Urodynamic changes in a noble rat model for nonbacterial prostatic inflammation

The Prostate 67:888 ^ 899 (2007)

UrodynamicChanges in aNoble RatModel forNonbacterial Prostatic Inflammation

Jenni Bernoulli,1* Emrah Yatkin,1 Eva-Maria Talvitie,1

Risto Santti,1 and Tomi Streng1,2

1Institute of Biomedicine,Departmentof Anatomy,Universityof Turku,Turku, Finland2DivisionofGeneticsand Physiology,Departmentof Biology, Laboratoryof Animal Physiology,

Universityof Turku,Turku, Finland

BACKGROUND. Chronic nonbacterial prostatitis (CP) associatedwith voidingdysfunction isa poorly understood clinical phenomenon. The goal of the present studywas to induce prostaticinflammation with estrogen and androgen treatment and to record associated urodynamicchanges in Noble rats.METHODS. Rats were treated with estradiol and testosterone implants to increase estradiolconcentration in serumwhile testosterone concentrationwasmaintained at or slightly above thecontrol level. Theurodynamical recordingswereperformedunder anesthesia after thehormonetreatments for 3 and 6 weeks. The dorsolateral lobes of the prostates were removed forhistopathological analysis after recordings.RESULTS. After the 3-week treatment, lymphocytes, mainly T-cells, were located around thecapillaries. During the following 3 weeks lymphocytes migrated into stroma and acini.Cytotoxic T-cells were seen intraepithelially, and neutrophiles inside the acini. Removal ofestrogen implant or treatment with anti-estrogen diminished inflammation. No changes invoiding pattern were seen after the 3-week treatment. Three weeks later, bladder weight andcapacity were increased, and the micturition time was prolonged.CONCLUSIONS. Elevated estrogen concentration was essential for the gradual developmentof prostatic inflammation. The profile and location of inflammatory cells suggest that prostaticvasculature is one of the sites of estrogen action. Urodynamic changes which developed inassociation with glandular inflammation indicated abnormal bladder function, reflecting anincipient obstruction. Prostate 67: 888–899, 2007. # 2007 Wiley-Liss, Inc.

KEY WORDS: prostatitis; estrogen; micturition

INTRODUCTION

Menwith symptoms of chronic prostatitis, indicatedtypically by pain or discomfort on ejaculation (chronicprostatitis/chronic pelvic pain syndrome (CP/CPPS),NIH Category III), have commonly voiding com-plaints. They have higher LUTS scores, and they differfrom those with presenting problem with LUTS only[1]. The voiding symptoms do not correlate withurodynamic findings, but a significant proportion ofthe patients treated for CP/CPPS may have bladderoutlet obstruction or bladder dysfunction [2].

Numerous mechanistic explanations have beenproposed for the association of human voidingdysfunctions and CP/CPPS. Among them, high

pressure voiding causing reflux of urine or semen intoprostatic ducts leading to an increased intraprostaticpressure is often mentioned [3]. Increased ductalpressure which would develop first causes destruction

Grant sponsor: National Technology Agency of Finland; Grantnumber: 40326/03; Grant sponsor: Hormos Medical Corporation(Turku, Finland); Grant sponsor: Orion Corporation Orion Pharma(Espoo, Finland).

*Correspondence to: Jenni Bernoulli, MSc, Institute of Biomedicine,Department of Anatomy, University of Turku, Kiinamyllynkatu 10,FIN-20520, Turku, Finland. E-mail: [email protected] 15 November 2006; Accepted 12 January 2007DOI 10.1002/pros.20567Published online 17 April 2007 in Wiley InterScience(www.interscience.wiley.com).

� 2007Wiley-Liss, Inc.

of glandularwalls and stromal leakage of inflammatoryfactors present in secretion. Inflammatory factorssuch as NGF would recruit nociceptive C fibers andcause chronic pain [4,5]. Another possibility is thatchronic prostatitis causes poorly relaxed pelvic floormuscles through neurogenic mechanisms. Experimen-tal evidence shows that irritation of prostate activatesafferent lumbosacral neurone overlapping nociceptiveneurons receiving inputs from pelvic floor [6,7]. Thismechanism may account for the referred perinealpain associated with CP/CPPS as well as dysfunctionsof pelvic floor muscles. In this case, nonbacterialprostatitis develops first (e.g., autoimmune or endo-crine basis), and voiding dysfunctions will follow.Therefore, the sequential relationship between pro-static inflammation and possible voiding dysfunctionsremains unclear.

The goal of the present study was to follow thegradual development of prostatic inflammation andconcomitant alterations in voiding of Noble rats.Although hormonally manipulated Noble rats havebeen used for a long time as a model for prostatecarcinogenesis [8], neither prostatic inflammation norchanges in urodynamics associated with chronicprostatitis in this strain have been characterized. Byusing estrogen and androgen implants, an elevatedestrogen concentration in serum essential for inductionof nonbacterial prostatitis was achieved, while serumandrogen concentrationwasmaintained at or above thecontrol level to rule out the possible effect of androgendeficiency. The present study indicates that prostaticinflammation and voiding dysfunction developedunder the same endocrinological conditions, andadditionally, that the gradual development of theinflammation from perivascular to glandular formtook place before the signs of incipient obstructivevoiding became evident in urodynamic recordings.Present findings do not clarify the questionwhether theinflammatory changes and voiding dysfunctions arecausally related.

