Phosphodiesterase Inhibitors as Potential Cognition ... - IngentaConnect

222

Current Topics in Medicinal Chemistry, 2010, 10, 222-230

Phosphodiesterase Inhibitors as Potential Cognition Enhancing Agents Christopher J. Schmidt* Neuroscience Research Unit, Pfizer, Inc., Groton, USA Abstract: As might be predicted for an organ designed for cell to cell communication, cyclic nucleotide signaling in the brain is highly organized and regulated. Augmentation of cyclic nucleotide signaling by means of phosphodiesterase inhibition appears to be a viable and tractable means of enhancing neuronal communication. Of the various CNS disorders that have been considered as target indications for phosphodiesterase inhibitors, no condition has received more attention than cognitive dysfunction. This review provides a background for understanding the expanding literature in this field as well as a brief update on the rationale driving the search for selective inhibitors of targets such as PDE1B, PDE2, PDE5 and PDE9.

INTRODUCTION The commercial success of PDE5 inhibitors and the growing appreciation of cyclic nucleotide phosphodiesterases (PDEs) as drug targets has lead to a surge in the research in this area and a corresponding increase in both the primary and review literature on PDEs [1-4]. Among diseases of the central nervous system, no PDE indication has received more attention than cognition. There have been a number of excellent reviews published recently on the potential to improve cognitive function through the use of PDE inhibitors [5-8] as well as reviews on specific PDEs as will be touched upon during the discussion of each enzyme. Mindful of these resources, this update will focus on the most recent publications while briefly describing some of the background material needed to navigate through this expanding literature. One goal of this review will be to highlight the importance of chemical tools in advancing our understanding of the PDEs in the brain. At least 5 of the 11 PDE families have received serious consideration as targets for cognitive dysfunction including PDE1B, 2, 4, 5 and 9. The expression pattern of PDE8 has been evaluated in Alzheimer’s disease; however, the absence of pharmacological tools has limited research on this enzyme. PDE10A inhibitors have also been shown to have some activity in specific cognition assays related to corticostriatal function [9,10] but have been most thoroughly examined for the treatment of psychosis [11]. The cyclic nucleotides function as ubiquitous intracellular signaling molecules in all mammalian cells. The ability of extracellular signals to modulate the intracellular concentration of cyclic nucleotides allows cells to respond to external stimuli across the boundary of the cell membrane. The cyclic nucleotides signaling cascades have been adapted to respond to a host of transduction systems ranging from transmembrane proteins such as the G-protein linked receptors (GPRs) to voltage and ligand gated ion channels and even diffusible gases. Changes in the intracellular *Address correspondence to this author at the MS8220-4142, Pfizer, Inc., Eastern Point Road, Groton, CT 06371; Tel: 860-441-8797; E-mail: [email protected]

1568-0266/10 $55.00+.00

concentration of just two molecules, adenosine 3’, 5’-cyclic monophosphate (cAMP) and guanosine 3’,5’-cyclic monophosphate (cGMP), serve as the second messengers for a bewildering array of intracellular signaling pathways. cAMP is synthesized from ATP by one of eight different adenylyl cyclase (AC) isozymes. AC exists in both cytoplasmic and membrane bound forms and each subtype can be selectively activated by small GTPases such as the receptor G-proteins or directly by Ca2+/CaM [12]. Guanylyl cyclase, the enzyme responsible for the conversion of GTP to cGMP, also exists in both soluble and particulate forms. Soluble quanylyl cyclase (sGC) is activated by NO derived from neuronal, endothelial or inducible nitric oxide synthase (nNOS, eNOS or iNOS, respectively), while the particular enzymes (pGC) are transmembrane receptors activated by naturetic peptides [13]. The cyclic nucleotides transmit their signal in the cell through a variant of tertiary elements. (see fig. 1). The best described of these are cAMP dependent protein kinase (PKA) and cGMP dependent protein kinase (PKG). The binding of the cyclic nucleotide to each enzyme enables their phosphorylation of downstream enzymes and proteins functioning as effectors or additional elements in the signaling cascade. In recent years, non-kinase targets of the cyclic nucleotides have received increased attention including cyclic nucleotide gated ion channels [14], PDEs and the small GTPase, EPAC (exchange protein activated by cAMP) [15]. The ability of cAMP and cGMP to function as the intracellular messengers for multiple signaling cascades in the same cell is largely due to the temporal and spatial restriction of these cascades. This “compartmentalization” is accomplished in many cases by embedding the transducer, the cyclase and its effectors in large macromolecular complexes. One of the best studied scaffolds for the building of such complexes are the A-kinase anchoring proteins or AKAPs [16]. The spatial and temporal shape of the cyclic nucleotide response is also regulated by the cyclic nucleotide PDEs [17]. These enzymes terminate the activity of the second messengers by hydrolysis of the phosphodiesterase bond in cAMP and cGMP to generate 5’-AMP or 5’-GMP, respectively. The importance of PDEs in controlling cyclic © 2010 Bentham Science Publishers Ltd.

Phosphodiesterase Inhibitors as Potential Cognition Enhancing Agents

Current Topics in Medicinal Chemistry, 2010, Vol. 10, No. 2

223

Fig. (1). Cyclic nucleotide signaling cascades.

nucleotide signaling is illustrated by the fact that most tissues contain 10-fold greater PDE than cyclase activity. Although most clearly demonstrated for members of the PDE4 family, it is assumed that most PDEs are also spatially restricted, often part of the same macromolecular complex as the cyclic nucleotide effector. The affinity and rate of cAMP or cGMP metabolism, also subject to regulation, allows PDEs to influence the magnitude and duration of the second messenger signal. The mammalian PDEs are a large group of closely related enzymes divided into 11 families based on substrate specificity, inhibitor sensitivity and more recently, on sequence homology. The 11 families are coded by 21 genes providing several of the families with multiple members. Each family is designated by an Arabic number with the isozymes within that family identified by a letter (e.g. PDE4B). Further diversity is afforded by the existence of a number of splice variants for each gene product, each designated by an additional number, e.g. PDE4B1. The species designation often precedes the name of the phosphodiesterase as in MM or RN for Mus musculus (mouse) or Rattus norvegicus (rat), respectively. All mammalian PDEs share a common domain structure composed of N-terminal regulatory units with a C-terminal catalytic domain. All are likely to function as dimers and occlusion of their active site may be important in their regulation [18]. Despite the similarity in the reaction catalyzed by

these enzymes, the homology of catalytic domain between families is only about 25-50 percent [19] but is close to 70 percent within families. The within family similarity may explain the relative ease with which it has been possible to develop active site inhibitors selective for entire PDE families as opposed to subtype selective agents. The Nterminal regulatory domains of the enzymes are the regions of greatest diversity being completely absent in some short forms or capable of complex regulatory functions in others. The regulatory domains of certain PDEs provides a means for crosstalk between signaling cascades (e.g. Ca2+-calmodulin for PDE1), crosstalk between cyclic nucleotides (e.g. PDE2), regulation by phosphorylation (e.g. PDE4) and intracellular targeting. PDE Inhibition for Improving Cognitive Function Cognitive deficits are a feature of many CNS disorders across the spectrum of neurodegenerative and psychiatric diseases. Cognition may be defined as the mental processes involved in the utilization of perception, reasoning and memory with each of these domains being subject to disruption to varying extents. Patients with Alzheimer’s disease show characteristically severe and progressive deficits in both immediate and long-term memory [20]. In contrast, the disruption of memory function in schizophrenia is less severe but additional deficits occur in attention, reasoning, and speed of processing [21]. Modest deficits in reasoning

