Diabetes mellitus, which affects about 171 million

Transcription

1 Journal of Andrology, Vol. 29, No. 5, September/October 2008 Copyright E American Society of Andrology Effect of Caffeine on Erectile Function via Up-Regulating Cavernous Cyclic Guanosine Monophosphate in Diabetic Rats RONG YANG,* JIULING WANG,* YUN CHEN,* ZEYU SUN,* RUN WANG,{ AND YUTIAN DAI* From the *Department of Urology, Affiliated Drum Tower Hospital, Nanjing University, School of Medicine, Nanjing, Jiangsu, China; and the ÀDepartment of Urology, University of Texas Medical School at Houston and MD Anderson Cancer Center, Houston, Texas. ABSTRACT: Erectile dysfunction (ED) is a common complication of diabetes mellitus. Phosphodiesterase-5 (PDE5) inhibitors, which inhibit the breakdown of intracellular cyclic guanosine monophosphate (cgmp), are used to treat diabetic ED. Caffeine, a nonselective PDE inhibitor used in our daily diet, is controversial regarding its effect on erectile function. To investigate the effect of caffeine on erectile function in diabetic rat models and explore the mechanism, male Sprague-Dawley rats were injected with streptozotocin to induce diabetes mellitus. The rats with blood glucose levels above 300 mg/dl were selected for the study. The rats were divided into 4 groups: group A (normal control rats), group B (diabetic rats treated with normal saline), group C (diabetic rats treated with caffeine, 10 mg/kg per day), and group D (diabetic rats treated with caffeine, 20 mg/kg per day). After 8 weeks of treatment, intracavernous pressure (ICP) was measured to assess erectile function. The radioimmunoassay was used to evaluate the level of cgmp in the cavernosum. The ICP and the cavernous cgmp decreased significantly in the diabetic rats compared with normal controls. An 8-week administration of caffeine at the given dosages increased the ICP and cavernous cgmp in diabetic rats. Caffeine consumption improved the erectile function of diabetic rats by up-regulating cavernous cgmp. Key words: Diabetes mellitus, erectile dysfunction, NO/cGMP pathway. J Androl 2008;29: Diabetes mellitus, which affects about 171 million people around the world, can cause severe health problems (Rathmann and Giani, 2004). Erectile dysfunction (ED) is one of the most common complications in male patients with diabetes mellitus (Blanco et al, 1990). The prevalence of ED is approximately 35% 75% in diabetic people, 3 times more than that in nondiabetic people (Richardson and Vinik, 2002). Pathophysiological mechanisms underlying diabetesassociated ED are largely due to endothelial dysfunction, which could lead to the inability of the endothelium to produce vasorelaxing messengers and to maintain vasodilation and vascular homeostasis (Musicki and Burnett, 2007). Nitric oxide (NO) is one of the most important vasorelaxing messengers in the process of erection. It mediates the formation of cyclic guanosine monophosphate (cgmp), which is degraded by phosphodiesterase-5 (PDE5; Bredt and Snyder, Correspondence to: Yutian Dai, Department of Urology, Gulou Hospital, Nanjing, Jiangsu , China ( ytdai@hotmail. com), and Run Wang, Department of Urology, University of Texas Medical School at Houston and MD Anderson Cancer Center, Houston, TX ( run.wang@uth.tmc.edu). Received for publication December 16, 2007; accepted for publication April 9, DOI: /jandrol ). The increased cgmp level will lead to the activation of protein kinases A and G, which subsequently phosphorylate the actin-myosin system proteins and Ca 2+ channels located within the cellular membrane and sarcoplasmatic reticulum membrane. These processes will cause a reduction of free cytoplasmatic Ca 2+ and finally the relaxation of smooth muscle, thereby triggering an erection. Therefore, the NO-cGMP pathway plays a key role in penile erection. Previous studies have demonstrated that in diabetes mellitus, the NOcGMP pathway is adversely affected, which correspondingly impairs the relaxation of corpus cavernosum (Saenz de Tejada et al, 1989). One acceptable therapeutic strategy for the diabetic ED is to preserve or enhance the NO-cGMP pathway. Now, the most effective oral drugs are the inhibitors of PDE5, which inhibit the breakdown of intracellular cgmp and subsequently help achieve and maintain erection. Caffeine is a common component in our everyday diet. It is a nonselective inhibitor of phosphodiesterases (Corbin and Francis, 2002), which can elevate the level of intracellular cgmp and cyclic adenosine monophosphate (camp). Caffeine could also decrease the influx of Ca 2+ and reduce the Ca 2+ sensitivity of contractile apparatus in some smooth muscles (Watanabe et al, 1992). These pharmacological effects lead to the 586

