Scientists are casting a wide net in their search for new therapeutic approaches to Alzheimer’s disease. One such expedition has now snagged sildenafil, better known as Viagra, a drug widely used to treat erectile dysfunction. In the December 6 Nature Aging, researchers led by Feixiong Cheng at the Cleveland Clinic in Ohio describe how their network-based analysis of AD genes fished through 1,608 approved drugs for possible connections to amyloid and tau pathology. A top hit was sildenafil, a phosphodiesterase 5 inhibitor that crosses the blood-brain barrier.

  • Molecular repurposing study links Viagra and AD pathology.
  • Viagra inhibits tau kinases and lowers p-tau in cultured human neurons.
  • Men who take it have a two-thirds lower risk of AD; a clinical trial is planned.

Previous data from rodents showed sildenafil could lessen tauopathy and improve memory, and the authors now report that it lowered p-tau in cultured human neurons. Analyzing medical records from several million people, they found sildenafil use correlated with a 69 percent lower risk of AD.

Cheng believes these data justify testing sildenafil in a clinical trial, and is planning a Phase 2 study in patients with early AD.

Fishing Strategy. Researchers looked for existing drugs (capsule) with the shortest paths (purple) between their targets (blue) and AD genes (red). [Courtesy of Fang et al., Nature.]

Others agreed. “In my opinion, with these findings it is mandatory to start a clinical trial,” Ana García-Osta at the University of Navarra in Pamplona, Spain, wrote to Alzforum. Meanwhile, David Knopman at the Mayo Clinic in Rochester, Minnesota, praised the methodology. “I was very impressed by the creative and focused approach to a network-based analysis to identify repurposed drugs for further study in dementia,” he told Alzforum.

Endophenotypes for Drug Discovery
In the last decade, network approaches have become popular, as funding agencies have encouraged systematic surveys of the molecular changes that occur during neurodegeneration as a way to identify new drug candidates (May 2013 webinar; Feb 2015 news). Specifically, Cheng and colleagues are interested in matching network analyses of disease genes with those of drug target effects as a way of identifying drugs that can be repurposed for new uses (Cheng et al., 2018; Cheng et al., 2019). 

To tackle AD, joint first authors Jiansong Fang and Yadi Zhou at the Cleveland Clinic, Pengyue Zhang at Indiana University in Indianapolis, and Chien-Wei Chiang at Ohio State University in Columbus compiled a list of 144 genes linked to amyloid or tau pathology. They combined these genetic data with transcriptomic and proteomic data from amyloidosis and tauopathy mouse models to build AD “endophenotype disease modules.” In brief, these are protein interaction networks associated with amyloid and tau pathology. The authors constructed three modules based primarily on genetic data, and 10 based on proteomic data.

Next, the authors searched for interactions between proteins in these endophenotype modules and the known targets of the 1,608 drugs. Interactions were typically indirect. The authors used existing protein-protein interactome data to predict which drug targets would have downstream effects on AD-related proteins in the modules. They prioritized drugs with the shortest “paths” (i.e., fewest intermediaries) between a given drug’s target and the module proteins (see image above). This generated a list of 66 drug candidates.

Prioritizing Candidates. In picking drugs for detailed study, researchers considered factors such as brain penetrance, preclinical data, whether the drug had been tested before for AD, and how connected it is to AD protein networks (z score). [Courtesy of Fang et al., Nature.]

Of the 66 candidates, 11 had been in AD trials, and for another 10, evidence from animal models suggests they could be useful for AD. Among the former were the ALS drug riluzole, the diabetes drug pioglitazone, the antibiotic minocycline, and the heart drug gemfibrozil (see table above). Some of these have posted negative results for AD, while others remain under investigation (Nov 2018 conference news; Nov 2019 news; Dec 2019 conference news). Donepezil also appeared on the list. These the authors set aside, and focused on the remaining candidates.

Indirect Connections. Sildenafil target proteins (blue) affect some AD genes (red), such as APP and the tau kinases GSK3β and CDK5. [Courtesy of Fang et al., Nature.]