MATERIALSANDMETHODS

Experimental Animals andDesign of the Experiments

The present study consists of three separate experi-ments. Young (10–12 weeks old) adult male Noble ratswere used. They were implanted under anesthesia(Rompun (20mg/ml); Bayer HealthCare AG, Leverku-sen, Germany and Ketaminol (50 mg/ml); Pfizer Oy,Esbo, Finland) with the estradiol and testosteronereleasing implants (21-day releasing implants, Inno-vative Research of America, Inc. Sarasota, Florida).Daily released amounts were 70 mg for estradiol and

240 mg for testosterone. All animals were housed (tworats per cage) under a 12 hr light–12 hr dark lightingcycle. They had free access to soy-free feeding pellets(SDS, Witham, Essex, England) and tap water. TheAnimal Care Organization and University EthicalCommittee of Turku University approved the studyprotocol. The rats were handled in accordance with theinstitutional animal care policies of the University ofTurku.

In the first experiment, the animals were implantedwith estradiol and testosterone implants for 3 (n¼ 10)and 6 weeks (n¼ 10). Thereafter, the urodynamicalrecordings and histopathological analysis were done.In the 6-week study, the hormone implants weresurgically removed and replaced with the new iden-tical implants after 3 weeks. Age-matched placebo(n¼ 10) animals had implants containing a biologicallyinert substance.

In the second experiment, two different doses ofestrogen (5 and 70 mg/day) (IRA implants, 0.1 mg/21-day and 1.5 mg/21-day) and the effect of an anti-estrogen (ICI 182,780; Tocris Cookson, cat number1047) were studied. The dose of testosterone was240 mg/day like in the first experiment. In another partof the second experiment, ICI 182,780 was dissolved inabsolute ethanol and diluted in rapeseed oil to a finalconcentration of 3.3 mg/ml [9]. The final volume ofethanol didnot exceed1%.All rats (n¼ 16)were treatedfor 3weekswith estradiol (70 mg/day) and testosterone(240 mg/day). In the beginning of the third week,animals were divided into two groups. Half of theanimals received ICI 182,780 s.c. in the flank twice perweek at a dose 3 mg/kg. The injections were given onDays 14 and 17 after implantation. Another half servedas controls and received injections of absolute ethanoldissolved in rapeseed oil without ICI 182,780.

In the third experiment, the reversibility of lympho-cyte infiltration was studied by using estradiol (70 mg/day) and testosterone (240 mg/day) implants preparedfor treatment of 3 weeks. The implants were removedafter the corresponding release time, and animals weremaintained for three more weeks without hormonesupplementation.

Histopathological Examination andSerumHormoneMeasurements

At the end of the treatment, animals were weighedand anesthetized with carbon dioxide. Bloodsamples were collected by heart puncture before neckdislocation. The testes and epididymis, the lowerurinary tract (bladder and proximal urethra), accessorysex glands (seminal vesicles and dorsolateral lobes ofthe prostate), and pituitary gland were rapidly excised

The Prostate DOI 10.1002/pros

Estrogen-Related Prostatitis 889

and weighed. Thereafter, organ samples were fixed in10% formalin overnight, and routinely processed priorto embedding in paraffin to prepare histologicalsections for subsequent stainings. Paraffin embeddedsections were cut at 5 mm thickness and stained withhematoxylin and eosin (H&E).

Serum samples were stored at �708C until themeasurements of hormone concentrations. Beforeestradiol and testosterone measurements, sampleswere extracted twice with diethylether, and theremaining organic solution was evaporated undernitrous gas in warm water bath. Hormones wereredissolved in zero-standard serum, and concentratedwhen necessary. Thereafter, enzyme immunoassay forthe in-vitro-diagnostic quantitative determination of17b-estradiol and testosterone were conducted accord-ing to instructions given by the manufacturer (17b-estradiol ELISA for human serumandplasmaRE52041,testosterone ELISA for human serum and plasmaRE52151, IBL-Hamburg, Germany). Serum prolactinwasmeasuredwith rat prolactin enzyme immunoassaykit without pre-treatment (Rat prolactin EIAkit, A05101, Spi-Bio, BERTIN group, Montigny LeBretonneux, France). Detection range for estradiol was25–2,000 pg/ml, for testosterone 0.2–16 ng/ml, and forprolactin 0.39–50 ng/ml. All sampleswere analyzed induplicates.

A standardized histopathological classification andcounting system of inflammation infiltrates proposedby Nickel et al. [14] was used with modifications.Infiltrate was considered to be perivascular whenmorethan ten inflammatory cells were found around thecapillary, stromal when inflammatory cells werescattered in the stroma or periglandular space, andglandular when inflammatory cells were found in theepithelium or lumen (Fig. 1A–C). Inflammatory infil-trates in each category were counted from the serialsections made from dorsolateral lobes of the prostate.Six sections from each animal were screened bytwo independent observers who worked blindly.This quantification method gave an estimate of infla-mmatory changes in different lobes and at differentlevels of each lobe. The computer-based quantitativemethod was used to study inflammatory responses toanti-estrogen treatment. In addition to the total numberof inflammation, the mean size of the infiltrates wasassessed.

