Role of the UHRF1–KLHL6–CORO2B axis in obesity‐related insulin resistance

ABSTRACT

Background

Obesity‐related insulin resistance (IR) involves mitochondrial dysfunction and ferroptosis dysregulation. Ubiquitin‐like with PHD and RING finger domains 1 (UHRF1), an epigenetic regulator of DNA methylation, mediates crucial functions in disease progression, though its mechanisms in IR remain unclear.

Methods

Using murine IR models and tumor necrosis factor‐α (TNF‐α)‐induced adipocyte systems, UHRF1's regulation of mitophagy and ferroptosis was assessed. Bioinformatics identified downstream targets Kelch‐like protein 6 (KLHL6) and coronin 2B (CORO2B), prompting examination of the UHRF1‐KLHL6‐CORO2B axis in obesity‐related IR.

Results

Obesity‐related IR mice exhibited increased body weight, impaired glucose and insulin tolerance, adipose tissue inflammation, impaired mitophagy, and alterations in ferroptosis‐related parameters, accompanied by reduced UHRF1 expression. In vitro, TNF‐α‐treated adipocytes showed decreased mitophagy, elevated oxidative stress, and changes consistent with ferroptosis, along with lower UHRF1 levels. UHRF1 overexpression was associated with changes in PINK1/Parkin‐related markers, cell viability, oxidative stress indicators, ferroptosis‐related parameters, and inflammatory responses. UHRF1 overexpression was associated with increased KLHL6 DNA methylation, decreased KLHL6 expression, and alterations in CORO2B ubiquitination.

Conclusions

UHRF1 overexpression was associated with decreased KLHL6 expression, altered CORO2B ubiquitination, and changes in autophagy‐, ferroptosis‐, and inflammation‐related markers. These results provide insights into potential pathogenic mechanisms and suggest that UHRF1, KLHL6, and CORO2B may serve as candidate targets for further investigation in obesity‐related IR.

In high‐fat diet (HFD)‐induced obese mice with insulin resistance, overexpression of UHRF1 may recruit DNMT1 to promote DNA methylation of KLHL6, leading to its downregulation. Subsequently, suppression of KLHL6 may decrease the ubiquitination of CORO2B, resulting in elevated CORO2B protein levels. This process may attenuate inflammation and ferroptosis in insulin‐resistant mice, ultimately ameliorating the disease. HFD, high‐fat diet; UHRF1, Ubiquitin (Ub)‐like with PHD and RING finger domains 1; KLHL6, Kelch‐like protein 6; DNMT1, DNA (cytosine‐5)‐methyltransferase 1; CORO2B, coronin‐2B; Ub, Ubiquitin.

INTRODUCTION

Obesity‐related insulin resistance (IR), a metabolic disorder characterized by diminished insulin responsiveness in peripheral tissues and consequent hyperglycemia [1, 2], is mechanistically linked to mitochondrial dysfunction [3]. Mitophagy, the selective autophagic clearance of damaged mitochondria, confers protection against IR by maintaining cellular homeostasis [4]. Conversely, ferroptosis, an iron‐dependent form of regulated cell death driven by lipid peroxidation, exacerbates mitochondrial impairment and accelerates IR progression [5]. Consequently, modulating the interplay between mitophagic cytoprotection and ferroptotic injury represents a critical therapeutic axis in IR. Elucidating key regulatory mechanisms governing this balance may yield novel insights for IR intervention strategies.

Ubiquitin (Ub)‐like with PHD and RING finger domains 1 (UHRF1) serves as a pivotal epigenetic regulator that sustains DNA methylation patterns [6], thereby precisely controlling gene expression [7, 8]. Emerging evidence has highlighted the significant role of UHRF1 in metabolic pathophysiology, exemplified by findings from Vakayil et al. demonstrating its capacity to promote adipogenesis while attenuating fibrosis through suppression of the GPNMB and TGF‐β signaling pathways, thereby preserving adipose tissue functionality [9]. These observations suggest that UHRF1 may alleviate obesity‐related IR by improving adipocyte performance and mitigating metabolic dysregulation.

Kelch‐like protein 6 (KLHL6) is an E3 Ub ligase that regulates substrate protein degradation via the Ub‐proteasome system [10], thereby maintaining proteostasis and intracellular homeostasis [11]. Recent research has revealed abnormal expression of KLHL6 in obesity‐related pathologies [12], which may significantly disrupt the activity and function of downstream effector molecules. Nevertheless, the precise contribution of KLHL6 to obesity‐related IR remains undefined. In parallel, coronin‐2B (CORO2B), a protein closely associated with cytoskeletal dynamics, has been preliminarily studied in some metabolic diseases. One study indicated that CORO2B expression was significantly reduced in diabetic nephropathy [13]. Its functional relevance to IR, however, constitutes uncharted territory in metabolic research.

In this study, we explored the UHRF1–KLHL6–CORO2B axis to investigate its potential role in the pathophysiology of obesity‐related IR, providing insights into underlying mechanisms and possible targets for further investigation.

METHODS

Animal modeling

Eight‐week‐old male C57BL/6J mice were purchased from Vital River Laboratory Animal Technology Co., Ltd. (Beijing) and acclimatized for 7 days. Mice were randomly assigned into the following groups: Control, IR, Adeno‐associated virus (AAV)‐oe‐NC, AAV‐oe‐UHRF1, AAV‐oe‐UHRF1 + AAV‐oe‐NC, AAV‐oe‐UHRF1 + AAV‐oe‐KLHL6, AAV‐sh‐NC, AAV‐sh‐KLHL6, AAV‐sh‐KLHL6 + AAV‐sh‐NC, and AAV‐sh‐KLHL6 + AAV‐sh‐CORO2B, with six mice per group. The Control group received a standard diet (10% fat), whereas all other groups were administered a high‐fat diet (HFD; 60% fat) to establish an obesity‐related IR model. On the day before model establishment, mice designated for viral intervention (oe‐NC, oe‐UHRF1, oe‐KLHL6, sh‐NC, sh‐KLHL6, and sh‐CORO2B groups) underwent anesthesia followed by intraperitoneal injection of 1.0 × 10¹⁰ v.g of the respective AAVRec2 construct [14]. During the 12‐week feeding period, body weight was monitored. Upon completion, omental adipose tissue (OMAT) samples were collected from the mice and stored for further analysis [15].

