Neurotransmitters: Central and Peripheral Mechanisms

Neurotransmitters: central and peripheral mechanisms

K-E Andersson1*

1Department of Clinical Pharmacology, University of Lund, Lund, Sweden

Re¯exive erection initiated by recruitment of penile afferents, involves both autonomic and somatic efferents. The re¯ex is mediated at the spinal cord level, modulated by supraspinal in¯uences, and may use several transmitters. Dopamine, acetylcholine, nitric oxide, and peptides, such as oxytocin and ACTH=a-MSH, seem to have a facilitatory role, whereas serotonin may be either facilitatory or inhibitory, and enkephalins are inhibitory. Peripherally, the balance between contractant and relaxant factors controls the degree of contraction of the smooth muscle of the corpora cavernosa, and determines the functional state of the penis. Noradrenaline contracts both corpus cavernosum and penile vessels via stimulation of a1-adrenoceptors. The role of endothelins in the control of penile smooth muscle tone is presently unclear. Neurogenic nitric oxide (NO) is considered the most important factor for relaxation of penile vessels and corpus cavernosum. The role of other mediators, released from nerves or endothelium has not been de®nitely established. International Journal of Impotence Research (2000) 12, Suppl 4, S26±S33.

Keywords: CNS pharmacology; erectile dysfunction; autonomic and somatic mechanisms

Introduction Central neuromediation

The central regulation of erection with processing The central mechanisms in¯uencing the erectile and integration of tactile, olfactory, auditory and response include spinal and supraspinal pathways. mental stimuli, involves many transmitters and Several regions within the central nervous system transmitter systems. This is also the case periph- may be involved.2 Much of the knowledge gained in erally, and the different steps involved in neuro- this area relates to morphological and pharmacolo- transmission, impulse propagation and intracellular gical studies in experimental animal models (eg transduction of neural signals in penile smooth rodents, primates) in which neurochemical pertur- muscles are only partly known. Re¯exive erection bations can be achieved and responses monitored in initiated by recruitment of penile afferents involves a reasonably meaningful way. For example, studies both autonomic and somatic efferents. The re¯ex is in rats have revealed that electrical stimulation of mediated at the spinal cord level and modulated the medial preoptic area,8 the paraventricular by supraspinal in¯uences, and involves several nucleus,9 or the hippocampal formation10 elicits transmitters.1±6 an erectile response. Peripherally, the balance between contractant and relaxant factors controls the degree of contraction of the smooth muscle of the corpora cavernosa, and determines the functional state of the penis: Hydroxytryptamine detumescence and ¯accidity, and tumescence and erection, respectively.6,7 Below a short review is given of some of the 5-hydroxytryptamine (5-HT; serotonin) is present in transmitters=modulators believed to be involved in supraspinal as well as spinal and peripheral sites the control of erectile mechanisms centrally and believed to be involved in sexual behavior. Neurons peripherally. containing 5-HT can be found in the medullary raphe nuclei and ventral medulla reticular formation, including the rostral nucleus paragigantocellu- laris. Bulbospinal neurons containing 5-HT project to the lumbar spinal cord in the rat and cat.2 Fibres immunoreactive for serotonin were largely and *Correspondence: K-E Andersson, Department of Clinical Pharmacology, Lund University Hospital, S-221 85 Lund, densely distributed in the dorsal horn, the dorsal Sweden. gray commissure, the sacral parasympathetic nu- E-mail: [email protected] cleus and the ventral horn. Some serotonergic ®bres Neurotransmitters in ED K-E Andersson S27 occurred in close apposition with retrogradely- lar nucleus, similar experiments have established labelled sacral preganglionic neurons and motor that D2 rather than D1 receptors primarily facilitate neurons and synapses were demonstrated at the erections.17 The mechanism of erection following ultrastructural level.11 paraventricular D2 receptor stimulation apparently 5-HT has been most strongly implicated in the involves oxytocinergic neurotransmission. In addi- spinal pharmacology of erectile function with tion, NO seems to be involved, since apomorphine participation in both sympathetic and parasympa- (given s.c.) increased NOS activity and NO produc- thetic out¯ow mechanisms. Experimental para- tion in the paraventricular nucleus.20 Both the digms in animals have indicated that 5-HT increase in NO and the erectile response to pathways exert a general inhibitory effect on male apomorphine were reduced by NOS inhibition. sexual behavior,12 although these pathways may be inhibitory or facilitatory depending upon the action of the amine at different 5-HT receptors located in the central nervous system.1,13 Noradrenaline Despite this general understanding, con¯icting reports of 5-HT agonists either enhancing or depres- Evidence for noradrenergic effects in the central sing sexual function were recognized and then neuromediation of penile erection is sparse. attributed to the action of multiple 5-HT receptors. However, the current data suggest that increased In accordance with the selective use of 5-HT nor-adrenergic activity stimulates, whereas de- receptor agonists and antagonists, components of creased noradrenergic activity inhibits, sexual male copulatory behavior were found to be dis- function.12,21,22 played variably. The sum of multiple studies suggests that 5-HT2 and 5-HT1A receptors mediate inhibitory effects on penile erection, whereas 5- HT1C receptors mediate facilitatory effects on this Gamma-amino butyric acid 14 function. Stimulation of 5-HT1C receptors increased circulating oxytocin,15 and nitric oxide synthase (NOS) inhibitors given by intracerebrovas- Cumulative data resulting from investigations on the role of gamma-aminobutyric acid (GABA) in penile cular administration prevented 5-HT1C-receptor mediated erectile responses,16 suggesting that both erection indicate that this neurotransmitter may function as an inhibitory modulator in the auto- oxytocin and NO were involved in the 5-HT1C- receptor mediated responses. nomic and somatic re¯ex pathways involved in Thus, 5-HT appears to serve various functions in penile erection.1 Recent investigations showed that male sexual function and likely acts as a major activation of GABA(A) receptors in the paraven- modulator of the central neuroregulatory control of tricular nucleus of the hypothalamus reduced penile erection. apomorphine-, N-methyl-D-aspartic acid (NMDA-), and oxytocin-induced penile erection and yawning in male rats.23 Dopamine

