Blackwell Science, LtdOxford, UKJSMJournal of Sexual Medicine1743-6095Journal of Sexual Medicine 2005200524478491Original ArticleCyclic Nucleotide SignalingLin et al.

478

ORIGINAL RESEARCH—BASIC SCIENCE Cyclic Nucleotide Signaling in Cavernous Smooth Muscle Ching-Shwun Lin, PhD, Guiting Lin, MD, PhD, and Tom F. Lue, MD Knuppe Molecular Urology Laboratory, Department of Urology, University of California, San Francisco, CA, USA DOI: 10.1111/j.1743-6109.2005.00080.x

ABSTRACT

Introduction. Penile erection depends on cavernous smooth muscle relaxation that is principally regulated by cyclic nucleotide signaling. It is hoped that a comprehensive review of publications relevant to this subject will be helpful to both scientists and clinicians who are interested in the sciences of erectile function/dysfunction. Aims. To review the roles of extracellular signaling molecules, their receptors, intracellular effectors, and phosphodiesterases in cyclic nucleotide signaling that leads to cavernous smooth muscle relaxation. The involvement of these molecules in the development of erectile dysfunction and the possibility of using them as therapeutic agents or targets are also discussed. Methods. Entrez, the search engine for life sciences, was used to search for publications relevant to the topics of this review. Keywords used in the searches included vascular, cavernous, penis, smooth muscle, signaling molecules (adenosine, nitric oxide, etc.), and key elements in the cyclic nucleotide signaling pathways (cAMP, cGMP, cyclases, PKG, PKA, etc.). Articles that are dedicated to the study of erectile function/dysfunction were prioritized for citation. Results. More than 1,000 articles were identified, many of which are studies of the vascular system and are therefore reviewed but not cited. Studies on erectile function have identified both cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP) signaling pathways in cavernous smooth muscle. Many signaling molecules of these two pathways have been shown capable of inducing erection when administered intracavernously. However, for sexually induced erection, nitric oxide (NO) is the responsible signaling molecule and it passes on the signal through soluble guanyl cyclase (sGC), cGMP, and protein kinase G (PKG). Conclusions. The NO/sGC/cGMP/PKG pathway is principally responsible for sexually stimulated erection. Detumescence is mainly carried out by the degradation of cGMP by phosphodiesterase 5. Both cAMP and cGMP signaling pathways are susceptible to genetic and biochemical alterations in association with erectile dysfunction. Several key elements along these pathways are potential therapeutic targets. Key Words. Cavernous Smooth Muscle; Penis; cAMP; cGMP; Cyclases; PKG; PKA; Phosphodiesterase; Nitric Oxide Synthase; PDE5 Inhibitors; Molecular Biologic Studies of Sexual Function; Male Erectile Disorder

Introduction

M

edicines that affect cyclic nucleotide signaling have been used to treat erectile dysfunction (ED) for, perhaps, thousands of years [1]. However, the importance of cyclic nucleotide signaling in erectile function was not fully appreciated until the introduction of sildenafil in 1998. The tremendous success of sildenafil has drawn many researchers into taking a closer look at the J Sex Med 2005; 2: 478–491

cyclic nucleotide signaling pathways in the erectile tissue. Their studies have unequivocally established that cyclic guanosine monophosphate (cGMP) signaling in cavernous smooth muscle cells (CSMC) plays a critical role in penile erection, while cyclic adenosine monophosphate (cAMP) signaling, a lesser role. Yet, both cGMP and cAMP signaling pathways are involved in the pathogenesis of ED and have been exploited for the treatment of ED. Therefore, this article

Cyclic Nucleotide Signaling intends to provide a comprehensive review of both the major and minor signaling pathways that are mediated by cAMP and cGMP. While highly analogous, especially within the intracellular compartment, these two signaling pathways share little in common at the extracellular signal–receptor interface. In the cGMP pathway, the signaling molecule binds to, and activates its effector, whereas in the cAMP pathway the signaling molecule does not bind to its effector. What follows is a summary of the three types of interactions between the extracellular signaling molecules and their receptors and effectors. Graphic representation of these interactions is shown in Figure 1. 1. Direct activation of the soluble guanyl cyclase (sGC) in the cytoplasm, as exemplified by nitric

479 oxide (NO). After its release from the endothelium or nerve endings, NO enters CSMC by diffusion. It then binds to, and activates the sGC that catalyzes the synthesis of cGMP. Thus, sGC acts as both a receptor and an effector for NO. 2. Direct activation of the particulate guanyl cyclase (pGC) in the cytoplasmic membrane, as exemplified by natriuretic peptides. These peptides are secreted from heart, brain, and the endothelium and act as endocrine or paracrine hormones. They bind to, and activate the pGC in the cytoplasmic membrane of CSMC. Thus, the pGC acts as both receptors and effectors for natriuretic peptides. 3. Indirect activation of the particulate adenyl cyclase (pAC) in the cytoplasmic membrane via

Figure 1 Cyclic nucleotide signaling cascades leading to cavernous smooth muscle relaxation. Arrows indicate binding or activation (see text for details). AC = adenyl cyclase; ANP = atrial natriuretic peptide; ATP = adenosine triphosphate; BNP = brain natriuretic peptide; cAMP = cyclic adenosine monophosphate; cGMP = cyclic guanosine monophosphate; CGRP = calcitonin gene-related peptide; CNP = C-type natriuretic peptide; GC = guanyl cyclase; GPCR = G proteincoupled receptor; GTP = guanosine triphosphate; IP3R = inositol 1,4,5-trisphosphate receptor; NO = nitric oxide; PDE = phosphodiesterase; PG = prostaglandin; PKA = protein kinase A; PKG = protein kinase G; sGC = soluble GC; VIP = vasoactive intestinal peptide.

J Sex Med 2005; 2: 478–491

480 direct activation of a G protein-coupled receptor (GPCR). Each of these signaling molecules binds to a specific receptor that belongs to the GPCR superfamily, causing the receptor to augment the exchange of guanosine triphosphate (GTP) for guanosine diphosphate (GDP) on the a subunit of the heterotrimeric G protein. Binding of GTP to Ga subunit causes it to dissociate from the Gbg subunits, and both the Ga subunit and the Gbg subunits are now able to activate the pAC, triggering the production of cAMP. Because the Ga subunit has an intrinsic GTPase activity that hydrolyzes GTP to GDP, it can convert itself from the GTP-bound form to the GDP-bound form. The GDP-bound, inactive Ga then reassociates with the Gbg subunits, thus returning the G protein to the resting state and terminating the transmission of the initial signal. While this signaling pathway is used by most cyclic nucleotide signaling molecules, there are instances when “activation” of adenyl cyclase (AC) can have an inhibitory rather than stimulatory effect [2]. For example, the a2 adrenoceptors are predominately coupled to the G protein containing an inhibitory (ai) rather than a stimulatory (as) a subunit. Therefore, their activation by epinephrine and epinephrine agonists results in the inhibition of AC [3]. Once activated, AC and guanyl cyclase (GC) convert adenosine triphosphate (ATP) to cAMP and GTP to cGMP, respectively. Subsequent binding of cAMP to protein kinase A (PKA) or cGMP to protein kinase G (PKG) results in the activation of the respective kinase, which then phosphorylates specific downstream targets such as ion channels, leading to relaxation of CSMC. Termination of the signaling event is carried out by phosphodiesterases (PDEs) that catalyze the hydrolysis of cAMP to AMP and cGMP to GMP. Like other signal transduction pathways (e.g., those of growth factors and cytokines), cyclic nucleotide signaling operates not in a linear fashion as described above, but rather in a complex network that encompasses many interweaving pathways. However, because the linear pathways are the best characterized and appear to be the most important in the physiological, as well as pathological, processes, this review is organized along the two linear (main) pathways. Whenever relevant and available, any branching pathways J Sex Med 2005; 2: 478–491

Lin et al. will be drawn at the appropriate junctions along the main pathways. It should also be pointed out that CSMC share many physiological and biochemical properties with vascular smooth muscle cells (VSMC), and it is with VSMC that a great deal of knowledge on cyclic nucleotide signaling was learned. Thus, many references used in this review are studies on the vasculature.