Among those, sildenafil drew the researchers’ attention. Its known downstream targets connected to the AD network modules, it enters the brain, and preclinical studies suggested potential for reducing AD pathology. Osta had previously shown that sildenafil lowered p-tau in Tg2576 mice, while others reported it suppressed plaques in APP/PS1 mice and improved memory in a rat model of vascular dementia (Cuadrado-Tejedor et al., 2011; Zhang et al., 2013; Venkat et al., 2019). Unlike some of the other candidates, sildenafil is widely used in people, at least of the male variety, so the scientists had epidemiological data to mine.

Myriad Mechanisms?
The authors confirmed sildenafil’s effect on p-tau in forebrain neurons coaxed from iPS cells taken from AD patients. Six days of sildenafil suppressed p-tau181 by 40 percent. It also stimulated neurite outgrowth.

How would a drug for erectile dysfunction affect tau? The authors noted sildenafil inhibits two kinases responsible for tau phosphorylation, GSK3β and CDK5 (see image above). These kinases are also linked to inflammation in microglia and macrophages (Yuskaitis and Jope, 2009; Na et al., 2015). The authors tested sildenafil on a human microglial cell line stimulated with the proinflammatory molecule lipopolysaccaride, and found the phosphodiesterase inhibitor prevented activation of the two kinases. The results hint the drug could counteract neuroinflammation as well as tau pathology.

Might sildenafil reduce a person’s risk for AD by improving cerebrovascular blood flow? Commenters were divided on this point. “Increased blood flow and glucose metabolism may contribute,” Osta speculated. However, Jos Prickaerts at the University of Maastricht in the Netherlands noted that at the doses typically used for erectile dysfunction, the effects on brain circulation are minor.

Other mechanisms may be in play. Osta noted that PDE5 inhibition boosts levels of cyclic GMP. This second messenger plays numerous roles within cells, among them activating CREB, which then turns on expression of several genes involved in memory formation. Low cGMP in the cerebrospinal fluid of AD patients correlates with cognitive decline, Osta found (Ugarte et al., 2015). 

Along the same lines, Ottavio Arancio at Columbia University Medical Center, New York, previously reported that both tau and Aβ oligomers dampen production of nitric oxide, which is needed to generate cGMP (Jun 2009 news; Puzzo et al., 2005; Acquarone et al., 2019). Thus, sildenafil may help counteract these oligomers' toxicity. “[These new] human data are consistent with the large amount of preclinical data published by my lab and others in the last 20 years,” Arancio wrote to Alzforum (full comment below).

What Does It Mean?
Do these effects in human cells translate into a clinical benefit? To get at this question, the authors analyzed U.S. medical records, comparing 116,412 people prescribed sildenafil, nearly all men, with an average age of 71, with 460,356 age- and sex-matched controls. It is unclear how often men with a prescription take the drug, and most men only take it prior to sexual intercourse, not regularly as is the case with most prescription drugs. Nonetheless, men prescribed sildenafil had about one-third the risk of AD, with an incidence of 0.08 percent versus 0.25 in controls.

Was this association confounded by other factors? The researchers attempted to address this by matching controls according to co-morbidities such as hypertension and diabetes, and adjusting for age and mild cognitive impairment. They found sildenafil’s benefit was independent of all these factors. Four other drugs used for hypertension or diabetes, namely diltiazem, losartan, metformin, and glimepiride, showed no association with AD incidence. The researchers were unable to assess how sildenafil affects women, because few take the drug off-label for pulmonary hypertension, amounting to only 2 percent of the cohort.

Despite the effort to weed out confounds, Knopman noted that sildenafil use may select for those men who remain the most active and vigorous with age. He cautioned against using epidemiological studies alone to determine drug candidates, because such data cannot determine causation. “Pharmacoepidemiology has repeatedly failed the field of Alzheimer disease therapeutics,” Knopman said. Prickaerts suggested that sildenafil might improve mood in men who take it, which could be another confounding factor. “Erectile dysfunction is highly correlated with depression, which itself is a risk factor for AD,” he wrote.

Putting It to the Test
Researchers agree the next step is to test sildenafil in a trial. Cheng said his study will enroll people at early stages of AD who have fluid biomarker evidence of p-tau and Aβ42/Aβ40 changes. The researchers have not yet nailed down where the cutoffs will be and whether participants with symptoms will be excluded. They plan to test a dose range of 25-75 mg daily, using an adaptive trial design that will allow them to adjust parameters as they go. For erectile dysfunction, men take 50 mg if they are under 60, and 25 mg if they are older.

