Effects of Plant Polyphenols on Obesity-Induced Inflammatory Depression: A Review

1. Introduction

According to the World Health Organization (WHO), 2.5 billion adults (18 years and older) are overweight globally in 2022, with 890 million of them suffering from obesity. Furthermore, WHO estimates that by 2035, there will be 1.9 billion obese people globally, with 1 in 4 persons on average being obese (World Health Organization (WHO), 2022).

Obesity substantially increases the risk of multiple chronic conditions, including cancer (Lamabadusuriya et al., 2025), diabetes (Milito et al., 2026), depression (Monsalve et al., 2025), respiratory syndrome (Zhao et al., 2026) and cardiovascular disease (Wang et al., 2025). In recent years, the association between obesity and depression has attracted growing attention. Meta-analyses indicate that obesity increases the risk of depression by 65% in older adults and 32% in the general adult population (Guo et al., 2023; Pereira-Miranda et al., 2017). Among children and adolescents, the prevalence of depression is 1.34 times higher in those with obesity than in their normal-weight peers. (Quek et al., 2017). Clinically, obesity-related depression often presents with atypical features, such as increased appetite or weight gain, hypersomnia, leaden paralysis, and heightened sensitivity to interpersonal rejection—symptoms that differ from the appetite loss and insomnia commonly observed in melancholic depression (Lamers et al., 2018). Epidemiological and clinical studies suggest that obesity-induced depression is closely linked to a chronic low-grade inflammatory state driven by immune activation, with neuroinflammation emerging as a central mechanism (Fulton et al., 2022). Current antidepressant therapies primarily target monoamine neurotransmitters, including serotonin (5-HT), norepinephrine (NE), and dopamine (DA). However, these treatments often show limited efficacy in patients with obesity-related depression (Kloiber et al., 2007), suggesting that mechanisms of obesity-related depression are different from those of neurotransmitter-mediated depression. Therefore, it is particularly important to explore in depth the relationship between obesity, neuroinflammation and depression.

Polyphenols, a diverse group of plant-derived bioactive compounds, have attracted considerable interest due to their anti-inflammatory, antioxidant, and neuroprotective properties (Koca et al., 2025). They modulate immune cell activity, regulate the expression of pro-inflammatory cytokines, and influence multiple signaling pathways implicated in depression (Mijailović et al., 2025). Although numerous reviews have examined the role of polyphenols in stress-induced depression, relatively few have focused specifically on obesity-associated inflammatory depression. Moreover, the effective doses required to achieve therapeutic benefits in this context remain unclear. Accordingly, this review synthesizes current evidence on the mechanistic interplay between obesity, neuroinflammation, and depression. It further evaluates the therapeutic potential and effective dosing of polyphenols in modulating inflammatory pathways and mitigating obesity-induced inflammatory depression. This review aims to offer new sights into dietary interventions for mood disorders, particularly obesity-induced inflammatory depression.

2. Causes of Obesity-Induced Neuroinflammation

Microglia are resident immune cells in the central nervous system (CNS) and play an important role in neuroinflammation. Under normal physiological conditions, microglia exist in a resting state. In response to tissue injury, pathogens, or metabolic stress, they become activated and adopt an amoeboid morphology (Cancela et al., 2026). Activated microglia are broadly categorized into a pro-inflammatory (M1-like) phenotype and an anti-inflammatory or reparative (M2-like) phenotype. M1-like microglia secrete pro-inflammatory cytokines that contribute to neuronal dysfunction, whereas M2-like microglia promote debris clearance, neuronal protection, and tissue repair (Jia et al., 2021). Several studies have demonstrated that obese patients show significant activation of microglia in the hippocampus and hypothalamic sites, accompanied by elevated inflammatory cytokines (Cope et al., 2018; Miller & Spencer, 2014). These findings suggest that obesity functions as a chronic immunometabolic stressor that skews microglia toward a sustained pro-inflammatory phenotype. In this section, we explore the main pathways by which obesity activates microglia and summarize the relevant mechanisms that may be involved (Fig. 1).

