. Phosphodiesterase 5 inhibition improves synaptic function, memory, and amyloid-beta load in an Alzheimer's disease mouse model. J Neurosci. 2009 Jun 24;29(25):8075-86. PubMed.


Please login to recommend the paper.


  1. This is a very interesting paper. It is a logical step in the line of investigation of this very good research group. Sildenafil has acute treatment effects, as well as long-lasting effects, after chronic treatment in a mouse model of Alzheimer disease (AD). Both synaptic plasticity (LTP) and memory performance are improved. Very importantly, Aβ load in the brain is decreased after chronic treatment. The authors suggest that this study is a proof of concept for PDE5 inhibitors as a new class of drugs for treatment of AD. This is, in a way, correct. However, whether PDE5 is eventually the right target for AD is not clear yet. It has been shown that the expression of PDE5 is strongly reduced in Alzheimer disease patients (Reyes-Irisarri et al., 2007). Along similar lines, PDE5 inhibition did not improve memory in aged rats (Domek-Lopacinska and Strosznajder, 2008). So, it might be that the target of sildenafil is reduced in the aged or Alzheimer’s brain. These mice were still relatively young (treated at three months and tested at six to eight months), so the target was still present sufficiently. Thus, as long as PDE5 is present it seems to be a useful target for treating memory deficits. Whether it can be applied to late stages of Alzheimer disease still needs to be established. The most important point, in my view, is that this study is proof that PDE inhibitions in general offer therapeutic targets for treatment of memory disorders (see also an earlier publication of the same group with rolipram, a PDE4 inhibitor: Gong et al., 2004). In this respect one should not focus on PDE5 only. However, PDE5 is the only PDE (out of a family of 11) which one can use to perform translational memory studies since there are already well-described and approved drugs on the market, e.g., Levitra (vardenafil) and Viagra (sildenafil). In this respect the choice of the authors for PDE5 is clear. Lastly, It is common that cognition enhancers do not work in healthy animals, unless one uses deficit models, e.g., drugs, lesions, aged, etc. This is because the cognitive performance of healthy animals is already at a maximal level. There is no room for further enhancement. Thus, the authors are correct in that “maximal levels are already induced in vehicle-treated WT mice.”


    . Expression of the cGMP-specific phosphodiesterases 2 and 9 in normal and Alzheimer's disease human brains. Eur J Neurosci. 2007 Jun;25(11):3332-8. PubMed.

    . 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 Jun 24;1216:68-77. PubMed.

    . Persistent improvement in synaptic and cognitive functions in an Alzheimer mouse model after rolipram treatment. J Clin Invest. 2004 Dec;114(11):1624-34. PubMed.

  2. Nitric oxide (NO) signaling is essential for normal physiological function of several systems and organs, such as the cardiovascular system and the brain. NO achieves most of its effects through activation of soluble guanylate cyclase (sGC) and production of the intracellular second messenger cGMP. There is consistent evidence that the NO/cGMP signaling pathway plays a key role in modulating synaptic neurotransmission and plasticity widely in the brain, especially in areas such as hippocampus and cortex, which are critical for learning and memory (reviewed in Thatcher et al., 2004). Within this context, this paper of Puzzo et al. provides further support for the role of this pathway in learning and memory and its potential involvement in neurodegenerative processes associated with Alzheimer disease. Indeed, the authors report that upregulation of the NO/cGMP signaling in APP/PS1 transgenic mice through inhibition of cGMP catabolism by the phosphodiesterase 5 (PDE5) inhibitor sildenafil (Viagra, Revatio) rescues fear conditioning learning, attenuates the deficits of spatial working memory in the radial arm maze test, and improves synaptic function in hippocampal slices in vitro. Notably, these effects appear rapidly after the administration of the drug, persist for several weeks after its withdrawal, and are accompanied by reduced β amyloid load and restoration of normal levels of CREB phosphorylation. The significance of these findings is twofold: they point to the NO/cGMP pathway as a relevant pathogenic mechanism in AD, and they indicate that its upregulation may be neuroprotective and have beneficial long-lasting effects on cognition. This view is consistent with previous data obtained in rodent models of stroke (i.e., middle cerebral artery occlusion) where PDE5 inhibitors increase cerebral levels of cGMP, rescue cognitive impairment, and have neuroprotective effects (Zhang et al., 2002; 2005; 2006; Ko et al., 2009). In addition, results achieved with NO-donating derivatives of anti-inflammatory and antioxidant compounds further support a beneficial role of the NO/cGMP signaling in AD neurodegeneration (reviewed in Gasparini et al., 2005). For example, it has been shown that chronic administration of HCT 1026 and NCX 2216, two NO-donating derivatives of the non-steroidal anti-inflammatory drug flurbiprofen, significantly reduces β amyloid load in APP/PS1 mutant double transgenic mice (Jantzen 2002; Van Groen and Kadish, 2005).