Commentators said other PDE5 inhibitors on the market—such as tadalafil (Cialis) and vardenafil (Levitra)—might be worth testing as well. Tadalafil improved memory in the J20 mouse model of amyloidosis (García-Barroso et al., 2013). Arancio noted that sildenafil has off-target effects on other phosphodiesterases besides PDE5, and suggested testing more specific compounds (Fiorito et al., 2013; Fiorito et al., 2017; Zuccarello et al., 2020). 

Cheng believes the methodology used in this study can help uncover other potential treatments. He plans to add additional endophenotypes such as inflammation and metabolic dysfunction into his AD modules to identify drugs that might be used for combination therapy. He will also use the same network approach to search for drugs to treat Parkinson’s, amyotrophic lateral sclerosis, and other neurodegenerative conditions.

Similar efforts are underway elsewhere. Madhav Thambisetty at the National Institute on Aging in Bethesda, Maryland, leads the Drug Repurposing for Effective Alzheimer’s Medicines (DREAM) study, which likewise uses omics data to identify druggable pathways (Desai et al., 2020). Coincidentally, Thambisetty told Alzforum he is preparing a manuscript describing his own study’s findings on sildenafil, though he would not elaborate on what he found.—Madolyn Bowman Rogers

Comments

  1. I read this manuscript with a lot of interest, as it shows human data that are consistent with the large amount of preclinical data published by my lab and others in the last 20 years, supporting the use of phosphodiesterase 5 inhibitors against Alzheimer’s disease. I still remember when I presented the first evidence at a meeting on phosphodiesterases. The major critique was that there is no phosphodiesterase 5 in the human brain. To address this issue, we published a manuscript (Teich et al., 2016) in which we demonstrated that phosphodiesterase 5 exists in human neurons and is a viable therapeutic target. I have dedicated more than 20 years to these studies; to see them supported by evidence in humans is very rewarding.

    Regarding the mechanisms of the protective effect, the manuscript does not discuss the issue in depth, probably due to the fact that there is a large amount of work published on the topic, and it would have been hard to synthesize it in the discussion. We showed that phosphodiesterase 5 inhibitors counteract the reduction in activity of the NO cascade due to elevation of tau oligomers, re-establishing normal phosphorylation of the memory related molecule, CREB (Acquarone et al., 2019). The same cascade is also affected after elevation of Aβ oligomers (Puzzo et al., 2005, 2009). 

    Thus, the most likely scenario is that extracellular tau and Aβ oligomers downregulate the NO cascade, possibly by interacting with a transmembrane protein. Our work showed that APP is necessary for the damage of synaptic plasticity and memory by Aβ and tau oligomers (Puzzo et al., 2017). 

    As the authors point out, due to the nature of the use of sildenafil (mostly in a male population, without a specific regimen), the study has some limitations that will be addressed by a randomized controlled trial, possibly using inhibitors of phosphodiesterase 5 that can be administered for prolonged periods of time, as it is required in a chronic condition such as Alzheimer’s. Indeed, a prolonged use of sildenafil might incur side effects due to interaction of the drug with phosphodiesterase 6, 1, and perhaps 9, 10, and 11. A more specific phosphodiesterase 5 inhibitor could serve the purpose (see, for instance, Fiorito et al., 2013, 2017; Zuccarello et al., 2020). 

    References:

    . PDE5 Exists in Human Neurons and is a Viable Therapeutic Target for Neurologic Disease. J Alzheimers Dis. 2016;52(1):295-302. PubMed.

    . Synaptic and memory dysfunction induced by tau oligomers is rescued by up-regulation of the nitric oxide cascade. Mol Neurodegener. 2019 Jun 27;14(1):26. PubMed. Correction.

    . 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.

    . LTP and memory impairment caused by extracellular Aβ and Tau oligomers is APP-dependent. Elife. 2017 Jul 11;6 PubMed.

    . Identification of a Novel 1,2,3,4-Tetrahydrobenzo[b][1,6]naphthyridine Analogue as a Potent Phosphodiesterase 5 Inhibitor with Improved Aqueous Solubility for the Treatment of Alzheimer's Disease. J Med Chem. 2017 Nov 9;60(21):8858-8875. Epub 2017 Oct 23 PubMed.

