INVESTIGATING NAT PRODUCTS & PHARMACEUTICALS TARGETING INFLAMMATION IN NEURODEGENERATIVE DISEASES

Alzheimer’s and Parkinson’s disease (AD and PD, respectively) are the two most common neurodegenerative diseases. Currently, no interventions have been successful in stopping the progression of these diseases. Here, we utilize two independent strategies: the use of natural products and repurposed pharmaceuticals to target neuroinflammation in models of neurodegeneration. Natural products have been used for their medicinal properties for centuries in traditional eastern medicine practices. Mucuna pruriens (Mucuna) is a well-known natural source of levodopa, typically prescribed in Ayurveda for PD. A novel levodopa reduced seed extract exhibited protective effects against oxidative stress and PD specific toxic agents, both in vitro and in vivo. Further phytochemical investigation of our extract led to the isolation and identification of seven newly reported compounds. Isolates failed to protect against toxin inducers of PD in cellular models. These data support that our novel levodopa reduced seed extract, but not isolated compounds, protect against toxin-induced models of PD. The Mediterranean diet, which is primarily composed of polyphenols, has gained considerable interest in the management of age-related diseases. Therefore, common polyphenols classes, isoflavones, and lignans with their gut-derived microbial metabolites were evaluated. Polyphenol microbial metabolites generally showed greater blood-brain barrier permeability and protection against oxidative stress, as compared to their parent compounds. Moreover, polyphenol microbial metabolites may heavily contribute to the beneficial effects of polyphenol enriched diets in disease prevention. Repurposing pharmaceuticals is an approach that allows for expedited drug discovery to fast track pharmaceuticals to a new targeted patient population. Here, we analyze direct thrombin inhibitor, dabigatran etexilate (Pradaxa®), against neuroinflammation in AD and PD models. In a Drosophila melanogaster transgenic model of PD, dabigatran treatment improved locomotor ability and reduced neuroinflammatory markers in males. Further, in a tau-based animal model of AD short-term treatment with dabigatran modulates expression of proteins related to antioxidants, mitogen-activated protein kinases, tau, thrombin, and oxidative stress. Taken together, these data indicate short-term treatment with a direct thrombin inhibitor modulates protein expression in the brains of aged Tg4510 mice. These findings support our hypothesis that targeting thrombin, a key mediator of neuroinflammation and neurotoxicity may be effective in reducing neuroinflammation in neurodegenerative diseases. Our results indicate two common drug discovery methods, namely, investigation of natural products and repurposing pharmaceuticals, may provide insights for targeting neuroinflammation in neurodegenerative diseases. Further research should focus on moving these therapeutics through the drug discovery pipeline to the patient population.

neuroinflammation in AD and PD models. In a Drosophila melanogaster transgenic model of PD, dabigatran treatment improved locomotor ability and reduced neuroinflammatory markers in males. Further, in a tau-based animal model of AD short-term treatment with dabigatran modulates expression of proteins related to antioxidants, mitogen-activated protein kinases, tau, thrombin, and oxidative stress.
Taken together, these data indicate short-term treatment with a direct thrombin inhibitor modulates protein expression in the brains of aged Tg4510 mice. These findings support our hypothesis that targeting thrombin, a key mediator of neuroinflammation and neurotoxicity may be effective in reducing neuroinflammation in neurodegenerative diseases.
Our results indicate two common drug discovery methods, namely, investigation of natural products and repurposing pharmaceuticals, may provide insights for targeting neuroinflammation in neurodegenerative diseases. Further research should focus on moving these therapeutics through the drug discovery pipeline to the patient population.
x LIST OF TABLES MANUSCRIPT 1: Table 1. Levodopa (L-dopa) content for each Mucuna pruriens seed extracts as determined by liquid chromatography electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS)……………………………………………………37 Table 2. Survival (median and maximum) of C. elegans (N2), as compared with 1-methyl-4-phenylpyridinium (MPP + ) treatment (750 μM)……………..…41  Table 1. In silico SwissADME predictive permeability of parent polyphenols and their microbial metabolites in the gut and blood-brain barrier (BBB)..….79          Alzheimer's Disease AD typically affects the elderly population, characterized by episodic memory deficits later progressing to full-blown dementia with language impairment and severe behavioral changes 4 . Specific causes of AD are not known; however, lifestyle, genetics, and environment are all identified as risk factors 5,6 . Pathologically, AD is identified postmortem by the presence of b-amyloid (Ab) plaques and neurofibrillary tangles, comprised of Ab fragments and hyperphosphorylated tau, respectively. The AD brain exhibits extensive neuronal death, believed to originate in the hippocampus and move to adjacent brain regions 4 . For decades, the amyloid hypothesis has dominated AD research. This hypothesis suggests there is a cascading event that beings with Ab accumulation leading to neurofibrillary tangles, which in turn disrupts synaptic and neuronal function 7 . Pharmaceutical interventions targeting disease pathogenesis have not been successful in stopping the progression of the disease.

MANUSCRIPT 2:
It has become more widely accepted that AD is much more complicated than the original amyloid hypothesis. Research has shifted to explore age-related decline through the influence of tau, inflammation, gut microbiota and vascular dysfunction in AD pathogenesis [8][9][10][11] . Our previous research has focused heavily on neuroinflammation induced by the activation of the vasculature. Driving this interest is the correlation of cardiovascular disease and vascular risk factors, such as hypertension and diabetes, with AD incidence rates 6 . Further, injury to the cerebrovasculature leads to the activation and dysfunction of brain endothelial cells 12 .
In the AD brain, vascular activation has detrimental consequences for neuronal viability. Many vascular-derived factors, such as thrombin, are neurotoxic and likely critical in the pathogenesis of AD 11,13,14 .
Currently, available medications target symptoms, but not disease pathology or progression. The two approved drug treatments for AD are inhibitors of cholinesterase and N-methyl-D-aspartate (NMDA) receptors; both exhibit modest beneficial effects on cognition 5,15 . Significant research has focused on identifying novel disease-modifying drugs, but there have been several failures in the clinic 16 .
Current trends for AD research are now focusing on identifying biomarkers for earlier diagnosis and identifying interventions that aim to modify disease progression 4,17 .
Continued AD research must identify new ways to modify disease progression and pathogenesis to drive the future of pharmaceutical developments.
Parkinson's Disease PD is characterized by irregular, uncontrolled movements, frequently presenting as resting tremors or dyskinesias. Pathological hallmarks include the loss of dopaminergic neurons in the substantia nigra and an abundance of a-synuclein aggregates, termed Lewy bodies, throughout the brain 18 . Direct causes of Parkinson's disease are not known; however, a variety of genetic predispositions, gender, and environmental factors have been identified as contributors. At least 17 genetic mutations have been identified as contributing factors in PD including a-synuclein, parkin, DJ-1, and leucine-rich repeat kinase 2 (LRRK2) [18][19][20] . Environmental exposure to pesticides, such as paraquat, and similarly structured chemical 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) can also result in symptoms indistinguishable from PD [18][19][20] . Interestingly, males are 1.5 times more likely to have PD than females, and incident rates in females increase dramatically after menopause, suggesting a protective role of estrogen 21,22 .
Pharmaceutical interventions for PD have mainly focused on the replenishment of dopamine to reduce motor impairments. These pharmaceutical classes include catechol-O-methyl transferase (COMT) inhibitors, monoamine oxidase type B (MAO-B) inhibitors, dopamine agonists, and levodopa paired with a dopamine decarboxylase inhibitor [23][24][25] . Levodopa combination therapy limits the production of dopamine in the periphery and attempts to restore dopamine imbalance in the brain. Levodopa is then converted to dopamine by dopamine decarboxylase, increasing levels of dopamine in the affected brain regions 26 . Levodopa paired with a dopamine decarboxylase inhibitor is currently the standard of care for PD; however, with prolonged treatment, levodopa becomes less effective, and motor deficits can increase 26 . The aforementioned therapies are effective in some cases, but new formulations and administration routes are being explored to overcome problems of prolonged efficacy 23 . Surgical interventions have also been approved as therapies for PD. Deep brain stimulation using high-frequency stimulation in the subthalamic nucleus and globus pallidus internus exhibits improved motor control 24,25 . Similar to AD, available interventions cannot stop or prevent disease progression; therefore, it is essential to identify and target disease-modifying mechanisms. Current literature proposes the investigation of genetic mutations, mitochondrial function, neuroinflammation, and oxidative stress, among others 18,26 .

