Testosterone Regulates PDE5 Expression and in vivo Responsiveness totadalafil in Rat Corpus Cavernosum

European Urology European Urology 47 (2005) 409 416 Testosterone Regulates PDE5 Expression and in vivo Responsiveness totadalafil in Rat Corpus Cavernosum Xin-hua Zhang a,1, Annamaria Morelli a,1, Michaela Luconi a, Linda Vignozzi a, Sandra Filippi b, Mirca Marini c, Gabriella Barbara Vannelli c, Rosa Mancina a, Gianni Forti a, Mario Maggi a, * a Andrology Unit, Department of Clinical Physiopathology, University of Florence, V.le G. Pieraccini, 6, 50139 Florence, Italy b Interdepartmental Laboratory of Functional and Cellular Pharmacology of Reproduction, Department of Pharmacology and Clinical Physiopathology, University of Florence, V.le G. Pieraccini, 6, 50139 Florence, Italy c Department of Anatomy, Histology and Forensic Medicine, University of Florence, V.le G. Pieraccini, 6, 50139 Florence, Italy Accepted 26 October 2004 Available online 18 November 2004 Abstract Objectives: To investigate the effect of testosterone on PDE5 expression and PDE5 inhibitor tadalafil in vivo responsiveness in a rat model. Methods: PDE5 expression was localized by immunohistochemistry in the rat corpus cavernosum (CC) and quantified by both real-time RT-PCR and Western blot analysis in several tissues. In the in vivo study, control, castrated and testosterone (T) supplemented castrated rats were treated with acute or chronic oral tadalafil. Erectile function was evaluated by monitoring intracavernous pressure (ICP) following electro-stimulation (ES) of the cavernous nerve and intracavernous injection of NO donor, sodium nitroprusside (SNP). Results: Rat CC expressed the highest PDE5 mrna level. PDE5 was specifically immunolocalized in endothelial and smooth muscle cells. Surgical castration induced a significant reduction of PDE5 gene and protein expression (p < 0.05), and ES response at all stimulation frequencies (p < 0.001). T supplementation completely restored PDE5 expression, erectile response to ES and responsiveness to PDE5 inhibitor. Both acute and chronic tadalafil treatment were ineffective in ameliorating the ES response in castrated rats. Injection of increasing concentrations of SNP in castrated rats resulted in a statistically significant increase in ICP/MAP ratio as that observed in intact rats. In addition, tadalafil did not amplify the SNP effect in castrated rats at all the doses tested (0.06 6 nmoles). Conclusions: Our findings demonstrate that testosterone positively regulates PDE5 expression and in vivo responsiveness to PDE5 inhibitor, tadalafil, in the rat CC. # 2004 Elsevier B.V. All rights reserved. Keywords: PDE5; Tadalafil; Corpus cavernosum; Castration; Testosterone; ICP; Rat 1. Introduction The erectile process has been shown to be strongly androgen-sensitive, at least in the rat model. Essentially, androgens play a dual role in the erectile process by controlling both pro-erectile and anti-erectile signalling pathways, making erection a discrete phenomenon strictly related to sexual arousal and reproduction * Corresponding author. Tel. +39 055 42 71 415; Fax: +39 055 42 71 413. E-mail address: m.maggi@dfc.unifi.it (M. Maggi). 1 The authors equally contributed to the work. [1,2]. In addition, androgens exert a positive role in the maintenance of the general architecture of the penile contractile structure. In fact, castration induces cavernous smooth muscle atrophy and contributes to a structural alteration of erectile tissue with a relative increase in connective tissue content [3 5]. In castrated rats, erectile response is markedly impaired while testosterone (T) reverses this deficiency [3,4]. Castration has also been associated with damage of veno-occlusive mechanism and increased cavernosal contraction to a-adrenergic stimuli [6,7]. However, the primary role of androgens in erectile function is 0302-2838/$ see front matter # 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.eururo.2004.10.021

