Mitochondria

Mitochondria from AD patients differ from those of non-AD individuals morphologically, functionally, and in terms of gene expression. ^{Reference Ridge and Kauwe31} Extensive data suggest that mitochondria instigate and/or mediate diverse AD pathologies. Debate continues over the origin of these AD mitochondrial changes. ^{Reference Swerdlow32} Some data suggest mitochondrial dysfunction temporally occurs upstream from proteopathies, indicating a primary mitochondrial pathology may supersede amyloid and tau pathologies; other data suggest that mitochondrial dysfunction is a consequence of amyloid proteopathy but nonetheless contributes to ongoing neural damage once triggered. In terms of specific molecular targets, recent work has implicated translocator protein (TSPO), an outer membrane mitochondrial protein that locates cytosol cholesterol to mitochondrial membranes. ^{Reference J.Jung33} TPSO is present in the glial cells that respond to neuroinflammation and regulate the opening of the mitochondrial permeability transition pores controlling entry of molecules necessary for mitochondrial function. TSPO is a pro-apoptotic protein which functions via its interaction with the voltage-dependent anion channel 1 (VDAC1) protein, the most abundant outer membrane mitochondrial protein; VDAC1 is the organelle’s gatekeeper for the passage of ions, nucleotides, and metabolites, playing a central role in apoptosis regulation through its interaction with apoptotic and anti-apoptotic proteins of the Bcl-2 (B-cell CLL/lymphoma 2) protein family. ^{Reference Shoshan-Barmatz, Krelin and Shteinfer-Kuzmine34} Notable, TSPO is upregulated in the brain of AD patients and is generally thought to be conclusively linked to AD.

In viral infections such as COVID-19, mitochondria are well recognized as pivotal organelles in controlling signaling pathways essential to the host response in restraining viral infections. ^{Reference Yuan, Shan and Yao35} Specifically, a major role in antiviral defense is played by mitochondrial antiviral signaling (MAVS) protein, an adaptor protein that coordinates the activation of interferon-inducing pathways and autophagy at the mitochondrial level by involving activation of NF-κB and (IRF3 leading to the production of type I interferons (IFN-α /β) and additional pro-inflammatory cytokines. ^{Reference Seth, Sun, Ea and Chen36,Reference Wang, Yuan and Jia37} Co-localized with Bcl-2 proteins and essential for antiviral innate immunity, MAVS is a 540 amino acid outer membrane mitochondrial protein that consists of three components, an N-terminal caspase activation recruitment domain, a proline-rich domain, and a transmembrane C terminal domain, which induce apoptosis in virally infected host cells by interacting with the caspase 8 protease. ^{Reference Refolo, Vescovo, Piacentini, Fimia and Ciccosanti38} Recently, it was demonstrated that mitochondrial-located MAVS protein mediates its pro-apoptotic activity by associating with VDAC1 and modulates VDAC1 protein stability via the ubiquitin-proteasome pathway.

The significance of mitochondrial changes in AD and COVID-19 is an evolving area of interest, and mitochondrial dysfunction represents a reasonable therapeutic target in the realm of AD/COVID-19 overlap, with particular focus on outer mitochondrial membrane proteins such as VDAC1.

Endoplasmic Reticulum

The ER is increasingly recognized as a molecular contributor to the pathogenesis of AD. ^{Reference Li, Yu, Jiang and Tan39} ER stress has been observed in postmortem AD brains, and ER stress is known to arise with the accumulation of misfolded or unfolded proteins, such as Aβ and tau. These mechanisms may be mediated by inositol-requiring enzyme 1 (IRE1), protein kinase R-like endoplasmic reticulum kinase (PERK), and activating transcription factor 6 (ATF6) stress sensors. ^{Reference Hedskog, Pinho and Filadi40} In addition to these stress responses, cross-talk between the ER and adjacent mitochondria may also be impacted in AD. Sub-compartments of ER are in physical and biochemical contact with mitochondria via raft-like lipid regions referred to as mitochondria-associated membranes (MAMs), which play important roles in lipid synthesis, calcium homeostasis, and apoptotic signaling. ^{Reference Area-Gomez and Schon41} Within the ER-mitochondria, MAMs bridging complex, inositol-1,4,5-triphosphate receptors facilitate biochemical cross-talk. Upregulated MAM-associated proteins are found in AD brains as well as (APP) Swe/Lon mouse models and can be detected before the appearance of plaques. ^{Reference Schreiner, Hedskog, Wiehager and Ankarcrona42}

