ORIGINAL PAPER

Tempol protects the gallbladder againstischemia/reperfusion

Pedro J. Gomez-Pinilla & Pedro J. Camello &

Jesus A. F. Tresguerres & María José Pozo

Received: 8 April 2010 /Accepted: 4 May 2010 /Published online: 23 June 2010# University of Navarra 2010

Abstract Impairment in gallbladder emptying, in-crease in residual volume, and reduced smoothmuscle contractility are hallmarks of acute acalculouscholecystitis and seem to be related to ischemia/reperfusion (I/R). This study was designed to deter-mine the effects of tempol, a general antioxidant, onI/R-induced changes in gallbladder contractile capac-ity, the mechanisms involved in the contractileprocess, and the level of inflammatory mediators.Experimental gallbladder I/R was induced in maleguinea pigs by common bile duct ligation for 2 days,then a deligation of the duct was performed and after2 days the animals were sacrificed. A group of animalswas treated with tempol, administered in the drinkingwater at 1 mmol/l for 10 days prior the bile ductligation and until animal sacrifice. Isometric tension

recordings showed that KCl and cholecystokinin-induced contractions were impaired by I/R, whichcorrelated with decreased F-actin content and detri-mental effects on Ca2+ influx. In addition, I/Rdepolarized mitochondrial membrane potential, asindicated by the reduction of the heterogeneityof the rhodamine123 fluorescence signal, and in-creased the expression of NF-κB, COX-2, and iNOS.Tempol treatment improved contractility via normal-ization of Ca2+ handling and improvement of F-actincontent. Moreover, the antioxidant ameliorated mito-chondrial polarity and normalized the expressionlevels of the inflammatory mediators. These resultsshow that antioxidant treatment protects the gallblad-der from I/R, indicating the potential therapeuticbenefits of tempol in I/R injury.

Keywords Tempol . Gallbladder .

Ischemia/reperfusion . Calcium handling .

Mitochondrial function

Introduction

Acute acalculous cholecystitis (AC) is a pathophysio-logical condition characterized by gallbladder inflam-mation without evidence of calculi or sludge. AC istraditionally known to occur in critically ill patients,following cardiac surgery, abdominal vascular surgery,severe trauma, burns, prolonged fasting, total parenteral

J Physiol Biochem (2010) 66:161–172DOI 10.1007/s13105-010-0021-y

P. J. Gomez-Pinilla : P. J. Camello :M. J. PozoDepartment of Physiology, Nursing School,University of Extremadura,10071 Caceres, Spain

J. A. F. TresguerresDepartment of Physiology, Medicine School,University Complutense of Madrid,28040 Madrid, Spain

M. J. Pozo (*)Department of Physiology, Nursing School,Avda Universidad s/n,10003 Cáceres, Spaine-mail: mjpozo@unex.es

nutrition, or sepsis [2], but de novo presentation in theabsence of critical illness or predisposing factors hasalso been reported. AC is believed to have a morefulminant course, as well as significantly highermorbidity and mortality than gallstone-associated acutecholecystitis [15].

Although its pathogenesis is unknown, gallbladderstasis, residual volume, and impaired empty rate arealways present in AC, probably as the result of thedeleterious neural and muscular actions of inflammatorymediators such as reactive oxygen species released asconsequence of ischemia/reperfusion (I/R) injury andeicosanoid proinflammatory mediators [29]. In animalmodels, it has been described that neural contractionsare impaired in inflamed gallbladder and after I/R [11,27]. Previous reports have described that the targets forinflammation in smooth muscle cells were ion channelsand G protein-coupled receptors located in the plasmamembrane [39]. We have recently shown that calciumsignaling is altered in this pathological condition [12].

Pharmacological treatment aimed to improve I/Rand AC-associated gallbladder stasis would be ofinterest to recover gallbladder function and to avoidgallbladder failure, especially in patients at highsurgical risk. Since inflammation is associated toAC, different anti-inflammatory agents have beentested in animal models to prevent the deleteriousimpact of inflammation. Thus, the prostaglandinsynthase inhibitor indomethacin partly reversed in-flammation and contractile dysfunction but did notreverse the AC-induced decrease in nerve-evokedcontractions [28]. Melatonin has been proposed as aprotective agent against macromolecular destructionassociated with inflammation. We have shown thatmelatonin treatment has prophylactic and therapeuticeffects on I/R-induced impairment in gallbladderneuromuscular function [11] and protects Ca2+ ho-meostatic mechanisms located both in the plasma andsarcoplasmic membranes [12]. The protective effectsof melatonin could be just related to the ability of theindoleamine to reduce oxidative stress, but also to itsimmunomodulatory role in inflammation. It would beinteresting to know whether general antioxidants exertbeneficial effects on I/R-induced gallbladder smoothmuscle dysfunction and can be used as preventiveagents in medical conditions associated to I/R.

Tempol (4-hydroxy-2,2,6,6-tetramethylpiperidine-N-oxyl or 4-hydroxy-tempo) has a relatively lowmolecular weight (172) and permeates biological

membranes [20]. As antioxidant tempol scavengessuperoxide anions [20], acts as a genuine“SOD-mimetic,” and reduces the formation of hy-droxyl radicals [19]. Beneficial effects of tempol havebeen described in medical conditions related toelevated oxidative stress such as renal failure [6],hypertension [33], myocardial infarct [23], andneoplastic process [10]. In this study we evaluate theimpact of tempol on altered gallbladder contractilityevoked by I/R and demonstrate the benefits of tempoltreatment in this condition.