Neuroinflammation in Alzheimer's and Parkinson's Diseases
Neuroinflammation has been recognized as a major factor in AD and PD pathogenesis 6,7,17,18,[26][27][28] . While these two diseases present quite differently, the neuroinflammatory response is similar 3 . Integral to the regulation of inflammation in the brain is the neurovascular unit (NVU). Astrocytes, microglia, neurons, pericytes, and endothelial cells collectively create the NVU that maintains normal cerebral blood flow and blood-brain barrier (BBB) functioning 29 . Under normal conditions, microglia, the major immune cells of the brain, work to detect pathogens or tissue damage and elicit a response to clear cellular debris and degenerating cells 17 . However, when microglia are continuously activated by external stimuli such as aggregated protein or foreign pathogens, this response can then exacerbate any present neuronal damage 30 . This toxic microglial inflammatory response includes the release of a variety of inflammatory mediators, such as chemokines, cytokines, cyclooxygenase-2 (COX-2), inducible nitric oxide synthase (iNOS) and reactive mediators, namely, reactive oxygen species (ROS) and nitric oxide species (NOS) 17,31,32 . The chronic production of these inflammatory mediators increases oxidative stress and contributes to neuronal death. Although complex, neuroinflammation plays a major role in neurodegenerative diseases and has been identified as a worthy target for pharmaceutical interventions in neurodegenerative diseases.

Therapeutic Strategies
In recent years, big pharmaceutical companies such as Pfizer have shut down their neuroscience research groups. A major part of this is due to the lack of success in developing drugs to treat neurological diseases. Since 1990, there has been a decline in nervous system drugs, with fewer Phase I trial starts, as well as a greater probability of failure in Phase III 33 . The investigation of new therapeutics for central nervous system diseases has proven to be extremely cost and time ineffective. Therefore, it is imperative to acknowledge and utilize methods that will reduce both factors. Herein, we investigate two commonly used drug discovery strategies: natural products and repurposing pharmaceuticals. Therapeutic Strategy I: Natural Products Plant-based approaches to medicine have been utilized for centuries. The generational knowledge of traditional Chinese medicine (TCM) and traditional Indian medicine (Ayurveda) have driven the investigation of natural sources to identify biologically active compounds 34 . Among these discoveries have come major medical advancements for cancer, hypertension, malaria, and pain, among others 35 . One of the most commonly known natural product derived pharmaceuticals is aspirin. Willow bark as a pain and fever reliever dates back more than 3,500 years. The active compound, salicylic acid, was initially identified in 1838. Scientists at Bayer then acetylated salicylic acid to create a more stable compound, now known as aspirin 36 .
This story is not unique to aspirin, approximately 51% of all newly approved drugs from 1981 to 2014 were derived from or structurally related to a natural product 37 .
Natural product, Gingko biloba, is an example that has been successful in mitigating the cognitive decline in animal models but has been controversial in AD clinical trials [38][39][40][41] . The success of natural products has been instrumental in the foundations of modern medicine and thus should not be ignored as a viable option for future therapeutic developments 35,42 .
Mucuna pruriens is a legume common to southern China and eastern India, used heavily as a food source, and for the treatment of tremors 43 . Ayurveda traditionally used Mucuna to treat PD, as the beans are naturally high in levodopa 44 .
This use of Mucuna pruriens led to further investigation of this medicinal plant. In two clinical trials, Mucuna pruriens preparations exhibited more rapid onset and more tolerable profiles as compared to levodopa paired with a dopamine decarboxylase inhibitor, suggesting potential advantages of the natural product to the common pharmaceutical 45,46 . Multiple studies suggest that the anti-PD activity of Mucuna is entirely due to the presence of levodopa [46][47][48][49] . However, newer evidence also suggests that there may be bioactive compounds other than levodopa could provide neuroprotection 50 .
Another interesting class of natural products, namely, polyphenols, exhibits a wide range of bioactive properties including antioxidant, anti-inflammatory, antiapoptotic and lipid-lowering properties 51,52 . Interestingly, the Mediterranean diet is characterized by a high intake of polyphenols and has been identified as a beneficial intervention for the management of cardiovascular disease, obesity and neurodegenerative diseases [53][54][55] . While polyphenols themselves may exhibit biological effects, it is generally considered that the consumed compounds get metabolized in the colon by microbiota to produce bioactive metabolites [56][57][58][59][60] . Two common classes of polyphenols: lignans and isoflavones are found primarily in flaxseed and soy, respectively. The common and prototypical dietary lignan, secoisolariciresinol (SECO), is known to be converted to the polyphenol microbial metabolites (PMM) enterodiol (ED) and enterolactone (EL) 61 . The isoflavones, genistein (GEN) and daidzein (DAI), are metabolized in the gut by intestinal microflora to produce equol (EQ) 62 . Isoflavones are structurally similar to estrogen, containing 3phenylchromen-4-one backbone, making them a curious class of compounds for PD research 63 . Additionally, GEN was identified as a potential nutraceutical for AD due to the ability to inhibit mitochondria-dependent apoptosis, and alleviate b-amyloid neurotoxicity, but failed in clinical trials [64][65][66] . Failing in the clinic may have been due to rapid metabolism into metabolites, therefore it would be of interest to examine gut microbial-derived metabolites.
Therapeutic Strategy II: Repurposing Pharmaceuticals Another promising strategy for new therapeutics is to repurpose "old" drugs.
Repurposing pharmaceuticals is defined as determining alternative uses beyond the scope of the original indication 67 . This approach is advantageous as these therapeutics have been extensively studied for safety, pharmacokinetic profiles, bulk manufacturing, and in vitro and in vivo screening 67 . This strategy reduces risk, development time and costs, and provides a fast-track to the patient population 68 .
Repurposing pharmaceuticals has been utilized to identify new therapies for cancer, depression, and schizophrenia, among others [68][69][70] . Aspirin is again a good example drug, as it has been re-purposed for several alternative therapeutic indications.
Aspirin was initially FDA-approved to reduce pain and fevers. Recently, aspirin has been extensively studied for antiproliferative and anticancer activities 71 . Aspirin is effective in reducing cell viability in multiple subtypes of breast cancer cells 72 . In colorectal cancer patients, aspirin has shown improved survival rates 73 . In addition to cancer, aspirin has also been analyzed in combating fungal infections 74 . Moreover, recent studies have looked at repurposing approved antibiotics to reduce protein aggregation in neurodegenerative diseases [75][76][77] . Using pharmaceuticals to target diseases alternative to initial indications has proven a worthy mechanism to identify new therapies and should be continuously explored.
Dabigatran etexilate (Pradaxa®, Boehringer Ingelheim) is a commonly prescribed anticoagulant that directly inhibits thrombin. Vascular activation results in injured endothelial cells which upregulate the production of thrombin both in vivo and in vitro 13,78,79 . Elevated levels of thrombin can be neurotoxic through the interaction with protease-activated receptor 1 (PAR-1), leading to the release of reactive oxygen species (ROS) and increases in a large array of inflammatory proteins 80,81 . Increased levels of both thrombin and PAR-1 have been identified in both AD and PD 82 .
Previous studies have identified that dabigatran treatment mitigates the neurotoxic effects of thrombin 13,83,84 . Dabigatran etexilate also provides neuroprotection in a rotenone-induced rat model of PD through nuclear receptor-related 1 85 .

Dissertation Objective
The following dissertation will propose new therapeutic strategies from natural products and pharmaceuticals to reduce neuroinflammation in AD and PD models.
Outcomes will provide new insight into the current state for the protection against neuroinflammation.