In COVID-19, vesicle trafficking within the ER of host cells is important in coronavirus replication; ^{Reference Zhao, Guo and Comunale43} the replicase-transcriptase machinery and other viral structural proteins assemble within the host ER, making it an essential cellular component for viral genome replication and capsid assembly in the formation of new virus particles, which germinate in the ER-Golgi intermediate compartment prior to fusion with the plasma membrane and viral release. ^{Reference Sawicki and Sawicki44,Reference Perrier, Bonnin and Desmarets45} This process induces ER stress analogous to that occurring in association with AD pathology, involving both MAVS and MAM proteins, and mimicking many pathognomonic elements of an unfolded protein response. ^{Reference Sureda, Alizadeh and Nabavi46}

In both AD and COVID-19, ER function and stress are intimately linked to autophagy, a catabolic process involving the engulfment of cellular material by a double-membrane structure, the phagophore, which subsequently closes into an autophagosome vesicle sequestering cargo and debris which is degraded following fusion with a lysosome. Autophagy is a bulk process that unselectively degrades cellular material as required for cellular upkeep. However, autophagy can also selectively target distinct organelles requiring turnover. The specific elimination of the ER via a selective form of autophagy is now recognized as a unique biochemical process termed ER-phagy. ^{Reference Wilkinson47} ER-phagy is specifically involved in both AD and COVID-19 and constitutes another potentially druggable area of biochemical overlap between these two disorders.

Curcumin

Derived from the rhizomatous, ginger-like plant Curcuma longa, curcumin is a natural polyphenol with well-evidenced anti-inflammatory and anti-microbial activities. Its anti-inflammatory properties are thought to derive from a series of synergistic mechanisms, ranging from free-radical scavenging, the modulation of antioxidant enzymes as well as the inhibition of free-radical generating systems including cyclooxygenases. ^{Reference Hewlings and Kalman48,Reference Menon and Sudheer49} It may also downregulate activation of NF-κB signaling cascades and thus mediate a broader, systemic anti-inflammatory effect. ^{Reference SSingh and Aggarwal50,Reference van't Land, Blijlevens and Marteijn51} In AD models, curcumin was also shown to be a potent inhibitor and destabilizer of neurotoxic amyloid species, even reducing the senile plaque burden in APPswe/PS1dE9 mouse models. ^{Reference He, Wang and Wei52,Reference Mishra and Palanivelu53} It may also be effective in the chelation of metal species and reducing cholesterol esters – both potent risk factors in the aggravation of AD protein and immune pathologies. ^{Reference Mishra and Palanivelu53}

These emerging associations, aided by a relative ease of access and the less stringent regulations surrounding natural supplements, have driven a sensationalized, and largely unevidenced, belief of a potential therapeutic benefit in AD. Yet, no trial has demonstrated any significant effect on either disease onset or prognosis with curcumin or its derivatives. This may be partly attributable to the poor bioavailability of the bulky curcumin molecule in the aqueous extracellular environment. These same characteristics also compromise its ability to cross the blood–brain barrier and thus limit its efficacy as a neurological agent.