2.1. Peripheral Inflammation and Blood-Brain Barrier Disruption

Obesity is known to be associated with inflammation of peripheral tissues due to altering the gut microbiota and thus elevating circulating levels of lipopolysaccharide (LPS) produced by gram-negative bacteria (Monsalve et al., 2025). Concurrently, the rapid expansion of adipose tissue in obesity induces local hypoxia, adipocyte death, and activation of hypoxia-inducible factor-1α (HIF-1α), which in turn promotes nod-like receptor family pyrin domain containing 3 (NLRP3) inflammasome activation (Gonzalez et al., 2019). These processes drive the release of pro-inflammatory cytokines and stimulate hepatic production of C-reactive protein (CRP) (Stanimirovic et al., 2022). Large cohort studies have demonstrated that depressed patients have higher CRP concentrations, which makes elevated CRP levels one of the best predictors of the onset of depression in obesity (Hickman et al., 2014). The blood-brain barrier (BBB) is the main pathway through which peripheral inflammatory and immune responses interact with the brain microenvironment. It has been shown that peripheral inflammatory cytokines can down-regulate the expression of tight junction proteins, such as claudin-5 and zonula occludens-1 (ZO-1), thus disrupting tight junctions between neighboring cells and leading to increased BBB permeability (Huang et al., 2021; Wu et al., 2020). Furthermore, it is discovered that LPS inhibites P-glycoprotein activity and induces the secretion of matrix metalloproteinases, which resultes in endoplasmic reticulum stress, mitochondrial damage, and apoptosis in endothelial cells (Bernhart et al., 2018). Impaired BBB integrity caused by endothelial cell apoptosis further exacerbates the infiltration of peripheral inflammatory cytokines in brain tissue. The continuous influx of peripheral inflammatory cytokines into the CNS stimulates microglia over time, leading them to adopt a pro-inflammatory phenotype and trigger neuroinflammation. It has been shown that the mechanism by which peripheral inflammation activates microglia is that inflammatory cytokines, such as interleukin-1β (IL-1β), stimulate the release of the microglia damage-associated molecular pattern (DAMP) molecule adenosine triphosphate (ATP), and then ATP-mediated Ca²⁺ signaling facilitates the assembly of the NLRP3 inflammasome, thus triggering a series of inflammatory cytokine production (Garaschuk, 2021). Furthermore, Toll-like receptor 4 (TLR4) on the surface of microglia can be bound by LPS, saturated fatty acids (SFA), and adipokine resistin, which might intensify neuroinflammation by starting an inflammatory cascade (An et al., 2020; Benomar & Taouis, 2019; Lu et al., 2021).

2.2. Dietary Saturated Fatty Acids as Direct Microglial Activators

Deliciousness and high energy density are the main features of the diets of obese people. These diets are rich in high saturated fats (e.g., lard or palm oil) that significantly elevate SFA levels in the blood (Nakajima et al., 2020). It is shown that consumption of saturated fats (Lai et al., 2016) and serum concentrations of the SFA palmitic acid (PA) (Tsuboi et al., 2013) are positively associated with depressive symptoms and serum levels of the peripheral inflammatory marker CRP. Indeed, SFA can cross the BBB via transporter- and receptor-mediated mechanisms, allowing them to directly interact with microglia (Sanchez-Cano et al., 2021). Zhuang et al. (2022) demonstrate that high-fat diet (HFD) can lead to PA deposition in the hippocampus of mice and lead to activation of microglia, as evidenced by a smaller microglia area and an increased immunological response to cluster of differentiation 68 (CD68). In vitro, PA and stearic acid (SA) cause BV2 microglia to produce pro-inflammatory cytokines through the TLR4/NF-ĸB pathway, which impaires neurogenesis (Wang et al., 2012). In addition, Beaulieu et al. (2021) demonstrate that PA can activate N9 microglia to trigger neuroinflammation via mediating the nuclear factor kappa-B (NF-ĸB) pathway and phosphorylation of the c-Jun. These findings indicate that dietary SFAs not only exacerbate systemic inflammation but also directly trigger neuroinflammatory signaling within the CNS.

