BJUI OBJECTIVE CONCLUSIONS

Transcription

1 21 THE AUTHORS; 21 Investigative Urology VIP AND ERECTILE FUNCTION IN CASTRATED RATS SHEN ET AL. BJUI Vasoactive intestinal polypeptide, an erectile neurotransmitter, improves erectile function more significantly in castrated rats than in normal rats Min-Guang Zhang 1, Zhou-Jun Shen 1 *, Cun-Ming Zhang 1, Wei Wu 2, Ping-Jin Gao 3, Shan-Wen Chen 4, Wen-Long Zhou 1 1 Department of Urology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, 2 Department of General Surgery, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, 3 Shanghai Institues for Biological Sciences, Chinese Academy of Sciences & Shanghai Jiaotong University School of Medicine, Institute of Health Sciences, 4 Department of Urology, 1 st Affiliated Hospital, School of Medicine, Zhejiang University, *Correspondence Author Accepted for publication 1 September 21 OBJECTIVE Androgen is essential for physiological erection. Vasoactive intestinal polypeptide (VIP) is an important erectile neurotransmitter. While previous studies demonstrated that VIP expression in the penis was androgen-independent, it remains controversial whether androgen has any effect on VIP-mediated erection. The present study aims to investigate the regulatory role of androgen in VIP-mediated erectile effect. MATERIALS AND METHODS Male SD rats were divided into a group, a group, and a with-testosterone-replacement group. Four weeks later, each group was subdivided into low and high-dose VIP subgroups and subjected to intracavernous injection of.5 and 2 μg VIP, respectively. Erectile function was tested by recording intracavernosal pressure (ICP) and mean arterial blood pressure (MAP) before and after VIP injection. The expressions of the VIP-receptor (VPAC2), G-protein stimulatory and inhibitory alpha subunits (Gs-α, Gi-α), and PDE3A in rat corpus cavernosum (CC) was qualified by real-time PCR and Western blot analysis. RESULTS Castration reduced erectile function while testosterone restored it. VIP improved erectile function in a dose-dependent manner. High-dose VIP significantly enhanced erectile function in castrated rats and there was no difference of ICP/MAP among three groups after injection of highdose VIP. Low-dose VIP also resulted in a higher improvement of erectile function in castrated rats, although the ICP/MAP was lower in these rats than in the other two groups. VPAC2 and Gs-α were up-regulated while Gi-α and PDE3A were down-regulated in CC of castrated rats. CONCLUSIONS VIP improves erectile function much more significantly in hypogonadal condition, mainly due to the higher expression of VPAC2, Gs-α, and lower expression of Gi-α and PDE3A in CC of castrated rats. Androgen may negatively regulate the erectile effect of VIP. KEYWORDS androgen, vasoactive intestinal polypeptide, erectile function,, signaling pathway INTRODUCTION Androgen plays an essential role in physiological erection and is critical for the development and growth of erectile organs. The NO/cGMP pathway, the principle erectile signalling pathway, has been proven to be strongly androgen-dependent, both in the expression and activity of nitric oxide synthase (NOS) and in the degradation of cgmp [1 3]. Phosphodiesterase type 5 (PDE5) inhibitors, the most predominant and effective drugs for erectile dysfunction (ED), have also been shown to be positively regulated by androgen [4]. As penile erections are present in infants and hypogonadal males [5] and 2 45% of medically or surgically castrated patients retain normal erections [6], however, it is clear that the relationship between androgen and erection is complicated. It also evokes the hypothesis that there may be an erectile signalling pathway differing from the NO/cGMP pathway which plays a major role in hypogonadism. Vasoactive intestinal peptide (VIP), a naturally occurring 28-amino acid neurotransmitter, was first isolated from the small intestines of pigs and found to have the potency of vascular dilation [7]. It was also found to be present in the male genital tract [8,9], acting as an important erectile neurotransmitter that regulates the relaxation of the smooth muscle cells of the corpus cavernosum (CC) via a signalling pathway different from the NO/cGMP pathway [1]. By binding and activating the VPAC2 receptor, a G-protein coupled receptor, VIP participates in the erectile process via the activation of the adenylyl cyclase (AC)/cAMP pathway. It was found that VIPergic nerves were depleted in 21 THE AUTHORS 21 doi:1.1111/j x x 1