2 Yang et al N Caffeine and Erectile Function in Diabetes 587 relaxation of cavernous smooth muscles and decrease the contractile effects of cavernous muscle strips induced by high K + solution, noradrenaline, and transmural electrical stimulation (Adebiyi and Adaikan, 2004). Recent studies suggest that long-term consumption of caffeine could improve the glucose homeostasis and the endothelial function in diabetic patients (Lopez-Garcia et al, 2006; Park et al, 2007). Thus, caffeine might enhance erectile function in these patients. However, current available epidemiology surveys were inconclusive regarding the effect of caffeine on erectile function. Some studies showed that caffeine had a positive effect on erectile function (Diokno et al, 1990; Akkus et al, 2002) and others indicated that caffeine had no effect in the general population (Berrada et al, 2003; Nicolosi et al, 2003). No epidemiology study is available to evaluate the effect of caffeine on ED in diabetic patients. To our knowledge, the effect of caffeine on ED in experimentally induced diabetic rats also has not been reported. This study was designed to investigate the effect of caffeine on erectile function and to explore the effect of caffeine on cavernosus NO/cGMP pathway in diabetic rats. Methods Animal and Treatment Forty-eight male Sprague-Dawley rats weighing g were obtained from Shanghai Slac Laboratory Animal Co Ltd (Shanghai, China). Forty of them were selected randomly and injected intraperitoneally with freshly prepared streptozotocin (STZ; 65 mg/kg). The remainder were treated with vehicle (0.1 mol/l citrate-phosphate buffer, ph 4.5) as normal controls (group A). All rats were kept in a temperature-controlled, air conditioned animal house with a 12- h light-dark cycle and were given free access to food and water. Procedures were performed according to the recommendations of the institutional animal care committee. Seventy-two hours after the rats were injected with STZ, the blood glucose level in each rat was monitored at regular intervals throughout the study and immediately before sacrifice. Of the 40 STZ-induced rats, those having blood glucose levels higher than 300 mg/dl (16.6 mmol/l) were selected for the study. They were randomly divided into 3 groups and given different treatment regimens intragastrically for 8 weeks: group B (n 5 8), diabetic controls, diabetic rats treated with normal saline; group C (n 5 9), low caffeine dose group, diabetic rats given caffeine with 10 mg/kg per day; group D (n 5 9), moderate caffeine dose group, diabetic rats given caffeine with 20 mg/kg per day. After 8 weeks of treatment, intracavernous pressure (ICP), mean systemic arterial pressure (MAP), and the ratio of peak ICP to MAP (ICP/MAP ratio) were measured. The cavernosus cgmp was assessed by radioimmunoassay. Erectile Function Evaluation The measurement of ICP, calculation of the ICP/MAP ratio, and method of cavernous nerve stimulation were described previously (El-Sakka et al, 2001; Chen et al, 2007). Under anesthesia with urethane (1.6 g/ kg intraperitoneally), each rat was placed on a heating pad to maintain body temperature. A PE-50 tube was inserted into the left carotid artery via a midline neck incision to record the MAP. Through a lower abdominal midline incision, the major pelvic ganglions, posterolateral to the prostate, were located. The pelvic nerves and the cavernous nerves were exposed and identified. The skin overlying the penis was incised and both penile crura were exposed by removing part of the overlying ischiocavernous muscle. A 23G needle connected to a PE-50 tube with heparinized saline (250 IU/mL) was carefully inserted into the right crus. The other end of the PE-50 tube was connected to a pressure sensor for measuring the intracavernous pressure. Electrostimulation was performed with a stainless steel bipolar hook electrode (each pole 0.2 mm in diameter; the 2 poles separated by 1 mm). The stimulus parameters were: current 5 ma, frequency 20 Hz, pulse width 0.2 ms, and duration 50 s. Pressure was measured and recorded with a Windows computer programmed multiplying channel physiograph and analyzed with RM6240B/C multichannel biosignal collection processing system (Chengdu Implement Company, Chengdu, China). Determine the Level of Cavernous cgmp The rat penis was amputated at the scarification. The corpora cavernosa weighing,50 mg were immediately frozen in liquid nitrogen after the corpus spongiosum was removed. The frozen sample was ground into powder and homogenized in 2 ml of cold sodium acetate buffer (50 mm, ph 4.75). cgmp was extracted by the following method. Each homogenized penis sample was mixed with 2 ml dehydrated alcohol followed by centrifugation at g for 15 min at 4uC, and the supernatant was collected. The remaining sediment was washed again with 75% ethanol and centrifuged, and the ethanolic phase was kept. These 2 supernatants were mixed and dried at 60uC (Seidler et al, 2002). Following redissolution, aliquots of the samples were assayed for cgmp by radioimmunoassay with an RIA kit (Shanghai University of