Oxytocin Dopaminergic neurons have been identi®ed as traveling from the caudal hypothalamus to innervate the lumbosacral spinal cord and dopamine may Hypothalamic supraoptic and paraventricular participate in the CNS regulation of the autonomic nuclei project to spinal centers in¯uencing erectile and somatic components that produce penile function.2 The oxytocinergic spinal projections are re¯exes. Supraspinal dopaminergic neurons also more likely to in¯uence sacral autonomic rather comprise the incertohypothalamic system with than somatic out¯ow.24 Plasma oxytocin concentra- projections to the medial preoptic area and para- tions are known to be elevated in humans following ventricular nucleus. Injection of apomorphine into sexual stimulation. these central nuclei has facilitatory effects on sexual When injected into the lateral cerebral ventricle, behavior,2,17 and produces the `penile erection- the paraventricular nucleus, or hippocampus of stretching-yawning' response. laboratory animals, oxytocin is a potent inducer of Two main types of dopamine receptors, D1 and penile erection. Oxytocin receptors are believed to D2, are associated with erectile function centrally, mediate the erections, since the erectile response is with the latter perceived to predominate in this blocked by the administration of oxytocin antago- effect. In the medial preoptic area, use of selective nists and by electrolytic lesion of the paraventricu- receptor agonists has established that low levels of lar nucleus. Immunoreactive oxytocin-containing dopaminergic stimulation, via D1 receptor in parti- spinal neurons associating with sacral preganglionic cular, facilitates erections.18,19 In the paraventricu- neurons con®rmed by retrograde labelling supports