Methods

Animals and tempol treatment

Male guinea pigs, weighing 400 to 600 g, wereused in the study. I/R was induced to animals bycommon bile duct ligation for 2 days anddeligation for another 2 days, as described previ-ously [11, 12]. This method was approved by theEthical Committee of University of Extremadura. Inbrief, after anesthesia with 20 mg/kg i.p. ketaminehydrochloride and 5 mg/kg i.p. xylacine, a laparot-omy was performed, and the distal end of thecommon bile duct was ligated. Two days after, thecommon bile duct was deligated under anesthesiawith microsurgical scissors, and 2 days later, theanimals were euthanized with deep halothane anes-thesia and cervical dislocation for tissue harvest. Ascontrol, a group of guinea pigs were sham-operated(n=4), which included all of the surgical steps, withthe exception of common bile duct ligation. A groupof five animals were treated with tempol beforethe induction of I/R. Tempol was administered in thedrinking water at 1 mmol/l for 14 days before theanimal was sacrificed. This dose of tempol waspreviously shown to be effective in this model of I/R[11].

Functional studies

Gallbladders were removed, and they were immedi-ately placed in ice-cold Krebs–Henseleit solution(K-HS; for composition, see the “Solutions anddrugs” section) at pH7.35. The gallbladder was cutin longitudinal full-thickness strips (3×10 mm) thatwere placed vertically in a 10-ml organ bath filled

162 P.J. Gomez-Pinilla et al.

with Krebs–Henseleit solution maintained at 37°Cand gassed with 95% O2, 5% CO2. Isometriccontractions were measured using force displace-ment transducers that were interfaced with aMacintosh computer using a MacLab hardware unitand software (ADInstruments, Colorado Springs,CO, USA). The muscle strips were placed under aninitial resting tension equivalent to a 1.5-g load. Thedirect effects of acetylcholine (ACh), cholecystoki-nin (CCK), ionomycin, 60 mM of KCl, or capacita-tive calcium entry (CCE) protocol on gallbladdertone were studied. At the end of each experiment thedry weight of the strips was measured to normalizethe gallbladder contractile responses.

Cell isolation

Gallbladder smooth muscle cells were dissociatedenzymatically using a previously described method[12]. Briefly, the gallbladder was cut into small piecesand incubated for 34 min at 37°C in enzyme solution(ES, for composition see the “Solutions and drugs”section) supplemented with 1 mg/ml bovine serumalbumin (BSA), 1 mg/ml papain, and 1 mg/mldithioerythritol (DTT). Next, the tissue was trans-ferred to fresh ES containing 1 mg/ml BSA, 1 mg/mlcollagenase, and 100 µM CaCl2 and incubated for9 min at 37°C. Single smooth muscle cells wereisolated by several passages of the tissue piecesthrough the tip of a fire-polished glass Pasteur pipette.The resultant cell suspension was kept in ES at 4°Cuntil use, generally within 6 h. Cell viability wasroutinely checked by trypan blue staining of cellsfounding a similar viability (∼90%) in all group ofanimals, and gallbladder smooth muscle cell lengthwas also similar in the different groups. All experi-ments involving isolated cells were performed atroom temperature (22°C).

Cell loading and [Ca2+]i determination

[Ca2+]i was determined by epifluorescence microsco-py using the fluorescent ratiometric Ca2+ indicatorfura 2. Isolated cells were loaded with 4 µM fura 2-AM at room temperature for 25 min. An aliquot ofcell suspension was placed in an experimentalchamber made with a glass poly-D-lysine-treatedcoverslip (0.17 mm thick) filled with Na+–HEPESsolution (for composition see the “Solutions and

drugs” section) and mounted on the stage of aninverted microscope (Eclipse TE2000-S; Nikon).After cell sedimentation, a gravity-fed system wasused to perfuse the chamber with Na+–HEPESsolution in the absence or presence of experimentalagents. Cells were illuminated at 340 and 380 nm by acomputer-controlled monochromator (Optoscan,Cairn Research) at 0.3–1 cycles/s, and the emittedfluorescence was selected by a 510/40-nm band-passfilter. The emitted fluorescence images were capturedwith a cooled digital charge-coupled device camera(ORCAII-ER; Hamamatsu Photonics) and recordedusing dedicated software (Metafluor, Universal Imag-ing). The ratio of fluorescence at 340 nm tofluorescence at 380 nm (F340/F380) was calculatedpixel by pixel and used to indicate the changes in[Ca2+]i. A calibration of the ratio for [Ca2+]i was notperformed in view of the many uncertainties related tothe binding properties of fura 2 with Ca2+ inside ofsmooth muscle cells.

F-actin content measurement

The F-actin content of gallbladder smooth muscle cellswas determined according to a previously publishedprocedure [12]. Briefly, samples of cell suspensions(200 µl) were placed in Na+–HEPES solution andquickly transferred to 200 µl ice-cold 3% (w/v)formaldehyde in phosphate-buffered saline solution(PBS; for composition see the “Solution and drugs”section) for 10 min. Fixed cells were permeabilized byincubation for 10 min with 0.025% (v/v) Nonidet P-40detergent dissolved in PBS. Cells were then incubatedfor 30 min with fluorescein isothiocyanate-labeledphalloidin (FITC-phalloidin; 1 µM) in PBS solutionsupplemented with 0.5% (w/v) BSA. After incubation,the cells were collected by centrifugation for 2 min at10,000×g and resuspended in PBS solution. Stainingof actin filaments was measured using a confocal laser-scanning system (model MRC-1024, Bio-Rad) withexcitation wavelength of 488 nm and emission at515 nm. The cellular F-actin content was quantified asarbitrary units of fluorescence using the ImageJsoftware.

Estimation of mitochondrial membrane potential

Mitochondrial membrane potential was measured inisolated gallbladder muscle cells using the cell-

Tempol in gallbladder I/R 163

permeant green fluorescent dye rhodamine 123 thatis readily sequestered by active mitochondria.Cells were loaded with 10 µM rhodamine 123 at37°C during 10 min, excited at 490 nm, andimaged (long-pass filter 510 nm) using a fluores-cence microscope (Eclipse TE2000-S; Nikon,Melville, NY, USA). After background subtrac-tion, average and standard deviation of the cellularfluorescence were measured to calculate thecoefficient of variation (CV=(SD/Mean)×100),which has been shown to correlate with themitochondrial membrane potential [40]. A higherpotential (hyperpolarization) drives rhodamine intothe mitochondrial matrix and generates a punctuateddistribution of the dye, increasing the coefficient ofvariation, while dissipation of the mitochondrialmembrane potential (depolarization) leads to moreeven distribution of the dye, decreasing thecoefficient.