Traditional Chinese Medicine, Shanghai, China) as per the manufacturer s manual. Each sample was tested in duplicate. The protein content of the pellets remaining after centrifugation was measured with BCA Protein Assay Kit (Pierce, Rockford, Illinois). The results were expressed in picomoles of cgmp per microgram protein (mean 6 SD). Statistical Assay All data were expressed as mean 6 SD. To compare among multiple groups, 1-way analysis of variance (ANOVA) was used. The least significant difference test (LSD) was employed for comparison between each 2 groups. All statistical analyses were processed through the Statistical Package for the Social Sciences (SPSS version 13.0 for Windows; SPSS Inc, Chicago, Illinois). P,.05 was considered statistically significant. Results General Data Of 40 STZ-induced rats, 26 had blood glucose higher than 300 mg/dl with significantly increased food and water intakes, hyperuresis, and weight loss compared with the normal controls. During this period, 2 rats in group B and 1 in group D died of complications (anorexia and infection). Body weight and blood glucose level of the rats are shown in Table 1.

3 588 Journal of Andrology N September ÙOctober 2008 Table 1. Rat body weight and blood glucose (mean 6 SD) Group A Group B Group C Group D Initial Body weight, g Blood glucose, mg/dl a a a 8 weeks Body weight, g a a a Blood glucose, mg/dl a a a a P,.01 compared with group A. ICP and ICP/MAP Ratio The results of ICP and ICP/MAP ratio with and without the administration of caffeine are shown in the Table 2 and Figure 1. There were no statistical differences of the baseline levels of ICP among all groups. With electrostimulation, significantly lower peak ICP values were observed in group B compared with group A ( vs mm Hg, P,.01). The peak ICP/MAP ratio was also significantly lower in group B compared with group A ( vs , P,.01). In caffeine-treated rats, the peak ICP values were partially restored ( mm Hg for group C and mm Hg for group D) compared with normal saline-treated rats ( mm Hg for group B, P,.01). However, peak ICP values in caffeine-treated rats did not return to the normal level. The peak ICP/MAP ratio was also greater in caffeine-treated groups ( in group C and in group D) compared with the normal saline-treated rats ( , P,.01) but was still less than that in the normal control rats. There were no significant differences of ICP and ICP/ MAP ratio identified between 10 and 20 mg/kg per day caffeine-treated groups (group C vs group D, P..05). Cavernous cgmp Table 3 showed the results of cavernous cgmp in each group. The cavernous cgmp decreased significantly in diabetic rats ( pmol/mg protein for group B, pmol/mg protein for group C, and pmol/mg protein for group D) when compared with normal control rats ( pmol/mg protein for group A, P,.01). However, when diabetic rats were treated with caffeine (group C and group D), the cgmp levels increased significantly compared with the diabetic rats treated with normal saline (group B, P,.05), even though the cgmp in caffeine-treated groups did not return to the normal level. No statistical differences of cgmp level were seen between 10 and 20 mg/kg per day caffeine-treated groups (Figure 2). Discussion It is well known that penile erection depends on decreased penile vascular resistance, which subsequently results in increased penile blood flow. The action consists of the relaxation of the penile helicine arteries and the cavernous smooth muscle that lines the cavernosal spaces (Andersson and Wagner, 1995). Research has demonstrated that reduced vasorelaxation or increased vasoconstriction under diabetic conditions accounts for the development of ED. The NO-cGMP pathway plays an important role in initiating and maintaining an erection (Kandeel et al, 2001; Seidler et al, 2002). Previous studies have shown significant impairment of vasorelaxant effect of the NOcGMP pathway in the diabetic condition, resulting in reduced erectile capability (Saenz de Tejada et al, 1989). In the STZ-induced diabetic rat models, a reduction of the ICP responses to nerve stimulation was observed 2 3 months after the STZ treatment (Rehman et al, 1997). In this study, we confirmed the measurable reductions of the peak ICP, the ICP/MAP ratio, and the level of cavernous cgmp at the 8-week mark in STZ-induced diabetic rats. These findings are similar to previously published results (Park et al, 2004; Bivalacqua et al, 2005). On the basis of knowledge about the impaired NOcGMP pathway in diabetes mellitus, PDE-5 inhibitors Table 2. Effect of caffeine on erectile function in diabetic rats (mean 6 SD) Group A (n 5 8) Group B (n 5 6) Group C (n 5 9) Group D (n 5 8) Baseline ICP, mm Hg Peak ICP, mm Hg a ab ab MAP, mm Hg Peak ICP/MAP a ab ab Abbreviations: ICP, intracavernous pressure; MAP, mean systemic arterial pressure. a P,.01 compared with group A. b P,.01 compared with group B.