2 SHEN ET AL. the corpus cavernosum of men with ED and diabetes-related ED [11]. We have previously reported that gene transfer of VIP into the cavernosum improved erectile response in diabetic rats [12]. Clinical studies also showed that Invicorp (Plethora Solutions, London, UK), a combination of VIP 25 μg and phentolamine mesylate 1 or 2 mg for intracavernosal injection (ICI), was effective in more than 8% of patients with ED, including those who were non-responsive to other therapies [13]. However, in normal men, ICI of VIP alone did not induce erection [14], indicating that VIP improves erectile response more significantly in pathological conditions than in normal conditions. Our previous study revealed that the expression of VIP at the mrna level was androgen-independent in rat CC [15]. Another study also found that the protein expression of VIP was androgen-independent in the penis [16]. However, up to now the relationship between androgen and VIP and the effect of VIP on erection is unclear, and the hypothesis that androgen has an effect on VIP-mediated erection remains controversial. The aim of the present study was to investigate the regulatory role of androgen in VIP-mediated erection and elucidate the underlying molecular mechanisms by examining the VIP-mediated signalling pathway. MATERIALS AND METHODS In all, 6 adult male Sprague Dawley rats, weighing 25 3 g, were obtained from Shanghai SLAC Laboratory Animal Co. Ltd. The rats were randomly divided into three groups: ; surgical ; and with-testosterone (T)-replacement (castrated but given testosterone undecanoate 1 mg/ kg per month by s.c. injection). After 4 weeks, each group was subdivided into low-dose and high-dose VIP subgroups by ICI of.5 μg and 2 μg VIP, respectively. After evaluation of erectile function, a blood sample was collected through an abdominal aorta cannula. The rats were killed by injecting air through the left carotid artery catheter and CC tissue samples were collected immediately. The connective tissues were removed and samples were frozen in liquid nitrogen and stored at 8 C until further analysis. The study was approved by the Animal Experiment Committee of Ruijin Hospital, Shanghai Jiaotong University School of Medicine. Erectile function was evaluated as previously reported [17]. Briefly, systemic mean arterial pressure (MAP) was continuously monitored via a PE-1 cannula placed in the left carotid artery. Each 22-gauge needle was inserted in the right and left penile crus for recording intracavernosal pressure (ICP) and drug administration. Both PE-1 tubing and the needle were connected to PE-5 tubing which was filled with heparin (25 IU/mL). ICP and MAP were recorded via pressure transducers connected to an MPA2 recorder (Shanghai Alcott Biotech. Co., Ltd., Shanghai, China). The cavernous nerve was exposed at the right side of the prostate and mounted on a bipolar platinum hook electrode connected to a JL-B electrical stimulator (Shanghai Jialong Teaching Instrument Manufacturer, Shanghai, China). Electrical stimulation (ES) parameters were 5 V, 25 Hz with a square-wave duration of 2 ms for 3 to 6 s, with a 5-min interval between subsequent stimulations. The erectile response elicited by electric stimulation or ICI of VIP was quantified by calculating maximal ICP/MAP. The ratio of ICP/MAP elicited by electric stimulation of the cavernous nerve