224 Current Topics in Medicinal Chemistry, 2010, Vol. 10, No. 2

and working memory are also reported in depressed patients [22]. The function of PDEs as a regulator of second messenger signaling has made these enzymes a target of interest for many diseases in which abnormal signaling may play a role or in which enhanced signaling may somehow mitigate the consequences of the pathophysiology. Nowhere is this more true that in disease of the central nervous system (CNS) and more specifically in the area of cognitive function. The tight reciprocal relationship between activity and structure in the CNS allows for abnormal function to affect structure and visa versa [23]. Thus the significant structural changes seen in diseases such as AD may have their genesis in synaptic dysfunction [8]. A significant portion of the burden of any CNS disease may therefore be traced to it effect on neurotransmission or synaptic dysfunction. Almost invariably, our current therapies for these disorders seek to normalize or compensate for this deficit. PDE inhibition can provide a mechanism for the augmentation of residual cell to cell signaling. In contrast to the effect of receptor agonists and antagonists, signal augmentation due to PDE inhibition can occur without markedly disrupting the fidelity of that signaling. Both in vitro and in vivo studies have linked alterations in cyclic nucleotide concentrations with biochemical or physiological process linked to cognitive function. cAMP and cGMP are known to be involved in processes which change the signaling characteristics of a synapse based upon its previous experience, processes collectively referred to as synaptic plasticity [24]. Signal intensity as well as the level of coincident activity at a synapse are established variables which can result in either potentiation or depotentiation of transmission at a particular synapse. Long term potentiation (LTP) is the best described of these processes and is known to be modulated by both the cAMP and cGMP signaling cascades. The reciprocal process of long term depression (LTD) is also subject to such regulation. Targeting Cognition Preclinically With a few notable exceptions, the majority of commonly used preclinical cognition models assess mnemonic function. Animal models utilized for assessing memory include passive avoidance, delayed alternation, novel object recognition, social recognition and maze tasks such as the Morris water maze and radial arm maze [25]. Novel object recognition may be considered the “workhorse” of preclinical cognition models currently [26]. This task takes advantage of the inherent tendency of animals to spend more time exploring a novel as opposed to a familiar object. No training is required and the animal is not food deprived or forced into a stressful situation which would increase the potential for nonspecific responses. Forgetting, spending equivalent time on the familiar versus the novel object, can be induced by extending the time between exposures to the target objects or by the administration of an amnesic agent such as scopolamine or MK-801. Compounds improving memory are expected to prevent the effect of time or the amnesic agent as shown by an increase in the percentage of time spent exploring the novel versus the familiar object. The social recognition assay replaces the objects with juvenile rats to take advantage of the greater

Christopher J. Schmidt

tendency of animals to explore non-threatening conspecifics. Delayed alternation also relies on the natural preference of foraging animals for exploring novel spaces. When given two routes, rats and mice will choose the alternative pathway to one recently traveled unless memory is disrupted by a long delay or some pharmacological manipulation. The 4 or 8 arm radial maze takes advantage of this same tendency although these more demanding memory tasks generally require training for animals to reach an acceptable degree of performance, i.e. visiting a majority of the arms before reentering any arm. In contrast, passive avoidance relies on memory to inhibit the natural tendency of an animal to seek a dark compartment when that compartment was previous associated with a small shock. Manipulations disrupting memory minimize the effect of the shock on the latency to return to the compartment whereas improvements in memory are associated with increases in latency. The Morris water maze is a task assessing spatial learning and memory and is therefore considered to reflect hippocampal function. Animals are trained to use spatial cues to swim to a platform hidden beneath the surface of an opaque pool. Results are commonly reported from probe trials which assess the percentage of time a swimming animal spends in the correct quadrant after removal of the platform. Specific PDEs as Targets for Cognition PDE1B The PDE1 enzymes were originally named based on their order of elution during the chromatographic separation of PDE activities. The family is composed of three dual substrate enzymes termed A, B and C and all are characterized by their ability to be activated by Ca2+/calmodulin (Ca2+/CaM). The PDE1 family thus provides a mechanism for cross-talk between cyclic nucleotide pathways and Ca2+ signaling. The isozymes differ in their regulatory properties, substrate affinities, sensitivity to Ca2+/CaM and in their tissue distribution [1]. Although both PDE1A and PDE1B are found in the brain, the latter is the subtype of greatest interest for cognition. The high levels of PDE1B in the brain and its activation by Ca2+/calmodulin were the characteristics which made it one of the first PDEs identified. PDE1B has a 10-fold higher affinity for cGMP over cAMP [2] although it is not clear which cyclic nucleotide is the endogenous substrate of PDE1B. Indeed, this is the case with all dual substrate PDEs, with the exception of PDE10A for which there is evidence that the enzyme metabolizes both cAMP and cGMP in vivo [11]. There are two Ca2+/CaM binding sites on each PDE1 monomer although occupancy of a single site is sufficient to reorientation the auto-inhibitory domain which lies between the two Ca2+/CaM sites thereby activating the enzyme. The binding of Ca2+/CaM to the enzyme results in an increase in the maximum velocity of hydrolysis (Vmax) without affecting the Km for either cyclic nucleotide. In addition to differences in the affinity of each isozyme for Ca2+/CaM, phosphorylation of the binding site further modifies the affinity of each isozyme for Ca2+/CaM . The affinity of