CAMP Signaling Pathway

This section first reviews cAMP signaling molecules that include adenosine, calcitonin generelated peptides (CGRP), prostaglandins, and vasoactive intestinal peptides (VIP), and their specific receptors.

Adenosine Adenosine is released from a variety of cells as a result of increased metabolic rates, and its actions on the vasculature are most prominent when oxygen demand is high [4]. However, the vascular response to the action of adenosine can be either relaxation or constriction, depending on which type of the adenosine receptor (AR) is activated. Currently, four AR subtypes (A1, A2A, A2B, and A3) belonging to the GPCR superfamily have been recognized [4]. In general, the A1 receptor is believed to be coupled to Gi and Go proteins of the G protein family, and its activation results in inhibition of AC and activation of phospholipase C, both of which lead to vasoconstriction. The A2 receptors are coupled to the Gs proteins, and their activation results in the stimulation of AC and thus vasorelaxation. The A3 receptor is coupled to Gi and Gq proteins, and its activation results in the activation of phospholipase C/D and the inhibition of AC, leading to vasoconstriction. The differential distribution of these AR subtypes largely determines whether a particular vessel relaxes or contracts as a result of adenosine stimulation [4]. Whether adenosine plays a role in physiological penile erection is unclear. Intracavernous injection of adenosine has been shown to induce full erection in the dog [5], and the A2A receptor was shown to mediate the relaxation effects of adenosine in human and rabbit corpus cavernosum [6]. However, pelvic nerve stimulation-induced tumescence in anesthetized dogs was not affected by AR inhibitor 8-SPT, suggesting the lack of involvement of adenosine or its receptors in physiological erection [7]. Furthermore, intracavernous injection of

Cyclic Nucleotide Signaling adenosine failed to induce erection in human volunteers [6].

CGRP Family CGRP, amylin, and adrenomedullin are members of the CGRP family. These short-chain peptides are potent vasodilators released from perivascular nerve fibers. They act through the calcitonin receptor-like receptor (CRLR) that belongs to the GPCR superfamily [8]. However, unlike other GPCR family members, the CRLR itself is nonfunctional unless aided by two additional proteins, namely, the CGRP-receptor component protein (RCP) and the receptor activity-modifying protein (RAMP1, RAMP2, or RAMP3). RCP interacts with the CRLR, RAMP1, and RAMP2 and facilitates signal transduction by adrenomedullin and CGRP. The RAMP proteins act as molecular chaperones for trafficking of the CRLR to the cell surface. They also influence the CRLR’s preferential binding to either CGRP or adrenomedullin— RAMP1 favors CGRP, and RAMP2 or RAMP3 favors adrenomedullin. The CGRP levels found in the penis, bladder, kidney, testes, and adrenal gland gradually increased up to maturity and then rapidly declined in the aging rats [9]. When CGRP is administered intracavernously in ED patients, a dose-related increase in penile arterial inflow (and erection) occurs [10]. Adenovirus-mediated gene transfer of CGRP also enhances erectile responses in aged rats, apparently through an increase of cAMP levels in the corpora cavernosa [11]. Intracavernous administration of adrenomedullin also results in cavernous relaxation; however, the effect is through an NO/cGMP instead of the cAMP pathway [12]. Prostaglandins Prostaglandins (PGs) are a family of eicosanoids capable of initiating numerous biological functions. The prime mode of PG action is through specific PG receptors that all belong to the GPCR family. There are at least nine known PG receptor subtypes in mouse and man, as well as several additional splice variants with divergent carboxyl termini [13]. Four of the subtypes (EP1–EP4) bind PGE2; two (DP1 and DP2) bind PGD2, and the other three subtypes (FP, IP, and TP) bind PGF2a, PGI2, and TXA2, respectively. On the basis of signaling attributes, the PG receptors are classified into three types. The “relaxant” receptors IP, DP1, EP2, and EP4 are coupled to an ascontaining G protein and therefore capable of

481 stimulating AC to increase intracellular cAMP. The “contractile” receptors EP1, FP, and TP are coupled to an aq-containing G protein, which activates phospholipase C instead of AC. Therefore, these contractile receptors do not signal through the cAMP pathway, and their signaling outcome is an increase of intracellular calcium. The EP3 receptor is also a contractile receptor, but it is coupled to an ai-containing G protein that inhibits AC to result in a decrease of cAMP formation. Animal and human corpus cavernosum produces several PGs, including PGF2a, PGE2, PGD2, PGI2, and TXA2 [14,15]. In studies on isolated human penile tissue, different PGs have been shown to elicit different effects on human corpus cavernosum, corpus spongiosum, and cavernous artery [16]. While PGF2a, PGI2, and TXA2 contract the corpus cavernosum and corpus spongiosum, PGE1 and PGE2 (but not PGI2) relax the corpus cavernosum and spongiosum that have been precontracted with norepinephrine or PGF2a. Likewise, PGE1, but not PGI2, increases the arterial blood flow when injected into the monkey’s corpus cavernosum [17]. Therefore, while PGI2 is the predominant vasorelaxant in blood vessels, its action in the erectile tissue is either contractile or neutral. This disparity in the action of PGI2 between blood vessels and the erectile tissue and the difference between the effects of PGI2 and PGE1/2 in the erectile tissue are most likely due to differences in the distribution of PG receptors. Indeed, recent studies showed that in the corpus cavernosum the relaxant effects of prostanoids are mediated by EP2- and/ or EP4-receptors (for PGE1 and PGE2), but not IP receptor (for PGI2) [18,19]. While the production of PGs and the expression of PG receptors in the erectile tissue have been clearly demonstrated, their roles in physiological erection are yet to be defined. On the other hand, the erectogenic effects of PGE1 as a pharmaceutical agent have been extensively documented. First described 16 years ago, intracavernous injection of PGE1 is one of the most effective and safest treatments for ED [20]. Transurethral application of PGE1 is also an effective alternative.

VIP The human or animal penis is richly supplied with nerves containing VIP [21]. The mature VIP peptide contains 28 amino acid residues and is derived from prepro-VIP, which consists of 170 amino acid residues. The primary structure of VIP is J Sex Med 2005; 2: 478–491

482 closely related to pituitary AC-activating polypeptide and, to a much lesser extent, secretin, glucagon, gastric inhibitor polypeptide, and helodermin-like peptides. Two subtypes of VIP receptors, VPAC1 and VPAC2, belonging to the GPCR family have been cloned from human and rat tissues. VPAC2, but not VPAC1, mRNA was identified in cultured rat CSMC [22]. Intracavernous injection of VIP was shown to induce penile erection in the dog [23]. In men, intracavernous injection of VIP did not produce a rigid erection but improves success rates when combined with papaverine and phentolamine [24]. However, it has been shown that VIP release is not essential for neurogenic relaxation of human cavernous smooth muscle [25] and VIP failed to stimulate cAMP production in cultured human CSMC [26]. Aging does not affect the expression level of VIP mRNA in rat cavernous smooth muscle [27]. AC

As discussed earlier, signaling molecules in the cAMP signaling pathway bind to, and activate specific cytoplasmic membrane receptors, which, through their coupled G proteins, activate ACs. To date, nine membrane-bound isoforms and one soluble form of mammalian AC have been cloned and characterized [2]. The structure of a membrane-bound AC includes a short variable amino terminus, followed by six transmembrane spans, a large cytoplasmic domain, a second set of six transmembrane regions, and another large cytoplasmic domain that includes the carboxyl terminus. The overall similarity among the different ACs is roughly 60%: the most conserved sequences are located in the two cytoplasmic domains and range from 50% to 90%. While different membranebound ACs are regulated differently, they are all stimulated by the GTP-bound form of the Ga subunit and are all (except AC9) stimulated by forskolin. Moreover, all members of the membrane-bound ACs are inhibited by P-site analogs, which are essentially adenine nucleoside 3¢polyphosphates. It has been shown that cAMP formation in response to forskolin was reduced in the corpus cavernosum of alloxan-induced diabetic rabbits, suggesting an impaired AC function in diabetes mellitus [28]. PKA