Immunohistochemical Staining (CD3þ,CD8þ)

Paraffin-embedded tissue sections for immunohis-tochemical staining were cut and placed on the super-frost glasses. The sectionswere stored atþ 48Cuntil thestaining was performed. Sections were deparaffinizedby xylene, alcoholic solutions (100, 96, and 70%), anddistilled water. Antigen retrieval was performed byboiling sections for 15 min in citric acid buffer (10 mM,pH 6.0). Endogenous peroxidase activity was blockedwith 1%H2O2 solution for 20min at room temperature.Sections were then incubated overnight at þ48C withthe primary antibody diluted in 0.05% tween. Anti-bodies for CD3 (1:100) andCD8 (1:100)were purchasedfromCaltag Laboratories (mousemonoclonal antibodyto rat CD3MR5300 andmousemonoclonal antibody torat CD8 MR5200S, Caltag Laboratories, Burlingame,Canada). For immunoreaction, peroxidase-conjugatedanti-mouse secondary antibody was used and immu-noreactivity visualized with the DAB chromagen(Envision System-HRP (DAB) with mouse primaryantibodies K4007, Dako Cytomation/Oy ALGOL Ab,Glostrup, Denmark).

UrodynamicalMeasurements

The methodological details of the urodynamicalmeasurements have been described previously [10].Animals were anesthetized using an i.p. injection ofchloral hydrate (0.9 g/kg, Sigma Chemical Co. St.Louis, MO), and i.v. injections of urethane (0.32 g/kg,Sigma Chemical Co.) were used to maintain anesthesiafor urodynamical measurements. The bladder anddistal urethra were exposed. A 20G i.v. cannula wasinserted through the bladder apex into the lumen forsaline (0.9% NaCl) infusion and bladder pressuremeasurements. Micturition was evoked by salineinfusion (10 ml/hr). Transvesical cystometry wasrecorded by using Statham pressure transducers(P23BC, Hato Ray, PR). The urine flow rate wasmeasured by using a flow meter (model T206) with anultrasonic flow probe (2,5SB178; Transonic Systems,Inc. Ithaca, NY) at the distal part of urethra.

Continuous recordings were obtained by using AcqKnowledge 3.5.3 program (Biopac Systems Inc., SantaBarbara, CA). When the micturitions had illustratedseveral similar and consecutive features for the rodent,three representative voidings were chosen for further


Fig. 1. Histological classification of theprostatic inflammation in the dorsolateral lobes.Classification isbasedon the anatomical locationoftheinflammatoryinfiltrate.A: Short-termestradiol- and testosterone-treatmentinducedlymphocyteinfiltration aroundthe capillaries, clas-sifiedasperivascularinflammation(H&E,250�).B:Morepronouncedinflammatoryinfiltration,classifiedas stromalinflammation,was seeninthe prostatic stroma and periglandular space after longer estradiol and testosterone treatment (H&E, 250�).C: Third category, glandularinflammation, consistedofmainlyneutrophiles, butalso lymphocytesinside theacini (H&E,250�).D: Immunohistochemical stainingrevealedthatperivascularinflammationfociconsistedofmainlyT-lymphocytes(CD3þcells)(250�).EandF:Specific subsetofT-cells,cytotoxic(CD8þ)cellswereobservedbothin stromalinfiltrates aswellasintraepithelially (250�).

890 Bernoulli et al.


Fig. 1.


analysis. According to Chien et al. [11], the bladderpressure wave during micturition of the rat may bedivided into four phases. During the first phase,intraluminal pressure of the bladder increases rapidlybut intraluminal pressure oscillations (IPHFOs) andurine flow are absent. In the second phase, flow peaksappear and are synchronized with the IPHFOs. Inthe third phase, the flow stops and bladder pressuredeclines. During the fourth phase bladder pressuredeclines to the basal bladder pressure and filling ofthe bladder by saline starts again (Fig. 2). Maximalbladder pressure in the first and second phases wasanalyzed. In addition to themaximal bladder pressure,the urine flow rate was measured from the highestpeak. Micturition time (the period from the first flowpeak until the flow had ceased), micturition intervals,bladder capacity (voided volume plus infusedvolume), and residual urine (voided volume minusinfused volume) were also measured. In addition,extracellular electrical activities of the proximal rhab-dosphincter were recorded by using a flexible suctionelectrode.

Statistical Analysis

The normal distribution of the data was checkedwith Shapiro–Wilk’s W normality test. In normally

distributed data, One-Way ANOVA was used tocompare the statistical differences between the groups.If the data were not normally distributed, Mann–Whitney U-test or Kruskal-Wallis test was used(Statistica, version 5.1, Stat Soft, Inc. Tulsa, OK andGraphPad Prism version 4.00 for Windows, GraphPadSoftware, San Diego, CA).

RESULTS

Gradual Developmentof the Prostatic Inflammation

Development of inflammatory infiltrates wasstudied in intact adult Noble rats treated with estrogenand testosterone for 3 and 6 weeks. Increased numberof inflammatory cells was seen in the dorsolateral lobesof the prostate at 3 weeks after hormone implantation.The inflammatory cells, mainly lymphocytes, were firstlocatedaround the capillaries.When the treatment timewas extended to 6 weeks, inflammatory cells were alsoseen in the stroma and the periglandular space andfinally in the epithelium and lumen of acini (Fig. 1A–C). Perivascular and stromal inflammation consistedmainly of lymphocytes and mast cells; glandularinflammation consisted of lymphocytes and polymor-phonuclear leucocytes, mainly of neutrophiles.The majority of the lymphocytes in the stroma were


Fig. 2. Simultaneousrecordingofbladderpressure (A),EMGof theproximalrhabdosphincter (B), andurine flowrate (C) inonemicturitioncycle of the placebo-treatedmale rat.During the first phase (Ph1), tonic bladder pressure increases.The second phase (Ph 2) is associatedbladder pressure oscillations, increase in amplitude of rhabdosphincter EMG, and urine flow.Before the third phase (Ph 3) urine flow ceasesand a temporary increase inbladder pressure is seen.During the fourthphase (Ph 4) thebladder pressure declines to the same or lower levelseenbeforemicturitioncontraction.After treatment for 6weekswith estradiol and testosterone, phase1maximalpressurewas significantlydecreased, but neither changes were seen in bladder pressure during phase 2 nor in the EMG of the proximal rhabdosphincter. A trend ofdecreasedflowratewas seeninhormone-treatedanimals.