Introduction
Parkinson's disease (PD) is a progressive neurodegenerative disease that leads to impaired motor function and is characterized by a loss of dopaminergic neurons in the substantia nigra and is second only to Alzheimer's disease in its prevalence [1].
The etiology and pathophysiology of PD are not very well understood and have consequently stifled the development of effective therapeutic interventions for PD.
Accumulating evidence suggests that elevated oxidative stress and neuroinflammation associated with microgliosis and intracellular aggregation of αsynuclein molecules may be responsible for dopaminergic neuronal atrophy and ultimately the clinical manifestation of PD [2][3][4].
Mucuna pruriens, commonly known as Mucuna or velvet bean, is native to eastern India and western regions of China. Mucuna seeds, a rich source of naturally occurring levodopa (L-dopa; 4-7% in Mucuna seeds) [5], have been used traditionally as an effective remedy for several brain related maladies, including reducing tremors (as seen in PD), as documented in the ancient treatise of Ayurveda, the Indian traditional system of medicine [6]. The lack of effective pharmaceutical treatments has stimulated research interest in Mucuna as a PD therapeutic agent in several animal studies and a limited number of human clinical trials [7][8][9]. For example, Mucuna, at a dosage of 17.5 mg/kg, improved motor function and reduced dyskinesia in patients with advanced PD with fewer adverse effects as compared with the conventional treatment of L-dopa paired with a dopamine decarboxylase inhibitor, namely Carbidopa [9]. Mucuna has also been reported to show protective effects against PD in rodent models by increasing the activity of brain mitochondrial complex-I [10] and reducing motor dysfunction [11,12].

Murine Microglia BV-2 Cells
The production of total nitric oxide species (NOS) was determined using the Griess reagent as previously reported by our group [14,20]. BV-2 cells were seeded in clear 24-well plates at 1 × 10 5 cells/mL in serum free media. Cells were treated with MPE (12.5, 25, and 50 μg/mL), Resv (20 μM), or solvent control (0.1% DMSO) for 1 h. The cells were exposed to inflammatory stress induced by treating with LPS (1 μg/mL) for 24 h. Next, culture media from each well were transferred to a 96-well plate and measured for total NOS using the Griess reagent kit (Promega, Fitchburg, WI, USA). Absorbance values were recorded using the SpectraMax M2 plate reader (Molecular Devices, Sunnyvale, CA, USA) at 535 nm.

Non-Contact Co-Culture Assay with BV-2 and SH-SY5Y Cells
The non-contact co-culture assay was performed according to protocols previously reported by our group [21]. Media from each treatment was collected and centrifuged at 15,000 rpm for 10 min.
After centrifugation, BV-2 cell supernatant was used to treat SH-SY5Y cells for 24 h.
Cellular viability of SH-SY5Y cells was determined using the CTG assay.

1-Methyl-4-Phenylpyridinium (MPP + ) Induced Dopaminergic Neurotoxicity in C.
elegans Wild type C. elegans (N2) were maintained on nematode growth media culture plates at 20 °C and age synchronized as previously reported by our group [21]. Then,  t10 seconds to calculate the climbing distance [24].

Statistical Analyses
All data are presented as mean ± standard errors of three separate biological

Preparation of Levodopa (L-dopa)-Reduced Mucuna pruriens Extract (MPE)
Mucuna pruriens is a medicinal plant that is well known to naturally contain Ldopa (4-7%) [5], which might be attributed to its neuroprotective effects against PD [6]. However, the presence of other phytochemicals in M. pruriens, including polyphenols (tannins, flavonoids, gallic acid, phenolic acids), saponins, terpenoids, alkaloids, and fatty acids, have been reported with various pharmacological activities (see Supplementary Materials; Figure S2 and Table S1) [6,[25][26][27][28][29]. Recent studies also suggest that phytochemicals apart from L-dopa may also contribute to the overall neuroprotective activities of M. pruriens [13,30]. Therefore, in this study, we prepared a M. pruriens seed extract (MPE) containing reduced L-dopa levels (<0.1%), which was subsequently evaluated for its neuroprotective effects using a panel of in vitro and in vivo assays. The seeds of M. pruriens were extracted/solvent-solvent partitioned in varying solvents to yield extracts, which were evaluated for L-dopa content by liquid chromatography electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS). As shown in Table 1, the L-dopa levels in the initial methanol M.
pruriens seeds extract was 28.0%, which was significantly reduced to 0.03% in the ethyl acetate M. pruriens extract (MPE). As even this low level (0.03%) of L-dopa could impart biological effects, we evaluated a pure L-dopa solution (<0.1%) in several of the in vitro assays. Our preliminary data showed that the MPE, but not this pure L-dopa was active in these assays (data shown in Supplementary Materials Figures S5 and S6). Therefore, this MPE extract was selected for further evaluation of its neuroprotective effects in a panel of cell-based and in vivo bioassays as described below.

Species (ROS) Production in Microglia BV-2 Cells
Microglia are the native immune cells of the and differentiated human SH-SY5Y neuronal cells that exposed to H2O2 [14].
Moreover, studies from other research groups also reported that Mucuna seeds powder (300 mg/kg/BW in diet) reduced oxidative stress in rodent sperm cells [31].

MPE Reduces Lipopolysaccharide (LPS)-Induced Nitric Oxide Species (NOS) Production in Microglia BV-2 Cells and Protects SH-SY5Y Cells in a Co-Culture
Model Elevated production of NOS leading to massive neuronal death has been implicated in PD [32]. All of the aforementioned Mucuna extracts (at 25 µg/mL) were evaluated for their protective effects against LPS-induced NO production in BV-2 cells. Among the extracts, MPE showed the highest ability to reduce NO production in BV-2 cells exposed to LPS (see Supplementary Materials Figure S3B). Therefore, MPE was evaluated for its protective effects against neuroinflammation induced by LPS in BV-2 cells and in a non-contact co-culture model with SH-SY5Y neuroblastoma cells [21]. As shown in Figure  reporting a reduction in nitrite levels induced by 1-methyl-4-phenyl-1,2,3,6tetrahydropyridine (MPTP) in the nigrostriatal region of Parkinsonian mice brain [33]. nigrostriatal tract of rats with PD symptoms [10,33].

MPE Reduces Lethality of MPP + Induced Dopaminergic Neurotoxicity in C. elegans
The neurotoxin, MPTP, is metabolized to MPP + by monoamine oxidase-B and is subsequently taken up by dopaminergic neurons, where it inhibits mitochondrial complex I, resulting in ATP depletion to induce neuronal death [35]. Therefore, we evaluated the effects of MPE against MPP + dopaminergic neurotoxicity in wild type C.
elegans. The effects of MPE in MPP + induced neurotoxic paralysis and lethality in C.
The median and maximum survival of worms after exposure to 750 μM MPP + was 72 h ( Table 2).
Treatment of MPP+ significantly reduced the median and maximum survival by 3.2-fold (72 h) and 3.5-fold (72 h), respectively, compared with worms in the control group ( Figure 4A). MPE at 20 μg/mL significantly increased (p < 0.001) the median and maximum survival by 1.3-fold (96 h) and 1.9-fold (138 h), respectively, compared with worms treated with MPP + alone ( Figure 4B). MPE at 40 μg/mL significantly increased the mean and maximum survival in C. elegans by 1.8-fold (132 h) and 2.25-fold (162 h) respectively, compared with worms treated with MPP + alone ( Figure   4C and Table 2).