Glucose (GTT) and insulin tolerance test (ITT)

After fasting for 12 h, mice underwent an intraperitoneal GTT via injection of glucose (0.75 g/kg body weight). Blood glucose levels were measured at 0, 30, 60, and 120 min postinjection. For ITT, mice were fasted for 2 h and then injected with insulin (HumulinR, 0.75 U/kg body weight). Blood glucose concentrations were similarly assessed at the same time points following insulin administration.

Hematoxylin and eosin (HE) staining

Adipose tissue samples were fixed, embedded in paraffin, and sectioned into 4 μm slices. The sections were deparaffinized in xylene, dehydrated in graded ethanol, and stained with hematoxylin followed by eosin. Following dehydration and clearing in xylene, the samples were mounted with neutral resin. Tissue pathology was observed under an optical microscope (CX43, Olympus, Tokyo, Japan).

Cell culture and treatment

Mouse 3T3‐L1 pre‐adipocytes (ATCC) were cultured in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS). Adipogenic differentiation was initiated when cells reached confluence. Cells were cultured in induction medium I (DMEM containing 10 μg/mL insulin, 0.5 mM 1‐methyl‐3‐isobutylxanthine, and 1 μg/mL dexamethasone) for 24 h, followed by incubation in fresh DMEM for an additional 24 h. Cells were then maintained in induction medium II (DMEM supplemented with 10 μg/mL insulin) for 48 h. After differentiation, cells were maintained in DMEM containing 10% FBS with medium changes every 2 days until the majority of cells exhibited a mature lipid‐laden morphology. To establish an IR model, mature adipocytes were treated with tumor necrosis factor‐α (TNF‐α, 10 ng/mL) for 24 h, as described previously [16]. For gene modulation, mature adipocytes were transiently transfected with overexpression plasmids (oe‐NC, oe‐UHRF1, oe‐KLHL6) or short hairpin RNA plasmids (sh‐NC, sh‐KLHL6, sh‐CORO2B) using Lipofectamine 3,000 reagent (GenePharma, Shanghai, China) according to the manufacturer's instructions.

Oil red O staining

Mature adipocytes were rinsed with phosphate‐buffered saline (PBS) and fixed in 4% paraformaldehyde for 30 min. Following fixation, cells were incubated for 30 min with Oil Red O working solution prepared via mixing the Oil Red O stock solution with distilled water (3:2 ratio), centrifugation at 250 × g for 4 min, and collection of the supernatant. Images were captured under a microscope (Olympus).

Glucose uptake assay

Cellular glucose uptake was measured using the Glucose Uptake Assay Kit™ (Promega, Cat. No. J1342) according to the manufacturer's instructions. Briefly, mature adipocytes were incubated with 50 μL of 1 mM 2‐deoxy‐D‐glucose (2DG) per well for 10 min. The reaction was terminated by adding stop buffer, followed by neutralization buffer and 2‐deoxyglucose‐6‐phosphate (2DG6P) detection reagent. After incubation for 1 h at room temperature, luminescence was measured using a microplate luminometer (TECAN).

Enzyme‐linked immunosorbent assay (ELISA)

Tissue concentrations of interleukin (IL)‐6 and IL‐1β, as well as serum insulin levels, were quantified using ELISA kits per manufacturer's protocols (IL‐6: 98027ES48, Yeasen; IL‐1β: D721017, Sangon; insulin: JL11459, JONLNBIO, Shanghai).

Cell Counting Kit‐8 (CCK‐8) assay

Cells were plated in 96‐well culture plates at a density of 5,000 cells/well. Following experimental treatment, 10% CCK‐8 reagent was added to each well followed by incubation for 3 h. Optical density (OD) was measured at 450 nm using a microplate reader.

Reverse transcription‐quantitative polymerase chain reaction (RT‐qPCR)

Total RNA was extracted using the Quick RNA Extraction Kit (DP419, Jiachu Biotechnology Co., Ltd). RT‐qPCR was performed using a fluorescence quantitative PCR kit (QR0100, Sigma‐Aldrich, St. Louis, USA), with glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) as the single internal reference gene. Relative mRNA expression levels were calculated via the 2‐ΔΔCt method. Primer sequences are provided in Table 1.

Table 1.

Primer sequences

Gene	Direction	Sequence (5′‐3′)
UHRF1	F	AGACCTCTCTGGCAACAAGCGT
	R	AGTCTTCAGCCTCCGCACCTTT
KLHL6	F	GCCAGGATGTACCACCTTTCTG
	R	CAGGTTACCTCTGCCACAAATCG
CORO2B	F	GGTGGAACACAAGACAGATCGC
	R	CTTTCCAGCCAGGTAGAGCATG
GAPDH	F	CATCACTGCCACCCAGAAGACTG
	R	ATGCCAGTGAGCTTCCCGTTCAG

F, Forward; R, Reverse.

Western blot (WB)

Protein lysates were extracted from tissues and cultured cells using RIPA lysis buffer (ab170197, Abcam) supplemented with protease inhibitors. Protein concentrations were determined using a bicinchoninic acid (BCA) protein assay kit (P0010, Beyotime). Equal amounts of protein were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and transferred onto polyvinylidene difluoride (PVDF) membranes (ab133411, Abcam) using a wet transfer system. Membranes were blocked with 5% nonfat dry milk or 5% bovine serum albumin (BSA), depending on the primary antibody. For phospho‐protein detection, 5% BSA was used as the blocking reagent. The membranes were then incubated overnight at 4°C with the following primary antibodies: anti‐UHRF1 (1:1,000, ab303552, Abcam), anti‐KLHL6 (1:100, KL‐16774R, Klbscience, Shanghai, China), anti‐CORO2B (1:1,000, 13,802‐1‐AP, Proteintech), anti‐Akt (1:1,000, 9,272, Cell Signaling Technology), anti‐phospho‐Akt (Ser473; 1:1,000, 9,271, Cell Signaling Technology), anti‐PTEN‐induced putative kinase 1 (PINK1; 1:1,000, 23,274‐1‐AP, Proteintech), anti‐Parkin RBR E3 ubiquitin protein ligase (Parkin; 2.5 μg/mL, 702,785, Thermo Fisher Scientific), anti‐LC3 II/I (1:1,000, 12,135‐1‐AP, Proteintech), anti‐glutathione peroxidase 4 (GPX4; 1:1,000, MA5‐32827, Thermo Fisher Scientific), anti‐solute carrier family 7 member 11 (SLC7A11; 1:1,000, PA1‐16893, Thermo Fisher Scientific), and anti‐GAPDH (1:2,000, ab181602, Abcam). After washing with TBST, membranes were incubated with the appropriate horseradish peroxidase–conjugated secondary antibody (1:2,000, ab205718, Abcam) for 2 h at room temperature. Protein bands were visualized using enhanced chemiluminescence (ECL) detection reagents (A38554, Thermo Fisher Scientific) and quantified using ImageJ software. GAPDH was used as a loading control, and relative protein expression levels were calculated accordingly.

