Life Sciences 91 (2012) 1281–1287

Contents lists available at SciVerse ScienceDirect

Life Sciences

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

Fibrosis-related gene expression in the prostate is modulated by doxazosin treatment

Flávia K. Delella a,b, Livia M. Lacorte b, Fernanda Losi A. Almeida c, Maeli Dal Pai b, Sérgio L. Felisbino b,⁎a Department of Structural and Functional Biology, Institute of Biology—University of Campinas (UNICAMP), Campinas, Sao Paulo, Brazilb Department of Morphology, Institute of Biosciences—Univ Estadual Paulista (UNESP), Botucatu, Sao Paulo, Brazilc Department of Morphological Sciences, State University of Maringa (UEM), Maringa, Parana, Brazil

⁎ Corresponding author at: Departamento deMorfologiEstadual Paulista (UNESP), 18618-970, Botucatu, SP, Brasi

E-mail address: felisbin@ibb.unesp.br (S.L. Felisbino)

0024-3205 © 2012 Elsevier Inc.http://dx.doi.org/10.1016/j.lfs.2012.09.017

Open access under the El

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 9 March 2012Accepted 26 September 2012

Keywords:DoxazosinBenign prostatic hyperplasiaCollagenTGF beta-1Alpha-adrenergic blockadeNeurotransmitter vesicles

Aims: To gain new insights into the molecular mechanisms of action of doxazosin, we investigated the pros-tatic stroma ultrastructure and the expression of genes involved with fibrosis, such as collagen type I and III(COL1A1 and COL3A1, respectively) and TGF-beta 1, in the rat ventral prostate.Main methods: Adult Wistar rats were treated with doxazosin (25 mg/kg/day), and the ventral prostateswere excised at 7 and 30 days after treatment. Untreated rats were controls. Ventral prostates were subjectedto ultrastructural, immunohistochemical, biochemical and molecular analyses.Key findings: Doxazosin-treated prostates showed thickened bundles of collagen fibrils, activated fibroblasts,enlarged neurotransmitter vesicles and increased tissue immunostaining for collagen type I and type III whencompared to untreated prostates. After 7 and 30 days of doxazosin treatment mRNA expression of COL1A1and COL3A1 was significantly increased and reduced, respectively, compared to the control group.

TGF-beta 1 mRNA and protein levels were increased after 7 days of doxazosin treatment, whereas onlymRNA levels remained increased after 30 days of treatment.Significance: Our data suggest that relaxation of smooth muscle cells by alpha-blockers interferes with themechanical dynamics of the prostatic stroma extracellular matrix components, generating a pro-fibroticeffect probably via the TGF-beta 1 signaling pathway. Long term treatment with doxazosin may also leadto a reduced turnover of extracellular matrix components. Our results add to a better understanding of themolecular mechanisms behind the effects of alpha-blockade on prostatic histoarchitecture and the responseto treatment for benign prostatic hyperplasia.

© 2012 Elsevier Inc. Open access under the Elsevier OA license.

Introduction

Although the major function of the prostate gland is performed bythe secretory epithelium, it is well known that the stromal compart-ment plays a crucial role in the structural support and in the mainte-nance of the differentiated state of epithelial cells (Marker et al., 2003).

Fibroblasts and smoothmuscle cells (SMC) synthesize, organize andmaintain a complex extracellular network in the prostate stroma (Knoxet al., 1994; Nagle, 2004). The collagen proteins are some of the mostimportant extracellular matrix (ECM) macromolecules (Hynes, 2009),and their rates of synthesis, accumulation and degradation are crucialfor prostate structure and function (Myllyharju and Kivirikko, 2001).The main collagen proteins found in the prostatic stroma are types I,III, IV, VI and VII (Sinha et al., 1991; Knox et al., 1994; de Carvalhoet al., 1997a; Vilamaior et al., 2000; Taboga and Vidal, 2003). Fibers ofthe elastic system, hyaluronan, proteoglycans and others glycoproteinsare also present (Knox et al., 1994; de Carvalho et al., 1997b) and

a, Instituto de Biosciencias, Univl. Tel./fax: +55 14 3880 0474..

sevier OA license.

contribute to the dynamic epithelium-stroma cross-talk via paracrinesignals that regulate the behavior of both epithelial and stromal cells(Marker et al., 2003). Similarly, epithelial and stromal cells dynamicallyinteract with ECMmolecules so that changes in the ECMmicroenviron-ment can lead to alterations in the physiology of both epithelial andstromal cells (Nelson and Bissell, 2006).

Stromal cells and extracellular matrix components are alteredin prostatic diseases (Knox et al., 1994; Lee and Peehl, 2004). It isknown that the fibrotic process is regulated by transforming growthfactor-beta 1 (TGF-beta 1), a growth factor that recruits proinflammatorycells andfibroblasts to thefibrotic area, stimulating these cells to producecytokines and extracellular matrix components, like type I collagen(Jimenez et al., 1994). However, the extent to which these alterationsare causes or consequence of prostatic diseases is unknown.