MPE Abrogates Chemically Induced Neurotoxicity in D. melanogaster
Changes in several behavioral phenotypes of D. melanogaster in response to genetically or chemically induced neurotoxicity have been exploited extensively to evaluate potential neuroprotective effects of therapeutics [36]. As MPE was significantly more neuroprotective at 40 μg/mL in reducing MPP + induced dopaminergic neurotoxicity in C. elegans (Figure 4), we used this dosage to determine its effect on climbing behavior (negative geotaxis) in D. melanogaster neurotoxin induced PD model. The two neurotoxins (6-OHDA and rotenone) used in our study induce a PD-like phenotype in D. melanogaster characterized by several behavioral changes including a muted innate negative geotaxis response due to locomotor defects. The aforementioned toxins generally injure dopamine neurons and cause behavioral defects including climbing, which can be measured by negative geotaxis assay. Similar to MPP + , rotenone is another mitochondrial complex I inhibitor that causes ATP impairment and ROS production and induces neuronal death [35]. In our study, D. melanogaster were exposed to 6-OHDA and rotenone to induce PD like phenotype. After 10 days, flies exposed to neurotoxins showed a highly muted climbing ability compared with control group. This loss of negative geotaxis ability was significantly ameliorated when flies were pre-treated with MPE.
The median climbing distance in 6-OHDA treated flies and rotenone treated flies climbing behavior. In the MPE + 6-OHDA treated flies, the median climbing distance was 54.5% (13.95 cm) higher as compared with flies that were treated with 6-OHDA alone ( Figure 5). In the MPE + rotenone treated flies, this was 48.7% (9.9 cm) higher than in flies that were exposed to rotenone only without any MPE pre-treatment ( Figure 5). Our results on the neuroprotective effects of MPE on neurotoxin induced PD models using C. elegans and D. melanogaster support previous studies with Mucuna in rodent models of PD using MPTP [33] and 6-OHDA [25] and provide further evidence on the neuroprotective effects of non-L-dopa bioactives in MPE.

Conclusions
In

Conflicts of Interest:
The authors declare no conflict of interest.            The seeds of this medicinal plant have been used traditionally to treat brain disorders associated with Parkinson's disease (PD) in the Indian traditional system of medicine, Ayurveda. Although levodopa is believed to be the major constituent in Mucuna responsible for its neuroprotective effects against PD symptoms [1,2], increasing evidence suggests that other phytochemicals may also contribute to the anti-PD effects of this natural product [3][4][5].
Our group has previously reported that a levodopa-reduced M. pruriens ethyl acetate (MPEA) seeds extract showed neuroprotective effects against PD in several in vitro and in vivo models including murine BV-2 microglia and human SH-SY5Y neuroblastoma cells, Caenorhabditis elegans, and Drosophila melanogaster [6].

General Experimental
The

Cell Culture
Human neuroblastoma (SH-SY5Y) cells were purchased from American Type Culture Collection (ATCC, Manassas, VA, USA) and maintained as previously described [6]. Compounds were dissolved in DMSO to achieve stock solutions (

Introduction
In the United States, obesity is a growing epidemic. There have been many studies that link obesity to an increased risk of developing other diseases, such as cardiovascular disease, diabetes, and Alzheimer's disease [1]. A quantitative systematic review estimated that the United States spent $113.9 billion dollars total, or 4.8% of all healthcare spending in 2008 on overweight and obesity care [2]. The standard American diet consisting of high levels of saturated fat, sodium and sugars, contributes to these staggering obesity numbers [3]. People abiding by alternative diets in countries such as Greece and Italy exhibit decreased risks of obesity as well as confounding diseases such as metabolic syndrome [4].
Increasing evidence supports the beneficial effects of a traditional Mediterranean diet in the management of cardiovascular disease, obesity and most recently, neurodegenerative diseases [5]. The Mediterranean diet is characterized by high intake of polyphenols and unsaturated fats, with most of the beneficial effects from this diet being attributed to the high polyphenol intake [6,7]. There has been an increased interest in polyphenols for the treatment of neurodegenerative diseases as they exhibit the ability to cross the blood-brain barrier (BBB) and have a wide range of bioactive properties including antioxidant, anti-inflammatory, anti-apoptotic and lipid-lowering properties [8,9].
Polyphenols are a large class of secondary metabolites produced via the shikimate-derived phenylpropanoid or polyketide pathways, and they are characterized by the presence of two or more benzene rings bearing hydroxyl group(s) and lack any nitrogenous functional group in their core structure [10].
Polyphenols are further divided into several subclasses including stilbenes, flavonoids, lignans, and phenolic acids [11]. Although polyphenols exhibit biological effects in a variety of assays, their poor bioavailability, extensive phase-2 metabolism, and whether they achieve physiologically relevant concentrations as their intact/parent forms to exert their protective effects, have been questioned. Rather, a growing consensus is that dietary polyphenols are metabolized by microbiota in the colon to yield bioactive gut microbial metabolites [12][13][14][15][16]. Herein, two common classes of dietary polyphenols were investigated, isoflavones and lignans, as well as their known polyphenol microbial metabolites (PMMs).
They exhibit the ability to serve as antioxidants, alleviate oxidative stress, and reduce the risk of hormone-dependent cancers [18][19][20]. GEN has been identified as a potential nutraceutical for Alzheimer's disease as it exhibits the ability to inhibit mitochondria-dependent apoptosis, and alleviate b-amyloid neurotoxicity, providing neuroprotection [21][22][23]. DAI also shows pro-apoptotic and neurotoxic effects against glutamate treatment in mouse hippocampal and cerebral cell cultures [24]. When metabolized in the gut by intestinal microflora, these isoflavones are converted into equol (EQ) (Figure 1) [25]. EQ has exhibited bioactive properties in cardiovascular disease, bone health, and cancers [26]. A recent study identified EQ as protective against oxidative stress in microglia, through the downregulation of neuronal apoptosis, and increased neurite growth [14].
Lignans often have complex structures made up of C6 and C3 units [27]. In plant tissues, lignans are often found as dimers and can be in the free state or sugar bound [28]. Data suggests a lignan-rich diet has numerous benefits such as prevention of hormone-dependent tumors, decrease in plasma cholesterol and glucose profiles, and delaying type 2 diabetes [29][30][31]. The common and prototypical dietary lignan, secoisolariciresinol (SECO), is known to be deglycosylated by bacteria and converted to the PMMs enterodiol (ED) and enterolactone (EL) (Figure 1) [32].
A common question regarding dietary polyphenols is whether they are present in systemic circulation in physiologically relevant concentrations to exert their biological effects. Polyphenol concentrations in the blood generally range from 0.1-1.0 μM [33].
However, certain phenolic metabolites such as pyrogallol sulfate and catechol sulfate reached plasma concentrations ranging from 5-20 μM, while their parent compounds were undetected [34]. Absorption through the GI tract is complex as the pH changes as compounds traverse through the stomach to the intestines, which changes the bioavailability of substances [35]. The high-throughput parallel artificial membrane permeability assay (PAMPA) is used to determine permeability properties related to the transcellular in vivo absorption process of large compound libraries [36]. PAMPA has become a robust, versatile method for predicting passive permeability of compounds through the gastrointestinal (GI) track, the BBB, and skin [37,38]. A major concern for the development of pharmaceuticals for CNS-related diseases is the ability of compounds to cross the BBB, therefore it is of high importance to evaluate this ability early in drug discovery [39]. The BBB is a lining of endothelial cells that protects the brain from the peripheral nervous system [40]. PAMPA uses a combination of phospholipids specific to the membrane being tested along with a microfiber filter to simulate the biological membrane [41].
Herein, we propose that certain PMMs derived from gut microflora metabolism of their parent polyphenols are gut and BBB permeable and may provide protection against neuroinflammatory stress. To explore this hypothesis, we first performed in silico screening for gut and BBB permeability and then PAMPA of the parent polyphenols and their respective PMMs. Neuroprotective activity for all of the compounds was then assessed by evaluating the levels of nitric oxide species (NOS) and inflammatory cytokines against LPS-induced inflammation in BV-2 murine microglia.

Compounds and Chemicals
Dimethylsulfoxide (

In Silico ADME Predictors
The in silico tool SwissADME was used as a measure to predict BBB and gut permeability. Predicted permeability was assessed for each individual compound and control, as previously explained [56].

Parallel Artificial Membrane Permeability Assay (PAMPA)
Parallel Artificial Membrane Permeability Assay was performed to analyze both blood-brain barrier (BBB) and gut passive permeability. All materials for PAMPA were

LPS Stimulation of Murine Microglia BV-2 Cells
As previously published, BV-2 cells were seeded at a density of 100,000 cells/mL in 24-well plates [57]. After reaching a confluency of 85%, cells were exposed to

Quantification of Nitric Oxide
Nitric Oxide (NO) was detected in BV-2 culture media following the stimulation of LPS for 24 h from the previously described experiment by way of the Griess Assay (Promega, Madison WI), according to the manufacturer's protocol.

Statistical Analysis
All data are reported as mean ± standard errors of at least three independent biological samples. The analysis of all cellular data was conducted by ANOVA followed by Dunnett's test or Tukey's test for multiple comparisons of group means.
The significance of the toxic agent compared to the control group is presented as p was used for all statistical analysis calculations and graphical representations.

SwissADME Predicts Polyphenol Microbial Metabolites Are Highly Gut and BBB Permeable
Utilizing SwissADME in silico modeling, BBB and gut permeability were predicted (Table 1). All six of the compounds (parent polyphenols and their corresponding PMMs; Figure 1) were identified as having high gut absorbance. The positive PAMPA gut controls, verapamil, ranitidine, ketoprofen and antipyrine, all exhibited high gut permeability in the SwissADME predictor. ED, DAI and EQ were predicted to have BBB permeability. The PAMPA BBB controls, both BBB permeable and impermeable, were additionally screened in SwissADME, predicting passive permeability for verapamil and corticosterone, but no BBB passive permeability by theophylline. After this predictive measure, compounds were experimentally evaluated for passive permeability in the gut and BBB.

SECO, GEN, DAI, EL, and EQ Exhibit High Permeability through PAMPA Gut
Compounds were evaluated for their permeability through simulation membranes of the gut using PAMPA at pHs 5.0, 6.2, and 7.4 ( Figure 2). Verapamil was used as the highly permeable control, ranitidine as the low permeability, and antipyrine was used as an intermediate control.  Figure 2).

GEN, EL, and EQ Show BBB Passive Permeability in PAMPA Assay
Isoflavones and lignans were evaluated for BBB penetration through PAMPA experiments ( Figure 3)

Isoflavones and Lignans Show No Cytotoxicity in Murine Microglia
Isoflavones were administered at 20 μM, while lignans were dosed at 10 μM.
Parent polyphenols and PMMs were evaluated for cytotoxicity in murine microglia after a 24 h incubation period ( Figure 4A,5A). There were no significant reductions in cellular viability at each of the test concentrations, thus indicating nontoxic levels.

Isoflavones Reduce NOS Production
Isoflavones were evaluated for their ability to reduce nitric oxide production after LPS induction, as determined by the Griess Reagent ( Figure 4B). Cells treated with LPS produced 41.5 ± 5.6 μM NOS, significantly more than the control (0.4 ± 0.02 μM). GEN significantly reduced nitric oxide at both 20 and 10 μM by 68% and 38%, respectively as compared to LPS alone. At 20 μM, DAI and EQ also significantly reduced nitric oxide compared to LPS by 24 and 22%, respectively.

Isoflavones Reduce Pro-Inflammatory Cytokine Release
TNF-a was increased significantly to 687.7 ± 8.0 pg/mL after LPS stimulation, compared to the unstimulated control, 115.4 ± 2.8 pg/mL ( Figure 4C). All three isoflavones significantly reduced TNF-a production compared to the LPS-induced treatment. GEN reduced TNF-a, by 29.1% and 16.6% at concentrations of 20 and 10 μM, respectively. DAI and EQ exhibited similar, albeit less potent, protective abilities at both the high (20 μM) and low (10 μM) concentrations ( Figure 4C). IL-6 production was significantly increased by LPS (103.7 ± 3.6 pg/mL) compared to the control (0.6 ± 0.05 pg/mL) ( Figure 4D). All isoflavones were able to significantly reduce IL-6 production compared to the LPS-induced treatment. GEN was the most effective,   Figure   5B).

Discussion
The overall health of humans and communities depends on many factors including genetics, various environmental factors, and diet. Increased adherence to the Mediterranean diet was recently correlated with a reduced risk of neurodegenerative diseases, including prodromal Parkinson's disease [5,42]. In this report, we sought to explore two common classes of polyphenols found in the Mediterranean diet, namely isoflavones and lignans, to investigate the bioavailability and bioactivity of these compounds and their metabolites.
Previously published literature suggests that for polyphenols to be bioactive, they must first be transformed in the colon by the gut microbiota into molecularly unique metabolites [16]. Compounds which are permeable through the gut mucosa enter the bloodstream and circulate in the body. The in vivo permeability may change if these substances are compatible with active transporters, as well as differences based on the individual's GI environment and diet [43]. A number of factors contribute to the absorption ability of a compound, namely, physicochemical (e.g., pKa, solubility, polarity), physiological (e.g., GI pH, GI blood flow), and dosage form (e.g., tablet, capsule) [44].
To investigate the role of lignans, isoflavones, and PMMs in systemic circulation, we explored gut permeability through in silico (SwissADME) and in vitro (PAMPA) models. Gut PAMPA results largely coincided with those of in silico predictions and showed that the test compounds, with the exception of ED, passively permeate the gut. The presence of O-methyls on SECO and lactone moiety on EL may provide these molecules with additional lipophilicity needed to cross the membrane ( Figure   1), whereas ED contains free hydroxyl groups, which may reduce passive permeability. These data are supported by previously published results such as the evaluation of GI absorption of alkaloids from Coptis [45]. PAMPA is a useful tool for predicting a compound's permeability through the human gut, however, it only considers the rate of passive transport [36].
The BBB serves as a protective barrier of the brain, comprised of endothelial cells that create tight junctions, allowing for extremely selective passive permeability [40]. This typically results in higher penetrability by lipophilic compounds through the BBB [46]. Compounds that pass through the BBB may influence the brain state and physiology. We sought to identify if the aforementioned isoflavones, lignans, and While there are many other factors that contribute to BBB permeability, such as plasma concentration, plasma binding, and metabolic modifications by barrier enzymes, PAMPA is a good predictor of passive movement across the BBB [35,46].
The relationship between gut and BBB permeability is further complicated by presystemic metabolism. For instance, previous studies have shown that following gut metabolism, EL predominantly exists as a glucuronide conjugate, certainly affecting its BBB permeability and overall bioavailability [48].
We identified that these PMMs, specifically EL and EQ, passively cross both the gut and BBB barriers. To further identify the potential bioactivities of these molecules, we examined their protective ability against inflammation in microglia. Microglia are the resident macrophages of the brain that work to eliminate debris from the brain [48]. However, continuous activation of microglia can lead to the over-production of inflammatory cytokines, as identified in postmortem brains of Alzheimer's and Parkinson's disease patients [49]. One mechanism of microglia activation is the increase in LPS levels in peripheral blood, due to poor diet [50,51]. The continued elevated levels of these pro-inflammatory cytokines, specifically IL-6 and TNF-a, can stimulate the recruitment and activation of other microglia, further inducing the production of reactive oxygen species and nitric oxide species [52]. NOS at low doses is critical for maintaining healthy microglia and neuron function, but at high doses can induce necrosis or apoptosis [53]. Phenolic compounds have previously been linked to reduced microglia-induced neuroinflammation [54]. At 20 µM, all tested isoflavones were able to significantly reduce nitric oxide production, and TNF-α and IL-6 concentrations compared to LPS alone. These results are consistent with previous reports of isoflavones [14]. However, GEN clearly showed the greatest antiinflammatory properties when compared to SECO and EQ. Furthermore, we subjected SECO and its microbial-derived metabolites to neuroprotective assays.
Among the lignans tested, only the gut-derived metabolite EL was able to significantly reduce IL-6 and TNF-a production. Additionally, EL showed significantly greater reduction in nitric oxide concentrations than its parent metabolite and ED. Recent studies with certain polyphenol metabolites, namely, gallic acid derivatives, demonstrated that these metabolites pass across BBB endothelium and provide neuroprotective effects through modulation of the NF-κB pathway [55]. It is clear from these previous studies and the work presented here that the neuroprotective potential of BBB-permeable metabolites is an important area for further investigation to understand the relationship between dietary intake of polyphenols and brain health.
In summary, the fate of microbial metabolites is largely governed by their ability to permeate through biological barriers. After investigating two classes of polyphenols often found in the Mediterranean diet, namely, isoflavones and lignans, our data suggest that their gut microbial metabolites, but not parent compounds, may enter the blood, cross the BBB and provide protection against neuroinflammation. The       suggests that gut microbial-derived metabolites are much more abundant 22 . Gut microbial metabolite, equol (EQ), has been shown to protect against cardiovascular disease, bone health, and specific cancers 23,24 . In our recent publication, we showed that of these isoflavones, EQ is most likely to cross the blood-brain barrier and protects against inflammation in murine microglia models 25 .
As phytoestrogens and known agonists of estrogen receptor beta (ER-b), we hypothesized that these isoflavones, EQ, in particular, will provide protection in

Non-Contact Co-Culture
The non-contact co-culture experiment was performed using treated media from BV-2 cells on human neuroblastoma, as previously described 27 . BV-2 were pretreated with isoflavones for 1 h, then induced with lipopolysaccharide (1mg/mL) for 23 h. Media was removed from BV-2 and placed on SH-SY5Y, seeded in whitewalled 96-well plates at 100,000 cells/mL. SH-SY5Y were then incubated for 24 h, cell viability was then assessed using CTG.

Cellular Viability against Toxic Agents
Human neuroblastoma were seeded in white-walled 96-well plates at 100,000 cells/mL for 24 h. Cells were incubated with corresponding treatments for 2 h in serum-free media, then induced with toxic agents for 24 h: 6-OHDA (100 μM) or MPP + (2 mM). Cellular viability was determined by CTG, illustrated as percentage of control.

Caenorhabditis elegans
C. elegans were maintained and assayed for lifespan against MPP + as previously

Climbing Assay
After 7 days of treatment, flies were assayed for climbing ability 28,29 . Briefly, flies were placed in empty vials marked at 5 cm and allowed to acclimate for 30 minutes. Flies were gently tapped to the bottom of the vial and a picture was taken after 10 seconds using a Nikon D2200 camera (Nikon, Tokyo, JP). Percent climbing was calculated by ((# flies above line / total # flies)*100).

Statistical Analysis
All data are reported as mean ±standard errors of at least three independent biological samples. Analysis of all cellular and Drosophila data were conducted by ANOVA followed by Dunnett's test for multiple comparisons of group means.

Cytotoxicity Screen
Compounds were tested for cytotoxicity at 20 μM in neuroblastoma (Figure 1).
No significant difference was seen between control and isoflavone treatment, indicating safe concentrations.

Effects of Isoflavones in Non-Contact Co-Culture
Treatment with LPS media reduced SH-SY5Y viability to 56.9 ± 1.2%, as compared to control, 100 ± 2.5% ( Figure 2). Isoflavones at all concentrations exhibited significantly increased cellular viability as compared to LPS treatment alone.

C. elegans Lifespan
We further evaluate EQ against toxic agent, MPP + , in the C. elegans model. MPP + treatment alone significantly reduced lifespan to 72 h (Fig. 4A). EQ at 17.5 and 35 μM significantly improved the median lifespan to 108 h (Fig. 4B &C).

Drosophila melanogaster Negative Geotaxis
D. melanogaster were evaluated for EQ (100 μM) protection against 6-OHDA toxicity in the negative geotaxis assay (Fig. 5). 6-OHDA treatment for one week failed to significantly reduce the climbing ability of male flies. EQ treatment exhibited increased climbing ability as compared to 6-OHDA, however not significantly.

Discussion
There has been an increased interest in flavonoids for the prevention of   ±standard error (n³3), reported as percentage compared to control. Statistical significance was evaluated using ANOVA followed with Dunnett's multiple comparison test, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 as compared to control cells. Toxin insult was compared to control, ## p < 0.01. Cells were treated, then exposed to 25 μM 6-OHDA (A) and 2 mM MPP + (B) for 24h, where cellular viability was assessed. Data are expressed as mean ±standard error, exhibited as percentage compared to control (n³6). Statistical significance was evaluated using ANOVA followed with Dunnett's multiple comparison test, *p < 0.05, ** p < 0.01, ***p < 0.001, and ****p < 0.0001. Toxin insult was compared to control, #### p < 0.0001.  Dabigatran treatment is yet to be explored in transgenic models of PD.
Therefore, we utilize a Drosophila melanogaster model of PD expressing mutant leucine-rich repeat kinase 2 (LRRK2). Drosophila melanogaster have been heavily used in neuroscience research as their neuronal structures and circuits and neurotransmitters, such as dopamine, are similar to those found in the human brain [23][24][25] . When dopamine is dysfunctional in Drosophila, locomotion is impaired, similar to symptoms seen in human PD 26 . Leucine-rich repeat kinase 2 (LRRK2) has been identified in both idiopathic and inherited PD, accounting for as high as 40% of familial cases 27 . Postmortem analysis revealed LRRK2 is heavily expressed in the Lewy bodies of the brain stem and cortex, as well as in the brain vasculature, axons, and neuronal cell bodies 28,29 . LRRK2 mutant Drosophila exhibit reduced lifespan, compromised motility, and mitochondria morphological changes as compared to wildtype 30 .
Herein, we hypothesize that dabigatran treatment in LRRK2 transgenic Drosophila melanogaster will protect against motor function deficits by reducing oxidative stress.

Toxicity Analysis
Toxicity of dabigatran at 25 μM was identified by recording deaths per vial every other day. At least 100 individual flies per treatment were monitored and evaluated over the course of one week.

Negative Geotaxis Assay
After 7 days of treatment, flies were assayed for climbing ability 31,32 . Briefly, flies were placed in empty vials marked at 5 cm and allowed to acclimate for 30 minutes. Flies were gently tapped to the bottom of the vial and a picture was taken after 10 seconds using a Nikon D2200 camera (Nikon, Tokyo, JP). Percent climbing was calculated by ((# flies above line / total # flies)*100) and standardized as a percentage to control (gender-specific).

Drosophila Head Homogenization
Drosophila were flash-frozen on dry ice and heads were collected through brass sieves, as previously reported 33 . Heads were counted and 10 μL/head of phosphate buffered saline was added for NOS and ROS. For western blots, fly heads were homogenized in phosphate buffered saline with 1X protease inhibitors. Samples were frozen immediately on dry ice and stored at -80°C. To homogenize the heads for assays, they were sonicated in a water bath for 5 min, centrifuged (10,000 RCF, 5 min, 4°C), then placed at -80°C until frozen. This process was repeated 3 times and the supernatant was collected.

Nitric Oxide Quantification
Nitric oxide species was quantified in fly head homogenates using the Griess Reagent System (Promega, Madison, WI, USA) 34 . tests were performed to identify significant differences between vehicle and dabigatran treatment. All other data were analyzed by unpaired one-tailed t-tests.

Toxicity Analysis
Throughout the experiment, fly deaths were recorded to evaluate toxicity related to dabigatran treatment ( Figure 1). Dabigatran did not induce significant death in either gender or genotype. Therefore, treatment at 25 μM was regarded as safe in this experiment.
Dabigatran did not exhibit any significant effects as compared to control in female flies.

Nitric Oxide Species Production
NOS production was evaluated in fly head homogenates using the Griess reagent ( Figure 3). Interestingly, LRRK2 mutant flies did not exhibit increased NOS as compared to wildtype control. In male LRRK2 flies, dabigatran exhibited a significant reduction in NOS as compared to control (control 38.8 ±5; dabigatran 26.8 ±3.3) ( Figure 3A). NOS was unaltered by dabigatran treatment in female LRRK2 flies ( Figure 3B).

Reactive Oxygen Species
ROS production in fly head homogenates was also investigated (Figure 4). flies exhibited no difference in ROS production with dabigatran treatment ( Figure 4B).

Western Blot Analysis
Tyrosine hydroxylase (TH; Figure 5

Discussion
Currently, there are no treatments that stop or even slow the progression of neurodegenerative diseases. Therefore, it is imperative to explore novel approaches aside from classical targets. Neuroinflammation has been identified as a major contributor in AD and PD that may begin years before an official diagnosis 5 . Activated brain vasculature can lead to an increase in disease severity [36][37][38] . Therefore, we performed western blotting to quantify tyrosine hydroxylase expression in fly head homogenates to identify the correlation between TH levels and the observed motor function. In transgenic male flies, there was a significant reduction in TH as compared to wildtype control. However, dabigatran treatment did not improve the levels of TH. This suggests that alternative processes may be involved.
The production of nitric oxide and reactive oxygen species are linked to both neuroinflammation and damaged cerebrovasculature in neurodegenerative diseases 5,42,43 . Dabigatran treatment has been shown to decrease oxidative stress markers in diseased models 14,20,21 . LRRK2 has also been associated with increases in these mediators of oxidative stress 44 . In Drosophila melanogaster NOS and ROS production can be altered with dietary modifications 35,45,46 . Here, surprisingly, NOS was reduced in transgenic flies as compared to wildtype. However, dabigatran treatment significantly reduced NOS production as compared to control in transgenic male flies. No differences were found in the females. Additionally, ROS was significantly reduced in male LRRK2 flies following dabigatran treatment. To further investigate these results, we analyzed the expression of proteins in the regulatory pathways of each NOS and ROS production.
NOS is essential to normal function, but the continuous production of NOS leads to excessive toxicity in the central nervous system 47 . iNOS is one of three isoforms that contributes to the production of NOS. Activated microglia induce cytokine activity, leading to iNOS production which has been deemed neurotoxic 48,49 .
We analyzed iNOS to get a better understanding of toxic NOS production. In Interestingly, our data reports differences in treatment response by gender. In human PD, the incidence, progression, and symptom severity present differently in males and females 53,54 . These differences have been largely attributed to hormonal differences 55,56 . Sex differences in Drosophila have frequently been overlooked.
However, reports on lifespan indicate the potential of extreme differences between gender in flies 57 .

Introduction
Cardiovascular disease and cardiovascular risk factors (CVRFs) are strongly associated with an increased risk of developing dementia, particularly Alzheimer's disease (AD). Hypertension, hypercholesterolemia, hyperhomocysteinemia, diabetes, atherosclerosis, and hypoxia are all linked with an increased risk of developing AD [1][2][3][4][5]. While the connection between CVRFs and AD is well-documented, the mechanism by which the disorders are connected is not well understood. It is likely there are several, pathological mediators involved in the progression of both disorders.
One such mediator is the serine protease thrombin, the main driver of the coagulation cascade [6]. Thrombin is indicated as a potential pathological mediator in both cardiovascular disease and AD. Thrombin has been implicated in atherosclerosis and diabetes-related pathology [7][8][9][10][11]. Thrombin is also elevated in the Alzheimer's brain. Past studies found increased thrombin and reduced levels of the thrombin inhibitor protease nexin-1 in the brain of Alzheimer's patients compared to healthy controls; this elevation was largely localized in vessels, amyloid deposits, and neurofibrillary tangles [12][13][14]. Thrombin is also elevated in AD patient-derived microvessels, and brain endothelial isolated from AD patients synthesize their own thrombin [15,16]. Thrombin is a mediator of vascular-derived oxidative stress and inflammation. Aside from its pro-coagulant activity, thrombin has well-documented pro-inflammatory and pro-oxidative effects on several cell types in the body, largely through its activation of protease-activated receptors (PARs) [17,18]. In the periphery, increased thrombin in disease is related to the formation of atherosclerotic plaques, endothelial cell activation, and an increase in both inflammatory cytokines and reactive oxygen species (ROS) [7,[19][20][21].
Thrombin is also associated with AD-related hallmarks in the brain, including tau, amyloid aggregation, and apolipoprotein E [40][41][42][43][44]. Rats treated with thrombin in vivo exhibit cognitive deficits along with cell death and glial scarring [35]. Thrombin may be involved in the altered processing and secretion of amyloid precursor protein (APP), and thrombin cleavage of apolipoprotein E4 results in a neurotoxic fragment [42][43][44]. Thrombin accumulation has been identified in neurofibrillary tangles, and thrombin induces rapid tau aggregation [40,41].
Thrombin is related to altered processing and aggregation of tau, but the role of thrombin as a pathological mediator has not yet been explored in a tau-based model of AD. Currently, a number of tauopathy animal models are being used to study AD. One such model is the Tg4510, which overexpresses human tau with a P301L mutation at 13:1 versus murine tau [45]. These mice exhibit profound tau pathology and neuronal loss in the hippocampus and cortex, as well as cognitive deficits and metabolic changes. The pathology starts early at about 2 to 4 months of age and progresses with age [45,46]. The pathological features, including tau hyperphosphorylation synapse loss, are more pronounced in females than males [47]. Additionally, the tg4510 mouse model displays blood vessel abnormalities accompanied by alterations in oxidative and inflammatory markers [48].
We suggest that thrombin is a mediator of cerebrovascular-derived inflammation and neurotoxicity in AD, and therefore targeting thrombin may be a worthy therapeutic strategy to combat Alzheimer's disease-related pathology. Our lab has focused on the direct thrombin inhibitor Dabigatran etexilate (Pradaxa®, Boehringer Ingelheim), a commonly prescribed anticoagulant that is administered orally. In a previous paper, we identified that dabigatran etexilate treatment for 34 weeks in 3xTg AD mouse model reduced levels of oxidative stress and markers of cerebrovascular inflammation in the brain [49].
The objective of this study is to investigate the role of thrombin as a pathological mediator and the potential therapeutic benefits of inhibiting thrombin in an animal model of tauopathy and AD. We explore the effects of short-term treatment with direct thrombin inhibitor, dabigatran etexilate, in aged Tg4510 mice. We hypothesize that that inhibiting thrombin will reduce oxidative stress and inflammation-related indicators corresponding to an overall reduction in tau-related dysfunction in the brain. Samples were centrifuged and supernatant was collected for analysis.

Mass Spectrometry Data Acquisition with SWATH-MS
Mass spectrometry was performed as previously described with minor modifications [50]. Samples were analyzed on a SCIEX TripleTOF® 5600 mass spectrometer using a DuoSpray™ ion source (SCIEX, Framingham, MA) coupled to

One-week dabigatran etexilate treatment reduces expression of coagulation related proteins
The expression of major proteins involved in coagulation, namely, thrombin, prothrombin and fibrin were evaluated by western blot. Thrombin levels remained unaltered in wild type as compared to Tg-Vehicle. Although not significant, dabigatran treatment decreased thrombin by 33% as compared to Tg-Vehicle ( Figure 1A).
Prothrombin was significantly lower in wild type mice by 44% compared to Tg-Vehicle ( Figure 1B). Dabigatran treatment significantly reduced prothrombin by 46.7% as compared to Tg-Vehicle. Fibrin levels were slightly reduced in both wildtype (54.9%) and dabigatran etexilate (44.4%) as compared to Tg-Vehicle ( Figure 1C).

Dabigatran etexilate reduces oxidative-stress related proteins
Inducible nitric oxide synthase (iNOS) and NADPH oxidase 4 (NOX4) are enzymes involved in the activation of oxidative stress by catalyzing the production of NOS and ROS. The expression of iNOS was significantly reduced in wild type and Tg-Dabigatran, as compared to Tg-Vehicle by 23.5% and 17.4%, respectively ( Figure   2A). Tg-Dabigatran also significantly reduced NOX4 expression by 24.7% as compared to Tg-Vehicle ( Figure 2B).
No differences were reported between wild type and Tg-Vehicle. Tg-Dabigatran significantly increased the expression of both SOD1 and SOD2 by 34.2% and 20.7%, respectively.

Effect of Dabigatran etexilate on neuroinflammatory proteins
Glial fibrillary acidic protein (GFAP) and ionized calcium binding adaptor molecule 1 (IBA1) are markers of activated astrocytes and microglia, respectively.