PDE1A and C are reduced by phosphorylation by PKA whereas the phosphorylation of PDE1B by CaM kinase II reduces its affinity for Ca2+/CaM by a factor of six. Dephosphorylated is mediated by the Ca2+/CaM dependent phosphatase, calcineurin. Thus PDE1B in particular is regulated in a complex fashion by Ca2+ signaling. Activation of the kinase, phosphatase and phosphodiesterase is dictated by the affinity of each for Ca2+/CaM. Preferential activation of CaM kinase II at low levels of Ca2+ would maintain PDE1B activity at a low level and allow cyclic nucleotide dependent signaling. However, elevations in Ca2+ could initiate a rapid increase in PDE activity subsequent to the dephosphorylation of PDE1B and binding of Ca2+/CaM to the enzyme. Although somewhat theoretical, such a scenario illustrates how PDE1B may be regulated to temporally shape the cyclic nucleotide signal in response to changes in intracellular Ca2+. Although PDE1B is found at low levels in various peripheral tissues, its highest levels are within the brain particularly in the striatum and the dentate gyrus of the hippocampus [27]. Subcellular localization studies indicate that PDE1B is found in the soluble fraction within the brain [28]. Unfortunately, progress in understanding the role of PDE1B in neuronal function has been hampered by the lack of potent and selective brain penetrant inhibitors of the enzyme. Much of what we do know about the function of PDE1B in the brain derives from studies of a PDE1B knockout mouse [29]. PDE1B KO mice demonstrate impaired spatial learning in the Morris water maze, increased levels of novelty induced locomotor activity as well as an increased sensitivity to the stimulant actions of methamphetamine. D1 dopamine agonists, presumably via their ability to increases cAMP or cGMP levels, produce an enhanced phosphorylation of the PKA/PKG substrate DARPP-32 in brain slices from KO mice. Phosphorylation of DARPP-32 by PKA or PKG generates a potent inhibitor of protein phosphatase 1, leading to increased phosphorylation and activation of a number of voltage and ligand gated ion channels. Interestingly, DARPP-32 KO mice show a reduced response to the stimulant effects of low doses of amphetamine presumably due to the absence of the DARPP32 mediated brake on phosphatase activity. A recent study using dual DARPP-32 /PDE1B KO animals showed that the exaggerated response of PDE1B knockouts to methamphetamine were absent in the dual KOs strongly supporting the hypothesis that this effect of the PE1B KO is in part mediated by enhanced DARPP-32 phosphorylation [30]. The effect of PDE1B KO on D1 signaling suggests the potential for PDE1B inhibitors to augment D1 dopamine receptor function. It would be critical to determine whether this may be the case in regions other than the striatum, e.g. the prefrontal cortex or hippocampus. D1 receptor and PKA activation have been shown to be critical for the induction of hippocampal LTP [31,32]. Although PDE1B KO mice have been reported to exhibit deficits in spatial learning, it would be of interest to test the effects of PDE1B inhibition on LTP acutely. A chemical tool to explore the function of PDE1B would have a significant impact on our understanding of this enzyme. Upregulation of PDE1B in the striatum after dopamine depletion has lead to speculation that PDE1B inhibitors may be beneficial in Parkinson’s disease [33] and


225

perhaps in those components of cognitive dysfunction believed to involve striatal dysfunction. A nonselective inhibitor of PDE1 activity, vinpocetine has been used in some studies to assess the potential effect of PDE1B on cognitive processes. A recent study used vinpocetine to reverse the cognitive deficits produced in a model of central oxidative stress by intracerebroventricular administration of streptozotocin [34]. Chronic treatment with vinpocetine (21 days) reversed the detrimental effect of streptozotocin treatment on passive avoidance and the Morris water maze. PDE2A PDE2A is the only member of the PDE2 family although the 3 described splice variants allow for soluble and membrane associated forms of the enzyme [35]. Like PDE1B, PDE2A is a dual substrate enzyme with slightly higher affinity for cGMP although it may metabolize either cAMP or cGMP depending on the tissue. Although expressed in the periphery, the highest expression levels of PDE2A are in the brain. Studies of the intracellular distribution of PDE2A indicate both membrane and cytoplasmic localization Palymitoylation may be responsible for the localization of cortical PDE2A in lipid rafts. PDE2A is also one of the five PDE families with tandem GAF domains in their N-terminal regulatory domain. GAF domains are a phylogenetically ancient small molecule binding motif generally believed to bind cGMP in vivo [36]. The binding of cGMP to PDE2A can increase the maximum enzyme activity by 30-fold providing a mechanism for the rapid termination of a cyclic nucleotide signal. It is likely that only the second or GAF-B domain of PDE2 binds cGMP while it is speculated that the GAF A domain may be involved in stabilizing dimer formation. Although most interest in PDE2A has focused on inhibition of the enzyme, recently, the concept of developing GAF domain ligands for the purpose of activating PDE2 has been reported [37]. The ability of PDE2A to metabolize cAMP in response to cGMP activation allows the enzyme to function as a site of cross-talk between the cAMP and cGMP cascades. In vivo evidence for this cross talk has been observed in the regulation of aldosterone release within the zona glomerulosa of the adrenal gland. In the latter system, ANP mediated elevation of cGMP reduces aldosterone secretion by activating PDE2A to reduce cAMP. The activation of PDE2A is sufficient to reduce aldosterone secretion even in the presence of continued cAMP synthesis [38]. This rapid termination of a cAMP signal again illustrates the ability of PDE activity to sculpt second messenger signaling. Interestingly, most studies of the role of PDE2A in brain tissue point to its modulation of cGMP. It is possible that the endogenous substrate of the enzyme substrate varies across tissues possibly as a function of subcellular localization and or splice variant. A recent immunohistochemical study demonstrated a consistent pattern of PDE2A expression in the brain across mammalian species included human [39]. The enzyme expression was prominent in regions associated with cognitive function including the cortex, striatum, hippocampus, amygdala and the habenula. Scattered but significant levels of expression are also observed in the important regions of


