European Journal of Pharmacology 707 (2013) 32–40

Contents lists available at SciVerse ScienceDirect

European Journal of Pharmacology

0014-29http://d

n CorrToulousfax: þ3

E-m

journal homepage: www.elsevier.com/locate/ejphar

Pulmonary, gastrointestinal and urogenital pharmacology

Relevance of the cyclophosphamide-induced cystitis model forpharmacological studies targeting inflammation and pain ofthe bladder

Céline Augé a,b,c,d,n, Gérald Chene e, Marc Dubourdeau e, Denis Desoubzdanne f,Bruno Corman f, Stefano Palea d, Philippe Lluel d, Nathalie Vergnolle a,b,c,Anne-Marie Coelho d

a INSERM, U1043, 31300 Toulouse, Franceb CNRS, U5282, 31300 Toulouse, Francec Université de Toulouse, UPS, Centre de Physiopathologie de Toulouse Purpan (CPTP), 31000 Toulouse, Franced UROsphere, Faculté des Sciences Pharmaceutiques, 31062 Toulouse, Francee Ambiotis, Incubateur Midi-Pyrénées, 29 rue Jeanne Marvig, 31400 Toulouse, Francef Profilomic, 31 rue d’Aguesseau, 92100 Boulogne-Billancourt, France

a r t i c l e i n f o

Article history:Received 10 September 2012Received in revised form1 March 2013Accepted 8 March 2013Available online 26 March 2013

Keywords:CystitisCyclophosphamideInflammationVisceral painUrinary biomarkerRat

99/$ - see front matter & 2013 Elsevier B.V. Ax.doi.org/10.1016/j.ejphar.2013.03.008

esponding author at: INSERM U1043, Centree-Purpan, BP3028, 31024 Toulouse Cedex 3, Fr3 562 74 45 28.ail address: celine.auge@inserm.fr (C. Augé).

a b s t r a c t

This work aimed at establishing the relevance of using the in vivo model of cyclophosphamide (CYP)-induced bladder inflammation in rats for in vivo pharmacological studies. Specifically, we measured visceralnociception, identified key inflammatory mediators and evaluated the effects of relevant pharmacologicaltreatments. Cystitis was induced in female rats by a single CYP injection. Sensitivity of the lower abdomento von Frey mechanical stimulation was determined as a nociceptive parameter. Bladders were assessed forweight, wall thickness and macroscopic damage. Inflammatory mediators were quantified in bladders andurines. The effects of aspirin, ibuprofen and morphine were investigated on all these parameters. A singleCYP injection increased nociceptive scores and decreased nociceptive threshold in response to mechanicalstimuli between 1 and 4 h post-administration. Increased bladder weight and wall thickness wereassociated with edema and hemorrhage. Bladder levels of IL-1β, IL-6, MCP-1 and VCAM, and urinary levelsof PGE2 were increased. In contrast, a decrease in the urinary metabolites, indoxyl sulfate and pantothenicacid, was observed. Aspirin, ibuprofen and morphine decreased CYP-induced referred visceral pain. Aspirinand ibuprofen also reversed the increased wall thickness, macroscopic damage and levels of IL-1β, IL-6 andPGE2, and the decreased panthotenic acid levels. In contrast, morphine increased wall thickness, edema,hemorrhage, and bladder IL-6 and MCP-1 levels. This work presents a new and reliable method to evaluatevisceral sensitivity in rats, and new relevant biomarkers identified in the bladder and urine to measureinflammation and pain parameters for in vivo pharmacological studies.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

Bladder pain syndrome is a chronic inflammatory disease ofunknown etiology characterized by urinary frequency, urgency andsuprapubic pain (Parsons, 2007). Cyclophosphamide (CYP)-inducedurinary bladder inflammation is a well-established experimentalmodel for bladder pain syndrome. Several studies have reportedthat CYP-induced cystitis results in a local inflammation (Ahluwaliaet al., 1994; Cox and Abel, 1979), associated with an increase invoiding frequency and a decrease in urine volume per void (Huet al., 2003; Maggi et al., 1992; Velasco et al., 2001). Some of thesereports have also shown that CYP induced an increase in painful

ll rights reserved.

de Physiopathologieance. Tel.: þ33 562 74 61 41;

behaviors characterized by eye closure, abnormal posture and adecrease in respiration rate (Boucher et al., 2000; Joshi et al., 2008;Lanteri-Minet et al., 1995; Saitoh et al., 2010). These behaviors peakshortly after CYP injection and persist up to 4 h (Boucher et al.,2000; Lanteri-Minet et al., 1995) and can be blocked by analgesicslike morphine (Boucher et al., 2000; Joshi et al., 2008). However,none of the observed behaviors can be considered as a directobservation of changes in nociception and pain perception. TheCYP-induced cystitis model is widely used in rats and mice tocharacterize new inflammatory and nociceptive pathways, andidentify new mediators (Arms et al., 2010; Malley and Vizzard,2002; Smaldone et al., 2009). However, a clear picture of thetherapeutic profile of anti-inflammatory and analgesic drugs onboth inflammatory and nociceptive parameters lacks in this model.

According to these different reports, the aim of the presentstudy was (1) to characterize different parameters and biomarkers

C. Auge et al. / European Journal of Pharmacology 707 (2013) 32–40 33

assessing urinary bladder inflammation and visceral pain in amodel of acute cystitis induced by CYP and (2) to assess therelevance of using this experimental model to test and developefficient therapeutic approaches for the treatment or relief of thesesymptoms. First, we wanted to develop a new approach toevaluate sensitivity of the lower abdomen in response to innoc-uous and noxious mechanical stimuli (von Frey filaments) infemale rats, this method having been previously used only in mice(Boudes et al., 2011), or in rat hind paws (Guerios et al., 2009;Merriam et al., 2011). Next, we tried to identify and quantifydifferent key inflammatory mediators in urinary bladder andurine, which may contribute to the development of inflammationand pain. Finally, we tested the effects of two non-steroidal anti-inflammatory drugs (NSAIDs), aspirin and ibuprofen, and anagonist of opioid receptors, morphine, against these inflammatoryand nociceptive parameters. The present study raises evidences forthe relevance of the use of the CYP-induced cystitis model in ratsto test and develop efficient therapeutic approaches for thetreatment of urinary bladder inflammation and pain.

2. Materials and methods

2.1. Animals

Female Sprague Dawley rats (Janvier, le Genest Saint Isle,France) weighing 225–250 g were used in this study. Rats werehoused 3–4 per cage, under constant humidity (63%) and tem-perature (21 1C), with free access to food and water and weremaintained on a 12 h/12 h light–dark cycle. All animals were alloweda minimum of 5 days period of acclimation to the animal facilitybefore the start of any experiments. At the end of procedures, ratswere humanely euthanized using CO2 inhalation (100%, 3 l/min),followed by cervical dislocation. All procedures were approved bythe Regional Institutional Animal Care Committee and veterinaryservices.