International Journal of Impotence Research Neurotransmitters in ED K-E Andersson S28 the role of oxytocin in the autonomic spinal that occurs in the paraventricular nucleus during circuitry that mediates penile erection.11,25 sexual activity.31 Morphine also prevents apomor- Central oxytocin seems to be involved in the phine-, oxytocin- and NMDA-induced penile expression of penile erection induced not only by erection and yawning by inhibiting NO synthase drugs but also by sexual physiological stimuli, since activity in the paraventricular nucleus through the non-contact erections in rats were reduced dose- stimulation of opioid receptors of the m subtype.32,33 dependently by a selective oxytocin receptor antagonist, given into the lateral ventricles.26 The oxytocinergic system may be in¯uenced by Nitric oxide the NO synthase signal transduction pathway, since inhibitors of this pathway prevent penile erection and yawning in rats induced by oxytocin, dopa- Evidence is accumulating that NO functions as a mine, and NMDA stimulation. Conversely, oxytocin neurotransmitter in the CNS, where NO can mod- increases NO production in the paraventricular ulate sexual behavior and penile erection.16,34 ± 37 nucleus.27 NO may act in several discrete brain regions, eg in the medial preoptic area36,37 as well as in the paraventricular nucleus.9,16 Injection of NOS inhi- ACTH and related peptides bitors intracerebroventricularly or in the paraventricular nucleus prevented penile erectile responses induced by dopamine agonists, oxytocin, and Peptides derived from the common 31 kDa precursor NMDA in rats. The inhibitory effect of NOS pro-opiomelanocortin, including adrenocortico- inhibitors was not observed when these compounds tropic hormone (ACTH) and a-melanocyte stimulat- were injected concomitantly with L-arginine, the ing hormone (a-MSH), have been associated with substrate for NO.16 NO production increased in the erectile responses. Both peptides can induce penile paraventricular nucleus of the hypothalamus of erection along with grooming, stretching and yawn- male rats during non-contact penile erections and ing following its intracerebroventricular injection.28 copulation, con®rming that nitric oxide is a physio- In humans, an a-MSH analogue, melanotan II, logical mediator of penile erection at the level of the was reported to initiate erection in men with paraventricular nucleus.35 29 psychogenic erectile dysfunction. NO may mediate the actions of ACTH and 5-HT1C ACTH and a-MSH act on melanocortin receptors, agonists which elicit erections when injected and the cloning of ®ve different subtypes of into the intracerebroventricular system according melanocortin receptor has recently opened up to mechanisms unrelated to oxytocinergic neuro- new possibilities for the development of drugs.30 transmission.16 Argiolas et al showed that the peptides act in the hypothalamic periventricular region, and that grooming, stretching and yawning, but not Peripheral neuromediation penile erection, are mediated by melanocortin-4 receptors.28 The different structures of the penis receive sympathetic, parasympathetic, somatic, and sensory in- 2,38 Opioid peptides nervation. The distribution of autonomic nerves within the corpora cavernosa has been studied by several investigators, but the density of nerves and Endogenous opioid peptides have long been the quantitative relations between the number of assumed to be involved in the regulation of male nerve terminals and and the number of smooth sexual responses, since sexual dysfunction has been muscle cells have not been established. The nerves observed clinically in men chronically using opi- contain different transmitters, and the nerve ates, and copulatory behavior in male rats can be populations have been categorized as adrenergic, experimentally depressed with the systemic admin- cholinergic, and non-adrenergic, non-cholinergic istration of morphine or other opioids. Current data (NANC). The latter nerves may contain not only support the hypothesis that opioid m receptor neuropeptides, but also transmitters and transmit- stimulation centrally prevents penile erection by ter=modulator generating enzymes, such as NOS inhibiting mechanisms that converge upon central (NOS) and heme oxygenases (HO). Thus, it seems oxytocinergic neurotransmission of this function.2 that one important population of nerves in the Morphine injected into the paraventricular nucleus corpora cavernosa contain not only acetylcholine, of the hypothalamus of rats prevented noncontact but also NOS, vasoactive intestinal polypeptide penile erections and impaired copulation. These (VIP), and neuropeptide Y.39,40 effects of morphine are apparently mediated by a The nerves and vasculature of the penis produce prevention of the increased nitric oxide production and release transmitters and modulators, producing