Alpha-1 adrenergic receptor blockers, such as doxazosin, inhibitthe binding of catecholamines (released from autonomous nervevaricosities) to alpha-1 receptors on the membrane of smooth musclecells. Thus, they lead to relaxation of prostatic smooth muscle cellsand relieve the symptoms associated with benign prostatic hyperpla-sia (BPH). Besides the direct action on the smooth muscle cells,doxazosin has been shown to induce apoptosis in stromal and epithe-lial prostatic cells, suggesting additional effects on the long-term

1282 F.K. Delella et al. / Life Sciences 91 (2012) 1281–1287

management of BPH (Anglin et al., 2002; Chiang et al., 2005; Kyprianouet al., 1998). Although the mechanisms by which doxazosin inducesapoptosis are not clear, some studies suggest that apoptosis occurs as fre-quently in normal cells as it does in tumor cells (Kyprianou et al., 2009).

Previous results from our laboratory demonstrated for the first timethat doxazosin treatment increases deposition of collagen fibers in therat prostate stroma (Justulin et al., 2008). Recently, Imamura et al.(2010) found that patients treated orally with α1-blockers had accu-mulation of collagen fibers in the prostatic stroma. The authorssuggested that this structural change could be one of the factorsresponsible for the development of resistance to this treatment. Werecently showed that fibroblasts become more active and that smoothmuscle cells shift from a predominantly contractile to a more syntheticphenotype under α-adrenergic blockade by doxazosin (Delella andFelisbino, 2010). However, doxazosin treatment also reduced theweight and volume of the prostate gland. Thus, additional studiesusing molecular approaches are needed to evaluate if the increasedarea of collagen fibers observed after doxazosin treatment is due torearrangement of pre-existing fibers in a smaller gland or de novosynthesis of collagen.

Because the epithelium-stroma cross-talk is vital to prostate homeo-stasis, and to gain new insights into themolecularmechanisms of actionof doxazosin on the prostatic stroma, we investigated the stroma ultra-structural morphology and the expression of some genes involved inthe fibrotic pathway, such as collagens type I and III and TGF-beta1 inthe rat ventral prostate.

Material and methods

Animals and doxazosin administration

Adult male 3-month-old Wistar rats (n=30) were maintained in acontrolled environment with free access to food and water. Theexperiment was performed according to the Guide for Care and Use ofLaboratory Animals. The animals were divided into two groups: controlrats and the doxazosin-treated group. Doxazosin-treated animalsreceived daily doses of doxazosin (Doxazosin mesylate, Pfizer, GalenaPharmaceutical and Chemistry, SP, Brazil) at 25 mg/kg of body weightdissolved in corn oil as vehicle by oral gavage. The dosage and durationof treatments were determined by consulting with Pfizer and have pre-viously been published (Yono et al., 2007). The control animals receivedonly the vehicle. After 7 and 30 days of treatment, 10 animals from eachgroup were euthanized by an overdose of pentobarbital. The ventralprostates were excised and immediately weighed, and the left andright lobes were snap frozen or immersed in fixative.

Transmission electron microscopy

The ventral prostates tissue fragments of animals from each ex-perimental group were immersed in 2.5% glutaraldehyde with 0.25%tannic acid solution in Millonig's buffer for 2 hours (Cotta-Pereiraet al., 1976), washed, and post fixed in 1% osmium tetroxide in thesame buffer for 1 hour. Tissue fragments were then washed again,dehydrated in graded acetone, and embedded in Araldite. Semithinsections were stained with Toluidine Blue and were analyzed underlight microscopy. Ultrathin silver sections were obtained with a dia-mond knife and were contrasted with 2% alcoholic uranyl acetatefollowed by 2% lead citrate in 1 N sodium hydroxide solution for10 minutes. The grids were examined under a Phillips transmissionelectron microscope operating at 80 kV.

Morphometric analysis

The transmission electron microscopy images of axonal varicositiesfrom each experimental group were used for morphometric analysisto determine the mean diameter of the neurotransmitter vesicles.

Random interactive measurements were performed for at least 30vesicles per image from two different images of five distinct animalsat a magnification of 60,000× (a minimum of 300 measurementsper experimental group) using Leica Q-win software Version 3 forWindowsTM.

Immunohistochemistry

The ventral prostate frozen sections (7 μm) were collected onsilanized glass slides andfixed in coldmethanol for 10 minutes. Sectionswere then blocked with 3% hydrogen peroxide in methanol for 15 mi-nutes (to block endogenous peroxidase activity) before blocking with3% bovine serum albumin in PBS for 1 hour at room temperature. Thesections were incubated with the following primary monoclonalantibodies for 2 hours at room temperature: Type I Collagen, 1:300,Genetex, GTX 26308; Type III Collagen, 1:300, Abcam, ab 6310;TGF-beta 1, 1:100, Santa Cruz, sc-146; Alpha Actin, 1:300, Santa Cruz,sc-32251. The primary antibody was detected using a secondaryperoxidase-conjugated antibody (Santa Cruz Biotechnology, SantaCruz, CA, USA), except for TGF-beta 1, which was detected using theMach 4 Universal HRP-Polymer Kit with DAB (BiocareMedical, Concord,CA, USA). Chromogen color development was accomplished with3.3′-diaminobenzidine tetrahydrochloride (Sigma, USA). Slides werecounterstained with Harris's hematoxylin. The negative control wasperformed by excluding the primary antibody incubation step.