2.2. Induction of experimental cystitis and treatments

Acute cystitis was induced by a single intraperitoneal (i.p.)injection of CYP at a final dose of 150 mg/kg. Control animalsreceived physiological saline at a volume of 5 ml/kg i.p. Aspirin,ibuprofen (both 300 mg/kg) or vehicle (5% Na2CO3 in distilledwater—5 ml/kg) were administered orally (p.o.) 5 min before CYPadministration. Morphine (3 mg/kg) or its vehicle (physiologicalsaline—5 ml/kg) were injected subcutaneously (s.c.) 5 min beforeCYP administration.

2.3. Nociceptive response to mechanical stimulation

Sixty rats (10 per treatment group) were used randomly in thispart of the study. Drug treatments were hidden to the observer ofnociception testing. Each rat was placed individually in a clearplastic box divided into 6 small chambers with a metal grid floor.Rats were allowed to acclimatize to the environment for amaximum of 30 min and during 2 days before starting experi-ments. Mechanical stimulation was performed using von Freymonofilaments (Bioseb, Vitrolles, France) with increasing bendingforces of 1, 2, 4, 6, 8, 10, 26 and 60 g. The filaments were applied tothe lower abdominal area close to the urinary bladder. Eachfilament was applied 3 times for 1–2 s once every 5 s in ascendingorder of forces. Care was taken not to stimulate the same point twicein succession to avoid learning or sensitization. A score was assignedbased on the animal response: 0¼no response, 1¼reaction of theanimal (e.g. retractation of the abdomen), 2¼reaction of the animaland change of position, 3¼reaction of the animal, change of position,

licking of the area stimulated and/or vocalization (Asfaha et al.,2007). Nociceptive threshold and nociceptive scores were thenevaluated at the basal point (15 min before any treatment) and atdifferent time points after induction of cystitis, ranging from 1 to 4 h.Nociceptive threshold was defined as the von Frey filament for whichwe get a first score of 1. Mechanical nociceptive score was expressedas the percentage of the maximal score for the 3 applications.Corresponding areas under the curves (AUC) were calculated byplotting individual nociceptive score against von Frey forces.

2.4. Assessment of inflammation

Sixty rats (10 per treatment group) were used randomly in thispart of the study. During 2 days before starting these experiments,rats were allowed to acclimatize to individual metabolic cagesduring 3 h, without access to food and water. On the day ofexperiments, urine specimens were collected from rats kept for4 h after induction of cystitis in their individual metabolic cages.Animals were sacrificed after urine collection. Urinary bladderswere rapidly dissected and lipoid tissue cleaned from around thewall. Inflammatory parameters such as bladder weight, bladderwall thickness measured with a caliper, edema and hemorrhagescores were assessed. Edema and bleeding were evaluated accord-ing to the criteria of Gray (Gray et al., 1986). Urine and half of theurinary bladder were immediately stored at −801C for quantifica-tion of inflammatory mediators or metabolites. The other half ofthe urinary bladder was placed in formaldehyde for histologicalanalysis.

2.5. Inflammatory mediator analysis in urinary bladder

Cytokines or inflammatory mediators (ICAM-1, IFNγ, IL-1β, IL-4,IL-5, IL-6, IL-10, IL-13, IL-17A, IP10, KC/Groα, MCP-1, MIP-1α, NGF,RANTES, TNFα, VCAM-1, VEGF) were determined by multiplexanalysis (xMAP). Individual bladders were homogenized in 500 mlof cell lysis buffer (Panomics Ozyme, Saint-Quentin Yvelines,France) supplemented with a cocktail of antiproteinases (Sigma–Aldrich, Saint-Quentin Fallavier, France) using a Precellys 24 tissuehomogenizer (Ozyme, Saint-Quentin Yvelines, France). Homoge-nates were centrifuged at 10,000 rpm for 5 min (D’Aldebert et al.,2011). The supernatants from each sample were analyzed on theLuminex 100IS using a Procarta cytokines assay kit 18-plex(Panomics Ozyme, Saint-Quentin Yvelines, France). The cytokineconcentrations provided by Luminex were normalized to theprotein concentration in bladder tissues and expressed in pico-grams per milligram of protein (pg/mg of proteins) or in percen-tage of increase compared to i.p. saline group (100%, control). Totalprotein content was determined using the BCA (bicinchoninicacid) protein assay kit (Pierce Thermo Scientific, Courtaboeuf,France).

2.6. Quantification of PGE2 in rat urine

Frozen urine samples were thawed at room temperature beforebeing analyzed using a specific enzyme immunoassay (EIA) forprostaglandin E2 (PGE2) according to the protocol provided by themanufacturer (Oxford Biomedical Research, Euromedex, Souffel-weyersheim, France). For this, samples were added to the wells ofa microplate pre-coated with an anti-PGE2. This test kit operateson the basis of competition between the enzyme conjugate andthe PGE2 in the sample for a limited number of binding sites on theantibody coated plate. After incubations and extensive washings,plates were read at 450 nm. By using the standard curve, theconcentration of each sample was determined in comparison tothe corresponding concentration of PGE2 standard. The PGE2concentrations provided by EIA were normalized to the creatinine

C. Auge et al. / European Journal of Pharmacology 707 (2013) 32–4034

concentration in urines and expressed in nanograms per milligramof creatinine (ng/mg of creatinine). Urine creatinine was deter-mined by using the QuantiChromTM Creatinine Assay Kit (R&DSystems, Lille, France)

2.7. Metabolomic analysis of urine samples

Each frozen urine sample was thawed at room temperature andvortexed. After centrifugation (10,000 rpm, 4 1C, 5 min), 100 μl ofsupernatant was collected and diluted 1/3 in deionized water (USFElga Maxima II). A volume of 10 ml of sample was injected byan Accela chromatographic device (Thermo Fisher Scientific, LesUlis, France). The chromatographic separation was performed on aHypersil Gold C18 1.9 μm, 2.1�150 mm column. The detectionwas achieved using a LTQ-Orbitrap mass spectrometer fitted withan electrospray source and operated in positive and negative ionmodes, as previously described (Roux et al., 2012). Raw data were

***

0

500

1000

1500

2000

2500

AUC

**

0

500

1000

1500

2000

2500

AUC

Noc

icep

tive

scor

e (%

)

0.5 1 2 4 8 16 32 640

10

20

30

40

50

60

Von frey force (g)

T = +1h

0.5 1 2 4 8 16 32 640

10

20

30

40

50

60

Noc

icep

tive

scor

e (%

)

Von frey force (g)