    . Development of novel phosphodiesterase 5 inhibitors for the therapy of Alzheimer's disease. Biochem Pharmacol. 2020 Jun;176:113818. Epub 2020 Jan 21 PubMed.

Make a Comment

To make a comment you must login or register.

References

Webinar Citations

  1. Can Network Analysis Identify Pathological Pathways in Alzheimer’s

News Citations

  1. Alzheimer's Disease Research Summit 2015: Expanding the Horizon
  2. Trials of Diabetes-Related Therapies: Mainly a Bust
  3. Minocycline Does Not Work in Mild Alzheimer’s Disease
  4. At CTAD, Early Failures and Hints of Success, from Small Trials
  5. NO Joke: Viagra Lessens Aβ, Cognitive Problems in AD Mice

Research Models Citations

  1. Tg2576
  2. APPPS1
  3. J20 (PDGF-APPSw,Ind)

Paper Citations

  1. . Network-based approach to prediction and population-based validation of in silico drug repurposing. Nat Commun. 2018 Jul 12;9(1):2691. PubMed.
  2. . A genome-wide positioning systems network algorithm for in silico drug repurposing. Nat Commun. 2019 Aug 2;10(1):3476. PubMed.
  3. . Sildenafil restores cognitive function without affecting β-amyloid burden in a mouse model of Alzheimer's disease. Br J Pharmacol. 2011 Dec;164(8):2029-41. PubMed.
  4. . Phosphodiesterase-5 inhibitor sildenafil prevents neuroinflammation, lowers beta-amyloid levels and improves cognitive performance in APP/PS1 transgenic mice. Behav Brain Res. 2013 Aug 1;250:230-7. PubMed.
  5. . Sildenafil treatment of vascular dementia in aged rats. Neurochem Int. 2019 Jul;127:103-112. Epub 2018 Dec 25 PubMed.
  6. . Glycogen synthase kinase-3 regulates microglial migration, inflammation, and inflammation-induced neurotoxicity. Cell Signal. 2009 Feb;21(2):264-73. Epub 2008 Oct 29 PubMed.
  7. . The early synthesis of p35 and activation of CDK5 in LPS-stimulated macrophages suppresses interleukin-10 production. Sci Signal. 2015 Nov 24;8(404):ra121. PubMed.
  8. . Decreased levels of guanosine 3', 5'-monophosphate (cGMP) in cerebrospinal fluid (CSF) are associated with cognitive decline and amyloid pathology in Alzheimer's disease. Neuropathol Appl Neurobiol. 2015 Jun;41(4):471-82. PubMed.
  9. . 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.
  10. . Synaptic and memory dysfunction induced by tau oligomers is rescued by up-regulation of the nitric oxide cascade. Mol Neurodegener. 2019 Jun 27;14(1):26. PubMed. Correction.
  11. . Tadalafil crosses the blood-brain barrier and reverses cognitive dysfunction in a mouse model of AD. Neuropharmacology. 2013 Jan;64:114-23. Epub 2012 Jul 7 PubMed.
  12. . Synthesis of quinoline derivatives: discovery of a potent and selective phosphodiesterase 5 inhibitor for the treatment of Alzheimer's disease. Eur J Med Chem. 2013 Feb;60:285-94. PubMed.
  13. . Identification of a Novel 1,2,3,4-Tetrahydrobenzo[b][1,6]naphthyridine Analogue as a Potent Phosphodiesterase 5 Inhibitor with Improved Aqueous Solubility for the Treatment of Alzheimer's Disease. J Med Chem. 2017 Nov 9;60(21):8858-8875. Epub 2017 Oct 23 PubMed.
  14. . Development of novel phosphodiesterase 5 inhibitors for the therapy of Alzheimer's disease. Biochem Pharmacol. 2020 Jun;176:113818. Epub 2020 Jan 21 PubMed.
  15. . Targeting abnormal metabolism in Alzheimer's disease: The Drug Repurposing for Effective Alzheimer's Medicines (DREAM) study. Alzheimers Dement (N Y). 2020;6(1):e12095. Epub 2020 Nov 26 PubMed.

Further Reading

Primary Papers

  1. . Endophenotype-based in silico network medicine discovery combined with insurance record data mining identifies sildenafil as a candidate drug for Alzheimer’s disease. Nat Aging. 1, 2021, pp. 1175–88. Nat Aging.