Circular RNA ZBTB46 Attenuates Apoptosis and Oxidative Stress in Lipopolysaccharide-Injured Human Endothelial Cells by Modulating ERBB2-AKT Signaling

Sepsis, a life-threatening systemic inflammatory cascade initiated by pathogenic invasion, frequently culminates in multi-organ failure¹. Within this critical context, SA-AKI emerges as a predominant and clinically significant complication in intensive care units². Characterized by abrupt deterioration of renal filtration capacity concurrent with systemic infection, this condition demonstrates mortality rates exceeding those of non-septic AKI by 38%-42% in clinical cohorts². Approximately 60% of individuals diagnosed with sepsis or septic shock are anticipated to develop SA-AKI³. Therefore, SA-AKI has become a major challenge in intensive care, contributing significantly to morbidity and mortality⁴. At present, the mechanism of SA-AKI remains incompletely understood. Although preclinical studies have involved microcirculatory dysfunction, alterations in cellular metabolism, and imbalances within the renin-angiotensin-aldosterone system (RAAS), mitochondrial dysfunction, and inflammatory dysregulation, the pathophysiological mechanism of SA-AKI remains elusive, and there is currently a lack of optimal measures to support early diagnosis and ongoing management^5,6,7,8. SA-AKI features heightened apoptosis, excessive oxidative stress, and inflammation⁹. Within this research framework, lipopolysaccharide (LPS) was employed in an in vitro system to mimic the manifestations of sepsis-associated AKI in human umbilical vein endothelial cells (HUVECs) for preliminary target exploration⁵.

In recent years, circular RNAs (circRNAs), a recently identified category of non-coding RNA molecules, have gained increasing attention for their role in various diseases¹⁰. CircRNAs possess a stable circular structure and participate in cellular physiological and pathological processes by serving as molecular sponges for microRNAs (miRNAs), regulating gene transcription, or engaging in protein interactions¹¹. Investigations have indicated that circRNAs are critically involved in modulating inflammation, oxidative stress, and apoptotic pathways, thereby generating increased research focus on their possible implications in sepsis and related conditions, including acute kidney injury¹². For instance, hsa_circ_0072463 has been identified as a promising diagnostic marker and therapeutic target for SA-AKI¹³. Circ_001653 was reported to alleviate SA-AKI through the recruitment of BUD13¹⁴. However, the functional significance of circRNAs in SA-AKI remains largely unexplored.

This study was designed to examine the influence of circZBTB46 on endothelial cell function, oxidative stress, and inflammatory responses using an in vitro SA-AKI system. By constructing an LPS-induced HUVEC model, we investigated endothelial injury under endotoxin stimulation. It should be noted that LPS-challenged HUVECs represent a reductionist endotoxin/TLR4-driven endothelial stress model that is relevant to, but not equivalent to, the multifactorial pathophysiology of sepsis-associated acute kidney injury. This model primarily captures TLR4-mediated inflammatory activation, oxidative stress, and apoptosis-related endothelial responses, while omitting key in vivo components such as hemodynamic alterations, complement and coagulation cascades, damage-associated molecular patterns, and heterogeneous microbial stimuli. Within this experimental framework, we investigated whether circZBTB46 alleviates LPS-induced oxidative stress and apoptosis by modulating the ERBB2 signaling pathway, thereby protecting endothelial cell function. To our knowledge, this study provides the first evidence linking circZBTB46 to endothelial stress responses in the context of sepsis-associated acute kidney injury. By elucidating the molecular mechanisms underlying circZBTB46-mediated endothelial protection, this research may deepen the comprehension of SA-AKI's pathogenic mechanisms and pave the way for innovative therapeutic approaches centered on circRNA.

This protocol describes the experimental procedures used to investigate the effects of circZBTB46 modulation on inflammatory signaling, oxidative stress, and ERBB2-AKT pathway activation in HUVECs.

Cell culture
The HUVECs cells used in this study were obtained from a commercial source. HUVECs were used as a representative endothelial cell model because of their wide availability, reproducible culture conditions, and high transfection efficiency, which facilitate genetic manipulation and mechanistic investigation of endothelial signaling pathways under inflammatory stress^15,16. HUVECs were cultured in DMEM supplemented with 10% (v/v) fetal bovine serum and antibiotic-antimycotic cocktail at 1% concentration. All cell cultures were maintained under standard physiological conditions in a humidified chamber (37 °C, 5% CO₂), which promoted sustained cellular proliferation while preserving >95% viability as confirmed by trypan blue exclusion assay. Although endothelial-specific media (e.g., EGM-2 or M199 supplemented with endothelial growth supplements) are commonly recommended for long-term HUVEC maintenance, DMEM was used in this study to ensure consistency across experimental treatments and compatibility with LPS stimulation and pharmacological inhibitor assays. For the AG-825 inhibitor assays, incubate HUVECs with fresh DMEM containing 10 µM AG-825 for 1 h. Then, add the LPS solution to the wells to achieve a final concentration of 10 µg/mL, and continue the culture for 24 h. To minimize potential phenotypic drift, HUVECs were used at low passage numbers and exposed to DMEM-based experimental conditions only for short-term assays. All functional experiments were performed within defined treatment windows rather than during prolonged culture.

Cell transfection
To elucidate the biological role of circZBTB46, overexpression plasmids (oe-circZBTB46) and matched negative control constructs (oe-NC) were synthesized and validated by GenScript. HUVECs were seeded into 6-well plates at a density of 1 x 10⁵ cells/well and cultured overnight to reach ~70%-80% confluence at the time of transfection. For each well, 2.5 µg plasmid DNA (oe-circZBTB46 or oe-NC) was diluted in 125 µL of reduced serum medium and mixed with 2.5 µL of P3000 reagent (Tube A). In parallel, 3.75 µL of Lipofectamine 3000 was diluted in 125 µL of reduced serum medium (Tube B). Tube A and Tube B were then combined, gently mixed, and incubated at room temperature for 15-20 min to allow complex formation (final complex volume: 250 µL/well). The complexes were added dropwise to the cells in antibiotic-free complete medium. After 6 h of incubation at 37 °C, 5% CO₂, the transfection medium was replaced with fresh complete growth medium. Cells were cultured for 48 h prior to subsequent assays. All procedures followed the manufacturer's instructions.

LPS-induced model
To generate an in vitro SA-AKI model, HUVECs (5 x 10⁵ cells/well) transfected with oe-circZBTB46 or NC were seeded in 96-well plates and maintained for 24 h until 80% confluency was achieved. Following this, these cells underwent treatment with LPS at 10 µg/mL for 24 h. HUVECs without LPS treatment served as the control group.

qRT-PCR
Total RNA extraction from HUVECs utilized a commercial reagent. Reverse transcription for cDNA synthesis was performed with the commercial reagent kit. Quantitative PCR (qPCR) was executed in triplicate using a Real-Time PCR System alongside a master mix. The expression quantities of target genes were calibrated against U6, with analysis conducted via the 2^−ΔΔCT method^17,18. For circRNA expression analysis, U6 small nuclear RNA was used as the internal reference for normalization, as it is commonly employed for quantification of small non-coding RNAs and has been reported to exhibit relatively stable expression across short-term in vitro experimental conditions. Primer sequences applied in this study are listed as follows:

circZBTB46-Forward: 5'-CGGCGCTCATGAGTAAGAAC-3'

circZBTB46-Reverse: 5'-CGCCTCTTCTACAGACTGGG-3'

U6-Forward: 5'-TGCTATCACTTCAGCAGCA-3'

U6-Reverse: 5'-GAGGTCATGCTAATCTTCTCTG-3'.

CCK-8 assay
HUVECs' viability was assessed using CCK-8 purchased commercially. In brief, cells were seeded into a 96-well plate, and the cells underwent adhesion culture for 24 h. After this, 10 µL of CCK-8 solution was added to each well, and the cells were further incubated at 37 °C for 2 h. A microplate reader was employed to measure the optical density (OD) at a wavelength of 450 nm. Preliminary observations demonstrated that 10 µg/mL of LPS significantly decreased the viability of HUVECs, leading to the selection of this concentration for subsequent experimental processes.

Flow cytometry
Initially, HUVECs were detached with EDTA-free trypsin according to standard procedures. In accordance with the manufacturer's guidelines for the Annexin V-APC/PI Apoptosis Assay Kit, cell suspensions were centrifuged at 1,000 x g for 5 min at 4 °C, after enzymatic dissociation. The pelleted cells were then washed 2x with ice-cold PBS under identical conditions and resuspended in 100 µL of 1x binding buffer. Each sample was treated with 5 µL of Annexin V-FITC conjugate and 5 µL of propidium iodide (PI) staining solution, followed by a 10 min incubation in darkness at room temperature (20-25 °C). Finally, 400 µL of 1x binding buffer was added to all samples, and apoptosis was analyzed within 1 h using a flow cytometer. Events were first gated by FSC/SSC to exclude debris. For apoptosis analysis, quadrant gates were then applied on the Annexin V-FITC vs PI plot using appropriate controls (unstained and single-stained/compensation controls) to define live (Annexin V^-/PI^-), early apoptotic (Annexin V⁺/PI^-), late apoptotic (Annexin V⁺/PI⁺), and necrotic (Annexin V^-/PI⁺) populations. The same gating template and thresholds were applied across all samples.

ELISA assay
The concentrations of inflammatory factors-tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), and interleukin-1β (IL-1β)-in the supernatant of cultured HUVECs were measured using commercially available ELISA kits. Specifically, the Human TNF-α ELISA Kit, Human IL-6 ELISA Kit, and Human IL-1β ELISA Kit were used according to the manufacturers' instructions. For each assay, 50 µL of undiluted cell culture supernatant was loaded per well. Standard curves were generated using the provided serial dilutions (TNF-α: 0-1000 pg/mL; IL-6: 0-500 pg/mL; IL-1β: 0-200 pg/mL). After sample and standard addition, plates were incubated at 37 °C for 90 min, followed by five washes with wash buffer. Biotin-conjugated detection antibody was then added and incubated at 37 °C for 60 min. After another washing step, streptavidin-HRP was added and incubated at 37 °C for 30 min. Following a final wash, 3,3',5,5'-tetramethylbenzidine (TMB) substrate was added and incubated at 37 °C in the dark for 15 min. The reaction was terminated with stop solution, and absorbance was immediately measured at 450 nm with a reference wavelength of 630 nm. Each sample and standard was assayed in duplicate, and all experiments included three independent biological replicates. The levels of oxidative stress markers-malondialdehyde (MDA), superoxide dismutase (SOD), and catalase (CAT)-in the supernatant were evaluated using the MDA ELISA Kit, SOD ELISA Kit, and CAT ELISA Kit, respectively, following the provided protocols^19,20,21. For MDA and SOD assays, 40 µL of supernatant was used per well without dilution; for CAT, samples were diluted 1:10 with assay buffer prior to loading. Standard curves were prepared according to kit specifications (MDA: 0-60 nmol/mL; SOD: 0-100 U/mL; CAT: 0-100 U/mL). Incubation steps were performed as follows: MDA - 60 min at 37 °C; SOD - 60 min at 37 °C; CAT - 30 min at 37 °C. Plates were washed 4x between key incubation steps. Color development was achieved by adding the respective substrates and incubating at 37 °C (MDA: 15 min; SOD: 20 min; CAT: 15 min). Absorbance was read at 450 nm for SOD and CAT, and at 532 nm for MDA. All measurements were performed in duplicate wells and repeated across three independent biological replicates.

RNA sequencing
Total RNA was extracted from HUVECs using commercial reagents according to the manufacturer's instructions. RNA purity and concentration were assessed using a spectrophotometer by measuring absorbance at 260 nm and 280 nm (A260/A280), and RNA integrity was evaluated with a Bioanalyzer. Only RNA samples with an RNA Integrity Number (RIN) ≥ 7 were used for subsequent library preparation. Ribosomal RNA was removed during library preparation using a standard rRNA depletion procedure done by a commercial company. Where indicated, RNase R treatment was applied prior to library construction to selectively digest linear RNAs and enrich circular RNA species. RNA fragmentation was then performed under elevated temperature conditions following the standard protocol of the RNA Library Preparation Kit manual to generate appropriately sized RNA fragments. Sequencing libraries were prepared using the RNA Library Preparation Kit according to the manufacturer's instructions and sequenced on an Illumina platform using a paired-end sequencing strategy with a read length of 150 bp (PE150). Sequencing depth was sufficient to support transcriptome-wide analysis. Raw sequencing reads were processed using standard quality control procedures to remove adaptor sequences and low-quality reads. Clean reads were aligned to the human reference genome, and gene expression levels were quantified accordingly^22,23. Differentially expressed genes (DEGs) between circZBTB46-overexpressing HUVECs and NC-treated HUVECs under LPS stimulation were identified using established statistical methods. Gene set enrichment analysis (GSEA) was performed to identify signaling pathways associated with circZBTB46 overexpression. Genes were ranked based on differential expression between experimental groups. Curated gene sets from the Molecular Signatures Database (MSigDB) were used for enrichment analysis. Enrichment significance was evaluated based on normalized enrichment score (NES) and false discovery rate (FDR), with FDR < 0.25 considered statistically significant. CircRNA sequencing and all associated bioinformatic analyses were outsourced.

Identification of DEGs
To screen DEGs, the expression levels between two distinct groups were contrasted via the DESeq (2012) R software package. The thresholds for statistical significance were defined as p < 0.05 with a |log2FC| value exceeding 0.58. The heatmap was generated by the pheatmap R package, and the volcano plot used the Goplot2 R package for drawing.

Functional enrichment analysis
GO and KEGG enrichment analyses derived from linear transcripts were conducted using the online platform EnrichR, and the Bubble Chart was drawn based on the GOplot R package. The GSEA software was utilized, based on the expression matrix and gene set files provided by the MSigDB database, with analysis parameters set (e.g., number of permutations and statistical methods). The enrichment of gene sets was evaluated using the normalized enrichment score (NES), P-value, and FDR value, providing an in-depth exploration of the functions and potential biological mechanisms of the differentially expressed genes.

Western blot
Total protein was extracted from HUVECs (approximately 1 x 10⁶ cells) using RIPA lysis buffer supplemented with protease and phosphatase inhibitors. Protein concentration was determined with a BCA Protein Quantification Kit. Equal amounts of protein (20 µg per lane) were resolved by electrophoresis on 10% sodium dodecyl sulfate-polyacrylamide gels (SDS-PAGE) at 80 V for 30 min, followed by 120 V for 60 min. Subsequently, proteins were transferred to polyvinylidene fluoride (PVDF) membranes using a wet-transfer system at 100 V for 90 min at 4 °C. Membranes were blocked with 5% non-fat milk in Tris-buffered saline with 0.1% Tween-20 (TBST) for 1 h at room temperature, followed by incubation with primary antibodies diluted in the same blocking buffer (typically 1:1000 for target proteins and 1:5000 for GAPDH) at 4 °C overnight. The primary antibodies used were as follows: ERBB2, p-AKT, AKT, and GAPDH. After incubation, membranes were washed 3x (10 min per wash) with TBST and then incubated with a horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG secondary antibody (1:5000 in blocking buffer) for 1 h at room temperature. Following three additional TBST washes (10 min each), protein bands were visualized using an Enhanced Chemiluminescence (ECL) Western Blotting Substrate Kit and imaged with an Imaging System. Band intensities were quantified using software.

Statistical analysis
Statistical evaluations were performed using GraphPad Prism 10. Data are displayed as means ± standard error of the means (SEMs). A p-value < 0.05 was defined to indicate statistical significance.

qRT-PCR analysis of circZBTB46 expression in an in vitro SA-AKI model
This set of experiments was conducted to test the hypothesis that circZBTB46 expression is altered under LPS stimulation and that modulation of circZBTB46 affects endothelial cell viability and apoptosis in an in vitro SA-AKI model. Initially, circZBTB46 expression was examined under control conditions and in an in vitro SA-AKI system. As illustrated in Figure 1A, circZBTB46 expression was markedly reduced in LPS-treated HUVECs compared with untreated control cells. This result suggests that the expression of circZBTB46 was inhibited under LPS treatment. In the next step, circZBTB46 was successfully overexpressed in the HUVECs for further investigation (Figure 1B).

CCK-8 assay and flow cytometric analysis of endothelial cell viability and apoptosis
The CCK-8 assay demonstrated that circZBTB46 overexpression significantly improved cell viability in LPS-treated HUVECs compared with NC-transfected cells under identical LPS stimulation (Figure 1C). Furthermore, the apoptosis rate of LPS-induced HUVECs was significantly reduced in the circZBTB46 overexpression group compared with the NC group at both 24 h and 48 h (Figure 1D). Together, these observations support the hypothesis that circZBTB46 modulation influences endothelial cell survival under LPS-induced injury conditions.

ELISA-based analysis of inflammatory cytokine secretion
To test whether circZBTB46 modulation is associated with changes in inflammatory responses and oxidative stress during LPS-induced endothelial injury, inflammatory cytokine production and oxidative stress indicators were assessed. As illustrated in Figure 2A, circZBTB46 overexpression significantly suppressed the secretion of pro-inflammatory mediators (IL-6, TNF-α, and IL-1β) in LPS-treated HUVECs compared with NC-transfected cells. These inflammatory mediators also exhibited a time-dependent decrease following circZBTB46 overexpression. This result indicated that the overexpression of circZBTB46 can alleviate LPS-induced inflammatory response.

Flow cytometric analysis of intracellular ROS and biochemical assays of oxidative stress markers
The levels of intracellular ROS were also significantly reduced in circZBTB46-overexpressing HUVECs relative to the NC group (Figure 2B). The circZBTB46 overexpression group exhibited a marked reduction in MDA levels at 24 h and 48 h, and a notable elevation in SOD and CAT activities relative to the NC group (Figure 2C). These results support the hypothesis that circZBTB46 overexpression is associated with attenuated inflammatory responses and improved oxidative stress status in LPS-induced endothelial injury.

RNA sequencing and pathway enrichment analysis
To reveal the underlying mechanism of overexpression of circZBTB46 in HUVECs of the SA-AKI model, RNA sequencing was performed. As depicted in Figure 3A, 118 differentially expressed genes (DEGs) were identified between LPS-treated HUVECs overexpressing circZBTB46 and LPS-treated NC-transfected HUVECs, consisting of 69 that were upregulated and 49 that were downregulated. The clustering heatmap displays the expression profiles of DEGs of circZBTB46 between the LPS-treated HUVEC overexpression group and the NC-treated HUVEC group (Figure 3B). Following this, the identified differentially expressed genes underwent enrichment analyses for Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Through GO enrichment analysis, pathways such as cellular response to stress and reaction to external stimulus were identified as potentially related to LPS-induced oxidative stress (Figure 3C). Additionally, KEGG enrichment analysis revealed that the NF-kappa B signaling pathway and the IL-17 signaling pathway play crucial roles in inflammatory responses and autoimmune diseases (Figure 3D). These transcriptomic findings support the hypothesis that circZBTB46 overexpression is associated with broad gene expression and pathway-level changes reflecting downstream inflammatory and oxidative-stress programs in the SA-AKI model.

Gene set enrichment analysis and Western blot analysis
To further investigate signaling pathways potentially responsible for the transcriptomic inflammatory and oxidative-stress programs identified above, we next focused on pathway-level regulators associated with endothelial survival and redox homeostasis. Although RNA-seq enrichment highlighted inflammatory and stress-response pathways such as NF-κB and IL-17signaling, these pathways largely represent downstream transcriptional programs activated under LPS-induced endothelial injury24,25. In contrast, ERBB2-AKT signaling is a well-established regulatory axis in endothelial cells that modulates cell survival, apoptosis, and oxidative stress responses. Therefore, we examined whether circZBTB46-associated ERBB2-AKT activation functionally correlates with the inflammatory and oxidative injury signatures observed at the transcriptomic level in the LPS-induced SA-AKI model. To test whether ERBB2/AKT signaling is involved in circZBTB46-associated endothelial responses in the context of the inflammatory and oxidative-stress programs identified above, pathway enrichment and functional validation experiments were performed. GSEA based on DEGs indicated significant enrichment of the ERBB2 signaling pathway in circZBTB46-overexpressing, LPS-treated HUVECs (Figure 4A). Consistently, protein expression analysis showed that ERBB2 and phosphorylated AKT (p-AKT) levels were markedly increased in the circZBTB46 overexpression group compared with the NC group (Figure 4B).

Pharmacological inhibition assay using ERBB2 inhibitor AG-825
Treatment with the ERBB2 inhibitor AG-825 significantly reduced ERBB2 and p-AKT expression compared with the circZBTB46 overexpression group without inhibitor treatment, indicating that circZBTB46 up-regulates ERBB2 expression and promotes AKT phosphorylation, while ERBB2 inhibition interrupts this process. CCK-8 assay analysis demonstrated that circZBTB46 promotes cellular growth, while the effect that was diminished upon ERBB2 inhibition (Figure 4C). ELISA results indicated that MDA content was markedly reduced in the circZBTB46 overexpression group relative to controls, whereas SOD and CAT activities were remarkably elevated. Following the administration of AG-825, MDA levels increased, and SOD and CAT levels decreased, indicating that circZBTB46 alleviates oxidative stress injury and augments antioxidative defense mechanisms, dependent on the ERBB2/AKT pathway (Figure 4D). Flow cytometry revealed that reactive oxygen species (ROS) production was notably reduced in the circZBTB46 overexpression group relative to the control cohort, whereas AG-825 administration resulted in a significant elevation of ROS levels. This further supports that circZBTB46 overexpression is associated with reduced intracellular ROS levels and attenuated oxidative stress, consistent with the involvement of ERBB2/AKT signaling (Figure 4E). Collectively, these results support the hypothesis that ERBB2/AKT signaling is involved in circZBTB46-associated modulation of oxidative stress, inflammatory responses, and endothelial cell function in the LPS-induced SA-AKI model.

Collectively, the results presented in this section demonstrate that circZBTB46 expression is altered under LPS-induced conditions and that circZBTB46 overexpression modulates endothelial cell viability, apoptosis, inflammatory responses, and oxidative stress in an in vitro SA-AKI model. Transcriptomic profiling and pathway enrichment analyses further revealed circZBTB46-associated gene expression changes, while ERBB2-AKT pathway analysis and pharmacological inhibition experiments provided functional evidence linking circZBTB46 to endothelial protective signaling mechanisms under endotoxin stress.

Data availability:
The raw data supporting the findings of this study, including original data from Western blot, CCK-8 assay, qRT-PCR, FACS, and ELISA experiments, are provided in the Supplementary Table 1 and Supplementary File 1. Due to institutional policies of Fuzhou University Affiliated Provincial Hospital, RNA sequencing data cannot be deposited in public repositories. However, all other raw experimental data generated in this study have been fully provided in the Supplementary Files. Additional information related to the sequencing analysis is available from the corresponding author upon reasonable request (sy1555@fzu.edu.cn).

Figure 1: CircZBTB46 expression and functional assessment in LPS-treated endothelial cells. (A) Relative expression of circZBTB46 in HUVECs following lipopolysaccharide (LPS) treatment was measured by quantitative real-time PCR. (B) The overexpression efficiency of circZBTB46 was verified after plasmid transfection. (C) Cell viability under the indicated conditions was assessed using the CCK-8 assay at the indicated time points. (D) Apoptosis of HUVECs was evaluated by flow cytometric analysis using Annexin V/PI staining. n = 3 independent biological samples, and all experiments were performed in triplicate. Statistical significance was assessed using Student's t-test or one-way ANOVA, as indicated. Data are presented as mean ± standard error of the means (SEMs), with *** p < 0.001. Please click here to view a larger version of this figure.

Figure 2: Effects of circZBTB46 on inflammatory response and oxidative stress in LPS-treated HUVECs. (A) The concentrations of TNF-α, IL-6, and IL-1β in cell culture supernatants were quantified by ELISA and are expressed as pg/mL. (B) Intracellular reactive oxygen species (ROS) levels were assessed by fluorescence-based flow cytometry, with representative histograms and corresponding quantitative analysis shown. (C) Oxidative stress-related indices, including malondialdehyde (MDA) content (nmol/mL) and superoxide dismutase (SOD) and catalase (CAT) activities (U/mL), were determined using commercial biochemical assay kits. All measurements were normalized to total protein content. For all assays, n = 3 independent biological samples, and each experiment was independently repeated at least three times. Statistical analyses were performed using Student's t-test or one-way ANOVA, as appropriate. Data are presented as mean ± standard error of the means (SEMs). p < 0.01 and p < 0.001. Please click here to view a larger version of this figure.

Figure 3: Transcriptomic profiling and functional enrichment analysis associated with circZBTB46 overexpression. (A) Volcano plot showing differentially expressed genes between circZBTB46-overexpressing and control HUVECs. (B) Heatmap of representative differentially expressed genes identified by RNA sequencing. (C) Gene Ontology (GO) enrichment analysis of differentially expressed genes. (D) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of differentially expressed genes. Please click here to view a larger version of this figure.

Figure 4: ERBB2/AKT pathway involvement in circZBTB46-associated endothelial responses. (A) Gene set enrichment analysis (GSEA) was performed to assess enrichment of ERBB-related signaling pathways in circZBTB46-overexpressing HUVECs under LPS stimulation compared with NC-transfected controls. (B) Protein expression levels of ERBB2, total AKT, and phosphorylated AKT (p-AKT) in HUVECs under the indicated treatment conditions were examined by Western blotting, with representative blots and densitometric quantification shown. (C) Oxidative stress-related indices, including malondialdehyde (MDA) content (nmol/mL) and superoxide dismutase (SOD) and catalase (CAT) activities (U/mL), were measured following pharmacological inhibition of ERBB2 signaling. All measurements were normalized to total protein content. (D) Intracellular reactive oxygen species (ROS) levels were assessed by fluorescence-based flow cytometry under the indicated treatments. n = 3 independent biological samples, and all experiments were performed in triplicate. Statistical significance was determined using one-way ANOVA with appropriate post hoc tests, as indicated. Data are presented as mean ± standard error of the means (SEMs), with * p < 0.05, ** p < 0.01, *** p < 0.001. Please click here to view a larger version of this figure.

Supplementary Table 1: Raw data of the figures. Please click here to download this File.

Supplementary File 1: Raw data of flow cytometer. Please click here to download this File.

During sepsis, pathogen-associated molecular patterns (PAMPs), such as LPS released by pathogens, activate Toll-like receptors (e.g., TLR4), initiating a systemic inflammatory response²⁶. This process triggers the secretion of multiple pro-inflammatory mediators, such as TNF-α, IL-6, and IL-1β, resulting in a cytokine storm²⁷. Simultaneously, LPS directly damages vascular endothelial cells, resulting in microcirculatory disturbances, increased vascular permeability, and thrombosis, which ultimately impair renal blood flow²⁷. Additionally, LPS-induced oxidative stress generates excessive ROS, further damaging renal tubular epithelial and endothelial cells, and promoting cell apoptosis through a mitochondrial-dependent pathway⁴. These processes collectively result in tubular dysfunction, decreased glomerular filtration rate, and ultimately acute kidney injury⁹. Therefore, the core mechanisms of SA-AKI encompass intricate interactions among inflammatory responses, endothelial dysfunction, oxidative stress, and apoptosis.

Previous studies have demonstrated that circRNAs exert critical regulatory functions in the pathogenesis of various diseases^10,28. In the field of sepsis and its related complications, although the functions of some circRNAs have been elucidated, the roles of most circRNAs in SA-AKI remain unclear. Investigations into the dysregulated functions of circRNAs in SA-AKI hold the potential to offer valuable clues for pinpointing therapeutic targets. Functioning as a novel non-coding RNA, circZBTB46 is generated through back-splicing of exons from the ZBTB46 gene, which encodes zinc finger and BTB domain-containing protein 46²⁹. Experimental evidence confirms that circZBTB46 knockdown induces hnRNPA2B1 polyubiquitination and proteasome-mediated breakdown via the AKT-mTOR signaling axis. This modulation, in turn, alleviates the progression of atherosclerosis³⁰. It was also reported that circZBTB46 confers protection against ferroptotic cell death in acute myeloid leukemia through upregulation of SCD³¹. Nevertheless, the role that circZBTB46 plays in the SA-AKI has not been elucidated yet.

In this study, we initially identified the protective role of circZBTB46 against LPS-induced damage in HUVECs and further elucidated its molecular mechanisms. This study is the first to reveal the protective role of circZBTB46 in HUVECs during SA-AKI. In the LPS-induced HUVEC injury model, overexpression of circZBTB46 significantly enhanced cell viability and decreased the apoptosis rate. It effectively mitigated the inflammatory response by decreasing the secretion of inflammatory factors such as TNF-α, IL-6, and IL-1β. Meanwhile, it improved the oxidative stress status by lowering ROS and MDA levels and enhancing SOD and CAT activities.

The ERBB2/AKT pathway is a crucial signaling route in cells, significantly involved in biological processes such as cell proliferation, survival, metabolism, and apoptosis²⁵. ERBB2 (also known as HER2), a member of the epidermal growth factor receptor (EGFR) family, belongs to receptor tyrosine kinases (RTKs)³². When ERBB2 binds to its ligand or gets activated, it triggers receptor dimerization and activates its intrinsic tyrosine kinase activity, thus transmitting signals by phosphorylating downstream molecules³³. AKT (also called protein kinase B, PKB), a key downstream effector of ERBB2, is a core component of the PI3K/AKT/mTOR signaling pathway³⁴. Once ERBB2 is activated, PI3K catalyzes the generation of PIP3, which draws AKT to the cell membrane for phosphorylation and activation³⁵. Activated AKT governs cell survival, proliferation, metabolism, and apoptosis by phosphorylating multiple downstream target molecules³⁶. For example, AKT can inhibit the activity of pro-apoptotic proteins like BAD and Caspase-9 9 while activating anti-apoptotic and pro-survival signaling pathways³⁷. Specifically, AKT suppresses the secretion of pro-inflammatory factors by restraining the NF-κB signaling pathway and lowers ROS levels by modulating antioxidant enzyme activities, including SOD and CAT³⁸. Moreover, AKT can decrease apoptosis by suppressing the mitochondrial-dependent apoptosis pathway³⁹.

In SA-AKI, we found that the activation of the ERBB2/AKT pathway is thought to play a vital protective role. Nevertheless, our data indicate that circZBTB46 overexpression is associated with cytoprotective effects that are consistent with increased ERBB2/AKT signaling activity. GSEA indicated that the overexpression of circZBTB46 activated the ERBB2 pathway, upregulated ERBB2 expression, and promoted the phosphorylation of AKT. After blocking this pathway with the ERBB2 inhibitor AG - 825, the beneficial effects of circZBTB46 on cell viability, inflammatory response, and oxidative stress were attenuated. Together, these results support the involvement of ERBB2/AKT signaling in the protective effects associated with circZBTB46 overexpression.

Several limitations of the present study should be acknowledged. First, all experiments were performed in an in vitro endothelial injury model using LPS-treated HUVECs, which captures key features of endotoxin-driven endothelial inflammation and oxidative stress but does not recapitulate the full complexity of SA-AKI in vivo, including multicellular interactions, hemodynamic alterations, and immune-organ crosstalk. Second, the absence of validation in animal models or clinical samples limits extrapolation of these findings to whole-organ and patient contexts. Third, although HUVECs provide a practical and experimentally tractable endothelial model, renal microvascular and glomerular endothelial cells represent the most physiologically relevant endothelial populations in SA-AKI and were not examined in this study. In addition, mechanistic analyses relied primarily on circZBTB46 overexpression, without loss-of-function or rescue experiments. Therefore, the current findings are best interpreted as being consistent with ERBB2-AKT involvement, while definitive causal relationships will require further validation in future studies.

Despite these limitations, this study advances understanding of circRNA involvement in SA-AKI by moving beyond descriptive expression analyses toward a functional and pathway-oriented framework. By integrating endothelial functional assays, transcriptomic profiling, pathway enrichment analysis, and pharmacological inhibition, this work provides a systematic evaluation of circZBTB46-associated endothelial responses under endotoxin stress, thereby offering mechanistic insight relevant to circRNA-mediated vascular injury in SA-AKI.

Importantly, the methodological strategy established in this study has broader applicability beyond circZBTB46. The LPS-induced endothelial injury model, combined with quantitative assessments of inflammation, oxidative stress, apoptosis, and transcriptomic changes, provides a versatile platform that can be readily adapted to investigate other circRNAs, signaling pathways, or forms of endothelial dysfunction relevant to sepsis, vascular inflammation, and kidney injury. Future studies should extend these findings through complementary loss-of-function approaches, such as circZBTB46 knockdown or genetic perturbation, validation in established animal models of sepsis-associated AKI, and analysis of patient-derived endothelial or kidney tissues. Additional work exploring circRNA-targeted delivery strategies or evaluating circZBTB46 as a potential biomarker in clinical cohorts may further clarify its translational relevance.