International Journal of Impotence Research (1996) 8, 47- 52
`© 1996 Stockton Press. All rights reserved
`0955-9930/96
`$12 00
`
`Sildenafil: an orally active type 5 cyclic GMP-specific
`phosphodiesterase inhibitor for the treatment of penile erectile
`dysfunction
`
`, Michael J Allen 1
`, Stephen A Ballard 1
`Mitradev Boolell 1
`, Sam Gepi-Attee 2
`, Ian H Osterloh 1 and Clive Gingell 2
`Alasdair M Naylor 1
`
`, Gary J Muirhead 1,
`
`1 Pfizer Central Res earch, Sandwich, Kent CT1 3 9N], UK ; 2 Urology Department, Southmead Ho s pital, Bristol, UK
`
`Sildenafil (Via gra ™, UK-92,480) is a novel oral agent under development for the treatment of
`peni_Ie erectil~ dysfunction. Erection is dependent on nitric. oxi.de and its second messenger,
`cychc guanosme monophosphate (cGMP). However, the relative Importance of phosphodiester(cid:173)
`ase (PDE) isozymes is not clear. We have identified both cGMP- and cyclic adenosine
`monophosphate-specific phosphodiesterases (PDE s) in human corpora cavernosa in 1•itro. The
`main PDE activity in this tissue was due to PDE5, with PDE2 and 3 also identified. Sildenafil
`is a selective inhibitor of PDE5 with a mean IC 50 of 0.0039 JlM. In human volunteers, we have
`shown sildenafil to have suitab le pharmacokinetic and pharmacodynamic prope•·ties (rapid
`absorption, relatively short half-life, no significant effect on heart rate and blood pressure) for
`an oral agent to be taken, as required, prior to sexual activity. Moreover, in a clinical study of
`12 patients with erectile dysfunction without an established organic cause, we have shown
`sildenafil to enhance the erectile response (duration and rigidity of erection) to visual sexual
`stimulation , thus highlighting the important role of PDE5 in human penile erection. Sildenafil
`hold s promise as a new effective oral treatment for penile erectile dysfunction.
`
`Keywords: penile erectile dysfunction; sildenafil; ora l treatment; phosphodiesterase type 5
`
`Introduction
`
`Penile erectile dysfunction is a common medical
`disorder. It has an estimated prevalence of 2°/)J in
`men aged 40 years, which increases to over 50%
`in men over the age of 70 years. 1 Penile erectile
`dysfunction has been defined as 'the inability to
`achieve and/ or sustain an erection for satisfactory
`sexual performance'. 2 Generally, it is accepted
`that this disorder adversely affects quality of life.
`Patients often report increasing anxiety, loss of
`self-esteem, lack of self-confidence, tension and
`difficulty in the relationship with their partner. 2
`Penile erection is a haemodynamic event which
`is dependent upon relaxation of
`the smooth
`muscle cells of the corpus cavernosum and of its
`associated arterioles, with consequ ential increase
`in arterial ±low into the trabecular spaces of the
`corpora cavernosa. 3
`.4 The increased blood ±low
`causes the lacunar spaces or sinusoids to become
`distended which results in compression of the
`small venules between
`the sinusoids and the
`tunica albuginea. The relative indistensibility of
`the tunica albuginea results in a vena-occlusive
`effect such that the penile pressure increases to
`
`Correspondenc e Dr M Boole ll
`Rece ived 25 Apri l ; accepted 15 May 1996
`
`approach mean arterial pressure and penile rigid(cid:173)
`ity develops.
`There is now ample evidence from both animal
`experiments and in vitro studies with human
`tissue to suggest that relaxation of the smooth
`muscle of the corpora cavernosa is mediated by
`nitric oxide via cyclic guanosine monophosphate
`(cGMP). s- s During sexual stimulation, nitric oxide
`is released from nerve endings and endothelial
`cells. Nitric oxide then stimulates the cytosolic
`enzyme guanylate cyclase to produce cGMP which
`results in a decrease in intracellular calcium and
`allows relaxation of smooth muscle cells . Cyclic
`nucleotide phosphodiesterase
`(PDE)
`isozymes,
`which are distributed in various tissues, specifi(cid:173)
`cally hydrolyse cyclic nucleotides, such as cGMP. 9
`Therefore, a pharmacological agent which inhibits
`the cGMP-specific phosphodiesterase
`isozyme,
`should enhance the action of nitric oxide/cGMP on
`penile erectile activity and have the potential to
`enhance penile erections during sexual stimu(cid:173)
`lation.
`To date, pharmacological therapy for penile
`erectile dysfunction has been largely based on the
`use of intracavernosal injections of vasoactive
`agents. Though efficacious, this form of therapy is
`associated with a high dropout rate for a variety
`of reasons. 10 Recent insights into the mechanism
`of penile erection have, however, led to the devel-
`
` DRL - EXHIBIT 1015
`
`