T = +3h

0

5

10

15

basal 1h 2h

*** ***## ##

Noc

icep

tivet

hres

hold

(g)

Fig. 1. Effect of intraperitoneal injection of cyclophosphamide (CYP) and saline on noci(AUC) (B) in response to mechanical stimuli at different time points post-administratplotting individual nociceptive score against von Frey forces. ## Po0.01 different fromSaline group.

processed using the Xcalibur 2.0.7 (Thermo Fischer Scientific, LesUlis, France) and the XCMS softwares (Smith et al., 2006), aspreviously described (Roux et al., 2012). The processed data weremean-centered and scaled to Pareto variance by using SIMCA-P11(Umetrics, Umea, Sweden). A PLS-DA was performed. Amongdiscriminating variables, pantothenic acid and indoxyl sulfatevariables were highlighted. The variable of histamine was targeted,according to the literature (el-Mansoury et al., 1994; van Ophovenand Hertle, 2007). A processing method with the Xcalibur 2.1 soft-ware was developed to integrate these three molecular peaks.Peaks detection was achieved by Genesis algorithm, a smoothingpoint of 1 and a S/N ratio of 5 were used. The m/z–RT (in Th andmin) pairs were 220.11724–4.65 ([MþH]þ), 212.0011–5.65/5.71(double peak, ([M–H]–) and 112.0861–0.78 ([MþH]þ), respec-tively. These pairs matched with those of the standard compounds(Sigma–Aldrich, Saint-Quentin Fallavier, France) prepared at 5 μg/mlin deionized water and analyzed as described above.

**

0

500

1000

1500

2000

2500

AUC

***

0

500

1000

1500

2000

2500

AUC

0.5 1 2 4 8 16 32 640

10

20

30

40

50

60

Noc

icep

tive

scor

e (%

)

Von frey force (g)

T = +2h

0.5 1 2 4 8 16 32 640

10

20

30

40

50

60

Noc

icep

tive

scor

e (%

)

Von frey force (g)

T = +4h

3h 4h

****** ##

Vehicle p.o./CYPVehicle p.o./Saline

##

ceptive threshold (A), nociceptive scores and corresponding areas under the curveion. Data represent mean7S.E.M. (n¼10 in each group). AUC were calculated by

Vehicle p.o./CYP basal value. **Po0.01, ***Po0.001 different from Vehicle p.o./

C. Auge et al. / European Journal of Pharmacology 707 (2013) 32–40 35

2.8. Histological evaluation

As previously described (Motta et al., 2011), bladders werefixed in 4% buffered formaldehyde and embedded in paraffin. Thinsections (5 mm) of the bladder were cut and stained with hema-toxylin and eosin (HE). The severity of inflammation was evaluatedusing an optical microscope Nikon E400 in each section accordingto 4 criteria: mucosal abrasion, hemorrhage, leukocyte infiltrationand edema (Juszczak, 2010).

2.9. Drugs

Cyclophosphamide monohydrate (CYP) was purchased fromFisher Scientific (Illkirch Cedex, France). Physiologic saline (NaCl0.9%) was purchased from Laboratoire Aguettant via Centravet(Lapalisse, France). Aspirin, ibuprofen sodium salt and sodiumcarbonate (Na2CO3) were purchased from Sigma–Aldrich (Saint-Quentin Fallavier, France). Morphine chlorydrate was provided byAguettant Laboratories (Lyon, France).

2.10. Statistical analysis

Data were presented as mean7standard error of the mean(S.E.M.). Analyses were done by running the GraphPad Prism

ns

Vehicle p.o./Saline

Vehicle p.o./CYP

0

5

10

15

***

Noc

icep

tive

thre

shol

d (g

)

**

$$ $

0

500

1000

1500

2000

2500

AU

C

ns

***

$$$

basal 1h

basal 1h

$$

Fig. 2. Effects of aspirin, ibuprofen or vehicle treatments on nociceptive threshold (Amean7S.E.M. (n¼10 in each group). AUC were calculated by plotting individual nociceptSaline group. $Po0.05, $$P o0.01, $$$Po0.001 different from Vehicle p.o./CYP group.

4.0 software (GraphPad, San Diego, CA, USA). For mechanicalnociceptive threshold and AUC, a one-way ANOVA with repeatedmeasures followed by Dunnett’s post test was performed tocompare each time point versus basal value for each treatmentgroup. Then, a non parametric Mann–Whitney t test was used tocompare CYP treatment versus saline, and then drugs versusvehicle, for each time point. For inflammatory parameters andmediators (multiplex analysis and EIA), a one-way ANOVA fol-lowed by Dunnett’s post test or a non parametric Mann–Whitneyt test were performed to compare CYP treatment versus saline, anddrugs versus vehicle. For metabolomic analysis, all data wereconverted to log values in order to perform multiple comparisonswithin groups by repeated-measures one-way ANOVA, followedby Tukey’s post test. Statistical significance was considered forPo0.05.

3. Results

3.1. CYP treatment induces visceral hypersensitivity

Nociceptive response to von Frey stimulation was measured atdifferent time points after i.p. administration of CYP or saline.A single i.p. injection of CYP, at the dose of 150 mg/kg, induced

Aspirin p.o./CYP

Ibuprofen p.o./CYP

2h

* *** ***

$$

$$

$

$

***

** **

$$$

2h

3h 4h

3h 4h

$$$$$

$$$$

) and nociceptive scores (AUC) (B) 4 h after CYP administration. Data representive score against von Frey forces. **Po0.01, ***Po0.001 different from Vehicle p.o./

0

5

10

15

20

25

30

35

40

45

0

500

1000

1500

2000

2500

Noc

icep

tive

thre

shol

d (g

)

2h

2h

basal 1h 3h 4h

basal 1h 3h 4h

AU

C

Vehicle s.c./CYP Morphine s.c./CYP

+ + +

+ + ++ + +

+ + ++ + ++ + +

Fig. 3. Effects of morphine or vehicle treatment on nociceptive threshold (A) andnociceptive scores (AUC) (B) 4 h after CYP administration. Data represent mean7S.E.M. (n¼10 in each group). AUC were calculated by plotting individual nociceptivescore against von Frey forces. þþþPo0.001 different from Vehicle s.c./CYP group.

C. Auge et al. / European Journal of Pharmacology 707 (2013) 32–4036

visceral hypersensitivity characterized by both mechanical allody-nia (a decrease in mechanical threshold in response to innocuousvon Frey forces) (Fig. 1A), and mechanical hyperalgesia (increasednociceptive scores in response to noxious von Frey forces) (Fig. 1B).These nociceptive responses were correlated with a significantincrease in the area under the curve (AUC) (Fig. 1B). The pro-nociceptive effect of CYP treatment was observed starting from 1 hafter its administration and was maintained for up to 4 h (Fig. 1Aand B), compared to saline-treated rats.