410 X.-h. Zhang et al. / European Urology 47 (2005) 409 416 postulated to be via stimulation of cgmp formation and degradation. Castration reduced nitric oxide (NO) and its generating enzyme NO synthase (NOS), as well as NOS-containing nerve fibers. As these events are completely reversed by T, it is generally assumed that androgens positively regulate NO synthesis and therefore cgmp formation [8 11]. We recently demonstrated that androgens also up-regulate the major mechanism responsible for cgmp degradation, i.e. phosphodiesterase type 5 (PDE5) [2]. We found that this PDE isoform is quantitatively abundant in the human male genital tract with the highest expression in the corpus cavernosum (CC) when compared with female reproductive and male non reproductive tissues. Moreover, in human and in rabbit model of hypogonadotropic hypogonadism, androgen ablation decreases while T supplementation restores CC PDE5 gene and protein expression, along with functional in vitro activity [2]. This castration-induced PDE5 depletion might, in turn, result in an hypo-responsiveness to PDE5 inhibitors, such as sildenafil. In fact, both in rabbits and humans, the hypogonadism-induced reduction of PDE5 expression leads to an in vitro decreased responsiveness to sildenafil [2]. A similar phenomenon has been described in hypogonadal patients, which were initially resistant to sildenafil therapy and became sensitive after appropriate T supplementation [12,13]. However, compelling evidences showing a direct in vivo effect of androgens on PDE5 expression and activity are still lacking. To fully understand the role of androgens in PDE5 biological function, we advanced our findings to an in vivo rat model. The primary goal of the present study was to verify whether rat CC expresses a high abundance of PDE5, and whether it is regulated by androgen. Thereafter, we tested whether acute or chronic responsiveness to a long-lasting PDE5 inhibitor, tadalafil, was affected by surgical castration and T replacement. 2. Materials and methods 2.1. Animals and tissues Male and female adult Sprague-Dawley rats (Harlan, Italy) weighing 275 330 g were used. For electrical stimulation (ES), male rats were separated into 3 groups: control, surgical castration, and castration with T (testosterone proprionate, Schering, Milan, Italy) replacement (100 mg/kg, subcutaneous injection after orchectomy, weekly for 2 weeks) [14]. Each group was further divided into 3 subgroups: no subtreatment, acute tadalafil (Eli Lilly, Sesto Fiorentino, Florence, Italy) dosing (one dose, 6 mg/kg by oral gavage, 2 h preceding the experiment) and chronic tadalafil therapy (2 mg/kg/day, dissolved in drinking water for 2 weeks) [15,16]. For intracavernous injection of sodium nitroprusside (SNP, Sigma, St. Louis, USA), rats were divided into the following 4 groups: control, control with acute tadalafil treatment, castration, and castration with acute tadalafil treatment. All experimental rats were kept 2 weeks. CC were collected after each experiment. Male genital tract and brain tissues were also harvested from control rats. Uterus was obtained from an additional control group of female rats. The Local Ethical Committee for Investigations in Animals of the University of Florence approved the study. Human CC samples, used as positive control for Western blot analysis, were obtained from patients undergoing penile prosthesis implantation (n = 4) with the approval of the Hospital Committee for Investigation in Humans and with informed consent. 