T-lymphocytes (CD3þ). Cytotoxic T-cells (CD8þ) werefound among stromal inflammatory cells and intrae-pithelially (Fig. 1D–E). Glandular forms were seenpredominantly in the lateral lobe at 6 weeks. Therewas a clear shift from the perivascular foci to moreadvanced forms, stromal and glandular foci, between3 and 6 weeks of the hormone treatment (Fig. 3). Nosigns of inflammation were seen in the periurethraland periductal areas. In addition, other organs werestudied to rule out possibility of inflammation. Yet, noinflammationwas found in other organs, e.g., kidney orbladder.

OrganWeights andHormoneConcentrations in Serum

Hormone treatments reduced body weight gain.Because of this, relative weights of the organs werecalculated (Table I). Serum hormone concentrationsconfirmed that estradiol concentration was increasedin serum at 3 weeks (Fig. 4A) while testosteroneconcentration remained at the control level. When thetreatment time was prolonged up to 6 weeks, testoster-one concentration increased slightly (Fig. 4B). Theincrease was statistically significant, and was reflectedin the increased relative weight of the seminal vesicles.Relative pituitary weights and serum prolactinconcentrations were significantly increased at 3 and6 weeks (Fig. 4C), likely due to estrogenic effect on thepituitary. The relative testis weight was significantlydecreased, possibly indicating diminishedproductionsof LH and FSH.


Fig. 3. Quantitative analysis of different lymphocyte infiltratesafter the 3- and 6-week treatmentwith estradiol and testosterone(Tþ E2). In the placebo animals, some focal perivascular inflamma-tionwas seen.However, 3-weekhormone treatmentinducedmoreperivascular infiltrates, andalso some stromalandglandular inflam-mation.Longerhormoneexposureappearedtomaintainperivascu-lar inflammation, butcausing further inflammation, seen as stromalandglandularinflammation.Valuesaremeans,nindicatesthenumberof theanimalsineachgroup.

TABLE

I.RelativeOrg

anWeigh

ts(g/g)of

Placebo-a

ndHor

mon

e-Treated

(TþE2)Animals

Placebo(n

¼10

)3W

eekTþE2

(n¼10

)6W

eekTþE2

(n¼10

)P-value

(placebovs.3w)

P-value

(placebovs.6w)

P-value

(3w

vs.6w)

Bladder

(n¼5þ

5)3.9�10

�4�0.5�10

�4

Novalues

5.8�10

�4�0.4�10

�4

Novalues

0.00

9(M

)Novalues

Sem

inal

vesicles

4.8�10

�4�0.7�10

�4

6.4�10

�4�0.8�10

�4

8.9�10

�4�1.8�10

�4

0.00

09(M

)0.00

03(M

)0.00

4(M

)Pituitary

3.1�10

�5�0.3�10

�5

4.8�10

�5�0.8�10

�5

7.4�10

�5�2.4�10

�5

0.00

02(M

)0.00

02(M

)0.01

02(M

)Testis

48.2�10

�4�0.7�10

�4

45.6�10

�4�0.5�10

�4

32.1�10

�4�0.5�10

�4

NS

0.00

016(M

)0.00

01(M

)

Values

aremed

ians�SD,statistical

analysissh

ownafterP-value(M

¼Man

n–W

hitney

U-test).N

S,n

on-significant.


UrodynamicsAfter theTreatment for 3 and 6Weeks

Representative recording of placebo-treated animalvoiding is shown in figure 2. No statistically significantdifferences were seen in any of the urodynamic para-meters recorded between the placebo and hormone-treated animals at 3 weeks. At 6 weeks, the bladdercapacity and the relative bladder weight were sig-nificantly increased in hormone-treated rats, whichwas reflected in the less frequent micturition andprolongedmicturition time (Tables I and II). Hormone-treated rats showed a trend of an increased volume of

residual urine. Maximal bladder pressure duringphase 1 was significantly decreased in hormone-treatedanimals. However, neither difference was seen in themaximal bladder pressure during voiding (phase 2),nor in the rhabdosphincter EMG (data not shown).The maximal urine flow rate of the hormone-treatedrats was decreased (nearly 10 ml/min less than incontrol rats), but the difference was not statisticallysignificant due to the high inter-individual variations.These findings suggest an altered function of thebladder at 6 weeks (Table II), reflecting an incipientobstruction.


Fig. 4. Serumhormone concentrations in theplacebo- andhormone-treatedanimals (Tþ E2).A: Serumestradiolconcentration increasedsignificantly already in 3weeks andwasmaintained athigh levelwithhormone treatment for 6weeks compared toplacebo animal.B: A slightdecreaseinserumtestosteroneconcentrationat3weeks,andanincreaseat6weeks.C:Serumprolactinincreasedsignificantlyafter treatmentfor 6weeks. Significant P-valuesbetween the groups are shown in the figures.Values aremeans� SD, statistical analysismethods denotedbycorresponding letters after P-value (A¼ ANOVA, andTukeyMultipleComparison test for normallydistributeddata; andK¼Kruskal-Wallistest andDunn’sMultipleComparison testfornonparametricdata),nindicates thenumberof theanimalsineachgroup.


