Neurodegenerative diseases (NDDs), such as Alzheimer's disease (AD) and Parkinson's disease (PD), represent a growing global health crisis, characterized by progressive cognitive and motor decline, imposing a significant burden on healthcare systems and society [1], [2]. A common pathological hallmark across these devastating conditions is a state of chronic imbalance, primarily driven by persistent oxidative stress and neuroinflammation [3], [4]. While conventional therapeutic strategies have encompassed symptomatic management, neurotransmitter modulation, anti-inflammatory approaches, and targeted interventions against protein aggregation, truly disease-modifying therapies that address the multifactorial pathogenesis of neurodegeneration remain elusive [5], [6]. This persistent gap underscores the urgent need for integrative, multi-target strategies capable of simultaneously modulating oxidative stress, inflammatory cascades, and metabolic dysfunction.
In recent years, research has increasingly illuminated the critical role of the gut-brain axis (GBA) in modulating neurological health and disease progression [7], [8]. This bidirectional communication network highlights how the composition and function of the gut microbiota can significantly influence brain homeostasis, metabolism, and inflammatory status [9]. Emerging evidence suggests that interventions targeting the gut environment may offer novel avenues for neuroprotection.
Two powerful, yet often separately studied, modulators of systemic health are dietary polyphenols and physical activity (PA). Polyphenols, secondary metabolites abundant in plant-based foods, are well-known for their antioxidant and anti-inflammatory properties [10], [11]. A significant portion of these compounds requires metabolism by the gut microbiota to become fully bioactive, leading to the production of beneficial metabolites like short-chain fatty acids (SCFAs) that exert neuroprotective effects via the GBA [12]. Similarly, physical activity has been consistently shown to favorably alter gut microbiota composition, increasing beneficial bacteria and reducing gut inflammation, which in turn supports brain health [13].
Crucially, these protective effects converge on key intracellular signaling pathways. The Nuclear factor-erythroid 2-related factor 2 (Nrf2) pathway stands out as a master regulator of cellular defense against oxidative stress and inflammation. Activation of Nrf2 leads to the transcription of numerous cytoprotective and antioxidant genes, offering a promising strategy to counteract the core pathologies of NDDs. Polyphenols are known activators of this pathway [14], [15], and exercise-induced changes in the gut milieu are hypothesized to contribute to this activation. This review aims to synthesize the current literature to establish a comprehensive model where Polyphenols and Physical Activity Stimulate Gut Microbiota Mediated Nrf2 Signaling to Combat Neurodegeneration. We will explore the synergistic relationship between diet, exercise, microbial metabolites, and the Nrf2 pathway, providing a framework for developing integrated, lifestyle-based therapeutic strategies against debilitating neurological disorders. Despite substantial progress in understanding the individual roles of polyphenols, physical activity, and gut microbiota in neurodegenerative diseases, these factors are often investigated in isolation in the existing literature. Previous reviews have primarily focused on either dietary polyphenols or exercise-induced neuroprotection, with limited integration of microbiota-mediated mechanisms and their downstream signaling pathways. In particular, the convergence of these lifestyle factors on gut microbiota–derived metabolites and their coordinated activation of the Nrf2 signaling pathway remains insufficiently addressed. Therefore, the novelty of this review lies in providing a unified mechanistic framework that integrates polyphenols, physical activity, and gut microbiota within the context of Nrf2-mediated neuroprotection. By emphasizing the role of microbiota-derived metabolites as key mediators linking lifestyle interventions to intracellular antioxidant and anti-inflammatory pathways, this work offers new insights into how combined lifestyle strategies may synergistically modulate neurodegenerative processes.
The nuclear factor erythroid 2-related factor 2 (Nrf2) signaling pathway stands as a pivotal endogenous defense mechanism against the multifaceted assaults characteristic of neurodegenerative diseases (Fig. 1) [14], [16].
Its central role stems from its function as a master transcription factor regulating the expression of a vast array of cytoprotective genes, which collectively orchestrate antioxidant responses, detoxification, mitochondrial biogenesis, and anti-inflammatory actions [17], [18]. In the context of age-dependent neurodegenerative disorders like Alzheimer's disease (AD) and Parkinson's disease (PD), where oxidative stress and chronic inflammation are not merely secondary consequences but core pathogenic drivers, the Nrf2 pathway represents a critical therapeutic target [19]. An abnormal Nrf2/ARE signaling pathway is strongly associated with the onset and progression of both AD and PD, making its activation a compelling strategy to mitigate key pathological hallmarks such as oxidative damage, mitochondrial dysfunction, and protein aggregation [16], [18]. The fundamental mechanism of Nrf2 regulation involves a delicate balance controlled by its cytoplasmic repressor, Kelch-like ECH-associated protein 1 (Keap1). Under normal physiological conditions, Keap1 binds to Nrf2 in the cytoplasm, facilitating its ubiquitination and subsequent proteasomal degradation, thereby maintaining low basal levels of the transcription factor. However, in response to electrophilic compounds or elevated levels of reactive oxygen species (ROS) both of which are prevalent in the diseased brain specific cysteine residues on Keap1 are modified [17]. This modification disrupts the Keap1-Nrf2 complex, releasing Nrf2 from its tether. The stabilized Nrf2 then translocates to the nucleus, where it forms heterodimers with small Maf proteins and binds to the Antioxidant Response Element (ARE), also known as the Electrophile Response Element (EpRE), located in the promoter regions of its target genes [20]. This binding event initiates the transcription of a comprehensive battery of protective enzymes, including heme oxygenase-1 (HO-1), NAD(P)H quinone oxidoreductase 1 (NQO1), superoxide dismutase (SOD), catalase (CAT), and various glutathione S-transferases [21]. The collective action of these enzymes enhances the cell's capacity to neutralize ROS, repair oxidative damage, and restore redox homeostasis, thereby conferring significant cytoprotection [18].
Beyond canonical Keap1-dependent degradation, Nrf2 activity is finely tuned through multiple Keap1-independent regulatory nodes that are highly relevant to neurodegeneration. Post-translational modifications, particularly phosphorylation by MAPK/ERK, PI3K/Akt, and PKC, can stabilize Nrf2 or enhance its nuclear translocation independently of cysteine oxidation on Keap1. Additionally, the autophagy adaptor p62/SQSTM1 competitively binds Keap1, sequestering it into autophagosomes and thereby liberating Nrf2 in a feed-forward loop that links proteostatic stress to antioxidant defense. The glycogen synthase kinase 3β (GSK-3β)/β-TrCP axis further modulates Nrf2 turnover, with GSK-3β phosphorylation promoting cytoplasmic retention and proteasomal degradation under basal conditions. Critically, Nrf2 exerts distinct, cell-type-specific functions within the central nervous system. Neurons maintain relatively low basal Nrf2 activity and rely heavily on astrocytic Nrf2-driven synthesis of glutathione (GSH) and NQO1 for redox buffering and neurotrophic support. In microglia, Nrf2 activation suppresses the pro-inflammatory M1 phenotype by inhibiting NF-κB-driven cytokine production and promoting a reparative M2 state. This cell-autonomous specialization underscores why global Nrf2 activation must be interpreted within the context of neurovascular unit dynamics. Furthermore, Nrf2 and NF-κB engage in extensive bidirectional crosstalk that dictates the redox-inflammatory balance in neurodegeneration. Nrf2 activation competes with NF-κB for shared transcriptional coactivators (e.g., CBP/p300) and directly impedes p65 nuclear translocation, while NF-κB can transcriptionally repress Nrf2 target genes under chronic inflammatory conditions. HO-1, a canonical Nrf2 target, generates carbon monoxide and bilirubin, which directly dampen NF-κB signaling and inflammasome activation. This mutual antagonism positions Nrf2 not merely as an antioxidant switch, but as a central integrator of cellular fate decisions that determine neuronal resilience versus degenerative decline.
The relevance of the Nrf2 pathway in combating neurodegeneration is underscored by extensive evidence from both genetic and pharmacological studies. In animal models, Nrf2 knockout has been shown to exacerbate AD-like pathologies, leading to increased numbers of reactive microglia, elevated levels of proinflammatory cytokines, and greater infiltration of immune cells into the brain [5]. Conversely, pharmacological activation of Nrf2 has demonstrated profound neuroprotective effects across various models of AD and PD. For instance, Nrf2 activation has been shown to alleviate key pathological features of PD, including oxidative stress, mitochondrial impairment, and neuroinflammation [3], [18]. Similarly, in AD models, Nrf2 activation mitigates oxidative damage and improves mitochondrial function, addressing two interconnected facets of AD pathogenesis [22]. The pathway's influence extends beyond direct antioxidant effects; it also plays a crucial role in modulating neuroinflammation. By reducing inflammasome-driven inflammation, Nrf2 offers significant potential for addressing the chronic neuroinflammatory aspects of AD. Furthermore, Nrf2 activity is tightly linked to synaptic plasticity and memory formation, with Nrf2 activators emerging as a promising novel therapeutic avenue for a range of neurodegenerative disorders [23]. The convergence of multiple pathological insults onto the Nrf2 pathway makes it an exceptionally attractive target, capable of simultaneously tackling oxidative stress, impaired protein degradation, and inflammatory cascades that drive neuronal death in conditions like AD and PD [14]. Importantly, the functional consequences of Nrf2 activation may vary across different brain cell types. In neurons, Nrf2 activation primarily supports mitochondrial function and resistance to oxidative damage, while in astrocytes it plays a key role in maintaining redox homeostasis and providing metabolic support to neurons. In microglia, Nrf2 activation can suppress pro-inflammatory signaling and modulate immune responses. This cell-type-specific complexity is critical for understanding how systemic interventions such as diet and exercise translate into neuroprotective outcomes.
Dietary polyphenols, a large and diverse class of plant-derived bioactive compounds, have garnered significant attention for their potential to combat neurodegeneration through the modulation of the Nrf2 signaling pathway [24]. Among the most studied are curcumin, resveratrol, and epigallocatechin gallate (EGCG), each demonstrating potent neuroprotective properties in preclinical models [25]. These compounds directly contribute to Nrf2 activation through several well-documented mechanisms. Curcumin, the principal polyphenol in turmeric, activates the Nrf2/ARE pathway, leading to the increased expression of antioxidant enzymes like HO-1 and NQO1, which are crucial for scavenging reactive oxygen species (ROS). These effects are mechanistically intertwined rather than independent; curcumin modulates upstream redox-sensitive kinases and electrophilic stress pathways that indirectly promote Nrf2 stabilization while concurrently dampening NF-κB signaling. Importantly, Nrf2 and NF-κB engage in extensive bidirectional crosstalk, competing for shared transcriptional coactivators (e.g., CBP/p300) and mutually inhibiting each other’s nuclear activity. Curcumin’s observed “dual” action likely reflects this complex network-level modulation rather than direct, linear activation or suppression of either pathway [26], [27]. Resveratrol, found abundantly in grapes and red wine, also potently activates the Nrf2 pathway, particularly through its interaction with sirtuin 1 (SIRT1). This activation leads to the upregulation of key antioxidant genes, including those encoding for SOD-1 and CAT, effectively bolstering cellular antioxidant defenses [28]. Epigallocatechin gallate (EGCG), the predominant catechin in green tea, similarly activates the Nrf2 pathway, enhancing the expression of HO-1 and other protective enzymes. In addition to its antioxidant properties, EGCG has been shown to inhibit the aggregation of amyloid-β (Aβ) and α-synuclein, regulate PI3K/Akt and ERK1/2 signaling pathways, and improve mitochondrial function, all of which are highly relevant to the pathologies of AD and PD [29]. Beyond their direct activation of Nrf2, polyphenols exhibit a 'prebiotic-like' effect, meaning they selectively promote the growth of beneficial gut bacteria while suppressing potentially harmful ones. Curcumin has been shown to increase the abundance of Lactobacillus and Bifidobacterium while reducing pathobionts like Enterobacteriaceae. Resveratrol enriches Akkermansia muciniphila, a bacterium associated with improved gut barrier integrity. EGCG, meanwhile, tends to promote overall microbiota diversity [30]. This selective modulation of the gut ecosystem is critical because it facilitates the transformation of poorly absorbed polyphenols into smaller, more bioavailable, and often more bioactive metabolites by the gut microbiota. These metabolites, such as urolithins from ellagic acid and equol from daidzein, can cross the blood-brain barrier more efficiently than their parent compounds and exert direct neuroprotective effects [31], [32]. This intricate interplay highlights that the neuroprotective benefits of polyphenols are not solely derived from their direct actions but are significantly amplified by their ability to reshape the gut microbiome into a healthier, more functional community (Table 1). While curcumin, resveratrol, and EGCG all converge on Nrf2/ARE activation, their translational potential diverges substantially due to pharmacokinetic constraints and microbiome-dependent metabolism. Curcumin demonstrates potent in vitro Nrf2 upregulation and NF-κB suppression, yet its clinical efficacy remains inconsistent, largely attributable to < 1% oral bioavailability, rapid hepatic glucuronidation, and poor blood-brain barrier penetration. Resveratrol’s neuroprotective effects appear highly contingent on microbial conversion to urolithins, creating a well-documented “responder vs. non-responder” dichotomy that complicates dose-standardization in human trials. EGCG exhibits superior intestinal absorption and more reproducible enrichment of SCFA-producing taxa (e.g., Akkermansia, Faecalibacterium), though high-dose supplementation raises dose-dependent hepatotoxicity concerns. Critically, most preclinical studies administer isolated compounds at supraphysiological doses that are rarely achieved in human dietary patterns, limiting direct clinical extrapolation. These discrepancies underscore that polyphenol efficacy cannot be evaluated in isolation; rather, it must be contextualized within host-specific microbiome capacity for metabolite generation, formulation technology (e.g., nanoencapsulation, phospholipid complexes), and synergistic lifestyle co-factors.
The neuroprotective efficacy of dietary polyphenols is fundamentally constrained by inter-individual variability in microbial metabolism, a phenomenon that has contributed to inconsistent clinical outcomes. Human populations exhibit distinct “metabotypes” based on their capacity to convert parent polyphenols into bioactive derivatives. For ellagitannins, individuals are classified as urolithin metabotype A (high urolithin A producers), metabotype B (isourolithin A dominant), or metabotype 0 (non-producers). These phenotypes are dictated by the presence and abundance of specialized taxa, notably Gordonibacter pamelaeae and Ellagibacter isourolithinifaciens, which possess the enzymatic machinery for sequential dehydroxylation and lactonization. Similarly, only 30–50% of individuals possess the gut microbial consortia (e.g., Slackia isoflavoniconvertens, Adlercreutzia equolifaciens) required to convert daidzein into equol, a metabolite with significantly higher estrogen receptor affinity and Nrf2-activating potency than its parent compound. Metabotype status is shaped by baseline enterotype composition, long-term dietary fiber intake, prior antibiotic exposure, host UGT polymorphisms, and aging-related microbiome shifts. Consequently, administering fixed-dose polyphenol supplements to unstratified cohorts often yields null clinical results, as non-producers cannot generate therapeutically relevant metabolite concentrations. This variability underscores the necessity of pre-intervention metabotyping and the development of next-generation formulations that either deliver downstream metabolites directly (e.g., urolithin A, equol) or co-administer targeted prebiotics to enrich metabolizer-capable taxa. Recognizing polyphenol metabolism as a host-microbiome co-dependent process is essential for designing precision nutrition strategies in neurodegenerative disease.
Physical activity serves as a powerful non-pharmacological intervention capable of inducing profound changes in both the gut microbiota and the body's intrinsic antioxidant systems, converging on the Nrf2 signaling pathway to confer neuroprotection (Fig. 2).
Schematic representation of how lifestyle factors such as exercise and dietary polyphenols influence gut microbiota composition and metabolite production. Increased microbial diversity promotes the generation of neuroactive metabolites including indolepropionic acid (IPA), short‑chain fatty acids (SCFAs), and urolithin A which cross the intestinal barrier into the bloodstream. These circulating metabolites interact with the blood–brain barrier (BBB) and contribute to neuroprotective effects, including reduced neuroinflammation, enhanced mitochondrial resilience, and decreased oxidative stress.
Regular exercise acts as a mild physiological stressor, a phenomenon known as hormesis, wherein the transient production of reactive oxygen species (ROS) during physical exertion triggers adaptive responses that ultimately enhance cellular resilience [48]. This process involves the activation of key signaling pathways, including the Nrf2 pathway, leading to the upregulation of a wide array of antioxidant enzymes [49]. Animal studies provide clear evidence for this mechanism: vigorous and prolonged aerobic exercise has been shown to increase the protein content of Nrf2 in the hippocampus and heme oxygenase-1 (HO-1) in the cortex [49], [50]. In models of Parkinson's disease, the neuroprotective effects of treadmill exercise against neurotoxins were critically dependent on the Nrf2 pathway; when Nrf2 was experimentally knocked down, the protective benefit of exercise was completely abolished [51]. Further supporting this, forced treadmill exercise in a rotenone-induced rat model of PD led to a significant upregulation of Nrf2 mRNA expression in the striatum, along with increased levels of its downstream targets, NQO.1 and TFAM, which is essential for mitochondrial biogenesis [52]. These findings establish a robust mechanistic link between physical activity and the enhanced activation of the Nrf2 antioxidant defense system in the brain. Beyond its direct effect on Nrf2, physical activity exerts a transformative influence on the gut ecosystem, which is increasingly recognized as a key mediator of its systemic health benefits [53], [54]. A growing body of evidence reveals a reciprocal relationship between exercise and the intestinal microbiota, where physical activity enhances gut microbial diversity and fosters a more favorable microbial profile. Systematic reviews and meta-analyses have confirmed that moderate exercise positively impacts the gut microbiome in adults, often resulting in increased microbial diversity and enrichment of beneficial bacterial taxa. For example, studies in mice have shown that voluntary wheel running increases the abundance of Lactobacillus and Bifidobacterium. In humans, brisk walking has been associated with an increase in Bacteroides species in healthy elderly women [55]. Exercise has also been shown to reverse diabetes-induced dysbiosis in mouse models, decreasing the Firmicutes-to-Bacteroidetes ratio and increasing the abundance of butyrate-producing bacteria like Ruminococcaceae and Bacteroidales [56]. This remodeling of the gut microbiota is functionally significant, as it enhances the production of beneficial microbial metabolites, such as short-chain fatty acids (SCFAs). These SCFAs, particularly butyrate, are known to strengthen the intestinal barrier, reduce systemic inflammation, and, importantly, contribute to the activation of the Nrf2 signaling pathway in the brain [57]. Fecal microbiota transplantation (FMT) experiments have elegantly demonstrated this gut-brain connection; transferring the gut microbiota from exercised mice to sedentary recipients was sufficient to transfer the cognitive benefits, linking the exercise-induced changes in the microbiome directly to improved brain function. The type and intensity of exercise appear to be critical determinants of these outcomes. While some studies show broad benefits from voluntary activity, others indicate that forced exercise may be more effective at inducing neuroprotective changes, highlighting the need for careful consideration of exercise protocols [58]. Ultimately, physical activity emerges not just as a tool for improving fitness but as a fundamental lifestyle intervention that reshapes the gut microbiome and primes the brain's antioxidant machinery via the Nrf2 pathway (Table 2).
The reported effects of physical activity on the gut–brain–Nrf2 axis are highly heterogeneous, reflecting substantial variability in exercise modality, intensity, duration, and adherence. Voluntary wheel running consistently enriches microbial diversity and SCFA production in rodent models, whereas forced treadmill protocols, while sometimes yielding stronger acute Nrf2 upregulation, can induce stress-mediated dysbiosis if intensity or duration is excessive. Human studies further reveal pronounced inter-individual variability: baseline microbiome composition, habitual diet, APOE genotype, age, and metabolic health significantly modulate both microbial remodeling and Nrf2 responsiveness to identical exercise regimens. For instance, individuals with low baseline microbial diversity or elevated systemic inflammation often exhibit blunted Nrf2 activation post-exercise, suggesting that “one-size-fits-all” prescriptions may be insufficient for neuroprotective microbiome adaptation. Furthermore, conflicting reports on exercise intensity thresholds for optimal SCFA production and Nrf2 priming highlight the need for precision dosing frameworks that account for host microbiome enterotypes and physiological stress tolerance.
The mechanisms by which physical activity reshapes the gut microbiome extend beyond simple taxonomic shifts and involve coordinated physiological adaptations. First, regular exercise accelerates intestinal transit time, reducing colonic substrate retention and altering fermentation kinetics; this selective pressure favors fast-growing, SCFA-producing taxa while limiting the proliferation of slow-metabolizing pathobionts. Second, exercise modulates bile acid homeostasis by increasing hepatic synthesis and intestinal secretion, thereby altering the primary-to-secondary bile acid ratio. Bile acids act as potent signaling ligands for the farnesoid X receptor (FXR) and G protein-coupled bile acid receptor (TGR5) on enterocytes and immune cells, which in turn regulate gut barrier integrity, antimicrobial peptide secretion, and microbial community composition. Third, exercise induces systemic and mucosal immune remodeling: transient increases in circulating IL-6 (derived from contracting muscle) exert anti-inflammatory effects in the gut, enhance secretory IgA production, and promote regulatory T-cell differentiation, collectively fostering a microenvironment conducive to beneficial microbial colonization. Additionally, exercise-induced myokines (e.g., irisin, lactate) and improved splanchnic perfusion during recovery phases enhance intestinal oxygenation and mucosal repair, further stabilizing microbial ecology. These interconnected pathways motility, bile acid signaling, immune modulation, and hemodynamic adaptation provide a mechanistic foundation for understanding why exercise consistently enriches microbial diversity and SCFA output, ultimately priming the gut-brain axis for Nrf2-mediated neuroprotection.
The gut microbiota–metabolite axis represents a dynamic interface through which dietary and lifestyle factors exert systemic effects on host physiology. Polyphenols and physical activity regulate this axis at multiple levels. Polyphenols serve as substrates for microbial biotransformation, leading to the generation of smaller, bioactive compounds such as urolithins and phenolic acids, while simultaneously modulating the composition of the microbial community. In parallel, physical activity enhances microbial diversity and functional capacity, promoting the enrichment of metabolite-producing taxa, including short-chain fatty acid–producing bacteria. Through these complementary mechanisms, both interventions shape the quantity and profile of microbiota-derived metabolites, including short-chain fatty acids, indole derivatives, and polyphenol-derived metabolites. These metabolites act as key signaling molecules that can cross physiological barriers and activate intracellular pathways such as Nrf2, thereby linking gut microbial activity to host antioxidant defenses and neuroprotection (Fig. 3).
The communication between the gut microbiota and the central nervous system is heavily reliant on a class of bioactive molecules known as microbial-derived metabolites. These compounds, synthesized from dietary precursors by gut bacteria, serve as crucial signaling agents that traverse the bloodstream to exert far-reaching effects on brain health, with the Nrf2 pathway being a prominent target. Among the most extensively studied are short-chain fatty acids (SCFAs), indole derivatives, and urolithins, each contributing uniquely to neuroprotection. SCFAs, primarily acetate, propionate, and butyrate, are produced by the fermentation of dietary fiber by commensal bacteria like Faecalibacterium, Roseburia, and Akkermansia [57]. Butyrate, in particular, has been shown to play a multifaceted role in neuroprotection. It can cross the blood-brain barrier and act as a histone deacetylase (HDAC) inhibitor, leading to epigenetic modifications that enhance the expression of neurotrophic factors like brain-derived neurotrophic factor (BDNF) and promote synaptic plasticity [68]. Critically, butyrate also contributes to Nrf2 activation by inhibiting its cytoplasmic repressor, KEAP1, thereby stabilizing Nrf2 and facilitating its translocation to the nucleus to boost antioxidant gene transcription. Acetate administration has also been shown to restore Nrf2 signaling and related antioxidant defenses [57]. Polyphenol supplementation has been demonstrated to increase the production of butyrate and acetate, reinforcing the concept of a prebiotic effect that enhances SCFA synthesis [69].
In addition to SCFAs, indole-3-propionic acid (IPA) has emerged as a particularly potent neuroprotective metabolite. IPA is synthesized by certain gut bacteria, such as Clostridium sporogenes, from dietary tryptophan [70]. IPA is a powerful antioxidant and a direct activator of the Nrf2 pathway [71]. In a mouse model of cardiac injury, FMT from healthy donors conferred protection against oxidative stress, an effect that was entirely dependent on Nrf2, as it was lost in Nrf2 knockout mice. The study identified IPA as the key mediator, showing that IPA intervention promoted Nrf2 nuclear translocation and upregulated its downstream antioxidant targets, HO1 and NQO1 [71]. IPA's neuroprotective actions extend to the brain, where it has been shown to protect microglia from inflammatory activation, thus breaking the vicious cycle of gut inflammation, systemic inflammation, and neuroinflammation that is implicated in AD and PD Another important class of polyphenol-derived metabolites is urolithins, which are produced by the gut microbiota from ellagitannins found in foods like pomegranates and nuts [31]. Urolithin A (UA) has gained significant interest for its ability to induce mitophagy, the selective clearance of damaged mitochondria, a process vital for neuronal health. UA also functions as a direct activator of the Nrf2 pathway, leading to the upregulation of antioxidant genes [72], [73]. In APP/PS1 mouse models of AD, long-term treatment with UA significantly improved learning and memory, prevented neuronal apoptosis, and enhanced neurogenesis, effects attributed in part to its Nrf2-mediated antioxidant and anti-inflammatory actions [74], [75]. UA further exerts anti-inflammatory effects in the brain by inhibiting Cathepsin Z, an enzyme involved in lysosomal degradation and inflammation, an action that appears to be linked to its modulation of the Nrf2 pathway [75]. Together, these microbial metabolites illustrate a sophisticated communication network where diet and exercise shape the gut microbiome to produce signaling molecules that travel to the brain, activate the Nrf2 pathway, and orchestrate a multi-pronged defense against neurodegeneration.
The synergistic interplay between polyphenols, physical activity, and the gut microbiota, culminating in Nrf2 activation, has been substantiated in numerous preclinical models of both Alzheimer's and Parkinson's diseases, providing a strong mechanistic rationale for their combined use. In models of Alzheimer's disease (AD), characterized by Aβ plaque deposition, tau pathology, and neuroinflammation, these interventions demonstrate significant promise. Aerobic exercise, specifically a 20-week treadmill training program in APP/PS1 mice, was found to delay the onset of cognitive impairment, as measured by improved performance in Morris water maze and eight-arm maze tests. This cognitive benefit was accompanied by distinct alterations in the gut microbiota, including a reduction in the phylum Bacteroidetes and an increase in the genus Faecalibaculum, a producer of short-chain fatty acids (SCFAs) [76]. The neuroprotective effects of fecal microbiota transplantation (FMT) from healthy wild-type mice further solidified this gut-centric view; FMT in APP/PS1 mice improved cognition, restored gut microbial dysbiosis, increased SCFA levels, and suppressed the pro-inflammatory TLR4/NF-κB signaling pathway in the brain [77]. Polyphenols also show robust effects in AD models. Curcumin has been shown to reduce Aβ plaque deposition and elevate BDNF levels in animal models of AD. Furthermore, urolithin A (UA), a gut-microbiota-derived metabolite, significantly improved learning and memory in APP/PS1 mice, an effect linked to its ability to prevent neuronal apoptosis and enhance neurogenesis, processes influenced by Nrf2 activation [74]. The combination of resveratrol with high-intensity interval training (HIIT) in aged rats led to beneficial effects in counteracting aging and oxidative stress in the hippocampus, suggesting a potential synergy between the two interventions [21].
Similarly, in preclinical models of Parkinson's disease (PD), which is defined by the progressive loss of dopaminergic neurons and α-synuclein pathology, the tripartite pathway demonstrates clear neuroprotective efficacy. The neuroprotective effects of exercise against neurotoxins like MPTP and MPP+ in rodent models are critically dependent on the Nrf2 pathway; knocking down Nrf2 expression completely abrogated the protective effect of treadmill exercise on nigrostriatal dopaminergic neurons [51].
Forced treadmill exercise in a rotenone-induced rat model of PD resulted in significant transcriptional upregulation of Nrf2 and its downstream targets (NQO1 and TFAM) in the striatum, an effect that correlated with marked improvements in motor function and preservation of dopaminergic neurons [52]. While these mRNA changes indicate robust transcriptional engagement of the Nrf2 axis, it is important to note that transcript-level upregulation does not always equate to proportional increases in functional protein expression or enzymatic activity, highlighting the need for complementary proteomic and functional validation in future exercise-intervention studies. Polyphenols have also proven effective in these models. Curcumin has been shown to exert neuroprotective effects in both MPTP and rotenone-induced mouse models of PD [78]. EGCG treatment in a PINK1-mutant Drosophila model of PD reduced brain iron accumulation and oxidative stress markers, while in an MPTP-induced mouse model, it restored motor function and protected dopaminergic neurons [79]. Sodium butyrate, a microbial metabolite, was found to protect against α-synuclein pathology in both the colon and substantia nigra of a rotenone-induced PD mouse model, an effect associated with its ability to remodel the gut microbiota and increase levels of the gut-brain axis hormone GLP-1 [80]. These collective findings from animal models of AD and PD strongly support the hypothesis that targeting the gut microbiota with polyphenols and/or physical activity can activate the Nrf2 pathway, thereby providing a powerful, multi-faceted defense against the core pathological processes driving neurodegeneration. Collectively, these studies reveal several common themes, including the central role of gut microbiota modulation, increased production of neuroactive metabolites, and activation of antioxidant and anti-inflammatory pathways such as Nrf2. Evidence from fecal microbiota transplantation studies further supports a causal role of the gut microbiota in mediating these effects, reinforcing the concept that microbiota-targeted interventions may have therapeutic potential.
Despite the compelling mechanistic evidence and promising results from preclinical models, the translation of polyphenol-based and exercise-based interventions into effective clinical therapies for neurodegenerative diseases faces significant hurdles. The most substantial challenge lies in the poor oral bioavailability of many key polyphenols, particularly curcumin and resveratrol [81]. Following oral administration, these compounds exhibit very low water solubility, leading to minimal absorption (<1%) [82]. They are rapidly and extensively metabolized in the intestine and liver into glucuronide and sulfate conjugates and are quickly eliminated from the systemic circulation, resulting in plasma concentrations that are often too low to elicit a meaningful biological response at the target site in the brain [83]. This pharmacokinetic limitation is widely considered the primary reason for the "disappointing" results observed in human clinical trials, which often fail to replicate the robust neuroprotective effects seen in animal studies. Clinical trials investigating curcumin and resveratrol for AD and PD have yielded mixed or negative outcomes, with few allowing for definitive conclusions about their therapeutic potential [84]. The limited number of human studies, many of which are small, short-term, or use complex nutraceutical formulations rather than purified compounds, further complicates the interpretation of existing data [85], [86]. Moreover, the optimal dosage, formulation, and duration of treatment remain unclear, and host factors such as genetics (e.g., apolipoprotein E genotype) can modulate the efficacy of these compounds, adding another layer of complexity [29].
Addressing these translational challenges requires a multi-pronged approach focused on overcoming bioavailability issues and designing more rigorous clinical trials. A significant area of research is the development of advanced delivery systems designed to enhance the absorption, stability, and targeted delivery of polyphenols. Strategies include the use of adjuvants like piperine, which inhibits the metabolic enzymes responsible for glucuronidation, thereby increasing systemic exposure to curcumin. Other innovative approaches involve encapsulating polyphenols within nanoparticles, liposomes, phospholipid complexes, or micelles [82]. For example, formulations like Theracurmin® and Longvida® have demonstrated substantially higher bioavailability compared to conventional curcumin supplements [82]. These advanced formulations aim to deliver therapeutically relevant concentrations of the active compound to the brain, bridging the gap between preclinical success and clinical efficacy. Concurrently, future research must prioritize well-designed human clinical trials that investigate the synergistic effects of combining polyphenol supplementation with structured physical activity regimens. A Phase 2 clinical trial (NCT01811381) is already underway to evaluate the combined effects of curcumin supplementation and aerobic yoga in individuals with Mild Cognitive Impairment, representing a step in the right direction [82]. Such studies should employ standardized methodologies, long-term follow-up periods, and focus on intermediate biomarkers (e.g., gut microbiota composition, levels of microbial metabolites, Nrf2 pathway activity) alongside clinical endpoints to better elucidate the mechanisms of action in humans [87]. In conclusion, while the convergence of polyphenols and physical activity on the gut microbiota to activate the Nrf2 pathway presents a highly promising, integrative strategy for combating neurodegeneration, its full therapeutic potential can only be realized by overcoming the significant pharmacokinetic barriers of dietary compounds and validating this synergistic approach in carefully designed human trials.
Despite robust preclinical evidence, several methodological and biological gaps impede clinical translation. First, the majority of studies rely on cross-sectional microbiome snapshots rather than longitudinal tracking of metabolite flux, Nrf2 nuclear translocation dynamics, and cognitive/motor endpoints. Second, the precise microbial consortia responsible for converting dietary polyphenols into Nrf2-activating metabolites (e.g., urolithins, IPA, equol) remain incompletely characterized, and inter-individual variations in these taxa are rarely stratified in trial design. Third, the additive versus synergistic effects of combined polyphenol-exercise interventions remain untested in controlled human cohorts, with most studies evaluating either diet or exercise in isolation. Microbiome-informed polyphenol dosing, guided by baseline enterotype profiling and metagenomic capacity for metabolite conversion, will significantly improve Nrf2 activation and cognitive outcomes compared to standard fixed-dose regimens. Moderate-intensity, microbiome-tailored exercise regimens will synergize with polyphenol-derived metabolites to amplify Nrf2-driven mitochondrial biogenesis and synaptic resilience in prodromal AD/PD, with optimal effects observed when exercise timing aligns with peak postprandial polyphenol metabolite circulation. Fecal metabolite profiling (e.g., urolithin/SCFA ratios, IPA levels) will serve as a more reliable, dynamic biomarker of intervention efficacy than parent compound plasma concentrations, enabling real-time adjustment of lifestyle prescriptions. Addressing these questions through multi-omics longitudinal trials, coupled with standardized exercise and polyphenol formulation protocols, will be essential to transition this integrative framework from mechanistic plausibility to precision neurotherapeutics.