Reactive oxygen species (ROS) detection

Tissue staining for ROS was performed using a ROS Detection Kit (KL1470516A, Klbio, Shanghai, China), while cellular ROS staining was conducted using a separate kit (MB4682, Meilunbio, Dalian, China).

Fe²⁺ content quantification

Tissue Fe²⁺ levels were measured using a colorimetric assay kit (E‐BC‐K304‐S, Elabscience, Wuhan, China), and cellular Fe²⁺ levels were assessed with an iron assay kit (EEA009, Thermo Fisher Scientific) according to the manufacturers' instructions. Briefly, samples were processed, reacted with a chromogenic reagent, and the resulting optical density (OD) was measured at the specified wavelength using a microplate reader to calculate Fe²⁺ concentrations.

Biochemical analysis

Superoxide dismutase (SOD) activity and malondialdehyde (MDA) concentration in tissue homogenates and cell lysates were assayed using the SOD Assay Kit (D799594, Sangon) and MDA Assay Kit (MAK085, Sigma‐Aldrich), respectively, following the manufacturer's instructions.

Methylated DNA immunoprecipitation‐qPCR (MeDIP‐qPCR)

Genomic DNA was isolated from cells using the Qiagen DNA Extraction Kit (QIAGEN, Germany). MeDIP was then performed with the MeDIP Kit (Active Motif, California, USA) as per the provided protocol. qPCR was used to amplify the immunoprecipitated DNA to assess the DNA methylation level of KLHL6.

ChIP‐qPCR

According to the instructions of the Simple ChIP Enzymatic Chromatin IP Kit (9,003, Cell Signaling Technology), cells were first cross‐linked with formaldehyde for 15 min, and the protein–chromatin cross‐linking was terminated with glycine. The chromatin was then digested with micrococcal nuclease and resuspended in ChIP buffer. Subsequently, immunoprecipitation was performed at 4°C using UHRF1 antibody (PA5‐29884, Thermofisher), DNMT1 antibody (5,032, Cell Signaling Technology), or IgG control antibody (A7016, Beyotime, Shanghai, China). Finally, chromatin was eluted, cross‐links were reversed, and DNA was purified for qPCR analysis.

Co‐immunoprecipitation (Co‐IP)

Cells were lysed in Pierce IP Lysis Buffer (87,787, Thermo Fisher Scientific) supplemented with a protease inhibitor cocktail (Roche, Switzerland). Equal amounts of total protein were subjected to immunoprecipitation using anti‐KLHL6 antibody, anti‐CORO2B antibody, or control IgG (1 μg, 30,000‐0‐AP, Proteintech), which were pre‐conjugated to Protein A/G PLUS‐Agarose beads (20,423, Thermo Fisher Scientific, MA, USA). The bead–antibody–protein complexes were collected, boiled in SDS loading buffer, and analyzed by Western blotting to assess protein–protein interactions in the immunoprecipitates. Input lysates were used as controls.

Ubiquitination detection

Cells were transfected or co‐transfected with Myc‐tagged CORO2B, hemagglutinin (HA)‐tagged ubiquitin (Ub), and Flag‐tagged KLHL6 expression plasmids. After 24 h, cells were lysed in RIPA buffer supplemented with protease inhibitors and centrifuged to remove cellular debris. The clarified supernatants were subjected to ubiquitination assays. Immunoprecipitation was performed using an anti‐HA antibody to enrich ubiquitinated proteins, followed by WB analysis using anti‐Myc (1:250, ab9106, Abcam) to detect ubiquitinated CORO2B and anti‐HA (1:1,000, ab236632, Abcam) to confirm ubiquitin conjugation. Anti‐Flag antibody (1:1,000, SAB4301135, Thermo Fisher Scientific, MA, USA) was used to assess KLHL6 expression in input samples. The ubiquitination level of CORO2B was compared in the presence or absence of KLHL6.

Protein stability assay

Cells were subjected to CHX (20 μg/mL) administration for varying durations (0, 15, 30, 60, 120, 240 min), and protein stability was assessed by WB.

Statistical analysis

Statistical analyses were conducted using GraphPad Prism version 9.0. Normality of all datasets was first analyzed using the Shapiro–Wilk test. A t‐test was used for comparisons between two groups, and one‐way analysis of variance (anova) followed by Tukey's post hoc test for comparisons among three or more groups. A significance threshold of P < 0.05 was applied throughout the study. All experiments incorporated at least three biological replicates, with quantitative data expressed as mean ± standard deviation (SD).

RESULTS

UHRF1 downregulation and its potential link to mitophagy and ferroptosis in obesity‐related IR

To elucidate the role of UHRF1 in obesity‐related IR, we characterized metabolic and molecular perturbations in murine models. Compared with the Control group, mice in the IR group exhibited significantly increased body weight (Figure 1a) accompanied by evident glucose metabolism disorders in GTT (Figure 1b) and reduced insulin sensitivity in ITT (Figure 1c). In addition, serum insulin levels were markedly elevated in the IR group relative to controls (Figure 1d). Histopathological analysis via HE staining revealed pronounced adipocyte hypertrophy in the IR group compared with the Control group (Figure 1e). ELISA assays showed significantly elevated levels of the inflammatory cytokines IL‐6 and IL‐1β in the serum of IR mice (Figure 1f). Furthermore, WB analysis indicated a marked reduction in the p‐Akt/Akt ratio in the IR group relative to controls (Figure 1g), suggesting impaired insulin signaling. WB analysis showed that the expression of mitophagy‐related proteins PINK1 and Parkin was downregulated in the adipose tissue of IR mice, along with a decreased LC3II/I ratio, indicating impaired mitophagy (Figure 1h). Concurrently, IR mice exhibited compromised antioxidant capacity with significantly reduced SOD activity alongside increased MDA, Fe²⁺ (Figure 1i), and ROS levels (Figure 1j). Downregulation of ferroptosis‐inhibitory proteins GPX4 and SLC7A11 was also observed (Figure 1k). Further RT‐qPCR (Figure 1l) and WB (Figure 1m) analyses confirmed a significant downregulation of UHRF1 expression in the adipose tissue of IR mice.