In the treatment of COVID-19, however, formulations of curcumin have shown some promise. When packaged in nanomicelles, Saber-Moghaddam et al. report 160 mg of curcuminoids daily was able to resolve COVID symptoms, including fever, chills, tachypnea, and myalgia significantly more expediently, and that hospitalization, supplemental oxygenation parameters as well as overall disease resolution were meaningfully improved. ^{Reference Saber-Moghaddam, Salari and Hejazi54} Multiple trials on various forms of curcumin remain ongoing. Though curcumin has yet to effective in the treatment of AD, the extensive data on its anti-inflammatory roles and its optimal formulation, both for efficacy and bioavailability, may be of relevance in the trials of COVID-19.

Pioglitazone

Pioglitazone is a thiazolidinedione, originally developed as an antihyperglycemic adjunct in the 1980s and 90s. ^{Reference Gillies and Dunn55} It targets peroxisome proliferator-activated receptors to reduce the effects of insulin resistance in tissues. In the brain, modulating these insulin signaling pathways was associated with a reduction of multiple inflammatory, cell death, and proliferation associated pathways, including IR, IRS-1, AKB, p-CREB, and Bcl-2. ^{Reference Ma, Shao and Wang56–Reference Thal, Heinemann, Luh, Pieter, Werner and Engelhard58} Crucially, in animal models, pioglitazone also mediated amyloid plaque clearance, reduced tau hyperphosphorylation, and aided synaptic plasticity. ^{Reference Hamano, Shirafuji and Makino59–Reference Quan, Qian, Li and Li61} These outcomes, along with robust safety and pharmacologic profiles, motivated a pair of large phase III trials; both, however, were terminated for lack of efficacy. It remains unclear why the trails failed, though in accord with the complexity and chronicity of all dementia trials, it is conceivable that multiple study and clinical parameters, especially the many years of indolent disease progression preceding the development of symptoms, have been contributory to trial failures.

In spite of these failures, pioglitazone is a well-established inflammatory mediator. Xie et al. demonstrated that even amongst individuals without hyperglycemia, pioglitazone significantly reduced IL-6 and TNF-α. ^{Reference Xie, Sinha and Yi62} With extended use, astrocyte, lymphocyte, and other inflammatory cytokine production modalities were attenuated. ^{Reference Quan, Qian, Li and Li61} Especially relevant in the management of SARS-CoV-2, pioglitazone may also act on pulmonary inflammation and fibrosis, with multiple studies observing diminished inflammatory markers in animal lung and lavage samples. ^{Reference Carboni and Carta63}

A major barrier in the treatment of COVID-19 is the management of acute symptomology. Typically, low even subclinical dosing of pioglitazone was administered for weeks to observe a gradual clinical effect. As it was intended for the treatment of chronic and progressive diseases, this is expected. However, the acute management of COVID-19 will inevitably require breakthrough dosing and emergent management. The secondary outcome profile will also be critical; if high-dose administration leads to uncontrolled hypoglycemia or hypersensitivity to insulin, these may complicate the management of infection.

Inositol

Inositol is a carbocyclic sugar that mediates cell signal transduction in response to a variety of chemical messengers, hormones, and growth factors. Structurally, inositol is a hexa-substituted alcohol of cyclohexane; epimerization of the six hydroxyl groups generates nine stereoisomers (see Figure 4). Myo-inositol (MI) is the most prominent stereoisomer and plays a central role as the structural platform for a number of inositol phosphate secondary messengers. It also serves as an important structural component of membrane phospholipids, such as phosphatidylinositol.

Figure 4: Molecular structures of inositol isomers. Inositol (A) is a collection of nine different stereoisomers of a hexa-substituted cyclohexane polyol. The most common isomer is myo-inositol (B), which is cis-1,2,3,5-trans-4,6-cyclohexanehexol. Scyllo-inositol (C) has undergone trials as an amyloid anti-aggregant in AD.