International Journal of Impotence Research Neurotransmitters in ED K-E Andersson S29 contraction or relaxation. These transmitters= function, and a1B-, and=or a1L-AR subtype selective modulators interact in their control of the contrac- antagonists were suggested to represent advantages tile state of the corpus cavernosum smooth muscle. in the treatment of erectile dysfunction.45 This may In addition they may also have other important be correct. However, the distribution of a1-AR functions, some of which are discussed later. subtypes in the penis and systemic vasculature may not be the same in rats and humans, and the method of study may in¯uence the results. For example, Hussain and Marshall found in vitro that the a1D-AR predominated in several systemic rat Contraction-mediating transmitters and receptors vessels,46 which may not be the case in humans.47 Tong and Cheng found a1A-ARs to be responsible for Noradrenaline and a-adrenoceptors: Penile ar- the contractile response of rat corpus cavernosum, teries, cavernosal smooth muscle, and the principal which does not seem to be in agreement with the in penile veins receive adrenergic innervation,41 and it vivo data.48 is generally accepted that the penis is kept in the Expression of mRNA for a2A-, a2B-, and a2C-ARs in ¯accid state mainly via a tonic activity in these whole human corpus cavernosum tissue has been 43 nerves through release of NA. Noradrenaline stimu- demonstrated, a2A-ARs predominating. Radioli- lates a-adrenoceptors (ARs) in the penile vascula- gand binding studies with a highly selective ligand ture contracting the helicine vessels, and in the for a2-ARs revealed speci®c a2-AR binding sites, and corpus cavernosum, contracting the trabecular functional experiments showed that the selective a2- smooth muscle.2 Noradrenaline stimulates not only AR agonist, UK 14,304, induced concentration- a-, but also b-ARs. However, in the human corpus dependent contractions of isolated strips of human cavernosum, receptor binding studies have revealed corpus cavernosum smooth muscle.49 Whether or that the density of a-ARs is almost 10 times higher not these a2-ARs are of importance for the contrac- than that of b-ARs. It is not only the number of tile regulation of tone in corpus cavernosum smooth receptors that is of importance. Androgens may muscle is still unclear. Prejunctional a2-ARs have regulate the a-AR responsiveness of cavernous been shown to modulate stimulus-evoked release of smooth muscle. Compared to normal rats, castrated NA from nerves in the human corpus cavernosum, animals showed an enhanced reactivity to a1-AR stimulation inhibiting the release of the amine. 42 stimulation. However, stimulation of prejunctional a2-ARs in Both a1- and a2-ARs have been demonstrated in horse penile resistance arteries was shown also to human corpus cavernosum tissue, but available inhibit NANC-transmitter release.50 This might be information supports the view of a functional one of the mechanisms by which NA maintains predominance of a1-ARs. This may also be the case detumescence. in the penile vasculature, although a contribution of a2-ARs to the contraction induced by NA and electrical stimulation of nerves cannot be excluded.2 An important question is whether or not one of Endothelins and endothelin receptors: Endothe- the a -AR subtypes is more important than the 1 lins (ETs) have been demonstrated in penile erectile others for the contractile effects of NA. The subtypes tissues and it is suggested that they contribute to the of a -AR with high af®nity for prazosin, currently 1 maintenance of corporal smooth muscle tone.2 designated as a , a ,anda have been demon- 1A 1B D Cultured endothelial cells from the human corpus strated in human corporal tissue. In human corporal cavernosum, but not non-endothelial cells, ex- tissue, mRNAs for a1A, a1B, and, a1D have been identi®ed, the a - and a -ARs predominating. pressed ET-1 mRNA. In the endothelium of human 1A 1D cavernous tissue, intense ET-like immunoreactivity However, at the protein level, only a1A-anda1B- ARs were expressed.43 On the other hand, func- was observed; immunoreactivity was also observed tional studies of human corpus cavernosum tissue, in the cavernous smooth muscle. Binding sites for using receptor binding and isometric tension ex- ET-1 were demonstrated by autoradiography in the periments showed the presence of a -, a -, and vasculature and cavernous tissue. Both ETA and ETB 1A 1B receptors have been found in human corporal a1D-ARs, and the NA-induced contraction in this tissue was suggested to be mediated by two or smooth muscle membranes, and it cannot be possibly all of the receptor subtypes.41 excluded that both receptor subtypes are functional. ET-1 potently induces slowly developing, long- An additional a1-AR subtype with low af®nity for prazosin (a1L), which is not yet fully characterized, lasting contractions in different penile smooth may occur in, for example, vascular smooth muscle. muscles: corpus cavernosum, cavernous artery, 2 The possibility that the a1L-AR subtype may be of deep dorsal vein, and penile circum¯ex veins. importance in human penile erectile tissues was Contractions can be evoked in human corpus 44 recently suggested. In rats, a1B-, and a1L-AR cavernosus tissue also by ET-2 and ET-3, although subtypes seem functionally relevant for erectile these peptides have a lower potency than ET-1.51

