Doi:10.1016/s0090-4295(02)02439-

PHOSPHODIESTERASES AS THERAPEUTIC TARGETS CHING-SHWUN LIN, ZHONG-CHENG XIN, GUITING LIN, AND TOM F. LUE The advent of sildenafil (Viagra) for the treat- morethanoneisoformthroughalternativesplicing
ment of erectile dysfunction has attracted or from multiple gene promoters. The recently widespread interests in phosphodiesterase type 5 completed human genome sequence3 suggests that (PDE5), the target of sildenafil,1 and other mem- all human PDE genes have been identified and bers of the PDE superfamily. PDEs are intracellular enzymes that catalyze the hydrolysis of cyclicadenosine monophosphate (cAMP) and cyclic PDE STRUCTURE, SUBSTRATE
guanosine monophosphate (cGMP). By counter- SPECIFICITY, AND TISSUE DISTRIBUTION
balancing adenyl and guanyl cyclases, which cata-lyze the formation of cAMP and cGMP, respec- Some PDEs are specific for the hydrolysis of cAMP or cGMP and others have mixed specificity concentrations of cAMP and cGMP, thereby influ- (Table I). Those that are specific for cAMP are encing cellular functions.2 In addition to sildenafil, PDE4, PDE7, and PDE8; those specific for cGMP hundreds of PDE inhibitors broadly or specifically are PDE5, PDE6, and PDE9; and those with mixed target various PDE isoforms. Those with specificity specificity are PDE1, PDE2, PDE3, PDE10, and and currently seeking approval from the United PDE11.2 Each PDE has a conserved catalytic do- States Food and Drug Administration (FDA) or un- main of some 270 amino acids near the carboxyl der clinical trials are the focus of this review. Em- terminal (Fig. 1). The amino terminals of certain phasis is given to those with urologic applications PDEs have regulatory domains that bind to cal- (eg, erectile dysfunction and benign prostatic hy- cium ions, calmodulin, proteins, and/or cGMP.2,4 perplasia [BPH]-associated urinary dysfunction).
Alternatively spliced amino terminals of PDE4Dvariants have been shown to direct membrane lo- HUMAN PDE FAMILIES
calization and interact with regulatory proteins.5,6Reversible binding to Mg2ϩ also gives rise to con- The superfamily of PDEs is divided into classes I formational variants in the PDE4 families.7 Most and II, which share no sequence homology. Class I PDE families are expressed in a wide variety of includes all known mammalian PDEs that share tissues, and PDE6 is expressed only in photorecep- 20% to 45% sequence homology in the conserved tors.8 Within each PDE family, the alternatively catalytic domain. The human genome contains 21 spliced isoforms appear to exhibit tissue specifici- distinct PDE genes that encode 11 PDE protein ty.8,9 Differences among PDE isoforms in tissue families (Table I and Fig. 1).2 PDE1, PDE3, PDE4, distribution, enzymatic activity, and drug sensitiv- PDE6, PDE7, and PDE8 are multigene families, ity offer the window of opportunity for the design and PDE2, PDE5, PDE9, PDE10, and PDE11 are of isoform-specific inhibitors that may provide the unigene families. Each PDE gene usually encodes highest therapeutic benefits/undesirable side ef-fects ratio.10,11 This work was supported in part by grants from the CaliforniaUrology Foundation (to C.-S. L.), Arthur Rock and the Rock HISTORY OF PDE INHIBITORS
Foundation, and National Institutes of Health (2R01-DK-45370,to T. F. L.). Natural substances that are now known to exert From the Knuppe Molecular Urology Laboratory, Department their medicinal effects through inhibition of PDEs of Urology, University of California, San Francisco, School ofMedicine, San Francisco, California; and Department of Urology, have been used for many centuries. For example, The 1st Hospital, Peking University, Beijing, China the Chinese herb Yin Yang Huo (Epimedii herba), Reprint requests: Ching-Shwun Lin, M.D., Knuppe Molecular which is traditionally used to enhance sexual func- Urology Laboratory, Department of Urology, University of Cal- tion, contains a PDE5-inhibitory compound, icar- ifornia, San Francisco, School of Medicine, 1657 Scott Street, rin.12 Coffee, which has been prescribed to treat Submitted: July 30, 2002, accepted (with revisions): November asthma in the 19th century, derives its PDE-inhib- itory effects from caffeine.13 In the late 1950s, UROLOGY 61: 685– 691, 2003 • 0090-4295/03/$30.00
TABLE I. Human PDE families and their specific inhibitors
Substrate
Chromosome
Inhibitor
Target Disease*
KEY: PDE ϭ phosphodiesterase; cAMP ϭ cyclic adenosine monophosphate; cGMP ϭ cyclic guanosine monophosphate; FDA ϭ Food and Drug Administration; ED ϭ erectiledysfunction.