Protein extraction and western blotting analysis for TGF-beta 1

The ventral prostates frozen samples were mechanically homoge-nized in 50 mM Tris–HCl buffer pH 7.5, 0.25% Triton X-100 and EDTAbymeans of a Polytron homogenizer (Kinematica, Lucerne, Switzerland)for 30 s at 4 °C. Following centrifugation of the homogenate, the proteinwas extracted from the supernatant and was quantified as described byBradford (1976). Equal amounts of protein (75 μg) from the ventralprostate frozen samples were heated at 95 °C for 5 minutes in thesample-loading buffer and were then subjected to SDS–PAGE underreducing conditions and were transferred to nitrocellulose membranes(Sigma Chemical Co., St. Louis, MO). The blots were blocked with 3%bovine serum albumin in TBST (10 mM Tris–HCl pH 7.5, 150 mMNaCl, 0.1% Tween-20) for 1 hour andprobed overnightwith the primaryantibody, anti-TGF-beta 1 (1:1,000; MAB240; R&D Systems, Minneapo-lis, USA). Goat anti-β-actin antibody (1:1,000; Santa Cruz Biotechnology,Santa Cruz, CA, USA) served as loading control. After incubation withthe corresponding horseradish peroxidase-conjugated secondary anti-bodies, the blots were detected by means of chemiluminescence(Pierce® ECL Western Blotting Substrate). TGF-beta 1 and β-actin pro-tein expression was quantified by densitometric analysis of the bandsand was expressed as integrated optical density (IOD). The TGF-beta 1expression was normalized to the β-actin values. Normalized data areexpressed as the means±SD.

Collagen I, collagen III and TGF-beta 1 mRNA expression

Total RNA was isolated from ventral prostate samples using TRIzolReagent (Invitrogen Life Technologies, Carlsbad, CA, USA), accordingto the manufacturer's specifications. The RNA concentration (μg/μL)was determined in each sample using a NanoDrop® ND-1000 UV–vis Spectrophotometer (Nanodrop Technologies, Wilmington, DE,USA). The RNA purity was ensured by obtaining a 260/280 nm ODratio equal to 2.0. Total RNA (2 μg) was treated with AmplificationGrade Deoxyribonuclease I according to the protocol provided byInvitrogen (Life Technologies Corporation, Carlsbad, CA, USA). Thepurified total RNA was reverse transcribed with random hexamerprimers using a High-Capacity cDNA Archive Kit (Life TechnologiesCorporation, Carlsbad, CA, USA). Type I collagen, type III collagenand TGF-beta 1 gene expression levels were detected for each

1283F.K. Delella et al. / Life Sciences 91 (2012) 1281–1287

treatment group using a RT-qPCR system (Applied Biosystems 7300,Foster City, CA, USA). Reactionswere performed in duplicate, includingno-template controls to confirm that themastermixwas free from con-taminants. The RT-qPCR conditions were 10 minutes at 95 °C followedby 40 cycles of denaturation at 95 °C for 15 seconds and annealing/extension at 60 °C for 1 minute. The FAM-conjugated gene-specificcustom assays were: Rn01463848_m1 for the type I Procollagen α-1gene, Rn01437683_m1 for the type III Procollagen α-1 gene andRn99999016_m1 for the Transforming Growth Factor 1 (TGF beta-1)gene. A custom GAPDH (glyceraldehyde 3-phosphate dehydrogenase)VIC-conjugated gene-specific assay (TaqMan 4352338E) was used asa reference to normalize quantification of mRNA targets. This geneshowed invariant expression across all samples (p=0.824).

RT-qPCR was performed for target and reference genes in thesame RT reaction using two replicates of each sample and a referencegene. An amplification plot graphically displayed the fluorescencedetected over the number of cycles that were performed. Using the2−ΔΔCt method, data were recorded as the fold-change in geneexpression normalized by the reference gene and relative to the cali-brator sample (Livak and Schmittgen, 2001). All samples then werenormalized to the ΔCt value of a calibrator sample to obtain a ΔΔCtvalue (ΔCttarget−ΔCtcalibrator). The control group was chosen as thecalibrator sample to evaluate target gene putative differential mRNAexpression relative to this group. Dissociation-melting curves con-firmed the specific amplification of the cDNA target and the absenceof nonspecific products.

Statistical analysis

One-way analysis of variance (ANOVA)was performed to determinewhether differences existed between groups (pb0.05) and the Tukey–Kramer multicomparison post-test was employed. A p-value of b0.05was considered significant. The statistical tests were performed usingInstat (version 3.0; Graph-Pad, Inc., San Diego, CA, USA).