PKA, also called cAMP-dependent kinase (cAK), is the principal receptor for cAMP, and it mediates J Sex Med 2005; 2: 478–491

Lin et al. the vast majority of the cellular effects of cAMP by phosphorylating a wide variety of downstream targets both in the cytoplasmic and the nuclear compartments [29]. PKA is composed of two regulatory (R) and two catalytic (C) subunits that form a tetrameric holoenzyme R2C2. Binding of cAMP to the R subunits causes the holoenzyme to dissociate into an R2(cAMP)4 dimer and two free catalytically active C subunits. The C subunits are encoded by three different genes, Ca, Cb, and Cg, while the regulatory subunits are expressed from four different genes, RIa, RIb, RIIa, and RIIb. Two forms of the PKA holoenzyme exist: type I, formed by RIa and RIb dimers, and type II, formed by RIIa and RIIb dimers. These isozymes may form from either homo- or heterodimers of the R subunits yielding holoenzyme complexes of PKA with a number of combinatorial configurations including RIa2C2, RIb2C2, RIIa2C2, RIIb2C2, and RIaRIbC2. The presence of multiple C subunit genes further adds to the diversity and complexity of the various holoenzyme complexes that differ in biochemical and functional properties as well as patterns of expression and localization. These differences among the isozymes contribute to the broad specificity of PKA in a wide variety of physiological processes in response to cAMP signaling.

PKA Substrates

More than 100 different cellular proteins have been identified as physiological substrates of PKA, with more than 90% (135 out of 145) being phosphorylated at serine while the remainder at threonine [30]. The predominant target sequence (>50%) is Arg-Arg-X-Ser, in which Ser is the phosphate acceptor. Three PKA substrate proteins are discussed below; additional PKA substrates that are also PKG substrates will be discussed in the “PKG Substrates” section. 1. PDEs: Among the 11 mammalian PDE families that will be discussed in a later section, PDE3 and PDE4 are known PKA substrates in VSMC. Phosphorylated PDE3 and PDE4 exhibit higher cAMP-catalytic activities than their unphosphorylated counterparts; therefore, phosphorylation of each of these two PDEs is likely to constitute a feedback mechanism that helps blunt excessive cAMP signaling [31,32]. 2. cAMP-responsive element binding protein: PKA mediates many of the long-term effects

Cyclic Nucleotide Signaling of cAMP by phosphorylating the ubiquitous transcription factor cAMP-responsive element binding protein (CREB), which binds as a dimer to the cAMP-responsive element (CRE), TGACGTCA, in the target gene. Phosphorylation of CREB at Ser-133 by PKA and other kinases has been shown to mediate angiotensin II-induced hypertrophy of VSMC [33]. 3. ATP-sensitive potassium channel: ATPsensitive potassium (KATP) channels are composed of four inwardly rectifying subunits and four regulatory subunits. Endogenous vasodilators such as adenosine, CGRP, PGI, and VIP activate KATP channels through PKA, which phosphorylates KATP channels at multiple sites [34]. The expression and characteristics of KATP channel subunits in CSMC have been reported [35].

CGMP Signaling Pathway

Signaling molecules in the cGMP pathway include natriuretic peptides and NO that activate pGC and sGC, respectively.

Natriuretic Peptides The natriuretic peptide family is involved in the regulation of cardiovascular homeostasis and consists of atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), and C-type natriuretic peptide (CNP) [36]. The mature ANP has 28 amino acid residues after cleavage from a precursor peptide. BNP and CNP have 32 and 22 amino acid residues, respectively, and are also cleavage products of precursor polypeptides [36]. Whereas ANP and BNP are ligands for the NPR-A receptor, CNP is a ligand for the NPR-B receptor. NPR-A and NPR-B receptors are members of the GC family and are therefore also called GC-A and GC-B. ANP, BNP, and CNP also bind to a common receptor, NPR-C, which is a truncated form of NPR-A. NPR-C lacks GC activity [37] and should not be confused with GC-C of the GC family. In many tissues, NPR-C is more abundant than NPR-A or NPR-B, and because of its lack of GC activity, NPR-C has been regarded as a “clearance” receptor whose primary function is the removal of natriuretic peptides from the system [37]. However, a recent study shows that CNP activates NPR-C, which is then coupled to Gi protein and the smooth muscle G protein-gated inwardly rectifying K+ channel, leading to vasorelaxation [38].

483 The effects of ANP, BNP, and CNP on cGMP production and smooth muscle relaxation in isolated human and animal corpus cavernosum and in cultured CSMC have been investigated [22,39,40]. The results indicate that CNP is the most potent natriuretic peptide and that it relaxes the isolated cavernous smooth muscle by binding to NPR-B. However, whether CNP and NPR-B play a role in physiological erection remains to be seen.

NO NO is a gaseous molecule produced from various tissues, in particular the endothelium and nerve. Because of its small size, NO can diffuse inside its target cell where it interacts with molecules that contain iron in either a heme or iron–sulphur complex. The most physiologically relevant receptor for NO is sGC, and the NO/sGC/cGMP pathway is responsible for the vasorelaxing effect of many endothelium-dependent vasodilators, including histamine, estrogens, insulin, and corticotrophin-releasing hormone, nitrovasodilators, and acetylcholine. This pathway is also principally responsible for penile erection. Synthesis of NO is catalyzed by NO synthase (NOS) that converts L-arginine and oxygen to Lcitruline and NO. NOS exists as three isoforms in mammals: neuronal NOS (nNOS) and endothelial NOS (eNOS) are preferentially expressed in neurons and endothelial cells, respectively, and inducible NOS (iNOS) in virtually all cell types. All three NOS isoforms have been identified in the corpus cavernosum, with nNOS and eNOS being considered responsible for initiating and sustaining erection, respectively [41]. Downregulation of nNOS expression has been found in the corpus cavernosum of aging rats [42], and NO-mediated corpus cavernous smooth muscle relaxation is impaired in aging rats [43]. Despite its secondary role in erectile function, the endotheliumdependent cavernous smooth muscle relaxation is attenuated in the aging rabbit, and this agingrelated defect is paradoxically accompanied with upregulated eNOS expression in both the endothelium and cavernous smooth muscle [44,45]. However, contradictory observation was made in another study that showed downregulation of eNOS in the corpus cavernosum of aging rats [46]. Gene transfer of nNOS or eNOS to the penis has been shown to augment erectile responses in aging rats [47,48]. Gene transfer of iNOS to the penis also resulted in enhanced intracavernous pressure [49]. However, despite these encouraging J Sex Med 2005; 2: 478–491

484 results, it should be noted that mice with disrupted nNOS or eNOS gene have normal erectile function [50,51]. Compensatory mechanisms, alternative splicing of the disrupted gene, and/or other unknown mechanisms are possibly involved in the preservation of erectile function in the knockout mice. GC

In mammals, seven membrane-bound (particulate) GC isoforms (GC-A to GC-G) and one soluble GC (sGC) have been identified [37,52]. Based on their ligand specificity, the pGCs have been classified as (i) natriuretic peptide receptors (GC-A and GC-B) that are activated by natriuretic peptides, including ANP, BNP, and CNP; (ii) intestinal peptide receptor (GC-C) that is activated by intestinal peptides including guanylin, uroguanylin, and lymphoguanylin; and (iii) orphan receptors (GC-D, GC-F, and GC-G). While the membrane-bound GC system is not known to play a role in physiological erection, expression of GCB in human and rat corpus cavernosum and induction of cavernous smooth muscle relaxation by CNP (ligand for GC-B) have been demonstrated recently [22,40]. sGC plays a pivotal role in erectile function as it provides the link between NO and cGMP that represent the extracellular and intracellular signaling molecules, respectively, in physiological erection [21]. sGC is a heterodimeric protein consisting of a and b subunits, each of which exists in two isoforms (a1, a2, and b1, b2) that are encoded by two separate genes [52]. Messenger RNAs of a1, a2, b1, and b2 subunits have been detected in human corpus cavernosum [53]. In animal studies, sGC activator YC-1 has been shown to cause erectile responses [54]. Immunohistochemical examination found similar sGC expression in the corpus cavernosum of potent and impotent patients [55]. PKG

PKG, also called cGMP-dependent kinase (cGK), is the principal receptor and mediator for cGMP signals. In mammals, PKG exists in two major forms, PKG-I and PKG-II, which are encoded by two separate genes. PKG-I, which is important for regulating cavernous smooth muscle tone, exists as a and b isoforms that arise through alternative splicing of the N-terminal region. The two isoforms are often co-expressed; whether they perform different physiological functions has not been determined. J Sex Med 2005; 2: 478–491

Lin et al. The PKG-I polypeptide contains three functional domains: the N-terminal, the regulatory, and the catalytic domains. The N-terminal domain has three functions: dimerization, autoinhibition, and localization. It contains a leucine zipper that holds two PKG-I molecules together as homodimers, it inhibits the catalytic domain in the basal state, and it interacts with specific cellular proteins that anchor PKG-I to different subcellular locations. The regulatory domain contains two tandem cGMP-binding sites, and the catalytic domain catalyzes the transfer of the g phosphate from ATP to a serine/threonine residue of the substrate protein. Binding of cGMP to the regulatory domain induces a conformational change that releases the autoinhibition of the catalytic domain by the N-terminal domain. Deletion of the N-terminal and regulatory domains results in a constitutively active PKG-I that retains substrate specificity, requires no cGMP for activation [56], and translocates to the nucleus [57]. The translocation of the PKG-I catalytic domain to the nucleus is reminiscent of the well-established nuclear translocation of the PKA catalytic subunits. Indeed, the PKG-I catalytic domain has been shown to activate CRE-dependent gene transcription [57]. However, whether the intact PKG-I is capable of nuclear translocation remains controversial [58]. Cavernous smooth muscle strips from PKG-I knockout mice cannot be relaxed by agents that raise cGMP levels and these mice have low ability to reproduce, presumably because of ED [59]. Interestingly, the AC activator forskolin was able to raise cAMP levels and subsequently relax the cavernous smooth muscle strips of the PKG-I knockout mice. These observations further affirm the essential role of the cGMP/PKG-I pathway in physiological erectile function. PKG Substrates

Far fewer PKG substrates have been identified compared with PKA substrates, and the majority of known PKG substrates are also phosphorylated by PKA. By screening peptide libraries, the optimal target sequence for PKG was determined to be Arg-Lys-X-(Ser/Thr), with Ser or Thr being the phosphate acceptor [60]. Some proven PKG substrates that are relevant to vascular and/or cavernous smooth muscle functions are discussed below: 1. Inositol 1,4,5-triphosphate (IP3) receptor: IP3 receptor mediates the rise of intracellular Ca2+

485

Cyclic Nucleotide Signaling

2.

3.

4.

5.

6.

7.

concentration and therefore smooth muscle contraction. It is phosphorylated by both PKG and PKA at two sites, resulting in inhibition of calcium release with subsequent smooth muscle relaxation [61]. Effects of IP3 on CSMC have been demonstrated [62]. IP3 receptor-associated PKG subtrate (IRAG): IRAG colocalizes with IP3 receptor and PKGI, and is preferentially phosphorylated by PKG-Ib [63]. Co-expression of IRAG and PKG-Ib in the presence of cGMP results in inhibition of calcium release. Phospholamban (PLB): PLB is a 5-kd protein that complexes with the sarcoplasmic Ca2+ATPase (SERCA) in heart, smooth muscle, and other tissues. Phosphorylation of PLB at Ser16 by PKA and PKG results in its dissociation from SERCA, thus allowing Ca2+ transport from the cytosol to the sarcoplasmic reticulum. PLB plays a major role in modulating smooth muscle intracellular Ca2+ but a minor role in cyclic nucleotide-mediated relaxation [64]. Calcium-activated potassium (BKCa) channel: Phosphorylation of BKCa channel by PKA or PKG leads to increased opening of this channel and decreased intracellular Ca2+ [65]. The expression and function of BKCa channel in CSMC have been demonstrated [66–68]. Transfection of the a (pore-forming) subunit of BKCa channel into corpus cavernosum has resulted in improved erectile function in rat ED models [69]. PDE5: PDE5, which is principally responsible for the hydrolysis of cGMP in VSMC and CSMC, is phosphorylated at Ser-92 by PKG and PKA. Phosphorylated PDE5 has higher cGMP-catalytic activity than unphosphorylated PDE5, suggesting a feedback mechanism regulating cGMP signaling [70]. Heat shock-related protein (HSP20): Phosphorylation of the 20-kd HSP20 is one of the major phosphorylation events associated with cAMP-induced vasorelaxation in carotid arterial smooth muscle. Phosphorylation of HSP20 at Ser-16 by PKA and PKG has been demonstrated in vitro. Transfection of the wild-type, but not the Ser-16, mutant into cultured VSMC leads to relaxation [71]. Transduction of phosphoHSP20 analogs also leads to dosedependent relaxation of carotid and coronary artery smooth muscle [71]. Myosin phosphatase (MP): MP, which dephosphorylates myosin light chain, leading to reduced smooth muscle tone, is a trimeric pro-

tein comprising a large (110–130 kd) regulatory myosin-binding subunit (MBS), a 37-kd catalytic subunit, and a 20-kd protein of unknown function. Activation of PKG-I by cGMP leads to MBS phosphorylation and MP activation. However, MP activation appears to depend not on MBS phosphorylation but rather on the presence of the leucine zipper in MBS [72,73]. 8. Phosphatase inhibitor-1 (PPI-1): In the presence of cGMP, PPI-1 is phosphorylated by PKG-I at Thr-35. It has been suggested that activated PPI-1 inhibits MP, leading to smooth muscle relaxation [74]. However, no difference in contractility of aorta or portal vein was noted between wild-type and PPI-1 knockout mice [75]. 9. GTPase RhoA: GTPase RhoA activates Rho kinase, which phosphorylates and inactivates MP, leading to smooth muscle contraction. Activity of GTPase RhoA is inhibited by PKG through phosphorylation, thus contributing to smooth muscle relaxation [76]. Cross-Activation

Cross-activation of PKG by cAMP has been shown in various vascular tissues, including rat aorta, pig coronary artery, and lamb pulmonary artery [77–79]. There is also substantial overlap between the two cyclic nucleotides in their ability to regulate CO2-induced cerebral vasodilatation of adult rat pial arteries [80]. Agents that stimulate cAMP production also cause activation of PKG in VSMC of coronary and pulmonary arteries [81,82]. Transfected PKG-Ia or PKG-Ib in COS7 fibroblasts was also similarly activated by cAMP and cGMP [83]. PKG-I in either aortic smooth muscle cells (AOSMC) or CSMC was activated at similar levels by cAMP and cGMP, except that PKG-I in CSMC of old rats was less activated by cGMP than by cAMP [83,84]. An earlier study showed that PKG was activated by cAMP in rat AOSMC, leading to a reduction in intracellular Ca2+ [85]. Overexpression of NOS, which leads to increased levels of NO and cGMP, was also shown to result in cross-activation of PKA [86]. Whether cross-activation of PKA and PKG plays physiological roles has been questioned in a study with PKG-I-deficient mice. In these mice, cGMP- but not cAMP-induced relaxation in aortic rings was impaired [87]. Furthermore, these mice were hypertensive, indicating PKG-I is the specific mediator of cGMP effects in regulating J Sex Med 2005; 2: 478–491

486 vascular muscle tone. However, in a more recent study with the same PKG-I-deficient mice, NO was able to relax vascular smooth muscle by increasing sGC activity and cGMP production with subsequent activation of PKA [88]. Thus, the newer data once more support the notion that PKA can be activated by cGMP and that this can lead to vasorelaxation. PDE

In each episode of cyclic nucleotide signaling, the increase of intracellular cAMP or cGMP concentration is typically two- to threefold over the basal level [89]. Decline of cyclic nucleotide levels occurs rapidly and often during the continued presence of the signaling hormone [89]. Termination of cyclic nucleotide signals is principally carried out by PDEs that catalyze the hydrolysis of cAMP and cGMP to AMP and GMP, respectively. Feedback mechanisms that increase PDE activities and/or expression by the increased cyclic nucleotide level facilitate the degradation of cyclic nucleotides [90–92]. The superfamily of mammalian PDEs consists of 11 families (PDE1 to PDE11) that are encoded from 21 distinct genes [1,93]. Each PDE gene usually encodes more than one isoform through alternative splicing or from alternate gene promoters. PDE1, PDE3, PDE4, PDE7, and PDE8 are multigene families, while PDE2, PDE5, PDE9, PDE10, and PDE11 are unigene families. PDE1, PDE2, PDE3, PDE10, and PDE11 hydrolyze both cAMP and cGMP. PDE4, PDE7, and PDE8 hydrolyze cAMP, while PDE5, PDE6, and PDE9 are specific for cGMP. Except for PDE6, which is specifically expressed in photoreceptor cells, all PDEs have been identified in the corpus cavernosum [94]. However, there is ample evidence that PDE5 is by far the principal PDE for the termination of cGMP signaling in the corpus cavernosum. For example, inhibition of the cGMP-catalytic activity of PDE5 by specific inhibitors has been shown to be highly effective in treating ED [95]. PDE3 appears to also play a role in penile erection, as demonstrated by the erectogenic effect of a PDE3-specific inhibitor milrinone [96]. Furthermore, while direct inhibition of PDE5 is the main mechanism through which sildenafil exerts its erectogenic effect, it has been shown that sildenafil also significantly increases cAMP concentration in isolated human cavernous tissue strips [97]. This effect is thought to involve PDE3 J Sex Med 2005; 2: 478–491

Lin et al. because cGMP, which is accumulated as a result of PDE5 inhibition by sildenafil, is capable of preventing cAMP degradation by competing for the same catalytic sites on the PDE3 molecules [89]. This attenuating effect of cGMP on the cAMPcatalytic activity of PDE3 is also believed to explain why inhibition of PKG could suppress the relaxing effect of forskolin in isolated human cavernous smooth muscle [98]. Incubation of cultured human CSMC with forskolin or PGE1 produced significant enhancement of cGMP accumulation [99]. Because forskolin and PGE1 are direct and indirect activators of AC, respectively, their effect on cGMP accumulation was surprising. By showing that cAMP acted as a competitive inhibitor for PDE5, the authors hypothesized that forskolin- or PGE1-induced cGMP accumulation was due to inhibition of PDE5 by increased cAMP concentration that resulted from AC activation. While it is clear that cAMP does not interact with PDE5 in an allosteric fashion, the mechanism through which cAMP inhibits PDE5 remains to be elucidated. Future Directions

As mentioned in the introduction, because of the similarity between cavernous and vascular smooth muscles, much of the knowledge in regard to cavernous smooth muscle function was learned from its vascular counterpart. Thus, this review would not be complete without at least mentioning two areas of research that have been pursued in the vascular, but not yet in the cavernous field. First, it has been shown that activation of ion channels by PKA in VSMC is regulated by A-kinase anchoring proteins [100,101], and G-kinase anchoring proteins for PKG have also been identified [102,103]. Second, while PKA and PKG are the principal effectors for cAMP and cGMP, respectively, alternative effectors have been identified. These include the cyclic nucleotide-gated channels, the cAMP-regulated guanine nucleotide exchange factors, PDE (particularly, PDE3), and aquaporin-1 [89,104,105]. Whether the kinase anchoring proteins or the alternative effectors are expressed in CSMC remains to be seen. Furthermore, while cyclic nucleotide signaling is undoubtedly important for cavernous smooth muscle relaxation, it cannot be ignored that the default state of the penis is flaccid, which, of course, is associated with cavernous smooth muscle contraction, and which is maintained through

487

Cyclic Nucleotide Signaling the Rho-kinase signaling pathway [106]. Thus, for erection to occur, is the Rho-kinase signaling pathway simply overridden by the NO/cGMP signaling pathway, or are there interactions between these two opposing pathways? Evidence for the latter scenario has been presented in a study in which PKG was shown to phosphorylate RhoA in vitro, and addition of PKG could inhibit RhoAinduced Ca+2 sensitization in permeabilized vascular smooth muscle [76]. More recent studies have also shown that PKA could inhibit RhoA activation [107] and the Rho-kinase pathway could antagonize the NO/cGMP pathway [108]. Thus, it can be reasonably expected that dynamic interactions exist between cyclic nucleotide signaling and Rho-kinase signaling pathways in the erectile tissue, and future research in this area will undoubtedly expand our understanding of the erectile mechanisms and provide new avenues for ED treatment. Conclusions

From the above review, it is clear that physiological (sexually induced) erection is mediated through the NO/sGC/cGMP/PKG pathway and detumescence is carried out principally by the degradation of cGMP by PDE5. On the other hand, the cAMP and the pGC pathways play auxiliary roles in erection, and so does PDE3 in bringing about detumescence. However, regardless of their relative importance in physiological erection, all of these pathways are susceptible to genetic and biochemical alterations, leading to, or resulting from ED. Furthermore, several key elements along these pathways have been exploited as therapeutic targets with varying degrees of success. While inhibitors of PDE5 are by far the most successful, they nevertheless are ineffective in roughly 30% of ED patients, exhibit several undesirable side effects, and offer only short-term cure. As such, continuing research in cyclic nucleotide signaling and other areas such as Rho-kinase signaling are needed to pave the way for better ED treatment. Acknowledgments

This work was supported in part by grants from the California Urology Foundation, Mr. Arthur Rock and the Rock Foundation, and the National Institutes of Health (2R01-DK-45370). Corresponding Author: Ching-Shwun Lin, PhD, Knuppe Molecular Urology Laboratory, Department of

Urology, University of California, San Francisco, CA 94143-1695, USA. Tel: 415-353-7205; Fax: 415-3539586; E-mail: [email protected] Conflict of Interest: None. References

1 Lin C-S, Xin ZC, Lin G, Lue TF. Phosphodiesterases as therapeutic targets. Urology 2003;61:685–91. 2 Patel TB, Du Z, Pierre S, Cartin L, Scholich K. Molecular biological approaches to unravel adenylyl cyclase signaling and function. Gene 2001;269:13–25. 3 Guimaraes S, Moura D. Vascular adrenoceptors: An update. Pharmacol Rev 2001;53:319–56. 4 Tabrizchi R, Bedi S. Pharmacology of adenosine receptors in the vasculature. Pharmacol Ther 2001;91:133–47. 5 Takahashi Y, Ishii N, Lue TF, Tanagho EA. Effects of adenosine on canine penile erection. J Urol 1992;148:1323–5. 6 Filippi S, Mancini M, Amerini S, Bartolini M, Natali A, Mancina R, Forti G, Ledda F, Maggi M. Functional adenosine receptors in human corpora cavernosa. Int J Androl 2000;23:210–7. 7 Noto T, Inoue H, Mochida H, Kikkawa K. Role of adenosine and P2 receptors in the penile tumescence in anesthetized dogs. Eur J Pharmacol 2001;425:51–5. 8 Conner AC, Hay DL, Howitt SG, Kilk K, Langel U, Wheatley M, Smith DM, Poyner DR. Interaction of calcitonin-gene-related peptide with its receptors. Biochem Soc Trans 2002;30:451–5. 9 Wimalawansa SJ. Age-related changes in tissue contents of immunoreactive calcitonin generelated peptide. Aging (Milano) 1992;4:211–7. 10 Stief CG, Wetterauer U, Schaebsdau FH, Jonas U. Calcitonin-gene-related peptide: A possible role in human penile erection and its therapeutic application in impotent patients. J Urol 1991;146: 1010–4. 11 Bivalacqua TJ, Champion HC, Abdel-Mageed AB, Kadowitz PJ, Hellstrom WJ. Gene transfer of prepro-calcitonin gene-related peptide restores erectile function in the aged rat. Biol Reprod 2001;65:1371–7. 12 Nishimatsu H, Hirata Y, Hayakawa H, Nagata D, Satonaka H, Suzuki E, Horie S, Takeuchi T, Ohta N, Homma Y, Minowada S, Nagai R, Kawabe K, Kitamura T. Effects of intracavernous administration of adrenomedullin on erectile function in rats. Peptides 2001;22:1913–8. 13 Narumiya S, FitzGerald GA. Genetic and pharmacological analysis of prostanoid receptor function. J Clin Invest 2001;108:25–30. 14 Daley JT, Watkins MT, Brown ML, Martinez V, Cuevas P. Saenz de Tejada I. Prostanoid producJ Sex Med 2005; 2: 478–491

488

15

16

17

18

19

20 21 22

23

24

25

26

Lin et al. tion in rabbit corpus cavernosum. II. Inhibition by oxidative stress. J Urol 1996;156:1169–73. Moreland RB, Albadawi H, Bratton C, Patton G, Goldstein I, Traish A, Watkins MT. O2-dependent prostanoid synthesis activates functional PGE receptors on corpus cavernosum smooth muscle. Am J Physiol Heart Circ Physiol 2001;281:H552– 8. Hedlund H, Andersson KE. Comparison of the responses to drugs acting on adrenoreceptors and muscarinic receptors in human isolated corpus cavernosum and cavernous artery. J Auton Pharmacol 1985;5:81–8. Bosch RJ, Benard F, Aboseif SR. Changes in penile hemodynamics after intra cavernous injection of PGE1 and prostaglandin I2 in pig-tailed monkeys. Int J Impot Res 1989;1:211–21. Angulo J, Cuevas P, La Fuente JM, Pomerol JM, Ruiz-Castane E, Puigvert A, Gabancho S, Fernandez A, Ney P, Saenz De Tejada I. Regulation of human penile smooth muscle tone by prostanoid receptors. Br J Pharmacol 2002;136: 23–30. Moreland RB, Kim N, Nehra A, Goldstein I, Traish A. Functional prostaglandin E (EP) receptors in human penile corpus cavernosum. Int J Impot Res 2003;15:362–8. Porst H. The rationale for prostaglandin E1 in erectile failure: A survey of worldwide experience. J Urol 1996;155:802–15. Andersson KE. Neurophysiology/pharmacology of erection. Int J Impot Res 2001;13(3 suppl):S8– 17. Guidone G, Muller D, Vogt K, Mukhopadhyay AK. Characterization of VIP and PACAP receptors in cultured rat penis corpus cavernosum smooth muscle cells and their interaction with guanylate cyclase-B receptors. Regul Pept 2002;108:63–72. Juenemann KP, Lue TF, Luo JA, Jadallah SA, Nunes LL, Tanagho EA. The role of vasoactive intestinal polypeptide as a neurotransmitter in canine penile erection: A combined in vivo and immunohistochemical study. J Urol 1987;138: 871–7. Kiely EA, Bloom SR, Williams G. Penile response to intracavernosal vasoactive intestinal polypeptide alone and in combination with other vasoactive agents. Br J Urol 1989;64:191–4. Pickard RS, Powell PH, Zar MA. Evidence against vasoactive intestinal polypeptide as the relaxant neurotransmitter in human cavernosal smooth muscle. Br J Pharmacol 1993;108:497–500. Palmer LS, Valcic M, Melman A, Giraldi A, Wagner G, Christ GJ. Characterization of cyclic AMP accumulation in cultured human corpus cavernosum smooth muscle cells. J Urol 1994;152: 1308–14.

J Sex Med 2005; 2: 478–491

27 Shen ZJ, Lu YL, Chen ZD, Chen F, Chen Z. Effects of androgen and ageing on gene expression of vasoactive intestinal polypeptide in rat corpus cavernosum. BJU Int 2000;86:133–7. 28 Sullivan M, Thompson CS, Mikhailidis DP, Morgan RJ, Angelini GD, Jeremy JY. Differential alterations of prostacyclin, cyclic AMP and cyclic GMP formation in the corpus cavernosum of the diabetic rabbit. Br J Urol 1998;82:578–84. 29 Johnson DA, Akamine P, Radzio-Andzelm E, Madhusudan M, Taylor SS. Dynamics of cAMPdependent protein kinase. Chem Rev 2001; 101:2243–70. 30 Shabb JB. Physiological substrates of cAMPdependent protein kinase. Chem Rev 2001; 101:2381–411. 31 Liu H, Maurice DH. Phosphorylation-mediated activation and translocation of the cyclic AMPspecific phosphodiesterase PDE4D3 by cyclic AMP-dependent protein kinase and mitogenactivated protein kinases. A potential mechanism allowing for the coordinated regulation of PDE4D activity and targeting. J Biol Chem 1999; 274:10557–65. 32 Conti M. Phosphodiesterases and cyclic nucleotide signaling in endocrine cells. Mol Endocrinol 2000;14:1317–27. 33 Funakoshi Y, Ichiki T, Takeda K, Tokuno T, Iino N, Takeshita A. Critical role of cAMP-response element-binding protein for angiotensin IIinduced hypertrophy of vascular smooth muscle cells. J Biol Chem 2002;277:18710–7. 34 Quinn KV, Giblin JP, Tinker A. Multisite phosphorylation mechanism for protein kinase A activation of the smooth muscle ATP-sensitive K+ channel. Circ Res 2004;94:1359–66. 35 Insuk SO, Chae MR, Choi JW, Yang DK, Sim JH, Lee SW. Molecular basis and characteristics of KATP channel in human corporal smooth muscle cells. Int J Impot Res 2003;15:258–66. 36 Matsuo H. Discovery of a natriuretic peptide family and their clinical application. Can J Physiol Pharmacol 2001;79:736–40. 37 Potter LR, Hunter T. Guanylyl cyclase-linked natriuretic peptide receptors: Structure and regulation. J Biol Chem 2001;276:6057–60. 38 Chauhan SD, Nilsson H, Ahluwalia A, Hobbs AJ. Release of C-type natriuretic peptide accounts for the biological activity of endothelium-derived hyperpolarizing factor. Proc Natl Acad Sci USA 2003;100:1426–31. 39 Kim SZ, Kim SH, Park JK, Koh GY, Cho KW. Presence and biological activity of C-type natriuretic peptide-dependent guanylate cyclasecoupled receptor in the penile corpus cavernosum. J Urol 1998;159:1741–6. 40 Kuthe A, Reinecke M, Uckert S, Becker A, David I, Heitland A, Stief CG, Forssmann WG, Magert

Cyclic Nucleotide Signaling

41

42

43 44

45

46

47

48

49

50

51

52

HJ. Expression of guanylyl cyclase B in the human corpus cavernosum penis and the possible involvement of its ligand C-type natriuretic polypeptide in the induction of penile erection. J Urol 2003;169:1918–22. Hurt KJ, Musicki B, Palese MA, Crone JK, Becker RE, Moriarity JL, Snyder SH, Burnett AL. Aktdependent phosphorylation of endothelial nitricoxide synthase mediates penile erection. Proc Natl Acad Sci USA 2002;99:4061–6. Carrier GO, Fuchs LC, Winecoff AP, Giulumian AD, White RE. Nitrovasodilators relax mesenteric microvessels by cGMP-induced stimulation of Caactivated K channels. Am J Physiol 1997;273:H76– 84. Cartledge J, Minhas S, Eardley I. The role of nitric oxide in penile erection. Expert Opin Pharmacother 2001;2:95–107. Haas CA, Seftel AD, Razmjouei K, Ganz MB, Hampel N, Ferguson K. Erectile dysfunction in aging: Upregulation of endothelial nitric oxide synthase. Urology 1998;51:516–22. Bakircioglu ME, Sievert KD, Nunes L, Lau A, Lin C-S, Lue TF. Decreased trabecular smooth muscle and caveolin-1 expression in the penile tissue of aged rats. J Urol 2001;166:734–8. Rajasekaran M, Kasyan A, Jain A, Kim SW, Monga M. Altered growth factor expression in the aging penis: The Brown-Norway rat model. J Androl 2002;23:393–9. Champion HC, Bivalacqua TJ, Hyman AL, Ignarro LJ, Hellstrom WJ, Kadowitz PJ. Gene transfer of endothelial nitric oxide synthase to the penis augments erectile responses in the aged rat. Proc Natl Acad Sci USA 1999;96: 11648–52. Magee TR, Ferrini M, Garban HJ, Vernet D, Mitani K, Rajfer J, Gonzalez-Cadavid NF. Gene therapy of erectile dysfunction in the rat with penile neuronal nitric oxide synthase. Biol Reprod 2002;67:20–8. Chancellor MB, Tirney S, Mattes CE, Tzeng E, Birder LA, Kanai AJ, De Groat WC, Huard J, Yoshimura N. Nitric oxide synthase gene transfer for erectile dysfunction in a rat model. BJU Int 2003;91:691–6. Burnett AL, Nelson RJ, Calvin DC, Liu JX, Demas GE, Klein SL, Kriegsfeld LJ, Dawson VL, Dawson TM, Snyder SH. Nitric oxide-dependent penile erection in mice lacking neuronal nitric oxide synthase. Mol Med 1996;2:288–96. Burnett AL, Chang AG, Crone JK, Huang PL, Sezen SE. Noncholinergic penile erection in mice lacking the gene for endothelial nitric oxide synthase. J Androl 2002;23:92–7. Andreopoulos S, Papapetropoulos A. Molecular aspects of soluble guanylyl cyclase regulation. General Pharmacol 2000;34:147–57.

489 53 Behrends S, Steenpass A, Porst H, Scholz H. Expression of nitric oxide-sensitive guanylyl cyclase subunits in human corpus cavernosum. Biochem Pharmacol 2000;59:713–7. 54 Mizusawa H, Hedlund P, Brioni JD, Sullivan JP, Andersson KE. Nitric oxide independent activation of guanylate cyclase by YC-1 causes erectile responses in the rat. J Urol 2002; 167:2276–81. 55 Klotz T, Bloch W, Zimmermann J, Ruth P, Engelmann U, Addicks K. Soluble guanylate cyclase and cGMP-dependent protein kinase I expression in the human corpus cavernosum. Int J Impot Res 2000;12:157–64. 56 Boerth NJ, Lincoln TM. Expression of the catalytic domain of cyclic GMP-dependent protein kinase in a baculovirus system. FEBS Lett 1994;342:255–60. 57 Collins SP, Uhler MD. Cyclic AMP- and cyclic GMP-dependent protein kinases differ in their regulation of cyclic AMP response elementdependent gene transcription. J Biol Chem 1999;274:8391–404. 58 Feil R, Gappa N, Rutz M, Schlossmann J, Rose CR, Konnerth A, Brummer S, Kuhbandner S, Hofmann F. Functional reconstitution of vascular smooth muscle cells with cGMP-dependent protein kinase I isoforms. Circ Res 2002;90:1080– 6. 59 Hedlund P, Aszodi A, Pfeifer A, Alm P, Hofmann F, Ahmad M, Fassler R, Andersson KE. Erectile dysfunction in cyclic GMP-dependent kinase Ideficient mice. Proc Natl Acad Sci USA 2000;97:2349–54. 60 Tegge W, Frank R, Hofmann F, Dostmann WR. Determination of cyclic nucleotide-dependent protein kinase substrate specificity by the use of peptide libraries on cellulose paper. Biochemistry 1995;34:10569–77. 61 Murthy KS, Zhou H. Selective phosphorylation of the IP3R-I in vivo by cGMP-dependent protein kinase in smooth muscle. Am J Physiol Gastrointest Liver Physiol 2003;284:G221–30. 62 Krall JF, Fittingoff M, Rajfer J. Characterization of cyclic nucleotide and inositol 1,4,5-trisphosphate-sensitive calcium-exchange activity of smooth muscle cells cultured from the human corpora cavernosa. Biol Reprod 1988;39:913–22. 63 Schlossmann J, Ammendola A, Ashman K, Zong X, Huber A, Neubauer G, Wang GX, Allescher HD, Korth M, Wilm M, Hofmann F, Ruth P. Regulation of intracellular calcium by a signalling complex of IRAG, IP3 receptor and cGMP kinase Ibeta. Nature 2000;404:197–201. 64 Lalli MJ, Shimizu S, Sutliff RL, Kranias EG, Paul RJ. [Ca2+]i homeostasis and cyclic nucleotide relaxation in aorta of phospholamban-deficient mice. Am J Physiol 1999;277:H963–70.

J Sex Med 2005; 2: 478–491

490 65 Schubert R, Nelson MT. Protein kinases: Tuners of the BKCa channel in smooth muscle. Trends Pharmacol Sci 2001;22:505–12. 66 Lee SW, Kang TM. Effects of nitric oxide on the Ca2+-activated potassium channels in smooth muscle cells of the human corpus cavernosum. Urol Res 2001;29:359–65. 67 Malysz J, Gibbons SJ, Miller SM, Gettman M, Nehra A, Szurszewski JH, Farrugia G. Potassium outward currents in freshly dissociated rabbit corpus cavernosum myocytes. J Urol 2001;166:1167– 77. 68 Archer SL. Potassium channels and erectile dysfunction. Vascul Pharmacol 2002;38:61–71. 69 Christ GJ, Day N, Santizo C, Sato Y, Zhao W, Sclafani T, Bakal R, Salman M, Davies K, Melman A. Intracorporal injection of hSlo cDNA restores erectile capacity in STZ-diabetic F-344 rats in vivo. Am J Physiol Heart Circ Physiol 2004; 287:H1544–53. 70 Mullershausen F, Russwurm M, Thompson WJ, Liu L, Koesling D, Friebe A. Rapid nitric oxideinduced desensitization of the cGMP response is caused by increased activity of phosphodiesterase type 5 paralleled by phosphorylation of the enzyme. J Cell Biol 2001;155:271–8. 71 Woodrum D, Pipkin W, Tessier D, Komalavilas P, Brophy CM. Phosphorylation of the heat shock-related protein, HSP20, mediates cyclic nucleotide-dependent relaxation. J Vasc Surg 2003;37:874–81. 72 Surks HK, Mochizuki N, Kasai Y, Georgescu SP, Tang KM, Ito M, Lincoln TM, Mendelsohn ME. Regulation of myosin phosphatase by a specific interaction with cGMP-dependent protein kinase Ialpha. Science 1999;286:1583–7. 73 Huang QQ, Fisher SA, Brozovich FV. Unzipping the role of myosin light chain phosphatase in smooth muscle cell relaxation. J Biol Chem 2004;279:597–603. 74 Tokui T, Brozovich F, Ando S, Ikebe M. Enhancement of smooth muscle contraction with protein phosphatase inhibitor 1: Activation of inhibitor 1 by cGMP-dependent protein kinase. Biochem Biophys Res Commun 1996;220:777– 83. 75 Carr AN, Sutliff RL, Weber CS, Allen PB, Greengard P, de Lanerolle P, Kranias EG, Paul RJ. Is myosin phosphatase regulated in vivo by inhibitor1? Evidence from inhibitor-1 knockout mice. J Physiol 2001;534:357–66. 76 Sauzeau V, Le Jeune H, Cario-Toumaniantz C, Smolenski A, Lohmann SM, Bertoglio J, Chardin P, Pacaud P, Loirand G. Cyclic GMP-dependent protein kinase signaling pathway inhibits RhoAinduced Ca2+ sensitization of contraction in vascular smooth muscle. J Biol Chem 2000;275: 21722–9.

J Sex Med 2005; 2: 478–491

Lin et al. 77 Jiang H, Colbran JL, Francis SH, Corbin JD. Direct evidence for cross-activation of cGMPdependent protein kinase by cAMP in pig coronary arteries. J Biol Chem 1992;267:1015–9. 78 Eckly-Michel A, Martin V, Lugnier C. Involvement of cyclic nucleotide-dependent protein kinases in cyclic AMP-mediated vasorelaxation. Br J Pharmacol 1997;122:158–64. 79 Dhanakoti SN, Gao Y, Nguyen MQ, Raj JU. Involvement of cGMP-dependent protein kinase in the relaxation of ovine pulmonary arteries to cGMP and cAMP. J Appl Physiol 2000;88:1637– 42. 80 Wang Q, Bryowsky J, Minshall RD, Pelligrino DA. Possible obligatory functions of cyclic nucleotides in hypercapnia-induced cerebral vasodilation in adult rats. Am J Physiol 1999;276:H480–7. 81 White RE, Kryman JP, El-Mowafy AM, Han G, Carrier GO. cAMP-dependent vasodilators crossactivate the cGMP-dependent protein kinase to stimulate BK (Ca) channel activity in coronary artery smooth muscle cells. Circ Res 2000;86:897– 905. 82 Barman SA, Zhu S, Han G, White RE. cAMP activates BKCa channels in pulmonary arterial smooth muscle via cGMP-dependent protein kinase. Am J Physiol Lung Cell Mol Physiol 2003;284:L1004–11. 83 Lin C-S, Liu X, Tu R, Chow S, Lue TF. Agerelated decrease of protein kinase G activation in vascular smooth muscle cells. Biochem Biophys Res Commun 2001;287:244–8. 84 Lin C-S, Liu X, Chow S, Lue TF. Cyclic AMP and cyclic GMP activate protein kinase G in cavernosal smooth muscle cells: Old age is a negative factor. BJU Int 2002;89:576–82. 85 Lincoln TM, Cornwell TL, Taylor AE. cGMPdependent protein kinase mediates the reduction of Ca2+ by cAMP in vascular smooth muscle cells. Am J Physiol 1990;258:C399–407. 86 Lincoln TM, Komalavilas P, Boerth NJ, MacMillan-Crow LA, Cornwell TL. cGMP signaling through cAMP- and cGMP-dependent protein kinases. Adv Pharmacol 1995;34:305–22. 87 Pfeifer A, Klatt P, Massberg S, Ny L, Sausbier M, Hirneiss C, Wang GX, Korth M, Aszódi A, Andersson KE, Krombach F, Mayerhofer A, Ruth P, Fässler R, Hofmann F. Defective smooth muscle regulation in cGMP kinase I-deficient mice. EMBO J 1998;17:3045–51. 88 Sausbier M, Schubert R, Voigt V, Hirneiss C, Pfeifer A, Korth M, Kleppisch T, Ruth P, Hofmann F. Mechanisms of NO/cGMP-dependent vasorelaxation. Circ Res 2000;87:825–30. 89 Francis SH, Turko IV, Corbin JD. Cyclic nucleotide phosphodiesterases: Relating structure and function. Prog Nucleic Acid Res Mol Biol 2001;65:1–52.

491

Cyclic Nucleotide Signaling 90 Corbin JD, Turko IV, Beasley A, Francis SH. Phosphorylation of phosphodiesterase-5 by cyclic nucleotide-dependent protein kinase alters its catalytic and allosteric cGMP-binding activities. Eur J Biochem 2000;267:2760–7. 91 Lin C-S, Chow S, Lau A, Tu R, Lue TF. Identification and regulation of human PDE5A gene promoter. Biochem Biophys Res Commun 2001;280:684–92. 92 Lin C-S, Chow S, Lau A, Tu R, Lue TF. Regulation of human PDE5A2 intronic promoter by cAMP and cGMP: Identification of a critical Sp1binding site. Biochem Biophys Res Commun 2001;280:693–9. 93 Montorsi F, Corbin JD, Philips S. Review of phosphodiesterases in the urogenital system: New directions for therapeutic intervention. J Sex Med 2004;1:322–36. 94 Küthe A, Wiedenroth A, Mägert HJ, Uckert S, Forssmann WG, Stief CG, Jonas U. Expression of different phosphodiesterase genes in human cavernous smooth muscle. J Urol 2001;165:280–3. 95 Eardley I, Cartledge J. Tadalafil (Cialis) for men with erectile dysfunction. Int J Clin Pract 2002;56:300–4. 96 Kuthe A, Montorsi F, Andersson KE, Stief CG. Phosphodiesterase inhibitors for the treatment of erectile dysfunction. Curr Opin Investig Drugs 2002;3:1489–95. 97 Stief CG, Uckert S, Becker AJ, Harringer W, Truss MC, Forssmann WG, Jonas U. Effects of sildenafil on cAMP and cGMP levels in isolated human cavernous and cardiac tissue. Urology 2000;55:146–50. 98 Uckert S, Hedlund P, Waldkirch E, Sohn M, Jonas U, Andersson KE, Stief CG. Interactions between cGMP- and cAMP-pathways are involved in the regulation of penile smooth muscle tone. World J Urol 2004;22:261–6. 99 Kim NN, Huang Y, Moreland RB, Kwak SS, Goldstein I, Traish A. Cross-regulation of intra-

100 101

102

103

104

105

106 107

108

cellular cGMP and cAMP in cultured human corpus cavernosum smooth muscle cells. Mol Cell Biol Res Commun 2000;4:10–4. Zhong H, Minneman KP. Alpha1-adrenoceptor subtypes. Eur J Pharmacol 1999;375:261–76. Hayabuchi Y, Dart C, Standen NB. Evidence for involvement of A-kinase anchoring protein in activation of rat arterial K (ATP) channels by protein kinase A. J Physiol 2001;536:421–7. Vo NK, Gettemy JM, Coghlan VM. Identification of cGMP-dependent protein kinase anchoring proteins (GKAPs). Biochem Biophys Res Commun 1998;246:831–5. Yuasa K, Omori K, Yanaka N. Binding and phosphorylation of a novel male germ cell-specific cGMP-dependent protein kinase-anchoring protein by cGMP-dependent protein kinase Ialpha. J Biol Chem 2000;275:4897–905. Matsuzaki T, Tajika Y, Tserentsoodol N, Suzuki T, Aoki T, Hagiwara H, Takata K. Aquaporins: A water channel family. Anat Sci Int 2002;77:85– 93. Chin KV, Yang WL, Ravatn R, Kita T, Reitman E, Vettori D, Cvijic ME, Shin M, Iacono L. Reinventing the wheel of cyclic AMP: Novel mechanisms of cAMP signaling. Ann NY Acad Sci 2002;968:49–64. Mills TM, Chitaley K, Lewis RW. Vasoconstrictors in erectile physiology. Int J Impot Res 2001;13(5 suppl ):S29–34. Qiao J, Huang F, Lum H. PKA inhibits RhoA activation: A protection mechanism against endothelial barrier dysfunction. Am J Physiol Lung Cell Mol Physiol 2003;284:L972–80. Bolz SS, Vogel L, Sollinger D, Derwand R, de Wit C, Loirand G, Pohl U. Nitric oxide-induced decrease in calcium sensitivity of resistance arteries is attributable to activation of the myosin light chain phosphatase and antagonized by the RhoA/Rho-kinase pathway. Circulation 2003;107: 3081–7.

J Sex Med 2005; 2: 478–491