2.3. Lipid Droplet Accumulation in Microglia

Recent evidence highlights lipid droplet (LD) accumulation in microglia as an emerging mechanism linking metabolic stress to neuroinflammation. HFD exposure during adolescence induces emotional behavioral abnormalities in mice and promotes LD accumulation in hippocampal microglia. Importantly, LD burden correlates positively with microglial activation status (Yao et al., 2024). Similar lipid-laden microglia have been identified in the aging brain and are considered representative of dysfunction and pro-inflammatory states in the aging brain (Marschallinger et al., 2020). In vivo and in vitro studies show that innate inflammation induces de novo synthesis of LDs in microglia through activation of the TLR4 signaling pathway (Khatchadourian et al., 2012; Marschallinger et al., 2020). Perilipin-2 (PLIN 2), located on the surface of LDs, can bind to the inflammation amplifying factor triggering receptor expressed on myeloid cells-1 (TREM 1), leading to an abnormal accumulation of TREM 1 inside microglia LDs. TREM 1 subsequently promotes neuroinflammation through the NLRP3 inflammasome pathway, and the elevated pro-inflammatory cytokines in turn activate the TLR4 signaling pathway to induce LD formation, thus constituting a vicious cycle (Li et al., 2023). This explains the long lasting neuroinflammatory state in the brain of obese patients (Henn et al., 2022). Interventions targeting LD metabolism support a causal role for LD accumulation in neuroinflammation. According to Khatchadourian et al. (2012), inhibition of long-chain acyl-CoA synthetase with triacsin C suppresses LD formation and reduces reactive oxygen species and cytokine production in microglia. Similar findings are reported by Li et al. (Li et al., 2023). They find that high glucose can induce lipophagy damage in microglia, leading to the LDs accumulation, and triggering an inflammatory cascade response. In contrast, rapamycin, an autophagy activator, is successful in reversing the defective lipophagy in microglia caused by high glucose, reducing LDs accumulation and inflammation production. These findings suggest that intracellular LD accumulation represents a critical metabolic driver of sustained microglial activation in obesity.

3. Pathways of Neuroinflammation Triggering Depression

3.1. Abnormalities of the Hypothalamic-Pituitary-Adrenal Axis

Neuroinflammation is closely linked to dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis, a central stress-response system that is frequently hyperactivated in obesity (Dymek et al., 2025). Pro-inflammatory cytokines released by activated microglia modulate HPA axis activity, with IL-1β acting as a primary driver. IL-1β stimulates the release of corticotropin-releasing hormone from the hypothalamus and adrenocorticotropic hormone from the pituitary gland, ultimately increasing glucocorticoid (GC) secretion—cortisol in humans and corticosterone in rodents (Dionysopoulou et al., 2021; Fu et al., 2023) (Figure 2). GC is the ultimate effector of the HPA axis, and elevated levels of GC are considered a biological risk factor for depression (Cheiran Pereira et al., 2022). Notably, the relationship between the HPA axis and obesity appears to be bidirectional. Exposure to high levels of cortisol can induce obesity through a variety of pathways: (a) increased appetite with a preference for energy-dense food; (b) stimulation of adipogenesis and adipocyte hypertrophy, especially in visceral fat; (c) inhibition of thermogenesis in brown adipose tissue with relative reduction of energy expenditure (Milano et al., 2020). It is obvious that there is a cycle between obesity and high levels of cortisol. This cycle increases the susceptibility of obesity to depression. Elevated corticosterone can decrease immature neuron activation and may have long-term effects on the integration or function of cells (Workman et al., 2015). In db/db mice, increased corticosterone suppresses hippocampal brain-derived neurotrophic factor (BDNF) expression via activation of glucocorticoid receptors (Wosiski-Kuhn et al., 2014). In contrast, lentiviral suppression of the glucocorticoid receptor expression restores BDNF expression and rescues hippocampal neurogenesis in db/db mice. Therefore, obesity-related depression may be associated with impaired hippocampal neurogenesis induced by elevated corticosterone.