`

`Sildenafi l
`M Boolell eta/
`
`[
`.
`TM
`; 1- 4-e thoxy-3-(6,7-
`opment of sildenafil (Vragr_a
`dihydro-1-methyl-7-oxo-3-piopyl-1H-pyrazolo [ 4,3 -
`d]
`pyrimidin-5-yl)
`phenylsulphonyl]-4-methyl(cid:173)
`piperazine, pfizer Central Research) a nov~ l,
`or ally-efficacious drug for the treatment of p emle
`erectile dysfunction.
`.
`.
`.
`.
`This review focuses on lll v1tro studies relatmg
`to
`the mode of action of sildenafil, phar(cid:173)
`macokinetic s tudies in human vo lunteers , and an
`early clinical study in patients with erectile dys-
`function.
`
`Methods
`
`Iso lulion of soluble ph os ph odie s te ru se act ivities from
`human ti ss ues
`
`Frozen human corpus cavernosal tissue (obtained
`from IIAM, Exton, Pennsylvania; donor age r ange
`48- 64 years) was thawed on ice, coarsely chopped,
`and
`then homogenised
`in approximately 4
`volumes of ice cold HEPES buffer (20 mM, contain(cid:173)
`ing 0.25 M sucrose, 1 mM EDT A, 1 mM phenyl(cid:173)
`methyl sulphonylfluoride [PMSF], pH 7.2) using
`an Ultra Turrax homogeniser at high setting. The
`homogenate was filt ered through two layers of
`su rgical gauze to remove any undisp er sed tiss ue
`and fibrous material. The filtrate was centrifuged
`at 100 000 g for 60 min at 4°C. The s up ernatant
`was filtered through a 0.2 ~LM filter and either
`u sed directly (see below) or s tored in liquid nitro (cid:173)
`gen prior to analysis . A similar procedure was
`used for preparation of soluble fractions from
`samples of human cardiac ventricle and human
`skeletal muscle (also obtained from IIAM).
`Phosphodiesterase activities in the soluble frac (cid:173)
`tions prepared from human tissues were separat(cid:173)
`ed using a Pharmacia FPLC system (Pharmacia
`Ltd, Milton Keyne s, UK) with a Mono Q anion
`exchange column (1 mL bed volume, Pharmacia
`Ltd). The Mono Q co lumn was pre-equilibrated
`with HEPES buffer
`(20 mM, containing 1 mM
`EDTA, 0.5mM PMSF, pH 7.2), before loading
`tissue soluble fractions (2- 5 mL). The column was
`then washed with 5 ml of the buffer and the phos (cid:173)
`phodies terase isozymes were eluted using a con-
`
`tinuous gradient of 0-500 mM NaCl in the same
`buffer (tota l vo lume 55 mL) at a flow rate of 1 mL/
`min, and 2 mL fra ctions wer e collec ted. Fractions
`comprising the main peaks of pho sp hodies tera se
`act ivity were pooled and sto red at - 80°C for use
`in characterisation a nd inhibition studies.
`
`Dete rmination oJ phos ph odi es terase activity and
`isozym e characteri sa tion
`The cyclic nucleotide phosphodiesterase activity
`in FPLC fraction s was d etermined u s ing a modifi(cid:173)
`cation of the two step radioisotopic procedure of
`Thomp son and Appleman. 11 The reaction mixture
`(total volume 100 ~tL) contained co lumn fraction
`(10 to 25 ~tL) , [ 3H]-cGMP or [ 3H}cAMP (500 nM,
`2 ~tCi/m L), bovine serum albumin (0.5 mg/ mL) and
`MgC1 2 (5 mM) in Tris HCl buffer (15 mM, pH 7.4).
`Reaction was initiated by addit ion of the ra diola(cid:173)
`belled substrate, and the samp les incubated in a
`water bath at 30"C for 30 min. The reaction was
`stopped by immers ing the samp le tube s in boiling
`water for 2 min. The 5'-mo nonucletides formed by
`hydrolysis of cyclic nucleotides were determined
`after their further conversion to nucleo sides using
`s nake venom (Oph iophagus hannah) nucleo tid ase
`activity as described by Thomp son and App le(cid:173)
`man.1 1
`test compounds were
`inhibitor studies,
`For
`added to incubation mixtures in dimethyl sul(cid:173)
`pho x ide (DMSO, final concentration 2°/cl v/v). The
`hydrolysis of cyclic nucleotides did not exceed
`15 1YcJ and und er these conditions, product forma (cid:173)
`tion increased linearly with time and amount of
`enzyme .
`from
`IC 50 values were determined
`sigmoidal curves, fitted to plots of enzyme activity
`vs
`log compound concentration using a curve
`fitting programme. 1 2
`Pho sphodies terase activities in FPLC fraction s
`were characterised based on their substrate spe(cid:173)
`cificities, effects of calcium calmodulin, the effect
`of cGMP on cAMP hydrolytic act ivity and the
`inhibitory potency of known selective inhibitors of
`the PDE isozymes (Tab le 1). 13
`16 Human cardiac
`-
`ventricle was used as a source of soluble PDE1
`and human s keletal muscle (gluteus m ax imus) for
`PDE4 .
`
`Table 1 Classification of PDE isozymes 1 3
`
`Iso zyme
`.fi11nily
`
`PDE I
`PDE 2
`PDE 3
`PDE 4
`PO E 5
`
`Suhstrat e
`specificity
`
`cA MP/ cGMP
`cAMP/ cGMP
`cAMP
`cAMP
`cG MP
`
`Effect of cG M P un
`cA MP hydrolysis
`
`Effect uj calcium
`ctllmorlulill
`
`N.A.
`s timulation
`inhibition
`no effect
`N.A.
`
`stimulation
`no effect
`no efi"ect
`no effect
`no efi.e ct
`
`Slamlarrl inhihil or
`
`vin pocetine 1 h
`none
`m ilrinone 16
`ro li pram 1
`''
`
`zaprinast 1 '', E4021 15
`
`N.A. - not ap propri ate for chara c terisation of these isozyme families
`PDE = phos phodi es ter ases
`cGMP =cyclic g uano s ine rnonopho s ph a te
`cAMP = cyclic adenos in e monopho sphate
`
`