the hypothalamus and brain stem. Interestingly, the majority of PDE2A immunoreactivity is associated with neuronal process rather than with cell bodies although rare immunopositive cells can be found throughout the brain. The preferential expression of PDE2A protein in neuronal processes is particularly evident in the hippocampus where both granule and pyramidal cells are virtually devoid of PDE2A protein despite their high levels of mRNA expression. This pattern is most striking in dentate granule cells which can be observed in coronal sections of the rat brain as a crescent of non-reacting cells adjacent to the mossy fiber projections which are among the most intensely stained structures in the brain. The presence of high levels of PDE2A in mossy fiber is interesting given the demonstration of a presynaptic, cAMP-dependent LTP at their synapses on to CA3 pyramidal cells [40]. One of the first studies to evaluate the function of PDE2A in neurons contrasted the effects of rolipram and the nonselective PDE2A inhibitor EHNA on primary cultures of rat cortical neurons [41]. Exposure of these cultures to NMDA produces an immediate Ca2+-mediated increase in intracellular concentrations of both cAMP and cGMP due to the presence of Ca2+ stimulated AC and NOS/sGC. The NMDA stimulated increase in cAMP was selectively enhanced by rolipram while the cGMP signal was selectively enhanced by EHNA. According this scheme, the intracellular response to NMDA receptor activation consists minimally of a Ca2+, cAMP and cGMP response with the cyclic nucleotide signals being subject to differential regulation by PDEs. These results suggest the possibility that diverging arms of the NMDA signaling cascade could be independently augmented by specific PDE inhibitors. At the very least, it seems clear that such inhibitors provide the means to isolate and identify the unique functions of each cyclic nucleotide cascade. The only selective PDE2A inhibitor described in the literature to date, Bay 60-7550, also preferentially increases cGMP in primary neuronal cultures and hippocampal slices. Bay 60-7550 also increases LTP induction in rat hippocampal slices. Consistent with its biochemical and electrophysiological effects, Bay 60-7550 was found to be active in novel object and social recognition tasks [42]. More recently, Bay 60-7550 was reported to reverse the deficit in object recognition produced by tryptophan depletion [43]. These results are interesting in light of the PDE2 positive cells identified in the dorsal raphe, a region known to contain the cell bodies of the serotonergic neurons projecting to the forebrain [39]. A similar study in aged rats demonstrated that the beneficial effect of Bay 60-7550 on object recognition could be reversed by a nNOS inhibitor, strongly suggesting that the effects of PDE2A inhibition in the CNS are due to alterations in the levels of cGMP [44]. Several recent studies indicate that PDE2A inhibition may have efficacy in the treatment of anxiety states [45]. Induction of oxidative stress in mice by depletion of central glutathione levels with buthionine sulfoximine (BSO) results in an increase in a number of anxiety like behaviors assessed by open field time and the elevated plus maze. These effects were reversed by treatment with Bay 60-7550. In primary cultures of rat cortical neurons, the increase in RSO levels


following exposure to BSO was reversed by BAY60-7550. The effect of the PDE2A inhibitor itself was prevented by the PKG inhibitor, KT-5823 but not the PKA inhibitor H-89. Additionally, the effects of Bay 60-7550 were associated with an increase in the phosphorylation of PKG substrate VASP-Ser239 in both the hypothalamus and amygdala. In a subsequent study, the anxiolytic effects of Bay 60-7550 were prevented by pretreatment with the guanylyl cyclase inhibitor ODQ. These data lend further support to the hypothesis that PDE2 is an important regulator of cGMP in the brain despite its established role as a cAMP phosphodiesterase in the periphery. The potential activity of PD2 inhibitors as both a cognitive enhancer and an anxiolytic suggests that such drugs may have a unique clinical profile in the treatment of the cognitive symptoms associated with affective disorders. PDE4 Inhibitors Of the PDEs, PDE4 has by far received the greatest attention as a target for improving cognitive function [5], [46]. The PDE4 family, encoded by four genes A –D, is the most complex and best studied of all the PDEs. Over 20 splice variants have been identified within the family with the majority involving differential splicing of the N-terminal domains generating long and short forms of the enzymes presumably allowing for differing subcellular localization and regulation [2]. PDE4 are all high affinity cAMP specific enzymes. PDEA, B and D are the dominant isozymes found in the brain although the regional pattern of subtype expression may vary between species [44]. Both PDE4A and B are expressed at high levels in the frontal and temporal cortex and hippocampus of the human brain [47]. The study of PDE4 has benefited from the early availability of the highly selective and brain penetrant inhibitor rolipram. Preclinical studies with rolipram have indicated PDE4 inhibition may be of use in the treatment of the cognitive dysfunction associated with Alzheimer’s disease [5], schizophrenia [48,49], depression and aging. In addition to improving memory function in unimpaired animals, rolipram and other PDE4 inhibitors have been shown to reverse memory disruption produced by scopolamine, MK801, tryptophan depletion and MAPK inhibition. Recently rolipram was shown to improve performance in an object retrieval task in nonhuman primates, a prefrontal task of executive function [54]. Executive function has been identified as a key cognitive domain showing deficits in schizophrenia. Beyond symptomatic treatment, the ability of rolipram to increase levels of the neurotrophin, BDNF, has lead to suggestions that PDE4 inhibition has the potential to halt or reverse the progression of some neurodegenerative conditions. A recent study of the effect of rolipram on object memory in rodents demonstrated beneficial effects of subchronic but not acute treatment on memory which persisted beyond the disappearance of drug from the plasma. The data support the contention that PDE4 inhibition can produce sustained changes in neuronal function most likely mediated through changes in gene transcription [50]. In addition to cognition but perhaps related to such effects on synaptic function, preclinical studies also suggest PDE4 inhibition can provide antipsychotic and antidepressant efficacy. Indeed, PDE4B is a major interacting protein of the


multifunctional scaffold protein/risk factor “disordered in schizophrenia 1” (DISC1), and disruption of its function has been suggested to play a significant role in the contribution of DISC1 variants to psychopathology [51]. At the synaptic level, PDE4 inhibition has been shown to augment hippocampal LTP and to enhance the phosphorylation of the transcription factor CREB, a key player in synaptic plasticity. PDE4 inhibitors reportedly improve longterm but not short-term memory formation perhaps in keeping with data indicating the former to be cAMP mediated while the latter is under the control of cGMP [52] Despite the immense body of the data supporting the use of PDE4 inhibitors for the treatment of cognition and years of pharmaceutical effort, the serious and up to now insurmountable side-effect issues associated with PDE4 inhibitors have halted the development every PDE4 inhibitor tested to date even before late stage clinical evaluation. The most significant of these toleration issues are vasculitis and emesis. While it has been argued that vasculitis may be a rodent specific problem that may be addressed in preclinical development, nausea and emesis have derailed a number of PDE4 inhibitors in the clinic [53]. An obvious strategy for circumventing this problem requires the development of subtype selective PDE inhibitor in the hope that therapeutic efficacy and side-effect issues might be dissociated. This remains a challenging approach due to the similar catalytic domain structure within the PDE4 family (but see below). Although emesis cannot be observed in rodents, studies using KO mice have suggested that the emetic activity of PDE4 inhibitors may be attributed to inhibition of PDE4D [54]. Indeed, unlike PDE4A and PDE4B, PDE4D is highly expressed in the area postrema, a region of the brain stem known to be involved in initiation of the emetic response [55] and as discussed elsewhere [3], the clinical trial of a 100-fold selective PDE4D inhibitor were terminated by Pfizer, Inc. due to early signs of emesis. The development of compounds lacking activity at PDE4D may therefore provide a path forward. Interestingly, the broad distribution of PDE4D in the CNS has lead some groups to consider alternative ways to target this variant. DeCode Genetics has reported the development of a series of allosteric PDE4 inhibitors based upon crystallographic analysis of the PDE4D and 4B regulatory domains (Company Website). An IND filing by the company was reported in 2008 for a noncompetitive inhibitor of PDE4D. The compound, DG071 has been reported to be active in a number of animal models of cognition while exhibiting a reduced risk of nausea and emesis based on studies in nonhuman primates. It is possible that regional differences in cAMP concentration could contribute to regionally difference degrees of PDE4D inhibition by competitive (active site) versus noncompetitive inhibitors. Such differences could lead to an improved sideeffect profile. The ability to produce PDE4D subtype selectivity by targeting allosteric regions of the enzyme suggests the same strategy could be used to develop selective inhibitors of PDE4A and B. Perhaps such a strategy will one day lead to the successful development of a PDE4 inhibitors capable of being used in the treatment of CNS disorders including cognitive dysfunction.