    Nevertheless, questions remain unanswered in this challenging scenario, especially in relation to the contribution of cardiovascular mechanisms in the observed effects. Indeed, all PDE5 inhibitors, including tadalafil (Cialis, Adcirca), seem to cross the blood-brain barrier to some extent, as indicated by the adverse events reported in humans following their clinical use (i.e., migraine, seizures, transient global amnesia, optic neuropathy) and by experimental evidence in rodents (Zhang et al., 2002, 2006; Prickaerts et al., 2002; Ko et al., 2009). The NO/cGMP pathway and, in particular, PDE5 inhibitors have remarkable effects on the cardiovascular system. Thus, possible contributions of vascular mechanisms cannot be ruled out based on current data with sildenafil. Further studies are warranted to clarify this important point and exploit the potential of this signaling pathway to design innovative therapeutic strategy for AD treatment.


    . Activity of flurbiprofen and chemically related anti-inflammatory drugs in models of Alzheimer's disease. Brain Res Brain Res Rev. 2005 Apr;48(2):400-8. PubMed.

    . Tadalafil improves short-term memory by suppressing ischemia-induced apoptosis of hippocampal neuronal cells in gerbils. Pharmacol Biochem Behav. 2009 Feb;91(4):629-35. Epub 2008 Oct 29 PubMed.

    . Microglial activation and beta -amyloid deposit reduction caused by a nitric oxide-releasing nonsteroidal anti-inflammatory drug in amyloid precursor protein plus presenilin-1 transgenic mice. J Neurosci. 2002 Mar 15;22(6):2246-54. PubMed.

    . Nitric oxide mimetic molecules as therapeutic agents in Alzheimer's disease. Curr Alzheimer Res. 2005 Apr;2(2):171-82. PubMed.

    . Transgenic AD model mice, effects of potential anti-AD treatments on inflammation and pathology. Brain Res Brain Res Rev. 2005 Apr;48(2):370-8. PubMed.

    . Delayed treatment with sildenafil enhances neurogenesis and improves functional recovery in aged rats after focal cerebral ischemia. J Neurosci Res. 2006 May 15;83(7):1213-9. PubMed.

    . Functional recovery in aged and young rats after embolic stroke: treatment with a phosphodiesterase type 5 inhibitor. Stroke. 2005 Apr;36(4):847-52. PubMed.

    . Sildenafil (Viagra) induces neurogenesis and promotes functional recovery after stroke in rats. Stroke. 2002 Nov;33(11):2675-80. PubMed.

    . Tadalafil, a long-acting type 5 phosphodiesterase isoenzyme inhibitor, improves neurological functional recovery in a rat model of embolic stroke. Brain Res. 2006 Nov 6;1118(1):192-8. PubMed.