Figure 1.

UHRF1 Downregulation and possible effects on mitophagy and ferroptosis in adipose tissue during obesity‐related insulin resistance. (a) Body weight monitoring of experimental mice; (b) glucose tolerance test (GTT) curves; (c) insulin tolerance test (ITT) curves; (d) serum insulin levels were measured. (e) hematoxylin and eosin (HE) staining of adipose tissues; (f) ELISA detection of IL‐6 and IL‐1β levels in serum; (g) the p‐Akt/Akt levels were measured by Western blot; (h) Western blot analysis of mitophagy‐related proteins PINK1, Parkin, and LC3 in adipose tissues; (i) colorimetric analysis of SOD activity, MDA content, and Fe²⁺ levels in adipose tissues; (j) ROS levels detected by fluorescence staining; (k) Western blot analysis of ferroptosis‐related proteins GPX4 and SLC7A11 in adipose tissues; (l) RT‐qPCR analysis of UHRF1 mRNA expression; (m) Western blot analysis of UHRF1 protein expression. N = 6, ****P < 0.0001.

Association of UHRF1 overexpression with autophagy‐ and ferroptosis‐related alterations in insulin‐resistant adipocytes

To model obesity‐related IR in vitro, mature adipocytes differentiated from 3T3‐L1 cells were treated with TNF‐α. The role of UHRF1 overexpression in this context was subsequently investigated. First, microscopic observation (Figure 2a) and Oil Red O staining (Figure 2b) confirmed successful differentiation of 3T3‐L1 cells into mature adipocytes. Glucose uptake experiments showed that TNF‐α treatment obviously impaired glucose uptake at 24 h (Figure 2c), indicating successful modeling. Further analysis revealed that TNF‐α significantly reduced cell viability as measured by CCK‐8 assay (Figure 2d). ELISA results showed a significant increase in the pro‐inflammatory cytokines IL‐6 and IL‐1β after TNF‐α treatment (Figure 2e). WB analysis indicated that TNF‐α markedly downregulated PINK1 and Parkin expression and decreased the LC3II/I ratio in adipocytes (Figure 2f), coinciding with increased intracellular ROS levels (Figure 2g). This inflammatory milieu was accompanied by reduced SOD activity and elevated MDA and Fe²⁺ levels (Figure 2h), alongside downregulation of ferroptosis‐related proteins GPX4 and SLC7A11 (Figure 2i). RT‐qPCR (Figure 2j) and WB (Figure 2k) further confirmed that UHRF1 expression was markedly reduced in TNF‐α‐treated cells, suggesting a negative regulatory role for UHRF1 in this model. To further elucidate the function of UHRF1, cells were transfected with an overexpression construct. RT‐qPCR (Figure 2l) and WB (Figure 2m) validated successful overexpression of UHRF1. Additionally, the autophagy inhibitor 3‐methyladenine (3‐MA) was added, which did not alter UHRF1 expression. CCK‐8 results showed that UHRF1 overexpression significantly improved cell viability, but this improvement was reversed by 3‐MA (Figure 2n). Furthermore, overexpression of UHRF1 significantly reduced IL‐6 and IL‐1β levels, while 3‐MA treatment abolished these effects (Figure 2o), indicating a potential anti‐inflammatory role. WB analysis demonstrated that UHRF1 overexpression significantly increased the levels of mitophagy‐related proteins PINK1 and Parkin, as well as the LC3II/I ratio, whereas these changes were abrogated by 3‐MA (Figure 2p). Concordantly, ROS analysis further showed that UHRF1 overexpression significantly reduced oxidative stress, but ROS levels increased again upon 3‐MA treatment (Figure 2q). Biochemical analysis showed that UHRF1 overexpression was associated with a significant increase in SOD activity and concomitant decreases in MDA and Fe²⁺ levels, whereas these effects were attenuated following 3‐MA treatment (Figure 2r). WB analysis showed that GPX4 and SLC7A11 expression was increased following UHRF1 overexpression but was reduced upon 3‐MA treatment (Figure 2s), suggesting that autophagy‐related processes may be involved in the protective effects of UHRF1 against TNF‐α–induced cellular injury.

Figure 2.

UHRF1 overexpression alleviates TNF‐α‐induced insulin resistance via mitophagy‐dependent ferroptosis suppression. (a) Microscopic observation of differentiated 3T3‐L1 adipocytes; (b) Successful differentiation of mature adipocytes confirmed by Oil Red O staining; (c) Glucose uptake assay evaluating TNF‐α‐induced insulin resistance; (d) CCK‐8 assay to detect cell viability; (e) ELISA to measure IL‐6 and IL‐1β levels; (f) Western blot analysis of mitophagy‐related proteins PINK1, Parkin, and LC3; (g) Fluorescence detection of intracellular ROS levels; (h) Biochemical analysis of SOD activity, MDA content, and Fe²⁺ levels; (i) Western blot detection of ferroptosis‐inhibitory proteins GPX4 and SLC7A11; (j) RT‐qPCR detection of UHRF1 mRNA expression; (k) Western blot detection of UHRF1 protein expression; (l) RT‐qPCR validating UHRF1 overexpression; (m) Western blot validating UHRF1 overexpression; (n) CCK‐8 assay measuring cell viability after UHRF1 overexpression and 3‐MA treatment; (o) ELISA detection of inflammatory cytokine changes; (p) Western blot detection of mitophagy proteins under different treatments; (q) Fluorescence intensity analysis of ROS levels; (r) Changes in SOD activity, MDA content, and Fe²⁺ levels; (s) Western blot detection of GPX4 and SLC7A11 expression. N = 3, ns P > 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