`

`

`

`

`

`

`

`Sildenafil
`M Boolell et at
`
`penile erectile activity in patients with erectile
`dysfunction during sexual s timulation. Experi(cid:173)
`mental evidence indicates that this
`is a
`local
`action of sildenafil on the corpus cavernosal
`smooth muscle cells where it enhances nitric oxide
`mediated relaxation. 21
`22 We believe that under
`'
`conditions of nitric oxide drive, type 5 PDE is the
`most important regulator of cGMP levels in the
`corpus cavernosal smooth muscle cell. The mecha(cid:173)
`nism of action of sildenafil suggests that it will
`induce or enhance erections during sexual stimu(cid:173)
`lation, but will not provide erections
`in
`the
`absence of any physiological or sexual stimulus.
`Sildenafil demonstrates ideal pharmacokinetics
`for an oral agent to be taken, as required, prior to
`sexual activity for the treatment of penile erectile
`dysfunction. It is rapidly absorbed when adminis(cid:173)
`tered orally, and has an onset of action of less
`t~an one hour. Furthermore, the drug has a rela(cid:173)
`tively. short plasma half life of approximately 4 h
`and IS not expected to accumulate on repeated
`single daily administration. In the dose range
`studied, sildenafil is generally well tolerated and
`has no significant effects on pulse rate and blood
`pressure.
`Sildenafil represents a new class of orally active
`and peripherally acting drug for the treatment of
`penile erectile dysfunction. The RigiScan study
`demonstrates that sildenafil enhances the erectile
`response to erotic stimuli under standardised con(cid:173)
`ditions . The results are reported in terms of mean
`duration of rigidity of greater than 60% . Although
`there is considerable debate as to the threshold
`degree of rigidity which correlates with sufficient
`hardness
`for penetrative
`intercourse, previous
`studies indicate that the vast majority of patients
`who achieve 60% rigidity will have an erection
`24 Its
`that is sufficient for sexual intercourse.23
`'
`efficacy in enhancing erections during stimulation
`in the natural and private setting and its safety on
`long-term adminstration are currently being inves(cid:173)
`tigated in large-scale clinical trials.
`
`Acknowledgements
`We wish to thank Mrs Lucy Short for secretarial assist(cid:173)
`ance and Ms J Pearson for help with the statistical
`analyses.
`
`References
`1 Furlow WL. Prevalence of Impoten ce in the United States.
`Ivied r\sp<~cts Hum Sexuality 1985; 19: 13- 16.
`2 NU-L Consensus Development Panel on Impotence. ]AMA
`1993; 280: 83- 90.
`
`3
`
`7
`
`14
`
`Lue TF, e l al. Haemodynamics of erection in the monkey. ]
`u· I 1983· 130:123 7- 1241 .
`.
`.
`.
`10
`'KE Wagner G. Physiology of pemle erectwn.
`·
`4 Anderson
`' •
`. · 1 n v 1995 .75:191- 236.
`I .
`P 1Ys 10
`e
`'
`Cl .· t G) A revi ew of erectile dys-
`r. L
`·ner SE Melman A, HI S
`·
`1·ns ight s and more questions. J Urol 1993;
`. el
`.
`. N'
`"
`fun ctwn.
`ew
`1491: 1246-Al255.on WJ RaJ'fer J Ignarro Lj. The L-arginine
`· 1 P A
`6 B
`rons
`•
`'
`us
`' .
`. . ··on in th e co rpu s cavernosum of human
`J V) · VJO V32
`n eurotr a n sm Jss J
`(
`. d . bbit Circulation 1993; 117 supp