Western blot analysis

Gallbladder smooth muscle was homogenized in lysissolution (for composition see the “Solutions and drugs”section) using a homogenizer (Ika-Werke, Staufen,Germany) and then sonicated for 5 s. Lysates werecentrifuged at 10,000 g for 15 min at 4°C to removenuclei and unlysed cells, and the protein concentrationwas measured. Protein extracts (30 µg) were heat-denatured at 95°C for 5 min with DTT, electrophoresedon 7.5% polyacrylamide-SDS gels, and then trans-ferred to a nitrocellulose membrane. Membranes wereblocked for 1 h at room temperature using 10% BSAand incubated overnight at 4°C with affinity-purified mouse polyclonal antibodies for NF-κBp65 (1:500, BD Bioscience), COX-2 (1:500, BDBioscience), and inducible-NOS (1:500, BD Bio-science). A mouse anti-α tubulin monoclonalantibody (1:1,000, Santa Cruz Biotechnology) wasused as load control. After washing, the mem-branes were incubated for 1 h at room temperaturewith anti-mouse IgG-horseradish peroxidase conju-gated secondary antibody (1:10,000, AmershamBiosciences). The blots were then detected withthe Supersignal West Pico ChemiluminescentSubstrate (Pierce, IL, USA). The intensity of thebands was quantified using ImageJ software (NIH,Bethesda, MD, USA) and normalized respect to α-tubulin content.

Solutions and drugs

The K-HS contained (in mM): 113 NaCl, 4.7 KCl, 2.5CaCl2, 1.2 KH2PO4, 1.2 MgSO4, 25 NaHCO3, and 11.5D-glucose. This solution had a final pH of 7.35 afterequilibration with 95% O2–5% CO2. The ES used todisperse cells was made up of (in mM): 10 HEPES, 55NaCl, 5.6 KCl, 80 sodium glutamate, 2 MgCl2, and 10D-glucose, with pH adjusted to 7.3 with NaOH. TheNa+–HEPES solution contained (in mM): 10 HEPES,140 NaCl, 4.7 KCl, 2 CaCl2, 2 MgCl2, and 10 D-glucose, with pH adjusted to 7.3 with NaOH. TheCa2+-free Na+–HEPES solution was prepared by sub-stituting EGTA (1 mM) for CaCl2. The PBS solutionused in F-actin studies contained (in mM): NaCl 137,KCl 2.7, Na2HPO4 5.62, NaH2PO4 1.09, and KH2PO4

1.47 with pH adjusted to 7.2. The lysis solutioncontained (in mM): 50 mM Tris/HCl pH7.4, 150 mMNaCl, 1% Triton X-100, 1% deoxycholate, 0.5% (w/v)NaN3, 1 mM EGTA, 0.4 mM EDTA, 10 mM benza-midine, 25 µg/ml leupeptin, and 1 mM phenylmethyl-sulfonyl fluoride. Drug concentrations are expressed asfinal bath concentrations of active species. Drugs andchemicals were obtained from the following sources:ACh, (±) BayK8644, caffeine, CCK-8 sulfated,FITC-phalloidin, ionomycin, 1,4-dithio-DLthreitol,nitrendipine, pinacidil, tempol (4-hydroxy-2,2,6,6-tetra-methylpiperidine-N-oxyl or 4-hydroxy-tempo), andthapsigargin were from Sigma Chemical (St. Louis,MO, USA); 2-aminoethoxydiphenylborane (2-APB)from Tocris (Bristol, UK); fura 2-AM and rhodamine123 were from Molecular Probes (Molecular ProbesEurope, Leiden, the Netherlands); collagenase was fromFluka (Madrid, Spain), and papain was from Worthing-ton Biochemical (Lakewood, NJ, USA). Other chem-icals used were of analytical grade from Panreac(Barcelona, Spain). Stock solutions of 2-APB, fura 2-AM, ionomycin, pinacidil, and thapsigargin were pre-pared in dimethyl sulfoxide (DMSO) and (±) BayK8644,FITC-phalloidin and nitrendipine were prepared inethanol. The solutions were diluted such that the finalconcentrations of DMSO or ethanol were ≤0.1% v/v.These concentrations of solvents did not interfere withfura 2 fluorescence nor affected to strip tone.

Quantification and statistics

Results are expressed as means±SEM of n cells,gallbladder strips, or blots. All results from [Ca2+]i

164 P.J. Gomez-Pinilla et al.

determinations are given as maximal amplitude of thepeak (ΔF340/F380). Gallbladder tension is given inmillinewtons (mN)/milligram of tissue. Statisticaldifferences between means were determined byanalysis of variance for repeated measures (ANOVAone-way), followed by Bonferroni post hoc test.Differences were considered significant at P<0.05.

Results

Effects of tempol on I/R-induced gallbladderdysfunction

As described previously [12], experimental I/R inducesimpaired gallbladder contractility in response to theneurotransmitter ACh (10 µM), the hormone CCK(10 nM), activation of Ca2+ influx through L-type Ca2+

channels by 60 mM KCl, and activation of CCE(Fig. 1). The protocol for activation of CCE consisted

in the depletion of the intracellular calcium stores byincubation with 1 µM of thapsigargin in a calcium-freemedium for 30 min and the following reintroduction ofextracellular Ca2+, which induces a sustained contrac-tion [24]. This impairment in contractility is due toinflammation and the concomitant oxidative stresssince the contractile response of sham-operated ani-mals does not differ from that found in control animals(ACh: control 3.235±0.288, sham 3.952 ± 0.489mN/mg; CCK: control 3.692 ± 0.527, sham 3.095±0.277 mN/mg; 60 mM of KCl: control 3.369±0.413,sham 3.617 ± 0.291 mN/mg; CCE: control3.722 ± 0.656 and sham 4.197±0.481 mN/mg;n=10–8 strips from eight and four animals, respec-tively), and the antioxidant tempol amelioratedI/R-induced hypo-contractility (Fig. 1).