International Journal of Impotence Research Neurotransmitters in ED K-E Andersson S30 The contractions induced by ET-1 may be depen- responses to electrical stimulation. It was suggested dent on both transmembrane calcium ¯ux (through that eNOS is essential for erection, not only in nNOS voltage-dependent and=or receptor-operated cal- deleted, but also in normal mice. eNOS is expressed cium channels) and on the mobilization of inositol in cavernosal smooth muscle and could be a main trisphosphate- (IP3-) sensitive intracellular calcium source of NO alongside nNOS. However, it was stores. ET-1 may function not only as a long-term shown that mice carrying a mutation in the nNOS regulator of corporal smooth muscle tone, but also as gene are still able to express an alternatively spliced modulator of the contractile effect of other agents, eg mRNA of nNOS, which could be the source of NO in NA, or as a modulator of cellular proliferation and nNOS mutant mice.54 Since NO is produced, and phenotypic expression. the guanylyl=cGMP=cGKI pathway apparently is Even if much of the in vitro information available intact in the corpus cavernosum of these animals, suggests that ETs may be of importance for erectile they are not suitable to test whether or not this physiology and pathophysiology, the role of the pathway is a prerequisite for normal penile erection. peptides in vivo is unclear. cGMP signals via three different receptors in eukaryotic cells, including ion channels, phospho- diesterases, and protein kinases. At present, however, the molecular targets which are activated by cGMP and ®nally execute the relaxation of penile Relaxation-mediating transmitters and receptors smooth muscle are not known. Two different cGMP- dependent protein kinases (cGK I and II) have been Acetylcholine and cholinergic receptors: The im- identi®ed in mammals. Inactivation of cGKI in mice portance of parasympathetic nerves for producing abolished both NO=cGMP-dependent relaxation of penile erection has been well established.2 Penile vascular and intestinal smooth muscle and inhibi- tissues from humans and several animal species are tion of platelet aggregation, causing hypertension, 55 rich in nerves staining for acetylcholine (ACh) intestinal dysmotility and abnormal hemostasis. esterase.38 From these nerves, ACh can be released cGKI-de®cient mice show a very low ability to by transmural electrical ®eld stimulation. In isolated reproduce. Corpus cavernosum tissue from these corpus cavernosum cells, carbachol consistently mice has an inability or markedly reduced ability to produced contraction. This means that relaxation relax in response to neuronally or endothelially 56 induced by ACh can be obtained either by inhibition released, or exogenously administered NO. Ana- of the release of a contractant factor, eg NA, and=or lysis of the NO=cGMP-induced relaxation clearly is produced by the release of a relaxation-producing showed that cGKI is the major mediator of the cGMP factor, eg NO. signaling cascade in mouse corpus cavernosum It is important to stress that parasympathetic tissue. Its absence cannot be compensated by the activity is not equivalent with the actions of ACh; cAMP signaling cascade that relaxes normal and other transmitters may be released from cholinergic cGKI-null penile erectile tissue to a similar extent. nerves.39,40 Parasympathetic activity may produce Taken together, these ®ndings suggest that activa- penile tumescence and erection by inhibiting the tion of cGKI is a key step in the signal cascade release of NA through stimulation of muscarinic leading to penile erection. receptors on adrenergic nerve terminals, and=or by releasing NO and, eg vasodilating peptides from nerves and endothelium. Vasoactive intestinal polypeptide (VIP) and VIP receptors: The penis of humans as well as animals Nitric oxide and the guanylate cyclase=cGMP is richly supplied with nerves containing VIP.38 The pathway: An important role for NO in the relaxa- majority of these nerves also contain immunoreac- tion of corpus cavernosum smooth muscle and tivity to NOS, and colocalization of NOS and VIP vasculature is widely accepted.2,52 In vitro, several within nerves innervating the penis of both animals investigators have shown that both acetylcholine- and humans has been demonstrated by many and neuronally mediated relaxation in animal and investigators. It seems that most of these NO- and human corpus cavernosum involves release of NO, VIP-containing neurons are cholinergic, since they or a NO-like substance.2 Both the endothelium also contain VAChT,39 which is a speci®c marker for and=or the nerves innervating the corpus caverno- cholinergic neurons. VIP receptors (types 1 and 2), sum may be the source of the NO, and thus, more linked via a stimulatory G-protein to adenylyl than one isoform of NOS can be involved. cyclase, are considered to mediate the actions of Mice lacking nNOS have erections, show normal the peptide.57 The importance of the different mating behaviour, and respond with erection to subtypes of VIP in penile tissues have not been electrical stimulation of the cavernous nerves.53 clari®ed. VIP related peptides, eg pituitary adenylyl Surprisingly, isolated corporal tissue from both wild cyclase-activating peptide (PACAP), which has been type and nNOS-deleted animals showed similar found to be colocalized with VIP in penile nerves,

International Journal of Impotence Research Neurotransmitters in ED K-E Andersson S31 seem to act through one of the VIP receptors. The partly known. Peripherally, the different steps stimulatory effect of VIP on adenylyl cyclase leads involved in neurotransmission, impulse propaga- to an increase in cAMP, which in turn activates tion and intracellular transduction of neural signals cAMP-dependent protein kinase. in penile smooth muscles also need further inves- Undeniably, VIP has an inhibitory and relaxation- tigation. This work was supported by grants from producing effect on strips of human corpus caver- the Swedish Medical Research Council (grant no nosum tissue and cavernosal vessels in vitro, but it 6837), and the Medical Faculty, University of Lund, has been dif®cult to convincingly show that VIP Sweden. released from nerves is responsible for relaxation of penile smooth muscle in vitro or in vivo.2,58 Whether or not VIP has a role as a neurotransmitter or modulator of neurotransmission in the penis has not References been established. Even if its physiological role in penile erection and in erectile dysfunction remains 1 de Groat WC, Booth AM. Neural control of penile erection. In: to be settled, VIP receptors in the penis are an Maggi CA, ed. The Autonomic Nervous System. Vol. 6, interesting therapeutic target. Chapter 13, Nervous Control of the Urogenital System. Harwood Academic Publishers: London, UK, 1993, pp 465 ± 524. Prostanoids and prostanoid receptors: Prostaglan- 2 Andersson K-E, Wagner G. Physiology of penile erection. Physiol Rev 1995; 75: 191 ± 236. dins (PGs) and thromboxanes are locally acting 3 Argiolas A, Melis MR. Neuromodulation of penile erection: an hormones derived from arachidonic acid by the overview of the role of neurotransmitters and neuropeptides. action of cyclooxygenases. Human corpus caverno- Prog Neurobiol 1995; 47: 235 ± 255. sum tissue has the ability to synthetize various 4 Rampin O, Bernabe J, Giuliano F. Spinal control of penile prostanoids, and also has the ability to locally erection. World J Urol 1997; 15: 2 ± 13. 5 McKenna KE. Central control of penile erection. Int J Impot 2 metabolize them. The production of prostanoids Res. 1998; 10(Suppl 1): S25 ± S34. can be modulated by oxygen tension and suppressed 6 Andersson K-E et al. Current Research and Future Therapies. by hypoxia. There are ®ve primary active prostanoid In: Jardin A, Wagner G, eds. 1st International Consultation on metabolites: PGD , PGE , PGF , PGI , and TXA , and Erectile Dysfunction. Plymbridge Distributors Ltd: UK, 2000. 2 2 2 2 2 7 Andersson K-E. Pharmacology of lower urinary tract smooth it has been proposed that there are ®ve major groups muscles and penile erectile tissues. Pharmacol Rev 1993; 45: of receptors, corresponding to these metabolites, 254 ± 308. that mediate their effects, namely DP, EP, FP, IP, and 8 Giuliano F et al. Stimulation of the medial preoptic area of the TP receptors. cDNAs encoding representatives of hypothalamus in the rat elicits increases in intracavernous each of these groups of receptors have been cloned, pressure. Neurosci Lett 1996; 209:1±4. 9 Chen KK, Chan SH, Chang LS, Chan JY. Participation of including several subtypes of EP receptors. The paraventricular nucleus of hypothalamus in central regulation prostanoid receptors are G-protein-coupled receptors of penile erection in the rat. J Urol 1997; 158: 238 ± 244. with differing transduction systems. 10 Chen KK et al. Elicitation of penile erection following Penile tissues may contain most of these groups activation of the hippocampal formation in the rat. Neurosci Lett 1992; 141: 218 ± 222. of receptors. However, their role in penile physiol- 11 Tang Y et al. Oxytocinergic and serotonergic innervation of ogy is still far from established. Prostanoids may be identi®ed lumbosacral nuclei controlling penile erection in involved in contraction of erectile tissues via PGF2a the male rat. Neuroscience 1998; 82: 241 ± 224. 12 Bitran D, Hull EM. Pharmacological analysis of male rat sexual and thromboxane A2, stimulating TX and FP receptors and initiating phosphoinositide turnover, behavior. Neurosci Biobehav Rev 1987; 11: 365 ± 389. 13 Rehman J et al. Modi®cation of sexual behavior of Long-Evans as well as in relaxation via PGE1 and PGE2, male rats by drugs acting on the 5-HT1A receptor. Brain Res stimulating EP receptors (EP2=EP4) and initiating 1999; 821: 414 ± 425. an increase in the intracellular concentration of 14 Bancila M et al. 5-Hydroxytryptamine2C receptors on spinal cAMP. Prostanoids may also be involved in inhibi- neurons controlling penile erection in the rat. Neuroscience. 1999; 92: 1523 ± 1537. tion of platelet aggregation and white cell adhesion, 15 Bagdy G, Kalogeras KT, Szemeredi K. Effect of 5-HT1C and 5- and recent data suggest that prostanoids and HT2 receptor stimulation on excessive grooming, penile transforming growth factor-b1 (TGF-b1) may have a erection and plasma oxytocin concentrations. Eur J Pharmacol role in modulation of collagen synthesis and in the 1992; 229: 9 ± 14. regulation of ®brosis of the corpus cavernosum.59 16 Melis MR, Argiolas A. Role of central nitric oxide in the control of penile erection and yawning. Prog Neuropsycho- pharmacol Biol Psychiatry 1997; 21: 899 ± 922. 17 Chen KK, Chan JY, Chang LS. Dopaminergic neurotransmission at the paraventricular nucleus of hypothalamus in Conclusions central regulation of penile erection in the rat. J Urol 1999; 162: 237 ± 242. 18 Hull EM et al. Opposite in¯uence of medial preoptic D1 and The central regulation of the erectile process D2 receptors on genital re¯exes: implications for copulation. involves several transmitters, including dopamine, Life Sci 1992; 51: 1705 ± 1713. 19 Markowski VP et al. A D1 agonist in the MPOA facilitates serotonin, noradrenaline, nitric oxide, and peptides, copulation in male rats. Pharmacol Biochem Behav 1994; 47: such as oxytocin and ACTH=a-MSH, but is still only 483 ± 486.

International Journal of Impotence Research Neurotransmitters in ED K-E Andersson S32 20 Melis MR, Succu S, Argiolas A. Dopamine agonists increase 39 Hedlund P, Alm P, Andersson K-E. NO synthase in nitric oxide production in the paraventricular nucleus of the cholinergic nerves and NO-induced relaxation in the rat hypothalamus: correlation with penile erection and yawning. isolated corpus cavernosum. Br J Pharmacol 1999; 127: 349 ± Eur J Neurosci 1996; 8: 2056 ± 2063. 360. 21 Giuliano F, Rampin O. Alpha receptors in the central nervous 40 Hedlund P, Ny L, Alm P, Andersson K-E. Cholinergic nerves system and its effects on erection. J Androl 1999; 20: 683 ± 687. in human corpus cavernosum and spongiosum contain nitric 22 Rampin O. Pharmacology of alpha-adrenoceptors in male oxide synthase and heme oxygenase. J Urol 2000, (in press). sexual function. Eur J Urol 1999; 36(Suppl 1): 103 ± 106. 41 Andersson K-E., Hedlund P, Alm P. Sympathetic pathways 23 Rosaria Melis M, Spano MS, Succu S, Argiolas A. Activation and adrenergic innervation of the penis. Int J Impot Res 2000; of gamma-aminobutyric acid(A) receptors in the paraventri- 12: 1±8. cular nucleus of the hypothalamus reduces apomorphine-, N- 42 Reilly CM, Stopper VS, Mills TM. Androgens modulate the methyl-D-aspartic acid- and oxytocin-induced penile erection alpha-adrenergic responsiveness of vascular smooth muscle in and yawning in male rats. Neurosci Lett 2000; 281: 127 ± 130. the corpus cavernosum. J Androl 1997; 18: 26 ± 31. 24 Tang Y, Rampin O, Giuliano F, Ugolini G. Spinal and brain 43 Goepel M et al. Characterization of alpha-adrenoceptor circuits to motoneurons of the bulbospongiosus muscle: subtypes in the corpus cavernosum of patients undergoing retrograde transneuronal tracing with rabies virus. J Comp sex change. Neurol 1999; 414: 167 ± 192. J Urol 1999; 162: 1793 ± 1799. 25 Veronneau-Longueville F et al. Oxytocinergic innervation of 44 Davis B, Chapple, C, Chess-Williams R. The a1L-adrenoceptor autonomic nuclei controlling penile erection in the rat. mediates contraction in human erectile tissue. Eur J Urol 1999; Neuroscience 1999; 93: 1437 ± 1447. 35(Suppl 2): 102 (abstract 406). 26 Melis MR, Spano MS, Succu S, Argiolas A. The oxytocin 45 Sironi G et al. Effects of intracavernous administration of antagonist d(CH2)5Tyr(Me)2-Orn8-vasotocin reduces non- selective antagonists of alpha(1)-adrenoceptor subtypes on contact penile erections in male rats. Neurosci Lett 1999; erection in anesthetized rats and dogs. J Pharmacol Exp Ther 265: 171 ± 174. 2000; 292: 974 ± 981. 27 Melis MR, Succu S, Iannucci U, Argiolas A. Oxytocin 46 Hussain MB, Marshall I. Characterization of alpha 1-adreno- increases nitric oxide production in the paraventricular ceptor subtypes mediating contractions to phenylephrine in nucleus of the hypothalamus of male rats: correlation with rat thoracic aorta, mesenteric artery and pulmonary artery. Br J penile erection and yawning. Regul Pept 1997; 69: 105 ± 111. Pharmacol 1997; 122: 849 ± 858. 28 Argiolas A, Melis MR, Murgia S, Schioth HB. ACTH- and 47 Rudner XL et al. Subtype speci®c regulation of human alpha-MSH-induced grooming, stretching, yawning and penile vascular alpha(1)-adrenergic receptors by vessel bed and age. erection in male rats: site of action in the brain and role of Circulation 1999; 100: 2336 ± 2343. melanocortin receptors. Brain Res Bull 2000; 5: 425 ± 431. 48 Tong YC, Cheng JT. Subtyping of alpha 1-adrenoceptors 29 Wessells H et al. Synthetic melanotropic peptide initiates responsible for the contractile response in the rat corpus erections in men with psychogenic erectile dysfunction: cavernosum. Neurosci Lett 1997; 228: 159 ± 162. double-blind, placebo controlled crossover study. J Urol 49 Gupta S et al. The expression of functional postsynaptic 1998; 160: 389 ± 393. alpha2-adrenoceptors in the corpus cavernosum smooth 30 Wikberg JE. Melanocortin receptors: perspectives for novel muscle. Br J Pharmacol 1998; 123: 1237 ± 1245. drugs. Eur J Pharmacol 1999; 375: 295 ± 310. 50 Simonsen U et al. Prejunctional alpha 2-adrenoceptors inhibit 31 Melis MR, Succu S, Spano MS, Argiolas A. Morphine injected nitrergic neurotransmission in horse penile resistance arteries. into the paraventricular nucleus of the hypothalamus prevents J Urol 1997; 157: 2356 ± 2360. noncontact penile erections and impairs copulation: involve- 51 Saenz de Tejada I et al. Endothelin: localization, synthesis, ment of nitric oxide. Eur J Neurosci 1999; 11: 1857 ± 1864. activity, and receptor types in human penile corpus caverno- 32 Melis MR, Succu S, Argiolas A. Prevention by morphine of N- sum. Am J Physiol 1991; 261: H1078 ± H1085. methyl-D-aspartic acid-induced penile erection and yawning: 52 Burnett AL. Nitric oxide in the penis: physiology and involvement of nitric oxide. Brain Res Bull 1997; 44: 689 ± 694. pathology. J Urol 1997; 157: 320 ± 324. 33 Melis MR, Succu S, Iannucci U, Argiolas A. Prevention by 53 Burnett AL et al. Nitric oxide-dependent penile erection in morphine of apomorphine- and oxytocin-induced penile mice lacking neuronal nitric oxide synthase. Mol Med 1996; 2: erection and yawning: involvement of nitric oxide. Naunyn 288 ± 296. Schmiedebergs Arch Pharmacol 1997; 355: 595 ± 600. 54 Eliasson MJ, Blackshaw S, Schell MJ, Snyder SH. Neuronal 34 Lorrain DS, Matuszewich L, Howard RV, Du J, Hull EM. Nitric nitric oxide synthase alternatively spliced forms: prominent oxide promotes medial preoptic dopamine release during male functional localizations in the brain. Proc Natl Acad Sci USA rat copulation. Neuroreport 1996; 8: 31 ± 34. 1997; 94, 3396 ± 3401. 35 Melis MR, Succu S, Mauri A, Argiolas A. Nitric oxide 55 Pfeifer A et al. Defective smooth muscle regulation in cGMP production is increased in the paraventricular nucleus of the kinase I-de®cient mice. EMBO J 1998; 17: 3045 ± 3051. hypothalamus of male rats during non-contact penile erections 56 Hedlund P et al. Erectile dysfunction in cyclic GMP- and copulation. Eur J Neurosci 1998; 10: 1968 ± 1974. dependent kinase I-de®cient mice. Proc Natl Acad Sci USA 36 Sato Y et al. Effect of the nitric oxide level in the medial 2000; 97: 2349 ± 2354. preoptic area on male copulatory behavior in rats. Am J 57 Fahrenkrug J. Transmitter role of vasoactive intestinal peptide. Physiol 1998; 274: R243 ± R247. Pharmacol Toxicol 1993; 72: 354 ± 363. 37 Sato Y et al. The effects of alterations in nitric oxide levels in 58 Okamura T, Ayajiki K, Toda N. Monkey corpus cavernosum the paraventricular nucleus on copulatory behavior and relaxation mediated by NO and other relaxing factor derived re¯exive erections in male rats. J Urol 1999; 162: 2182 ± 2185. from nerves. Am J Physiol 1998; 274: H1075 ± H1081. 38 Dail WG. Autonomic innervation of male reproductive genita- 59 Moreland RB et al. PGE1 suppresses the induction of collagen lia. In: Maggi CA, (ed). The Autonomic Nervous System. Vol. 6, synthesis by transforming growth factor beta 1 in human Chapter 3, Nervous Control of the Urogenital System. Harwood corpus cavernosum smooth muscle. J Urol 1995; 153: 826 ± Academic Publishers: London, UK, 1993, pp 69 ± 101. 834.