TheDose-Dependencyof EstrogenActionandtheAnti-InflammatoryAction of an

Anti-Estrogen (ICI182,780)

The effects of two different daily doses of estradiolwere compared (5 and 70 mg), while using the samedose of testosterone (240 mg/day). The treatment timewas 3 weeks. Mean estradiol concentration in serumwas 29 pg/ml in the higher estradiol dose group, whilethe concentration in the lower estradiol dose groupremained under detection limit (estimated value2.5 pg/ml). The pituitary weight showed a dose-dependent increase as expected. The number ofinflammation infiltrates was dependent on the dose ofestradiol i.e., it was smaller in the low estradiol group(Fig. 5).

ICI 182,780 treatment of the estradiol- and testoster-one-treated rats attenuated significantly estradiol-induced inflammation in 3 weeks. The total numberof inflammation infiltrates relative to prostate areawas decreased. The mean size of the inflammationinfiltrates remained unaltered. No major changes wereseen in serum hormone concentrations or in the organweights between the groups (data not shown).

Reversibilityof Inflammation

Reversibility of the estrogen-induced inflammationwas studied by removing estradiol and testosteroneimplants after the treatment for 3 weeks. The animalswere analyzed following 3-week period after theimplant removal.Histology of prostates showed fadingof the inflammation. There were no glandular inflam-mation left and also the number of perivascularand stromal inflammation infiltrate was diminished(Fig. 6).

DISCUSSION

Exposure of Noble rats to excessive amounts ofestradiol induced gradually developing, reversibleglandular prostatic inflammation in the dorsolateralprostate in 6 weeks. Increased bladder weight andcapacity, and alterations in bladder function, reflectingincipient obstructive voiding were recorded concomi-tantly with the glandular inflammation. Perivascularand stromal inflammation preceded the first changes inthe lower urinary tract function. These findings are notsupportive of the idea that obstructive voiding causeslymphocyte infiltration in our current prostatitismodel.

Cellular composition and inflammatory pattern(perivascular, stromal, and periglandular)were similarto those described in earlier animal studies [12,13]


TABLE II. Urodynamic Parameters of Placebo- andHormone-Treated (TþE2) Animals

Placebo (n¼ 9) 6 Week TþE2 (n¼ 10) P-value

Max BP phase 1 31.1� 4.57 21.8� 2.84 0.00005 (A)Max BP phase 2 55.4� 7.51 57.3� 8.90 NSBC 1.7� 0.38 2.3� 0.47 0.01 (A)Max FR 30.4� 18.10 20.8� 10.72 NSMT 14.1� 7.15 16.5� 9.80 NSM Int 1.8� 0.33 2.3� 0.51 0.03 (A)RU 1.0� 0.39 1.4� 0.49 (0.06) (M)

Values are means� SD, statistical analysis shown as P-values (A¼ANOVA and M¼Mann–Whitney U-test). Abbreviations used in the table: Max BP, maximal bladder pressure (cmH2O);BC, bladder capacity (voided volume plus infused volume) (ml); Max FR, flow rate (ml/min);MT, micturition time (sec); M Int, micturition interval (min); RU, residual urine (ml); NS, non-significant.

Fig. 5. The dose-dependency of estrogen action. Both groupsreceived same dose of testosterone butdifferentdosages of estra-diol (groups respectivelyTþ E2 5 mg and Tþ E2 70 mg).The meannumberof lymphocyteinfiltratesper animalwasusedas a quantita-tive parameter of estrogen action. A clear quantitative differencewas seen between the low-dose and high-dose estradiol groups.The lower dose (5 mg/day) induced few perivascular infiltrates.Thehigher dose (70 mg/day) induced more infiltrates in each category.Values are means, n indicates the number of the animals in eachgroup.


and in human prostatitis [14]. More detailed immuno-histochemical study of the samples showed thatperivascular and stromal inflammatory infiltrates con-sisted of T (CD3þ) lymphocytes. When the treatmentperiod was extended up to 6 weeks, inflammationproceeded to glandular form and cytotoxic T-cells(CD8þ) were observed in the stroma but also intrae-pithelially as in the more severe forms of humanprostatitis [15]. Neutrophiles were found in theglandular lumina as a sign of acute inflammation.Perivascular and stromal inflammatory infiltrates con-sisted exclusively of lymphocytes showing chronicsigns of inflammation. Prostatitis patients often showsimilar kind of inflammatory pattern with both acuteand chronic forms [14].

Increased concentration of estrogen in serum hasbeen considered to be an important etiologic factorfor the experimental nonbacterial prostatitis in rats[12,16,17]. However, it is also known that castration oreven decreased androgen concentration may causeacute and chronic inflammation, which is reducedwith testosterone treatment [18]. Therefore, serumtestosterone level was maintained at or above thecontrol level in the present study to confirm the role ofexcessive estrogen. Excessive amount of estradiol wasnot only needed for the induction of lymphocyteinfiltration but also for the maintenance of theinflammation. When both testosterone and estradiolimplants were removed after 3-week treatment, sub-sequent reduction in inflammation was observed,providing evidence for the necessity of continuous

excessive estrogen exposure for the lymphocyteinfiltration. Additionally, lymphocyte infiltration mea-sured as the number of inflammatory infiltrates wasdependent on the amount of estradiol released dailyfrom the implants. There was a clear differencebetween the two daily doses (5 and 70 mg/day) testedin the current study. Moreover, the multiplicity of thefoci was reduced by anti-estrogen treatment, implyingan estrogen receptor (ER)-mediated process. Theinformation supporting the role of sex hormones inthe development of nonbacterial prostatitis in human isscanty [19]. The most suggestive evidence comes fromthe clinical trial with mepartricin which decreasesboth estrogen concentration in serum and the totalNIH-CPSI score [20].