In a previous report we described that tempoltreatment decreases the oxidative stress induced byI/R [11]. Here we show that this decrease runs inparallel with a significant (P<0.01) improvement in

60 s

1 m

N/m

g

10 μM ACh

60 s

1 m

N/m

g

60 s

1 m

N/m

g

60 s

1 m

N/m

g

10 nM CCK

60 mM KCl CCE

Control I/R I/R +Tempol

**

****

**

10 μM Acetylcholine

Control I/R I/R+Tempol0

1

2

3

4

5

Ten

sio

n (

mN

/mg

)

60 mM KCl

Control I/R I/R+Tempol0

1

2

3

4

5

Ten

sio

n (

mN

/mg

)

CCE

Control I/R I/R+Tempol0

1

2

3

4

5

Ten

sio

n (

mN

/mg

)

10 nM Cholecystokinin

Control I/R I/R+Tempol0

1

2

3

4

5

Ten

sio

n (

mN

/mg

)

ba

dc

Fig. 1 Tempol treatment improves myogenic gallbladdercontractile response impaired by ischemia/reperfusion (I/R).Gallbladder smooth muscle strips from three groups of animals:control, common bile duct ligated and then deligated (I/R), andI/R treated with tempol for 14 days (I/R+tempol) werechallenged with a 10 µM acetylcholine (ACh), b 10 nMcholecystokinin (CCK), c 60 mM KCl, or d capacitative

calcium entry protocol (CCE), and isometric contraction wasrecorded. Gallbladder I/R induced a clear impairment in thecontraction and tempol treatment increased contractility for alltested agonists. Traces are representative of 6–12 strips fromeight, eight, and five animals, respectively. Histograms showthe gallbladder contractile response for the three experimentalgroups (mean±SEM). **P<0.01 I/R vs I/R+tempol

Tempol in gallbladder I/R 165

the myogenic gallbladder contractile response(ACh: I/R 30.80% of control and I/R + tempol56.99% of control; CCK: I/R 19.18% of control andI/R + tempol 37.56% of control; 60 mM of KCl: I/R28.06% of control and I/R + tempol 54.01% ofcontrol; CCE: I/R 25.62% of control and I/R + tempol62.58% of control; n=12–4 strips from eight and fiveanimals, respectively; Fig. 1). The rest of the paperwas designed to determine the mechanisms throughwhich tempol exerts its beneficial effects on gallblad-der contractility.

Effects of tempol on calcium handling

[Ca2+]c is the main second messenger that modulatessmooth muscle contraction. Thus, it could be possiblethat tempol treatment avoided damage of Ca2+

homeostasis induced by AC resulting in the increase

in [Ca2+]î available for contraction. As shown inFig. 2 tempol treatment normalized calcium influxthrough voltage-operated channels (60 mM KCl;control 0.230±0.014, I/R 0.103±0.013, and I/R+tempol 0.241 ± 0.017 ΔF340/F380, n=29–12 cellsfrom six, six, and five animals, respectively, P<0.01I/R vs I/R+tempol, Fig. 2a). This normalization isaccompanied with the reestablishment of the sensitiv-ity of the channels to nitrendipine and pinacidil(inhibitor of L-type Ca2+ channels and opener ofadenosine triphosphate (ATP)-sensitive K+channels,respectively; Table 1), lost in I/R. Similar effects wereobtained when L-type Ca2+ channels were selectivelyactivated with 1 µM of (±) BayK8644 (control 0.185±0.010, I/R 0.072 ± 0.006, and I/R+tempol 0.189±0.012 ΔF340/F380, n=33–15 cells from six, six, and fiveanimals, respectively, P<0.01, I/R vs I/R+ tempol).(±) BayK8644-evoked calcium influx was reduced by

40 s

0.1

ΔF34

0/F

380

Control

I/R

I/R +Tempol

60 mM KCl

0.05

ΔF

340/

F38

0

40 s

Control

I/R

I/R+Tempol

CCE

**

60 mM KCl

Control I/R I/R +Tempol0.0

0.1

0.2

0.3Δ

F34

0/F

380

CCE

Control I/R I/R+Tempol0.00

0.05

0.10

0.15

ΔF 3

40/F

380

**

a

b

Fig. 2 Tempol treatment restores calcium influx after I/Rinjury. Original recording of calcium influx in response to adepolarizing solution containing a 60 mM KCl and b CCEprotocol in cells from control, I/R, and I/R+tempol. I/R reducedthe sustained calcium influx evoked by the activation ofvoltage-operated and store-operated calcium channels and

tempol treatment normalize both. Traces are typical of 20–12cells from six, six, and five animals, respectively. Histogramsshow summary data of ΔF340/F380 from experiments in theabove described conditions (mean±SEM). **P<0.01 I/R vs I/R+tempol

166 P.J. Gomez-Pinilla et al.

pinacidil in control (79.39% inhibition) and in thetempol-treated group (77.61% inhibition) but unaffect-ed in cholecystic cells (−4.46% inhibition).

When Ca2+ influx through capacitative channels wasactivated by depletion of Ca2+ stores, tempol preventedthe impairment of influx under inflammation conditions(CCE; control 0.090±0.003, I/R 0.051 ± 0.004, andI/R+tempol 0.120±0.021 ΔF340/F380, n=49–40 cellsfrom six, six, and five animals, respectively, P<0.01I/R vs I/R+tempol, Fig. 2b). In addition, tempolreestablished the sensitivity of CCE to nitrendipinealthough Ca2+ influx was still insensitive to the blocker2-APB (Table 2). These results indicate that tempolnormalizes Ca2+ influx, the main source of Ca2+ forcontraction.