UHRF1 affects KLHL6 expression

Bioinformatic analysis was conducted using obesity‐related IR datasets from GEO GSE162653 (differentially expressed genes) and GSE67024 (DNA methylation) under stringent criteria (P < 0.01, |logFC| > 0) (Figure 3a). Intersection starting from either GSE162653 or GSE67024 yielded 164 common genes (Figure 3b). Taking the top 20 factors from each dataset for intersection resulted in two overlapping genes: TGOLN2 and KLHL6 (Figure 3c). Given KLHL6's established role as an E3 Ub ligase with emerging implications in diabetes pathogenesis, especially type 1 diabetes [12], and its undefined relationship with obesity‐related IR or epigenetic regulation by UHRF1, we prioritized it for further investigation. In IR models, both RT‐qPCR (Figure 3d) and WB (Figure 3e) analyses showed that KLHL6 expression was significantly increased in adipose tissues. TNF‐α‐treated adipocytes similarly exhibited elevated KLHL6 expression, while UHRF1 overexpression suppressed this induction (Figure 3f,g), indicating UHRF1‐mediated repression of KLHL6. Further MeDIP‐qPCR analysis confirmed enhanced KLHL6 promoter methylation following UHRF1 overexpression (Figure 3h). ChIP‐qPCR results showed that the binding of DNMT1 and UHRF1 to the KLHL6 promoter was significantly increased in the oe‐UHRF1 group compared with the oe‐NC group (Figure 3i), whereas the IgG control showed no enrichment.

Figure 3.

UHRF1 was associated with KLHL6 expression. (a) Volcano plots of differential gene expression (GSE162653) and DNA methylation (GSE67024) in obesity‐related insulin resistance; (b) Venn diagram showing the intersection of differentially expressed genes from both datasets; (c) Overlap of the top 20 significantly altered candidates identifying TGOLN2 and KLHL6; (d) RT‐qPCR analysis of KLHL6 mRNA expression in adipose tissues of insulin‐resistant (IR) mice; (e) Western blot analysis of KLHL6 protein expression in adipose tissues of IR mice; (f) RT‐qPCR analysis of KLHL6 mRNA levels in mature adipocytes after TNF‐α treatment and UHRF1 overexpression; (g) Western blot analysis of KLHL6 protein levels in mature adipocytes after TNF‐α treatment and UHRF1 overexpression; (h) MeDIP‐qPCR analysis of DNA methylation levels in the KLHL6 promoter region. (i) ChIP‐qPCR detection of DNMT1 binding at the KLHL6 promoter. IgG served as the negative control. N = 3, ****P < 0.0001.

Association of UHRF1 and KLHL6 with cellular ferroptosis

In vitro results showed that compared with the oe‐UHRF1 + oe‐NC group, the oe‐UHRF1 + oe‐KLHL6 group exhibited significantly increased KLHL6 mRNA and protein expression levels, without markedly altering UHRF1 expression, confirming successful KLHL6 overexpression (Figure 4a,b). Subsequent assessment via CCK‐8 assay demonstrated that KLHL6 overexpression substantially decreased cell viability (Figure 4c). Concomitantly, inflammatory cytokines IL‐6 and IL‐1β levels were significantly elevated after KLHL6 overexpression (Figure 4d). Furthermore, the expression of key mitophagy‐related proteins PINK1, Parkin, and LC3II/I significantly decreased following KLHL6 overexpression (Figure 4e), while intracellular ROS levels markedly increased (Figure 4f). Biochemical analysis showed that KLHL6 overexpression significantly reduced SOD activity while increasing MDA and Fe²⁺ levels (Figure 4g), accompanied by a significant downregulation of ferroptosis‐inhibitory proteins GPX4 and SLC7A11 (Figure 4h).

Figure 4.

UHRF1 is associated with ferroptosis‐related cellular injury in relation to KLHL6. (a) RT‐qPCR analysis of UHRF1 and KLHL6 mRNA expression in 3T3‐L1 cells to validate KLHL6 overexpression efficiency and its effect on UHRF1 expression; (b) Western blot analysis of UHRF1 and KLHL6 protein expression in 3T3‐L1 cells; (c) CCK‐8 assay to evaluate the effect of KLHL6 overexpression on cell viability; (d) ELISA to detect the expression levels of inflammatory cytokines IL‐6 and IL‐1β; (e) Western blot analysis of mitophagy‐related proteins PINK1, Parkin, and LC3II/I; (f) measurement of intracellular ROS levels; (g) biochemical assays of SOD activity, MDA content, and Fe²⁺ levels in cells; (h) Western blot analysis of ferroptosis‐inhibitory proteins GPX4 and SLC7A11. N = 3, ns P > 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Overexpression of KLHL6 is associated with insulin resistance, ferroptosis, and mitochondrial dysfunction in mice

In vivo, compared with the AAV‐oe‐NC group, the AAV‐oe‐UHRF1 group showed significantly elevated UHRF1 mRNA and protein levels in adipose tissues, with a marked decrease in KLHL6 expression. Subsequent KLHL6 overexpression restored KLHL6 levels without significant effect on UHRF1 expression (Figure 5a,b). Body weight records showed that mice in the AAV‐oe‐UHRF1 group weighed significantly less than those in the AAV‐oe‐NC group, whereas KLHL6 overexpression reversed this weight reduction (Figure 5c). GTT and ITT results indicated that UHRF1 overexpression significantly improved whole‐body glucose metabolism and insulin sensitivity in mice, while these effects were abrogated by simultaneous KLHL6 overexpression (Figure 5d,e). In addition, serum insulin levels were markedly reduced by UHRF1 overexpression, and this reduction was reversed upon KLHL6 overexpression (Figure 5f). Further HE staining revealed that adipocyte hypertrophy was ameliorated in the AAV‐oe‐UHRF1 group compared with AAV‐oe‐NC, but KLHL6 overexpression re‐induced adipocyte hypertrophy (Figure 5g). Meanwhile, UHRF1 overexpression effectively reduced IL‐6 and IL‐1β levels in tissues, but this effect was reversed by KLHL6 overexpression (Figure 5h). UHRF1 overexpression markedly increased the p‐Akt/Akt ratio, whereas KLHL6 overexpression abolished this effect (Figure 5i). Additionally, mitophagy‐related proteins PINK1, Parkin, and LC3II/I were significantly upregulated in the AAV‐oe‐UHRF1 group compared with AAV‐oe‐NC, but downregulated following KLHL6 overexpression (Figure 5j). ROS levels decreased significantly after UHRF1 overexpression, and this reduction was counteracted by KLHL6 overexpression (Figure 5k). Furthermore, compared with AAV‐oe‐NC, SOD activity increased and MDA and Fe²⁺ levels decreased significantly in the AAV‐oe‐UHRF1 group; however, these beneficial trends were likewise reversed by KLHL6 overexpression (Figure 5l). Finally, ferroptosis‐inhibitory proteins GPX4 and SLC7A11 were significantly increased after UHRF1 overexpression but subsequently decreased upon KLHL6 overexpression (Figure 5m).