3.2. Pharmacological blockade of CYP-induced referred visceral pain

Because the i.p. injection of CYP caused a decrease in nocicep-tive threshold and an increase in nociceptive scores in response toa mechanical stimulus, we wanted to investigate if a treatmentwith NSAIDs or an opioid agonist was able to reduce the bladderhypersensitivity induced by CYP. A single oral administration ofaspirin or ibuprofen, at the dose of 300 mg/kg, significantlyreduced the increase in nociceptive responses caused by CYP. Thisreversal was observed between 1 and 4 h after both treatment andwas associated with an increased mechanical threshold (Fig. 2A)and a decreased nociceptive score AUC (Fig. 2B).

Similarly, inhibition of nociceptive responses was observed whenrats received a single s.c. injection of morphine at the dose of 3 mg/kg.Morphine was able to significantly reverse the CYP-induced hyper-sensitivity, as demonstrated by an increased mechanical threshold inmorphine-treated rats (Fig. 3A) and a decreased nociceptive score(Fig. 3B) up to 3 h post-administration, compared to vehicle (s.c.saline)-treated rats. The anti-nociceptive effects of morphine werelost 4 h after its administration (Fig. 3A and B).

3.3. Effect of NSAIDs and morphine on CYP-induced bladderinflammation

In order to correlate the effects of CYP administration on bladderhypersensitivity with bladder inflammation, urinary bladders fromthe different groups were harvested and scored for different para-meters. As shown in Table 1, urinary bladder weight was significantlyincreased 4 h after CYP administration, compared to saline-treatedrats. NSAIDs or morphine treatments did not affect the increase onurinary bladder weights. Urinary bladder wall thickness was alsosignificantly increased after CYP treatment compared to control(Table 1). Both aspirin and ibuprofen, at the dose of 300 mg/kg,significantly reduced the wall thickness compared to vehicle-treatedrats. In contrast, s.c. injection of morphine induced a significantincrease in bladder wall thickness compared to vehicle (s.c. saline)-treated rats. Finally, edema and hemorrhage scores were increased4 h after CYP administration (Table 1). This increase was not reversedby a pre-treatment with aspirin or ibuprofen. In contrast, morphineinduced a significant increase in edema and hemorrhage scorescompared to vehicle (s.c. saline)-treated rats.

3.4. Histological modifications induced by CYP administration andpharmacological evaluation

Histologic evaluation of the bladder with HE staining 4 h afterCYP treatment showed clear signs of inflammation, such as severemucosal abrasion and edema (Fig. 4B). Compared to controlbladders (Fig. 4A), CYP-treated rats presented an alteration ofbladder wall structure, with a loss of epithelium barrier and astrong submucosal edema (black arrows). In NSAIDs-treated rats(Fig. 4C and D), edema seemed to be less important compared tovehicle-treated animals. Compared to ibuprofen (Fig. 4D), bladderharvested from aspirin-treated rats presented a normal epitheliumand less edema (Fig. 4C). Morphine treatment (Fig. 4E) had noeffect on submucosal edema or on bladder wall abrasion compared

to its vehicle (s.c. saline) (Fig. 4F). All these results are consistentwith macroscopic scores presented in Table 1.

3.5. Pharmacological changes in tissue inflammatory mediators afterCYP-induced cystitis

The effects of a single administration of CYP on levels of cytokines,chemokines and adhesion molecules in bladder tissues were profiled4 h after its administration. Over the 18 different cytokines orinflammatory mediators we investigated (IL-1β, ICAM, IL-6, IL-5,IL-17, TNF-α, IL-4, IL-10, KC, IL-13, NGF, RANTES, IFN-γ, MIP-1α,MCP-1, VEGF, IP-10 and VCAM), only 5 were detected in significantamounts in bladder tissues: IL-1β, IL-6, IL-10, MCP-1 and VCAM(Fig. 5). CYP treatment induced a significant increase of IL-1β, MCP-1,VCAM and IL-6, compared to saline-treated group. A pre-treatmentwith ibuprofen significantly decreased CYP-induced increase of IL-1βand IL-6 levels. A pre-treatment with aspirin seemed to decreaseIL-1β and IL-6 levels, but this effect did not reach significance. Neitheraspirin nor ibuprofen pre-treatment had an effect on MCP-1 andVCAM levels. IL-10 expression was weak but detectable in thosetissues and was not modified by any of the treatments (Fig. 5).Compared to vehicle (s.c. saline)-treated rats, s.c. injection of mor-phine induced a significant increase in levels of IL-6 and MCP-1 inbladder tissues of cystitic rats (Table 2). No effect of morphinetreatment was observed in the tissue levels of IL-1β and VCAM.

3.6. Markers of cystitis in urines

Different metabolites and mediators were quantified in the urinesof rats to potentially identify markers for the cystitis condition in this

Table 1Effects of drug treatments on inflammatory parameters in urinary bladder.

Groups Urinary bladder weight (mg) Urinary bladder wall thickness (mm) Edema Hemorrhage

Vehicle p.o./Saline 77.5872.54 0.3570.03 0.0070.00 0.0070.00Vehicle p.o./CYP 105.1076.59aa 0.6470.03aaa 2.4070.27 2.0070.26Aspirin p.o./CYP 114.3079.07 0.4770.04bb 1.8070.36 1.1070.23Ibuprofen p.o./CYP 113.2075.71 0.4470.04bb 1.7070.45 1.2070.36Vehicle s.c./CYP 92.5076.48 0.7670.05 1.8070.33 1.1070.10Morphine s.c./CYP 91.1674.09 0.9570.05c 3.0070.00 2.3070.26cc

Data represent mean7S.E.M. (n¼10 in each group). aaPo0.01, aaaPo0.001 different from Vehicle p.o./Saline group. bbPo0.01 different from Vehicle p.o./CYP group.cPo0.05, ccPo0.01 different from Vehicle s.c./CYP group.

x100

epithelium

Mucosa

x100

epithelium

Mucosa

Mucosa

epithelium

x100

Mucosa

epithelium

x100

Mucosa

epithelium

x100

Mucosa

epithelium

x100

Fig. 4. Histological appearance of urinary bladders in saline (A) and in CYP-treated rats that were also treated with p.o. vehicle (B), p.o. aspirin (C), p.o. ibuprofen (D), s.c.saline (E) and s.c. morphine (F) (HE staining). Black arrows show edema.

C. Auge et al. / European Journal of Pharmacology 707 (2013) 32–40 37

model. First, PGE2 was detected by EIA in urines and was foundsignificantly increased in cystitic rats (CYP-treated) (Table 3). Bothaspirin and ibuprofen, but not morphine, treatments significantlydecreased PGE2 levels in urines (Table 3). Histamine content was alsodetected in urines but levels in CYP-treated rats were not alteredcompared to controls (Table 3). Drug treatments did not affect levels ofurinary histamine in rats with cystitis. Indoxyl sulfate was significantlyreduced in the urine of CYP-treated rats (Table 3). After aspirin oribuprofen treatments, this significant decrease was not modified. Incontrast, morphine treatment reversed this CYP-induced decrease of

indoxyl sulfate compared to vehicle (s.c. saline) (Table 3). Finally, thelevel of pantothenic acid was significantly down-regulated in theurines of rats treated with CYP. Aspirin, ibuprofen and morphinetreatments were able to reverse the increase of this biomarkerinduced by CYP, compared to their corresponding vehicles (Table 3).

4. Discussion

The present study provided a clear picture of the parametersand mediators that can be assessed in the experimental model of

Bla

dder

tiss

ue c

ytok

ine

leve

ls(in

per

cent

age

of in

crea

se c

ompa

red

to sa

line)

MCP-1

IL-1ββ

Saline CYP

0

150

300

450

600

750

900***

$

IL-6

Saline CYP

0

150

300

450

600

750

900***

$$

IL-10

Saline CYP

0

25

50

75

100

125

Saline CYP0

100

200

300

400 ***VCAM

Saline CYP0

50

100

150

200

250 **Vehicle p.o./Saline

Vehicle p.o./CYP

Aspirin p.o./CYP

Ibuprofen p.o./CYP

Fig. 5. Changes in IL-1β, IL-6, IL-10, MCP-1 and VCAM, levels in bladder tissues following aspirin, ibuprofen and vehicle treatment in CYP-induced cystitis. Data representmean7S.E.M. (n¼6–10 in each group). Values are expressed as percentage of increase compared to Vehicle p.o./Saline group (100%, control). **Po0.01, ***Po0.001different compared to Vehicle p.o./Saline group. $Po0.05, $$Po0.01 different from Vehicle p.o./CYP group.

Table 2Changes in IL-1β, IL-6, IL-10, MCP-1 and VCAM, levels in bladder tissues following s.c. saline or morphine treatment in CYP-induced cystitis (4 h).

Groups IL-1β IL-6 IL-10 MCP-1 VCAM

Vehicle s.c./CYP 15.8178.22 60.54736.00 0.4070.16 241.607138.80 35.40715.68Morphine s.c./CYP 18.8476.72 113.60724.90ccc 0.5670.55 601.60762.06ccc 39.36716.19

Data represent mean7S.E.M. (n¼9–10 in each group). Mediator levels are expressed in pg/mg of proteins, cccPo0.001 different from Vehicle s.c./CYP group.

Table 3Changes in PGE2, indoxyl sulfate, histamine, and pantothenic acid content in urinesof rats 4 h after saline or CYP treatment, and in animals that were dosed withvehicle, aspirin, ibuprofen or morphine.

Urine Content

Groups PGE2 (ng/mgof creatinin)

Indoxylsulfate (�108)

Histamine(�106)

Panthotenicacid (�107)

Area (arbitrary unit)

Vehicle/Saline 2.270.5 4.370.6 5.670.9 10.870.9Vehicle/CYP 40.8715.2a 2.070.5aa 3.871.0 4.771.1aa

Aspirin/CYP 4.970.9b 2.170.4 4.770.7 13.371.8bb

Ibuprofen/CYP 5.4 70.6b 2.270.4 5.570.5 9.871.2b

Saline/CYP 42.379.2 1.170.1 4.871.3 3.070.4Morphine/CYP 27.275.2 2.770.5cc 6.970.9 6.171.0c

Data represent mean7S.E.M. (n¼3–10 in each group). PGE2 values are expressedin ng per mg of creatinine. Metabolite levels represent arbitrary value (area).aPo0.05, aaPo0.01 different from Vehicle/Saline group. bPo0.05, bbPo0.01different from Vehicle/CYP group. cPo0.05, ccPo0.01 different from Saline/CYPgroup.

C. Auge et al. / European Journal of Pharmacology 707 (2013) 32–4038

CYP-induced acute cystitis in female rats, measuring both inflam-mation and visceral pain. We demonstrated the relevance of usingthis model to evaluate and characterize new mediators and path-ways involved in these symptoms, and the pharmacologicalinterest to test new therapeutic approaches for the treatments ofbladder inflammation and pain with this model. By establishingthe profile of nociceptive responses to mechanical stimulation

using von Frey filaments, we reported, for the first time, a new andreliable method to evaluate nociceptive disturbances associatedwith urinary bladder inflammation in rats. Reproducible andquantifiable changes in referred nociceptive threshold (allodynia)and nociceptive intensity (hyperalgesia) were observed andreversed by anti-inflammatory and analgesic drugs. Therefore,the results presented here further characterized the acute CYPmodel as a valid in vivo model of bladder pain syndrome. This isparticularly important as it is nowadays recognized that bladderpain or discomfort is the most important criterion for a differentialdiagnosis of bladder pain syndrome (van de Merwe et al., 2008).We further identified cytokines, chemokines, adhesion molecules,lipid mediators and urine metabolites as mediators differentiallyexpressed in this bladder pain syndrome model. Taken together,the results presented here validate the use of the acute CYP modelin female rats as an in vivomodel of bladder pain syndrome, wherethe effects of anti-inflammatory and analgesic treatments can betested on parameters of both inflammation and visceral pain.

Overall, NSAIDs treatment relieved from both inflammationand visceral hypersensitivity in this acute model of cystitis. Weshowed here that both aspirin and ibuprofen were able to inhibitreferred visceral allodynia and hyperalgesia induced by CYP.Regarding urinary bladder inflammation, it is interesting to notethat not all the inflammatory parameters were modulated byNSAIDs. Ibuprofen treatment was able to reduce the increasedbladder wall thickness, the tissue levels of the pro-inflammatorycytokines IL-1β and IL-6, and also PGE2 level in the urine. Incontrast, the increase of bladder weight, macroscopic damage, or

C. Auge et al. / European Journal of Pharmacology 707 (2013) 32–40 39

tissue content for the chemokine MCP-1 and adhesion moleculeVCAM were not affected by this drug. Similarly, aspirin treatmentwas able to reduce the increased bladder wall thickness and PGE2levels, without affecting the overall weight, edema and hemor-rhage scores of the bladder. However, compared to ibuprofen,aspirin treatment failed to reduce the tissue levels of IL-1β andIL-6, but seemed to be more protective at the microscopic level,with a very good conservation of the epithelial structure and closeto the total absence of microscopic edema. These differencesbetween aspirin and ibuprofen can potentially be explained bytheir different modes of action. While the fixation of ibuprofen onthe cyclooxygenases (COX) is reversible and thus might not lastvery long, aspirin irreversibly acetylates COX enzymes (Loll et al.,1995). These differences can affect platelet physiology, whichmight thus play a role in the present model. Another differencebetween the mode of action of aspirin and ibuprofen is thespecificity for the different COX isoforms (Smith et al., 2000).While aspirin is equipotent at inhibiting COX-1 and COX-2,ibuprofen has demonstrated a fivefold greater inhibition of COX-2(Mitchell et al., 1994; Vane and Botting, 1996). Therefore, differencesin the inflammatory parameters modulated by aspirin versusibuprofen could be mediated by COX-1 inhibition. Finally, aspirinis known to act on the inflammatory response by facilitating therelease of lipoxins and resolvins, which is not driven by ibuprofentreatment. Considering the demonstrated protective effects oflipoxins and resolvins on mucosal architecture (Campbell et al.,2007; Fierro et al., 2003; Fiorucci et al., 2003), those metabolitesreleased upon aspirin treatment might also play a role in thepreservation of epithelial structures observed in cystitic rats treatedwith aspirin.

Although morphine and opioids in general have been reportedin a number of tissues for exerting anti-inflammatory properties(Smith, 2008), we observed here that morphine treatment did notdecrease any of the inflammatory parameters. In contrast, mor-phine increased bladder wall thickness and hemorrhage scores.However, concomitantly, morphine was able to completely inhibitCYP-induced hyperalgesia for the first 3 h after CYP administra-tion, and even further increased the nociceptive threshold inresponse to von Frey filaments during that same period. We knowthat opioids can alter the control of bladder activity, in generalassociated with urinary retention in humans (Malinovsky et al.,1998) and rats (Kontani and Kawabata, 1988). These effects onbladder function can then alter local inflammatory processes,associated with increased levels of some inflammatory mediatorssuch as IL-6 and MCP-1, and so increased bladder wall thicknessand hemorrhage scores. The opposing effects of morphine onnociception and inflammation may then be attributed to differentsites of action of opioids on peripheral versus central pathways.

Although allodynia and hyperalgesia are clearly linked to andoriginate from the inflammatory response, the results raised hereshowed that the nociceptive response can be separated from theinflammatory parameters. Our results suggest that both inflam-mation and nociception have to be considered in models of cystitisin order to define the best therapeutic treatment.

Among the medical challenges of bladder pain syndrome arethe diagnosis and the follow-up of the disease. It was reported thaturinary IL-6 was increased in bladder pain syndrome patients byseveral groups (Erickson et al., 2002; Lamale et al., 2006; Peterset al., 1999). Here, we have also shown the increased presence ofIL-6 in bladder tissues from CYP-treated rats. In addition, increasedlevels of IL-1β and MCP-1 were observed in those tissues. Thesechanges of IL-1β and IL-6 levels are consistent with previouspublished studies, where an increase in those proteins expressionwas observed 4 h after cystitis in rats (Girard et al., 2011; Malleyand Vizzard, 2002). In human, epithelial IL-1β staining was absentin normal subjects but positive in cells from bladder pain

syndrome patients (Hang et al., 1998). We report here increasedlevels of MCP-1 and VCAM in this rat model of acute cystitisinduced by CYP, in accord with the observation by Lv et al. (2010)of elevated MCP-1 expression in both urine and bladder tissues inpatients with bladder pain syndrome. Regarding VCAM, there areno preclinical or clinical reports describing its alteration in bladderpain syndrome. Only ICAM expression (another adhesion mole-cule) was detected in bladder samples of bladder pain syndromepatients (Green et al., 2004). Our results add both MCP-1 andVCAM to the list of potential biomarkers of bladder pain syn-drome. Our observations combined to the knowledge reported inthe literature on tissues from patients suggest that the presentrodent model of bladder pain syndrome is relatively close to thehuman pathology, and that a number of biomarkers are commonto the human disease and to this model.

In humans, it has been suggested that histamine is involved inthe pathogenicity of bladder pain syndrome, where its presencewas detected in urines (el-Mansoury et al., 1994). Here, wereported that urinary histamine levels were unchanged whetherthe animals were submitted to CYP-induced cystitis or not. Sinceour measurements were performed only few hours after theinduction of cystitis, the number of mast cells is most likelyunchanged at this early time of the disease development, poten-tially explaining the lack of difference in levels of the mast cellmarker histamine. In contrast, we showed that indoxyl sulfate, ametabolite of tryptophan (Niwa and Ise, 1994), was significantlyreduced in CYP-treated rats. Interestingly, the amounts of thismolecule were not as much decreased when animals receivedNSAIDs treatments, while an increase was observed after mor-phine treatment. In 2009, Rubio-Diaz et al. (2009) have shownthat serum from cats with feline interstitial cystitis had higherconcentration of tryptophan than did serum from healthy cats.In another study, Hsieh et al. (2007) also reported the presence oftryptophan in freeze-dried water samples voided from bladderpain syndrome patients. In our model, levels of indoxyl sulfate canreflect abnormalities in the metabolism or the intestinal absorp-tion of tryptophan following cystitis induced by CYP. Up to date,abnormal urinary excretion of metabolites of the amino acidtryptophan has never been reported in patients with bladder painsyndrome, except for tryptophan itself. Further studies investigat-ing the presence of indoxyl sulfate in the urines of bladder painsyndrome patients and healthy controls would shed some impor-tant and definitive light on the possible use of this metabolite as abiomarker in this disease. Lastly, we reported that levels of urinarypantothenic acid were also significantly decreased after CYPtreatment. This effect was completely abolished in rats treatedwith aspirin, and only partially abolished by ibuprofen or mor-phine treatments. Pantothenic acid is a water-soluble vitamin, alsoknown as vitamin B5. This finding is consistent with the conceptthat vitamin B5 could participate to the control of mucosalinflammation. Indeed, previous reports demonstrated that vitaminB5 enhances epithelial barrier functions and reduces inflammationin the skin (Proksch and Nissen, 2002). Here again, further studiesidentifying vitamin B5 in human urines would be necessary toconfirm its potential use as a biomarker of bladder pain syndrome.

5. Conclusions

Taken together, all the data shown in the present studyprovided a clear picture of the parameters and mediators involvedin urinary bladder inflammation and pain that could be measuredin a model of cystitis induced by systemic CYP in rodents. Thepresent study highlighted relevant information about the acuteCYP model in rats through (1) a new and reliable method to evaluatevisceral sensitivity, (2) new relevant biomarkers identified in the

C. Auge et al. / European Journal of Pharmacology 707 (2013) 32–4040

urinary bladder (MCP-1 and VCAM), and (3) relevant key metabolites(indoxyl sulfate and pantothenic acid) dosed in urine. Moreover, wevalidated the use of this model and different parameters of diseaseactivity for in vivo pharmacological studies, aiming at testing differenttherapeutic approaches for the treatment of urinary bladder inflam-mation and pain.

Acknowledgments

We thank Aurore Desquesnes for Luminex Analysis (platform inLife Sciences Anexplo of Toulouse Genopole, France).

References

Ahluwalia, A., Maggi, C.A., Santicioli, P., Lecci, A., Giuliani, S., 1994. Characterizationof the capsaicin-sensitive component of cyclophosphamide-induced inflamma-tion in the rat urinary bladder. Br. J. Pharmacol. 111, 1017–1022.

Arms, L., Girard, B.M., Vizzard, M.A., 2010. Expression and function of CXCL12/CXCR4 in rat urinary bladder with cyclophosphamide-induced cystitis. Am. J.Physiol. Renal. Physiol. 298, F589–600.

Asfaha, S., Cenac, N., Houle, S., Altier, C., Papez, M.D., Nguyen, C., Steinhoff, M.,Chapman, K., Zamponi, G.W., Vergnolle, N., 2007. Protease-activated receptor-4:a novel mechanism of inflammatory pain modulation. Br. J. Pharmacol. 150,176–185.

Boucher, M., Meen, M., Codron, J., Coudore, F., Kemeny, J., Eschalier, A., 2000.Cyclophosphamide-induced cystitis in freely-moving conscious rats behavioralapproach to a new model of visceral pain. J. Urol. 164, 203–208.

Boudes, M., Uvin, P., Kerselaers, S., Vennekens, R., Voets, T., De Ridder, D., 2011.Functional characterization of a chronic cyclophosphamide-induced overactivebladder model in mice. Neurourol. Urodyn. 30, 1659–1665.

Campbell, E.L., Louis, N.A., Tomassetti, S.E., Canny, G.O., Arita, M., Serhan, C.N.,Colgan, S.P., 2007. Resolvin E1 promotes mucosal surface clearance of neutro-phils: a new paradigm for inflammatory resolution. Faseb J. 21, 3162–3170.

Cox, P.J., Abel, G., 1979. Cyclophosphamide cystitis. Studies aimed at its minimiza-tion. Biochem. Pharmacol. 28, 3499–3502.

D’Aldebert, E., Cenac, N., Rousset, P., Martin, L., Rolland, C., Chapman, K., Selves, J.,Alric, L., Vinel, J.P., Vergnolle, N., 2011. Transient receptor potential vanilloid4 activated inflammatory signals by intestinal epithelial cells and colitis inmice. Gastroenterology 140, 275–285.

el-Mansoury, M., Boucher, W., Sant, G.R., Theoharides, T.C., 1994. Increased urinehistamine and methylhistamine in interstitial cystitis. J. Urol. 152, 350–353.

Erickson, D.R., Xie, S.X., Bhavanandan, V.P., Wheeler, M.A., Hurst, R.E., Demers, L.M.,Kushner, L., Keay, S.K., 2002. A comparison of multiple urine markers forinterstitial cystitis. J. Urol. 167, 2461–2469.

Fierro, I.M., Colgan, S.P., Bernasconi, G., Petasis, N.A., Clish, C.B., Arita, M., Serhan, C.N., 2003. Lipoxin A4 and aspirin-triggered 15-epi-lipoxin A4 inhibit humanneutrophil migration: comparisons between synthetic 15 epimers in chemo-taxis and transmigration with microvessel endothelial cells and epithelial cells.J. Immunol. 170, 2688–2694.

Fiorucci, S., Distrutti, E., de Lima, O.M., Romano, M., Mencarelli, A., Barbanti, M.,Palazzini, E., Morelli, A., Wallace, J.L., 2003. Relative contribution of acetylatedcyclo-oxygenase (COX)-2 and 5-lipooxygenase (LOX) in regulating gastricmucosal integrity and adaptation to aspirin. Faseb J. 17, 1171–1173.

Girard, B.M., Cheppudira, B.P., Malley, S.E., Schutz, K.C., May, V., Vizzard, M.A., 2011.Increased expression of interleukin-6 family members and receptors in urinarybladder with cyclophosphamide-induced bladder inflammation in female rats.Front. Neurosci. 5, 20.

Gray, K.J., Engelmann, U.H., Johnson, E.H., Fishman, I.J., 1986. Evaluation ofmisoprostol cytoprotection of the bladder with cyclophosphamide (Cytoxan)therapy. J. Urol. 136, 497–500.

Green, M., Filippou, A., Sant, G., Theoharides, T.C., 2004. Expression of intercellularadhesion molecules in the bladder of patients with interstitial cystitis. Urology63, 688–693.

Guerios, S.D., Wang, Z.Y., Boldon, K., Bushman, W., Bjorling, D.E., 2009. Lidocaineprevents referred hyperalgesia associated with cystitis. Neurourol. Urodyn. 28,455–460.

Hang, L., Wullt, B., Shen, Z., Karpman, D., Svanborg, C., 1998. Cytokine repertoire ofepithelial cells lining the human urinary tract. J. Urol. 159, 2185–2192.

Hsieh, T.F., Yu, K.J., Lin, S.Y., 2007. Possible application of Raman microspectroscopyto verify the interstitial cystitis diagnosis after potassium sensitivity test:phenylalanine or tryptophan as a biomarker. Dis. Markers 23, 147–152.

Hu, V.Y., Malley, S., Dattilio, A., Folsom, J.B., Zvara, P., Vizzard, M.A., 2003. COX-2 andprostanoid expression in micturition pathways after cyclophosphamide-induced cystitis in the rat. Am. J. Physiol. Regul. Integr. Comp. Physiol. 284,R574–585.

Joshi, S.K., Mikusa, J.P., Weaver, B., Honore, P., 2008. Morphine and ABT-594 (anicotinic acetylcholine agonist) exert centrally mediated antinociception in therat cyclophosphamide cystitis model of visceral pain. J. Pain 9, 146–156.

Juszczak, K., 2010. Functional, histological structure and mastocytes alterations inrat urinary bladders following acute and chronic cyclophosphamide treatment.J. Physiol. Pharmacol. 61, 477–482.

Kontani, H., Kawabata, Y., 1988. A study of morphine-induced urinary retention inanesthetized rats capable of micturition. Jpn. J. Pharmacol. 48, 31–36.

Lamale, L.M., Lutgendorf, S.K., Zimmerman, M.B., Kreder, K.J., 2006. Interleukin-6,histamine, and methylhistamine as diagnostic markers for interstitial cystitis.Urology 68, 702–706.

Lanteri-Minet, M., Bon, K., de Pommery, J., Michiels, J.F., Menetrey, D, 1995.Cyclophosphamide cystitis as a model of visceral pain in rats: model elabora-tion and spinal structures involved as revealed by the expression of c-Fos andKrox-24 proteins. Exp. Brain. Res. 105, 220–232.

Loll, P.J., Picot, D., Garavito, R.M., 1995. The structural basis of aspirin activityinferred from the crystal structure of inactivated prostaglandin H2 synthase.Nat. Struct. Biol. 2, 637–643.

Lv, J.W., Luo, Y., Leng, J., Xue, W., Liu, D.M., Huang, Y.R., 2010. Aberrant expression ofmonocyte chemoattractant protein-1 (mcp-1) in interstitial cystitis patients.Sci. Res. Essays 5, 663–667.

Maggi, C.A., Lecci, A., Santicioli, P., Del Bianco, E., Giuliani, S., 1992. Cyclopho-sphamide cystitis in rats: involvement of capsaicin-sensitive primary afferents.J. Auton. Nerv. Syst. 38, 201–208.

Malinovsky, J.M., Le Normand, L., Lepage, J.Y., Malinge, M., Cozian, A., Pinaud, M.,Buzelin, J.M., 1998. The urodynamic effects of intravenous opioids and keto-profen in humans. Anesth. Analg. 87, 456–461.

Malley, S.E., Vizzard, M.A., 2002. Changes in urinary bladder cytokine mRNA andprotein after cyclophosphamide-induced cystitis. Physiol. Genomics 9, 5–13.

Merriam, F.V., Wang, Z.Y., Hillard, C.J., Stuhr, K.L., Bjorling, D.E., 2011. Inhibition offatty acid amide hydrolase suppresses referred hyperalgesia induced by bladderinflammation. BJU. Int. 108, 1145–1149.

Mitchell, J.A., Belvisi, M.G., Akarasereenont, P., Robbins, R.A., Kwon, O.J., Croxtall, J.,Barnes, P.J., Vane, J.R., 1994. Induction of cyclo-oxygenase-2 by cytokines inhuman pulmonary epithelial cells: regulation by dexamethasone. Br. J. Phar-macol. 113, 1008–1014.

Motta, J.-P., Magne, L., Descamps, D., Rolland, C., Squarzoni-Dale, C., Rousset, P.,Martin, L., Cenac, N., Balloy, V., Huerre, M., Fr\widehathlich, L.F., Jenne, D.,Wartelle, J., Belaaouaj, A., Mas, E., Vinel, J.-P., Alric, L., Chignard, M., Vergnolle,N., Sallenave, J.-M., 2011. Modifying the protease, antiprotease pattern by elafinoverexpression protects mice from colitis. Gastroenterology 140, 1272–1282.

Niwa, T., Ise, M., 1994. Indoxyl sulfate, a circulating uremic toxin, stimulates theprogression of glomerular sclerosis. J. Lab. Clin. Med. 124, 96–104.

Parsons, C.L., 2007. The role of the urinary epithelium in the pathogenesis ofinterstitial cystitis/prostatitis/urethritis. Urology 69, 9–16.

Peters, K.M., Diokno, A.C., Steinert, B.W., 1999. Preliminary study on urinarycytokine levels in interstitial cystitis: does intravesical bacille Calmette-Guerin treat interstitial cystitis by altering the immune profile in the bladder?Urology 54, 450–453.

Proksch, E., Nissen, H.P., 2002. Dexpanthenol enhances skin barrier repair andreduces inflammation after sodium lauryl sulphate-induced irritation. J. Der-matolog. Treat. 13, 173–178.

Roux, A., Xu, Y., Heilier, J.F., Olivier, M.F., Ezan, E., Tabet, J.C., Junot, C., 2012.Annotation of the human adult urinary metabolome and metabolite identifica-tion using ultra high performance liquid chromatography coupled to a linearquadrupole ion trap-Orbitrap mass spectrometer. Anal. Chem. 84, 6429–6437.

Rubio-Diaz, D.E., Pozza, M.E., Dimitrakov, J., Gilleran, J.P., Giusti, M.M., Stella, J.L.,Rodriguez-Saona, L.E., Buffington, C.A., 2009. A candidate serum biomarker forbladder pain syndrome/interstitial cystitis. Analyst 134, 1133–1137.

Saitoh, C., Yokoyama, H., Chancellor, M.B., de Groat, W.C., Yoshimura, N., 2010.Comparison of voiding function and nociceptive behavior in two rat models ofcystitis induced by cyclophosphamide or acetone. Neurourol. Urodyn. 29,501–505.

Smaldone, M.C., Vodovotz, Y., Tyagi, V., Barclay, D., Philips, B.J., Yoshimura, N.,Chancellor, M.B., Tyagi, P., 2009. Multiplex analysis of urinary cytokine levels inrat model of cyclophosphamide-induced cystitis. Urology 73, 421–426.

Smith, C.A., Want, E.J., O’Maille, G., Abagyan, R., Siuzdak, G., 2006. XCMS: processingmass spectrometry data for metabolite profiling using nonlinear peak align-ment, matching, and identification. Anal. Chem. 78, 779–787.

Smith, H.S., 2008. Peripherally-acting opioids. Pain Physician 11, S121–132.Smith, W.L., DeWitt, D.L., Garavito, R.M., 2000. Cyclooxygenases: structural, cellular,

and molecular biology. Annu. Rev. Biochem. 69, 145–182.van de Merwe, J.P., Nordling, J., Bouchelouche, P., Bouchelouche, K., Cervigni, M.,

Daha, L.K., Elneil, S., Fall, M., Hohlbrugger, G., Irwin, P., Mortensen, S., vanOphoven, A., Osborne, J.L., Peeker, R., Richter, B., Riedl, C., Sairanen, J., Tinzl, M.,Wyndaele, J.J., 2008. Diagnostic criteria, classification, and nomenclature forpainful bladder syndrome/interstitial cystitis: an ESSIC proposal. Eur. Urol. 53,60–67.

van Ophoven, A., Hertle, L., 2007. The dual serotonin and noradrenaline reuptakeinhibitor duloxetine for the treatment of interstitial cystitis: results of anobservational study. J. Urol. 177, 552–555.

Vane, J.R., Botting, R.M., 1996. Mechanism of action of anti-inflammatory drugs.Scand. J. Rheumatol. Suppl. 102, 9–21.

Velasco, C., Angelico, P., Guarneri, L., Leonardi, A., Clarke, D.E., Testa, R., 2001. Effectsof the nuclear factor-kappaB inhibitors 2-hydroxy-4-trifluoromethylbenzoicacid and aspirin on micturition in rats with normal and inflamed bladder. J.Urol. 166, 1962–1968.