Smooth muscle cells also use intracellular Ca2+ frominternal stores. When the size of the pool of calciumreleasable was determined by the application of lowdose of ionomycin (50 nM) in a free-calcium medium,we found similar results in all the experimental groups(control 0.227±0.036, I/R 0.247 ± 0.016, and I/R+tempol 0.184±0.013 ΔF340/F380, n=20–13 cells fromsix, six, and five animals, respectively). Similar to thesize of Ca2+ internal stores, we did not find changes incalcium released by CCK (through IP3- and ryanodine-

receptors) in the three animal groups analyzed (data notshown).

Effects on contractile machinery

The increase in KCl- and CCE-evoked calcium influxevoked by tempol treatment indicates that the improve-ment in the gallbladder contractility is partially supportedby calcium-dependent mechanisms. However, the find-ing that tempol also improves CCK-evoked contractionwithout changes in the CCK-evoked calcium releasesuggests that tempol also operates by a mechanismindependent on Ca2+ mobilization. This possibility wastested by application of 1 µM ionomycin in presence ofextracellular Ca2+ (at this concentration ionomycinraises [Ca2+]c independently of channels and receptors).Ionomycin caused similar elevation in [Ca2+]c in all theexperimental groups (control 0.305±0.025, I/R0.272 ± 0.036, and I/R+ tempol 0.299±0.042ΔF340/F380, n=32–17 cells from six, six, and fiveanimals, respectively) but the contractile responseinduced by 1 µM of ionomycin was smaller in I/Rstrips and tempol treatment improved this contraction(control 3.050±0.252, I/R 0.457±0.103, and I/R+tempol 1.84±0.19 mN/mg, n=7–4 strips from six, six,and five animals, respectively). These data support thehypothesis that inflammation could alter directly thecontractile machinery and that tempol beneficial effectscan also be mediated through the protection of thecontractile apparatus. To investigate this possibility, wedetermined the total amount of the contractile proteinF-actin by labeling isolated gallbladder smooth musclecells with FITC-conjugated phalloidin. As shown inFig. 3, I/R resulted in a statistically significant (P<0.01,control vs I/R) decrease in F-actin content that waspartially prevented by tempol treatment (P<0.01, I/R vsI/R+tempol).

Effects on mitochondrial function

The possible effect of I/R and tempol treatment onmitochondrial function was investigated. In additionto ATP production mitochondria are the major sourceof ROS, and consequently, a mitochondrial dysfunc-tion can have harmful oxidative stress-related effectsat cellular level. Here, we explore mitochondria statususing rhodamine 123, an amphiphilic cation whichaccumulates within active, polarized mitochondriafollowing Nernstial equilibrium. Thus, in cells with

Table 2 Effect of acute inflammation and tempol treatment onCCE-evoked calcium influx pharmacology

Control I/R I/R+Tempol

Nitrendipine(1 µM)

42.17 −9.88** 56.37

2-APB (100 µM) 48.90 −14.22** −3.54**Both 91.65 −11.82** 59. 40**

Data are expressed as percent of inhibition for each drug in thethree experimental groups; n=6, 6, and 5 animals, respectively

**P<0.01 vs control

Table 1 Effect of acute inflammation and tempol treatment onKCl-evoked calcium influx pharmacology

Control I/R I/R+Tempol

Nitrendipine(1 µM)

94.68 4.82** 80.00

Pinacidil (10 µM) 85.93 5.11** 87.64

Data are expressed as percentage of inhibition for each drug inthe three experimental groups; n=6, 6, and 5 animals,respectively

**P<0.01

Tempol in gallbladder I/R 167

low mitochondrial potential the dye will redistributeto the cytosol and the fluorescence will be spatiallyhomogeneous compared to cells with more polarizedmitochondria. Therefore, the coefficient of variationof cell fluorescence has been successfully used inseveral models to compare the overall mitochondrialpotential of different cell batches [5, 40]. As shownFig. 4, after I/R, cells presented lower mitochondrialfunctionality than control cells (P<0.01, control vsI/R), noted by a reduction in coefficient of variation.Gallbladder smooth muscle cells from tempol-treatedanimals show an increase in the coefficient (P<0.01,I/R vs I/R+tempol), indicative of higher mitochon-drial functionality than I/R cells.

Effects on inflammation markers

Injurious and inflammatory stimuli, such as freeradicals, activate NF-κB-mediated overproduction ofkey proinflammatory mediators as chemokines, cyto-kines, COX-2, and iNOS, which are attributed to theinitiation and progression of inflammatory gastroin-testinal diseases in human and animal models. Weassayed the expression of NF-κB, COX-2, and iNOSin guinea pig gallbladder smooth muscle by Westernblot. As observed in Fig. 5, NF-κB, COX-2, andiNOS expression were significantly increased in I/Rgallbladder smooth muscle (P<0.01). Tempol treat-ment reduced the expression of these inflammatory

markers, which supports the beneficial effects oftempol in I/R injury.

Discussion

The current paper shows that the impairment inguinea pig gallbladder myogenic contraction evokedby inflammation can be ameliorated by tempoltreatment. Tempol exerts its beneficial effects throughcalcium-dependent and independent mechanisms suchas contractile machinery and mitochondria function.The expression pattern of classical inflammationmarkers as NF-κB, COX-2, and iNOS suggests thattempol treatment avoids inflammation too.

In this paper, we use a model of ischemia/reperfusion-induced inflammation. Bile duct ligatureis a commonly used model of acute acalculouscholecystitis [13, 25–27] which has provided wideinformation about the effect of inflammation ongallbladder. However, in this experimental model thegallbladder suffers ischemia and increasing stretch asthe result of bile duct obstruction and the continuousbile output. This makes it difficult to see anyimprovement in the gallbladder contractility whenpossible treatments are assayed. To solve this, weintroduced a variation of the experimental model inwhich, after common bile ligation, we deligated thebile duct and after 2 days we harvested the tissue [11].