  3. This paper by Puzzo and colleagues presents some novel and interesting findings with respect to PDE5 inhibition in an APP/PS1 transgenic mouse model. The authors show that both short-term and long-term administration of sildenafil to APP/PS1 mice improves learning and memory, reverses LTP deficits, and lowers total brain Aβ levels, both Aβ40 and Aβ42. The mechanism of action with respect to these effects remains somewhat unclear, although altered cGMP profiles seems one likely explanation (see in depth discussion in comment by Colton et al. below) resulting in restoration of CREB phosphorylation. The authors have, indeed, previously shown that exogenously applied Aβ lowers CREB phosphorylation, a phenomenon that can be reversed by NO donors and cGMP analogs (Puzzo et al., 2005). The restoration of LTP and reversal of memory deficits are important findings in the current study. The impressive effect of a single dose administration on the behavior of the APP/PS1 mice would, indeed, indicate a rapid mechanism such as the increased CREB phosphorylation.

    The reductions in Aβ by sildenafil treatment are intriguing, and there are many potential mechanisms to be considered including action of degrading enzymes, modification of the microglial responses to Aβ, or secretase modification. It is unclear in the current report whether a particular species of Aβ was modified to account for the profound behavioral and synaptic responses, or whether neuroprotection and Aβ reduction were independent mechanisms. It is not particularly surprising that Aβ levels remained reduced three to five months after the cessation of sildenafil treatment. A recent report by Karnoski and colleagues of Dave Morgan’s laboratory found that suppression of amyloid deposition by passive immunotherapy resulted in long-term reductions in Aβ four months after the termination of antibody administration and clearance of the antibody (Karlnoski et al., 2009). The current report by Puzzo and colleagues confirms these findings using a different mechanism to lower Aβ. An important continuation of this study will be a detailed study of pathological changes in the treated mice including assessment of amyloid species, inflammatory states, and neurovascular health. Together, though, the present data from Puzzo et al. provide us with an interesting new potential target for the treatment of Alzheimer disease.


    . Suppression of amyloid deposition leads to long-term reductions in Alzheimer's pathologies in Tg2576 mice. J Neurosci. 2009 Apr 15;29(15):4964-71. PubMed.

    . Amyloid-beta peptide inhibits activation of the nitric oxide/cGMP/cAMP-responsive element-binding protein pathway during hippocampal synaptic plasticity. J Neurosci. 2005 Jul 20;25(29):6887-97. PubMed.

    . Phosphodiesterase 5 inhibition improves synaptic function, memory, and amyloid-beta load in an Alzheimer's disease mouse model. J Neurosci. 2009 Jun 24;29(25):8075-86. PubMed.

  4. This study on sildenafil by Puzzo, Arancio, and others provides intriguing insight into the role of guanyl cyclase (GC) and cGMP in Aβ-mediated pathology. Because nitric oxide (NO) is the primary activator of GC-coupled NO receptors and plays a critical role in hippocampal synaptic processing, the data also point to the importance of NO in pathophysiology associated with Alzheimer disease.

    The prolonged biological effect of sildenafil treatment is particularly interesting and difficult to explain. One potential explanation for these findings is a changed profile in cGMP levels. As shown by John Garthwaite’s group at the Wolfson Institute for Biomedical Research in London, an NO-mediated increase in cGMP in neurons has a unique time course that depends on the level of NO, the presence and activity of GC-coupled NO receptors, GC receptor desensitization, and the level of phosphodiesterases (Mo et al., 2004). Receptor desensitization provides an interesting (and potentially modifiable) means to change the pattern of cGMP accumulation within a cell from a fast, but low-level phasic change (seen when PDE levels are high) to a large, sustained plateau in cGMP levels when PDE activity is low. Mixtures of these two profiles are also possible. In many neurons, the production of cGMP activated by NO is commonly associated with a phasic change in cGMP levels due to the combined presence of GC receptor desensitization and the activity of PDE5 that alters GC activity (Mo et al., 2004). In contrast, astrocytes that have low PDE levels demonstrate a rapid increase in cGMP levels to a maintained plateau level (Bellamy et al., 2000). The presence of an inhibitor of PDE5, as shown here with sildenafil, potentially changes the profile of NO-mediated GC activity in neurons from a transient to a tonic type response. It follows, then, that downstream phosphorylation events may also be prolonged or initiated depending on the levels and types of targets. Whether these types of changes that are clearly seen over seconds to minutes can produce the type of day-to-week changes seen by Puzzo and collaborators is not known. They could result, however, in altered gene transcription that promotes long-term growth/survival mechanisms.

    What is particularly exciting to our research group, however, is the relationship of the observed changes in PDE5 to Aβ. Our data support the concept that decreased NO favors Aβ accumulation and disease progression. The mechanism of NO’s effects may be cGMP-dependent or independent and may be regulated at the transcriptional or post-translational level. Lisa Ridnour from David Wink’s lab at NIH has shown that production and release of tissue inhibitor of metalloprotease-1 (TIMP-1) is suppressed by NO/cGMP (Ridnour et al., 2008). In turn, the lower levels of TIMP-1 promote increased activity of MMP9 (Ridnour et al., 2007). Several authors have previously published that MMPs degrade Aβ proteins (Qiao et al., 1997; Yan et al., 2006) and, consequently, increased MMP activity could lower Aβ levels. When NO is reduced as in our genetic mouse models, the failure to elicit appropriate changes in TIMP-1 levels and activity may facilitate the observed accumulation of Aβ in our APPSw/NOS2-/- mice (Colton et al., 2006). Thus, this type of NO/sGC/cGMP-dependent mechanism may be coupled to the changes observed in Aβ levels, and sildenafil, by increasing the sGC/cGMP profile, would act like increased NO to restrict Aβ accumulation.


    . Rapid desensitization of the nitric oxide receptor, soluble guanylyl cyclase, underlies diversity of cellular cGMP responses. Proc Natl Acad Sci U S A. 2000 Mar 14;97(6):2928-33. PubMed.

    . Degradation of amyloid beta-protein by a metalloprotease secreted by microglia and other neural and non-neural cells. J Biol Chem. 1997 Mar 7;272(10):6641-6. PubMed.

    . Matrix metalloproteinase-9 degrades amyloid-beta fibrils in vitro and compact plaques in situ. J Biol Chem. 2006 Aug 25;281(34):24566-74. PubMed.

    . Molecular mechanisms for discrete nitric oxide levels in cancer. Nitric Oxide. 2008 Sep;19(2):73-6. PubMed.

    . Nitric oxide regulates matrix metalloproteinase-9 activity by guanylyl-cyclase-dependent and -independent pathways. Proc Natl Acad Sci U S A. 2007 Oct 23;104(43):16898-903. PubMed.

    . Kinetics of a cellular nitric oxide/cGMP/phosphodiesterase-5 pathway. J Biol Chem. 2004 Jun 18;279(25):26149-58. PubMed.

  5. Nitroglycerin is said to easily cross the blood-brain barrier and enhance cGMP expression (1). Perhaps you'd expect to see a lower incidence of AD in those using the Nitrodur patch.

    Of further relevance to AD is the study by Brink and colleagues which reports that sildenafil increases muscarinic acetylcholine receptor signaling in human neuroblastoma cells (2). Is the induction of NOS downstream of mAChR? I refer to my previous post in which I ask whether nitroglycerin (sp) has been used as a NO donor in the treatment of AD (3).

    See also:

    Comment by Mary Reid


    . Nitroglycerin enhances cGMP expression in specific neuronal and cerebrovascular structures of the rat brain. J Chem Neuroanat. 2004 Mar;27(1):23-32. PubMed.

    . Appearance of antidepressant-like effect by sildenafil in rats after central muscarinic receptor blockade: evidence from behavioural and neuro-receptor studies. J Neural Transm. 2008;115(1):117-25. PubMed.

    View all comments by Mary Reid

Make a Comment

To make a comment you must login or register.

This paper appears in the following:


  1. NO Joke: Viagra Lessens Aβ, Cognitive Problems in AD Mice