4 Yang et al N 589 Caffeine and Erectile Function in Diabetes Figure 1. (A) Electrostimulation of the cavernous nerve in a normal rat with normal pattern of intracavernous pressure (ICP). (B) Electrostimulation of the cavernous nerve in a diabetic rat with significant decrease of ICP. (C) Electrostimulation of the cavernous nerve in a diabetic rat treated with 10 mg/kg caffeine with partial restoration of ICP. (D) Electrostimulation of the cavernous nerve in a diabetic rat treated with 20 mg/kg caffeine with partial restoration of ICP. The red lines below are electronic stimulation duration (50 seconds). (sildenafil, vardenafil, tadalafil, and udenafil) are used to treat diabetic ED. The question is whether caffeine, as a nonselective PDE inhibitor used in our daily diet, is beneficial to penile erection in the diabetic condition. A study has shown that caffeine can increase the intracellular level of cgmp and camp and inhibit Ca2+ signaling (Moreland et al, 2001). It exhibits relaxant activities on various smooth muscles, including cavernous smooth muscle (Watanabe et al, 1992; Apaydin et al, 1998; Lindaman et al, 2002). Unfortunately, the role of caffeine on erectile function in epidemiology studies is controversial (Shiri et al, 2004). There is no study to evaluate the effect of caffeine for ED in diabetic patients. Therefore, study of the caffeine effect on erectile function in diabetic animal model can provide valuable information. In our study, diabetic rats were treated with 2 doses of caffeine according to the 2000 report published by the Australia New Zealand Food Authority (ANZFA): low dose (10 mg/kg per day) and moderate dose (20 mg/kg per day) (Smith et al, 2000). The 10 mg/kg per day dosage in rats equates to 250 mg/person per day (Corbin and Francis, 2002), which is the average intake of adults every day. The 20 mg/kg dose is at the upper end of the moderate dose in rats. Most literature suggests the positive effects of caffeine, such as increased energy, alertness, and motivation, with the low to moderate dosages. The high levels of caffeine consumption (.500 mg/kg per day) were reported to have negative effects, such as increasing anxiety and impaired sleep, which are risk factors of ED (Sasaki et al, 2005; Quek et al, 2007). Therefore, the high dose was not included in our study. Table 3. Effect of caffeine on the level of cavernous cgmp in diabetic rats (mean 6 SD) cgmp, pmol/mg protein Group A Group B Group C Group D a ab ab Abbreviation: cgmp, guanosine monophosphate. a b P,.01 compared with group A. P,.05 compared with group B.