It is not known how excessive estrogen exposurepromotes prostatic inflammation. In agreement withpresent findings, gene expression study using Wistarrats showed that even short exposure to estrogenupregulated proinflammatory transcripts, well beforeany inflammatory cells were observed in the prostate[13]. Estradiol induced prolactin release from pituitarygland and increased the weight of the pituitary gland.An increased prolactin secretion is in agreement withearlier findings suggesting that prolactin mediatesat least partly the estrogen-induced inflammatoryaction in the prostate [21]. In the transgenic mousewith prostate-specific expression of prolactin, chronicinflammation (primarily lymphocytes and macro-phages) was frequently observed in the prostate butalso an increased distribution of ERa in stromal cellswas evident [22]. Therefore, it is difficult to get anunequivocal evidence to support the idea that eitherprolactin or estrogen alone would cause inflammationsince blocking of prolactin release by bromocriptineonly partially prevented estradiol-induced inflamma-tion [23]. In addition to pituitary release, prolactin canbe produced in prostate itself [24], and it may actnot only as a hormone but also as a cytokine [25],which further complicates the interpretation of thefindings.

Accumulation of lymphocytes in perivascular spacewas the earliest sign of estrogen action. This suggeststhat an estrogen-induced increase in vascular perme-ability has an important role in the inflammatoryresponse as in other autoimmune diseases. As anevidence for the autoimmune nature, the estradiol-induced prostatic inflammation could be transferred tonaive syngenic recipient by adoptive transfer of CD3þT-cells [26]. Only T-cells from the estradiol-treatedanimals demonstrated prostate-specific inflammation.Besides thedirect effects of estrogens on the endothelialcells, estrogens may also have influence on vascularsmooth muscle cells. They contain estrogen receptorsand therefore are potential targets for estrogen action


Fig. 6. Reversibility of the inflammation. In the animals treatedfor 6 weeks with estradiol and testosterone (Tþ E2 6 week),three different forms of inflammationwere observed.However, inanimals treated first for 3 weeks with estradiol and testosteroneand following 3 weeks without hormone supplementation (Tþ E23þ 0¼ 6 week), attenuation of the inflammation was seen.Valuesaremeans,nindicates thenumberof theanimalsineachgroup.


[27]. In the stromal smooth muscle cells or epithelialcells, estrogens may further increase synthesis ofchemokines e.g., MIP-2 which in turn attracts inflam-matory cells toward acini [13]. It has also beensuggested that estradiol treatment permits the pre-sentation of previously sequestered antigen(s) [28].

The association of the lower urinary tract dysfunc-tions with nonbacterial prostatitis has not been studiedin animal models until recently. The role of increasedurethral pressure has been tested in rats with partialurethral obstruction caused by nylon ligature onurethra [29]. A nonbacterial prostatitis was developedin the ventral lobes. The authors did not give a detaileddescription of their ligation technique. It is difficult toavoid closing of the collecting ducts or accompanyingblood vessels, of the ventral lobes by the ligatures. Theducts lie on the anterior wall of urethra before openinginto urethral lumen. According to our unpublishedfindings, closing of theducts or arteries has been shownto cause prostatitis. Our opinion is that the ligaturemodel is not relevant for the discussion of the possibleconsequences of prostatic inflammation on voiding,because of obvious changes in bladder function by theligature narrowing the urethral lumen. Nakano [30]injected hydrochloric acid beneath the prostatic cap-sule of the rat inducing tissue damage and conse-quently a nonbacterial prostatitis and lower urinarytract disorders within 4 to 8 days. The urinationintervals and voiding volumes were decreased andresidual volumes and bladder weights were increased.The bladder pressures and urine flow rates were notrecorded but the changes could be early signs ofobstructive voiding. There was no evidence that tissuedamage extended from the prostate via urethra tobladder. It must be that other mechanisms have to beinvolved. In agreement with the findings of Nakano[30], nonbacterial prostatitis induced with hormonaltreatments was associated with alterations of the lowerurinary tract in the current study. Both inflammatoryinfiltrates and the voiding dysfunction developedsomewhat slower than in the model of Nakano [30],without any tissue damage. The increases in theweightand volume of the bladder as well as prolongedvoiding times are signs of the changes in the bladderfunction and indicate incipient obstructive voiding. It isintriguing that regardless of the trigger (local hydro-chloric acid injection or hormone treatment), the timingof urodynamic changes in relation to prostatic inflam-mation and the nature of the urodynamic changesappear to be similar.

In the current study, estrogen-induced prostatitiswas associated with some alterations of LUT functionsat 6weeks,while voiding at 3weekswas unaltered. Theincreases in the weight and capacity of the bladder aresigns of incipient and obstructive voiding which may

manifest more clearly after longer treatment period.Interestingly, the first phase of voiding showedsignificant decrease in the bladder pressure. Thissuggests that estrogens may have effects on purino-ceptor function because the first phase of voiding isbased on purinoceptor function [31].