Figure 5.

Effects of KLHL6 overexpression on insulin resistance, ferroptosis, and mitochondrial function in mice. (a) RT‐qPCR analysis of UHRF1 and KLHL6 mRNA expression in mouse adipose tissues; (b) Western blot analysis of UHRF1 and KLHL6 protein expression in mouse adipose tissues; (c) Body weight change curves during the experimental period; (d) Glucose tolerance test (GTT) to evaluate glucose metabolism; (e) Insulin tolerance test (ITT) to evaluate insulin sensitivity; (f) Serum insulin was measured. (g) HE staining to observe morphological changes in adipose tissues; (h) ELISA to detect IL‐6 and IL‐1β levels in adipose tissues; (i) p‐Akt/Akt levels were measured by Western blot. (j) Western blot analysis of mitophagy‐related protein expression in adipose tissues; (k) Detection of ROS levels in tissue samples; (l) Measurement of SOD activity, MDA content, and Fe²⁺ levels in adipose tissues; (m) Western blot analysis of GPX4 and SLC7A11 protein expression in adipose tissues. N = 6, ns P > 0.05, ***P < 0.001, ****P < 0.0001.

Association of KLHL6 with CORO2B ubiquitination and protein stability

First, prediction using the Ubibrowser database suggested that KLHL6 might regulate the ubiquitination modification of CORO2B (Figure 6a). In vivo, WB analysis of mouse adipose tissues showed that compared with the Control group, CORO2B protein expression was significantly decreased in the IR group. UHRF1 overexpression substantially increased CORO2B expression, whereas subsequent KLHL6 overexpression led to a marked decrease in CORO2B levels (Figure 6b). Consistent with these findings, in vitro experiments in 3T3‐L1 adipocytes demonstrated that TNF‐α stimulation significantly reduced CORO2B expression relative to the Control group. UHRF1 overexpression significantly rescued CORO2B expression, while KLHL6 overexpression significantly suppressed it (Figure 6c). Mechanistically, Co‐IP experiments confirmed a direct interaction between KLHL6 and CORO2B (Figure 6d). Furthermore, KLHL6 markedly promoted the ubiquitination modification of CORO2B (Figure 6e). Supporting this functional relationship, CHX chase assays further demonstrated that KLHL6 overexpression accelerated CORO2B protein degradation, significantly shortening its half‐life (Figure 6f).

Figure 6.

KLHL6 was associated with CORO2B ubiquitination and protein degradation. (a) Prediction of CORO2B as a potential ubiquitination substrate of KLHL6 based on the Ubibrowser database; (b) Western blot analysis of CORO2B protein expression in mouse adipose tissues to assess the effects of UHRF1 and KLHL6 on CORO2B expression; (c) Western blot analysis of CORO2B protein levels in 3T3‐L1 cells; (d) co‐immunoprecipitation (Co‐IP) assay confirming a direct interaction between KLHL6 and CORO2B; (e) Co‐IP assay to verify the effect of KLHL6 on CORO2B ubiquitination levels; (f) CHX chase assay to evaluate the effect of KLHL6 on CORO2B degradation. N = 3, ns P > 0.05, *P < 0.05, ****P < 0.0001.

CORO2B silencing promotes mitochondrial dysfunction and ferroptosis in 3T3‐L1 cells

Initially, RT‐qPCR results showed that shRNA constructs targeting KLHL6, sh‐KLHL6‐1, sh‐KLHL6‐2, and sh‐KLHL6‐3 significantly decreased KLHL6 mRNA levels, with sh‐KLHL6‐3 showing the most effective knockdown and thus selected for subsequent experiments (Figure 7a). Similarly, sh‐CORO2B‐1, sh‐CORO2B‐2, and sh‐CORO2B‐3 efficiently suppressed CORO2B expression, with sh‐CORO2B‐1 being the most effective and used in follow‐up experiments (Figure 7b). WB analysis confirmed that KLHL6 knockdown significantly reduced its protein expression while concomitantly increasing CORO2B protein levels. Conversely, CORO2B silencing effectively reduced CORO2B expression without markedly affecting KLHL6 levels (Figure 7c). Functionally, compared with the sh‐NC group, cells with KLHL6 knockdown exhibited significantly increased viability (Figure 7d) and decreased secretion of inflammatory cytokines IL‐6 and IL‐1β (Figure 7e). Notably, CORO2B knockdown abrogated these protective effects, decreasing cell viability and markedly elevating inflammatory cytokine levels. Furthermore, mitophagy‐related proteins PINK1, Parkin, and the LC3II/I ratio were significantly upregulated in the sh‐KLHL6 group compared with sh‐NC (Figure 7f). This increase was attenuated following CORO2B knockdown, indicating that CORO2B may be associated with alterations in mitophagy‐related markers. Correspondingly, ROS levels decreased significantly after KLHL6 knockdown (Figure 7g), with increased SOD activity and reduced MDA and Fe²⁺ levels (Figure 7h). In contrast, CORO2B knockdown resulted in elevated ROS, suppressed SOD activity, and significant increases in MDA and Fe²⁺ accumulation. Finally, ferroptosis‐inhibitory proteins GPX4 and SLC7A11 were significantly higher in the sh‐KLHL6 group compared with sh‐NC but markedly lower in the combined sh‐KLHL6 + sh‐CORO2B group compared with the sh‐KLHL6 + sh‐NC group (Figure 7i), indicating that CORO2B knockdown promotes ferroptosis.