Dabigatran etexilate treatment alters tau phosphorylation but not overall expression
Total tau and phosphorylated forms S396, S404 and S416 were evaluated by western blot ( Figure 5). Tg-Vehicle expressed 75.9% more total tau than wild type mice ( Figure 5A).

LC-MS/MS SWATH acquisition identifies significant proteomic differences between wild type and Tg-Vehicle mice
Major proteins involved in coagulation, AD, inflammation and tau-related pathology were identified and evaluated using LC-MS/MS SWATH acquisition (Table   1). Significant differences in protein expression were first evaluated between wildtype and transgenic mice, visualized with a volcano plot ( Figure 6A). Coagulation related proteins exhibited significant differences in FIBB, ICAM5, KPCD, MMP9, and RACK1.
Differences were seen in inflammation related proteins DLG4, GFAP, NOS3 and TLR4. Alzheimer's disease related proteins that were significantly altered in Tg4510 mice were A4, APOE, BACE1, TAU (Human), TAU (Mouse), and TAU (Total). A recent publication identified proteomic signatures in the human brain affected by tau

Discussion
Increasing evidence has shown elevated levels of thrombin and thrombinrelated proteins of the coagulation cascade in the brains of Alzheimer's disease patients [12-14, 40-44, 51]. Further, thrombin accumulation is co-localized with tau aggregation [40,41]. Herein, we propose that inhibiting thrombin in a tau-pathology mouse model may identify an alternative approach to combat Alzheimer's diseaserelated pathology.
Our data shows the efficacy of treatment with a direct thrombin inhibitor, dabigatran etexilate, in a tau mouse model. As expected, the drug treatment reduces thrombin and proteins increased by thrombin activity (prothrombin, fibrin), confirming normal drug action. In addition to coagulation, dabigatran etexilate treatment also produced alterations in the expression of proteins involved in oxidative stress, inflammation, and tau-related pathology. These results suggest that inhibition and reduction of thrombin may affect aspects of AD pathology through the alteration of different thrombin-mediated signaling mechanisms in the brain.
Thrombin's pro-inflammatory effects throughout the body, including in the brain, are largely mediated through the signaling of its receptors, protease-activated receptors (PARs) 1, 3, and 4. PARs represent a unique family of G-protein coupled receptors that are activated by a self-ligand [17]. PARs have been found in a number of cell types in the periphery and the brain, including endothelial cells, neurons, astrocytes, and microglia [52]. When cleaved and activated by thrombin, PAR1 initiates a variety of intracellular signaling. Among others, PAR1 activates the mitogen-activated protein kinase (MAPK) pathway which can lead to cell growth, proliferation or migration [53]. MAPK related proteins were analyzed by LC-MS/MS SWATH analysis, identifying significant differences between Tg-Vehicle and Tg-Dabigatran etexilate in ANT3, ITB2, KPCB, MK08, MP2K1 and MP2K2. Interestingly, Dabigatran etexilate treatment did not alter proteins to wild type levels.
Thrombin activation of PARs is responsible for pro-inflammatory and oxidative effects in both the periphery and brain. Here, we identified significantly elevated levels of iNOS, GFAP, and IBA1 in transgenic mice as compared to wild type.
Dabigatran etexilate treatment reduced iNOS, and NOX4, as compared to vehicle.
Dabigatran etexilate significantly increased COX2, DYN1, SOD1, and SOD2. These increases may be due to the timing in the antioxidant cycle [54]. Together, these results indicate a shift towards reduced oxidative stress following dabigatran etexilate treatment.
Total tau and related phosphorylated (S396, S404, S416) tau species were significantly increased in transgenic mice compared to wild type, as was expected.
Dabigatran treatment slightly reduced expression of total tau, significantly decreased S396, S416; but led to an increase in S404. S396 and S404 phosphorylation are found early in the disease course of AD [55,56] and are related to destabilization of microtubules [57,58]. S396 is also linked with abnormal truncation of the tau protein, indicating altered functionality [55]. Phosphorylation at S416 by CamKII is largely found within the neuronal soma, rather than localized to microtubules, and has been found to be associated with the promotion of AD-related cell death [59,60]. Together, our findings indicate that dabigatran etexilate treatment may reduce AD-related tau dysfunction through altered phosphorylation, particularly decreased phosphorylation at S396 and S416.
To further explore tau pathology, we performed LC-MS/MS SWATH analysis on a variety of previously identified tau related proteins. ANS1B, NCKP1, SV2B and SYGP were significantly altered with dabigatran etexilate treatment, compared to Tg-Vehicle. These proteins were previously found to be downregulated in tau pathology in AD brains across multiple proteomic studies [61]. These data indicate additional changes in tau-related dysfunction following dabigatran etexilate treatment.
It's important to note that there may be alternative explanations for the effects seen with dabigatran etexilate treatment, particularly for the decreased expression of oxidative and inflammatory mediators. Fibrin, a coagulation protein found downstream of thrombin, has also been identified as a potential pathological mediator in Alzheimer's disease [62]. Higher levels of fibrin have been identified in the AD brain compared to healthy controls [51]. Fibrin accumulation in AD, similar to thrombin, has been linked with increases in inflammation and oxidative stress, as well as alterations in both amyloid and tau pathology [62]. Our findings showed a decrease in fibrin expression as a result of dabigatran etexilate treatment, which is expected given thrombin catalyzes the conversion of fibrinogen into fibrin. It is possible that some of the other changes identified with dabigatran etexilate treatment, including decreases in oxidative stress-related and inflammatory proteins, are the result of decreased fibrin accumulation rather than reduced thrombin signaling activity.
Alternatively, there are anti-coagulant proteins, such as activated protein C (APC), that act in opposition to thrombin. These proteins may alternatively catalyze PARs and subsequently alter the signaling mechanism activated, producing antiinflammatory and cytoprotective effects [63]. Further studies have shown antiinflammatory effects of APC-like ligands mediated through PAR signaling [64]. Just as reduced thrombin activity may result in reduced fibrin accumulation and therefore reduced inflammation, reduced levels of thrombin may allow for APC and other anticoagulant proteins to bind to PARs and activate anti-inflammatory signaling processes.

CONCLUDING REMARKS
An exponential increase in the prevalence of neurodegenerative diseases has led to a worldwide health crisis. The two most common neurodegenerative diseases, Alzheimer's and Parkinson's disease constitute the largest majority of cases.
Although research has been ongoing, no successful interventions have been discovered that stop the progression of the disease. Focus has shifted from targeting disease characteristic protein aggregations, namely b-amyloid and a-synuclein, to identifying targets earlier in disease progression, such as neuroinflammation or cognitive decline. Herein, we utilize two independent strategies: the use of natural products and repurposed pharmaceuticals to target neuroinflammation in models of Alzheimer's and Parkinson's disease. After isolation, no single compound exhibited these protective effects, supporting a potential synergistic effect in the extract. Further, polyphenol microbial metabolites, but not parent compounds showed protective effects in vitro against toxins. Equol, a gut-derived metabolite, was also effective in vivo against Parkinson's disease-specific toxic agents. These natural products exhibit protective effects in Parkinson's disease models, suggesting potential therapeutic in neurodegenerative diseases.

182
A newer approach to identify new therapeutics is to repurpose previously approved pharmaceuticals for new indications. Repurposing pharmaceuticals is a drug discovery approach that may fast track pharmaceuticals to a new targeted patient population. We analyzed direct thrombin inhibitor, dabigatran etexilate (Pradaxa®), against neuroinflammation in AD and PD models. In a transgenic While our results indicate these interventions may reduce neuroinflammation in models of neurodegeneration, extensive research is still necessary. Further elucidation of the mechanism of action in additional animal models is necessary before clinical trials can be conducted. Realistically, these natural products may be extremely far from reaching the clinic. However, dabigatran may be much closer to the targeted population. Moreover, these results support our initial hypothesis that natural products and repurposing pharmaceuticals are strategies that can target neuroinflammation in neurodegenerative diseases.