**

*

Control I/R I/R + Tempol

50 µm

Control I/R I/R+Tempol0

2500

5000

7500

F-A

ctin

Co

nte

nt

(AU

F)

a b

Fig. 3 Altered contractile machinery by I/R is improved bytempol treatment. a Gallbladder smooth muscle cell from thethree experimental groups were stained with 1 µM FITC-phalloidin to determine the F-actin content as index ofcontract i le machinery status. b Histograms showinflammation-related decrease in F-actin content (expressed as

arbitrary units, mean±SEM) in gallbladder cells and thebeneficial effect of the tempol treatment. Note the improvementin F-actin content when the I/R group was treated with tempol.n=96–144 cells from four, four, and three animals, respective-ly; *P<0.05 and **P<0.01 vs control. Scale bar indicates50 µm

168 P.J. Gomez-Pinilla et al.

This maneuver worsened the myogenic gallbladdercontractility [12] as result of an increase in the tissueoxidative stress [11, 12] probably due to reperfusion,but we found a recovery of contractility underpharmacological treatment [11, 12].

Tempol is a stable piperidine nitroxide of lowmolecular weight that permeates biological mem-branes and accumulates in the cytosol [20]. In avariety of animal models, deleterious effects ofreperfusion injury on both local and remote organshave been demonstrated. Tempol administration sig-nificantly reduced organ injuries caused by I/R ingastric mucosa [1], intestine [37], myocardium [22],kidney [9], liver [35], and brain [17]. We have alsoreported protective effects of tempol in the gallblad-der neurotransmission after I/R injury through thereduction of the increase in gallbladder oxidativestress evoked by inflammation [11], which is inkeeping with rising evidence that ROS contributes togallbladder injury [7, 11, 12, 29] and that tempolscavenges superoxide anions and acts as a genuine“SOD-mimetic” [19].

Here, we describe that tempol treatment also protectsintracellular pathways related to smooth muscle con-traction. According to our results, calcium channels inthe plasma membrane are targets for ROS generatedafter reperfusion. Tempol protects these channels,maintaining the permeability and the sensitivity topharmacological blockade. This beneficial effect could

explain the improvement in the contractile response todepolarization and Ca2+ store depletion. In addition,our study proves that calcium-independent mechanismsfor contraction can also be protected by tempoltreatment. Thus, the improvement of CCK- andionomycin-evoked contractile responses in tempol-treated animals, despite similar Ca2+ mobilization inthe experimental groups, can be related to theprotection of the contractile machinery by tempol. Inkeeping with this, it has been shown that the persistentpresence of intracellular ROS, in neuronal cells,provoked oxidative damage to actin coinciding withthe increase in proinflammatory cytokines TNF-α andIL-1β [3]. Similarly, H2O2 and peroxynitrite inhibitmyofibrillar protein function, contributing to contrac-tile dysfunction in pathologic states in which ROS areliberated [30, 36].

Mitochondria are the intracellular organelles respon-sible for the synthesis of ATP via oxidative phosphor-ylation, where electrons flowing from reducedsubstrates in the respiratory chain are occasionallytransferred to oxygen to form superoxide anion andderived products that act as signaling molecules inseveral cellular processes [4, 38]. A very originalfinding in the present study is that mitochondriamembrane potential, essential for normal function, isprotected by tempol in I/R injury. The reduction inmembrane potential described in I/R group could beresponsible of the impairment in the contractile

Control I/R I/R + Tempol

50 µm

*

**

Control I/R I/R+Tempol0

10

20

30

Var

iati

on

co

effi

cien

t (%

)

a b

Fig. 4 I/R evoked mitochondria dysfunction and tempoltreatment improves it. a Gallbladder smooth muscle cell fromthe three experimental groups were stained with Rhodamine123 to determine mitochondrial membrane potential status. bHistograms show mitochondrial membrane potential expressedas coefficient of variation (mean±SEM). Membrane depolar-

ization causes a decrease in the coefficient of variation (see the“Methods” section). Note the decrease in membrane potentialafter I/R and that tempol improves mitochondria functionality.n=105–203 cell from four, four, and three animals, respective-ly; *P<0.05 and **P<0.01 vs control. Scale bar indicates50 µm

Tempol in gallbladder I/R 169

response since it leads to the reduction of ATPsynthesis [14]. It has been shown that ROS generationduring ischemia reperfusion injury causes mitochon-drial membrane depolarization and that antioxidanttreatment prevents it and cell death [18, 21]. We havepreviously shown that tempol-reduced ROS generation[11] and we report now that tempol treatment improvesmitochondrial membrane potential. The reestablish-ment of membrane potential could also lead not only torestoration of energetic supply for contraction but alsoto normalization of altered calcium signal, given theimportance of mitochondria in calcium handling [5].

The transcription factor NF-kB is known to be aredox-sensitive factor that plays a central role inimmune responses and inflammation. NF-kB istrapped in the cytoplasm in stimulated cells andtranslocated into the nucleus in response to several

stimuli, including oxidative stress. Reactive oxygenspecies enhance the signal transduction pathways forNF-kB activation in the cytoplasm and translocationinto the nucleus, where it regulates gene expressionby binding to promoter regions of a large number ofgenes involved in the inflammatory response, such asIL-1, TNF-α, IL-6, COX-2, and iNOS. Here, weshow that gallbladder inflammation associated to I/Rruns in parallel with an increase in p65 NF-kB subunitexpression, essential for nuclear localization and bind-ing to specific DNA sequences. The increase in p65subunit could be related with the increase in theexpression of COX-2 and iNOS found in inflamedgallbladder. Many studies have shown that nitric oxideand prostaglandin (synthesized by iNOS and COX-2,respectively) are the main inflammatorymediators in thepathogenesis of gastrointestinal tract diseases that