Figure 7.

Effects of CORO2B on Mitochondrial Dysfunction and Ferroptosis in 3T3‐L1 Cells. (a, b) RT‐qPCR analysis of the effects of different shRNA sequences on KLHL6 (a) and CORO2B (b) mRNA expression to identify the most effective knockdown construct for subsequent experiments; (c) Western blot analysis of KLHL6 and CORO2B protein expression; (d) CCK‐8 assay to evaluate changes in cell viability across groups; (e) ELISA to detect expression levels of inflammatory cytokines IL‐6 and IL‐1β; (f) Western blot analysis of mitophagy‐related proteins PINK1, Parkin, and LC3II/I; (g) measurement of intracellular ROS levels; (H) biochemical analysis of SOD activity, MDA content, and Fe²⁺ levels; (i) Western blot analysis of ferroptosis‐inhibitory proteins GPX4 and SLC7A11. N = 3, ns P > 0.05, ***P < 0.001, ****P < 0.0001.

CORO2B knockdown was associated with improvements in mitochondrial function and ferroptosis‐related alterations in obesity‐related insulin resistance

WB analysis showed that compared with the AAV‐sh‐NC group, the AAV‐sh‐KLHL6 group exhibited significant downregulation of KLHL6 protein expression accompanied by a notable increase in CORO2B expression in adipose tissues. Conversely, knockdown of CORO2B led to a significant decrease in CORO2B levels without altering KLHL6 expression (Figure 8a). Body weight monitoring revealed that the AAV‐sh‐KLHL6 group had significantly reduced body weight compared with the AAV‐sh‐NC group, indicating that KLHL6 knockdown alleviates obesity. This protective effect was reversed by CORO2B knockdown, which restored elevated body weight, suggesting that CORO2B knockdown weakens the weight loss induced by KLHL6 silencing (Figure 8b). GTT and ITT results confirmed enhanced insulin sensitivity and significantly improved glucose tolerance in the AAV‐sh‐KLHL6 group compared with the AAV‐sh‐NC group. These benefits were markedly abrogated following CORO2B knockdown (Figure 8c,d). In addition, serum insulin levels were significantly reduced in the AAV‐sh‐KLHL6 group, and this reduction was reversed by CORO2B knockdown (Figure 8e). Histological examination via HE staining showed that adipocyte hypertrophy was significantly reduced and adipose tissue structure improved in the AAV‐sh‐KLHL6 group compared with AAV‐sh‐NC, whereas CORO2B silencing caused a re‐enlargement of adipocyte size and structural deterioration (Figure 8f). Inflammatory cytokine analysis revealed significantly lower levels of IL‐6 and IL‐1β in the AAV‐sh‐KLHL6 group compared with AAV‐sh‐NC. In contrast, the anti‐inflammatory effect was nullified by CORO2B knockdown, with the combined AAV‐sh‐KLHL6 + AAV‐sh‐CORO2B group displaying significantly increased IL‐6 and IL‐1β levels compared with the AAV‐sh‐KLHL6 + AAV‐sh‐NC group (Figure 8g). Additionally, the enhanced p‐Akt/Akt ratio observed in the AAV‐sh‐KLHL6 group was abolished by CORO2B knockdown (Figure 8h). Mitophagy‐related proteins PINK1, Parkin, and LC3II/I were significantly upregulated in the AAV‐sh‐KLHL6 group relative to AAV‐sh‐NC, while CORO2B knockdown markedly suppressed their expression (Figure 8i). ROS assays showed that ROS levels were markedly decreased in the AAV‐sh‐KLHL6 group compared with AAV‐sh‐NC (Figure 8j), accompanied by increased SOD activity and decreased MDA and Fe²⁺ levels (Figure 8k). CORO2B knockdown reversed these redox improvements, causing elevated ROS, reduced SOD activity, and increased MDA and Fe²⁺ accumulation. Finally, WB results demonstrated that ferroptosis‐inhibitory proteins GPX4 and SLC7A11 were significantly increased following KLHL6 knockdown but downregulated after CORO2B silencing (Figure 8l).

Figure 8.

CORO2B knockdown impairs mitochondrial function and promotes ferroptosis in obesity‐related insulin resistance. (a) Western blot analysis of KLHL6 and CORO2B protein expression in mouse adipose tissues; (b) body weight change curves; (c) glucose tolerance test (GTT) to assess glucose metabolic function; (d) insulin tolerance test (ITT) to evaluate insulin sensitivity; (e) serum insulin was measured; (f) H&E staining to observe adipose tissue structure and changes in cell size; (g) ELISA detection of IL‐6 and IL‐1β levels in adipose tissues; (h) p‐Akt/Akt levels were measured by Western blot; (i) Western blot analysis of mitophagy‐related proteins PINK1, Parkin, and LC3II/I; (j) detection of ROS levels in adipose tissues; (k) biochemical assays were used to measure SOD activity, MDA content, and Fe²⁺ levels; (l) Western blot analysis of ferroptosis‐inhibitory proteins GPX4 and SLC7A11. N = 6, ns P > 0.05, ****P < 0.0001.

DISCUSSION

Obesity‐related IR profoundly compromises patients' quality of life [17]. Current therapeutic approaches, including lifestyle modifications and pharmacological interventions [18], frequently demonstrate limited efficacy while carrying adverse effects [19]. Consequently, identifying novel therapeutic targets and elucidating their regulatory mechanisms is of paramount importance for advancing disease management. Among numerous potential candidates, UHRF1, a pivotal epigenetic regulator governing DNA methylation and gene expression [20, 21], has garnered significant interest. Beyond its established roles in cell proliferation, differentiation, and genetic stability maintenance, UHRF1 exhibits emerging significance in metabolic regulation [22, 23], particularly concerning energy homeostasis, mitochondrial function, and oxidative stress responses [24, 25]. Investigating UHRF1's role in obesity‐related IR promises to enhance our molecular understanding of pathogenesis while establishing a foundation for novel UHRF1‐targeted therapies.