Results

Transmission electron microscopy investigation

The ultrastructural analysis showed that the ventral prostatefrom control rats is composed by a thin stroma structured in a well-organized pattern (Fig. 1a). The collagen fibrils of the subepithelial stro-mawere thin and sparsely distributed adjacent to fibroblast extensions.One or two layers of smoothmuscle cells surrounded the acini. Delicateelastic fibers and bundles of collagen fibers filled the stroma between

Fig. 1. (a) Ultrastructure of the ventral prostate from control rats showing few and thin bundlecells (smc) and fibroblast cytoplasmic extensions (f). Elaunicfibers (white arrows). (b) Ventral pwith doxazosin for 30 days. Note the dense bundles of collagen fibrils (black arrows) around s

fibroblasts and smooth muscle cells. The smooth muscle cells showeda regular and smooth outline (Fig. 1b).

Seven days of doxazosin treatment promoted activation of fibro-blasts (Fig. 1b), which contained large nuclei with loose chromatinand evident nucleoli. The bundles of collagen fibrils were denserbetween the cells (Fig. 1b). At day 30 of dox treatment, fibroblastsand smooth muscle cells showed many cytoplasm extensions closelyassociated with thickened bundles of collagen fibrils (Fig. 1c).

Neurotransmitter vesicle diameter determination

The ultrastructural analysis showed many axonal varicosities filledwith neurotransmitter vesicles and mitochondria dispersed amongsmooth muscle cells in the ventral prostate from control rats (Fig. 2a),and from ventral prostate of rats treated with doxazosin for 7 (Fig. 2b)and 30 days (Fig. 2c). Morphometrical analysis showed that 30 daysof doxazosin treatment significantly increased the mean diameter ofthe neurotransmitter vesicles (Fig. 3). At day 7, there was a trendtowards an increase in the diameter of neurotransmitter vesicles, butwithout statistically significant difference.

Type I and type III collagen analyses

To further examine the effects of doxazosin on major componentsof the extracellular matrix, we assessed the mRNA and protein ex-pressions of collagens type I and III. Type I collagen was found to bedispersed throughout the stroma, surrounding glandular acini andducts in the ventral prostate from control rats (Fig. 4a). Doxazosintreatment increased the density of type I collagen at day 7 (Fig. 4d)and at day 30 of treatment (Fig. 4f), mainly around acini. Expressionof the mRNA for collagen type I (COL1A1) was transiently increasedat day 7, and decreased significantly at 30 days of doxazosin treat-ment (Fig. 7). Type III collagen was found distributed in the intersti-tial stroma, filling epithelial infoldings and around blood vessels(Fig. 4c) of the ventral prostate stroma from control rats. The α-1adrenergic blockade also increased the distribution of collagen typeIII at day 7 (Fig. 4e) and day 30 (Fig. 4g) of the treatment, mainlybelow the epithelium. Expression of the mRNA for collagen type III(COL3A1) was also transiently increased at day 7, and was significant-ly decreased at day 30 (Fig. 7).

TGF-beta 1 expression

To evaluate TGF-beta 1 localization, mRNA and protein expressionin the ventral prostate from control and doxazosin-treated rats, weused immunohistochemistry, real time quantitative PCR and western

s of collagen fibrils (black arrows) above the epithelium (ep) and around smooth musclerostate from a rat treatedwith doxazosin for 7 days. (c) Ventral prostate from a rat treatedmooth muscle cells (smc) and active fibroblasts (f). Scale bars=1 μm.

Fig. 2. Ultrastructure of the rat ventral prostate from control (a), doxazosin-treated for 7(b) and 30 days (c) showing neurotransmitter vesicles inside the axonal varicositiesdispersed among smooth muscle cells. Note the increased diameter of neurotransmittervesicles after doxazosin treatment (arrows). Scale bars=400 nm.

Fig. 3.Mean diameter of neurotransmitter vesicles of the ventral prostate from control,doxazosin day 7 and doxazosin day 30 rats, assessed by morphometry. Values aremean±SD. *Statistically different from control and from doxazosin day 7 with pb0.05.

1284 F.K. Delella et al. / Life Sciences 91 (2012) 1281–1287

blotting, respectively. TGF-beta 1 was weakly detected in both epithe-lial and stromal compartments in the ventral prostate from controlrats (Fig. 5a). Doxazosin treatment increased TGF-beta 1 immuno-staining at day 7 (Fig. 5b), which returned to levels similar to thatfound in ventral prostate from control rats at day 30 (Fig. 5c).

Fig. 6a shows a representative western blot for TGF-beta 1 fromthe ventral prostate of control and the doxazosin-treated rats. Thespecific immunoreactive band for TGF-beta 1 (25 kDa) was more in-tense in the ventral prostate from rats treated with doxazosin for7 days (pb0.05) when compared to controls (Fig. 6b).