65 kDa

55 kDa

NF-κB

α-Tubulin

Cont I/R I/R +Tempol

75 kDa

55 kDa

COX-2

α-Tubulin

Cont

130 kDa

55 kDa

iNOS

α-Tubulin

Cont

*

*

*

I/R I/R +Tempol

I/R I/R +Tempol

I/R I/R + Tempol0

1

2

Control

NF

-ΚB

Exp

ress

ion

I/R I/R + Tempol0

1

2

3

Control

CO

X-2

Exp

ress

ion

I/R I/R + Tempol0

1

2

ControlIN

OS

Exp

ress

ion

a

b

c

Fig. 5 I/R increased theexpression of NF-κB,COX-2, and iNOS andtempol treatment normalizesit. Typical example orWestern blot analysis ofa NF-κB, b COX-2, andc iNOS protein expressionin gallbladder from the threeexperimental groups. 30 µgof protein was loaded ineach well, and α-tubulinwas used as loading control.Histograms show I/R-related increase in protein ofinterest expression as foldincreases with respect tocontrol tissue (dotted line).Data are means±SEM offour to six experiments withproteins from six, six, andfour animals, respectively.*P<0.05 vs control

170 P.J. Gomez-Pinilla et al.

correlate with enhanced expression of COX-2, iNOS,and NF-κB [8, 16, 31, 32, 34]. The role of NF-κB as arate-limiting step in the inflammatory cascade hasmade it an attractive target for drug treatment. Tempoltreatment reduces the expression of NFκB, COX-2,and iNOS, supporting the hypothesis that NF-κB islocated up-stream of both markers of inflammation andthat a decrease in its expression leads to inflammationresolution. This is the first report showing tempol-evoked reduction in the expression of markers ofinflammation which converts tempol in a goodcandidate to be tested as a no steroidal anti-inflammatory drug.

In conclusion, tempol protects gallbladder contrac-tility from I/R injury preventing mitochondria dys-function and the increased expression of inflammationmarkers. This results in normalization of calciumhomeostasis and protection of contractile machineryand almost normal contraction. These findings evi-dence that tempol has a clinical potential as aprotective drug in inflammatory and I/R conditions.

Acknowledgments The authors thank Purificación Delgadofor technical assistance. This work was supported byMinisterio deEducacion y Ciencia (BFU 2007-60563), Junta de Extremadura(PRI07A069), FEDER and Instituto de Salud Carlos III(RETICEF: RD06/0013/1012 and RD06/0013/0002)

References

1. Abdallah DM, El Abhar HS, Abdel-Aziz DH (2009)TEMPOL, a membrane-permeable radical scavenger,attenuates gastric mucosal damage induced by ischemia/reperfusion: a key role for superoxide anion. Eur JPharmacol 603:93–97

2. Barie PS, Eachempati SR (2003) Acute acalculous chole-cystitis. Curr Gastroenterol Rep 5:302–309

3. Barth BM, Stewart-Smeets S, Kuhn TB (2009) Proinflam-matory cytokines provoke oxidative damage to actin inneuronal cells mediated by Rac1 and NADPH oxidase. MolCell Neurosci 41:274–285

4. Camello-Almaraz C, Gomez-Pinilla PJ, Pozo MJ, CamelloPJ (2006) Mitochondrial reactive oxygen species and Ca2+signaling. Am J Physiol Cell Physiol 291:C1082–C1088

5. Camello-Almaraz C, Gomez-Pinilla PJ, Pozo MJ, CamelloPJ (2008) Age-related alterations in Ca2+ signals andmitochondrial membrane potential in exocrine cells areprevented by melatonin. J Pineal Res 45:191–198

6. Chatterjee PK, Cuzzocrea S, Brown PA, Zacharowski K,Stewart KN, Mota-Filipe H, Thiemermann C (2000)Tempol, a membrane-permeable radical scavenger, reducesoxidant stress-mediated renal dysfunction and injury in therat. Kidney Int 58:658–673

7. Cullen JJ, Conklin JL, Ephgrave KS, Oberley LW (1999)The role of antioxidant enzymes in the control of opossumgallbladder motility. J Surg Res 86:155–161

8. Dong WG, Mei Q, Yu JP, Xu JM, Xiang L, Xu Y (2003)Effects of melatonin on the expression of iNOS and COX-2in rat models of colitis. World J Gastroenterol 9:1307–1311

9. Fujii T, Takaoka M, Ohkita M, Matsumura Y (2005)Tempol protects against ischemic acute renal failure byinhibiting renal noradrenaline overflow and endothelin-1overproduction. Biol Pharm Bull 28:641–645

10. Gariboldi MB, Lucchi S, Caserini C, Supino R, Oliva C,Monti E (1998) Antiproliferative effect of the piperidinenitroxide TEMPOL on neoplastic and nonneoplastic mam-malian cell lines. Free Radic Biol Med 24:913–923

11. Gomez-Pinilla PJ, Morales S, Camello-Almaraz C, MorenoR, Pozo MJ, Camello PJ (2006) Changes in guinea piggallbladder smooth muscle Ca2+ homeostasis by acuteacalculous cholecystitis. Am J Physiol Gastrointest LiverPhysiol 290:G14–G22

12. Gomez-Pinilla PJ, Camello PJ, Pozo MJ (2007) Effects ofmelatonin on gallbladder neuromuscular function in acutecholecystitis. J Pharmacol Exp Ther 323:138–146

13. Gomez-Pinilla PJ, Camello PJ, Pozo MJ (2008) Protectiveeffect of melatonin on Ca2+ homeostasis and contractilityin acute cholecystitis. J Pineal Res 44:250–260

14. Huttemann M, Lee I, Pecinova A, Pecina P, Przyklenk K,Doan JW (2008) Regulation of oxidative phosphorylation,the mitochondrial membrane potential, and their role inhuman disease. J Bioenerg Biomembr 40:445–456

15. Johnson LB (1987) The importance of early diagnosis ofacute acalculus cholecystitis. Surg Gynecol Obstet164:197–203

16. Kankuri E, Asmawi MZ, Korpela R, Vapaatalo H,Moilanen E (1999) Induction of iNOS in a rat model ofacute colitis. Inflammation 23:141–152