227

PDE5 Selective PDE5 inhibitors have enjoyed great commercial success as a treatment for erectile dysfunction and pulmonary hypertension. It has been the availability of these selective inhibitors that has driven the study of PDE5 as a potential cognition target. This interest has continued to develop despite some degree of controversy about the expression level of PDE5 in the brain [56]. PDE5A is the single member of this family although multiple promoters generate N-terminal splice variants to allow both cAMP and cGMP to regulation expression of the enzyme. PDE5A is a cGMP specific PDE with two GAF binding domains for cGMP. Occupancy of the GAF-A domain by cGMP increases enzyme activity by approximately 10-fold and increases phosphorylation by PKG. The phosphorylation of PDE5 by PKG has been shown to prolong the activation state of the enzyme by increasing the affinity of the GAF domain for cGMP [2]. The first studies illustrating the potential for PDE5 inhibition to improve cognition were conducted with the nonselective cGMP PDE inhibitor zaprinast (see an excellent review by Reneerkens et al., 2009) utilizing the object recognition task. The results of these studies have subsequently been replicated with more selective inhibitors such as sildenafil and vardenafil. In addition to improving baseline function, PDE5A inhibition has been shown to prevent memory deficits produced in rodents by nNOS inhibitors, scopolamine, hyperammonemia, electroshock and diabetes. More recently, Rutten et al., [57] have compared the effect of PDE4 inhibition with rolipram and PDE5 inhibition with sildenafil on an executive function task in cynomolgus monkeys. In this object retrieval task, both rolipram and sildenafil improved performance from approximately 50% correct to 75%. Although the effects of both cAMP and cGMP on synaptic plasticity, including LTP, have been used to rationalize the efficacy of PDE4 and 5 inhibitors in memory tasks, the mechanistic explanation for their effect on prefrontal functions such as executive function remains to be determined. The effect of cGMP on synaptic plasticity has lead to the evaluation of PDE5 inhibitors in animal models of Alzheimer’s disease [6]. Although the mature pathology in Alzheimer’s disease involves both amyloid plaques and neurofibrillary tangles, the earliest symptoms of the disease, particularly memory loss, may be the result of synaptic dysfunction produced by accumulation of the amyloid peptide, A. Both LTP and CREB phosphorylation can be disrupted by A and the peptide has been reported to interfere with elements of the NO/cGMP signaling cascade [58]. Based on these observations, studies have examined the effect of PDE5 inhibition on synaptic function in mice over expressing the amyloid precursor protein [59]. Sildenafil improved LTP in hippocampal slices from mutant mice following acute bath application or chronic administration in vivo. PDE5 inhibition also improved performance in tests of fear conditioning, reference memory and spatial working memory. The authors also provide evidence that PDE5 inhibition restores normal levels of CREB phosphorylation.


The data are consistent with the hypothesis that by enhancing synaptic activity, elevated levels of cGMP may strengthen synapses and counter the damaging effects of A. However, despite the number of preclinical studies examining the effect of PDE5 inhibitors on cognitive function, the expression level of PDE5 in the brain remains a controversial issue. While PDE5A is expressed at high levels in the smooth muscle of the cerebrovasculature, the Purkinje cells of the cerebellum are the only neuronal population showing appreciable expression of PDE5 protein. A recent immunohistochemical study by [56] confirmed these findings and noted only scattered cells with low levels of PDE5A protein in other brain regions. Predictably, these observations have drawn attention to the well known vascular effects of PDE5 inhibitors however, a recent study using vardenafil found the beneficial effects of this compound in novel object recognition were independent of changes cerebral blood flow [60]. The availability of selective and safe PDE5 inhibitors has lead to some limited clinical testing. Goff et al., [61] recently reported no effect of acute sildenafil on cognitive performance in a small group of schizophrenic patients stabilized on antipsychotic medication. Unfortunately, the PDE5 inhibitors currently marketed were not developed for CNS indications and the issue of achieving good central exposure hangs over all such trials. In addition, given that it is currently unclear what duration of treatment might be required to improve cognitive function in schizophrenia, the outcome of short duration trials must be viewed with caution. PDE9A Like the PDE5, PDE9A is cGMP specific enzyme. Unlike PDE5, little is known about the regulation of PDE9 and its N-terminal region lacks any known regulatory elements. PDE9A has the highest affinity for cGMP among all the PDEs (Km = 0.17 uM) leading to the suggestion that the enzyme may control basal levels of the cyclic nucleotide [62]. PDE9 is encoded by a single gene (PDE9A) although 21 splice variants have been identified. Although there are no published studies of PDE9A protein localization, message levels have been studied in rats and human [63, 64]). In humans, the highest message levels are found in brain, spleen and small intestine. Within the brain, PDE9A message is widely distributed at low levels throughout the cortex, hippocampus and striatum. High levels of PDE9A mRNA are found in the Purkinje neurons, and to a lesser extent, in the granule cells of the cerebellum. The established role of cGMP in the regulation of synaptic function suggests PDE9 inhibitors may have significant effects in the CNS. Until recently, the only selective PDE9A inhibitor described in the neuroscience literature was Bay 73-6691 [65]. In addition to some modest effects on hippocampal LTP, Bay 73-6691 demonstrated activity in several cognition models including the object and social recognition tasks. Bay 730-691 also reversed the effect of scopolamine in passive avoidance and appeared to have a trend effect for reversal of MK-801-induced deficits in delayed alternation. We have recently described studies using both wild-type and PDE9A KO mice and a new series


of PDE9A inhibitors. Mice with a null form of PDE9A show constitutive increases in cGMP in several brain regions including the cortex, hippocampus striatum and cerebellum. In addition, CSF concentrations of cGMP were also increased in the KOs. Consistent with the selectivity of PDE9A, there were no changes in cAMP in any region examined. Inhibition of PDE9A with the selective inhibitor PF-4181366 produced dose-dependent elevation in cGMP in the same brain regions including in CSF [66]. The availability of agents such as Bay 73-6691 and PF-4181366 should have a dramatic effect on our understanding of the role of PDE9A on CNS function over the next several years. More importantly, these studies will inform us as to the potential utility of PDE9A inhibitors as therapeutic agents. SUMMARY The substantial and ubiquitous modulation of neurotransmission by cyclic nucleotides accounts for the high concentration of PDEs in the brain as well as the significant impact of PDE inhibition on CNS function. The PDEs represent a very “drug-able” family of enzymes and of the PDEs generally considered to be potential targets for cognition, selective inhibitors now exist for PDE2A, 5A and 9A. The need to inhibit subtypes within the PDE1and 4 families has presented a challenge to the development of agents capable of probing the CNS function of PDE1B and PDE4A/B. Although a limited number of agents have been available for preclinical research, it is clear that the modulation of cyclic nucleotide signaling in the brain represents a powerful approach to the manipulation of neuronal activity. Hopefully, the expanding list of selective PDE inhibitors presages a rapid growth in our understanding of the function and therapeutic potential of this family of enzymes. It seems safe to predict that the next decade will see the clinical evaluation of a number of PDE inhibitors for a variety of CNS indications including cognition. ACKNOWLEDGEMENT The author wishes to thank Shelley Erb for her help in completing this manuscript REFERENCES [1] [2]

[3] [4]

[5] [6]

[7]

Goraya, T. A.; Cooper, D. M. Ca2+-calmodulin-dependent phosphodiesterase (PDE1): current perspectives. Cell Signal, 2005, 17, 789-797. Bender, A. T.; Beavo, J. A. Cyclic nucleotide phosphodiesterases: molecular regulation to clinical use. Pharmacol. Rev., 2006, 58, 488-520. Menniti, F. S.; Faraci, W. S.; Schmidt, C. J. Phosphodiesterases in the CNS: targets for drug development. Nat. Rev. Drug Discov., 2006, 5, 660-670. Lugnier, C. Cyclic nucleotide phosphodiesterase (PDE) superfamily: a new target for the development of specific therapeutic agents. Pharmacol. Ther., 2006, 109, 366-398. Ghavami, A.; Hirst, W. D.; Novak, T. J. Selective phosphodiesterase (PDE)-4 inhibitors: a novel approach to treating memory deficit? Drugs R D, 2006, 7, 63-71. Puzzo, D.; Sapienza, S.; Arancio, O.; Palmeri, A. Role of phosphodiesterase 5 in synaptic plasticity and memory. Neuropsychiatr Dis. Treat., 2008, 4, 371-387. Reneerkens, O. A.; Rutten, K.; Steinbusch, H. W.; Blokland, A.; Prickaerts, J. Selective phosphodiesterase inhibitors: a promising target for cognition enhancement. Psychopharmacology, (Berl) 2009, 202, 419-443.

Phosphodiesterase Inhibitors as Potential Cognition Enhancing Agents [8]

[9]

[10]

[11]

[12]

[13]

[14] [15] [16]

[17] [18]

[19]

[20] [21]

[22] [23]

[24] [25] [26]

[27] [28]

Bales, K.R.; Menniti, F.S. Promoting synaptic resilience in Alzheimer’s disease patients through PDE inhibition. In: Emerging Drugs and Targets for AD. Martinez, A., Ed, Springer, in press 2009. Rodefer, J. S.; Murphy, E. R.; Baxter, M. G. PDE10A inhibition reverses subchronic PCP-induced deficits in attentional set-shifting in rats. Eur. J. Neurosci., 2005, 21, 1070-1076. Grauer, S. M.; Pulito, V. L.; Navarra, R. L.; Kelly, M. P.; Kelley, C.; Graf, R.; Langen, B.; Logue, S.; Brennan, J.; Jiang, L.; Charych, E.; Egerland, U.; Liu, F.; Marquis, K. L.; Malamas, M.; Hage, T.; Comery, T. A.; Brandon, N. J. Phosphodiesterase 10A inhibitor activity in preclinical models of the positive, cognitive, and negative symptoms of schizophrenia. J. Pharmacol. Exp. Ther., 2009, 331, 574-590. Schmidt, C. J.; Chapin, D. S.; Cianfrogna, J.; Corman, M. L.; Hajos, M.; Harms, J. F.; Hoffman, W. E.; Lebel, L. A.; McCarthy, S. A.; Nelson, F. R.; Proulx-LaFrance, C.; Majchrzak, M. J.; Ramirez, A. D.; Schmidt, K.; Seymour, P. A.; Siuciak, J. A.; Tingley, F. D., 3rd; Williams, R. D.; Verhoest, P. R.; Menniti, F. S. Preclinical characterization of selective phosphodiesterase 10A inhibitors: a new therapeutic approach to the treatment of schizophrenia. J. Pharmacol. Exp. Ther., 2008, 325, 681-690. Sadana, R.; Dessauer, C. W. Physiological roles for G proteinregulated adenylyl cyclase isoforms: insights from knockout and overexpression studies. Neurosignals, 2009, 17, 5-22. Garbers, D. L.; Chrisman, T. D.; Wiegn, P.; Katafuchi, T.; Albanesi, J. P.; Bielinski, V.; Barylko, B.; Redfield, M. M.; Burnett, J. C., Jr. Membrane guanylyl cyclase receptors: an update. Trends Endocrinol. Metab., 2006, 17, 251-258. Biel, M. Cyclic nucleotide-regulated cation channels. J. Biol. Chem., 2009, 284, 9017-9021. Borland, G.; Smith, B. O.; Yarwood, S. J. EPAC proteins transduce diverse cellular actions of cAMP. Br. J. Pharmacol., 2009, 158, 7086. Dessauer, C. W. Adenylyl cyclase-a-kinase anchoring protein complexes: the next dimension in cAMP signaling. Mol. Pharmacol., 2009, 76, 935-941. Baillie, G. S. Compartmentalized signalling: spatial regulation of cAMP by the action of compartmentalized phosphodiesterases. FEBS J., 2009, 276, 1790-1799. Pandit ,J.; Forman, M.D.; Fennell, K.F.; Dillman, K.S.; Menniti, F.S.Mechanism for the allosteric regulation of phosphodiesterase 2A deduced from the X-ray structure of a near full length construct. Proc. Natl. Acad. Sci. USA, 2009, 106(43), 18225-8230. Beavo, J. A.; Francis, S. H.; Houslay, M. D. Cyclic Nucleotide Phosphodiesterases in Health and Disease; CRC Press: USA, 2006. Zakzanis, K. K. Quantitative evidence for neuroanatomic and neuropsychological markers in dementia of the Alzheimer's type. J. Clin. Exp. Neuropsychol., 1998, 20, 259-269. Heinrichs, R. W.; Zakzanis, K. K. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology, 1998, 12, 426-245. Zakzanis, K. K.; Leach, L.; Kaplan, E. On the nature and pattern of neurocognitive function in major depressive disorder. Neuropsychiatry Neuropsychol. Behav. Neurol., 1998, 11, 111-119. Holtmaat, A.; Svoboda, K. Experience-dependent structural synaptic plasticity in the mammalian brain. Nat. Rev. Neurosci., 2009, 10, 647-658. Citri, A.; Malenka, R. C. Synaptic plasticity: multiple forms, functions, and mechanisms. Neuropsychopharmacology, 2008, 33, 18-41. Lee, Y. S.; Silva, A. J. The molecular and cellular biology of enhanced cognition. Nat. Rev. Neurosci., 2009, 10, 126-140. Dere, E.; Huston, J. P.; De Souza Silva, M. A. The pharmacology, neuroanatomy and neurogenetics of one-trial object recognition in rodents. Neurosci. Biobehav. Rev., 2007, 31, 673-704. Polli, J. W.; Kincaid, R. L. Molecular cloning of DNA encoding a calmodulin-dependent phosphodiesterase enriched in striatum. Proc. Natl . Acad. Sci. USA, 1992, 89, 11079-11083. Xie, Z.; Adamowicz, W. O.; Eldred, W. D.; Jakowski, A. B.; Kleiman, R. J.; Morton, D. G.; Stephenson, D. T.; Strick, C. A.; Williams, R. D.; Menniti, F. S. Cellular and subcellular localization of PDE10A, a striatum-enriched phosphodiesterase. Neuroscience, 2006, 139, 597-607.

Current Topics in Medicinal Chemistry, 2010, Vol. 10, No. 2 [29]

[30]

[31]

[32]

[33]

[34]

[35]

[36] [37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45] [46] [47]

229

Reed, T. M.; Repaske, D. R.; Snyder, G. L.; Greengard, P.; Vorhees, C. V. Phosphodiesterase 1B knock-out mice exhibit exaggerated locomotor hyperactivity and DARPP-32 phosphorylation in response to dopamine agonists and display impaired spatial learning. J. Neurosci., 2002, 22, 5188-5197. Ehrman, L. A.; Williams, M. T.; Schaefer, T. L.; Gudelsky, G. A.; Reed, T. M.; Fienberg, A. A.; Greengard, P.; Vorhees, C. V. Phosphodiesterase 1B differentially modulates the effects of methamphetamine on locomotor activity and spatial learning through DARPP32-dependent pathways: evidence from PDE1BDARPP32 double-knockout mice. Genes Brain Behav., 2006, 5, 540-551. Stramiello, M.; Wagner, J. J. D1/5 receptor-mediated enhancement of LTP requires PKA, Src family kinases, and NR2B-containing NMDARs. Neuropharmacology, 2008, 55, 871-877. Granado, N.; Ortiz, O.; Suarez, L. M.; Martin, E. D.; Cena, V.; Solis, J. M.; Moratalla, R. D1 but not D5 dopamine receptors are critical for LTP, spatial learning, and LTP-Induced arc and zif268 expression in the hippocampus. Cereb. Cortex., 2008, 18, 1-12. Giorgi, M.; D'Angelo, V.; Esposito, Z.; Nuccetelli, V.; Sorge, R.; Martorana, A.; Stefani, A.; Bernardi, G.; Sancesario, G. Lowered cAMP and cGMP signalling in the brain during levodopa-induced dyskinesias in hemiparkinsonian rats: new aspects in the pathogenetic mechanisms. Eur. J. Neurosci., 2008, 28, 941-950. Deshmukh, R.; Sharma, V.; Mehan, S.; Sharma, N.; Bedi, K. L. Amelioration of intracerebroventricular streptozotocin induced cognitive dysfunction and oxidative stress by vinpocetine -- a PDE1 inhibitor. Eur. J. Pharmacol., 2009, 620, 49-56. Russwurm, C.; Zoidl, G.; Koesling, D.; Russwurm, M. Dual acylation of PDE2A splice variant 3: targeting to synaptic membranes. J. Biol. Chem., 2009, 284, 25782-25790. Martinez, S. E.; Beavo, J. A.; Hol, W. G. GAF domains: twobillion-year-old molecular switches that bind cyclic nucleotides. Mol. Interv., 2002, 2, 317-323. Jager, M.; Russwurm, C.; Koesling, D.; Russwurm, M. GAF domain-induced activation of phosphodiesterase 2 and 5. BMC Pharmacology, 2009, 9(Suppl 1), P28. Nikolaev, V. O.; Gambaryan, S.; Engelhardt, S.; Walter, U.; Lohse, M. J. Real-time monitoring of the PDE2 activity of live cells: hormone-stimulated cAMP hydrolysis is faster than hormonestimulated cAMP synthesis. J. Biol. Chem., 2005, 280, 1716-1719. Stephenson, D. T.; Coskran, T. M.; Wilhelms, M. B.; Adamowicz, W. O.; O'Donnell, M. M.; Muravnick, K. B.; Menniti, F. S.; Kleiman, R. J.; Morton, D. Immunohistochemical localization of phosphodiesterase 2A in multiple mammalian species. J. Histochem. Cytochem., 2009, 57, 933-949. Kobayashi, K. Targeting the hippocampal mossy fiber synapse for the treatment of psychiatric disorders. Mol. Neurobiol., 2009, 39, 24-36. Suvarna, N. U.; O'Donnell, J. M. Hydrolysis of N-methyl-Daspartate receptor-stimulated cAMP and cGMP by PDE4 and PDE2 phosphodiesterases in primary neuronal cultures of rat cerebral cortex and hippocampus. J. Pharmacol. Exp. Ther., 2002, 302, 249-256. Boess, F. G.; Hendrix, M.; van der Staay, F. J.; Erb, C.; Schreiber, R.; van Staveren, W.; de Vente, J.; Prickaerts, J.; Blokland, A.; Koenig, G. Inhibition of phosphodiesterase 2 increases neuronal cGMP, synaptic plasticity and memory performance. Neuropharmacology, 2004, 47, 1081-1092. van Donkelaar, E. L.; Rutten, K.; Blokland, A.; Akkerman, S.; Steinbusch, H. W.; Prickaerts, J. Phosphodiesterase 2 and 5 inhibition attenuates the object memory deficit induced by acute tryptophan depletion. Eur. J. Pharmacol., 2008, 600, 98-104. Domek-Lopacinska, K.; Strosznajder, J. B. The effect of selective inhibition of cyclic GMP hydrolyzing phosphodiesterases 2 and 5 on learning and memory processes and nitric oxide synthase activity in brain during aging. Brain Res., 2008, 1216, 68-77. Masood, A.; Nadeem, A.; Mustafa, S. J.; O'Donnell, J. M. Reversal of oxidative stress-induced anxiety by inhibition of phosphodiesterase-2 in mice. J. Pharmacol .Exp. Ther., 2008, 326, 369-379. Monti, B.; Contestabile, A. Memory-enhancing drugs: a molecular perspective. Mini Rev. Med. Chem., 2009, 9, 769-781. Perez-Torres, S.; Miro, X.; Palacios, J. M.; Cortes, R.; Puigdomenech, P.; Mengod, G. Phosphodiesterase type 4 isozymes expression in human brain examined by in situ hybridization histochemistry and[3H]rolipram binding autoradiography. Com-


[48]

[49]

[50]

[51]

[52]

[53] [54]

[55]

[56]

[57]

parison with monkey and rat brain. J. Chem. Neuroanat., 2000, 20, 349-374. Halene, T. B.; Siegel, S. J. PDE inhibitors in psychiatry--future options for dementia, depression and schizophrenia? Drug Discov. Today, 2007, 12, 870-878. Siuciak, J. A.; Chapin, D. S.; McCarthy, S. A.; Martin, A. N. Antipsychotic profile of rolipram: efficacy in rats and reduced sensitivity in mice deficient in the phosphodiesterase-4B (PDE4B) enzyme. Psychopharmacology (Berl), 2007, 192, 415-424. Rutten, K.; Prickaerts, J.; Schaenzle, G.; Rosenbrock, H.; Blokland, A. Sub-chronic rolipram treatment leads to a persistent improvement in long-term object memory in rats. Neurobiol. Learn Mem., 2008, 90, 569-575. Millar, J. K.; Mackie, S.; Clapcote, S. J.; Murdoch, H.; Pickard, B. S.; Christie, S.; Muir, W. J.; Blackwood, D. H.; Roder, J. C.; Houslay, M. D.; Porteous, D. J. Disrupted in schizophrenia 1 and phosphodiesterase 4B: towards an understanding of psychiatric illness. J. Physiol., 2007, 584, 401-5. Rutten, K.; Prickaerts, J.; Hendrix, M.; van der Staay, F. J.; Sik, A.; Blokland, A. Time-dependent involvement of cAMP and cGMP in consolidation of object memory: studies using selective phosphodiesterase type 2, 4 and 5 inhibitors. Eur. J. Pharmacol., 2007, 558, 107-112. Spina, D. PDE4 inhibitors: current status. Br. J. Pharmacol., 2008, 155, 308-15. Robichaud, A.; Stamatiou, P. B.; Jin, S. L.; Lachance, N.; MacDonald, D.; Laliberte, F.; Liu, S.; Huang, Z.; Conti, M.; Chan, C. C. Deletion of phosphodiesterase 4D in mice shortens alpha(2)adrenoceptor-mediated anesthesia, a behavioral correlate of emesis. J. Clin. Invest, 2002, 110, 1045-1052. Cherry, J. A.; Davis, R. L. Cyclic AMP phosphodiesterases are localized in regions of the mouse brain associated with reinforcement, movement, and affect. J. Comp. Neurol., 1999, 407, 2872301. Menniti, F. S.; Ren, J.; Coskran, T. M.; Liu, J.; Morton, D. G.; Sietsma, D. K.; Som, A.; Stephenson, D. T.; Tate, B. A.; Finklestein, S. P. PDE5A inhibitors improve functional recovery after stroke in rats: optimized dosing regimen with implications for mechanism. J. Pharmacol. Exp. Ther., 2009, 331(3), 842-850. Rutten, K.; Basile, J. L.; Prickaerts, J.; Blokland, A.; Vivian, J. A. Selective PDE inhibitors rolipram and sildenafil improve object retrieval performance in adult cynomolgus macaques. Psychopharmacology (Berl), 2008, 196, 643-648.

Received: November 10, 2009

Revised: December 4, 2009

Christopher J. Schmidt [58]

[59]

[60]

[61]

[62]

[63] [64]

[65]

[66]

Puzzo, D.; Vitolo, O.; Trinchese, F.; Jacob, J. P.; Palmeri, A.; Arancio, O. Amyloid-beta peptide inhibits activation of the nitric oxide/cGMP/cAMP-responsive element-binding protein pathway during hippocampal synaptic plasticity. J. Neurosci., 2005, 25, 6887-6897. Puzzo, D.; Staniszewski, A.; Deng, S. X.; Privitera, L.; Leznik, E.; Liu, S.; Zhang, H.; Feng, Y.; Palmeri, A.; Landry, D. W.; Arancio, O. Phosphodiesterase 5 inhibition improves synaptic function, memory, and amyloid-beta load in an Alzheimer's disease mouse model. J. Neurosci., 2009, 29, 8075-8086. Rutten, K.; Van Donkelaar, E. L.; Ferrington, L.; Blokland, A.; Bollen, E.; Steinbusch, H. W.; Kelly, P. A.; Prickaerts, J. H. Phosphodiesterase inhibitors enhance object memory independent of cerebral blood flow and glucose utilization in rats. Neuropsychopharmacology, 2009, 34, 1914-1925. Goff, D. C.; Cather, C.; Freudenreich, O.; Henderson, D. C.; Evins, A. E.; Culhane, M. A.; Walsh, J. P. A placebo-controlled study of sildenafil effects on cognition in schizophrenia. Psychopharmacology (Berl), 2009, 202, 411-417. Fisher, D. A.; Smith, J. F.; Pillar, J. S.; St Denis, S. H.; Cheng, J. B. Isolation and characterization of PDE9A, a novel human cGMPspecific phosphodiesterase. J. Biol. Chem., 1998, 273, 1555915564. Andreeva, S. G.; Dikkes, P.; Epstein, P. M.; Rosenberg, P. A. Expression of cGMP-specific phosphodiesterase 9A mRNA in the rat brain. J. Neurosci., 2001, 21, 9068-9076. Reyes-Irisarri, E.; Markerink-Van Ittersum, M.; Mengod, G.; de Vente, J. Expression of the cGMP-specific phosphodiesterases 2 and 9 in normal and Alzheimer's disease human brains. Eur. J. Neurosci., 2007, 25, 3332-3338. van der Staay, F. J.; Rutten, K.; Barfacker, L.; Devry, J.; Erb, C.; Heckroth, H.; Karthaus, D.; Tersteegen, A.; van Kampen, M.; Blokland, A.; Prickaerts, J.; Reymann, K. G.; Schroder, U. H.; Hendrix, M. The novel selective PDE9 inhibitor BAY 73-6691 improves learning and memory in rodents. Neuropharmacology, 2008, 55, 908-918. Schmidt, C. J.; Harms, J. F.; Tingley, F. D.; Schmidt, K.; Adamowicz, W. O.; Romegialli, A.; Kleiman, R. J.; Barry, C. J.; Coskran, T. M.; O'Neill, S. M. PDE9A-mediated regulation of cGMP: Developing a biomarker for a novel therapy for Alzheimer's disease. Alzheimer's Dementia, 2009, 5, 331-331.