KLHL6 serves as an essential E3 Ub ligase regulating diverse biological processes mainly through substrate ubiquitination and proteasomal degradation, thereby modulating cellular homeostasis and signaling pathways [26, 27]. Within metabolic disorders, evidence suggests that KLHL6 may exacerbate obesity by dysregulating lipid metabolism and inflammatory responses [12]. Nevertheless, the specific function and regulatory dynamics of KLHL6 in obesity‐related IR remain incompletely characterized. Interrogation of GEO datasets revealed that the DNA methylation level of KLHL6 is significantly reduced in obesity, implying that its expression may be regulated by epigenetic mechanisms. However, whether this reduction in methylation is related to UHRF1 remains to be explored. Building upon this observation, we further investigated the regulatory role of UHRF1 on KLHL6 and its functional consequences in obesity‐related IR pathogenesis. Our data indicate that KLHL6 is upregulated in obesity‐related IR models, which may be linked to decreased DNA methylation. UHRF1 could potentially regulate KLHL6 expression through DNA methylation. This mechanistic cascade suggests that the loss of UHRF1 may disrupt epigenetic control and potentially contribute to the progression of IR through KLHL6 upregulation.

Moreover, aberrant KLHL6 expression was associated with alterations in mitochondrial integrity and cell death‐related processes. Our data showed that KLHL6 overexpression was accompanied by reduced mitophagy‐related markers, including decreased PINK1 and Parkin expression and a lower LC3II/I ratio, together with enhanced oxidative stress and ferroptosis‐related changes. These mitochondrial and redox alterations may contribute to impaired insulin signaling. In contrast, UHRF1 overexpression was associated with increased DNA methylation at the KLHL6 locus and attenuation of its aberrant expression, coinciding with an improvement in obesity‐related IR. Collectively, these findings suggest a potential role of epigenetic regulation involving UHRF1 and KLHL6 in obesity‐related IR and provide a framework for further investigation of their involvement in metabolic dysregulation.

This study further delineates the central role of KLHL6 via its regulation of CORO2B in obesity‐related IR. We observed consistent CORO2B downregulation across in vivo and in vitro IR models, and this downregulation was significantly associated with the impairment of mitochondrial autophagy and the activation of ferroptosis. Silencing KLHL6 resulted in increased CORO2B protein expression, suggesting a potential involvement of ubiquitin–proteasome‐dependent regulation. Collectively, our data suggest a potential relationship involving UHRF1, KLHL6, and CORO2B in obesity‐related IR. These factors may be associated with alterations in autophagy‐ and ferroptosis‐related processes, although the underlying mechanisms require further experimental investigation. Together, these findings provide a framework for future studies aimed at elucidating the role of UHRF1‐ and KLHL6‐related regulation in metabolic dysregulation.

However, this study is inherently limited by its exclusive reliance on murine models and in vitro systems, with no clinical validation involved. In particular, intraperitoneal administration of AAV vectors may preferentially affect the liver and pancreas, and the observed gene expression changes in adipose tissue may not fully reflect direct viral effects. Moreover, relying solely on adipose tissue gene expression is insufficient to comprehensively explain the overall alterations in glucose tolerance and insulin sensitivity observed in the mice. Previous studies have reported adipose tissue‐specific knockout of UHRF1 in mice [23], as well as systemic knockout of KLHL6 [28] and CORO2B [29] in mice, highlighting their potential roles in metabolic regulation. While our data provide novel insights into the functions of UHRF1, KLHL6, and CORO2B in obesity‐related IR, their actual clinical efficacy and therapeutic targetability in human pathophysiology still require rigorous exploration. Future research should prioritize clinical validation of these factors in human cohorts, investigate their mechanistic interplay within more complex physiological and pathological environments, and perform a comprehensive assessment of their translational potential as therapeutic targets. Additionally, systematic analysis of other key metabolic organs, such as liver, pancreas, and skeletal muscle should be incorporated to better understand the systemic effects and enhance the reliability of in vivo findings.

CONCLUSION

This study suggests an epigenetic regulatory relationship involving UHRF1, KLHL6, and CORO2B. Our findings indicate that UHRF1 expression is associated with changes in KLHL6 expression, while KLHL6 is linked to alterations in CORO2B ubiquitination. In the context of obesity‐related IR, UHRF1 overexpression was accompanied by improvements in ferroptosis‐ and mitochondrial dysfunction‐related phenotypes, whereas KLHL6 overexpression or CORO2B silencing was associated with aggravated pathological alterations.

FUNDING

This research is supported by Huxiang TCM Physique Intervention Clinical Research Center (2023SK4061); Hunan Province ‘14th Five‐Year Plan’ key specialty of TCM (Hunan TCM Letter 2023 No.4); Excellent Youth Program of Hunan Education Department (24B0346); Hunan Provincial Natural Science Foundation for Young Scientists (2025JJ60626); Hunan University of Chinese Medicine and Hospital Joint Foundation (2023XYLH019; 2024XYLH365).

ETHICS STATEMENT

All animal procedures were approved by the Institutional Animal Care and Use Committee of Hunan Evidence‐based Biotechnology Co., Ltd (Ethics Number: XZ202303035).

AUTHOR CONTRIBUTIONS

Xu Deng and Xiaoping Li contributed equally to the conceptualization and design of the study. Xu Deng was primarily responsible for data collection and analysis, while Xiaoping Li played a key role in interpreting the results and drafting the manuscript. Jialuo Cai and Xinyu Chen both contributed to data analysis and interpretation and participated in writing and revising the manuscript. They also provided valuable feedback on the study's methodology and analysis. Zelin Xu and Yuntao Luo contributed to the project's overall management and provided critical feedback on the study's design and implementation. They also contributed to writing and revising the manuscript. All authors played significant roles in the development of the study and approved the final version of the manuscript.

DISCLOSURE

The authors state that they have no financial or commercial ties to other entities that could be seen as a conflict of interest in the research.

Approval of the research protocol: Not applicable.

Informed consent: Not applicable.

Approval date of registry and the registration no. of the study/trial: Not applicable.

Animal studies: All animal procedures were approved by the Institutional Animal Care and Use Committee of Hunan Evidence‐based Biotechnology Co., Ltd (Ethics Number: XZ202303035).

DATA AVAILABILITY STATEMENT

The data used and/or analyzed during the current study are available from the corresponding author.