International Journal of Impotence Research Neurotransmitters in ED K-E Andersson S33 Appendix Dr Porst: I have some psychogenic patients who use oxytocin nasal spray and are quite happy. No pharmaceutical company will go into this market Open discussion following Dr Andersson's because it's patent free, but I agree that oxytocin has presentation a role, especially in psychogenic patients.

Dr Heaton: Do you think that blocking vasoconstric- Dr Wyllie: How important is central control in the tive signals in the penis is going to be as appropriate erectile process? Many visceral functions operate a modality for managing erectile dysfunction as largely independently of central controls; is there vasodilatory agents like sildena®l? something different in erectile tissue? Dr Andersson: Yes, if we think that sympathetic Dr Andersson: Yes, it's important not to isolate the activity is one of the factors that causes ED, it is erectile function from sexual behavior. Central control, appropriate to counteract it; for example, in psycho- that is, oxytocin, is involved in many interesting genic impotence. In young men with an increased functions that could be important for sexuality. If you sympathetic tone, it would be appropriate to try to interfere with central mechanisms, you also affect reduce that tone in the corpora cavernosa. And it's bladder and gut, and it will be dif®cult to have important to ®nd the right alpha blocker, because, as selectivity and avoid side effects. We will probably you know, they have a lot of other effects. endupwithcombinationtherapieswherewestimulate centrally and enhance target organ sensitivity, that is, the corpora cavernosa. We can achieve individually tailored, good therapies for patients without very many side effects.

International Journal of Impotence Research