Dysregulation of the KLF9/TXNRD2 Axis Leads to Mitochondrial DNA Oxidation and NLRP3 Inflammasome Activation in Ulcerative Colitis

ABSTRACT

This study investigated the roles and mechanism of the transcription factor KLF9 and its downstream target gene TXNRD2 in ulcerative colitis (UC). An in vitro inflammatory model was established by stimulating human normal colonic mucosal epithelial cells (NCM460) with LPS and ATP. KLF9 or TXNRD2 expression was knocked down using RNA interference, and rescue experiments were performed using the NLRP3‐specific agonist nigericin or by co‐transfection with si‐TXNRD2. A 3% DSS‐induced UC model was established in C57BL/6 mice. Interventions were performed by tail vein injection of sh‐KLF9 adeno‐associated virus and/or intraperitoneal injection of nigericin. In DSS‐induced UC mouse colon tissues and LPS/ATP‐stimulated NCM460 cells, KLF9 was upregulated. KLF9 directly bound to the TXNRD2 promoter. Silencing KLF9 restored TXNRD2 expression, alleviated LPS/ATP‐induced oxidative stress, decreased the accumulation of cytoplasmic ox‐mtDNA and mtDNA, and inhibited NLRP3 inflammasome activation, as well as reduced pyroptosis, LDH release, and pro‐inflammatory cytokine secretion. si‐TXNRD2 or nigericin partially reversed the protective effects of KLF9 silencing. Colon‐specific KLF9 knockdown improved DSS‐induced weight loss, colonic shortening, disease activity index, and histopathological damage in mice, while simultaneously upregulating TXNRD2, inhibiting the NLRP3 pathway, and reducing oxidative stress and cytoplasmic mtDNA accumulation. Nigericin intervention partially offset the protective effects of KLF9 knockdown. KLF9 suppresses TXNRD2 transcription, leading to increased mitochondrial oxidative stress and the release of ox‐mtDNA in intestinal epithelial cells, which in turn triggers NLRP3 inflammasome activation, driving pyroptosis and intestinal inflammation.

In ulcerative colitis, the transcription factor KLF9 is abnormally overexpressed and inhibits TXNRD2 transcription. As a key mitochondrial antioxidant enzyme, the loss of TXNRD2 leads to disruption of mitochondrial redox homeostasis, resulting in excessive accumulation of mtROS and mitochondrial DNA damage. Upon release of oxidized mtDNA into the cytoplasm, it triggers the assembly and activation of the NLRP3 inflammasome, which in turn leads to the maturation and secretion of pro‐inflammatory cytokines IL‐1β and IL‐18. The resulting disruption of the epithelial barrier and amplification of the inflammatory response drive the onset and progression of ulcerative colitis. Targeting the KLF9/TXNRD2 axis may offer a new therapeutic strategy for UC.

1. Introduction

Ulcerative colitis (UC) is a complex, immune‐mediated chronic inflammatory bowel disease affecting approximately 5 million people worldwide. It is characterized by a chronic, relapsing course, with clinical manifestations primarily including abdominal pain, rectal bleeding, chronic diarrhea, and weight loss [1, 2]. Currently, treatment for UC primarily includes 5‐aminosalicylic acid preparations, corticosteroids, immunosuppressants, and biologics; however, the 1‐year clinical remission rate among patients is only approximately 40% [3], suggesting that existing treatment regimens still have significant limitations and that there is an urgent need to explore new therapeutic strategies.

Mitochondria are membrane‐bound organelles that serve as the primary site of reactive oxygen species (ROS) production. However, mitochondrial dysfunction can lead to the explosive production of ROS, resulting in mitochondrial oxidative stress [4]. This oxidative stress not only damages mitochondrial structure itself but, more importantly, causes oxidative damage to mitochondrial DNA (mtDNA) [5]. During oxidative stress, mtDNA is oxidized into oxidized mitochondrial DNA (ox‐mtDNA). Upon release into the cytoplasm, ox‐mtDNA acts as a damage‐associated molecular pattern and can directly bind to the LRR domain of the NLRP3 inflammasome [6], thereby triggering inflammasome activation. Upon activation, the NLRP3 inflammasome stimulates the caspase‐1‐dependent release of the pro‐inflammatory cytokines IL‐1β and IL‐18 and triggers pyroptosis via gasdermin D (GSDMD) [7]. An excess of pyroptosis can result in unabated and exaggerated inflammation, playing a role in the onset of inflammatory diseases [8]. Therefore, targeting the axis comprising mitochondrial oxidative stress, ox‐mtDNA release, NLRP3 inflammasome activation, and pyroptosis may hold promise for the treatment of inflammatory diseases.

Krüppel‐like factor 9 (KLF9) is a transcription factor involved in regulating various biological processes, such as mitochondrial homeostasis [9] and NLRP3‐mediated inflammation and pyroptosis [10, 11]. An analysis of diagnostic biomarkers and immune cell infiltration patterns in UC suggests that KLF9 may be involved in UC development [12]; however, the specific molecular mechanisms remain unclear. Thioredoxin reductase 2 (TXNRD2) is a selenocysteine‐containing enzyme that plays a key role in mitochondrial oxygen radical scavenging. Through its redox activity, TXNRD2 effectively counteracts mitochondrial‐dependent cell death induced by various oxidative stressors by participating in the detoxification of peroxides [13]. Selective inhibition of TXNRD2 in mitochondria leads to the accumulation of mtROS; the resulting decline in mitochondrial antioxidant capacity further exacerbates mitochondrial oxidative stress [14]. TXNRD2 expression can be suppressed by KLF9, leading to increased ROS‐mediated cell death [15], suggesting the potential of targeting the KLF9/TXNRD2 axis to control disease development by mitigating ROS‐mediated cell death.

Based on the above background, the following scientific hypothesis is proposed: KLF9 maintains mitochondrial redox homeostasis through transcriptional regulation of TXNRD2. Disruption of the KLF9/TXNRD2 axis leads to the accumulation of ox‐mtDNA and its subsequent release into the cytoplasm, thereby activating the NLRP3 inflammasome. This cascade in turn drives intestinal inflammation and promotes UC development.

2. Materials and Methods

2.1. Laboratory Animals and Model Establishment

Adult male C57BL/6 mice (6–8 weeks old, 18–22 g) were obtained from Hunan SJA Laboratory Animal Co. Ltd. (Shanghai, China). They were housed under a 12‐h light/dark cycle at 25°C with 50% relative humidity, with free access to food and water. Following a 1‐week acclimatization period in clean, well‐ventilated cages, all animal procedures were conducted. The study protocol was approved by the Animal Ethics Committee of the School of Medicine, Jiaxing University (Ethical Approval Number: JUMC2025‐009), and all experiments complied with the ARRIVE guidelines.

C57BL/6 mice were randomly assigned to control and dextran sulfate sodium (DSS) groups. UC was induced in the DSS group by providing drinking water with 3% (w/v) DSS (9011‐18‐1, AbMole, Shanghai, China) for 7 days, while the control group was given an equal volume of purified water [16]. Disease severity was evaluated by monitoring body weight, fecal consistency, and stool blood. Colon tissue samples were collected from all mice after euthanasia at the end of the experiment.

2.2. Animal Grouping and Drug Administration

C57BL/6 mice were randomly divided into 5 groups (6 mice per group). Control group: Free access to sterile distilled water; no other intervention; DSS group: Free access to sterile water containing 3% (w/v) DSS for 7 consecutive days; DSS + sh‐NC group: 1 week prior to DSS modeling, sh‐NC recombinant adeno‐associated virus (AAV) was administered via tail vein injection, followed by 7‐day DSS treatment; DSS + sh‐KLF9 group: 1 week prior to DSS modeling, KLF9‐targeted recombinant AAV (100 μL) was administered via tail vein injection, followed by 7‐day DSS treatment [17]; DSS + sh‐KLF9 + Nigericin group: 1 week prior to DSS modeling, sh‐KLF9 recombinant AAV (100 μL) was administered via tail vein injection. On days 3, 5, and 7 of the DSS modeling, nigericin (HY‐127019, MCE, 1 mg/kg, dissolved in DMSO and then diluted with sterile PBS to a final concentration of 1 mg/mL) was administered via intraperitoneal injection [18], followed by 7‐day DSS treatment.

The disease activity index (DAI) was assessed daily based on body weight, stool consistency, and fecal blood. Following euthanasia by cervical dislocation at the study endpoint, colon tissues were collected, and their lengths were recorded. Tissue specimens were either preserved in 4% paraformaldehyde for histopathological examination or frozen in liquid nitrogen and stored at −80°C.

2.3. Cell Culture and Transfection

NCM460 cells (ZQ1079, Cell Research, Shanghai, China) were cultured in MEM (11095080, Gibco) containing 10% fetal bovine serum (FBS; A5256701, Gibco) and 1% penicillin‐streptomycin (C0222, Beyotime, Shanghai, China) under conditions of 37°C and 5% CO₂.

NCM460 cells were seeded in 6‐well plates (5 × 10⁵ cells/well). At 50%–60% confluence, cells were transfected with a blend of 5 μg siRNA (diluted in 250 μL of Opti‐MEM®) and 5 μL of Lipofectamine™ 2000 (diluted in 250 μL of Opti‐MEM®). After 6 h, the transfection medium was replaced with 2 mL of fresh complete medium, and cells were cultured for another 48 h. Subsequently, cells were harvested for efficiency assays, including RT‐qPCR and Western blot. The siRNA sequences are as follows: si‐NC: sense 5′‐GATACCGTAGCTCGTGCAAGT‐3′, antisense 5′‐ACTTGCACGAGCTACGGTATC‐3′; si‐KLF9: sense 5′‐GGUUCAAAUAGGACUACUAGA‐3′, antisense 5′‐UAGUAGUCCUAUUUGAACCAU‐3′. si‐TXNRD2: sense 5′‐CGCAGGCGAAGUUACUCAAGG‐3′, antisense 5′‐UUGAGUAACUUCGCCUGCGUU‐3′.

2.4. Cell Grouping

Control group: Cells were cultured under standard conditions with an equal volume of PBS added; LPS + ATP group: Cells were treated with 1 μg/mL LPS (ST1470, Beyotime) for 5 h, followed by incubation with 5 mM ATP (C008, Novoprotein) for 12 h to induce cytotoxicity [19]; LPS + ATP + si‐NC group: Cells were transfected with negative control siRNA (si‐NC) for 48 h and treated with 1 μg/mL LPS for 5 h, followed by 5 mM ATP for 12 h; LPS + ATP + si‐KLF9 group: Cells were transfected with siRNA targeting KLF9 (si‐KLF9) for 48 h and treated with 1 μg/mL LPS for 5 h, followed by 5 mM ATP for 12 h; LPS + ATP + si‐KLF9 + Nigericin group: Cells were transfected with si‐KLF9 for 48 h and treated with 1 μg/mL LPS for 5 h and 5 mM ATP for 12 h; the NLRP3‐specific agonist nigericin (20 μM) was added 6 h before the end of LPS stimulation [20]; LPS + ATP + si‐KLF9 + si‐TXNRD2 group: Cells were co‐transfected with si‐KLF9 and TXNRD2‐targeting siRNA (si‐TXNRD2) for 48 h and treated with 1 μg/mL LPS for 5 h, followed by 5 mM ATP for 12 h.

2.5. Western Blot

Proteins were isolated from tissues or cells using PMSF‐supplemented RIPA buffer, and their concentrations were quantified by BCA assay. After mixing with loading buffer and denaturation, equal amounts of protein (25 μg/lane) were subjected to SDS‐PAGE and transferred to PVDF membranes. The membranes were blocked with 5% skim milk for 1 h, then incubated overnight at 4°C with primary antibodies against KLF9 (1:1000, ab227920, Abcam), TXNRD2 (1:1000, 16360‐1‐AP, Proteintech), NLRP3 (1:1000, ab263899, Abcam), GSDMD (1:1000, PD00‐18, Invitrogen), Cleaved Caspase‐1 (1:1000, ab179515, Abcam), and GAPDH (1:2000, ab9485, Abcam). Following this, the membranes were incubated with HRP‐labeled goat anti‐rabbit IgG secondary antibody at room temperature for 2 h. Protein bands were visualized using 300 μL each of ECL‐A and ECL‐B solutions for 3 min, with exposure times of 10‐20 s, followed by development for 1 min and scanning using an Epson scanner. Grayscale values of protein bands were analyzed using ImageJ software (v1.8.0).

2.6. RT‐qRCR

Total RNA was extracted using the RNAsimple Total RNA Extraction Kit (DP419, TIANGEN), with RNA concentration measured with a spectrophotometer. RNA was reverse‐transcribed into cDNA using the Reverse Transcription Kit (KR136, TIANGEN). PCR was conducted on a 7500 Real‐Time PCR System (ABI, USA) using the SYBR Premix EX Taq Kit (DRR041A, TAKARA). The thermal cycling protocol consisted of an initial denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. GAPDH served as the endogenous reference, and relative gene expression was quantified using the 2^−ΔΔCt method. All primer sequences are provided in Table 1.

Table 1.

Primers.

Genes	Primer sequences (5′–3′)
Human KLF9 Bank ID 59853224c1	Forward: GCCGCCTACATGGACTTCG
	Reverse: GGATGGGTCGGTACTTGTTCA
Mouse KLF9 Bank ID 146134344c1	Forward: GCCGCCTACATGGACTTCG
	Reverse: GGTCACCGTGTTCCTTGGT
Human TXNRD2 Bank ID 87196331c1	Forward: CGGCTTCGACCAGCAAATG
	Reverse: ACAGGACGGTGTCAAAGGTG
Mouse TXNRD2 Bank ID 12841921a1	Forward: GATCCGGTGGCCTAGCTTG
	Reverse: TCGGGGAGAAGGTTCCACAT
Human IL‐1β Bank ID 27894305c1	Forward: ATGATGGCTTATTACAGTGGCAA
	Reverse: GTCGGAGATTCGTAGCTGGA
Mouse IL‐1β Bank ID 118130747c1	Forward: GAAATGCCACCTTTTGACAGTG
	Reverse: TGGATGCTCTCATCAGGACAG
Human IL‐6 Bank ID 224831235c1	Forward: ACTCACCTCTTCAGAACGAATTG
	Reverse: CCATCTTTGGAAGGTTCAGGTTG
Mouse IL‐6 Bank ID 13624310c1	Forward: CTGCAAGAGACTTCCATCCAG
	Reverse: AGTGGTATAGACAGGTCTGTTGG
Human TNF‐α Bank ID 25952110c1	Forward: CCTCTCTCTAATCAGCCCTCTG
	Reverse: GAGGACCTGGGAGTAGATGAG
Mouse TNF‐α Bank ID 7305585a1	Forward: CCCTCACACTCAGATCATCTTCT
	Reverse: GCTACGACGTGGGCTACAG
Human IL‐18 Bank ID 342349317c1	Forward: TCTTCATTGACCAAGGAAATCGG
	Reverse: TCCGGGGTGCATTATCTCTAC
Mouse IL‐18 Bank ID 6680413a1	Forward: GACTCTTGCGTCAACTTCAAGG
	Reverse: CAGGCTGTCTTTTGTCAACGA
Human GAPDH Bank ID 378404907c1	Forward: GGAGCGAGATCCCTCCAAAAT
	Reverse: GGCTGTTGTCATACTTCTCATGG
Mouse GAPDH Bank ID 6679937a1	Forward: AGGTCGGTGTGAACGGATTTG
	Reverse: TGTAGACCATGTAGTTGAGGTCA

2.7. Dual Luciferase Assay

The human TXNRD2 promoter sequence containing the predicted KLF9 binding site was cloned into the pGL3‐Basic luciferase reporter vector (E1751, Promega) to generate the wild‐type reporter. A mutant reporter with mutated binding sites was constructed using gene mutation techniques. NCM460 cells were seeded in 24‐well plates and, upon reaching 70% confluence, co‐transfected with 200 ng of pGL3‐TXNRD2 promoter reporter, 50 ng of pRL‐TK internal control (E2241, Promega), and 200 ng of thKLF9 overexpression vector (pcDNA3.1‐KLF9) or an empty vector (pcDNA3.1) using Lipofectamine 3000 (L3000015, Invitrogen). After 48 h, firefly and Renilla luciferase activities were measured using the Dual‐Luciferase Reporter Assay System (E1910, Promega) on a GloMax 20/20 luminometer (Promega).

2.8. CCK‐8 Assay

Cell proliferation was assessed using the CCK‐8 assay kit (C0038, Beyotime). Log‐phase NCM460 cells were digested with 0.25% trypsin and resuspended in complete medium containing 10% FBS. The cell suspension (2 × 10⁵ cells/mL) was seeded into 96‐well plates at 100 μL/well (2 × 10⁴ cells), followed by treatment according to the experimental design. A volume of 10 μL of CCK‐8 solution was added to each well, followed by a 2‐h incubation. Absorbance readings at 450 nm were obtained using a Varioskan™ LUX microplate reader (Thermo Fisher Scientific).

2.9. LDH Analysis

The LDH release was measured using the LDH assay kit (BC0680, Solarbio, Beijing, China). Cells from each group were treated with cell lysis buffer at room temperature for 15–20 min. Then, cells were centrifuged at 1000 × g for 10 min, and the supernatant was taken to measure absorbance values at 490 nm using a microplate reader (Varioskan™ LUX, Thermo Fisher Scientific).

2.10. Flow Cytometry

Pyroptosis was assessed by FAM‐FLICA Caspase‐1/PI staining. Following 48‐h culture at 37°C with 5% CO₂, cells were centrifuged and resuspended in 200 μL of binding buffer. The suspension was mixed with 10 μL of FAM‐FLICA Caspase‐1 and 5 μL of PI, diluted with 300 μL of binding buffer, and incubated for 15 min at room temperature in the dark. Samples were then analyzed on a CytoFLEX S flow cytometer (Beckman) using FlowJo software. Gating strategy was set: debris was excluded using forward scatter area (FSC‐A) versus side scatter area (SSC‐A) plots. Single cells were then selected based on FSC height (FSC‐H) versus FSC‐A plots. Caspase‐1‐positive (FAM‐FLICA⁺) and PI⁺ cells were identified using untreated and single‐staining controls. The percentage of Caspase‐1⁺/PI⁺ cells was quantified and used to represent the pyroptotic cell population [21].

For ROS detection, cells were incubated with DCFH‐DA (S1105S, Beyotime) at 37°C for 30 min in the dark, digested with trypsin, and centrifuged at 500 × g for 5 min at 4°C. The cell pellet was washed twice with pre‐chilled PBS, fixed, and analyzed by flow cytometry.

2.11. Enzyme‐Linked Immunosorbent Assay (ELISA)

Human cytokine levels (IL‐1β, IL‐18, IL‐6, and TNF‐α) in cell supernatants or mouse colon homogenates were measured with the Aptplex™ Human Intestinal Inflammation 6‐Plex Panel ELISA Kit (MPA006, Elabscience). Mouse cytokines were assessed using commercial ELISA kits for IL‐1β (E‐EL‐M0037), IL‐18 (E‐EL‐M0730), IL‐6 (E‐EL‐M0044), and TNF‐α (E‐EL‐M3063) (Elabscience). Absorbance was read at 450 nm.

2.12. DAI Assessment

The DAI was assessed daily beginning on the first day of DSS administration. The DAI comprised three subscores: weight loss (0 = none, 1 = 1%–5%, 2 = 5%–10%, 3 = 10%–15%, 4 ≥ 15%), fecal consistency (0 = normal, 2 = loose, 4 = watery), and fecal blood (0 = none, 2 = slight, 4 = gross). The total DAI was the sum of these three scores [22].

2.13. HE Staining

Fresh mouse colon tissues were fixed in 4% paraformaldehyde for 24 h, dehydrated through a graded ethanol series, embedded in paraffin, and sectioned. The sections were then deparaffinized in xylene and rehydrated in graded ethanol before undergoing HE staining. Histological examination was performed under a light microscope. Two pathologists blinded to the group assignments evaluated the colon samples based on clinical and pathological criteria [23].

2.14. Extraction and Detection of mtDNA

Cell membranes were selectively disrupted using pre‐chilled HB homogenization buffer (containing 250 mM sucrose, 10 mM HEPES, pH 7.4, and 1 × protease‐phosphatase inhibitor), followed by differential centrifugation to separate the cytosolic fraction. Total DNA was extracted from the cytosolic fraction using the QIAamp DNA Mini Kit (51304, Qiagen). The copy numbers of the mitochondria‐specific genes d‐loop and Cox1 were measured by RT‐qPCR, with nuclear genes (e.g., GAPDH, representing non‐Numt sequences) amplified as controls to exclude nuclear DNA contamination. The relative abundance of cytoplasmic mtDNA was calculated using the 2^−ΔΔCt method [24].

2.15. Quantification of ox‐mtDNA

To detect ox‐mtDNA, DNA was extracted from the isolated cytosolic or mitochondrial fractions as described above. The extracted DNA was then analyzed using the 8‐Hydroxy‐2′‐deoxyguanosine (8‐OH‐dG) ELISA Kit (ab285254, Abcam). Absorbance was measured at 450 nm using a microplate reader, and the concentration of 8‐OH‐dG was calculated. Results were expressed as 8‐OH‐dG concentration (ng/mL) or as relative levels normalized to total mtDNA.

2.16. Statistical Analysis

Statistical analysis was performed using GraphPad Prism 9.0 software. Comparisons between two groups were conducted using the unpaired Student's t‐test. Comparisons among multiple groups were conducted using one‐way analysis of variance (ANOVA), followed by Tukey's post‐hoc test when variances were homogeneous. All data were obtained from at least three independent replicates and are expressed as mean ± standard deviation. *p < 0.05 was considered statistically significant.

3. Results

3.1. KLF9 Is Upregulated in UC Models

A mouse model of UC was first established using DSS induction. KLF9 mRNA and protein expression was found to be increased in the colon tissue of DSS‐induced mice, as determined by Western blot and RT‐qPCR (Figure 1A–C, p < 0.001). Significant elevations in the pro‐inflammatory cytokines IL‐1β, IL‐18, IL‐6, and TNF‐α were also observed (Figure 1D,E, p < 0.001). Furthermore, LPS/ATP stimulation of human normal colonic mucosal epithelial cells (NCM460) resulted in increased intracellular KLF9 expression (Figure 1F–H, p < 0.001), accompanied by increased secretion of inflammatory cytokines (Figure 1I,J, p < 0.001). These data indicate that dysregulation of KLF9 is closely associated with UC and related inflammatory response.

Figure 1.

Expression of KLF9 in UC models. (A) RT‐qPCR analysis of KLF9 mRNA expression in mouse colon tissue; (B, C) Western blot analysis of KLF9 protein expression levels in mouse colon tissue; (D) RT‐qPCR analysis of IL‐1β, IL‐18, IL‐6, and TNF‐α mRNA expression in mouse colon tissue; (E) ELISA analysis of IL‐1β, IL‐18, IL‐6, and TNF‐α expression in mouse colon tissue; (F) RT‐qPCR analysis of KLF9 mRNA expression in NCM460 cells; (G, H) Western blot analysis of KLF9 protein expression levels in NCM460 cells; (I) RT‐qPCR analysis of IL‐1β, IL‐18, IL‐6, and TNF‐α mRNA expression in NCM460 cells; (J) ELISA analysis of IL‐1β, IL‐18, IL‐6, and TNF‐α expression in NCM460 cells; Panel A–D: Animal experiments (n = 6); Panel E–H: Cell experiments were repeated independently three times (N = 3). Results were presented as mean ± standard deviation and analyzed using an independent samples t‐test. *p < 0.05, **p < 0.01, ***p < 0.001.

3.2. Silencing KLF9 Expression Reverses LPS/ATP‐Induced Inflammatory Responses

To further investigate the functional role of KLF9 in the inflammatory response of intestinal epithelial cells, LPS/ATP‐stimulated NCM460 cells were subjected to RNA interference targeting KLF9. Western blot and RT‐qPCR confirmed the efficiency of si‐KLF9 in knocking down KLF9 expression (Figure 2A–C, p < 0.001). Functionally, KLF9 knockdown significantly restored LPS/ATP‐impaired cell proliferation (Figure 2D, p < 0.001) and significantly suppressed the production of pro‐inflammatory cytokines such as IL‐1β, IL‐18, IL‐6, and TNF‐α (Figure 2E–L, p < 0.001). These results indicate that inhibiting KLF9 expression effectively alleviates LPS/ATP‐induced inflammatory damage in intestinal epithelial cells.

Figure 2.

Silencing of KLF9 reverses LPS/ATP‐induced inflammatory responses. (A) RT‐qPCR analysis of KLF9 mRNA expression levels in NCM460 cells; (B, C) Western blot analysis of KLF9 protein expression levels in NCM460 cells; (D) CCK‐8 assay for NCM460 cell proliferation; (E–H) RT‐qPCR analysis of IL‐1β, IL‐18, IL‐6, and TNF‐α mRNA expression in NCM460 cells; (I–L) ELISA analysis of IL‐1β, IL‐18, IL‐6, and TNF‐α levels; Cell experiments were performed in triplicate (N = 3). Results were expressed as mean ± standard deviation. Comparisons among multiple groups were performed using one‐way ANOVA, followed by Tukey's multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001.

3.3. Silencing KLF9 Reverses LPS/ATP‐Induced Oxidative Stress and NLRP3 Inflammasome Activation‐Mediated Pyroptosis

Dysregulated inflammatory responses are often accompanied by stress responses [25]. Given this, changes in oxidative stress markers following KLF9 knockdown were investigated. ELISA detected that following LPS/ATP treatment, SOD and GSH levels in NCM460 cells decreased, while MDA levels increased, indicating that LPS/ATP stimulation aggravates oxidative stress in NCM460 cells (Figure 3A, p < 0.001). Concurrently, flow cytometry analysis revealed a significant increase in ROS generation in NCM460 cells following LPS/ATP treatment (Figure 3B,C, p < 0.001), and ELISA results showed a marked increase in cellular LDH release (Figure 3D, p < 0.001), suggesting that LPS/ATP stimulation induced pyroptosis in NCM460 cells. Furthermore, flow cytometry analysis revealed a significant increase in pyroptosis rates in NCM460 cells following LPS/ATP treatment (Figure 3E,F, p < 0.001), supporting that LPS/ATP stimulation induces oxidative stress damage. Knocking down KLF9 effectively reversed all of the above changes, increasing SOD and GSH levels, reducing MDA and intracellular ROS, and decreasing pyroptosis rates and LDH release (Figure 3A–F, p < 0.001). Rescue experiments were subsequently performed using the NLRP3‐specific activator nigericin on KLF9‐knockdown cells. The addition of nigericin significantly reversed the attenuation of oxidative stress and the inhibition of pyroptosis mediated by KLF9 knockdown (Figure 3A–F, p < 0.001).

Figure 3.

Silencing of KLF9 reverses LPS/ATP‐induced oxidative stress. (A) ELISA assay for SOD, GSH, and MDA in NCM460 cells; (B, C) Flow cytometry analysis of intracellular ROS levels; (D) LDH concentration in NCM460 cells; (E, F) Flow cytometry analysis of pyroptosis rates; Cellular experiments were performed in triplicate (N = 3). Results were expressed as mean ± standard deviation. Comparisons among multiple groups were performed using one‐way ANOVA, followed by Tukey's multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001.

Oxidative stress‐mediated mitochondrial damage and excessive ROS production are key upstream signals that activate NLRP3 inflammasomes and trigger pyroptosis in intestinal epithelial cells, while the release of ox‐mtDNA and its accumulation in the cytoplasm can further amplify inflammatory responses [26]. Additionally, NLRP3 is associated with susceptibility to UC [27]. Therefore, proteins related to the NLRP3 inflammasome pathway were analyzed using Western blot. Following LPS/ATP treatment, NLRP3 protein expression was significantly upregulated in NCM460 cells, cleaved caspase‐1 increased, and GSDMD was cleaved (elevated GSDMD‐N levels), indicating that LPS/ATP stimulation significantly activates the NLRP3 inflammasome pathway in NCM460 cells (Figure 4A,B, p < 0.001). ELISA reported that levels of pyroptosis‐associated cytokines, including IL‐1β, IL‐18, IL‐6, and TNF‐α, in the supernatant of NCM460 cells were elevated following LPS/ATP treatment (Figure 4C–F, p < 0.001). Furthermore, ELISA analysis revealed that LPS/ATP treatment increased both total 8‐OH‐dG levels and cytoplasmic 8‐OH‐dG levels in NCM460 cells (Figure 4G,H, p < 0.001), suggesting that LPS/ATP exacerbates oxidative damage in both cells and the cytoplasm. RT‐qPCR results showed that KLF9 knockdown also reduced LPS/ATP‐induced accumulation of mtDNA in the cytoplasm, specifically manifested as a significant decrease in the relative ratio of D‐loop mtDNA to COX1 mtDNA in the cytoplasm (Figure 4I, p < 0.001). Preliminary experiments confirmed that, under baseline conditions without any inflammatory stimulation, knocking down KLF9 alone did not alter the expression of proteins in the NLRP3 inflammasome pathway and had no significant effect on cell viability (Supplementary Figure 1A–C). These results demonstrate that KLF9 knockdown exerts its protective effects by inhibiting NLRP3 inflammasome activation, a process closely associated with mitochondrial oxidative stress, ROS production, ox‐mtDNA, and mtDNA release into the cytoplasm.

Figure 4.

Knocking down KLF9 inhibits the pyroptotic effects of LPS/ATP on NCM460 cells. (A, B) Western blot analysis of NLRP3, Cleaved‐caspase‐1, GSDMD‐FL, and GSDMD‐N in NCM460 cells; (C–F) ELISA for IL‐1β, IL‐18, IL‐6, and TNF‐α in NCM460 cells; (G) ELISA for total 8‐OH‐dG in NCM460 cells; (H) ELISA for 8‐OH‐dG content in the cytoplasm of NCM460 cells; (I) RT‐qPCR analysis of D‐ring mitochondrial DNA, COX1 mitochondrial DNA, or non‐Numt mitochondrial DNA; Cell experiments were performed in triplicate (N = 3). Results were expressed as mean ± standard deviation. Comparisons among multiple groups were performed using one‐way ANOVA, followed by Tukey's multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001.

3.4. Silencing KLF9 Promotes TXNRD2 Expression, Thereby Reversing LPS/ATP‐Induced Oxidative Stress and Pyroptosis Mediated by the Activated NLRP3 Inflammasome Pathway

TXNRD2, as an antioxidant gene, can be inhibited by KLF9, leading to increased ROS‐driven cell death [15]. Therefore, it was hypothesized that TXNRD2 may be a key downstream target molecule through which KLF9 regulates oxidative stress and NLRP3 inflammasome activation in intestinal epithelial cells. Results from dual‐luciferase assays showed that KLF9 overexpression significantly inhibited luciferase activity of the wild‐type TXNRD2 promoter, while having no significant effect on the activity of the mutant promoter (Figure 5A, p < 0.001). Rescue experiments demonstrated that, on the basis of KLF9 knockdown, co‐transfection with si‐TXNRD2 effectively reduced TXNRD2 mRNA and protein levels without altering KLF9 expression (Figure 5B,C, p < 0.001). The effects of KLF9 knockdown alone, including suppression of the NLRP3 inflammasome pathway, reduced protein levels of NLRP3 and GSDMD‐N, and decreased secretion of IL‐1β, IL‐18, IL‐6, and TNF‐α, were partially reversed upon co‐knockdown of TXNRD2 (Figure 5C–H, p < 0.001). Furthermore, upon co‐knockdown of TXNRD2, total 8‐OH‐dG, cytoplasmic 8‐OH‐dG levels, and cytoplasmic mtDNA accumulation all rebounded (Figure 5I–K, p < 0.001), indicating that the absence of TXNRD2 disrupts the protective effect of KLF9 silencing on mitochondrial integrity. Also, co‐knockdown with TXNRD2 impaired the oxidative stress‐alleviating effects of KLF9 knockdown, with SOD and GSH levels decreasing and MDA and ROS levels increasing (Figure 6A–C, p < 0.001). Moreover, co‐knockdown TXNRD2 increased LDH release (Figure 6D, p < 0.001), as well as pyroptosis rate (Figure 6E,F, p < 0.001) in KLF9‐knockdown cells. Furthermore, under basal conditions without any stimuli, either the silencing of TXNRD2 alone or the combined silencing of KLF9 and TXNRD2 did not induce significant changes in the levels of NLRP3, cleaved caspase‐1, GSDMD‐N protein levels, nor did they significantly affect cell viability (Supplementary Figure 2A–C), further confirming that the regulation of the NLRP3 inflammasome by this signaling axis depends on inflammatory stress conditions rather than nonspecific effects under steady‐state conditions. These results collectively demonstrate that KLF9 may influence mitochondrial homeostasis by negatively regulating TXNRD2 expression. When KLF9 is inhibited, TXNRD2 expression is restored, thereby alleviating mitochondrial oxidative damage, inhibiting mtDNA release into the cytoplasm, and ultimately suppressing NLRP3 inflammasome‐mediated pyroptosis.

Figure 5.

Silencing of KLF9 promotes TXNRD2 expression and reverses LPS/ATP‐induced pyroptosis mediated by the NLRP3 inflammasome pathway. (A) Dual‐luciferase reporter assay; (B) RT‐qPCR analysis of TXNRD2 and KLF9; (C–D) Western blot analysis of TXNRD2, KLF9, NLRP3, GSDMD‐FL, and GSDMD‐N protein levels; (E–H) ELISA for pyroptosis‐related cytokines (IL‐1β, IL‐18, IL‐6, and TNF‐α) in NCM460 cells; (I) ELISA for total 8‐OH‐dG in NCM460 cells; (J) ELISA for 8‐OH‐dG levels in the cytoplasm of NCM460 cells; (K) RT‐qPCR analysis of D‐ring mitochondrial DNA, COX1 mitochondrial DNA, or non‐Numt mitochondrial DNA; Cell experiments were performed in triplicate (N = 3). Results were expressed as mean ± standard deviation. Comparisons among multiple groups were performed using one‐way ANOVA, followed by Tukey's multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 6.

Silencing KLF9 promotes TXNRD2 expression and reverses LPS/ATP‐induced oxidative stress. (A) ELISA for SOD, GSH, and MDA in NCM460 cells; (B, C) Flow cytometry analysis of intracellular ROS levels; (D) Cellular LDH concentration; (E, F) Flow cytometry analysis of pyroptosis rates. Cell experiments were performed in triplicate (N = 3). Results were expressed as mean ± standard deviation. Comparisons among multiple groups were performed using one‐way ANOVA, followed by Tukey's multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001.

3.5. Knocking Down KLF9 Reduces Inflammation in Mice With UC

The pathological role of KLF9 was validated in the mouse model of UC. By administering AAV carrying KLF9‐shRNA via tail vein injection, KLF9 was knocked down locally in the colons of mice concurrently with DSS modeling. KLF9 knockdown significantly ameliorated the disease phenotype, as evidenced by reduced weight loss (Figure 7A, p < 0.001), restored colon length (Figure 7B,C, p < 0.001), decreased DAI (Figure 7D, p < 0.001), and alleviated histopathological damage (Figure 7E,F, p < 0.001). At the molecular level, mRNA expression of pro‐inflammatory factors in colon tissue was also significantly reduced (Figure 7G, p < 0.001). Notably, treatment with nigericin partially offset these protective effects of KLF9 knockdown (Figure 7A–G, p < 0.001). These in vivo data strongly confirm that targeted inhibition of KLF9 can effectively alleviate UC by suppressing NLRP3 inflammasome activation.

Figure 7.

KLF9 knockdown alleviates inflammation in mice with DSS‐induced UC. (A) Body weight of mice with UC; (B, C) Colon length in mice with UC; (D) DAI scores; (E, F) Representative HE‐stained colonic sections; (G) RT‐qPCR analysis of IL‐1β, IL‐18, IL‐6, and TNF‐α in mouse colonic tissues. n = 6. Results were expressed as mean ± standard deviation. Comparisons among multiple groups were performed using one‐way ANOVA, followed by Tukey's multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001.

3.6. KLF9 Knockdown Regulates the TXNRD2/NLRP3 Axis and Alleviates Oxidative Damage in Mice With UC

To elucidate the molecular mechanisms underlying the protective effects of KLF9 knockdown in mice, changes in relevant signaling pathways were examined in colon tissues. KLF9 knockdown effectively reversed DSS‐induced reduction in colonic antioxidant capacity (as indicated by increased SOD and GSH) and lipid peroxidation damage (as indicated by decreased MDA) (Figure 8A–C, p < 0.001). Western blot analysis revealed that sh‐KLF9 significantly reduced KLF9 protein levels in colon tissue while upregulating TXNRD2 expression (Figure 8D,E, p < 0.001). Consistent with this, the protein levels of key components of the NLRP3 inflammasome pathway (NLRP3) and its downstream effector molecules (Cleaved Caspase‐1 and GSDMD‐N) were also significantly suppressed (Figure 8D,E, p < 0.001). More importantly, elevated levels of ox‐mtDNA were detected in the cytoplasm of colonic epithelial cells from mice in the DSS group, along with increased total and cytoplasmic levels of 8‐OH‐dG (Figure 8F,G, p < 0.001). Concurrently, the relative proportions of D‐loop, COX1, and non‐Numt mtDNA in the cytoplasm were also significantly elevated (Figure 8H, p < 0.001). Notably, nigericin partially counteracted the protective effects of KLF9 knockdown and fully abrogated its protective effects on mitochondria (Figure 8A–H, p < 0.001). Together, these data indicate that in DSS‐induced UC in mice, targeted knockdown of KLF9 exerts a therapeutic effect by restoring TXNRD2 expression, alleviating mitochondrial oxidative stress, and reducing the release of ox‐mtDNA, thereby inhibiting the excessive activation of NLRP3 inflammasomes.

Figure 8.

Knockdown of KLF9 regulates the TXNRD2/NLRP3 axis and alleviates oxidative damage in the colon of mice with UC. (A–C) ELISA for SOD, GSH, and MDA in mouse colon tissue; (D, E) Western blot analysis of TXNRD2, NLRP3, and proteins in the NLRP3 inflammasome pathway (NLRP3, Cleaved Caspase‐1, and GSDMD); (F) ELISA for total 8‐OH‐dG content in cells; (G) ELISA for 8‐OH‐dG levels in the cytoplasm of cells; (H) RT‐qPCR analysis of d‐ring mitochondrial DNA, COX1 mitochondrial DNA, or non‐Numt mitochondrial DNA. n = 6. Results were expressed as mean ± standard deviation. Comparisons among multiple groups were performed using one‐way ANOVA, followed by Tukey's multiple comparisons test. *p < 0.05, **p < 0.01, ***p < 0.001.

4. Discussion

UC is a chronic disease characterized by recurrent and remitting mucosal inflammation [28]. Although current treatments have demonstrated some efficacy, clinical remission rates remain suboptimal [2], underscoring an urgent need to elucidate its pathogenesis and develop novel therapeutic strategies. In recent years, the central role of NLRP3 inflammasome‐mediated excessive inflammatory responses in UC has been partially appreciated [29], but its upstream regulatory mechanisms remain incompletely understood. This study focused on the molecular link between mitochondrial oxidative stress and NLRP3 inflammasome activation, incorporating the KLF9/TXNRD2/ox‐mtDNA/NLRP3 axis into the UC pathogenesis, and elucidated the signaling pathway from transcriptional regulation to inflammatory effectors.

This study found that KLF9 expression was upregulated in UC model mice and LPS/ATP‐stimulated intestinal epithelial cells, while TXNRD2 expression was downregulated. Abnormal expression patterns of KLF9 [30, 31] and TXNRD2 [32] have been confirmed in inflammatory conditions. KLF9 exerts diverse regulatory functions in various physiological and pathological processes through regulating target gene transcription or interacting with other transcription factors and coactivators/inhibitors [33]. TXNRD2, a key enzyme in the mitochondrial‐specific thioredoxin system, protects mitochondrial DNA from oxidative damage by maintaining mitochondrial redox homeostasis. It is experimentally validated that knocking down KLF9 reduces ROS production in ischemic cardiomyocytes by upregulating TXNRD2 expression [15]. This study consistently supported KLF9‐mediated TXNRD2 transcription, linking KLF9‐mediated transcriptional regulation to mitochondrial function.

Downregulation of KLF9 can reduce trophoblast apoptosis mediated by oxidative stress and NLRP3 inflammasome activation [11]. Furthermore, cardiac‐specific knockout of KLF9 alleviates inflammatory responses and oxidative stress [34]. These studies suggest that KLF9 participates in the regulation of oxidative stress and inflammatory responses across different tissues and pathological contexts. As a key regulator of mitochondrial redox homeostasis, functional loss of TXNRD2 has been shown to exacerbate oxidative stress‐induced damage. Studies indicate that knockdown of TXNRD2 reverses the LXA4‐induced suppression of ROS and NLRP3 inflammasomes [32]. TXNRD2 deficiency also exacerbates lipid peroxidation levels and oxidative stress damage in rats with cerebral hemorrhage [35]. These findings further establish the central role of TXNRD2 in maintaining mitochondrial redox balance and regulating NLRP3 inflammasome activation. The present study systematically investigated the regulatory relationship between KLF9 and TXNRD2 in UC and their effects on NLRP3 inflammasome activation. Mechanistic studies revealed a signaling cascade in which KLF9, by negatively regulating TXNRD2, induces mitochondrial oxidative stress and the release of ox‐mtDNA, thereby activating NLRP3 inflammasome‐mediated pyroptosis. In rescue experiments, TXNRD2 silencing or nigericin partially reversed the protective effects of KLF9 knockdown, collectively confirming that KFLF9 exerts its effects through a pathway that involves inhibiting TXNRD2 and subsequently activating NLRP3 inflammasomes.

In vivo, localized knockdown of KLF9 in the colon significantly improved disease manifestations in DSS‐induced UC mice, including reduced body weight, shortened colon, elevated DAI, and alleviated histopathological damage. KLF9 knockdown upregulated TXNRD2 expression, inhibited the activation of NLRP3 pathway‐related proteins, and reduced oxidative stress and the accumulation of cytoplasmic mtDNA. Treatment with nigericin partially counteracted the protective effects of KLF9 knockdown, further confirming the critical role of this signaling axis in UC pathogenesis. The high consistency of in vivo and in vitro experimental results provides strong evidence for the feasibility of the KLF9/TXNRD2 axis as a therapeutic target for UC.

Limitations should also be acknowledged in this study design. This study elucidates the downstream mechanisms of KLF9; however, the upstream regulatory factors responsible for the upregulation of KLF9 itself in UC, specifically, which inflammatory signals or epigenetic mechanisms are involved, remain unclear and warrant further investigation. Although the focus is on ox‐mtDNA, the synergistic role of other ROS or metabolites from mitochondria in this process cannot be entirely ruled out. Further validation could be conducted using mitochondria‐targeted drugs or more refined gene‐editing models. In this study, unstimulated NCM460 cells exhibited basal expression of NLRP3 inflammasome pathway proteins. This may be attributed to the inherent innate immune characteristics of the intestinal epithelial cell line [29, 36], as well as the culture conditions, which included 10% FBS. It is possible that certain bioactive components or trace stimulants present in the serum exerted a mild influence on the baseline activation status of the inflammasome. Importantly, all quantitative conclusions regarding pyroptosis‑related markers in this study are based on relative changes between treatment groups following LPS/ATP stimulation, rather than on absolute expression levels under basal conditions. Therefore, the observed baseline expression does not affect the interpretation of stimulation‑induced effects. Nevertheless, the potential confounding effect of serum components could not be entirely excluded in the current experimental setup. Future studies should include comparative experiments under serum‑free or serum‑reduced conditions to more precisely evaluate the basal pyroptosis status of NCM460 cells in a truly quiescent state.

In conclusion, this study presents the critical role of the KLF9/TXNRD2 axis in UC. This discovery not only deepens our understanding of the pathogenesis of UC but also provides a theoretical basis and experimental foundation for developing UC treatment strategies targeting the KLF9/TXNRD2 axis. Future research may further explore the clinical relevance of the KLF9/TXNRD2 axis in UC patients and evaluate the potential and safety of interventions targeting this signaling axis in UC treatment, with the aim of providing new therapeutic options for UC patients.

Author Contributions

Yijia Fan: conceptualization, funding acquisition, validation, visualization, project administration, supervision. Lingling Dai: data curation, resources, formal analysis, visualization, writing – review and editing, funding acquisition, investigation. Feng Zhu: conceptualization, visualization, validation, formal analysis, project administration, data curation. Mingfang Ping: writing – original draft, visualization, validation, formal analysis, resources. Xiaofeng Zhu: investigation, writing – original draft, methodology, visualization, software, project administration, data curation. Junguo Chen: investigation, funding acquisition, methodology, writing – review and editing, software, supervision.

Ethics Statement

All animal experiments were approved by the Animal Ethics Committee of the School of Medicine, Jiaxing University. (Ethical Approval Number: JUMC2025‐009). The animal experiments have been carried out in accordance with the ARRIVE guidelines.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Effects of KLF9 knockdown on the NLRP3 inflammasome pathway in NCM460 cells under basal conditions. (A, B) Representative Western blot images and quantitative analysis of protein levels of NLRP3, Cleaved‐caspase‐1, GSDMD‐FL, and GSDMD‐N in the NLRP3 inflammasome pathway; (C) Cell viability assessed using the CCK‐8 assay. Cell experiments were performed in triplicate (N = 3). Results were expressed as mean ± standard deviation. Comparisons among multiple groups were performed using one‐way ANOVA, followed by Tukey's multiple comparisons test.

Figure S2: Effects of TXNRD2 knockdown on the NLRP3 inflammasome pathway in NCM460 cells under basal conditions. (A, B) Representative Western blot images and quantitative statistical analysis of protein levels of NLRP3, Cleaved‐caspase‐1, GSDMD‐FL, and GSDMD‐N in the NLRP3 inflammasome pathway; (C) Cell viability assessed using the CCK‐8 assay. Cell experiments were performed in triplicate (N = 3). Results were expressed as mean ± standard deviation. Comparisons among multiple groups were performed using one‐way ANOVA, followed by Tukey's multiple comparisons test.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.