2.2. Isolation of RNA and cdna synthesis Total RNA was extracted from frozen tissues using TRIZOL (Invitrogen, Carlsbad, CA, USA). For each sample 400 ng of total RNA were reverse-transcribed to cdna in a final volume of 80 ml using TaqMan Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA) under the following conditions: 10 min at 25 8C; 30 min at 48 8C, 5 min at 95 8C. 2.3. Real-time quantitative RT-PCR The mrna quantitative analysis was performed according to the fluorescent TaqMan methodology [2]. PCR primers and probes for PDE5 mrna were purchased as an Assay-on-Demand (AoD) gene expression product from Applied Biosystems. The beta-2 microglobulin gene was chosen as reference gene and the corresponding AoD product was provided by Applied Biosystems. The PCR mixture (25 ml final volume) consisted of 1 final concentration of AoD mix, 1 final concentration of Universal PCR Master Mix (Applied Biosystems), and 25 ng of cdna. Amplification and detection were performed with the ABI Prism 7700 Sequence Detection System with the following thermal cycler conditions: 2 min at 50 8C, 10 min at 95 8C, and 40 cycles at 95 8C for 30 s and 60 8C for1min.each measurement was carried out in duplicate. The analysis of the results was based on the comparative Ct method according to the manufacturer s instructions (Applied Biosystems), where Ct represents the cycle number at which the fluorescent signal, associated with an exponential increase in PCR products, crosses a given threshold. The maximum change in Ct values of the sample was determined by subtracting the average of duplicate Ct values of the reference gene from the average of duplicate Ct values of the target gene. 2.4. SDS-PAGE and Western blot analysis As previously described [2], frozen samples were directly suspendedinlysisbufferandhomogenizedforproteinanalysis. Thirtymg of protein extracts were separated on 10% denaturing SDS-polyacrylamide gel. After protein transfer to nitrocellulose membranes, the PDE5 protein was revealed by the previously characterized [2,17] anti-pde5 primary antibody (1:1000 in 10% blocking solution) followed by peroxidase-conjugated IgG (1:3000). 2.5. Immunohistochemistry Immunohistochemical studies were performed as previously described [2]. Briefly, the slides were stained for indirect immunoperoxidase technique using polyclonal anti-pde5 (1:100 vol/ vol) [18] as primary antibody. Controls were performed as previously described [2,17]. 2.6. Evaluation of erectile function As reported [19], rats were anesthetized with pentobarbital 45 mg/kg intraperitoneal injection. Mean arterial pressure (MAP) was continuously monitored via the femoral artery. Both

crura were exposed and each was perforated with a 26-gauge needle connected to PE-50 tubing filled with heparin (250 IU/ml) for ICP recording and drug delivery. MAP and ICP were recorded via a pressure transducer (World Precision Instruments Inc, Sarasota, FL, USA) connected to a TA240 recorder (Gold Inc, Cleveland, OH, USA). For ES, a bipolar platinum electrode attached to ST6 stimulator (Biomedica Mangoni Inc, Pisa, Italy) was mounted on the cavernous nerve. ES (width 5 ms, duration 30 s, 2.5 V) at varying frequencies (1, 2, 4, 8, 16, 32 Hz) was performed with a 5- min interval between each stimulation. Saline or increasing cumulative doses (0.06 6 nmoles) of SNP in a final volume of 60 ml were intracavernously administered at 15-min intervals. The erectile response elicited by electro- or pharmaco-stimulation was quantified by calculating maximal ICP/MAP 100 ratio. X.-h. Zhang et al. / European Urology 47 (2005) 409 416 411 2.7. Statistical analysis Results are expressed as mean SEM. Statistical analysis was performed with one-way ANOVA test followed by Student s t-test when appropriate. p < 0.05 was taken as significant. The computer program ALLFIT [20] was used for the analysis of frequencydependent ICP/MAP ratios that resulted sigmoid (S-shaped) after logarithmic transformation. ALLFIT uses the constrained fourparameter logistic model to analyze families of sigmoid curves, to obtain estimates of ED 50 values and to compare them using an F- test and tests of randomness of the residuals around the fitted curves. In particular, the statistical hypothesis that two or more curves share ED 50 can be tested by first forcing the curves to share this parameter ( shared ED 50 ) and then verifying that such constraints have only minimal (non significant) effect on several indicators of goodness of fit. 3.Results 3.1. Expression of penile PDE5 and modulation by testosterone Fig. 1, panel A shows quantitative expression of PDE5 mrna detected by real-time RT-PCR in several rat tissues, including male genital tract. As previously described for human tissues [2], the highest PDE5 mrna levels were detected in rat CC and vas deferens (VD), whilst the lowest was found in the uterus. The rat brain expresses similar PDE5 mrna abundance as VD. The analysis of PDE5 protein expression by Western blot provides similar results as those observed in gene expression (Fig. 1, panel B). In particular, an intense band of the expected molecular weight (approximately 98KDa) was found in human CC (positive control) and rat CC, VD and brain. The effect of surgical castration and T replacement on PDE5 expression is displayed in Fig. 2. Castration induced a significant reduction of both PDE5 gene and protein expression (p < 0.05 for mrna and p < 0.005 for protein), which was restored by T supplementation. Immuno-distribution of PDE5 in rat CC is shown in Fig. 3. We found that smooth muscle cells and endothelial cells of penile cavernae and blood vessels were intensively labeled (Panels A-C). By omitting the Fig. 1. (A) Quantitative detection of mrna of PDE5 in intact rat tissues by real-time RT-PCR. Data are expressed as mean SEM of arbitrary unit (a.u.), calculated according to the comparative Ct method and using the beta-2 microglobulin gene for normalization. (B) Western blot analysis of PDE5 in rat tissues. Thirty micrograms of proteins separated by 10% SDS- PAGE were probed with anti-pde5 antibody. A specific major band of the expected size (98 kda) is evident in all lanes. Human corpus cavernosum tissue (CC) was run as positive control [2]. Molecular weight marker (kda) is indicated to the left of the blot. primary antiserum, labeling was completely prevented (Panel D). 3.2. MAP and baseline ICP (bicp) As shown in Table 1, the MAP and bicp in control were 130.5 2.97 mmhg and 12.1 0.48 mmhg, respectively. Castration reduced bicp to 10.2 0.73 mmhg with no impact on MAP. On average, tadalafil dosing lowered MAP by 9 11 mmhg in control and T-replaced groups (but not in the castrated group), although this decrease was not significant. Castration significantly lowered bicp, which was completely prevented by T replacement, whilst tadalafil showed no influence on bicp in all groups. 3.3. Effect of testosterone and tadalafil on erectile response to ES Erectile response was quantified by ICP/MAP for each stimulation. Typical tracings at low stimulation frequency (4 Hz) are depicted in Fig. 4, while cumulative ICP/MAP results obtained in all the experimental groups are reported as sigmoid relationships in Fig. 5. In intact rats, ES of the cavernous nerve produced a frequency-dependent (1 32 Hz) ICP/MAP increase with ED 50 = 2.66 0.38 Hz and E max = 62.2

412 X.-h. Zhang et al. / European Urology 47 (2005) 409 416 2.7 mmhg. Castration depressed erectile function at all stimulation frequencies with unaffected ED 50 (ED 50 = 3.6 1 Hz), but significantly reduced maximal responsiveness (E max = 43.7 3.8 mmhg, p < 0.001), which was completely reinstated by T supplementation (ED 50 = 2.13 0.38 Hz, E max = 60.4 3 mmhg). Both modalities of tadalafil administration (acute or chronic) increased erectile response to ES (p < 0.001, vs. control) with comparable effectiveness (shared ED 50 = 1.29 0.17 Hz and shared E max = 72.7 2.5 mmhg). Conversely, in castrated rats, both acute and chronic tadalafil treatment was ineffective in increasing penile responsiveness to ES. Response to acute or chronic tadalafil administration was completely rescued in T-replaced castrated rats (shared ED 50 = 1.13 0.16 Hz and shared E max = 70.6 2.6 mmhg). These data indicate that surgical castration abolished and T replacement restored the tadalafil enhancement of erectile response to ES. Our observations suggest that the hampering of tadalafil pro-erectile capacity was due to a castration-induced PDE5 decrease. However, these results can not exclude an androgen modulation of NO production or a more general depressive effect of castration on the contractile apparatus. Fig. 2. (A) Quantitative detection of mrna for PDE5 in penile tissues from experimental rats by real-time RT-PCR. Intact animals (n = 5); castrated animals (n = 4); castrated plus T replaced animals (n = 3). Data are expressed as mean SEM of arbitrary unit (a.u.) calculated according to the comparative Ct method and using the beta-2 microglobulin gene for normalization. (B) Western blot analysis of PDE5 protein expression in rat CC after different treatments. Upper panel, densitometric evaluation of PDE5 protein expression calculated in percentage of band intensity over the control taken as 100%. Intact animals (n = 4); castrated animals (n = 4); castrated plus T replaced animals (n = 3). Data are mean SEM. Lower panel, a representative western blot. Human CC was run as positive control [2]. Molecular weight markers (kda) are indicated to the left of the blot. *p < 0.05, **p < 0.005, vs. castration. 3.4. Effect of castration and tadalafil on erectile response to SNP To rule out the androgen effect on NO formation, we studied the erectile response in terms of ICP/MAP by directly injecting SNP (0.06 6 nmoles) into the rat CC (Fig. 6). As shown in Fig. 6, increasing concentrations of SNP induced a dose-dependent increase in ICP/MAP in intact rats, which became statistically significant at 0.6 nmoles (p < 0.01). Tadalafil strongly enhanced SNP-induced penile erection at all the doses tested (0.06 6 nanomoles, p < 0.001 vs. control or any other groups) in intact Table 1 MAP and baseline ICP in experimental rat groups Groups MAP (mmhg) Baseline ICP (mmhg) Control (n = 10) 130.5 2.97 12.1 0.48 Castration (n = 10) 129.9 2.33 10.2 0.73 * Castration + testosterone (n = 5) 133 3.30 12.2 0.66 Control + tadalafil one dose (n = 12) 119.46 1.85 12.36 0.75 Castration + tadalafil one dose (n = 12) 132.32 3.83 10.5 0.60 * Castration + testosterone + tadalafil one dose (n = 6) 121.88 1.75 12.75 0.63 Control + tadalafil 2 weeks (n = 5) 121.2 5.11 12.5 0.79 Castration + tadalafil 2 weeks (n = 5) 131.15 3.55 10.3 0.92 * Castration + testosterone + tadalafil 2 weeks (n = 5) 123.65 3.59 12.5 0.45 Values are expressed as mean SEM. n = numbers of animals. * p < 0.05 vs. control.

X.-h. Zhang et al. / European Urology 47 (2005) 409 416 413 Fig. 3. Localization of cells immunolabelled with the anti-pde5 antibody in transversal sections of control rat penile tissue. Panels A C show an intense immunopositivity for PDE5 in the smooth muscle (black arrows) and endothelial (arrowheads) cells of the arteriolar wall and lacunar spaces of CC (A: magnification 50; B: magnification 50; C: magnification 200); (D, magnification 50): control section obtained by omitting the primary PDE5 antibody and counterstained with haematoxylin. Fig. 4. Representative ICP tracings. ICP increase was induced by electrical stimulation (2.5 V, 5 ms, 30 s) of the rat cavernous nerve at low stimulation frequency (4 Hz). A: control; B: castration; C: castration + T; D: control + tadalafil one dose; E: castration + tadalafil one dose; F: castration + T + tadalafil one dose; G: control + tadalafil 2 weeks; H: castration + tadalafil 2 weeks; I: castration + T + tadalafil 2 weeks.

414 X.-h. Zhang et al. / European Urology 47 (2005) 409 416 Fig. 5. Frequency-dependent erectile response. Erection was elicited by electrical stimulation (2.5 V, 5 ms, 30 s) of the rat cavernous nerve at varying stimulation frequency (1, 2, 4, 8, 16, 32 Hz). Erectile function was quantified by calculating max ICP/MAP 100 ratio for each stimulation. The baseline ICP/MAP (%) value (the beginning point) is also represented. Values are expressed as mean SEM. N =5 7 animals/experimental group. rats. Interestingly, the tadalafil amplification of SNP effect was inversely related to the concentration of SNP (r = 0.99, p < 0.05), as calculated by linear regression analysis. Injection of increasing concentrations of SNP in castrated rats produced a statistically significant increase (p < 0.001 vs. saline) in the ICP/MAP ratio similar to that in control, while tadalafil did not enhance them at all. 4. Discussion This study contributes to the concept that penile responsiveness to PDE5 inhibitors is androgen-dependent [5,12,13,21,22], just because the expression of their molecular target, i.e. PDE5, is positively regulated Fig. 6. Erectile response induced by intracavernous injection of sodium nitrosprusside (SNP). Erectile function evaluated by max ICP/MAP 100. Erectile response was elicited by a volume of 60 ml saline and cumulative doses (0.06 6 nmoles) of SNP intracavernously administered at 15-min period. Values are expressed as mean SEM in 5 separate rats for each group. **p < 0.01 vs. any other groups; 8p < 0.001 vs. saline. by T [2]. In the rat, as already found in rabbits and humans [2], PDE5 immunoreactivity is almost confined to the smooth muscle cells of blood vessels and cavernous spaces, where it regulates the amount of cgmp available for muscular relaxation. The finding of PDE5 staining in penile endothelial cells confirms a previous observation [2], although its functional significance is still unclear. Also the high abundance of PDE5 in VD is in keeping with that previous report [2]. Preliminary results obtained in our lab indicated that in VD PDE5 is androgen regulated and involved in cgmp metabolism and smooth muscle activity [23]. In rat penile extracts, the PDE5 gene and protein are highly expressed and strictly T regulated, being significantly reduced by surgical castration and completely restored by T supplementation. Hence, the present results confirm previous findings in other animal species [2]. However, in this study we show, for the first time, that the androgen milieu alteration can modulate the in vivo erectile response through changing the PDE5 availability. In fact, in two different experimental models, castration abolishes tadalafil responsiveness. In the first series of experiments we tested the capacity of tadalafil to increase the physiological erectile response elicited by ES of the cavernous nerve. This stimulation is supposed to increase vaso-relaxant neurotransmitter release, including NO, which finally results in an increased ICP and penile erection. We indeed found that in our experimental model, the increasing frequency of ES induces various degrees of ICP increase which are point by point enhanced by blocking cgmp degradation through tadalafil. Different modalities of tadalafil administration (daily

X.-h. Zhang et al. / European Urology 47 (2005) 409 416 415 delivery for 2 weeks or acute exposure) did not cause different amplification of ES-induced erection, suggesting that PDE5 activity is not altered by repeating the administration of the long-lasting inhibitor, tadalafil. However, molecular change in PDE5 expression under different modalities of tadalafil dosing was not further studied, because it is beyond the aim of this research. Interestingly, neither an acute nor a chronic exposure to tadalafil ameliorates castration-induced hypo-responsiveness to ES, while T supplementation completely rescues it. This indicates that T reinstates the molecular background underlying PDE5 inhibitor responsiveness. Although this finding fits well with our hypothesis of a positive effect of T on PDE5 expression, it does not rule out the possibility that this effect is mediated by an androgen-dependent increase in bio-availability of cgmp, the essential substrate for PDE5. Since it has been previously demonstrated that androgens positively modulate penile NO formation [8-11], we decided to bypass NO formation by directly injecting an NO donor, SNP, into CC. Even in this experimental condition, tadalafil did not amplified NOinduced ICP increase in castrated rats, while its proerectile effect was clearly evident in intact rats. This is most probably attributed to a down-regulation of PDE5, the molecular target of tadalafil, by androgen-deprivation. In fact, in the study of SNP, we first noticed that the responsiveness to NO was not different between castrated and intact rats. However, the response to ES was definitively reduced in the castrated group, most probably because NO formation is impaired in castrated rats [8-11] and the contractile machinery is somehow damaged [3 5,24]. The present results are in keeping with a previous finding of our group [2], showing that in hypogonadal rabbits and humans both PDE5 expression and activity are reduced and therefore there is a higher in vitro sensitivity of CC strips to NO donors, and a lack of enhancement by sildenafil. They also are in line with several clinical reports demonstrating a null effect of hypogonadism on penile erection [25 28], most probably because androgen reduction produces both decreased cgmp formation (effect on NO) and degradation (effect on PDE5). Finally, our results explain why penile erections are present in infants and eunuchs and even long-lasting in hypogonadal subjects [27] due to the relative deficiency of PDE5. 5. Conclusions Our findings support the hypothesis that testosterone positively regulates PDE5 inhibitor responsiveness in vivo, because it up-regulates the expression of the molecular target PDE5. This may have clinical significance: unresponsiveness to PDE5 inhibitors might underlie an androgen deficiency which should be carefully checked. Accordingly, recent studies indicate that a lack of clinical response to sildenafil in hypogonadal patients can be reversed by an adequate T supplementation [12,13]. Acknowledgements We thank Prof. M. Giorgi (Department of Experimental Medicine, University of L Aquila, Italy) for kindly providing the anti-pde5 rabbit polyclonal antibody, Prof. E.A. Jannini (Department of Experimental Medicine, University of L Aquila, Italy) for constructive criticisms during the course of the study, Dr. Nicola Mondaini (Department of Urology, University of Florence, Italy) for providing human corpus cavernosum samples. We also thank Mrs Mary Forrest for manuscript revision. This paper was supported by an unrestricted research grant from Ely-Lilly, Sesto Fiorentino, Florence, Italy and by Centro di Ricerca, Trasferimento e Alta Formazione MCIDNENT of the University of Florence. References [1] Maggi M, Filippi S, Ledda F, Magini A, Forti G. Erectile dysfunction: from biochemical pharmacology to advances in medical therapy. Eur J Endocrinol 2000;143:143 54. [2] Morelli A, Filippi S, Mancina R, Luconi M, Vignozzi L, Marini M, et al. Androgens regulate phosphodiesterase type 5 expression and functional activity in corpora cavernosa. Endocrinology 2004; 145:2253 63. [3] Shabsigh R. The effects of testosterone on the cavernous tissue and erectile function. World J Urol 1997;15:21 6. [4] Aversa A, Isidori AM, Greco EA, Giannetta E, Gianfrilli D, Spera E, et al. Hormonal supplementation and erectile dysfunction. Eur Urol 2004;45:535 8. [5] Traish AM, Park K, Dhir V, Kim NN, Moreland RB, Goldstein I. Effects of castration and androgen replacement on erectile function in a rabbit model. Endocrinology 1999;140:1861 8. [6] Mills TM, Lewis RW, Stopper VS. Androgenic maintenance of inflow and veno-occlusion during erection in the rat. Biol Reprod 1998;59:1413 8.

416 X.-h. Zhang et al. / European Urology 47 (2005) 409 416 [7] Reilly CM, Stopper VS, Mills TM. Androgens modulate the alphaadrenergic responsiveness of vascular smooth muscle in the corpus cavernosum. J Androl 1997;18:26 31. [8] Lugg JA, Rajfer J, Gonzalez-Cadavid NF. Dihydrotestosterone is the active androgen in the maintenance of nitric oxidemediated penile erection in the rat. Endocrinology 1995;136: 1495 501. [9] Reilly CM, Zamorano P, Stopper VS, Mills TM. Androgenic regulation of NO availability in rat penile erection. J Androl 1997;18:110 5. [10] Chamness SL, Ricker DD, Crone JK, Dembeck CL, Maguire MP, Burnett AL, et al. The effect of androgen on nitric oxide synthase in the male reproductive tract of the rat. Fertil Steril 1995;63:1101 7. [11] Baba K, Yajima M, Carrier S, Akkus E, Reman J, Nunes L, et al. Effect of testosterone on the number of NADPH diaphorase-stained nerve fibers in the rat corpus cavernosum abd dorsal nerve. Urology 2000;56:533 8. [12] Aversa A, Isidori AM, Spera G, Lenzi A, Fabbri A. Androgens improve cavernous vasodilation and response to sildenafil in patients with erectile dysfunction. Clin Endocrinol 2003;58:632 8. [13] Shabsigh R, Kaufman JM, Steidle C, Padma-Nathan H. Randomized study of testosterone gel as adjunctive therapy to sildenafil in hypogonadal men with erectile dysfunction who do not respond to sildenafil alone. J Urol 2004;172:658 63. [14] Callies F, Kollenkirchen U, von zur Muhlen C, Tomaszewski M, Beer S, Allolio B. Testosterone undecanoate: a useful tool for testosterone administration in rats. Exp Clin Endocrinol Diabetes 2003;111:203 8. [15] Corbin JD, Francis SH. Molecular biology and pharmacology of PDE- 5-inhibitor therapy for erectile dysfunction. J Androl 2003;24:S38 41. [16] Montorsi F, Verheyden B, Meuleman E, Jünemann KP, Moncada I, Valiquette L, et al. Long-term safety and tolerability of tadalafil in the treatment of erectile dysfunction. Eur Urol 2004;45:339 45. [17] Giordano D, De Stefano ME, Citro G, Modica A, Giorgi M. Expression of cgmp-binding cgmp-specific phosphodiesterase (PDE5) in mouse tissues and cell lines using an antibody against the enzyme amino-terminal domain. Biochim Biophys Acta 2001;1539:16 27. [18] D Amati G, di Gioia CR, Bologna M, Giordano D, Giorgi M, Dolci S, et al. Type 5 phosphodiesterase expression in the human vagina. Urology 2002;60:191 5. [19] Martinez-Pineiro L, Brock G, Trigo-Rocha F, Hsu GL, Lue TF, Tanagho EA. Rat model for the study of penile erection: pharmacologic and electrical-stimulation parameters. Eur Urol 1994;25: 62 70. [20] De Lean A, Munson PJ, Rodbard D. Simultaneous analysis of families of sigmoidal curves: applications to biossay, radioligand assay, and physiological dose response curves. Am J Physiol 1978;235:E97 E102. [21] Guay AT, Perez JB, Jacobson J, Newton RA. Efficacy and safety of sildenafil citrate for treatment of erectile dysfunction in a population with associated organic risk factors. J Androl 2001;22:793 7. [22] Yassin A, Diede HE, Saad F, Traish A. Combination therapy of tadalafil and testosterone in hypogonadal non-responders. Int J Impot Res 2003;S6:S27. [23] Mancina R, Filippi S, Morelli A, Vignozzi L, Salonia A, Montorsi F, et al. PDE5 inhibition in vas deferens improves the NO-donor relaxant effectabstracts of the 3rd European Congress of Andrology and the 16th Congress of the German Society of Andrology. Munster, Germany, 11 14 September 2004 Andrologia 2004;36:198 9. [24] Mills TM, Dai Y, Stopper VS, Lewis RW. Androgenic maintenance of the erectile response in the rat. Steroids 1999;64:605 9. [25] Bancroft J, Tennent G, Loucas K, Cass J. The control of deviant sexual behaviour by drugs. I. Behavioural changes following oestrogens and anti-androgens. Br J Psychiatry 1974;125:310 5. [26] Bancroft J, Wu FC. Changes in erectile responsiveness during androgen replacement therapy. Arch Sex Behav 1983;12:59 66. [27] Kwan M, Greenleaf WJ, Mann J, Crapo L, Davidson JM. The nature of androgen action on male sexuality: a combined laboratory-self-report study on hypogonadal men. J Clin Endocrinol Metab 1983;57:557 62. [28] Corona G, Mannucci E, Mansani R, Petrone L, Bartolini M, Giommi R, et al. Aging and pathogenesis of erectile dysfunction. Int J Impot Res 2004, May 6 [Epub ahead of print]. Editorial Comment F. Montorsi, Milan, Italy montorsi.francesco@hsr.it The study comes from a well-known group of basic science experts which demonstrates in an in vivo animal model the relationship between testosterone and the response to tadalafil, a long-acting PDE5 inhibitor. The clinical translation of this research highlights the importance of the assessment of the androgen profile in every patient complaining of erectile dysfunction. In the aging population where erectile dysfunction is most common, serum testosterone level tends to decrease. I believe that the question of importance deals with the impact of testosterone supplementation in patients with low-normal testosterone values and on treatment with a PDE5 inhibitor. The current study suggests that to optimize the response to a PDE5 inhibitor, a normal serum testosterone should first be re-established. Specifically designed studies in appropriate patient populations are thus urgently needed to clarify the best treatment strategy.