*The listed target diseases are those currently under clinical trial or seeking FDA approval; sildenafil is already approved for treating ED; exisulind and CP461 inhibited PDE5and PDE2, and their clinical trial information is posted at http://www.cellpathways.com. through publications by Sutherland and Rall,14,15 trials are listed in Table I. Although not listed, the physiologic effects of cAMP and PDE inhibitors a PDE2-specific inhibitor, erythro-9-(2-hydroxyl- began to be appreciated. In the past decade, hun- 3-nonyl)adenine,16 has been tested in various dreds of patents covering various PDE inhibitors tissues, but its clinical usefulness has yet to be have been issued. However, few of these initially reported. PDE6 polypeptides (PDE6␣, ␣Ј, and promising compounds ever passed clinical trials ␤) are essential for vision17 and, without any because of undesirable side effects, which can be known benefit to suppress their activities, are attributed to various PDEs’ overlapping enzymatic unlikely to be targets for drug development.
activity, drug sensitivity, and tissue distribution.
Because of PDE7’s predominant expression in With increasingly better understanding of the bi- T cells, it is hoped that its specific inhibitors ology of PDEs, efforts are underway to identify may be useful for treating immunologic and “second-generation” PDE inhibitors with im- inflammatory diseases, such as asthma.18 PDE proved therapeutic benefits/undesirable side ef- families 8 to 11 have been identified more re- cently, and reports of their specific inhibitorsare not yet available. PDE inhibitors pertaining CURRENT PDE INHIBITORS
to urologic applications are discussed in the fol- Examples of specific PDE inhibitors currently lowing sections, with an emphasis on PDE5 and seeking FDA approval or undergoing clinical UROLOGY 61 (4), 2003
direct the transcription of a pre-mRNA containingall three isoform-specific sequences.9 A secondpromoter is located between the A3 and A2-spe-cific exons and therefore can direct only the tran-scription of a pre-mRNA specifically encodingPDE5A2.9 By in vitro promoter assays, the activityof the PDE5A promoter is approximately threefoldthat of the PDE5A2 promoter.9 Both promoters ex-hibit increasing promoter activities with increasingcAMP or cGMP concentrations.9 Although the reg-ulation by cGMP is possibly a feedback mecha-nism, that by cAMP is more difficult to explain.
PDE5-SPECIFIC INHIBITORS AND
ERECTILE DYSFUNCTION
Sildenafil is the only FDA-approved PDE inhibi- tor currently available. It is highly specific forPDE5 inhibition with relatively minor cross-reac-tivity with PDE6.19 It has a chemical structure sim-ilar to cGMP (Fig. 2) and exerts its inhibitory effectby binding to the cGMP-catalytic sites of PDE5,thus allowing the accumulation of cGMP in theerectile tissue.20 As erotic stimuli trigger the releaseof nitric oxide in the erectile tissue and thus in- Regulatory and catalytic domains of mam- crease the production of cGMP, which is able to malian PDE families. The size of each PDE and its as- escape the destruction by PDE5 owing to the pres- sociated domains are drawn approximately to scale. ence of sildenafil, the concentration of cGMP thus The open box in the catalytic domain of PDE3 repre- reaches a level sufficient for a satisfactory erection, sents an ϳ44 amino acid insertion that is characteristic even in many patients with erectile dysfunction.21 of this family. The functions of the regulatory domainsare described in more detail in two reviews.2,4 Although many nonpenile tissues also expressPDE5 (Table II),9 they do not have the same mech-anism as penile tissue has that results in height- EXPRESSION AND REGULATION OF PDE5
ened production of cGMP after sexual stimulation.
PDE5 is expressed in a wide variety of tissues Such a difference between penile and nonpenile (Table II). The human PDE5A gene encodes three tissues is believed to be the reason sildenafil has a isoforms that differ only in the 5Ј end of their mes- satisfactory therapeutic benefits/undesirable side senger RNAs and the amino terminal of their pro- effects ratio.20,21 However, if a patient is taking ni- teins.9 PDE5A1 and A2 isoforms are expressed in trates (nitroglycerin, amyl nitrate, etc.) for isch- all tissues surveyed, and PDE5A3 appears to be emic coronary artery disease, he should not be tak- restricted to smooth muscles (Table III). In cul- ing Viagra because the combined effects of nitric tured cells, PDE5A1 expression is low or undetect- oxide release (from nitrates) and PDE5 inhibition able, PDE5A2 is in all cells surveyed, and PDE5A3 (by Viagra) could lower the blood pressure to a is still restricted to smooth muscles (Table IV).
When assayed with lysates of transfected COS-7 Two other PDE5-specific inhibitors (Fig. 2) tar- fibroblasts, the three isoforms have similar cGMP geting erectile dysfunction are currently under K of around 6 ␮M. When assayed for inhibition of FDA review and are expected to reach the market their cGMP-catalytic activities by sildenafil, the A1 within 1 year. Vardenafil, co-developed by Bayer isoform is more resistant (50% inhibitory concen- and GlaxoSmithKline, has been shown in a rabbit tration 28 nM) than the other two isoforms (14 and model to increase intracavernosal pressure more 13, respectively).9 Whether these isoforms possess quickly, at larger magnitude, and for a longer pe- distinct cellular functions has not yet been deter- riod than sildenafil.23 It has also been shown in clinical trials to have a high efficacy and low ad- Expression of the three PDE5 isoforms is dic- verse-event profile in a population with mixed tated by three isoform-specific first exons that are erectile dysfunction etiologies.24 Tadalafil, co-de- arranged in the order of A1-A3-A2 on the human veloped by Icos Corporation and Eli Lilly, has been PDE5A gene.9 Upstream from the A1-specific exon shown in clinical trials to have a long half-life of 17 is the PDE5A gene promoter that is expected to UROLOGY 61 (4), 2003
TABLE II. Tissue distribution of PDE5
Detection Method
Northern
Activity*
KEY: PDE ϭ phosphodiesterase; PCR ϭ polymerase chain reaction; ISH ϭ in situ hybridization; IHC ϭ immunohistochem-istry.
*cGMP catalysis. TABLE III. Tissue distribution of PDE5
TABLE IV. Cellular distribution of PDE5
isoforms in humans
isoforms
Cell-Lines/Strains
KEY: PDE ϭ phosphodiesterase. EXPRESSION OF OTHER PDES IN THE
CORPUS CAVERNOSUM
KEY: PDE ϭ phosphodiesterase. In addition to PDE5, at least 13 other PDEs are expressed at the RNA level in human corpus cav- PDE7A, PDE8A, PDE9A, and PDE10A.26 Such a diverse expression of most known mammalian PDE2A, PDE3A, PDE4A, PDE4B, PDE4C, PDE4D, PDEs implies a complex network of cross-talks UROLOGY 61 (4), 2003
strips was reversed by forskolin, sodium nitroprus-side, and inhibitors of PDE4 and 5. Thus, they sug-gested the possible use of inhibitors of PDE4 and 5for treating urinary obstruction secondary to BPH.
URINARY INCONTINENCE
Urinary urge incontinence has been treated with muscarinergic receptor blockers with limited effi-cacy. This is probably because neurotransmissionin the unstable detrusor is at least partly regulatedby nonadrenergic, noncholinergic pathways.31 Be-cause controlling cAMP and cGMP concentrationsis essential to nonadrenergic, noncholinergic path-ways, PDEs are now considered as potential thera-peutic targets for urge incontinence. The humandetrusor smooth muscle expresses PDE1, PDE2,PDE3, PDE4, and PDE5 and, likely, certain otherPDEs.32 Initial tests have indicated PDE1 to be themost promising target, and vinpocetine, a PDE1inhibitor, has been shown to improve clinical Comparison of structures of cGMP, silde- symptoms of nearly 60% (11 of 19) of patients who did not respond to standard pharmacologictherapy.33 among different PDE-regulatory pathways in thecorpus cavernosum. Indeed, sildenafil has been RENAL DISEASES
shown to significantly increase cAMP in isolated PDEs also play important roles in the physiology human cavernous tissue strips.27 This effect is and pathology of the kidney.34 In rat mesangial thought to involve PDE3, because cGMP, which is cells, PDE3 and PDE4 are linked to signaling path- accumulated as a result of PDE5 inhibition by sil- ways that modulate mitogenesis and generation of denafil, is capable of preventing cAMP degradation reactive oxygen metabolites. Increased activity of by competing for the same catalytic sites on the PDE5 in glomeruli and the cells of collecting ducts PDE3 molecules. Thus, increased cAMP owing to in sodium-retaining states, such as nephrotic syn- indirect inhibition of PDE3 may also contribute to drome, accounts for renal resistance to atriopeptin.
the erectogenic effect of sildenafil.
Abnormally high PDE4 activity in collecting ducts Animal models, such as rat and rabbit, are often is associated with unresponsiveness to vasopressin used in preclinical trials; therefore, differences in in mice with hereditary nephrogenic diabetes in- the PDE expression profile in the corpus caverno- sipidus. Several selective PDE inhibitors have been sum between animals and men must be taken into studied in animal models of various renal diseases, account when testing erectogenic PDE inhibitors.
including acute renal failure of different origins.34 For example, it has been shown that PDE3 is sig- The PDE5 inhibitor zaprinast has been shown to nificantly more expressed in human than in rabbit accelerate recovery from ischemic acute renal fail- corpus cavernosum, and this difference apparently ure in the rat.35 The PDE4 inhibitor Ro 20-1724 explains why milrinone, a PDE3 inhibitor, is able attenuated endotoxin-induced acute renal fail- to relax human, but not rabbit, corpus caverno- ure,36 and the PDE3 inhibitor cilostamide or cilostazol, together with the PDE4 inhibitor rolip-ram or the potent inhibitor of PDE3 lixazinone BPH-ASSOCIATED LOWER URINARY
alone, almost completely blocked the tubular cell TRACT SYMPTOMS
proliferation in association with folic acid-inducedacute renal failure.37 For the treatment of lower urinary tract symp- toms associated with, or suggestive of, BPH, FUTURE DIRECTIONS AND
Andersson et al.29 have suggested several potential CONCLUDING REMARKS
targets, including PDEs. Uckert et al.30 detectedRNA transcripts encoding PDE1, 2, 4, 5, 7, 8, 9, In addition to targeting at the specific PDE and 10 in various anatomic regions of the human isozymes directly, gene-based strategies have also prostate. The investigators also found that alpha- been proposed that seek to downregulate (eg, by adrenergic receptor-mediated tension in prostatic antisense oligonucleotides) the expression of spe- UROLOGY 61 (4), 2003
cific PDE genes at the transcription and/or transla- 13. Persson CGA: On the medical history of xanthines and tion levels.38,39 These are attractive alternatives be- other remedies for asthma: a tribute to HH Salter. Thorax 40:
cause, owing to the unique sequences of PDE 14. Sutherland EW, and Rall TW: The properties of an ad- isoform-specific exons and mRNAs, cross-reactiv- enine ribonucleotide produced with cellular particles, ATP, ity with related PDEs is not expected to be an issue.
Mgϩϩ, and epinephrine or glucagon. J Am Chem Soc 79:
Indeed, antisense-mediated downregulation of PDE5 has been shown to cause apoptosis of cancer 15. Sutherland EW, and Rall TW: Fractionation and char- cells in vitro (data presented at Gordon Research acterization of a cyclic adenine ribonucleotide formed by tis-
sue particles. J Biol Chem 232: 1077–1091, 1958.
16. Mery PF, Pavoine C, Pecker F, et al: Erythro-9-(2-hy- The complexity of the PDE superfamily poses droxy-3-nonyl)adenine inhibits cyclic GMP-stimulated phos- daunting challenges to our efforts to understand its phodiesterase in isolated cardiac myocytes. Mol Pharmacol biology; trying to design drugs on the basis of this 48: 121–130, 1995.
limited knowledge is even more difficult. However, 17. Arshavsky VY, Lamb TD, and Pugh EN Jr: G proteins the potential of using PDE inhibitors in urologic and phototransduction. Annu Rev Physiol 64: 153–187, 2002.
18. Nakata A, Ogawa K, Sasaki T, et al: Potential role of diseases is enormous. Additional studies in the bi- phosphodiesterase 7 in human T cell function: comparative ology and tissue distribution of various PDE iso- effects of two phosphodiesterase inhibitors. Clin Exp Immu- forms may identify agents that can specifically nol 128: 460 –466, 2002.
modify the function of kidney, bladder, testis, 19. Laties AM, and Fraunfelder FT: Ocular safety of Viagra prostate, and so forth and usher in a new era of (sildenafil citrate). Trans Am Ophthalmol Soc 97: 115–128,
1999.
medical management in urology. The surprising 20. Corbin JD, and Francis SH: Cyclic GMP phosphodies- success story of sildenafil is just the beginning.
terase-5: target of sildenafil. J Biol Chem 274: 13729 –13732,
1999.
21. Carter AJ, Ballard SA, and Naylor AM: Effect of the 1. Boolell M, Allen MJ, Ballard SA, et al: Sildenafil: an selective phosphodiesterase type 5 inhibitor sildenafil on erec- orally active type 5 cyclic GMP-specific phosphodiesterase in- tile dysfunction in the anesthetized dog. J Urol 160: 242–246,
hibitor for the treatment of penile erectile dysfunction. Int J Impot Res 8: 47–52, 1996.
22. Goldstein I, Lue TF, Padma-Nathan H, et al: Oral silde- 2. Francis SH, Turko IV, and Corbin JD: Cyclic nucleotide nafil in the treatment of erectile dysfunction. N Engl J Med phosphodiesterases: relating structure and function. Prog Nu- 338: 1397–1404, 1998.
cleic Acid Res Mol Biol 65: 1–52, 2001.
23. Saenz de Tejada I, Angulo J, Cuevas P, et al: The phos- 3. Venter JC, Adams MD, Myers EW, et al: The sequence phodiesterase inhibitory selectivity and the in vitro and in vivo of the human genome. Science 291: 1304 –1351, 2001.
potency of the new PDE5 inhibitor vardenafil. Int J Impot Res 4. Soderling SH, and Beavo JA: Regulation of cAMP and 13: 282–290, 2001.
cGMP signaling: new phosphodiesterases and new functions.
24. Porst H, Rosen R, Padma-Nathan H, et al: The efficacy Curr Opin Cell Biol 12: 174 –179, 2000.
and tolerability of vardenafil, a new, oral, selective phospho- 5. Jin SL, Bushnik T, Lan L, et al: Subcellular localization diesterase type 5 inhibitor, in patients with erectile dysfunc- of rolipram-sensitive, cAMP-specific phosphodiesterases: dif- tion: the first at-home clinical trial. Int J Impot Res 13: 192–
ferential targeting and activation of the splicing variants de- rived from the PDE4D gene. J Biol Chem 273: 19672–19678,
25. Eardley I, and Cartledge J: Tadalafil (Cialis) for men with erectile dysfunction. Int J Clin Pract 56: 300 –304, 2002.
6. Yarwood SJ, Steele MR, Scotland G, et al: The RACK1 26. Ku¨the A, Wiedenroth A, Ma¨gert HJ, et al.: Expression of signaling scaffold protein selectively interacts with the cAMP- different phosphodiesterase genes in human cavernous specific phosphodiesterase PDE4D5 isoform. J Biol Chem 274:
smooth muscle. J Urol 165: 280 –283, 2001.
27. Stief CG, Uckert S, Becker AJ, et al: Effects of sildenafil 7. Laliberte´ F, Han Y, Govindarajan A, et al: Conforma- on cAMP and cGMP levels in isolated human cavernous and tional difference between PDE4 apoenzyme and holoenzyme.
cardiac tissue. Urology 55: 146 –150, 2000.
Biochemistry 39: 6449 –6458, 2000.
28. Qiu Y, Kraft P, Lombardi E, et al: Rabbit corpus caver- 8. Loughney K, and Ferguson K: Identification and quan- nosum smooth muscle shows a different phosphodiesterase tification of PDE isozymes and subtypes by molecular biolog- profile than human corpus cavernosum. J Urol 164: 882–886,
ical methods, in Schudt C, Dent G, and Rabe KF (Eds): Phos- phodiesterase Inhibitors. San Diego, Academic Press, 1996, pp 29. Andersson KE, Chapple CR, and Hofner K: Future drugs for the treatment of benign prostatic hyperplasia. World 9. Lin CS, Chow S, Lau A, et al: Human PDE5A gene J Urol 19: 436 –442, 2002.
encodes three PDE5 isoforms from two alternate promoters.
30. Uckert S, Kuthe A, Jonas U, et al: Characterization and Int J Impot Res 14: 15–24, 2002.
functional relevance of cyclic nucleotide phosphodiesterase 10. Beavo JA: Cyclic nucleotide phosphodiesterases: func- isoenzymes of the human prostate. J Urol 166: 2484 –2490,
tional implications of multiple isoforms. Physiol Rev 75: 725–
31. Andersson KE, Chapple C, and Wein A: The basis for 11. Huang Z, Ducharme Y, MacDonald D, et al: The next drug treatment of the overactive bladder. World J Urol 19:
generation of PDE4 inhibitors. Curr Opin Chem Biol 5: 432–
32. Truss MC, Uckert S, Stief CG, et al: Cyclic nucleotide 12. Xin ZC, Kim E, Tian Z, et al: Icarrin on relaxation of phosphodiesterase (PDE) isoenzymes in the human detrusor corpus cavernosum smooth muscle. Chinese Sci Bull 46:
smooth muscle. I. Identification and characterization. Urol Res 24: 123–128, 1996.
UROLOGY 61 (4), 2003
33. Truss MC, Stief CG, Uckert S, et al: Initial clinical ex- 42. Arai T, Takano Y, Mizukami S, et al: The effect of olpri- perience with the selective phosphodiesterase-I isoenzyme in- none administered after cardio-pulmonary bypass during hibitor vinpocetine in the treatment of urge incontinence and open heart surgery, evaluated by its plasma concentrations low compliance bladder. World J Urol 18: 439 –443, 2000.
and hemodynamic changes. Masui 48: 1083–1090, 1999.
34. Dousa TP: Cyclic-3Ј,5Ј-nucleotide phosphodiesterase 43. Strandness DE Jr, Dalman RL, Panian S, et al: Effect of isozymes in cell biology and pathophysiology of the kidney.
cilostazol in patients with intermittent claudication: a ran- Kidney Int 55: 29 –62, 1999.
domized, double-blind, placebo-controlled study. Vasc Endo- 35. Guan Z, Miller SB, and Greenwald JE: Zaprinast accel- vascular Surg 36: 83–91, 2002.
erates recovery from established acute renal failure in the rat.
44. Park SW, Lee CW, Kim HS, et al: Effects of cilostazol on Kidney Int 47: 1569 –1575, 1995.
angiographic restenosis after coronary stent placement. Am J 36. Begany DP, Carcillo JA, Herzer WA, et al: Inhibition of Cardiol 86: 499 –503, 2000.
type IV phosphodiesterase by Ro 20-1724 attenuates endotox- 45. Sturton G, and Fitzgerald M: Phosphodiesterase 4 in- in-induced acute renal failure. J Pharmacol Exp Ther 278:
hibitors for the treatment of COPD. Chest 121: 192S–196S,
2002.
46. Baraniuk JN, and Tai CF: Inhibition of phosphodiester- 37. Matousovic K, Tsuboi Y, Walker H, et al: Inhibitors of ase 4 in allergic rhinitis. Curr Allergy Asthma Rep 2: 191–192,
cyclic nucleotide phosphodiesterase isozymes block renal tu- bular cell proliferation induced by folic acid. J Lab Clin Med 47. Timmer W, Leclerc V, Birraux G, et al: The new phos- 130: 487–495, 1997.
phodiesterase 4 inhibitor roflumilast is efficacious in exercise- 38. Epstein PM: Antisense inhibition of phosphodiesterase induced asthma and leads to suppression of LPS-stimulated expression. Methods 14: 21–33, 1998.
TNF-alpha ex vivo. J Clin Pharmacol 42: 297–303, 2002.
39. Rybalkin SD, Rybalkina I, Beavo JA, et al: Cyclic nucle- 48. Prasad S, Wilkinson J, and Gatzoulis MA: Sildenafil in otide phosphodiesterase 1C promotes human arterial smooth primary pulmonary hypertension. N Engl J Med 343: 1342,
muscle cell proliferation. Circ Res 90: 151–157, 2002.
40. Feigin VL, Doronin BM, Popova TF, et al: Vinpocetine 49. Berman JR, Berman LA, Lin H, et al: Effect of sildenafil treatment in acute ischaemic stroke: a pilot single-blind ran- on subjective and physiologic parameters of the female sexual domized clinical trial. Eur J Neurol 8: 81–85, 2001.
response in women with sexual arousal disorder. J Sex Marital 41. Yamaura K, Akiyoshi K, Irita K, et al: Effects of olpri- Ther 27: 411–420, 2001.
none, a new phosphodiesterase inhibitor, on gastric intramu- 50. Liu L, Li H, Underwood T, et al: Cyclic GMP-dependent cosal acidosis and systemic inflammatory responses following protein kinase activation and induction by exisulind and hypothermic cardiopulmonary bypass. Acta Anaesthesiol CP461 in colon tumor cells. J Pharmacol Exp Ther 299: 583–
Scand 45: 427–434, 2001.
UROLOGY 61 (4), 2003

Source: http://www.wlsz.cn/~guitinglin/lunwen/c.pdf