`., .'
`-
`. . .
`~7m 1~ A~adzoi KM, Goldstein I, De 1ejada IS. A mtr~c
`f
`tor mediates non -adrene rgic, non-cholmergJc
`'d 1'.k
`ox1 e- 1 e ac
`.1
`·.
`>laxation of pem e corp u s cavernosum
`neurogenic
`re
`l
`leI Clin In ves l1991 ; 1111:11 2- 118.
`smoo t 1 mAuLsc , ·el i Nitric oxide: a physiologic mediator of
`8 Burnett
`. e
`·
`nile erection. Scienc:e 1992 ; 257:401-403.
`.
`pe
`. C K mas N Modulation of vascular cycl1c nucleo-
`'h 0diester~ses by cyclic GMP: role in vasodilata-
`9 LudgnJel r
`ti e p wsp o
`)
`. !-!ea rl I 1993; 14 (Suppl1 : 141- 148.
`.
`E
`twn. UISB · _,1 cii Side effects of self-administration of
`10 Levme
`• L ·
`1
`·
`f.
`h
`t
`·
`·
`. u s papaverine and phento amme or t e trea -
`.
`mtra cavei no
`ment of impotence. 1 Ural 1989; 141: 54- 57.
`.
`.
`WJ Appleman MM. Mult1ple cychc nucleot1de
`h
`11 T ompson
`·
`b
`· B · 1
`· l
`phosphodiesterase activities from rat
`ram. · JOe 1em1s .ry
`USA 1971; 10: 311-3 16.
`.
`..
`12 De Lean A, Munson A, Rodbard D. S~ult a neous analys1s
`of fam ilies of sigmoidal curves:. apphcatwn to bwassay,
`radioligand assay and physwlog1cal dose-response curves.
`Am J Physiol 1978; 235: E97- E102.
`.
`.
`.
`13 Beavo )A, Conti M, Heaslip RJ. Multiple cychc nucleotide
`h
`. hodiestera ses. Mol Phorllloc:ol 1994; 46: 399- 405 .
`PS'losp P)
`, 1 al. Cyclic GMP IJotentiation by WIN 58237, a
`I ve r
`, v
`I d'
`. h'b't
`]
`novel cyclic nucleotide phosp 10 1es tera se m 1 1 or.
`Plwrnw co l Exp Til e r 1994; 271: 1143- 1149:
`.
`. .
`15 Saeki T, e l ol. Selective type V phosphodiesterase mhibl(cid:173)
`tor, E4021, dilates porcine large co ron ary artery. ] Pl!ar-
`macol Exp Th er 1995 ; 272:825- 831.
`.
`.
`16 Beavo JA, Reifs nyder DH. Prrmary se quence of cychc.
`nucleo tide phosphodiesterase enzymes an d th e design of
`selective inhibitors. Tre nd s Plwramc:ol Sc1 1990; 11: 150-
`17 ~~~1-espie PG . Beavo ]A . Inhibition a~d s timulation of
`photoreceptor pho s phod1e s teras~~ .by d.Jpyndamole and M
`& B 22 948. Mol Phurmo col 1989 , .16. 773- 781.
`18 Stieff 'cG e l ol. Phosophodiesterase isoenzyme of the
`human cavernous tissue and its fun ctional significance.
`Akt Urol1995; 26:22-24.
`19 Beavo )A. Cyclic nucle?tide phosphodiesterases : function (cid:173)
`al implications of multiple 1soforms. Physwl Hev 1995; 75:
`725-748.
`20 Sonnenburg WK. Beavo JA. Cyclic GMP and regulat ion of
`cyclic nu cleotide hydrolys is. Adv Pharmacal 1994; 26:
`87- 114.
`21 Ballard SA e t ol. In vitro profile of UK-92,480, an inhibitor
`of cyclic GMP s pecific phosp~odie s tera se 5 for the trea t(cid:173)
`ment of male erectile dysfunctiOn. J Uro/1996; 155: 676A.
`22 T ang K, Ballard SA, Turner LA, Naylor AM. Effects of the
`novel phosphodiesterase type 5 inhibitor, s ildenafil, on
`metha choline induced relaxation of iso lated rabbit corpus
`caverno s um. Br I Pharmocol 1996 (in press).
`23 Kaneko S, Bradley WE. Evaluation of erectile dysfunction
`with continuous monitoring of penile rigidity. I Ural 1986;
`1:16: 1026-1029.
`24 Ogrinc FG, Lin et 01. Evaluation of real-time RigiScan
`monitoring in pharmacological erec tion; I Urol 1995; 154 :
`·1356- 1359.
`
`