before ICI of VIP, called the bratio, represented basal erectile function. ICI of.5 μg VIP decreased MAP by 2% and MAP was elevated to normal levels within 2 min. Electric stimulation of the cavernous nerve was performed at 5, 1 and 15 min after ICI of.5 μg VIP and the maximal ratio of ICP/MAP, called the pratio, represented erectile function after ICI of.5 μg VIP. ICI of 2 μg VIP decreased MAP by 3 4% and MAP was elevated to normal levels within 3 5 min. A spontaneous increase of ICP was found within 5 min after injection and reached its peak about 2 min later. ES of the cavernous nerve, therefore, was not performed in these subgroups and the pratio represented erectile function after ICI of 2 μg VIP. We then took pratio/bratio as the indicator of the in vivo responsiveness to VIP. Serum testosterone level was determined by ELISA using a commercial kit (Westang Biotech Co., Ltd., Shanghai, China) and an ELISA plate reader (Multiskan Ex Primary EIA V. 2.3, Thermo, Vantaa, Finland). Real-time quantitative reverse transcriptase (RT)-PCR was performed as previously described [2]. For each sample 5 ng of total RNA were reverse transcribed to cdna in a final volume of 1 μl using the PrimeScript RT reagent Kit (Takara, Kyoto, Japan). Real-time quantitative RT-PCR was performed using an ABI 75 FAST fluorescent quantitation system with the following thermal cycler conditions: 1 cycle of pre-incubation at 95 C for 1 s, 4 cycles of amplification at 95 C for 5 s and 6 C for 34 s. Each measurement was carried out in triplicate. The PCR mixture (2 μl final volume) consisted of 2 the final concentration of SYBR Green Realtime PCR Master Mix (Takara). Beta-actin was chosen as a reference gene. The primer sequences are shown below. The results were analysed using the comparative Ct method according to the manufacturer s instructions (Applied Biosystems, Foster City, CA, USA). VPAC2 forward 5 TGAGCAGCATCCACCCAGAAT3 reverse 5 ACCGCTGCAAGCTCTGTGATT3 Gs-α forward 5 AAGGAGCAACAGCGATGGAGT 3 reverse 5 GGCAATTCCAGCATGCAAAA3 Gi-α forward 5 CAACCCAGCAGGATGTTCTCA 3 reverse 5 GCCTCCCACGTCAAACATTT T 3 PDE3A forward 5 TGGAGTTGATGGCCCTGTATG 3 reverse 5 AACGGTCATTGTACAGCACGG 3 betaactin forward 5 CAGAAGGACTCCTACGTG3 reverse 5 GCTCGGTCAGGATCTTCATG3 Frozen CC samples were lysed as described previously [2]. Ten μg of protein extracts were separated on 1% denaturing SDSpolyacrylamide gel and transferred to polyvinylidene fluoride membranes, which were probed with monoclonal anti-vpac2 antibodies (Abcam, Cambridge, UK, 1:1), anti-pde3a antibodies (Abcam, 1:1), anti-gi-α antibodies (Cell Signaling Technology, Danvers, MA, USA, 1:1) and anti-beta-actin (Sigma, St Louis, MO, USA, 1:8), followed by peroxidase-conjugated IgG (1:3) incubation and final detection by electrochemiluminescence (ECL) plus reagents (GE Healthcare, Little Chalfont, UK). Densitometry was used to measure the expression of VPAC2, Gi-α and PDE3A relative to beta-actin with Bio-Rad Gel doc. Software (Bio-Rad Laboratories, Hercules, CA, USA). Data are expressed as mean ± SD. Comparisons between two groups were carried out using an unpaired t-test, among more than two groups using a one-way ANOVA and least significant difference (LSD) test in case of homogeneity of variance, and Dunnett s T3-test otherwise. Two-tailed P values <.5 were defined as significant. Data management and statistical calculations were conducted using the SPSS software for Windows version 16. (SPSS, Inc., Chicago, IL, USA) THE AUTHORS 21