In AD, a ¹H magnetic resonance spectroscopy study demonstrated elevated brain MI levels in the pre-dementia in adults with Down’s syndrome, suggesting a role for MI as an AD diagnostic. ^{Reference Huang, Alexander and Daly64} SI, another stereoisomer that is relatively rare in nature, has also been considered as a therapeutic for AD. SI has been reported to stabilize the nontoxic oligomers of Aβ and to inhibit their toxic aggregation, by coating the surface of Aβ protofibrils and disrupting their stacking into fibrillar aggregates; an analog series of SI derivatives was synthesized and evaluated revealing that all six inositol hydroxyl groups were involved in fibrillar aggregation inhibition. ^{Reference Sun, Zhang, Hawkes, Shaw, McLaurin and Nitz65} In the late 1990s, McLaurin and coworkers pursued a clinical trial of SI in AD which failed to show improvement. ^{Reference McLaurin, Golomb, Jurewicz, Antel and Fraser66}

In COVID-19, Bizzarri et al. have postulated that MI could be used to ameliorate the toxic pulmonary inflammatory response. ^{Reference Bizzarri, Laganà, Aragona and Unfer67} This suggestion is based on the observation that MI has been successfully used to treat newborn respiratory distress syndrome, achieving this goal by downregulating the inflammatory response via reduction of IL-6 levels known to mediate the inflammatory cascade. Since MI is essentially devoid of major side effects, they have speculated regarding its utility in the treatment of critically ill COVID-19 patients.

Glycosaminoglycan Mimics

Glycosaminoglycans (GAGs) are long linear mucopolysaccharides consisting of repeating disaccharide units; the repeating unit typically consists of an amino sugar, along with a uronic sugar or galactose. GAGs are essential molecules, covalently connecting to proteins to form proteoglycans. GAGs are highly negatively charged polymers that can also sequester physiologically important proteins and strongly bind water and ions. Heparan sulfate is a prototypic GAG with a polyanionic structure arising from multiple geometrically positioned sulfate groups.

In AD, GAGs (specifically with sulfate moieties) are important molecular co-conspirators facilitating protein misfolding and oligomerization. They facilitate interactions between monomeric or oligomeric Aβ and neuronal/glial cell surfaces possibly involving the serpin-enzyme complex receptor, the alpha7nicotinic acetylcholine receptor (alpha7nAChR), the receptor for advanced glycosylation end-products, and formyl peptide receptor-like 1. ^{Reference Ariga, Miyatake and Yu68} Our group further observed that at an atomistic level, Aβ may interact directly with GAGs via its HHQK domain to mediate portion of neuronal membranes. ^{Reference Stephenson, Heyding and Weaver69}

Since heparin is an already available GAG mimetic, various groups have demonstrated its capacity to bind to Aβ and prevent subsequent oligomerization. However, heparin is a potent anticoagulant and inappropriate for chronic use in AD’s elderly cohorts. Accordingly, in the 1990s, we synthesized numerous polysulfonated small molecule GAG mimetics, ultimately pursuing clinical trials with tramiprosate (3-amino-1-propanesulfonic acid) for AD and eprodisate (1,3-propanedisulfonic) for renal failure in systemic amyloidosis – both failed to show efficacy in Phase III human trials (see Figure 5). ^{Reference Kisilevsky, Szarek and Weaver70}

Figure 5: Molecular structures of heparan sulfate, heparin sulfate, eprodisate. Heparan sulfate (A) (HS) is a linear polysaccharide that occurs as a proteoglycan (HSPG) in which two or three HS chains are attached in close proximity to cell surface or extracellular matrix proteins. Heparin (B) is a smaller glycosaminoglycan polysaccharide polymer, structurally related to heparan, and consists of a variably sulfated repeating disaccharide unit; the most common disaccharide unit is composed of a 2-O-sulfated iduronic acid and 6-O-sulfated, N-sulfated glucosamine, IdoA(2S)-GlcNS(6S). Eprodisate (C) (1,3-propanedisulfonate) is a negatively charged, sulfonated small molecule that has structural similarities to heparin and heparan sulfate; it is a glycosaminoglycan mimetic with sulfate group positioned geometrically to mimic those in heparin and heparan.

For COVID-19, Mycroft-West et al. have suggested repurposing heparin as a GAG mimetic for uses as a coronavirus antiviral. ^{Reference Mycroft-West, Su and Pagani71} They put forth this suggestion after using surface plasmon resonance and circular dichroism to measure the interaction between the SARS-CoV-2 Spike S1 protein receptor-binding domain (SARS-CoV-2 S1 RBD) and heparin. ^{Reference Song, Gui, Wang and Xiang72} Coronavirus contains four structural proteins, including spike (S), envelope (E), membrane (M), and nucleocapsid (N) proteins, with the S protein mediating viral entry into host cells by binding to the host receptor through the RBD in the S1 subunit and then fuzing the viral and host membranes through the S2 subunit. ^{Reference Letko, Marzi and Munster73} SARS-CoV recognizes the ACE2 enzyme as its receptor. However, full pathological expression of coronavirus attachment and entry requires not only ACE2 but also viral binding to host cell heparan sulfate GAG adjacent to and part of the expanded ACE2 receptor complex. Mycroft-West et al. exploited structural similarities between heparan sulfate and heparin, using heparin as a small molecule GAG mimetic to bind to the virus, thereby blocking its capacity to bind to host cells. Based upon this molecular interaction they postulated the rapid implementation of a first-line therapeutic by repurposing heparin whilst tailor-made, GAG-mimetic antivirals are being developed. Heparin is a well-known agent, but its use is complicated by its significant anticoagulant activity, especially in individuals critically ill with COVID-19. Accordingly, there may be a place for failed GAG-mimetic agents, such as eprodisate, as safer substitutes for heparin.

Furosemide

Furosemide is a Food and Drug Administration approved loop diuretic used in the treatment of hypertension and associated edema in cardiac, renal, and hepatic failures. ^{Reference Boles Ponto and Schoenwald74} Through inhibition of the Na(+)-K(+)-2Cl(-) cotransporter (NKCC2), ^{Reference Carone, Oxberry, Twycross, Charlesworth, Mihalyo and Wilcock75} furosemide works by blocking sodium and chloride tubular reabsorption in the proximal and distal tubules, and the thick ascending loop of Henle, resulting in decreased extracellular accumulation of fluid in cardiac and renal pathologies. ^{Reference Huang, Dorhout Mees, Vos, Hamza and Braam76}

Previous reports have indicated that antihypertensive use is associated with a reduced risk of AD and similar dementing disorders. ^{Reference Khachaturian, Zandi and Lyketsos77} In in vivo models of AD, furosemide was found to enhance kidney-mediated clearance of Aβ, rescue cognitive impairments, and attenuate astrogliosis and neurodegeneration. ^{Reference Tian, Cheng and Zhuang78} Moreover, in vitro studies using Tg2576 mice identified furosemide to reduce oligomerization of Aβ40 and Aβ42 and dissociate aggregated oligomers of Aβ42 to prevent AD pathologies. ^{Reference Zhao, Wang, Ho, Ono, Teplow and Pasinetti79} Wang et al. further observed the potential of furosemide as a probe molecule in alleviating neuroinflammation in AD. ^{Reference Wang, Vilekar, Huang and Weaver80} Furosemide induced the anti-inflammatory microglial M2 phenotype through reduced production of pro-inflammatory markers such as TNF-α, IL-6, NO, COX-2, and inducible nitric oxide synthase (iNOS), promotion of phagocytosis, and elevated secretion of anti-inflammatory IL-1RA and arginase. They then synthesized and optimized furosemide analogs to inhibit Aβ aggregation and neuroinflammation, demonstrating the therapeutic potential of furosemide-like drugs in AD. ^{Reference Wang, Vilekar, Huang and Weaver80}

Following COVID-19-induced hypercytokinemia, excessive production of pro-inflammatory cytokines, including IL-6 and TNF-α, underlies the resulting multi-organ pathologies. Furosemide treatment on peripheral blood mononuclear cells (PBMCs) derived from normal subjects was shown to reduce production levels of pro-inflammatory cytokines IL-6, IL-8, and TNF-α. ^{Reference Yuengsrigul, Chin and Nussbaum81} Moreover, increasing doses of furosemide were found to reduce secretion of IL-6 and TNF-α by placentas and PBMCs in normal pregnancy. ^{Reference Brennecke, Villar and Wang82} It has also been reported that furosemide drives macrophages towards an anti-inflammatory cytokine profile, hinting at its immunomodulatory effects. ^{Reference Bryniarski, Nazimek and Marcinkiewicz83} The beneficial clinical effects of furosemide are further supported by multiple clinical trials which have identified alleviated production of pro-inflammatory cytokines in patients with pulmonary pathologies including chronic lung disease, bronchopulmonary dysplasia, and tachypnea. ^84–88

Considering that pulmonary edema is thought to be due to the cytokine storm, a retrospective observational study was conducted on patients with COVID-19; tomographic evidence of pulmonary edema and volume overload justified a standard treatment using furosemide and a Negative Fluid Balance (NEGBAL) approach. ⁸⁹ Promising clinical responses to NEGBAL have been reported. ^{Reference Santos, Zanardi and Alo90} Moreover, Kevorkian et al. have suggested repurposing furosemide in combination with early short-course corticosteroids for use in non-critically ill COVID-19 patients to reduce the risk of mechanical ventilation and/or mortality. ^{Reference K.Kevorkian, Riveline and Vandiedonck91} In addition, an ongoing Phase 2/3 clinical trial is assessing the efficacy of nebulized furosemide for treatment of pulmonary inflammation and respiratory failure in intubated and mechanically ventilated COVID-19 patients. ⁹²

The need for a widely available therapeutic to address the urgent need for ameliorating COVID-19 hypercytokinemia is growing rapidly. While current biological therapeutics such as siltuximab (an IL-6 antagonist) may target similar inflammatory pathways as furosemide, they have a relatively high immunogenic potential. ^{Reference Harding, Stickler, Razo and DuBridge93} Therefore, as serum levels of IL-6 and TNF-α are dominant predictors of COVID-19 severity and death ^{Reference Del Valle, Kim-Schulze and Huang94} and the fact that furosemide has been reported to be an inhibitor of both, repurposing this approved drug renders an optimistic therapeutic approach.

Non-Steroidal Anti-Inflammatory Drugs

Non-steroidal anti-inflammatory drugs (NSAIDs) such as ibuprofen are used in the treatment of mild-to-moderate pain and inflammation. ^{Reference Rainsford95} Mostly via nonselective and reversible inhibition of the cyclooxygenase enzymes COX-1 and COX-2, NSAIDs exert their analgesic and anti-inflammatory properties. A study by Zhang et al. found that ibuprofen treatment in a mouse model of AD led to suppression of inflammatory factors normally upregulated in AD including TNF-α, IL-1β, and NF-κB. They also found that through suppression of inflammation, the function of P-glycoprotein (P-gp) causing Aβ efflux was also increased, suggesting that ibuprofen may reduce AD pathology through a P-gp-mediated mechanism. ^{Reference Zhang, Qin and Zheng96} Interestingly, in an aged mouse model of AD, ibuprofen treatment led to a reduction in the levels of oxidative damage and ultimately reduced microglial activation and plaque deposition. ^{Reference Wilkinson, Cramer and Varvel97} Moreover, the beneficial effects of ibuprofen on cognition have also been demonstrated using a transgenic mouse model of AD in which treated mice achieved similar scores as control normal mice on complex visual-spatial learning tasks. ^{Reference Van Dam, Coen and De Deyn98} A bioinformatic analysis also revealed that ibuprofen treatment was associated with altered expression of genes associated with AD, suggesting that it may be a beneficial long-term therapeutic. ^{Reference Zamanian-Azodi and Rezaei-Tavirani99}

While during early stages of the pandemic there were concerns of exacerbated COVID-19 through use of NSAIDs, recent evidence suggests otherwise. As COVID-19 pathogenesis is largely driven by mediators of inflammation, repurposing NSAIDs which have anti-inflammatory effects is a potential therapeutic approach. A number of studies have suggested that early administration of NSAIDs may reduce the COVID-19 hyper-inflammatory response. ^{Reference Kelleni100,Reference Kelleni101} Moreover, as NSAIDs have been shown to reduce numerous pro-inflammatory cytokines involved in the initiation of cytokine storm, ^{Reference Chen, Alfajaro and Chow102} their use in COVID-19 may be of benefit. Other mechanisms by which NSAIDs may be beneficial include its effects on dampening of the NF-κB pathway, ^{Reference Mahdy, Galley, Abdel-Wahed, el-Korny, Sheta and Webster103} inhibition of caspases, ^{Reference Smith, Soti, Jones, Nakagawa, Xue and Yin104} reducing prostaglandin-mediated edema, ^{Reference Singh, Rastogi and Bansal105} and via modulatory effects on iNOS inflammatory cascades. ^{Reference Berg, Fellier, Christoph, Grarup and Stimmeder106} Repurposing NSAIDs for therapeutic use in both AD and COVID may therefore be a rational approach.

Sildenafil

Sildenafil is a selective cGMP-specific phosphodiesterase-5 inhibitor used primarily for the treatment of erectile dysfunction and pulmonary arterial hypertension. ^{Reference Barbas, Llinas and Prohens107} Considering that SARS-CoV-2 disrupts pulmonary perfusion regulation, oral administration of sildenafil to reduce pulmonary vascular resistance and inflammation may be a rational therapeutic approach in mild to severe cases. ^{Reference SSantamarina, Beddings and Lomakin108–Reference Isidori, Giannetta and Pofi110} A recent study observed that sildenafil treatment for COVID-19-induced acute respiratory distress syndrome was well tolerated and led to enhanced cardiac biomarkers as well as echocardiographic outcomes. ^{Reference McFadyen, Garfield and Mancio111} Thus, sildenafil may be a potential therapeutic in pulmonary complications of COVID-19 due to its easier use compared to nebulized vasodilator therapies such as inhaled NO which are unstable and difficult to administer. ^{Reference SSantamarina, Beddings and Lomakin108}

As sildenafil affects vascular function through its activation of the NO signaling cascade, ^{Reference Balarini, Leal and Gomes112} it may be a promising pharmaceutical intervention in AD. A single use of sildenafil in AD patients was demonstrated to improve cerebral oxygen metabolism and function. ^{Reference SSheng, Lu and Liu113} Using an endophenotype disease methodology for AD drug repurposing, it was identified that sildenafil use was associated with a 69% decreased risk of AD, even in patients with coronary artery disease, hypertension, and type 2 diabetes. Sildenafil was also shown to reduce expression of phospho-tau in neuron models derived from AD patients. ^{Reference Fang, Zhang and Zhou114} In a mouse model of AD, sildenafil was shown to reduce hippocampal levels of Aβ, reverse memory, and cognitive deficits and induce an anti-inflammatory response to prevent neuroinflammation. ^{Reference Zhang, Guo and Zhao115} A separate study also reported similar findings using an AD mouse model in which sildenafil reverse cognitive deficits reduced hippocampal tau hyperphosphorylation and increased the expression of brain-derived neurotrophic factor. ^{Reference Cuadrado-Tejedor, Hervias and Ricobaraza116} Additionally, in rats it was identified that sildenafil reduced levels of vascular cell adhesion molecule-1 (VCAM-1), TNF-α, and oxidative stress, while increasing levels of vascular endothelial growth factor, thereby hinting at its potential modulatory effects. ^{Reference Ibrahim, Haleem, AbdelWahab and Abdel-Aziz117} Thus, there is strong evidence for the dual beneficial effects of sildenafil on AD pathology and COVID-19 and may be a rational drug repurposing strategy.