Citrus Flavonoids as Selective Aryl Hydrocarbon Receptor Modulators in Obesity: Transcriptomic and Docking-Based Insights

1. Introduction

The aryl hydrocarbon receptor is a ligand-activated transcription factor that plays a pivotal role in xenobiotic metabolism and immune responses [1,2]. The aryl hydrocarbon receptor, a member of the basic helix-loop-helix/Per-Arnt-Sim (bHLH/PAS) family, functions as a sensor, integrator, and transcription factor [3,4]. Among transcriptional regulators involved in this immune-metabolic imbalance, the aryl hydrocarbon receptor (AhR) has emerged as a key integrator of environmental and endogenous signals [5,6]. Its activation by diverse ligands, such as environmental toxins, endogenous metabolites, and microbial products, can lead to pro- or anti-inflammatory outcomes depending on the specific tissue context and ligand characteristics [5,7]. Notably, aberrant AhR signaling links to obesity-induced inflammation and metabolic dysfunction [2,8].

Obesity, a multifactorial metabolic disorder, has become a major global health concern. Beyond the accumulation of excess adipose tissue, it is characterized by systemic low-grade inflammation and profound metabolic dysregulation [9,10,11]. Obesity increases the risk of insulin resistance, type 2 diabetes, cardiovascular disease, nonalcoholic fatty liver disease, and certain cancers [9,12]. These widespread effects are driven by immune and hormonal imbalances, including altered adipokines signaling and immune cell activation, which impact multiple organs [10,11,12,13]. Adipose tissue functions as both an energy reservoir and an endocrine organ that secretes adipokines such as leptin, adiponectin, and resistin, which regulate appetite, insulin sensitivity, and immune responses [14,15,16]. In obesity, adipocyte hypertrophy triggers a shift in macrophage polarization from anti-inflammatory M2 to pro-inflammatory M1 phenotypes, promoting secretion of cytokines such as TNF-α, IL-6, and MCP-1 [17,18]. This pro-inflammatory microenvironment exacerbates systemic insulin resistance and disrupts lipid and glucose homeostasis.

Recent evidence suggests that certain dietary flavonoids, particularly those derived from citrus fruits, may modulate AhR activity [19,20]. Flavonoids such as hesperidin (HES), narirutin (NAR), nobiletin (NOB), sinensetin (SIN), and tangeretin (TAN) have demonstrated antioxidant, anti-inflammatory, and metabolic regulatory properties in preclinical studies [21,22]. However, the mechanisms by which these flavonoids interact with AhR and influence gene networks related to adipocyte inflammation and metabolism remain incompletely understood.

To address this gap, we investigated the regulatory effects of citrus flavonoids on AhR signaling in a cellular model of adipogenesis and inflammation. Transcriptomic profiling via RNA sequencing was used to characterize gene expression changes in 3T3-L1 adipocytes treated with each flavonoid. In parallel, molecular docking simulations were performed to examine the structural interactions between flavonoids and the AhR ligand-binding domain. By integrating transcriptomic and computational approaches, we aimed to elucidate how citrus flavonoids regulate gene networks and receptor interactions relevant to metabolic and inflammatory processes.

The present study used RNA sequencing and molecular docking to explore how citrus flavonoids influence aryl hydrocarbon receptor (AhR) signaling in 3T3-L1 adipocytes. Our integrated approach identified flavonoids as selective AhR modulators (SAhRMs), providing new insights into their potential role in reducing obesity-associated inflammation and supporting nutraceutical development for metabolic disorders.

2. Results

2.1. Differentially Expressed Genes in Flavonoid-Treated 3T3-L1 Adipocytes

RNA sequencing analysis identified 23,282 genes expressed in 3T3-L1 adipocytes treated with citrus flavonoids, including HES, NAR, NOB, SIN, and TAN. Differential expression analysis revealed distinct gene regulation patterns between the DIM group and Normal group. The DIM group exhibited the most extensive transcriptional changes, uniquely regulating 2,138 genes—1,470 upregulated and 668 downregulated—underscoring its substantial impact on adipocyte gene expression.

Comparison of DEGs between HES and NAR treatments identified 639 commonly upregulated genes, 747 commonly downregulated genes, and 231 contra-regulated genes, suggesting both shared and divergent transcriptomic responses (Figure 2A). Additionally, DIM group showed a separate cluster of 1,647 DEGs (1,124 upregulated and 523 downregulated), further confirming its distinct regulatory role. Intersection analysis of the polymethoxylated flavone subgroup (NOB, SIN, TAN) revealed 495 commonly upregulated genes, 754 downregulated genes, and 112 contra-regulated genes, indicating functional convergence within this subclass (Figure 2B).

Scatter plots of normalized gene expression (Figure S1) highlighted several DEGs with significant changes across treatment groups. Among them, Ahr, Adipoq, Pparg, IL-6, Il1a, Ccl2, and Cxcl10 displayed prominent treatment-specific expression profiles, reflecting modulation of genes involved in AhR signaling, adipogenesis, and inflammation.

Notably, a subset of 45 DEGs associated with AhR signaling, inflammation, and obesity exhibited distinct clustering patterns across flavonoid treatments. AhR related genes formed a separate cluster from those involved in metabolic and inflammatory responses, underscoring the central role of AhR in mediating flavonoid-induced transcriptomic changes in adipocytes (Figure 2C). Collectively, these findings underscore the diverse and complementary effects of citrus flavonoids on adipocyte transcriptomes, particularly through modulation of AhR -associated and immunometabolic signaling pathways.

2.2. Gene Ontology Analysis

Gene Ontology (GO) and differential gene expression analyses revealed that citrus flavonoids modulate biological processes, molecular functions, and cellular components associated with adipogenesis, inflammation, and AhR signaling. Among the experimental groups, the DIM group predominantly influenced adipogenesis-related genes (36%), HES NAR, and NOB exhibited balanced effects on adipogenic and inflammatory pathways (Figure 3A).

Notably, in the DIM group, AhR expression was significantly suppressed (0.416-fold) compared to the Normal group. However, treatment with HES and NAR restored AhR expression levels closer to those observed in the Normal group (0.988-fold and 1.214-fold, respectively), suggesting that these flavonoids may counteract differentiation-associated suppression of the AhR axis (Table 1). Key metabolic genes associated with obesity—including Adipoq, Apoa4, and Ppara—were significantly modulated. In particular, the DIM group exhibited a dramatic increase in Adipoq expression (~200-fold) and affected Pparg expression, indicating a potential role in improving adipocyte metabolic profiles. Notably, treatment with citrus flavonoids such as HES and NAR modulated the expression levels of these genes, suggesting that these compounds may influence adipocyte metabolism by regulating key adipogenic markers (Table 2). Inflammation-related genes such as Cxcl1, Cxcl5, Ccl20, and Il1a were highly upregulated by HES and NAR, while the DIM group showed suppression of pro-inflammatory targets including Cxcl5, Ccl17, Ccl2, Ccl5, and Il33 (Table 3).

GO enrichment analyses further supported these findings, highlighting significant regulation of biological processes (e.g., cellular response to stimuli, signal transduction, metabolic regulation; FDR ≤ 1.0e−18) (Figure 3B), molecular functions (e.g., cytokine activity, receptor binding, ligand interaction; FDR ≤ 1.0e−24) (Figure 3C), and cellular components (e.g., extracellular matrix, receptor complexes, AhR complex; FDR ≤ 1.0e−15) (Figure 3D).

2.3. Pathway Enrichment Analysis

Pathway enrichment analyses identified several biological pathways modulated by citrus flavonoids in 3T3-L1 adipocytes. KEGG analysis revealed significant enrichment in cytokine–cytokine receptor interaction, viral protein–cytokine interaction, and chemokine signaling pathways (FDR ≤ 1.0e−24) (Figure 4A). Reactome mapping showed activation of G protein-coupled receptor (GPCR) ligand binding, downstream signaling, and chemokine receptor pathways (FDR ≤ 1.0e−20) (Figure 4B). WikiPathways analysis emphasized chemokine signaling, adipogenesis, cytokine response, and inflammation (FDR ≤ 1.0e−16) (Figure 4C). STRING-based cluster enrichment identified modules related to CXCR signaling, JAK–STAT signaling, and the aryl hydrocarbon receptor (AhR) nuclear complex (FDR ≤ 1.0e−22) (Figure 4D). Collectively, these findings highlight the involvement of immune–metabolic pathways, with AhR signaling emerging as a key node in mediating flavonoid-induced responses in adipocytes.

2.4. Protein-Protein Interaction Network Analysis

STRING analysis identified gene networks associated with AhR, inflammation, and obesity (Figure 5). Central hub genes— AhR, Arnt2, Ccl20, Cxcl10, IL-6, and Adipoq—formed dense networks with regulatory overlap. AhR cofactors (Aip, Hsp90ab1, and Arnt) exhibited strong modular interactions, consistent with their known roles in xenobiotic metabolism. Metabolic regulators such as Pparg, Adipoq, Ppara, and Ramp3 displayed direct links with inflammatory genes IL-6, Il33, and Ccl2, are indicating immune–metabolic crosstalk. Notably, IL-6 emerged as a key connecting node. This network reinforces the convergence of AhR -driven gene regulation across inflammation and metabolism.

2.5. Molecular Docking Analysis

Molecular docking studies demonstrated that HES and NAR exhibited the strongest binding affinities to AhR, with docking scores of −8.9 and −8.7 kcal/mol, respectively. In contrast, NOB, SIN, and TAN showed comparatively lower affinities, ranging from −6.6 to −6.8 kcal/mol (Table 4; Figure 6).

HES and NAR formed hydrogen bonds with key amino acid residues within the AhR binding pocket, including Ser140, Gln150, Lys80, Phe136, and Tyr137. All tested flavonoids engaged in hydrophobic interactions with conserved residues such as Ser139, Tyr76, Ser75, Phe56, Ala79, Ile154, Leu118, Leu119, and Phe83. Notably, interactions with conserved residues like Tyr76, Phe83, and Phe136 suggest ligand specificity and potential functional relevance (Figure 6A and 6E). These structural interactions align with transcriptomic and network analyses, indicating that citrus flavonoids, particularly HES and NAR, may modulate AhR activity directly through binding. This mechanism potentially underpins their observed anti-inflammatory and anti-obesity effects, warranting further mechanistic exploration.

3. Discussion

Our study demonstrates that citrus-derived flavonoids modulate AhR-mediated signaling in 3T3-L1 adipocytes, thereby influencing downstream inflammatory and metabolic responses. Through transcriptomic analysis, flavonoids—particularly HES and NAR—were found to regulate gene clusters associated with adipogenesis, cytokine signaling, and AhR pathways [2,25]. Flavonoid treatment led to the upregulation of Arnt2 and partial restoration of Ahr expression, which had been suppressed during adipocyte differentiation. This regulatory pattern suggests that HES and NAR may function as selective AhR modulators (SAhRMs), consistent with prior findings on dietary polyphenols [26,27]. Importantly, these compounds downregulated pro-inflammatory mediators such as IL-6, Il1a, Ccl2, and Ccl20, confirming their anti-inflammatory potential [27,28]. Notably, Adipoq, Pparg, and Ppara—key genes involved in lipid metabolism and insulin sensitivity—were also upregulated, indicating improved metabolic homeostasis [29,30,31]. These effects were further supported by GO and pathway enrichment analyses, which identified cytokine–cytokine receptor interactions, GPCR signaling, and AhR nuclear complex as significantly enriched pathways. In particular, GPCR ligand binding was recognized as an enriched Reactome pathway, highlighting the role of GPCR-mediated signal transduction in adipocyte inflammatory and metabolic regulation [29]. This is supported by previous reports demonstrating that GPCRs transmit extracellular cues, including chemokines and hormones, into intracellular cascades linked to obesity and inflammation [32]. Taken together with cytokine receptor and JAK–STAT pathway enrichment, these findings suggest a convergence of AhR signaling with canonical immune regulatory axes, such as NF-κB and JAK–STAT [33,34], both of which are central to metabolic inflammation. Beyond the AhR nuclear complex, STRING-based cluster enrichment also implicated chemokine receptor binding, CXCR signaling, and the JAK–STAT axis [5,35]. This network-level overlap illustrates a multifaceted interplay between AhR modulation by flavonoids and immune signaling cascades, suggesting that citrus flavonoids regulate both inflammatory and metabolic responses through convergent transcriptional hubs [36,37]. STRING-based PPI network analysis revealed IL-6 and Pparg as central hubs bridging inflammatory and adipogenic modules. It was demonstrated that AhR cofactors, including Arnt, Aip, and Hsp90ab1, form a cohesive regulatory unit. This supports the hypothesis that AhR is a key integrator of immune-metabolic responses in adipocytes [38,39,40,41]. Docking studies further validated that HES and NAR directly bind to AhR’s ligand-binding domain [42], forming hydrogen bonds and hydrophobic contacts with conserved residues Ser140, Gln150, and Tyr137. Previous studies have shown that the natural flavonoid Urolithin A interacts with the AhR [43]. These structural interactions align with transcriptomic trends, thereby reinforcing the proposed mechanism of action [43,44,45,46]. Together, our data support a model in which citrus flavonoids act as SAhRMs, modulating AhR signaling to suppress obesity-associated inflammation while promoting adipocyte metabolic health.

4. Materials and Methods

4.1. Chemicals and Reagents

Dulbecco’s Modified Eagle Medium (DMEM), fetal bovine serum (FBS), and penicillin-streptomycin were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Flavonoids including HES, NAR, NOB, SIN, and TAN were obtained from Sigma-Aldrich (St. Louis, MO, USA). Other reagents such as 3-isobutyl-1-methylxanthine (IBMX), dexamethasone, bovine serum albumin (BSA), and insulin were also procured from Sigma-Aldrich. All reagents used were of analytical grade.

4.2. 3T3-L1 Cell Culture and Differentiation

Mouse 3T3-L1 preadipocytes (KCLB-42835, Korean Cell Line Bank, Seoul, Korea) were seeded in 6-well plates at a density of 3 × 10³ cells/well and maintained in DMEM containing 10% FBS and 1% penicillin-streptomycin under standard conditions (37°C, 5% CO₂). Upon reaching confluence, adipocyte differentiation was induced using a two-step protocol [23]. First, cells were treated for 24 h with differentiation induction medium composed of DMEM supplemented with 10% FBS, 0.5 mM IBMX, 0.5 µM dexamethasone, and 10 µg/mL insulin. This was followed by 48 h of treatment with differentiation progression medium containing DMEM, 10% FBS, and 10 µg/mL insulin. Cells were then maintained in DMEM with 10% FBS, refreshed every 2 days, until full differentiation. In this study, Cells maintained in an undifferentiated state were referred to as the 'Normal' group, whereas those subjected to the full adipogenic differentiation protocol were designated as the 'DIM' group. Flavonoid treatment groups (e.g., HES group) received the respective compound during the differentiation period. Chemical structures of the flavonoids are presented in Figure 1A, and the experimental timeline is illustrated in Figure 1B.

4.3. RNA Sequencing and Data Analysis

Total RNA was extracted using the Easy-BLUE Total RNA Extraction Kit (iNtRON Biotechnology, Seongnam, Korea). RNA quality was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Library preparation was performed [23] with the QuantSeq 3′ mRNA-Seq Library Prep Kit (Lexogen, Vienna, Austria), and sequencing was conducted on an Illumina NextSeq 500 platform (Illumina, CA, USA). Raw reads were aligned using Bowtie2 (v2.1.0), and differential gene expression analysis was performed using the EdgeR package (v3.2.2) in R (Bioconductor). Data normalization was conducted using the quantile method in Genowiz software (version 4.0.5.6, Ocimum Biosolutions, India). Genes were categorized based on annotation information from the Medline database provided by the National Center for Biotechnology Information (NCBI, Maryland, USA).

4.4. Gene Ontology and Pathway Enrichment Analysis

Differentially expressed genes (DEGs) were identified using the EdgeR package (v3.2.2) in R (Bioconductor) and filtered with thresholds of false discovery rate (FDR) ≤ 0.001 and absolute log₂ fold change ≥ 2. DEGs were subjected to Gene Ontology (GO) and pathway enrichment analyses using the Kyoto Encyclopedia of Genes and Genomes (KEGG), Reactome, WikiPathways, and local network cluster enrichment via the STRING database (v11.0; https://string-db.org). Enrichment was assessed within GO level 4 terms, spanning biological process, molecular function, and cellular component categories. Significance was determined using Fisher’s exact test followed by Holm–Bonferroni correction, and the top 10 enriched GO terms and pathways were reported. Gene expression clustering was performed using the k-means method based on relative expression levels.

4.5. Protein–Protein Interaction Network Analysis

DEGs identified in trait-linked hub modules were subjected to protein–protein interaction (PPI) network construction using the web-based STRING database (v11.0; https://string-db.org). Interactions were constructed using a medium confidence threshold (interaction score > 0.4) and visualized using Cytoscape software (v3.8.2). Hub genes were identified based on network centrality metrics, and overlapping DEGs across treatment groups were designated as hub DEGs. These hub genes were subsequently used to define functional gene clusters.

4.6. Molecular Docking Analysis

The crystal structure of the AhR ligand-binding domain (PDB ID: 5NJ8) was retrieved from the RCSB Protein Data Bank (https://www.rcsb.org/). Water molecules and unwanted heteroatoms were removed to prepare the receptor structure. Three-dimensional structures of HES, NAR, NOB, SIN, and TAN were downloaded from PubChem (https://pubchem.ncbi.nlm.nih.gov) in SDF format and converted using Open Babel [23]. Molecular docking was conducted using PyRx software (v0.8, The Scripps Research Institute, CA) with AutoDock Vina [24]. A grid box (25 Å × 29 Å × 35 Å) was defined around the ligand-binding site based on key active residues. Best poses were selected based on the lowest binding energies and stable interactions. Hydrogen bonding and hydrophobic interactions were visualized using PyMOL (v2.5.0; Schrödinger, NY), UCSF Chimera, and LigPlot+ software.

5. Conclusions

This study provides compelling evidence that citrus flavonoids—especially hesperidin and narirutin—selectively modulate AhR signaling and downstream gene networks in adipocytes. Through an integrated analysis of RNA sequencing, pathway enrichment, PPI networks, and molecular docking, we identified a dual mechanism by which these compounds suppress inflammation and enhance lipid metabolism. These findings advance our understanding of how dietary flavonoids, through AhR -centered immunometabolic regulation, may serve as nutraceuticals for managing obesity-associated inflammation and metabolic dysfunction.