3.2. Defects in Adult Hippocampal Neurogenesis

Rodent studies demonstrate that obesity-induced neuroinflammation inhibits the proliferation, differentiation and survival of newborn cells in the hippocampal dentate gyrus, which in turn prevents the production of newborn neurons (Baghdadchi et al., 2026; Kwon et al., 2026) (Figure 2). Furthermore, HFD-fed obese mice exhibit significantly fewer DCX (a marker of newly generated neurons within the last 2-3 weeks) and Ki67 (a proliferating cell marker) positive cells in the hippocampus, indicating reduced neurogenesis. These structural deficits are accompanied by depression-like behaviors, including reduced sucrose preference and altered exploratory behavior, supporting a functional link between impaired neurogenesis and mood disturbances (Yao et al., 2022). It is reported that pro-inflammatory cytokines (IL-6, IL-1β and TNF-α) appear to inhibit adult hippocampal neurogenesis by activating the signal transducer and activator of transcription 3 (STAT3) pathway in hippocampal neural progenitor cells (Chen et al., 2013; Kong et al., 2019), whereas interferon ɑ and interferon γ inhibit adult hippocampal neurogenesis by activating the STAT1 pathway (Borsini et al., 2018; Zhang et al., 2020). Meanwhile, in a model of LPS-induced neuroinflammation, pharmacological attenuation of microglial activation restores hippocampal neurogenesis (Domínguez-Rivas et al., 2021). Excessive microglial activation not only releases inflammatory mediators but also increases phagocytosis of newborn neurons (Domínguez-Rivas et al., 2021). In HFD-fed mice, hyperactivated microglia excessively engulf immature neurons, further impairing neurogenesis (Yao et al., 2022). Treatment with anti-inflammatory agents such as minocycline reduces microglial overactivation, rescues neurogenesis, and alleviates depressive-like behaviors (Bassett et al., 2021). These findings indicate that neuroinflammation-driven disruption of hippocampal neurogenesis represents a key mechanism linking obesity to depression

3.3. Resistance of Central Insulin and Leptin

Central insulin and leptin resistance is another mechanism connecting metabolic dysfunction to depressive symptoms. Large cohort studies demonstrate that reduced insulin sensitivity correlates with increased depression and anxiety symptoms in individuals with obesity (Perry et al., 2020). Preclinical studies also support this concept because it has been repeatedly demonstrated that HFD-induced insulin resistance in mice or rats leads to anxiety and depression-like behaviors (Hassan et al., 2019). Moreover, it has been demonstrated that knockdown of insulin receptor (InsR) in astrocytes also promotes the development of anxiety and depression-like behaviors in mice (Cai et al., 2018). These studies highlight the critical role that central insulin resistance plays in the association between obesity and depression. It has been shown that inflammatory cytokines in the brain can trigger hypothalamic insulin resistance by impairing insulin-dependent phosphorylation and downstream signaling at InsR via the IKKβ/NF-ĸB signaling pathway (Benzler et al., 2015), the JNK signaling pathway (Feng et al., 2020) and the TLR4 signaling pathway (Benomar & Taouis, 2019) (Figure 2). In rodents, HFD intake is considered an important factor in obesity-induced insulin resistance. Dietary SFA can activate TLR4/MyD88 signaling to enhance endoplasmic reticulum stress and hypothalamic insulin resistance (Li et al., 2020). Leptin is a receptor signal from adipose tissue to the brain that conveys energy storage. It regulates the body's energy balance and exerts antidepressant-like effects by acting on the leptin receptor (LepR) in the hypothalamus (Peng et al., 2021). Studies have shown that obesity-induced inflammation can inhibit the expression of leptin transporter proteins in the BBB, leading to a reduced capacity of the leptin transport system. This prevents high levels of leptin in the periphery from entering the brain to exert antidepressant activity (Sáinz et al., 2015). Meanwhile, obesity-induced inflammation can also impair hippocampal leptin receptor activity, resulting in the inability of leptin to activate downstream signals to exert antidepressant effect. Yamada et al. (2011) demonstrate that diet-induced obese mice with impaired hippocampal leptin receptor activity fail to respond to leptin administration in terms of the forced swimming test (FST) and the biochemical changes in the hippocampus. These studies suggest that inflammation-driven insulin and leptin resistance disrupt central metabolic signaling and neurotrophic support, thereby contributing to the development of obesity-associated depression.

4. Mechanisms of Polyphenols Ameliorating Obesity-Induced Inflammatory Depression

Polyphenols are one of the most abundant and widely distributed natural products in plants. Although their structures vary widely, they all have aromatic rings with one or more hydroxyl substituents (Mehra & Mittal, 2026). Phenolic compounds are traditionally divided into two groups: flavonoids and non-flavonoids (Figure 3). In chemical structure, flavonoids have three rings (C6-C3-C6) as their basic skeleton (labeled A, B, and C in Figure 3) (Shen et al., 2022). Based on the number of phenolic rings and the structural components linking them, flavonoids can be subdivided into several subclasses, including flavanols, flavones, flavonols, anthocyanidins, flavanones, isoflavones, and chalcones (Bessa et al., 2021). Non-flavonoid polyphenols primarily include phenolic acids, stilbenes, lignans, and hydrolysable tannins. Among them, phenolic acids are the most significant non-flavonoid polyphenols and can be further classified into hydroxybenzoic and hydroxycinnamic acids (Wang et al., 2022). Growing evidence indicates that polyphenols exert antidepressant and neuroprotective effects (Fekete et al., 2025). The subsequent sections explore how phenolic compounds alleviate inflammatory depression via the specific signaling pathways (Figure 4).

4.1. Modulation of Hippocampal NF-ĸB Pathway

Rodent studies demonstrate that HFD feeding increases hippocampal expression of NF-κB-related proteins (p-IKKα/β, p-IκBα, p65, p-p65) and is accompanied by a marked prolongation of immobility time in both the tail suspension test and the forced swimming test (Xie et al., 2024). Furthermore, LPS, TNF-α, and PA are all found to activate the microglial NF-κB signaling pathway and trigger neuroinflammation in cellular models (Brás et al., 2020; Evans et al., 2023). These studies confirm the important role of the NF-κB pathway in obesity-induced neuroinflammation. Polyphenols can ameliorate obesity-induced depressive-like behaviors by modulating the expression of the NF-κB signaling pathway in microglia. Mechanistically, this suppression is associated with reduced HPA axis hyperactivity, restoration of hippocampal BDNF expression, and normalization of serotonin metabolism. For example, apple polyphenols inhibit NF-κB-related gene expression in hippocampal microglia, reduce circulating corticosterone and adrenocorticotropic hormone levels, and alleviate depressive-like behaviors in obese mice (Xie et al., 2024). Baicalein (Liu et al., 2022) and salidroside (Zhu et al., 2015) enhance hippocampal BDNF expression through NF-κB inhibition, thereby improving neurogenesis and synaptic plasticity associated with depression. Fisetin suppresses NF-κB activation and indoleamine 2,3-dioxygenase-1 (IDO-1) expression, restoring serotonin synthesis pathways and exerting antidepressant effects (Choubey et al., 2019).

4.2. Modulation of MAPK Pathway in CNS

Mitogen-activated protein kinase (MAPK) is a group of evolutionarily conserved serine-threonine kinases. It contains four classical pathways: ERK1/2, p38, JNK1/2/3, and ERK5 (Behl et al., 2022). HFD and high-fat/high-fructose diet (HFFD) activate JNK and p38 pathways in the brain, leading to neuroinflammation and cognitive impairment (Li et al., 2018; Lu et al., 2021). In LPS-induced depression models, pharmacological inhibition of p38 restores inflammatory markers and ameliorates depressive-like behaviors (Y.-W. Zhao et al., 2018). This result reveals the fact that inhibition of the MAPK pathway can ameliorate inflammatory depression. Activation of the MAPK pathway is a cascading phosphorylation process, and polyphenols can act at various steps in the activation process. Phenolics compounds such as quercitrin (Sun et al., 2021), baicalin (Li et al., 2022), sinapic acid (Huang et al., 2023) and sophoraflavanone G (Guo et al., 2016) all down-regulate the phosphorylation of JNK, p38 and ERK, thereby inhibiting the MAPK signaling pathway and exerting anti-neuroinflammatory effects. In LPS-activated BV2 microglial cells, fisetin reduces the release of inflammatory cytokines, inducible nitric oxide synthase (iNOS) and cyclooxygenase (COX)-2, by inhibiting phosphorylation of ERK but not JNK (Cho et al., 2013). This suggests that the inhibition of the MAPK pathway by polyphenols is specific. Certain polyphenols simultaneously target MAPK and NF-κB pathways. For instance, isorhamnetin attenuates HFFD-induced neuroinflammation by inhibiting the activation of JNK, p38, and NF-ĸB signaling pathways (Mulati et al., 2021). Punicalagin similarly reduces phosphorylation of ERK, JNK, and p38 while inhibiting NF-κB activity. (Lo et al., 2022). Additionally, similar to the findings in the NF-ĸB pathway, the modulation of the MAPK pathway by polyphenols can promote the expression of the CREB/BDNF pathway, which ameliorates the depressive-like behaviors triggered by neuroinflammation (Sun et al., 2021).

4.3. Modulation of Hippocampal PI3K/Akt Pathway

The phosphatidylinositol-3-kinase (PI3K)/protein kinase B (Akt) pathway integrates metabolic and inflammatory signaling and is closely associated with obesity-related depression. Under normal physiological conditions, insulin activates insulin receptor substrates (IRS), initiating PI3K/Akt signaling to support neuronal survival and plasticity (Savova et al., 2023). But in obesity, insulin resistance reduces the activation of the IRS/PI3K/Akt signaling pathway, which exacerbates the risk of obesity suffering from depression (De Paiva et al., 2024). It is shown that genistein (Li et al., 2020) and ferulic acid (Wang et al., 2017) can exert neuroprotective effects in obese diabetic mice and rats by modulating the levels of p-IRS, p-Akt, and p-GSK3β. In addition, epigallocatechin gallate (EGCG) (Mi et al., 2017) and sea-buckthorn flavonoids (Mulati et al., 2020) can significantly ameliorate HFFD-induced insulin resistance and neuronal damage in obese mice by up-regulating IRS/PI3K/Akt and ERK/CREB/BDNF signaling pathways. In inflammatory depression models, PI3K/Akt signaling also interacts with NF-κB. Intraperitoneal injection of LPS activates the hippocampal PI3K/Akt/NF-ĸB and MEK/ERK signaling pathways, which triggers neuroinflammation and impairs hippocampal plasticity. Curcumin (Cianciulli et al., 2016), EGCG (Payne et al., 2023), and eriodictyol (He et al., 2018) can attenuate LPS-induced pro-inflammatory cytokines release in BV2 microglial cells by inhibiting the PI3K/AKT/NF-ĸB pathway in a dose-dependent manner, and the effect of curcumin is comparable to that of the PI3K inhibitor LY294002. However, due to the low solubility of curcumin in plasma and the high rate of degradation in acidic, alkaline and oxidative environments, clinical studies have not yielded good therapeutic results (Parikh et al., 2008).

4.4. Modulation of Other Pathways

In addition to the previously mentioned NF-ĸB, MAPK, and PI3K/Akt pathways, polyphenols can also ameliorate obesity-induced inflammatory depression through specific pathways. Gut microbial composition has been suggested as a potential contributor to the neurobehavioral abnormalities associated with HFD consumption (Paiva et al., 2024). EGCG can restore HFD-induced gut dysbiosis and exert antidepressant effects through the gut-brain axis. Specifically, EGCG can ameliorate HFD-induced intestinal inflammation by increasing the abundance of Muribaculaceae and Alloprevotella, which prevents SCFAs, PPAR-γ, and NF-ĸB secreted by the intestine from entering the CNS through the circulation of blood to affect the expression of hypothalamic neurotransmitters (Zhou et al., 2023). By modulating the gut-brain axis, (-)-epicatechin (EC) also ameliorates obesity-induced depression-like behaviors. However, the efficacy of EC is mainly achieved by increasing the abundance of Lactobacillus and Enterobacter (Kang et al., 2022). Lactobacillus has been known to play a role in HFD-induced depression. It can exert antidepressant effects by modulating hippocampal γ-aminobutyric acid (GABA) expression and HPA axis activity via the vagus nerve (Bravo et al., 2011). Additionally, blueberry polyphenol extract inhibits LPS-induced neuroinflammation and depressive-like behaviors in mice through modulation of brain enzymes activities, such as monoamine oxidase-A (MAO-A), acetylcholinesterase (AChE) and Na⁺/K⁺-ATPase (Spohr et al., 2023). This finding is further confirmed in a study conducted by Luduvico et al (2022), who discover that tannic acid can exert neuroprotective and antidepressant effects by modulating brain Na⁺/K⁺-ATPase and Ca²⁺-ATPase activities. However, the modulating effect of polyphenols on brain enzyme activities has only been reported in the LPS-induced depression model, and still needs to be further explored in the HFD-induced depression model.

5. Effective Doses of Polyphenols in Ameliorating Obesity-Induced Inflammatory Depression

Although numerous studies have confirmed the anti-inflammatory and neuroprotective potential of polyphenols, their effective doses in ameliorating obesity-related depression remains unclear. The dose-response relationship is a core parameter for evaluating the pharmacological activity of polyphenols and guiding the development of dietary supplements. Determining the effective doses of polyphenols not only help to explain the discrepancies in results across different studies but also provide a theoretical basis for future personalized and precision interventions (Table 1).

5.1. In Vitro Model

In in vitro models, the ameliorative effects of polyphenols on obesity-related depression are primarily evaluated by simulating neuroinflammation, such as by stimulating microglia with LPS or PA. As shown in Table 1, the effective doses of different polyphenolic compounds in in vitro models vary significantly, and this variation is likely closely related to their chemical structures. For example, the flavonoid baicalin (Li et al., 2022) and its aglycone baicalein (Ran et al., 2021) exhibit different inhibitory potencies against LPS-induced IL-1β release in BV2 microglia. Baicalin significantly reduced IL-1β release at 22.5 μM, whereas baicalein required 45 μM to achieve comparable effects. This result revealed an intrinsic relationship between the activity of flavonoids and their structural characteristics, namely that glycosylation (forming glycosides) may enhance the biological activity of the molecule. Further analysis of the effective doses of polyphenols reveals that the effective doses of polyphenol extracts are generally higher, such as blueberry polyphenol extracts (Carey et al., 2013). This is likely due to their complex composition and the relatively low content of active ingredients. Regarding dose-effect relationships, most studies confirmed that polyphenols inhibited the production of pro-inflammatory factors in microglia in a dose-dependent manner. For instance, sinapic acid (Huang et al., 2023), within the range of 5-20 μM, showed increasing inhibitory effects on iNOS and IL-6 expression as the concentration rised. However, some polyphenols exhibit hormetic effects, characterized by beneficial actions at low concentrations and cytotoxicity at higher doses. EGCG significantly inhibited BV-2 cell proliferation at 175-300 μM (Payne et al., 2023) and hesperetin reduced microglial viability at 200 μM (Evans et al., 2023). These results indicate that the biological effects of polyphenols are not simply linear and may shift from protective to toxic once a certain concentration threshold is exceeded.

5.2. In Vivo Model

The ameliorative effects of polyphenols on obesity-related depression have been validated in various rodent models, including high-fat/high-sucrose diet induced depression and diabetes-associated depression. The effective administration doses, intervention durations, and primary mechanisms of action vary among different polyphenolic compounds. Similar to the patterns observed in in vitro models, the effective doses of different polyphenols in animal models exhibit a wide range. Generally, crude extracts (such as apple polyphenol extract and sea-buckthorn flavonoids) require relatively higher effective doses, typically between 120-500 mg/kg/d, which is attributed to their complex composition and relatively low content of active ingredients (Mulati et al., 2020). In contrast, purified monomeric compounds (such as EGCG and hydroxytyrosol) have relatively lower effective doses, mostly in the range of 25-100 mg/kg/d, benefiting from their defined chemical structures and higher target affinity (Liu et al., 2024; Zhou et al., 2023). Regarding the dose-effect relationship, multiple studies confirmed that polyphenols ameliorated obesity-related depressive-like behaviors and molecular biomarkers in a dose-dependent manner. In an HFD-induced obese mouse model, 20 mg/kg EC produced greater improvements in open-field behavior, hippocampal BDNF expression, and glucocorticoid receptor regulation compared with 2 mg/kg (Kang et al., 2022). Similarly, hydroxytyrosol at 100 mg/kg was more effective than 25 mg/kg in improving cognition and reducing neuroinflammation (Liu et al., 2024). These findings not only confirm the dose-dependent effects of polyphenols but also provide important references for determining the optimal effective dosage window. It is noteworthy that the administration method and intervention duration also significantly impact the effective dose of polyphenols. Currently, most studies employ oral gavage, with intervention periods ranging from 4 to 14 weeks, a timeframe typically aligned with the duration of diet-induced obesity. For instance, continuous intragastric administration of apple polyphenol extract (500 mg/kg/d) for 8 weeks significantly ameliorated depressive-like behaviors in obese mice induced by a high-sucrose diet (Xie et al., 2024). In contrast, sea-buckthorn flavonoids required a longer intervention period of 14 weeks (administered via diet incorporation at 0.06%-0.31% w/w) to exert significant protective effects in a model of cognitive impairment induced by a high-fat and high-fructose diet (Mulati et al., 2020). Furthermore, the impact of administration timing (preventive versus therapeutic administration) on effective dose warrants attention. Theoretically, therapeutic administration may require higher doses or longer intervention periods to achieve effects comparable to preventive administration. However, systematic studies directly comparing the dosage requirements for these two administration timings are still relatively limited. Future research needs to further clarify the optimal dose windows for different administration timings.

6. Conclusion and Future Perspectives

With the rapid socio-economic development and the drastic evolution of dietary patterns, overweight, obesity and related metabolic problems caused by unbalanced diets have emerged. The incidence of depression triggered by obesity is also increasing annually. Due to the lack of response to conventional antidepressants in obese depressed patients, exploring a more targeted approach to ameliorate obesity-induced depression is particularly important. This review elucidated the pathophysiological mechanisms by which obesity triggers neuroinflammation and thus depression. Meanwhile, we summarized the findings from both in vivo and in vitro studies on the ameliorative effects of polyphenols against obesity-related depression, as well as their effective doses. So, given their significant anti-neuroinflammatory properties, polyphenols could be a promising therapeutic for treating obesity-induced depression.

However, the detailed and more targeted of obesity-induced depression are needed to be explored in the future. More work on the difference of obesity-induced depression and conventional depression are also needed. In addition, although polyphenols are reported to have significant anti-neuroinflammatory effects, direct evidence of polyphenols on ameliorating obesity-induced depression are limited. Meanwhile, since neurological disorders are often accompanied by the co-involvement of different brain regions and cells, in vitro studies employing polyphenols to intervene in microglia are no longer sufficient for the exploration of specific neurological diseases. In recent years, brain organoids have been a popular area of study. The integration of microglia with different neuronal cell types can be used to construct specific neurological disease models (Sabate-Soler et al., 2024). Hence, it may be more accurate to explore the therapeutic mechanism of polyphenols for inflammatory depression by constructing the microglia containing brain organoids. Last but not least, although animal and cellular experiments have assessed the antidepressant efficacy of phenolic compounds, clinical studies have not yet demonstrated the ameliorative effect of polyphenols in patients with obesity-induced depression. Therefore, more comprehensive clinical studies are still needed to verify the antidepressant efficacy of polyphenols.: