Dual miRNA‐Mediated Axes Act in Concert in the Antioxidant Action of Eicosapentaenoic Acid: let‐7c‐3p/TFAM Drives Mitochondrial Biogenesis While miR‐34c‐5p/SIRT1 Potentiates Enzymatic Defenses Authors: Haiqi Wang, Yanqi Lin, Yuanyuan Li, Xiaowen Wu, Jingyu Lin, Shiying Huang, Guiling Li Published: 2026 ABSTRACT ABSTRACTPrevious transcriptomic analysis revealed that eicosapentaenoic acid (EPA) alters miRNA expression in HepG2 cells. Two key miRNA‐mRNA axes mediating EPA's antioxidant effects were identified. EPA was found to downregulate let‐7c‐3p, which directly targets mitochondrial transcription factor A (TFAM). Inhibiting let‐7c‐3p or overexpressing TFAM enhanced antioxidant capacity, reduced reactive oxygen species, improved mitochondrial function, and promoted mitochondrial biogenesis. In parallel, EPA was found to upregulate miR‐34c‐5p, which directly targets NAD‐Dependent Protein Deacetylase Sirtuin (SIRT1). This repression of SIRT1 is associated with increased activities of antioxidant enzymes, including catalase, superoxide dismutase, and glutathione peroxidase. These findings indicate a dual‐miRNA mechanism through which EPA alleviates oxidative stress via coordinated enhancement of mitochondrial biogenesis and enzymatic defenses. Introduction Oxidative stress (OS), characterized by a systemic imbalance between the generation of reactive oxygen species (ROS) and the capacity of the antioxidant defense system, constitutes a fundamental pathophysiological mechanism underlying a wide spectrum of human diseases [1, 2]. Under physiological conditions, ROS, including superoxide anion (O2•−), hydrogen peroxide (H2O2), and hydroxyl radicals (•OH), function as crucial signaling molecules regulating cellular processes such as proliferation and survival. However, their aberrant accumulation inflicts damage upon cellular macromolecules including lipids, proteins, and DNA, ultimately leading to functional impairment, genetic mutations, and the pathogenesis of various chronic conditions, including neurodegenerative disorders, cardiovascular diseases, cancer, and metabolic syndromes [3, 4]. The mitochondrion, often termed the cellular powerhouse, is a primary site for both ROS generation and elimination. The mitochondrial electron transport chain (ETC), essential for aerobic ATP synthesis, is a major source of endogenous ROS, primarily due to electron leakage during oxidative phosphorylation [5]. Consequently, mitochondrial dysfunction and oxidative stress are inextricably linked in a vicious cycle: impaired ETC efficiency increases electron leakage and ROS production, while excessive ROS, in turn, damages mitochondrial components—including lipids of the inner membrane and mitochondrial DNA (mtDNA)—further exacerbating mitochondrial dysfunction [6]. This self‐perpetuating cycle is a critical contributor to cellular senescence and the progression of numerous age‐related diseases [7]. Therefore, strategies aimed at enhancing mitochondrial function and breaking this detrimental cycle hold significant therapeutic promise for mitigating oxidative stress‐associated pathologies. Given the central role of mitochondrial‐derived oxidative stress in disease pathology, nutritional interventions targeting this axis have garnered significant interest. Among the most promising candidates are omega‐3 polyunsaturated fatty acids (PUFAs), notably eicosapentaenoic acid (EPA). EPA is an essential PUFA renowned for its diverse biological activities, including anti‐inflammatory, anti‐cancer, and cardioprotective effects [8]. Accumulating evidence also underscores its potent antioxidant capabilities. For instance, EPA has been shown to reduce ROS generation in human aortic endothelial cells [9] and enhance the activities of antioxidant enzymes such as catalase (CAT) and superoxide dismutase (SOD) in rat pheochromocytoma PC12 cells, thereby protecting them from oxidative damage [10]. Our previous work further demonstrated that EPA can elevate mitochondrial antioxidant enzyme activities and attenuate cellular ROS levels, an effect concomitant with improved mitochondrial function [11]. However, the precise molecular mechanisms through which EPA orchestrates its antioxidant defense remain incompletely understood and paradoxical. While EPA can act as an antioxidant, its highly unsaturated structure also renders it susceptible to peroxidation, leading to pro‐oxidant effects under certain conditions, as observed in human prostate cancer cells [12, 13]. This duality highlights the complexity of its action. Current evidence suggests that EPA may exert its protective effects by modulating membrane proteins, cell signaling pathways, and gene expression to influence mitochondrial redox status. For example, EPA can activate the NRF2 pathway, a master regulator of antioxidant gene expression [14], and has been reported to affect the expression of mitochondrial dynamics factors like MIEF1 [15] and transcription factors like TFAM [16]. These findings point to a multifaceted mode of action, yet a clear, upstream regulatory mechanism that integrates these diverse effects is still lacking. In particular, the potential role of epigenetic regulators, such as microRNAs (miRNAs), in mediating the antioxidant effects of EPA is an emerging and underexplored area [17, 18]. Elucidating such a mechanism would provide a unifying framework for understanding how EPA fine‐tunes cellular antioxidant responses. MicroRNAs (miRNAs) are a class of short, non‐coding RNA molecules that function as critical post‐transcriptional regulators of gene expression. By binding to complementary sequences in the 3′‐untranslated regions (3′‐UTRs) of target mRNAs, miRNAs typically induce translational repression or mRNA degradation, thereby fine‐tuning numerous cellular processes [19]. A growing body of evidence has firmly established the involvement of specific miRNAs in the regulation of cellular oxidative stress [20]. These miRNAs can either be influenced by ROS levels or, in turn, directly modulate the expression of genes central to the antioxidant defense system, such as those encoding antioxidant enzymes and components of the mitochondrial ETC [21, 22, 23]. Notably, miRNAs are not only present in the cytoplasm but also reside within mitochondria, where they can directly influence mitochondrial function and genome integrity, further positioning them as ideal candidates for mediating communication between nutritional cues and mitochondrial redox balance. Given the emerging role of miRNAs in post‐transcriptional gene regulation, a key question is whether EPA's antioxidant effects might be mediated through miRNA‐dependent mechanisms. Critically, bioactive food components, including PUFAs like EPA, have been recognized as potent modulators of miRNA expression, representing a novel layer of post‐transcriptional regulation [17]. For instance, EPA has been shown to influence the expression of miR‐155‐3p in macrophages and miR‐30b in the liver, contributing to its anti‐inflammatory and antioxidant effects [24]. These findings suggest that miRNA‐mediated pathways are a plausible mechanism for EPA's bioactivity. However, a systematic and mechanistic investigation into the role of miRNAs in EPA's antioxidant action, particularly one that connects specific EPA‐regulated miRNAs to definitive mitochondrial targets and functional outcomes, is still lacking. Despite these observations, the upstream regulatory mechanisms linking EPA to mitochondrial redox regulation remain largely unknown. To address this gap, transcriptomic profiling of HepG2 cells treated with EPA was first performed. Among the differentially expressed miRNAs, let‐7c‐3p and miR‐34c‐5p were identified as the most significantly altered. These two miRNAs were prioritized for further functional validation based on two criteria: (1) their predicted targeting of TFAM and SIRT1, respectively, both of which are central regulators of mitochondrial function and redox homeostasis; and (2) their potential relevance to mitochondrial biology as suggested by the literature. While individual miRNA‐mediated pathways have been implicated in EPA's antioxidant effects, the question of whether EPA can simultaneously coordinate multiple distinct miRNA axes to achieve a multi‐layered antioxidant defense has not been addressed. The present study was designed to bridge this gap by identifying and functionally validating the key miRNA‐mRNA axes through which EPA exerts its antioxidant effects. The findings demonstrate that EPA alleviates oxidative stress by suppressing let‐7c‐3p to derepress the mitochondrial transcription factor A gene, TFAM, thereby driving mitochondrial biogenesis, while concurrently upregulating miR‐34c‐5p to repress NAD‐Dependent Protein Deacetylase Sirtuin‐1 (SIRT1), thereby potentiating enzymatic antioxidant defenses. Protein‐level validation and functional rescue experiments were further performed to confirm the regulatory roles of these miRNA‐target axes. This dual‐axis framework provides a new conceptual understanding of how a single nutritional intervention can coordinate multiple epigenetic mechanisms to achieve a comprehensive antioxidant response. Materials and Methods Cell Culture and EPA Treatment The human hepatoma cell line HepG2 (RRID: CVCL_0027) was obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) in 2023. According to the supplier's documentation, the cell line has been authenticated by short tandem repeat (STR) profiling upon deposition at the cell bank and was certified as mycoplasma‐free upon purchase. The cells were used within 25 passages after thawing and were maintained under standard culture conditions without any signs of contamination during the experimental period. Cells were maintained in high‐glucose Dulbecco's Modified Eagle's Medium (DMEM; HyClone, USA) supplemented with 10% (vol/vol) heat‐inactivated fetal bovine serum (FBS; Gemini, USA) and 1% (vol/vol) penicillin‐streptomycin (HyClone, USA) at 37°C in a humidified incubator with 5% CO2. Eicosapentaenoic acid (EPA; CAS: 10417‐94‐4, Sigma‐Aldrich, USA) was dissolved in absolute ethanol to prepare a stock solution. For treatment, the EPA stock was diluted directly into the cell culture medium to achieve the specified final concentrations (e.g., 100 µM and 150 µM). An equivalent volume of ethanol (final concentration 1%, vol/vol) was added to the control groups. Unless otherwise specified for viability assessment, all experiments were conducted following a 24‐h EPA treatment period under conditions that did not significantly affect cell viability, as confirmed by preliminary assays. All cells used in all experiments were within 25 passages to ensure genetic stability and consistent response. Luciferase Reporter Vector Construction The 3′‐untranslated regions (3′‐UTRs) of TFAM and SIRT1, each containing the predicted wild‐type binding sites for let‐7c‐3p and miR‐34c‐5p respectively, were amplified by PCR from HepG2 cDNA. The SIRT1 3‘‐UTR amplicon was similarly cloned into the psiCHECK‐2 vector using the XhoI and NotI sites, generating psiCHECK2‐SIRT1‐wt. The resulting amplicons were subsequently cloned into the psiCHECK‐2 vector (Promega, USA) downstream of the Renilla luciferase gene using the XhoI and NotI restriction sites, generating the wild‐type reporter plasmids psiCHECK2‐TFAM‐wt and psiCHECK2‐SIRT1‐wt. Site‐directed mutagenesis of the miRNA seed binding regions within the respective 3′‐UTRs was performed using an overlapping PCR strategy to create the mutant constructs, psiCHECK2‐TFAM‐mut and psiCHECK2‐SIRT1‐mut. All constructed plasmids were verified by Sanger sequencing prior to use in transfection experiments. The primer sequences used for the amplification and mutagenesis were as follows: For TFAM 3′‐UTR: Wild‐type TFAM 3′‐UTR: F: 5′‐CTCGAGGGATGATGACTATGGAAAGAG‐3′; R: 5′‐GCGGCCGCAAAGCAGCCATAGACAAG‐3′ Mutant TFAM 3′‐UTR: F: 5′‐GTTGATGGCTATGCGTATAAGTGATTGCTG‐3′; R: 5′‐CAGCAATCACTTATAGGCATAGCCATCAAC‐3′ For SIRT1 3′‐UTR: Wild‐type SIRT1 3′‐UTR: F: 5′‐CTCGAGGAGCACTCGGTTGTCTTTAC‐3′; R: 5′‐GCGGCCGCCACCAAATCCTCCCAATG‐3′ Mutant SIRT1 3′‐UTR: F: 5′‐CACAAGTATTAAATTACTAAAATGTGAATATGC‐3′; R: 5′‐GCATATTCACATTTTAGTAATTTAATACTTGTG‐3′ miRNA and siRNA Transfection Cells were grown in a culture plate for overnight to approximately 80% confluency. miRNA (RiboBio, Guangzhou, China) or siRNA (GenePharma, Shanghai, China) was then transfected into cells to a final concentration of 40 nM using Lipofectamine 2000 reagent (ThermoFisher, Rockford, IL, USA) following the manufacturer's instructions. The transfected cells were harvested 48 h later for subsequent assays. The si‐TFAM sequences (sense: 5′‐GGACACTCGATTTATTCCTGGATT‐3′ and antisense: 5′‐ACATCTCCATGAACAATTCTCAGTTT‐3′) were synthesized by GenePharma. For combined EPA and miRNA/siRNA transfection experiments, cells were first treated with 100 µM EPA for 24 h, followed by transfection with miRNA mimic, inhibitor, or siRNA. EPA was maintained in the culture medium throughout the transfection period (an additional 48 h). Dual‐Luciferase Reporter Assay To verify the direct targeting of TFAM and SIRT1 by let‐7c‐3p and miR‐34c‐5p, respectively, a dual‐luciferase reporter assay was performed. Briefly, HepG2 cells were seeded in a 96‐well plate at a density of 1 × 104 cells per well and cultured for 24 h. Upon reaching 70%–80% confluency, the cells were co‐transfected with 0.25 µg of either the wild‐type (psiCHECK2‐TFAM‐wt or psiCHECK2‐SIRT1‐wt) or mutant (psiCHECK2‐TFAM‐mut or psiCHECK2‐SIRT1‐mut) reporter plasmid, along with 5 pmol of the corresponding miRNA mimic (let‐7c‐3p or miR‐34c‐5p) or negative control miRNA (miR‐NC), using Lipofectamine 2000. After 36 h of transfection, luciferase activities were assessed by the Dual‐Luciferase Reporter Assay System (Promega, USA) according to the manufacturer's instructions. The Renilla luciferase signal was normalized to the internal control Firefly luciferase signal for each well to account for variations in transfection efficiency. The relative luciferase activity in cells transfected with the negative control miRNA (miR‐NC) and the wild‐type reporter plasmid was set as 1.0 (100%) for comparison. Measurement of Intracellular ROS Levels Intracellular reactive oxygen species (ROS) levels were assessed using the fluorescent probe 2′,7′‐dichlorodihydrofluorescein diacetate (DCFH‐DA) (Beyotime, China), following the manufacturer's instructions. In brief, harvested cells were incubated with 10 µM DCFH‐DA diluted in Opti‐MEM serum‐free medium (Thermo Fisher Scientific, USA) at 37°C for 30 min in the dark. Subsequently, the cells were washed three times with Opti‐MEM to remove excess probe. The green fluorescence intensity, proportional to the intracellular ROS levels, was immediately analyzed using a Guava EasyCyte flow cytometer (Millipore, USA) and the GuavaSoft 3.1.1 software. Measurement of Mitochondrial ROS Levels Mitochondrial superoxide production was assessed using MitoSOX Red mitochondrial superoxide indicator (Thermo Fisher Scientific, USA), following the manufacturer's protocol. In brief, harvested cells were incubated with 5 µM MitoSOX Red reagent at 37°C for 30 min in the dark. After incubation, the cells were washed three times with pre‐warmed phosphate‐buffered saline (PBS) to remove the residual dye. The cells were then resuspended in PBS, and the red fluorescence intensity, indicative of mitochondrial superoxide levels, was immediately analyzed using a Guava EasyCyte flow cytometer (Millipore, USA). Measurement of Total Cellular Antioxidant Capacity Harvested cells were lysed by ultrasonication (five cycles of 10‐second pulses with 20‐second intervals on ice) and centrifuged at 12,000g for 10 min at 4°C. The resulting supernatant was collected, and the protein concentration was determined using a bicinchoninic acid (BCA) assay kit (Beyotime, China). The total cellular antioxidant capacity of the lysates was evaluated using two complementary methods: the ABTS radical cation scavenging assay and the ferric reducing antioxidant power (FRAP) assay, performed with commercial kits (Beyotime, China) according to the manufacturer's instructions. The antioxidant capacity was normalized to the total protein content and expressed as Trolox equivalents (for ABTS) and FeSO4 equivalents (for FRAP) per milligram of protein. Measurement of Antioxidant Enzyme Activities and ADP/ATP Ratio The activities of key antioxidant enzymes were assessed using commercial assay kits according to the manufacturers' protocols. Catalase (CAT) activity was determined using a kit from Beyotime (China). The activities of superoxide dismutase (SOD) and glutathione peroxidase (GPx) were assessed by kits from JianCheng Bioengineering Institute (Nanjing, China). All enzyme activities were normalized to the total protein concentration determined by the BCA method. The cellular ADP/ATP ratio, an indicator of cellular energy status, was quantified using an ADP/ATP Ratio Assay Kit (Sigma‐Aldrich, USA). In brief, cells were lysed, and the lysate was incubated with the ATP reagent. The resulting luminescence (RLUA) represents the original ATP content. After a 10‐min incubation, the background luminescence (RLUB) was recorded. Subsequently, the ADP reagent was added to convert ADP to ATP, and the luminescence was measured again (RLUC), representing the total of background and ADP‐derived ATP. The ADP/ATP ratio was calculated as follows: (RLUC ‐ RLUB)/RLUA. Measurement of Mitochondrial Membrane Potential (MMP) Mitochondrial membrane potential (MMP) was assessed using the fluorescent, cationic dye JC‐1 (Beyotime, China), following the manufacturer's instructions. In brief, harvested cells were incubated with 5 μg/mL JC‐1 at 37°C for 20 min in the dark. The cells were then washed twice with ice‐cold JC‐1 staining buffer to remove unincorporated dye, resuspended in fresh buffer, and immediately analyzed using a Guava EasyCyte flow cytometer (Millipore, USA). The MMP was quantified and presented as the ratio of red fluorescence intensity (from J‐aggregates in healthy mitochondria) to green fluorescence intensity (from JC‐1 monomers in depolarized mitochondria). Quantitative Real‐Time PCR (qRT‐PCR) Analysis Total RNA was extracted from harvested cells using TRIzol Reagent (Thermo Fisher Scientific, USA) following the manufacturer's protocol. RNA concentration and purity were determined by spectrophotometry. For miRNA analysis, first‐strand cDNA was synthesized from total RNA using the stem‐loop reverse transcription method with the M‐MLV Reverse Transcriptase system (Promega, USA). For mRNA analysis, cDNA was synthesized using the ReverTra Ace qPCR RT Master Mix with gDNA Remover (TOYOBO, Japan) to eliminate genomic DNA contamination. Quantitative real‐time PCR was performed using the SYBR Green Premix Pro Taq HS qPCR Kit (Accurate Biology, China) on an ABI 7300 Real‐Time PCR System (Applied Biosystems, USA). Each reaction was carried out in technical triplicates. The relative expression levels of target genes were normalized to the internal control genes GAPDH (for mRNA) or U6 snRNA (for miRNA) and calculated using the comparative 2‐ΔΔCt method. The primer sequences for genes analyzed in this study are listed in Table 1. Primer sequences for other targets (e.g., mitochondrial‐encoded genes ND1, COX1, etc.) were as previously described [10]. Table 1Sequences for qRT‐PCR primers used in this study.GenePrimer (5′–3′) TFAM F: GGACACTCGATTTATTCCTGGATT R: ACATCTCCATGAACAATTCTCAGTTT PGC1‐α F: ACTCTCCGACACCAAACTGC R: CCACTTCATCCACCCAGAAA NRF1 F: AGAGAACAGTTCAACAGCAGCAA R: TGGGTAGGTTGGTGTTGATGG NRF2 F: TCTCCTCGCTGGAAAAAGAA R: AATGTGCTGGCTGTGCTTTA TFB1M F: TCTTCGACAGCACCAACAGC R: GGCAGGTAGATGAGGCACAG TFB2M F: CCTGGGCTACATCACCAACT R: CAGCATCCACAGCATCACAT ERRα F: TCCAGCATCCTGACCTATGC R: GGTAGGCGATCTTGGCATCT SIRT1 F: TAGCCTTGTCAGATAAGGAAGGA R: ACAGCTTCACAGTCAACTTTGT COX4 F: CAGGGTATTTAGCCTAGTTGGC R: GCCGATCCATATAAGCTGGGA John Wiley & Sons, Ltd. Sequences for qRT‐PCR primers used in this study. F: GGACACTCGATTTATTCCTGGATT R: ACATCTCCATGAACAATTCTCAGTTT F: ACTCTCCGACACCAAACTGC R: CCACTTCATCCACCCAGAAA F: AGAGAACAGTTCAACAGCAGCAA R: TGGGTAGGTTGGTGTTGATGG F: TCTCCTCGCTGGAAAAAGAA R: AATGTGCTGGCTGTGCTTTA F: TCTTCGACAGCACCAACAGC R: GGCAGGTAGATGAGGCACAG F: CCTGGGCTACATCACCAACT R: CAGCATCCACAGCATCACAT F: TCCAGCATCCTGACCTATGC R: GGTAGGCGATCTTGGCATCT F: TAGCCTTGTCAGATAAGGAAGGA R: ACAGCTTCACAGTCAACTTTGT F: CAGGGTATTTAGCCTAGTTGGC R: GCCGATCCATATAAGCTGGGA Western Blot Analysis The primary antibodies used in this study were: anti‐CAT (1:1000, #12980, Cell Signaling), anti‐SOD2 (1:1000, ab13533, Abcam), anti‐GPx1 (1:500, GTX116040, GeneTex), anti‐MFN1 (1:2000, 66635‐2‐Ig, Proteintech), anti‐FIS1 (1:2000, 66776‐1‐Ig, Proteintech), anti‐FLAG (1:2000, M30971, Boster), anti‐GAPDH (1:5000, AF7021, Affinity), anti‐TFAM (1:1000, Boster, PA1936), and anti‐SIRT1 (1:1000, Boster, A00018‐1). HRP‐conjugated secondary antibodies (anti‐rabbit IgG, 1:10,000, BA1054; anti‐mouse IgG, 1:5000, BA1050; Boster) were used for detection. Statistical Analysis All experiments were independently repeated at least three times (n ≥ 3) with separate biological samples. Data are presented as the mean ± standard deviation (SD). Statistical analyzes were performed using SPSS software (version 16.0; IBM, USA). After confirming the homogeneity of variances, differences between multiple groups were assessed by one‐way analysis of variance (ANOVA), followed by Duncan's post hoc test for multiple comparisons. A p‐value of less than 0.05 was considered statistically significant, and specific p‐values are indicated in the figure legends. Ethical Statement This study did not involve human participants or animal experiments. The human hepatoma cell line HepG2 used in this research was obtained from a commercial source (Cell Bank of the Chinese Academy of Sciences, Shanghai, China) and is not of direct human or animal origin requiring ethical approval. No human subjects or animals were involved in this study. Results EPA Downregulates let‐7c‐3p to Mediate Its Antioxidant Activity Accumulating evidence has implicated let‐7c‐3p in the cellular stress response, and its dysregulation is associated with various diseases, including cancers and metabolic disorders [20, 21]. However, its specific role, particularly in the context of nutritional intervention, remains largely unexplored. In this study, let‐7c‐3p was identified as one of the most significantly downregulated miRNAs in HepG2 cells treated with 100 µM or 150 µM EPA for 48 h (Figure 1A), suggesting its potential involvement in EPA's biological activities. To investigate whether the downregulation of let‐7c‐3p is functionally required for EPA's antioxidant effect, a gain‐of‐function approach was employed by overexpressing let‐7c‐3p. HepG2 cells were pretreated with 100 µM EPA for 24 h, followed by transfection with a let‐7c‐3p mimic for 48 h. As shown in Figure 1B, EPA treatment alone significantly reduced cellular ROS levels. Conversely, overexpression of let‐7c‐3p alone led to a marked accumulation of ROS, indicating its pro‐oxidant role. Critically, the antioxidant effect of EPA was completely abolished when let‐7c‐3p was simultaneously overexpressed, as evidenced by the unchanged ROS level in the co‐treatment group compared to the control. It was hypothesized that the ROS reduction induced by EPA (via let‐7c‐3p downregulation) was attributable to an enhancement of cellular antioxidant capacity. This was confirmed by both ABTS radical‐scavenging and FRAP assays. EPA treatment significantly boosted total antioxidant capacity, while let‐7c‐3p overexpression substantially weakened it (Figure 1C,D). Notably, the enhancement of antioxidant capacity induced by EPA was fully blocked by co‐transfection with the let‐7c‐3p mimic. In conclusion, these results demonstrate that downregulation of let‐7c‐3p is an essential upstream event for EPA to enhance cellular antioxidant capacity and mitigate oxidative stress. Downregulation of let‐7c‐3p mediates the antioxidant effect of EPA in HepG2 cells. (A) Relative expression level of let‐7c‐3p in HepG2 cells after treatment with 100 µM or 150 µM EPA for 48 h, as determined by qRT‐PCR. U6 snRNA was used as an internal control. (B) Intracellular ROS levels upon EPA treatment (100 µM, 24 h) and transfection with let‐7c‐3p mimic. (C and D) Total cellular antioxidant capacity measured by the ABTS radical‐scavenging assay (C) and the FRAP assay (D) under the same treatment conditions as in (B). Data are presented as the mean ± SD (n ≥ 3). *p < 0.05, **p < 0.01 versus control group; different letters (a, b, c and d) indicate significant differences (p < 0.05) among groups as determined by one‐way ANOVA with Duncan's test. EPA Upregulates miR‐34c‐5p to Enhance Antioxidant Defense Having established the role of let‐7c‐3p downregulation, the function of another EPA‐regulated miRNA, miR‐34c‐5p, was next investigated. In contrast to let‐7c‐3p, qRT‐PCR analysis revealed that the expression level of miR‐34c‐5p was significantly upregulated in HepG2 cells after treatment with 100 µM or 150 µM EPA for 48 h (Figure 2A). To determine whether the upregulation of miR‐34c‐5p is functionally involved in EPA's antioxidant effect, cells were transfected with a miR‐34c‐5p inhibitor. Inhibition of miR‐34c‐5p alone markedly increased intracellular ROS levels. While EPA treatment effectively reduced ROS in control cells, this effect was completely abolished when miR‐34c‐5p was inhibited (Figure 2B), indicating that the upregulation of miR‐34c‐5p is indispensable for EPA to mitigate cellular oxidative stress. To assess whether this effect was mediated through the augmentation of cellular antioxidant capacity, total antioxidant capacity was measured. Consistent with the ROS data, EPA treatment significantly enhanced total antioxidant capacity, as measured by both ABTS and FRAP assays. Conversely, inhibition of miR‐34c‐5p substantially attenuated basal antioxidant capacity. Importantly, the enhancement of antioxidant capacity induced by EPA was entirely blocked by co‐transfection with the miR‐34c‐5p inhibitor (Figure 2C,D). Collectively, these results demonstrate that EPA‐induced upregulation of miR‐34c‐5p is a crucial event for the enhancement of the cellular antioxidant defense system. Upregulation of miR‐34c‐5p is essential for the antioxidant activity of EPA. (A) Relative expression level of miR‐34c‐5p in HepG2 cells treated with 100 µM or 150 µM EPA for 48 h. U6 snRNA was used for normalization. (B) Analysis of intracellular ROS levels following EPA treatment (100 µM, 24 h) and transfection with miR‐34c‐5p inhibitor. (C and D) The total antioxidant capacity of cells was evaluated using the ABTS (C) and FRAP (D)assays after the indicated treatments. Data are presented as the mean ± SD (n ≥ 3). *p < 0.05, **p < 0.01 versus control group; different letters (a, b, c, and d) indicate significant differences (p < 0.05) among groups as determined by one‐way ANOVA with Duncan's test. The let‐7c‐3p/TFAM Axis Drives Mitochondrial Biogenesis and Function To elucidate the downstream mechanism by which let‐7c‐3p mediates EPA's effects, its direct targets were sought. Bioinformatic prediction suggested that the 3′‐untranslated region (3′‐UTR) of TFAM, a master regulator of mitochondrial biogenesis, contains a highly conserved binding site for let‐7c‐3p. A dual‐luciferase reporter assay was performed to validate this interaction. Co‐transfection of let‐7c‐3p mimic with the wild‐type TFAM 3′‐UTR reporter (psiCHECK2‐TFAM‐wt) significantly suppressed luciferase activity compared to the negative control. This suppression was completely abolished when the binding site was mutated (psiCHECK2‐TFAM‐mut), confirming the direct and specific binding of let‐7c‐3p to the TFAM 3′‐UTR (Figure 3A). To confirm that this targeting regulates TFAM expression in a cellular context, transfection experiments were performed. Transfection of let‐7c‐3p mimic into HepG2 cells led to a significant decrease in the mRNA level of TFAM (Figure 3B). Conversely, inhibition of endogenous let‐7c‐3p resulted in a marked increase in TFAM expression (Figure 3C), demonstrating that let‐7c‐3p acts as a tonic repressor of TFAM. The let‐7c‐3p/TFAM axis drives mitochondrial biogenesis and function. (A) Validation of let‐7c‐3p binding to the TFAM 3′‐UTR using a dual‐luciferase reporter assay. The luciferase activity of the wild‐type reporter (TFAM‐wt) was significantly suppressed upon co‐transfection with the let‐7c‐3p mimic, while mutation of the binding site (TFAM‐mut) abolished this effect. (B and C) Let‐7c‐3p inversely regulates TFAM expression at the mRNA level. (B) Transfection with a let‐7c‐3p mimic significantly decreased TFAM mRNA. (C) Conversely, inhibition of endogenous let‐7c‐3p increased TFAM mRNA. (D) Inhibition of let‐7c‐3p or overexpression of TFAM increased mitochondrial DNA (mtDNA) copy number. (E) Let‐7c‐3p inhibition upregulates the expression of mitochondrial‐encoded genes (ND1, ND5, COX1, COX3) and key nuclear‐encoded transcription factors (PGC‐1α, NRF1, TFAM). (F and G) Let‐7c‐3p inhibition or TFAM overexpression enhances mitochondrial function. (F) The ADP/ATP ratio is elevated, indicating improved cellular energy status. (G) Mitochondrial superoxide levels are reduced, demonstrating attenuated intramitochondrial oxidative stress. Data are presented as the mean ± SD (n ≥ 3). *p < 0.05, **p < 0.01 versus control group. Given the pivotal role of TFAM in mitochondrial genome maintenance and transcription, the functional consequences of this regulatory axis on mitochondrial biogenesis were investigated. Inhibition of let‐7c‐3p, which elevates TFAM expression, significantly increased mitochondrial DNA (mtDNA) copy number (Figure 3D). Consistent with enhanced biogenesis, the expression of key mitochondrial‐encoded genes (e.g., ND1, COX1) and nuclear‐encoded transcription factors governing mitochondrial biogenesis (e.g., PGC‐1α, NRF1) were also upregulated upon let‐7c‐3p inhibition (Figure 3E). To determine whether this boost in mitochondrial biogenesis translated into improved functional output, mitochondrial function was assessed. Cells with inhibited let‐7c‐3p exhibited a higher ADP/ATP ratio, indicating improved energy charge, and a reduction in mitochondrial superoxide production, demonstrating attenuated oxidative stress within the organelles (Figure 3F,G). In summary, these data delineate a clear pathway wherein EPA, by downregulating let‐7c‐3p, relieves the repression of TFAM. The subsequent upregulation of TFAM drives mitochondrial biogenesis and augments mitochondrial function, which constitutes a fundamental mechanism for EPA's antioxidant action. The miR‐34c‐5p/SIRT1 Axis Mediates EPA's Antioxidant Effect by Directly Enhancing Enzymatic Defenses Data in Section 3.2 established that EPA upregulates miR‐34c‐5p, and that this upregulation is necessary for EPA to reduce intracellular ROS levels and enhance total cellular antioxidant capacity (Figure 2B–D). To elucidate the downstream mechanism, the direct target of miR‐34c‐5p was sought. Bioinformatic analysis predicted a conserved binding site for miR‐34c‐5p in the 3′‐UTR of SIRT1. A dual‐luciferase reporter assay confirmed this direct interaction, as co‐transfection of the miR‐34c‐5p mimic with the wild‐type SIRT1 3′‐UTR reporter (psiCHECK2‐SIRT1‐wt) significantly suppressed luciferase activity compared to the negative control. This suppression was completely abolished when the predicted binding site was mutated (psiCHECK2‐SIRT1‐mut) (Figure 4A). Consistent with this, overexpression of miR‐34c‐5p in HepG2 cells led to a significant decrease in the mRNA level of SIRT1 (Figure 4B), confirming that miR‐34c‐5p directly represses SIRT1 expression. The miR‐34c‐5p/SIRT1 axis mediates EPA's antioxidant effect by enhancing enzymatic defenses. (A) miR‐34c‐5p directly targets the SIRT1 3′‐UTR. Dual‐luciferase reporter assay showing relative luciferase activity in HepG2 cells co‐transfected with wild‐type (SIRT1‐wt) or mutant (SIRT1‐mut) reporter and miR‐34c‐5p mimic or miR‐NC. **p < 0.01 versus miR‐NC + SIRT1‐wt. (B) miR‐34c‐5p represses SIRT1 mRNA expression. qRT‐PCR analysis of SIRT1 mRNA in HepG2 cells after miR‐34c‐5p mimic transfection. *p < 0.05 versus miR‐NC. (C and D) Activation of the miR‐34c‐5p/SIRT1 axis enhances antioxidant enzyme activities. CAT, SOD, and GPx activities in HepG2 cells after miR‐34c‐5p overexpression or SIRT1 knockdown (si‐SIRT1). Different letters indicate significant differences (p < 0.05) among groups. (E) miR‐34c‐5p inhibition abolishes EPA‐induced increases in CAT, SOD, and GPx activities. HepG2 cells were treated with 100 µM EPA for 24 h and transfected with miR‐34c‐5p inhibitor for 48 h. ##p < 0.01 versus control; **p < 0.01 versus EPA. C, control; inh, miR‐34c‐5p inhibitor; E + I, EPA + inhibitor. To determine whether activation of the miR‐34c‐5p/SIRT1 axis is sufficient to recapitulate the enhancement of the cellular antioxidant defense system, loss‐ and gain‐of‐function experiments were performed. Overexpression of miR‐34c‐5p alone, or direct knockdown of its target SIRT1 using siRNA, significantly increased the activities of key antioxidant enzymes, including catalase (CAT), superoxide dismutase (SOD), and glutathione peroxidase (GPx) (Figure 4C,D). In conclusion, these data delineate a coherent pathway for EPA's antioxidant action: EPA upregulates miR‐34c‐5p, which in turn directly represses SIRT1. This repression is sufficient to activate the core enzymatic antioxidant defense system, thereby enhancing cellular resistance to oxidative stress. miRNA‐Target Regulation at the Protein Level To confirm miRNA‐target regulation at the protein level, western blot analysis was performed for both axes. As shown in Figure 5A, let‐7c‐3p mimic decreased TFAM protein levels compared to the negative control. Conversely, EPA treatment increased TFAM protein levels in a concentration‐dependent manner (Figure 5B). For the miR‐34c‐5p/SIRT1 axis, miR‐34c‐5p mimic decreased SIRT1 protein levels (Figure 5C), and EPA treatment decreased SIRT1 protein levels in a concentration‐dependent manner (Figure 5D). These results provide direct evidence that both miRNAs regulate their respective target genes at the protein level. miRNA‐target regulation at the protein level. (A) Representative western blot images showing TFAM protein levels upon transfection with let‐7c‐3p mimic or NC mimic. (B) TFAM protein levels upon EPA treatment at 100 µM and 150 µM for 24 h, or vehicle control. (C) Representative western blot images showing SIRT1 protein levels upon transfection with miR‐34c‐5p mimic or NC mimic. (D) SIRT1 protein levels upon EPA treatment at 100 µM and 150 µM for 24 h. GAPDH served as loading control. Discussion The present study provides mechanistic insights into the antioxidant activity of eicosapentaenoic acid (EPA) by identifying and functionally validating two distinct but complementary miRNA‐mediated pathways in HepG2 cells. Our findings demonstrate that EPA alleviates oxidative stress through the coordinated action of the let‐7c‐3p/TFAM and miR‐34c‐5p/SIRT1 axes, which respectively enhance mitochondrial biogenesis and bolster enzymatic antioxidant defenses. The let‐7c‐3p/TFAM axis represents a primary mechanism by which EPA targets mitochondrial function. EPA was shown to downregulate let‐7c‐3p, which was validated as a direct post‐transcriptional repressor of TFAM, a master regulator of mitochondrial biogenesis. Consistent with this, inhibition of let‐7c‐3p or overexpression of TFAM recapitulated the effects of EPA, leading to increased mtDNA copy number, elevated expression of mitochondrial‐encoded genes, and improved mitochondrial function, as evidenced by a higher ADP/ATP ratio and reduced mitochondrial ROS production. These results align with previous reports demonstrating that TFAM overexpression enhances mitochondrial function and attenuates oxidative stress in various cell types. The work extends these findings by identifying let‐7c‐3p as an upstream epigenetic regulator that mediates EPA's effects on the mitochondria, effectively reducing ROS production at its primary intracellular source. Consistent with these functional observations, western blot analysis confirmed that let‐7c‐3p mimic decreased TFAM protein levels, while EPA treatment increased TFAM protein in a concentration‐dependent manner (Figure 5A–B), further supporting the role of this axis at the protein level. In parallel, the miR‐34c‐5p/SIRT1 axis was delineated as a complementary pathway that enhances the cellular capacity to neutralize ROS. EPA was found to upregulate miR‐34c‐5p, which was confirmed to directly target the 3′‐UTR of SIRT1. Activation of this axis, either by miR‐34c‐5p overexpression or SIRT1 knockdown, significantly increased the activities of key antioxidant enzymes, including CAT, SOD, and GPx (Figure 4C–D). This finding is particularly intriguing given the complex and context‐dependent role of SIRT1 in oxidative stress regulation. This interpretation is further supported by protein‐level and rescue evidence. miR‐34c‐5p mimic significantly reduced SIRT1 protein levels, and EPA treatment decreased SIRT1 protein in a concentration‐dependent manner (Figure 5C–D). Importantly, miR‐34c‐5p inhibition completely abolished EPA‐induced increases in CAT, SOD, and GPx activities (Figure 4E), confirming that miR‐34c‐5p is required for EPA to enhance enzymatic antioxidant defenses. Although SIRT1 is canonically considered a positive regulator of antioxidant defenses, its function is highly context‐dependent. Direct evidence shows that SIRT1 inhibition can, under certain conditions, reduce oxidative stress and enhance cell survival [25]. A plausible mechanism involves NRF2, as SIRT1 inhibition promotes NRF2 nuclear accumulation and transcriptional activity by reducing its acetylation [26]. Consistent with this, studies have shown that SIRT1 inhibition can enhance NRF2 activity and promote antioxidant enzyme expression in certain cellular contexts [27]. These results suggest that the miR‐34c‐5p/SIRT1 axis operates as a switch to amplify the antioxidant enzyme arsenal, thereby increasing the cell's capacity to clear existing ROS. While the precise mechanism linking SIRT1 repression to increased CAT, SOD, and GPx activities remains to be fully elucidated, the context‐dependent nature of SIRT1 function, as supported by the above literature, provides a reasonable framework for interpreting our findings. Future studies directly examining NRF2 acetylation status and nuclear translocation in this system would further strengthen the mechanistic understanding. The mechanistic architecture of EPA's antioxidant action is substantiated by complementary loss‐ and gain‐of‐function experiments. EPA's effect is functionally dependent on miRNA changes: overexpression of let‐7c‐3p abolishes EPA's ROS‐lowering effect, while inhibition of miR‐34c‐5p similarly blocks EPA's enhancement of antioxidant capacity. This causality is extended to the direct mRNA targets. The effects of let‐7c‐3p inhibition, such as increased mtDNA copy number and reduced mitochondrial ROS, are phenocopied by TFAM overexpression. Conversely, miR‐34c‐5p‐mediated enhancement of enzymatic defenses is recapitulated by SIRT1 knockdown. This dual‐layer evidence, showing that manipulation of either the miRNA or its downstream target yields the same outcome, validates that these axes are instrumental for EPA's antioxidant properties. These findings illustrate how EPA orchestrates a multi‐layered antioxidant response through coordinated modulation of distinct gene networks, extending beyond previous reports of individual miRNA‐mediated pathways. Despite these advances, several questions remain. The upstream signaling pathways through which EPA regulates the expression of let‐7c‐3p and miR‐34c‐5p are yet to be elucidated. Potential involvement of transcription factors such as NF‐κB or NRF2, or epigenetic modifiers responsive to lipid signals, warrants further investigation [28]. Additionally, while the data demonstrate the sufficiency of each axis, exploring the potential synergistic effects of their combined modulation could have therapeutic implications [29, 30]. Most importantly, the physiological relevance of these findings needs to be validated in vivo using animal models of oxidative stress‐related diseases, such as metabolic dysfunction‐associated steatotic liver disease (MASLD) [31, 32]. Future studies should also investigate whether miR‐34c‐5p‐mediated SIRT1 repression affects other downstream targets (e.g., FOXO3a, PGC‐1α) and whether other bioactive food components share similar miRNA‐mediated mechanisms [33]. Several limitations of this study should be acknowledged. First, all experiments were conducted exclusively in HepG2 cells, a human hepatoma cell line. While HepG2 cells are a well‐established model for studying hepatic oxidative stress and lipid metabolism, the findings may not fully recapitulate the responses of primary human hepatocytes or other cell types. Nevertheless, the high transfection efficiency and consistent responses of HepG2 cells make them particularly suitable for mechanistic investigations of the type presented here. Future studies using primary hepatocytes or in vivo models are warranted to validate the physiological relevance of these miRNA‐mediated axes. Second, the EPA concentrations used (100‐150 µM) are higher than typical plasma levels in humans, which are in the low micromolar range. However, such concentrations are consistent with those commonly employed in in vitro mechanistic studies to achieve detectable effects within a short treatment window (24‐48 h). Dose‐response studies using lower, more physiologically relevant concentrations would be valuable to confirm the translational potential of these findings. Third, while rescue experiments have confirmed the necessity of the miR‐34c‐5p/SIRT1 axis for EPA‐induced enzymatic antioxidant defenses, the potential synergistic effects between the two axes and their upstream regulatory mechanisms remain to be explored. Nevertheless, the complementary loss‐ and gain‐of‐function approaches presented in this study collectively establish the necessity and sufficiency of each miRNA‐target axis. Further investigations incorporating epistatic rescue designs would provide additional causal evidence. Conclusion In conclusion, this study reveals an epigenetic mechanism underlying the antioxidant activity of EPA. EPA orchestrates a coordinated, multi‐faceted antioxidant response by modulating two distinct miRNA‐mRNA axes: downregulation of let‐7c‐3p relieves repression of TFAM to drive mitochondrial biogenesis and reduce ROS production at its source, while upregulation of miR‐34c‐5p represses SIRT1 to enhance the activity of core antioxidant enzymes. Protein‐level validation and rescue experiments further establish the functional necessity of these axes in mediating EPA's antioxidant effects. These findings advance the understanding of how omega‐3 PUFAs exert their beneficial effects and highlight the potential of targeting specific miRNA pathways for nutritional and therapeutic interventions against oxidative stress‐related diseases. Author Contributions Haiqi Wang: methodology, writing original draft, writing review and editing. Yanqi Lin: methodology, investigation, data curation, formal analysis. Yuanyuan Li: investigation, data curation, formal analysis. Xiaowen Wu: investigation, data curation, formal analysis. Jingyu Lin: investigation, formal analysis. Shiying Huang: conceptualization, supervision, writing review, funding acquisition. Guiling Li: conceptualization, supervision, writing review and editing, funding acquisition. Conflicts of Interest The authors have nothing to report.