There are several possible explanations for thedevelopment of urodynamic alterations in associationwith the estrogen-induced prostatic inflammation.Inflammatory cells secrete cytokines that can exerttheir effects locally or systemically. They are found influid secreted by the prostate in men [32]. Cytokinessuch as IL-6 were most prevalent in men with prostaticinflammation [33]. The amounts paralleled in leukocytecounts. In the current study, perivascular and stromalinfiltratesweredistributed randomlyall over the lateralanddorsal lobeswhile the inflamedglandsweremostlylocated in the lateral lobes. The urodynamic changesdeveloped concomitantlywith glandular inflammationshowing intraepithelially located lymphocytes andluminal neutrophiles. It can be speculated that cyto-kines present inside the acini and ducts may influencethe musculature of the lower urinary tract and causeurodynamic changes. Another possibility is that uro-dynamic changes are caused by non-inflammatoryhormonal effects of estrogens on the LUT.

The mechanisms and the sites of estrogen actions inthe male LUT have not been established [34]. There areonly a few ERa-positive cells in the male LUT [35].However, ERa-positive cells have been found inautonomic and spinal ganglia and various structuresof the CNS innervating the LUT [36,37]. ERb is thepredominant form of ER in the male LUT and prostaticganglia [37]. Even though the role of ERb is stillunknown its presence in several LUT structures couldmake direct effects of estrogens possible. The effects ofcastration on the pelvic autonomic ganglia have beendescribed [38] but in the present study the possibleeffects of androgen deficiency were eliminated.Furthermore, experimental evidence shows that irrita-tion of prostate activates afferent lumbosacral neuronsinnervating pelvic floor. This mechanism couldaccount for muscular dysfunctions of the LUT asso-ciated with chronic prostatitis. Other experimentaldesigns such as the use of anti-inflammatory drugs orof immuno-deficient animals are needed to illuminatethe mechanisms.

CONCLUSIONS

Ourdata suggest the following sequence of events inthe development of prostatic inflammation and asso-ciated urodynamic alterations. Exposure to excessiveestrogen in the presence of physiological amounts ofandrogens induces prostatic inflammation in the



dorsolateral lobes of the Noble rats. Inflammation isestrogen dependent and it can be prevented by a non-estrogenic anti-estrogen. Inflammation evolves fromthe perivascular accumulation of T-cells to the stromaldistribution, and subsequently, proceeds into glandu-lar form when alterations of voiding were recordedfor the first time. Voiding alterations may representseparate non-prostatic estrogenic effects. A possibilityremains that either cytokines secreted by inflammatorycells systematically are involved when released in theglandular inflammation into the secretion or neuro-genicmechanismswould influencemuscular functionsof the LUT and cause urodynamic changes.

ACKNOWLEDGMENTS

Our thanks go to Professor Sirpa Jalkanen for thediscussions of immunology, and to Mrs. NataliaHabilainen-Kirillov and Ms. Tuula Tanner for theirexcellent technical assistance in preparation of thehistology sections. Natalia has been instrumental withanimal experiments.Mr. RikuAnttila is appreciated forhis thorough work with quantitative analysis of ICI182,780 sections.

REFERENCES

1. Nickel JC. The overlapping lower urinary tract symptoms ofbenign prostatic hyperplasia and prostatitis. Curr Opin Urol2006;16(1):5–10.

2. Gonzalez RR, Te AE. Is there a role for urodynamics in chronicnonbacterial prostatitis? Curr Urol Rep 2006;7:335–338.

3. Mehik A, Leskinen MJ, Hellstrom P. Mechanisms of pain inchronic pelvic pain syndrome: Influence of prostatic inflamma-tion. World J Urol 2003;21(2):90–94.

4. Lee JC, Yang CC, Kromm BG, Berger RE. Neurophysiologictesting in chronic pelvic pain syndrome: A pilot study. Urology2001;58(2):246–250.

5. Miller LJ, Fischer KA, Goralnick SJ, Litt M, Burleson JA,Albertsen P, Kreutzer DL. Nerve growth factor and chronicprostatitis/chronic pelvic pain syndrome. Urology 2002;59(4):603–608.

6. Ishigooka M, Zermann DH, Doggweiler R, Schmidt RA.Similarity of distributions of spinal c-Fos and plasma extravasa-tion after acute chemical irritation of the bladder and theprostate. J Urol 2000;164(5):1751–1756.

7. Chen Y, Song B, Jin XY, Xiong EQ, Zhang JH. Possiblemechanism of referred pain in the perineum and pelvisassociatedwith theprostate in rats. JUrol 2005;174(6):2405–2408.

8. Noble RL. The development of prostatic adenocarcinoma in Nbrats following prolonged sex hormone administration. CancerRes 1977;37(6):1929–1933.

9. Thompson CJ, Tam NN, Joyce JM, Leav I, Ho SM. Geneexpression profiling of testosterone and estradiol-17 beta-induced prostatic dysplasia in Noble rats and response to theantiestrogen ICI 182,780. Endocrinology 2002;143(6):2093–2105.

10. Streng T, Santti R, Talo A, Similarities and differences in femaleand male rat voiding. Neurourol Urodyn 2002;21(2):136–141.

11. Chien CT, Yu HJ, Lin TB, Chen CF. Neural mechanisms ofimpaired micturition reflex in rats with acute partial bladderoutlet obstruction. Neuroscience 2000;96(1):221–230.

12. Robinette CL. Sex-hormone-induced inflammation and fibro-muscular proliferation in the rat lateral prostate. Prostate 1988;12(3):271–286.

13. Harris MT, Feldberg RS, Lau KM, Lazarus NH, Cochrane DE.Expression of proinflammatory genes during estrogen-inducedinflammation of the rat prostate. Prostate 2000;44(1):19–25.

14. Nickel JC, True LD, Krieger JN, Berger RE, Boag AH, Young ID.Consensus development of a histopathological classificationsystem for chronic prostatic inflammation. BJU Int 2001;87(9):797–805.

15. Nickel JC. Textbook of Prostatitis. Oxford: Isis Medical Media;1999. 134 p.

16. Naslund MJ, Strandberg JD, Coffey DS. The role of androgensand estrogens in the pathogenesis of experimental nonbacterialprostatitis. J Urol 1988;140(5):1049–1053.

17. Keith IM, Jin J, Neal D Jr, Teunissen BD, Moon TD. Cellrelationship in a Wistar rat model of spontaneous prostatitis.J Urol 2001;166(1):323–328.

18. Lundgren R, Holmquist B, Hesselvik M, Muntzing J. Treatmentof prostatitis in the rat. Prostate 1984;5(3):277–284.

19. Pontari MA, Ruggieri MR. Mechanisms in prostatitis/chronicpelvic pain syndrome. J Urol 2004;172(3):839–845.

20. De Rose AF, Gallo F, Giglio M, Carmignani G. Role ofmepartricin in category III chronic nonbacterial prostatitis/chronic pelvic pain syndrome: A randomized prospectiveplacebo-controlled trial. Urology 2004;63(1):13–16.

21. Tangbanluekal L, Robinette CL. Prolactin mediates estradiol-induced inflammation in the lateral prostate of Wistar rats.Endocrinology 1993;132(6):2407–2416.

22. Kindblom J, Dillner K, Sahlin L, Robertson F, Ormandy C,Tornell J,WennboH. Prostate hyperplasia in a transgenicmousewith prostate-specific expression of prolactin. Endocrinology2003;144(6):2269–2278.

23. Lane KE, Ricci MJ, Ho SM. Effect of combined testosterone andestradiol-17 beta treatment on the metabolism of E2 in theprostate and liver of noble rats. Prostate 1997;30(4):256–262.

24. NevalainenMT, Valve EM, Ingleton PM,NurmiM,MartikainenPM, Harkonen PL, Prolactin and prolactin receptors areexpressed and functioning in human prostate. J Clin Invest1997;99(4):618–627.

25. Ben-JonathanN, LibyK,McFarlandM,ZingerM. Prolactin as anautocrine/paracrine growth factor in human cancer. TrendsEndocrinol Metab 2002;13(6):245–250.

26. Keetch DW, Humphrey P, Ratliff TL. Development of a mousemodel for nonbacterial prostatitis. J Urol 1994;152(1):247–250.

27. Koh KK. Effects of estrogen on the vascular wall: Vasomotorfunction and inflammation. Cardiovasc Res 2002;55(4):714–726.

28. SeethalakshmiL, BalaRS,MalhotraRK,Austin-RitchieT,Miller-GrazianoC,MenonM, Luber-Narod J. 17 beta-estradiol inducedprostatitis in the rat is an autoimmunedisease. JUrol 1996;156(5):1838–1842.

29. Takechi S, YokoyamaM, Tanji N, Nishio S, Araki N. Nonbacter-ial prostatitis caused by partial urethral obstruction in the rat.Urol Res 1999;27(5):346–350.



30. Nakano K, Kirimoto T, Hayashi Y, Oka T. Nonbacterialprostatitis model animal. 2006 United States Patent ApplicationPub. No. US 2006/0107336 A1.

31. Streng T, Talo A, Andersson KE. Transmitters contributing tothe voiding contraction in female rats. BJU Int 2004;94(6):910–914.

32. JangTL, SchaefferAJ. The role of cytokines in prostatitis.World JUrol 2003;21(2):95–99.

33. Drachenberg DE, Elgamal AA, Rowbotham R, Peterson M,Murphy GP. Circulating levels of interleukin-6 in patients withhormone refractory prostate cancer. Prostate 1999;41(2):127–133.

34. Streng TK, Talo A, Andersson KE, Santti R. A dose-dependentdual effect of oestrogen on voiding in the male mouse? BJU Int2005;96(7):1126–1130.

35. Makela S, Strauss L, Kuiper G, Valve E, Salmi S, Santti R,Gustafsson JA. Differential expression of estrogen receptorsalpha and beta in adult rat accessory sex glands andlower urinary tract.Mol Cell Endocrinol 2000;164(1–2):109–116.

36. VanderHorst VG,Meijer E, Holstege G. Estrogen receptor-alphaimmunoreactivity in parasympathetic preganglionic neuronsinnervating the bladder in the adult ovariectomized cat.Neurosci Lett 2001;298(3):147–150.

37. Salmi S, Santti R, Gustafsson JA, Makela S. Co-localizationof androgen receptor with estrogen receptor beta in thelower urinary tract of the male rat. J Urol 2001;166(2):674–677.

38. Keast JR, Saunders RJ. Testosterone has potent, selective effectson the morphology of pelvic autonomic neurons which controlthe bladder, lower bowel and internal reproductive organs of themale rat. Neuroscience 1998;85(2):543–556.



Urodynamic changes in a noble rat model for nonbacterial prostatic inflammation

Documents

Transcript of Urodynamic changes in a noble rat model for nonbacterial prostatic inflammation