3 VIP AND ERECTILE FUNCTION IN CASTRATED RATS TABLE 1 Weight, baseline ICP, MAP and erectile responses to ES of the cavernous nerve in the experimental rat groups Before ES of cavernous nerve After ES of cavernous nerve Group Weight, g bicp, mmhg bmap, mmhg ICP max, mmhg MAP max, mmhg bratio, % Control ± 27.9* 16.9 ± 5.4* 111. ± ± 15.* 17. ± ± 9.3* Castration ± ± ± ± ± ± Castration-with-T-replacement ± 27.5* 18. ± 5.6* 12.3 ± ± 19.* ± ± 1.4* Values are expressed as Mean ± SD. bicp, baseline ICP; bmap, baseline MAP; ICP max, maximal ICP after ES of cavernous nerve; MAP max, MAP corresponding to ICP max after ES of cavernous nerve. One-way ANOVA, LSD test: *P <.1 compared with group; P<.1 compared with -with-t-replacement group. TABLE 2 Erectile responses to ES of the cavernous nerve after ICI of.5 μg VIP in the experimental rat groups Before ICI Erectile response to ES of cavernous nerve after ICI.5 μg VIP Group bratio, % ICP max, mmhg MAP max, mmhg pratio, % pratio/bratio Control 68.6 ± 9.3* 9 ± 15.2* ± ± 9.4* 1.6 ±.14* Castration 36. ± 63.4 ± ± ± ±.27 Castration-with-T-replacement 64.6 ± 1.4* 83.8 ± 13.2** 123. ± ± ±.16* Values are expressed as mean ± SD. ICP max, maximal ICP after ES of cavernous nerve after ICI of.5 μg VIP; MAP max, MAP corresponding to ICP max after ICI of.5 μg VIP. One-way ANOVA, LSD test: *P<.1 compared with group; **P<.5 compared with group. FIG. 1. Erectile response before and after intracavernous injection of different doses of VIP in normal rats. The index of erectile response were determined as described in Materials and Methods. ICP: intracavernous pressure; MAP: mean arterial pressure; bratio: baseline-icp/map; pratio: post-ici ICP/MAP. One-way ANOVA, LSD test. *P <.5 compared with bratio; **P <.1 compared with pratio. ICP/MAP, % RESULTS saline VIP dose, μg The group had the lowest serum testosterone concentration (ng/ml) compared with the group and bratio, % pratio, % -with-t-replacement group (.467 ±.124 vs ±.484 and.994 ±.473; P =. and P =.1, respectively). T replacement recovered T concentration but the concentration was still lower than that of the group (P =.32). As shown in Table 1, significantly reduced the weight, baseline ICP, and bratio of the rats in this group compared with the rats, and T replacement counteracted these effects. There was no difference in MAP between the and groups, but T replacement after elevated MAP by 1 15 mmhg ( group and group vs -with-treplacement group, P =.7 and P =., respectively). Intracavernous injection of VIP has been shown to induce erection in rats, but the doses of VIP ranged from 1( 8) to 4 1( 5)M [18,19]. Therefore, we needed to determine the proper dose of VIP and examine the erectile response to different doses of VIP (.5 1 μg). As shown in Fig. 1, VIP induced a dose-dependent increase in ICP/MAP elicited by ES of the cavernous nerve 5 15 min after injection. Moreover, the induced increase became statistically significant when VIP was.5 μg. Meanwhile, 2 μg of VIP induced a spontaneous increase in ICP/MAP in all rats, which started 5 min after injection, reached its peak 2 min after injection, and decreased gradually over the next 1 min. We 21 THE AUTHORS 21 3

4 SHEN ET AL. therefore took.5 μg and 2 μg as the low and high dose of VIP, respectively. Next we compared the ICP/MAP before and after VIP injection to evaluate the in vivo responsiveness of rats to VIP. As shown in Tables 2 and 3 and Fig. 2, the in vivo responsiveness (pratio/bratio) was significantly higher in the group, both in low-dose VIP ( group vs group and -with-treplacement group, 4 ±.27 vs 1.6 ±.14 and 1.6 ±.16, P =.1 and P =.1, respectively) and high-dose VIP subgroups ( group vs group and -with-t-replacement group, ±.36 vs 1.19 ±.14 and 1.12 ±.16, P =. and P =., respectively). Although in the low-dose VIP subgroups, the pratio was lower in the group than in other groups ( group vs group and -with-t-replacement group, 55.6 ± 9.8 vs 73. ± 9.4 and 68.5 ± 1.6, P =.8 and P =.38, respectively); in the high-dose VIP subgroups, there was no difference in pratio among the three groups ( group vs group and -with-t-replacement group, 73.7 ± 12.9 vs 81.6 ± 9.6 and 72.2 ± 1.3, P =.336). TABLE 3 Spontaneous erectile responses to ICI of 2 μg VIP in the experimental rat groups Before ICI Spontaneous erectile response to ICI of 2 μg VIP Group bratio, % ICP max, mmhg MAP max, mmhg pratio, % pratio/bratio Control 68.6 ± 9.3* 12.6 ± ± ± ±.14* Castration 36. ± 76. ± ± ± 12.9 ±.36 Castration+T 64.6 ± 1.4* 91.1 ± ± ± ±.16* Values are expressed as mean ± SD; ICP max, maximal ICP after ICI of 2 μg VIP, MAP max, MAP corresponding to ICP max after ICI of 2 μg VIP. One-way ANOVA, LSD test: *P<.1 compared with group. FIG. 2. Erectile response of experimental rats after intracavernous injection of VIP. The index of erectile response was determined as described in Materials and Methods. ICP: intracavernous pressure; MAP: mean arterial pressure; bratio: baseline-icp/map; pratio: post-ici ICP/MAP, the maximal ratio of ICP/MAP after ICI of VIP; pr/br: pratio/bratio. One-way ANOVA, LSD test. *P <.5 compared with group; **P <.1 compared with group. pratio, % a P =.336 Control Castration Castration + T As VPAC2, Gs-α, Gi-α, and PDE3A are components of the VIP/cAMP erectile pathway that mediates VIP response, we next examined their expression in differently treated rats. Compared with the, the mrna level of VPAC2 was higher in the and -with-treplacement groups ( group vs and -with-treplacement groups, 1.39 vs and 2.138, P =. and P =., respectively; Fig. 3, panel a). Moreover, the mrna level of VPAC2 was higher in the group than in the -with-t-replacement group (P =.9). This differential pattern of VPAC2 expression was also revealed at the protein level by Western blot analysis (Fig. 4, panels a and b). We further found that resulted in increased expression of Gs-α at the mrna level ( group vs and -with-t-replacement groups, vs 1.13 and.868, P =.1 and P =., respectively) (Fig. 3, panel b). Protein expression of Gs-α was not examined because the anti-gs-α antibody was pratio/bratio, % b VIP.5 μg VIP.5 μg Subgroup Subgroup VIP 2 μg VIP 2 μg Control Castration Castration + T 4 21 THE AUTHORS 21

5 VIP AND ERECTILE FUNCTION IN CASTRATED RATS FIG. 3. mrna expressions of VPAC2, Gs-α, Gi-α and PDE3A in corpus cavernosum of experimental rats. Panel a, b, c, d: mrna levels for VPAC2, Gs-α, Gi-α and PDE3A gene in penile tissues were quantitated by real-time RT-PCR. Values were calculated according to the comparative Ct method and using beta-actin as reference gene. One-way ANOVA, LSD test. **P <.1 compared with group. a VPAC2 mrna c Gi-alpha mrna unavailable. We also found that reduced the expression of Gi-α at the mrna level, while T replacement significantly increased its expression ( group vs and -with-t-replacement groups,.431 vs 1.28 and 3.14, P =. and P =., respectively) (Fig. 3, panel c). A similar pattern of Gi-α expression at the protein level was revealed by Western blot analysis (Fig. 4, panels c and d). Castration also resulted in decreased expression of PDE3A at the mrna level ( group vs and -with-t-replacement groups,.324 vs 1.34 and 12, P =.4 and P =., respectively) (Fig. 3, panel d) and protein level (Fig. 4, panels e and f). DISCUSSION + T + T In the present study, we showed that VIP significantly improved erectile function in castrated rats. Furthermore, we found that the levels of key components of the VIP/ camp erectile pathway, VPAC2, Gs-α, Gi-α, and PDE3A, were related to serum T concentration in rat CC; levels of VPAC2 and Gs-α were higher while those of Gi-α and PDE3A were lower in castrated rats, suggesting that the changes in expression b Gs-alpha mrna d PDE3A mrna T + T pattern may mediate the enhanced effect of VIP on erection. Androgen plays a critical role in physiological erection and is essential for the development, growth, and maintenance of penile-tissue function [1]. The NO/cGMP pathway, the principle erectile pathway, has been shown to be highly dependent on androgen [2,4,2]. However, physiological erection is a complicated neurovascular process regulated by multiple factors. It has been confirmed that there was no correlation between total T levels and ED, or with the severity of ED [21,22]. Although T-replacement therapy (TRT) has attracted increasing attention and has been used in prostate patients with ED after radical prostatectomy [23], meta-analysis has revealed that the efficacy of hormonal treatments in men with ED was inconclusive [24], and the American College of Physicians does not recommend routine use of hormonal blood tests or TRT for patients with ED [25]. It is therefore worth investigating the relationship between androgen and other erectile signalling pathways. Vasoactive intestinal peptide is an important erectile neurotransmitter which was first isolated from the small intestines of pigs and found to have the potency of vascular dilation [7]. The injection of VIP induces erection in several animal species, including humans [26,27]. The VIP-induced vasodilation, which results in tumescence, is mediated by camp upon the activation of AC after the binding of VIP to its receptor [1]. Interestingly several studies, as well as the present study, have shown that VIP expression in CC is androgenindependent, both in rats and humans [15,16]. In the present study, we aimed to investigate further the relationship between androgen and the effect of VIP on erection. By surgical and T replacement, we established rat models with different T concentrations and basal erectile function (Table 1). To evaluate the in vivo erectile responsiveness to VIP, we used the ratio of post-ici ICP/MAP to baseline-icp/map (pratio/bratio) as the indicator of the in vivo erectile responsiveness to VIP and found that this responsiveness was significantly higher in the group than in the other two groups, both in lowdose and high-dose VIP. Furthermore, highdose VIP restored the reduced erectile function in castrated rats to the same level as that in the and castrated-with-treplacement rats. This suggests that the effect of VIP on erection is independent of androgen, at least in rats. Based on a previous report that ICI of 25 μg VIP was safe and effective in the treatment of ED and that low doses of VIP (up to 1.2 μg) could not induce useful erection in impotent men [13], we tested the effective dose of VIP in rats in the dose range from.5 μg to 1 μg. The dose of VIP used in the present study was consistent with (.5 μg), or higher than (2 μg), that used in the human body according to body weight, because of species differences. As an important erectile neurotransmitter, VIP regulates the relaxation of cavernous smooth muscle cells through the VIP/VPAC2/AC/ camp/protein kinase A erectile signalling pathway. Binding of VIP to the VPAC2 receptor activates AC, promoting the production of camp and activation of campdependent protein kinase A, which results in the relaxation of smooth muscle cells. Gs-α and Gi-α are two distinct G-proteins which mediate the stimulation and inhibition of AC and thus are involved in the regulation of cavernous smooth muscle tone [28]. camp is further regulated by its degradation by camp- 21 THE AUTHORS 21 5

6 SHEN ET AL. binding PDEs (PDE3A and PDE4), and PDE3A is highly expressed in CC [29 32]. Thus, the relaxation effect of VIP on cavernous smooth muscle depends on VPAC2, G-proteins, and PDE3A. To explain why VIP improves erectile function more significantly in castrated rats than in normal rats, we examined the expression of VPAC2, Gs-α, Gi-α, and PDE3A in CC by quantitative real-time PCR and Western blot. As shown in Fig. 3, VPAC2 and Gs-α were upregulated, while Gi-α and PDE3A were downregulated in castrated rats, and T replacement reversed the changes in expression. Overexpression of VPAC2 is consistent with higher sensitivity to VIP in castrated rats, while changed expression of Gs-α and Gi-α proteins is associated with reduction of cavernous smooth muscle tone. The reduced expression of PDE3A may decrease the degradation of camp, thus indirectly enhancing the effect of VIP. Taken together, we speculate that the reason that the effect of VIP on erection was more significant in castrated rats could be because of higher expression of VPAC2 and Gs-α, as well as lower expression of Gi-α and PDE3A in CC of these rats. To our knowledge, the present study establishes for the first time that androgen may negatively regulate an important erectile pathway, the VIP/cAMP pathway, to modulate erectile response. Therefore, we propose that in the process of penile erection, the VIP/cAMP signalling pathway serves as a complementary mechanism to the main NO/cGMP pathway, and VIP could play a dominant role in erection in hypogonadal conditions when the NO/cGMP pathway is inhibited by androgen deprivation. Our findings may have important clinical significance for patients with prostate cancer who have ED and are unresponsive to PDE5 inhibitors because of androgen deficiency, and who are unsuitable for androgen replacement. Our study may also be helpful for the treatment of late onset hypogonadal-related erectile dysfunction. ACKNOWLEDGEMENTS This work was supported by the National Natural Science Foundation of China ( ), and also partly supported by the Shanghai Science and Technology FIG. 4. Western blot analysis of VPAC2, Gi-α and PDE3A expression in corpus cavernosum of experimental rats. Molecular weight markers (kda) were indicated to the left of the blot and beta-actin was used for normalization. Relative band intensities of VPAC2, Gi-α and PDE3A against beta-actin in group are set as 1, and the corresponding values in and with T replacement group are compared by ANOVA, LSD test. Values for 4 samples in each group are expressed as the ratio ± SD. *P <.5 compared with contral group; **P <.1 compared with contral group; P <.1 compared with group. a VPAC2 relative expression c Gi-alpha relative expression e PDE3A relative expression Commission of major sub-topics (9DJ1442). We thank Drs Wei-Qing Han, Yong-Jie Wu and Wen-Quan Niu (Shanghai Key Laboratory of Vascular Biology and Division of Hypertension, Shanghai, China) for their expert technical assistance. We thank Medjaden Bioscience Ltd for assisting in the preparation of this manuscript. CONFLICT OF INTEREST None declared. REFERENCES + T 1 Traish AM, Guay AT. Are androgens critical for penile erections in humans? + T + T b 47 KDa VPAC2 42 KDa d + T + T beta-actin 4 KDa Gi-alpha 42 KDa f + T beta-actin 75 KDa PDE3A 42 KDa beta-actin Examining the clinical and preclinical evidence. J Sex Med 26; 3: Morelli A, Filippi S, Mancina R et al. Androgens regulate phosphodiesterase type 5 expression and functional activity in corpora cavernosa. Endocrinology 24; 145: Marin R, Escrig A, Abreu P, Mas M. Androgen-dependent nitric oxide release in rat penis correlates with levels of constitutive nitric oxide synthase isoenzymes. Biol Reprod 1999; 61: Zhang XH, Morelli A, Luconi M et al. Testosterone regulates PDE5 expression and in vivo responsiveness to tadalafil in 6 21 THE AUTHORS 21

7 VIP AND ERECTILE FUNCTION IN CASTRATED RATS rat corpus cavernosum. Eur Urol 25; 47: 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: Rousseau L, Dupont A, Labrie F, Couture M. Sexuality changes in prostate cancer patients receiving antihormonal therapy combining the antiandrogen flutamide with medical (LHRH agonist) or surgical. Arch Sex Behav 1988; 17: Said SI, Mutt V. Polypeptide with broad biological activity: isolation from small intestine. Science 197; 169: Larsen JJ, Ottesen B, Fahrenkrug J, Fahrenkrug L. Vasoactive intestinal polypeptide (VIP) in the male genitourinary tract: concentration and motor effect. Invest Urol 1981; 19: Willis EA, Ottesen B, Wagner G, Sundler F, Fahrenkrug J. Vasoactive intestinal polypeptide (VIP) as a putative neurotransmitter in penile erection. Life Sci 1983; 33: Lin CS, Lin G, Lue TF. Cyclic nucleotide signaling in cavernous smooth muscle. J Sex Med 25; 2: Gu J, Polak JM, Lazarides M et al. Decrease of vasoactive intestinal polypeptide (VIP) in the penises from impotent men. Lancet 1984; 2: Shen ZJ, Wang H, Lu YL, Zhou XL, Chen SW, Chen ZD. Gene transfer of vasoactive intestinal polypeptide into the penis improves erectile response in the diabetic rat. BJU Int 25; 95: Dinsmore WW, Wyllie MG. Vasoactive intestinal polypeptide/phentolamine for intracavernosal injection in erectile dysfunction. BJU Int 28; 12: Kiely EA, Bloom SR, Williams G. Penile response to intracavernosal vasoactive intestinal polypeptide alone and in combination with other vasoactive agents. Br J Urol 1989; 64: Shen ZJ, Lu YL, Chen ZD, Chen F, Chen Z. Effects of androgen and ageing on gene expression of vasoactive intestinal polypeptide in rat corpus cavernosum. BJU Int 2; 86: Cormio L, Gesualdo L, Maiorano E et al. Vasoactive intestinal polypeptide (VIP) is not an androgen-dependent neuromediator of penile erection. Int J Impot Res 25; 17: Gur S, Sikka SC, Chandra S et al. Alfuzosin attenuates erectile dysfunction in rats with partial bladder outlet obstruction. BJU Int 28; 12: Maher E, Bachoo M, Elabbady AA et al. Vasoactive intestinal peptide and impotence in experimental diabetes mellitus. Br J Urol 1996; 77: Suh JK, Mun KH, Cho CK, Shin HC, Kim YS, Park TC. Effect of vasoactive intestinal peptide and acetylcholine on penile erection in the rat in vivo. Int J Impot Res 1995; 7: Traish AM, Goldstein I, Kim NN. Testosterone and erectile function: from basic research to a new clinical paradigm for managing men with androgen insufficiency and erectile dysfunction. Eur Urol 27; 52: Rhoden EL, Telöken C, Sogari PR, Souto CA. The relationship of serum testosterone to erectile function in normal aging men. J Urol 22; 167: Rhoden EL, Telöken C, Mafessoni R, Souto CA. Is there any relation between serum levels of total testosterone and the severity of erectile dysfunction? Int J Impot Res 22; 14: Khera M. Androgens and erectile function: a case for early androgen use in postprostatectomy hypogonadal men. J Sex Med 29; 6 (Suppl. 3): Tsertsvadze A, Fink HA, Yazdi F et al. Oral phosphodiesterase-5 inhibitors and hormonal treatments for erectile dysfunction: a systematic review and meta-analysis. Ann Intern Med 29; 151: Qaseem A, Snow V, Denberg TD et al. Hormonal testing and pharmacologic treatment of erectile dysfunction: a clinical practice guideline from the American College of Physicians. Ann Intern Med 29; 151: Ottesen B, Wagner G, Virag R, Fahrenkrug J. Penile erection: possible role for vasoactive intestinal polypeptide as a neurotransmitter. Br Med J (Clin Res Ed) 1984; 288: Guidone G, Muller D, Vogt K, Mukhopadhyay AK. Characterization of VIP and PACAP receptors in cultured rat penis corpus cavernosum smooth muscle cells and their interaction with guanylate cyclase-b receptors. Regul Pept 22; 18: Anand-Srivastava MB, Di Fusco F. Redox modulation of Gi protein expression and adenylyl cyclase signaling: role of nitric oxide. Antioxid Redox Signal 24; 6: Küthe A, Eckel H, Stief CG et al. Molecular biological characterization of phosphodiesterase 3A from human corpus cavernosum. Chem Biol Interact 1999; : Küthe A, Montorsi F, Andersson KE, Stief CG. Phosphodiesterase inhibitors for the treatment of erectile dysfunction. Curr Opin Investig Drugs 22; 3: Uckert S, Hedlund P, Waldkirch E et al. Interactions between cgmp- and camppathways are involved in the regulation of penile smooth muscle tone. World J Urol 24; 22: Küthe A, Magert H, Uckert S, Forssmann WG, Stief CG, Jonas U. Gene expression of the phosphodiesterases 3A and 5A in human corpus cavernosum penis. Eur Urol 2; 38: Correspondence: Zhou-Jun Shen, Department of Urology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, 197 2nd Ruijin Road, Shanghai, 225, PR China. Tel: ; Fax: shenzj6@sina.com, zhangmg197@ hotmail.com Abbreviations: NOS, nitric oxide synthase; PDE5, phosphodiesterase type 5, ED, erectile dysfunction; VIP, vasoactive intestinal peptide; CC, corpus cavernosum; AC, adenylyl cyclase; ICI, intracavernosal injection; MAP, mean arterial pressure; ICP, intracavernosal pressure; ES, electrical stimulation; RT, reverse transcriptase. 21 THE AUTHORS 21 7