17. Kato N, Yanaka K, Hyodo K, Homma K, Nagase S, Nose T(2003) Stable nitroxide Tempol ameliorates brain injury byinhibiting lipid peroxidation in a rat model of transientfocal cerebral ischemia. Brain Res 979:188–193

18. Kim JS, Jin Y, Lemasters JJ (2006) Reactive oxygenspecies, but not Ca2+ overloading, trigger pH- andmitochondrial permeability transition-dependent death ofadult rat myocytes after ischemia-reperfusion. Am J PhysiolHeart Circ Physiol 290:H2024–H2034

19. Krishna MC, Russo A, Mitchell JB, Goldstein S, Dafni H,Samuni A (1996) Do nitroxide antioxidants act as scav-engers of O2-. or as SOD mimics? J Biol Chem271:26026–26031

20. Laight DW, Carrier MJ, Anggard EE (1997) Investigationof role for oxidant stress in vascular tolerance developmentto glyceryl trinitrate in vitro. Br J Pharmacol 120:1477–1482

21. Levraut J, Iwase H, Shao ZH, Vanden Hoek TL, SchumackerPT (2003) Cell death during ischemia: relationship tomitochondrial depolarization and ROS generation. Am JPhysiol Heart Circ Physiol 284:H549–H558

22. McCormick J, Barry SP, Sivarajah A, Stefanutti G, TownsendPA, Lawrence KM, Eaton S, Knight RA, Thiemermann C,Latchman DS, Stephanou A (2006) Free radical scavenginginhibits STAT phosphorylation following in vivo ischemia/reperfusion injury. FASEB J 20:2115–2117

Tempol in gallbladder I/R 171

23. McDonald MC, Zacharowski K, Bowes J, Cuzzocrea S,Thiemermann C (1999) Tempol reduces infarct size inrodent models of regional myocardial ischemia andreperfusion. Free Radic Biol Med 27:493–503

24. Morales S, Camello PJ, Alcon S, Salido GM, Mawe G,Pozo MJ (2004) Coactivation of capacitative calcium entryand L-type calcium channels in guinea pig gallbladder. AmJ Physiol Gastrointest Liver Physiol 286:G1090–G1100

25. Myers SI, Haley-Russell D, Parks L, Husband K (1988)Common bile duct ligation in rabbit: a new model of acutecholecystitis description of histology and bile analysis. JSurg Res 45:556–564

26. Parkman HP, Bogar LJ, Bartula LL, Pagano AP, ThomasRM, Myers SI (1999) Effect of experimental acalculouscholecystitis on gallbladder smooth muscle contractility.Dig Dis Sci 44:2235–2243

27. Parkman HP, James AN, Bogar LJ, Bartula LL, ThomasRM, Ryan JP, Myers SI (2000) Effect of acalculouscholecystitis on gallbladder neuromuscular transmissionand contractility. J Surg Res 88:186–192

28. Parkman HP, James AN, Thomas RM, Bartula LL, RyanJP, Myers SI (2001) Effect of indomethacin on gallbladderinflammation and contractility during acute cholecystitis. JSurg Res 96:135–142

29. Pozo MJ, Camello PJ, Mawe GM (2004) Chemicalmediators of gallbladder dysmotility. Curr Med Chem11:1801–1812

30. Qian Y, Luo J, Leonard SS, Harris GK, Millecchia L, FlynnDC, Shi X (2007) Hydrogen peroxide formation and actinfilament reorganization by Cdc42 are essential for ethanol-induced in vitro angiogenesis. Nihon Arukoru YakubutsuIgakkai Zasshi 42:605–609

31. Rogler G, Brand K, Vogl D, Page S, Hofmeister R, AndusT, Knuechel R, Baeuerle PA, Scholmerich J, Gross V(1998) Nuclear factor kappaB is activated in macrophages

and epithelial cells of inflamed intestinal mucosa. Gastro-enterology 115:357–369

32. Sakamoto C (1998) Roles of COX-1 and COX-2 ingastrointestinal pathophysiology. J Gastroenterol 33:618–624

33. Schnackenberg CG, Wilcox CS (1999) Two-week admin-istration of tempol attenuates both hypertension and renalexcretion of 8-Iso prostaglandin f2alpha. Hypertension33:424–428

34. Schreiber S, Nikolaus S, Hampe J (1998) Activation ofnuclear factor kappa B inflammatory bowel disease. Gut42:477–484

35. Sepodes B, Maio R, Pinto R, Marques C, Mendes-do-ValeJ, McDonald MC, Thiemermann C, Mota-Filipe H (2004)Tempol, an intracelullar free radical scavenger, reducesliver injury in hepatic ischemia-reperfusion in the rat.Transplant Proc 36:849–853

36. Snook JH, Li J, Helmke BP, Guilford WH (2008)Peroxynitrite inhibits myofibrillar protein function in anin vitro assay of motility. Free Radic Biol Med 44:14–23

37. Teke Z, Kabay B, Ozden A, Yenisey C, Bir F, DemirkanNC, Bicakci T, Erdem E (2008) Effects of tempol, amembrane-permeable radical scavenger, on local andremote organ injuries caused by intestinal ischemia/reper-fusion in rats. J Surg Res 149:259–271

38. Turrens JF (2003) Mitochondrial formation of reactiveoxygen species. J Physiol 552:335–344

39. Xiao ZL, Chen Q, Biancani P, Behar J (2001) Abnormalitiesof gallbladder muscle associated with acute inflammation inguinea pigs. Am J Physiol Gastrointest Liver Physiol 281:G490–G497

40. Xiong J, Camello PJ, Verkhratsky A, Toescu EC (2004)Mitochondrial polarisation status and [Ca2+]i signalling inrat cerebellar granule neurones aged in vitro. NeurobiolAging 25:349–359

172 P.J. Gomez-Pinilla et al.