The results of the q-RT-PCR for TGF-beta 1 are shown in Fig. 7.Doxazosin treatment significantly increased TGF-beta 1 mRNA ex-pression at day 7 (pb0.01), which returned to levels observed in con-trols at day 30.

Discussion

Alpha 1-adrenergic receptor blockers obstruct the adrenergic re-ceptors, which are abundant in smooth muscle cells of the prostateand bladder, resulting in a reduction of smooth muscle tone and

have been a safe and efficacious pharmacological treatment optionfor patients suffering from BPH (AUA, 2003; Caine, 1986).

The quilinazole-based class of α-adrenoceptor blockers, such asdoxazosin, has the additional effect of inducing apoptosis of prostatecells, which has been demonstrated to be involved in the long-term ef-ficacy of treatments for BPH (Anglin et al., 2002; Chiang et al., 2005;Kyprianou, 2003; Kyprianou et al., 1998; Yang et al., 1997). The induc-tion of apoptosis in prostate cells by doxazosin has been attributed toan increased expression of TGF beta-1 (Glassman et al., 2001; Ilioet al., 2001; Partin et al., 2003; Yang et al., 1997). However, Zhao et al.(2005), using a molecular approach, showed an involvement of TNF-alpha in the doxazosin molecular mechanism of action and disprovedthe participation of the TGF-beta 1 pathway in doxazosin-inducedapoptosis in stromal cells. Thus, more studies are needed to betterunderstand the direct and indirect effects of doxazosin treatment onprostatic tissue cells apoptosis.

Although the involvement of TGF-beta 1 in doxazosin-induced apo-ptosis is still controversial, there is no doubt that this multifunctionalcytokine regulates extracellular matrix production and degradation(Fleisch et al., 2006;Mauviel, 2005; Roberts et al., 1990). Our laboratoryhas been focused on the changes in the prostatic stroma induced bydoxazosin treatment, with special attention to the extracellular matrixcomponents. We have previously demonstrated that doxazosin treat-ment increases both collagen and elastic system fiber volume fractionsin the stroma of the rat prostatic complex (Delella and Felisbino, 2010;Justulin et al., 2008).

In the present study, we showed that 7 days of doxazosin treatmentsignificantly increased the mRNA and protein expressions of collagentype I and III and TGF beta-1 in the rat ventral prostate. Our results arein agreementwith previous results of total collagen fibers quantificationby picrosirius staining coupled tomorphometric analyses (Justulin et al.,2008). Moreover, the ultrastructural findings clearly showed thickenedbundles of collagen fibrils around stromal cells in the ventral prostatefrom doxazosin-treated rats. Together, these data suggest an earlyprofibrotic effect of doxazosin on prostate stroma via TGF beta-1 signal-ing. Although rodent models present some limitations and no entirelyresults can be applied for studying human BPH, they provide interestinginformation about the behavior of the prostatic tissue under several dif-ferent experimental conditions and shed light on the molecular mecha-nisms behind it. In this sense, our results using rats corroborate thefindings of Imamura et al. (2010) that have addressed the effectof doxazosin treatment on collagen fibers deposition in the humanhyperplasic prostate gland. As observed in our study, these authorshave found an increased amount of collagen fibers in a group of patientstreated by doxazosin and suggested that this stromal alteration may berelated to reduction in the long-term effectiveness of the treatment.

Fig. 4. Immunohistochemistry for collagens type I and III on sections from the ventral prostate of rats from control (a–c), doxazosin day 7 (d and e) and doxazosin day 30 groups (f and g).b: Negative control. a, d and f: Type I collagen. c, e and g: Type III collagen. Positive reaction for collagen types was observed in the interstitial stroma (asterisks) and in the thin periacinarstroma (arrows). Note the increased intensity of immunoreaction for collagen types in the prostate from doxazosin-treated rats. Scale bars: a, c–g=50 μm; b=10 μm.

1285F.K. Delella et al. / Life Sciences 91 (2012) 1281–1287

Notably, mRNA expression for both collagens I and III was signifi-cantly decreased after 30 days of doxazosin treatment, whereasTGF-beta 1 mRNA, but not the protein, remained increased. This lateeffect of doxazosin on collagen mRNA expression suggests a reduc-tion in the turnover of these proteins by stromal cells. These resultsreinforce the importance of evaluating different times of exposureto a better understanding of doxazosin effects on the prostate gland.The consequence of this late down regulation of collagen mRNAexpression by doxazosin for the maintenance of normal prostaticstroma homeostasis and for the epithelium-stroma interactions stillneeds to be determined and is currently under investigation by ourlaboratory.

The morphology of stromal cells was also affected by alpha1blockade. We showed previously that prostatic smooth muscle cellsexhibited a synthetic phenotype (Delella and Felisbino, 2010) afterdoxazosin treatment. Smooth muscle cells are known to reversiblyswitch phenotypes between a synthetic and a contractile state

(Halayko et al., 2008; Zhou et al., 2009). Smith et al. (1999) showedthat doxazosin decreased the contractility and expression of actinand myosin proteins in prostatic smooth muscle cells in vitro. Thissuggests that alpha1-blockers may induce dedifferentiation ofsmooth muscle cells into fibroblasts and myofibroblasts. However, ifthis possible dedifferentiation occurs in vivo remains to be deter-mined. In the present study, we observed activated fibroblasts withenlarged cytoplasm and loose chromatin after 7 days of doxazosintreatment. Activated fibroblasts are in agreement with the up regula-tion of collagen mRNA and TGF-beta 1 protein observed at this periodof the treatment. Furthermore, fibroblast phenotype at 30 days ofdoxazosin treatment is also in agreement with the down regulationof collagen mRNA and the normalization of TGF-beta 1 proteins. To-gether, these results reinforce our previous hypothesis (Delella andFelisbino, 2010; Justulin et al., 2008) that the reduction of smoothmuscle cell tonus induced by doxazosin treatment promotes changesin the prostatic tissue mechanical properties, leading to extracellular

Fig. 5. Immunohistochemistry for TGF-beta 1 protein in the rat ventral prostate. (A) Control; (B) doxazosin day 7; (C) doxazosin day 30. Note the positive immunostaining forTGF-beta 1 in the epithelial and stromal compartments (arrows), which is increased in the doxazosin day 7 group. Bars=10 μm.

1286 F.K. Delella et al. / Life Sciences 91 (2012) 1281–1287

matrix remodeling by fibroblasts and smooth muscle cells. We pro-pose that this stromal remodeling following doxazosin treatmentmay be detrimental for BPH therapy, since it increases the static com-ponent of the prostatic tissue that does not respond to alpha blockers(Imamura et al., 2010). Since quinazoline derivatives has alpha-1adrenoceptor independent effects, future studies from our laboratorywill evaluate our hypothesis using Tamsulosin, a non-quinazolineselective α-adrenergic antagonist that has more specificity for α-1Aand α-1D receptor subtypes.

Curiously, phenylephrine treatment, a selective α1-adrenergicreceptor agonist, did not exhibit opposite effect than doxazosin withregards to fibrosis. Previous studies showed that phenylephrine treat-ment induced atypical prostatic hyperplasia and mild fibrosis in therat ventral prostate (Golomb et al., 1998; Rosenzweig et al., 2004;Rosenzweig-Bublil and Abramovici, 2006). These effects have beenrelated to a tissue repair process occurring subsequent to the inflam-matory exudates that takes place during early period of phenyleph-rine treatment.

Fig. 6. (A) Representative western blot analyses of TGF-beta 1 and β-actin (Actin) inhomogenates from ventral prostates (lane 1=control; lane 2=doxazosin day 7;lane 3=doxazosin day 30). Seventy five micrograms of protein was loaded into eachlane (15 μg from five different individual homogenates). (B) Relative expression(IOD integrated optical density) of TGF-beta 1 and β-actin (loading control) (Fig. a).TGF-beta 1 was detected in the ventral prostate from all groups and was increased inthe doxazosin day-7 group (Fig. B). Data are as mean±SD. *Statistically significantdifferent from control with pb0.05.

As the human prostate, the rat ventral prostate is largely innervatedby both the sympathetic and parasympathetic nervous system (Vaalastiand Hervonen, 1979, 1980). The contraction of the smoothmuscle cellsthat surround acini and ducts is under sympathetic control, and itsactivation initiates the contractile function of the genital duct systemand the prostatic muscular element (Bruschini et al., 1978). It is wellknown that doxazosin is a selective alpha 1 blocker that occupiesthe alpha 1 adrenoceptor site for norepinephrine, thereby inhibitingsmooth muscle contraction (Lepor, 2007). Our study showed for thefirst time that alpha blockade increases the size of neurotransmittervesicles inside the axonal varicosities dispersed along smooth musclecells. It is known that after exocytosis, norepinephrine may betransported back into synaptic vesicles by the norepinephrine trans-porter and the vesicular monoamine transporter (Eisenhofer, 2001).However, how doxazosin is involved in this process remains to bedetermined. Our results encourage future studies to investigate ifdoxazosin and other alpha-blockers may increase norepinephrinereuptake into axonal varicosities and vesicles.

Conclusion

In summary, our data suggest that relaxation of smoothmuscle cellsby treatment with an alpha-blocker modulates collagen mRNA expres-sion in the prostatic stroma via the TGF-beta 1 signaling pathway,leading to a marked remodeling of cellular and extracellular elementsof the prostate stroma. Our results contribute to a better understanding

Fig. 7. Collagens type I (COL1A1) and III (COL3A1) and TGF-beta 1 mRNA expression asquantified by qRT-PCR in the ventral prostate of rats treated with doxazosin for 7 and30 days. Levels of COL1A1 and COL3A1 and TGF-beta 1 mRNA were significantly upregulated at day 7 of treatment and down regulated at day 30. The target genes werenormalized to the level of the housekeeping gene GAPDH. Data are significantly differ-ent from control group at pb0.05 (*), pb0.01 (**) and pb0.001 (***).

1287F.K. Delella et al. / Life Sciences 91 (2012) 1281–1287

of the molecular mechanisms of action of alpha-blockade on prostatehistoarchitecture and the response to BPH treatment.

Conflict of interest statement

The authors declare that they have no conflit of interest.

Acknowledgments

The authors thank the Electron Microscopy Center (Institute ofBiosciences, UNESP, Botucatu, SP) for the transmission electronmicroscopy protocol procedures and for the use of Phillips electronmicroscope. This work was funded by FAPESP (São Paulo StateResearch Foundation; grants 06/60114-6, to S.L.F. and 06/60115-2,to F.K.D); by FUNDUNESP (Foundation for Development of UnivEstadual Paulista), by CNPq, and by CAPES do Brazil. This paper repre-sents part of the PhD thesis presented by F.K.D. to the University ofCampinas—UNICAMP, Brazil.

References

Anglin IE, Glassman DT, Kyprianou N. Induction of prostate apoptosis by alpha1-adrenoceptor antagonists: mechanistic significance of the quinazoline component.Prostate Cancer Prostatic Dis 2002;5:88–95.

AUA Practice Guidelines Committee. AUA guideline on management of benign prostatichyperplasia. Chapter 1: Diagnosis and treatment recommendations. J Urol 2003;170:530–47.

BradfordMM. A rapid and sensitivemethod for the quantitation of microgram quantities ofprotein utilizing the principle of protein-dye binding. Anal Biochem 1976;72:248–54.

Bruschini H, Schmidt RA, Tanagho EA. Neurologic control of prostatic secretion in thedog. Invest Urol 1978;15:288–90.

Caine M. The present role of alpha-adrenergic blockers in the treatment of benign pros-tatic hypertrophy. J Urol 1986;136:1–4.

Chiang CF, Son EL, Wu GJ. Oral treatment of the TRAMP mice with doxazosin sup-presses prostate tumor growth and metastasis. Prostate 2005;64:408–18.

Cotta-Pereira G, Rodrigo FG, David-Ferreira JF. The use of tannic acid-glutaraldehyde inthe study of elastic and elastic related fibers. Stain Technol 1976;51:7-11.

de Carvalho HF, Taboga SR, Vilamaior PS. Collagen type VI is a component of the extracel-lular matrix microfibril network of the prostatic stroma. Tissue Cell 1997a;29:163–70.

de Carvalho HF, Vilamaior PS, Taboga SR. Elastic system of the rat ventral prostate andits modifications following orchiectomy. Prostate 1997b;32:27–34.

Delella FK, Felisbino SL. Doxazosin treatment alters stromal cell behavior and increaseselastic system fibers deposition in rat prostate. Microsc Res Tech 2010;73:1036–44.

Eisenhofer G. The role of neuronal and extraneuronal plasma membrane trans-porters in the inactivation of peripheral catecholamines. Pharmacol Ther2001;91:35–62.

Fleisch MC, Maxwell CA, Barcellos-Hoff MH. The pleiotropic roles of transforminggrowth factor beta in homeostasis and carcinogenesis of endocrine organs. EndocrRelat Cancer 2006;13:379–400.

Glassman DT, Chon JK, Borkowski A, Jacobs SC, Kyprianou N. Combined effect ofterazosin and finasteride on apoptosis, cell proliferation, and transforming growthfactor-beta expression in benign prostatic hyperplasia. Prostate 2001;46:45–51.

Golomb E, Kruglikova A, Dvir D, Parnes N, Abramovici A. Induction of atypical prostatichyperplasia in rats by sympathomimetic stimulation. Prostate 1998;34:214–21.

Halayko AJ, Tran T, Gosens R. Phenotype and functional plasticity of airway smoothmuscle: role of caveolae and caveolins. Proc Am Thorac Soc 2008;5:80–8.

Hynes RO. The extracellular matrix: not just pretty fibrils. Science 2009;26:1216–9.Ilio KY, Park II, Pins MR, Kozlowski JM, Lee C. Apoptotic activity of doxazosin on pros-

tate stroma in vitro is mediated through an autocrine expression of TGF-beta 1.Prostate 2001;48:131–5.

Imamura T, Ishii K, Kanda H, Arase S, Yoshio Y, Hori Y, et al. Structural changes ina1-adrenoceptor antagonist-treated human prostatic stroma. Clin Exp Med 2010;10:99-106.

Jimenez SA, Varga J, Olsen A, Li L, Diaz A, Herhal J, et al. Functional analysis of humanalpha 1(I) procollagen gene promoter. Differential activity in collagen-producingand -nonproducing cells and response to transforming growth factor beta 1.J Biol Chem 1994;269:12684–91.

Justulin Jr LA, Delella FK, Felisbino SL. Doxazosin reduces cell proliferation and increasescollagen fibers in rat prostatic lobes. Cell Tissue Res 2008;332:171–83.

Knox JD, Cress AE, Clark V, Manriquez L, Affinito KS, Dalkin BL, et al. Differential expres-sion of extracellular matrix molecules and the alpha 6-integrins in the normal andneoplastic prostate. Am J Pathol 1994;145:167–74.

Kyprianou N. Doxazosin and terazosin suppress prostate growth by inducing apoptosis:clinical significance. J Urol 2003;169:1520–5.

Kyprianou N, Litvak JP, Borkowski A, Alexander R, Jacobs SC. Induction of prostateapoptosis by doxazosin in benign prostatic hyperplasia. J Urol 1998;159:1810–5.

Kyprianou N, Vaughan TB, Michel MC. Apoptosis induction by doxazosin and otherquinazoline alpha1-adrenoceptor antagonists: a new mechanism for cancer treat-ment? Naunyn Schmiedebergs Arch Pharmacol 2009;380:473–7.

Lee KL, Peehl DM. Molecular and cellular pathogenesis of benign prostatic hyperplasia.J Urol 2004;172:1784–91.

Lepor H. Alpha blockers for the treatment of benign prostatic hyperplasia. Rev Urol2007;9:181–90.

Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-timequantitative PCR and the 2(−Delta Delta C(T)) Method. Methods 2001;25:402–8.

Marker PC, Donjacour AA, Dahiya R, Cunha GR. Hormonal, cellular, and molecular con-trol of prostatic development. Dev Biol 2003;253:165–74.

Mauviel A. Transforming growth factor-beta: a key mediator of fibrosis. Methods MolMed 2005;117:69–80.

Myllyharju J, Kivirikko KI. Collagens and collagen-related diseases. Ann Med 2001;33:7-21.Nagle RB. Role of the extracellular matrix in prostate carcinogenesis. J Cell Biochem

2004;91:36–40.Nelson CM, Bissell MJ. Of extracellular matrix, scaffolds, and signaling: tissue architecture

regulates development, homeostasis, and cancer. Annu Rev Cell Dev Biol 2006;22:287–309.

Partin JV, Anglin IE, Kyprianou N. Quinazoline-based alpha 1-adrenoceptor antagonistsinduce prostate cancer cell apoptosis via TGF-beta signalling and I kappa B alphainduction. Br J Cancer 2003;88:1615–21.

Roberts AB, Heine UI, Flanders KC, Sporn MB. Transforming growth factor-beta. Majorrole in regulation of extracellular matrix. Ann N Y Acad Sci 1990;580:225–32.

Rosenzweig-Bublil N, Abramovici A. Stromal fibrosis reaction in rat prostates inducedby alpha 1 adrenergic stimulation. J Androl 2006;27:276–84.

Rosenzweig N, Horodniceanu J, Abramovici A. Phenylephrine-induced neurogenicprostatitis facilitates the promotion of PIN-like lesions in rats. Prostate 2004;59:107–13.

Sinha AA, Gleason DF, DeLeon OF, WilsonMJ, Limas C, Reddy PK, et al. Localization of typeIV collagen in the basement membranes of human prostate and lymph nodes byimmunoperoxidase and immunoalkaline phosphatase. Prostate 1991;18:93-104.

Smith P, Rhodes NP, Ke Y, Foster CS. Influence of the a1-adrenergic antagonist, doxazosin,on noradrenalin-induced modulation of cytoskeletal proteins in cultured hyperplasticprostatic stromal cells. Prostate 1999;38:216–27.

Taboga SR, Vidal BC. Collagen fibers in human prostatic lesions: histochemistry andanisotropies. J Submicrosc Cytol Pathol 2003;35:1–6.

Vaalasti A, Hervonen A. Innervation of the ventral prostate of the rat. Am J Anat 1979;154:231–43.

Vaalasti A, Hervonen A. Autonomic innervation of the human prostate. Invest Urol1980;17:293–7.

Vilamaior PS, Felisbino SL, Taboga SR, Carvalho HF. Collagen fiber reorganization in therat ventral prostate following androgen deprivation: a possible role for smoothmuscle cells. Prostate 2000;45:253–8.

Yang G, Timme TL, Park SH, Wu X, Wyllie MG, Thompson TC. Transforming growthfactor beta 1 transduced mouse prostate reconstitutions: II. Induction of apoptosisby doxazosin. Prostate 1997;33:157–63.

Yono M, Yamamoto Y, Yoshida M, Ueda S, Latifpour J. Effects of doxazosin on blood flowand mRNA expression of nitric oxide synthase in the spontaneously hypertensive ratgenitourinary tract. Life Sci 2007;81:218–22.

Zhao H, Lai F, Nonn L, Brooks JD, Peehl DM. Molecular targets of doxazosin in humanprostatic stromal cells. Prostate 2005;62:400–10.

Zhou Y, Xiao XQ, Chen LF, Yang R, Shi JD, Du XL, et al. Proliferation and phenotypicchanges of stromal cells in response to varying estrogen/androgen levels in castratedrats. Asian J Androl 2009;11:451–9.