However,another related study by Geraghty and Burcher [44] showed a 25%decrease in the number of NK1 receptors in homogenates of ratbladder 2 weeks after capsaicin (150 mg/kg) administration. Eitherway, further studies to determine conclusively the effects of vanilloidadministration on NK1 receptors in the bladder are warranted.

TRPV1 and SP mRNA in the bladder were significantly increased24 h after vanilloid administration. TRPV1 mRNA may be increased tocompensate for the loss of actual TRPV1 channels (i.e., TRPV1-ir) afterbinding to the vanilloids. Similarly, the increase in SP mRNAtranscription may be to replace the loss of SP within the cells. Therewere no significant changes in NK1 receptor mRNA after theadministration of either vanilloid despite an increase in NK1-ir aftercapsaicin administration. Hence, capsaicin, but not RTX, appears tostimulate NK1 receptor translation, but not mRNA transcription.

5. Conclusions and perspective

Low doses of capsaicin and RTX appear to reduce the expression ofTRPV1 channels on both central (spinal cord) and peripheral(bladder) terminals of capsaicin-sensitive sensory fibres within 24 hof administration. This is associated with a decrease in the SP contentof these terminals. Although the functional consequences areunknown, the loss of TRPV1, and the subsequent loss of SP, maycontribute to decreased spinal reflex activity, irritation and mechan-ical stimulation in the bladder. A loss of TRPV1 channels and reducedSP on the bladder urothelium was also noted after vanilloidadministration. Therefore, in addition to diminished neuronalstimulation caused by vanilloids, urothelial cells may also becomeless sensitive to noxious stimuli. TRPV1 and SPmRNAwere apparently‘up-regulated’ in both spinal cord and urinary bladder 24 h aftervanilloid administration, possibly due to increased translation ofTRPV1 and SP as a compensatory mechanism for the loss of TRPV1channels (and/or neuronal terminals) and the decreased afferentinput from intracellular loss of SP. Future studies will determinewhether the acute alterations in SP, NK1 and TRPV1 expression inresponse to vanilloid administration reported here lead to long termchanges in urinary bladder responses to SP and TRPV1 agonists.

Acknowledgements

Many thanks to Dr. Iain Robertson for his assistance withstatistical analysis and Dr. Amanda Crawford for her help with themolecular biology work. This study was supported by a grant to DPGeraghty from the National Health and Medical Research Council ofAustralia (grant number 159401) and the University of Tasmania,Australia.

References

[1] Dhaka A, Uzzell V, Dubin A, Mathur J, Petrus M, Bandell M, Patapoutian A. TRPV1 isactivated by both acidic and basic pH. J Neurosci 2009;29:153–8.

[2] Gunthorpe M, Benham C, Randall A, Davis J. The diversity in the vanilloid (TRPV)receptor family of ion channels. Trends Pharmacol Sci 2002;23:183–91.

[3] Cortright D, Szallasi A. Biochemical pharmacology of the vanilloid receptor TRPV1.An update. Eur J Biochem 2004;271:1814–9.

[4] Geppetti P, Trevisani M. Activation and sensitisation of the vanilloid receptor:role in gastrointestinal inflammation and function. Br J Pharmacol 2004;141:1313–20.

[5] SanchezM, Sanchez A, Collado B, Malagarie-Cazenave S, Olea N, CarmenaM, PrietoJ, Diaz-Laviada I. Expression of the transient receptor potential vanilloid 1 (TRPV1)in LCaP and PC-3 prostate cancer cells and in human prostate tissue. Eur JPharmacol 2005;515:20–7.

[6] Szallasi A. Vanillid (capsaicin) receptors in health and disease. Am J Clin Pathol2002;118:110–21.

[7] Otsuka M, Yanagisawa M. Does substance P act as a pain transmitter? TrendsPharmacol Sci 1987;8:506–10.

[8] Jessell T, Iversen L, Cuello A. Capsaicin induced depletion of substance P fromprimary sensory neurons. Brain Res 1978;152:183–8.

[9] Szallasi A, Blumberg P. Vanilloid (capsaicin) receptors and mechanisms.Pharmacol Rev 1999;51:159–212.

[10] Szolcsanyi J. Actions of capsaicin on sensory receptors. In: Wood J, editor.Capsaicin in the study of pain. London: Academic Press; 1993. p. 1–27.

[11] Avelino A, Cruz C, Nagy I, Cruz F. Vanilloid receptor 1 expression in the rat urinarytract. Neuroscience 2002;109:787–98.

[12] Birder L, Kanai A, de Groat W, Kiss S, Nealen M, Burke N, Dineley K, Watkins S,Reynolds I, Caterina M. Vanilloid receptor expression suggests a sensory role forurinary bladder epithelial cells. Proc Nat Acad Sci USA 2001;98:13396–401.

[13] Imamura T, Ishizuka O, Zhong C, Ogawa T, Nakayama T, Kurizaki Y, Tanabe T,Nishizawa O, Andersson KE. An extract (THC-002) of Ba-Wei-Die-Huang-Waninhibits expression of tachykinins, and P2X3 and TRPV1 receptors, and inhibitsATP-induced detrusor overactivity in spontaneously hypertensive rats. NeurourolUrodyn 2009;28:529–34.

[14] Yamada T, Ugawa S, Ueda T, Ishida Y, Kajita K, Shimada S. Differential localizationsof the transient receptor potential channels TRPV4 and TRPV1 in the mouseurinary bladder. J Histochem Cytochem 2009;57:277–87.

[15] National Health and Medical Research Council of Australia. Australian code ofpractice for the care and use of animals for scientific purposes; 2004.

[16] de Sousa B, Horrocks L. Development of rat spinal cord: I.Weight and lengthwith amethod for rapid removal. Neuroscience 1979;2:115–21.

[17] Paxinos G, Watson C, editors. The rat brain in stereotaxic coordinates. UnitedStates of America: Academic Press Inc; 1986.

[18] Chen SR, Pan HL. Loss of TRPV1-expressing sensory neurons reduces spinal muopioid receptors but paradoxically potentiates opioid analgesia. J Neurophysiol2006;95:3086–96.

[19] Jeffry JA, Yu SQ, Sikand P, Parihar A, Evans MS, Premkumar LS. Selective targetingof TRPV1 expressing sensory nerve terminals in the spinal cord for long lastinganalgesia. PLoS ONE 2009;4:e7021.

[20] Cruz CD, Charrua A, Vieira E, Valente J, Avelino A, Cruz F. Intrathecal delivery ofresiniferatoxin (RTX) reduces detrusor overactivity and spinal expression ofTRPV1 in spinal cord injured animals. Exp Neurol 2008;214:301–8.

[21] Mantyh P, Gates T, Mantyh C, Maggio J. Autoradiographic localisation andcharacterisation of tachykinin receptor binding sites in the rat brain andperipheral tissues. J Neurosci 1989;9:258–79.

[22] Pan H, Kahan G, Alloway K, Chen S. Resiniferatoxin induces paradoxical changes inthermal and mechanical sensitivities in rats: mechanism of action. J Neurosci2003;23:2911–9.

[23] Szallasi A, Nilsson S, Farkas-Szallasi T, Blumberg P, Hokfelt T, Lundberg J. Vanilloid(capsaicin) receptors in the rat: distribution in the brain, regional differences inthe spinal cord, axonal transport to the periphery, and depletion by systemicvanilloid treatment. Brain Res 1995;703:175–83.

[24] Zhou HY, Chen SR, Chen H, Pan HL. The glutamatergic nature of TRPV1-expressingneurons in the spinal dorsal horn. J Neurochem 2009;108:305–18.

[25] Abbadie C, Brown J, Mantyh P, Basbaum A. Spinal cord substance P receptor-immunoreactivity increases in both inflammatory and nerve injury models ofpersistent pain. Neuroscience 1996;70:201–9.

[26] GamseR, LacknerD,GamseG, LeemanS.Effect of capsaicinpretreatment oncapsaicinevoked release of immunoreactive somatostatin and substance P from primarysensory neurons. Naunyn-Schmiedeberg's Arch Pharmacol 1981;316:38–41.

[27] Hayes A, Tyers M. Effects of capsaicin on nociceptive heat, pressure and chemicalthresholds and on substance P levels in the rat. Brain Res 1980;189:561–4.

[28] Velazquez R, McCarson K, Cai Y, Kovacs J, Shi Q, Evensji M, Larson A. Upregulationof neurokinin-1 receptor expression in rat spinal cord by an N-terminal metaboliteof substance P. Eur J Neurosci 2002;16:229–41.

[29] Garland A, Grady E, Payan D, Vigna S, Bunnett N. Agonist-induced internalizationof the substance P (NK1) receptor expressed in epithelial cells. Biochem J1994;303:177–86.

[30] Mantyh P, Allen C, Ghilardi J, Rogers S, Mantyh C, Liu H, Basbaum A, Vigna S,Maggio J. Rapid endocytosis of a G protein-coupled receptor: substance P-evokedinternalization of its receptor in the rat striatum in vivo. Proc Natl Acad Sci USA1995;92:2622–6.

[31] Mantyh P, DeMaster E, Malhotra A, Ghilardi J, Rogers S, Mantyh C, Liu H, BasbaumA, Vigna S, Maggio J, Simone D. Receptor endocytosis and dendrite reshaping inspinal neurons after somatosensory stimulation. Science 1995;268:1629–32.

257Y.J. Heng et al. / Regulatory Peptides 167 (2011) 250–258

Author's personal copy

[32] Bowden J, Garland A, Baluk P, Lefevre P, Grady E, Vigna S, Bunnett N, McDonald D.Direct observation of substance P-induced internalization of neurokinin 1 (NK1)receptors at sites of inflammation. Proc Natl Acad Sci USA 1994;91:8964–8.

[33] Szallasi A, Farkas-Szallasi T, Tucker J, Lundberg J,Hokfelt T, Krause J. Effects of systemicresiniferatoxin treatment on substance P mRNA in rat dorsal root ganglia andsubstance P receptor mRNA in the spinal dorsal horn. Brain Res 1999;815:177–84.

[34] Blumberg P, Szallasi A, Acs G. Resiniferatoxin— an ultrapotent capsaicin analogue.In: Wood J, editor. Capsaicin in the study of pain. London: Academic Press; 1993.p. 45–62.

[35] Cruz F. Vanilloid receptor and detrusor instability. Urology 2002;59:51–60.[36] Chancellor M, de Groat W. Intravesical capsaicin and resiniferatoxin therapy:

spicing up the ways to treat the overactive bladder. J Urol 1999;162.[37] Garret C, Carruette A, Fardin V, Moussaoui S, Peyronel J, Blanchard J, Laduron P.

Pharmacological properties of a potent and selective nonpeptide substance Pantagonist. Proc Nat Acad Sci 1991;88:10208–12.

[38] Everaerts W, Sepulveda MR, Gevaert T, Roskams T, Nilius B, De Ridder D. Whereis TRPV1 expressed in the bladder, do we see the real channel? Naunyn-Schmiedeberg's Arch Pharmacol 2009;379:421–5.

[39] Liu L, Mansfield KJ, Kristiana I, Vaux KJ, Millard RJ, Burcher E. The molecular basisof urgency: regional difference of vanilloid receptor expression in the humanurinary bladder. Neurourol Urodyn 2007;26:433–8 discussion 9; discussion51–3.

[40] Banasiak D, Burcher E. Effect of capsaicin on distribution of binding sites fortachykinins and calcitonin gene-related peptide in rat urinary bladder: aquantitative autoradiographic study. Peptides 1994;15:333–9.

[41] Burcher E. The study of tachykinin receptors. Clin Exp Pharmacol Physiol 1989;16:539–43.

[42] Quartara L, Maggi C. The tachykinin NK1 receptor. Part II: Distribution andpathophysiological roles. Neuropeptides 1998;32:1–49.

[43] Morrison J, Nimmo A, Whitaker E. The effect of neonatal capsaicin treatment onthe distribution of neurokinin binding sites in rat bladder. J Physiol 1989;423:79.

[44] Geraghty D, Burcher E. Effect of chronic capsaicin treatment on tachykinin NK1binding sites in the rat. Peptides 1992;13:409–11.

258 Y.J. Heng et al. / Regulatory Peptides 167 (2011) 250–258