5 590 Journal of Andrology N September ÙOctober 2008 Figure 2. The effect of caffeine on the level of cavernous cyclic guanosine monophosphate (cgmp) in diabetic rats. The level of cavernous cgmp decreased significantly in group B compared with group A (P,.01). With 8-week administration of caffeine, the level of cgmp in both group C and group D increased significantly when compared with group B (P,.05). There was no statistical significance between group C and group D. Our study demonstrated the beneficial effect of caffeine given at the low and moderate dosages on erectile function in diabetic rats. The ICP and ICP/MAP ratio are well accepted parameters to assess erectile function in animal studies. Our data showed that an 8- week administration of caffeine could partially restore the erectile function as demonstrated by increased ICP and ICP/MAP ratio in diabetic rats, however, the caffeine given at the low and moderate dosages did not completely restore normal erection. We did not find a dose-dependent effect at the given low and moderate caffeine dosages. These dosages were used according to ANZFA s report, and the gap between the given dosages might be too narrow to demonstrate the difference in this animal model. It appears to be a drawback of our study that we did not include a highdose treatment arm. Future study will incorporate the high-dose arm to elucidate the dose-dependent effect of caffeine on erectile function in this animal model. Studies in gallbladder and aortic smooth muscles (Watanabe et al, 1992; Lindaman et al, 2002) demonstrate that the relaxant effect of caffeine on smooth muscles relates to the increased cgmp level. Because cgmp plays an important role in penile erection (Bredt and Snyder, 1994), we assumed that the cavernous cgmp levels in the diabetic rats with and without caffeine treatment might be different. Therefore, we measured the cavernous cgmp levels of the rats with radioimmunoassay to clarify the relationship between caffeine intake and the NO/cGMP pathway. Our study showed a significant up-regulation of cgmp level in the caffeine-treated groups compared with the diabetic controls. This finding can explain the mechanism of the beneficial effect of caffeine on erectile function in diabetic rats. Caffeine can also increase the level of camp in gallbladder, aortic, and bladder smooth muscle tissues (Watanabe et al, 1992; Sekiguchi et al, 2002; Yi et al, 2006). camp also can trigger vasodilation, resulting in erection. This possibility was not included in this study. In theory, the elevated cavernous camp by caffeine might also contribute to the restoration of erectile function in diabetic rats. In this study, we demonstrated the beneficial effect of caffeine to enhance erectile function in diabetic rats by up-regulating cgmp in cavernous tissue. The etiology of diabetic ED is multifactorial (Moore and Wang, 2006). Despite the discovery of effective oral medications for ED, diabetic ED remains the medical condition most difficult to treat. At present, no single treatment strategy is effective for all diabetic ED cases. Combination therapies to target the multimolecular levels involved in diabetic ED are the future of managing this difficult condition (Chen et al, 2008). Caffeine, as shown in our study, cannot completely reverse ED in diabetic rats. However, as a daily diet, it might potentiate the effect of other erectile-genic medications, such as NO donators in the treatment of diabetic ED. Acknowledgment We thank Dorothy Stradinger for her editorial assistance. References Adebiyi A, Adaikan PG. Effect of caffeine on response of rabbit isolated corpus cavernosum to high K+ solution, noradrenaline and transmural electrical stimulation. Clin Exp Pharmacol Physiol. 2004;31: Akkus E, Kadioglu A, Esen A, Doran S, Ergen A, Anafarta K, Hattat H. Prevalence and correlates of erectile dysfunction in Turkey: a population-based study. Eur Urol. 2002;41: Andersson KE, Wagner G. Physiology of penile erection. Physiol Rev. 1995;75: Apaydin S, Gonen C, Guven H. The probable role of nitric oxide on the relaxations obtained by caffeine and aminophylline in rat uterus. Pharmacol Res. 1998;38: Berrada S, Kadri N, Mechakra-Tahiri S, Nejjari C. Prevalence of erectile dysfunction and its correlates: a population-based study in Morocco. Int J Impot Res. 2003;15(suppl 1):S3 S7. Bivalacqua TJ, Usta MF, Kendirci M, Pradhan L, Alvarez X, Champion HC, Kadowitz PJ, Hellstrom WJ. Superoxide anion production in the rat penis impairs erectile function in diabetes: influence of in vivo extracellular superoxide dismutase gene therapy. J Sex Med. 2005;2: Blanco R, Saenz de Tejada I, Goldstein I, Krane RJ, Wotiz HH, Cohen RA. Dysfunctional penile cholinergic nerves in diabetic impotent men. J Urol. 1990;144: Bredt DS, Snyder SH. Nitric oxide: a physiologic messenger molecule. Annu Rev Biochem. 1994;63:

6 Yang et al N Caffeine and Erectile Function in Diabetes 591 Chen Y, Dai Y, Wang R. Treatment strategies for diabetic patients suffering from erectile dysfunction. Expert Opin Pharmacother. 2008;9: Chen Y, Li SX, Yao LS, Wang R, Dai YT. Valsartan treatment reverses erectile dysfunction in diabetic rats. Int J Impot Res. 2007;19: Corbin JD, Francis SH. Pharmacology of phosphodiesterase-5 inhibitors. Int J Clin Pract. 2002;56: Diokno AC, Brown MB, Herzog AR. Sexual function in the elderly. Arch Intern Med. 1990;150: El-Sakka A, Yen TS, Lin CS, Lue TF. Traumatic arteriogenic erectile dysfunction: a rat model. Int J Impot Res. 2001;13: Kandeel FR, Koussa VK, Swerdloff RS. Male sexual function and its disorders: physiology, pathophysiology, clinical investigation, and treatment. Endocr Rev. 2001;22: Lindaman BA, Hinkhouse MM, Conklin JL, Cullen JJ. The effect of phosphodiesterase inhibition on gallbladder motility in vitro. J Surg Res. 2002;105: Lopez-Garcia E, van Dam RM, Qi L, Hu FB. Coffee consumption and markers of inflammation and endothelial dysfunction in healthy and diabetic women. Am J Clin Nutr. 2006;84: Moore CR, Wang R. Pathophysiology and treatment of diabetic erectile dysfunction. Asian J Androl. 2006;8: Moreland RB, Hsieh G, Nakane M, Brioni JD. The biochemical and neurologic basis for the treatment of male erectile dysfunction. J Pharmacol Exp Ther. 2001;296: Musicki B, Burnett AL. Endothelial dysfunction in diabetic erectile dysfunction. Int J Impot Res. 2007;19: Nicolosi A, Moreira ED Jr, Shirai M, Bin Mohd Tambi MI, Glasser DB. Epidemiology of erectile dysfunction in four countries: crossnational study of the prevalence and correlates of erectile dysfunction. Urology. 2003;61: Park CS, Ryu SD, Hwang SY. Elevation of intracavernous pressure and NO-cGMP activity by a new herbal formula in penile tissues of aged and diabetic rats. J Ethnopharmacol. 2004;94: Park S, Jang JS, Hong SM. Long-term consumption of caffeine improves glucose homeostasis by enhancing insulinotropic action through islet insulin/insulin-like growth factor 1 signaling in diabetic rats. Metabolism. 2007;56: Quek KF, Sallam AA, Ng CH, Chua CB. Prevalence of sexual problems and Its association with social, psychological and physical factors among men in a Malaysian population: a crosssectional study. J Sex Med. 2008;5(1): Rathmann W, Giani G. Global prevalence of diabetes: estimates for the year 2000 and projections for Diabetes Care. 2004;27: Rehman J, Chenven E, Brink P, Peterson B, Walcott B, Wen YP, Melman A, Christ G. Diminished neurogenic but not pharmacological erections in the 2- to 3-month experimentally diabetic F-344 rat. Am J Physiol. 1997;272:H1960 H1971. Richardson D, Vinik A. Etiology and treatment of erectile failure in diabetes mellitus. Curr Diab Rep. 2002;2: Saenz de Tejada I, Goldstein I, Azadzoi K, Krane RJ, Cohen RA. Impaired neurogenic and endothelium-mediated relaxation of penile smooth muscle from diabetic men with impotence. N Engl J Med. 1989;320: Sasaki H, Yamasaki H, Ogawa K, Nanjo K, Kawamori R, Iwamoto Y, Katayama S, Shirai M. Prevalence and risk factors for erectile dysfunction in Japanese diabetics. Diabetes Res Clin Pract. 2005;70: Seidler M, Uckert S, Waldkirch E, Stief CG, Oelke M, Tsikas D, Sohn M, Jonas U. In vitro effects of a novel class of nitric oxide (NO) donating compounds on isolated human erectile tissue. Eur Urol. 2002;42: Sekiguchi F, Miyake Y, Kashimoto T, Sunano S. Unaltered caffeineinduced relaxation in the aorta of stroke-prone spontaneously hypertensive rats (SHRSP). J Smooth Muscle Res. 2002;38: Shiri R, Koskimaki J, Hakama M, Hakkinen J, Huhtala H, Tammela TL, Auvinen A. Effect of life-style factors on incidence of erectile dysfunction. Int J Impot Res. 2004;16: Smith PF, Smith A, Miners J, McNeil JAP. Report From the Expert Working Group on the Safety Aspects of Dietary Caffeine. Canberra: Australian New Zealand Food Authority; Watanabe C, Yamamoto H, Hirano K, Kobayashi S, Kanaide H. Mechanisms of caffeine-induced contraction and relaxation of rat aortic smooth muscle. J Physiol. 1992;456: Yi CR, Wei ZQ, Deng XL, Sun ZY, Li XR, Tian CG. Effects of coffee and caffeine on bladder dysfunction in streptozotocin-induced diabetic rats. Acta Pharmacol Sin. 2006;27: