The Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/gds/) provided the sepsis-related dataset GSE89376. A total of 12 sepsis and 12 control samples were analyzed. Differential expression was assessed using the “limma” package in R (version 4.0; R Foundation for Statistical Computing, Vienna, Austria), with thresholds set at FC $>$1.3 for upregulation, FC 0.7 for downregulation, and p 0.05 for statistical significance [19]. To further identify hub genes, the dataset GSE67652 was also obtained from the GEO database for validation. This dataset included samples from an additional 12 septic patients and 12 healthy controls.

DEGs were subjected to functional enrichment employing the Database for Annotation, Visualization, and Integrated Discovery (DAVID; https://davidbioinformatics.nih.gov/summary.jsp). Gene ontology, along with Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, was identified and visualized.

To explore interactions among DEGs, a PPI network was established using the STRING platform (https://string-db.org/). Key modules were identified in Cytoscape (version 3.8.1; Cytoscape Consortium San Diego, CA, USA) via CytoHubba with Degree, Maximum Neighborhood Component (MNC), and Maximal Clique Centrality (MCC) algorithms. Overlapping genes across modules were identified through the Venn analysis tool (https://bioinformatics.psb.ugent.be/webtools/Venn/). Their expression patterns were subsequently validated in the GSE89376 and GSE67652 datasets by comparing septic and control samples. Data analysis and visualization were performed with Sangerbox (http://vip.sangerbox.com/home.html).

A total of nine participants were enrolled in this investigation, including three patients with sepsis-induced ALI, three patients with sepsis without pulmonary involvement, and three non-septic control patients without pulmonary inflammation. Sepsis patients were admitted to the Department of Intensive Care Medicine. In contrast, control patients were recruited from the Department of Hepatobiliary Surgery, the Third Affiliated Hospital of Naval Military Medical University. All participants were age- and gender-matched. Sepsis was identified using the Sepsis-3 criteria, which include life-threatening organ dysfunction and a Sequential Organ Failure Assessment (SOFA) score rise of $\ge$2 caused by a dysregulated host response to infection. Control participants were hospitalized for non-infectious hepatobiliary conditions and showed no evidence of systemic disease, sepsis, or lung injury at the time of sampling. All participants and their legal guardians provided written informed consent. The study protocol was evaluated and approved by Academic Committee of the Third Affiliated Hospital of Naval Medical University (Approval Number: KY2024081), and it followed the Declaration of Helsinki. Since the patient’s condition is fixed, random allocation is not possible. All subjects had peripheral venous blood drawn within 24 hours of either elective admission (for controls) or intensive care unit (ICU) admission (for septic patients). After centrifugation of blood specimens (3000 rpm, 10 min, 4 °C), serum was collected, portioned, and maintained at –80 °C pending analysis. Serum concentrations of CD3D, MyD88, high-mobility group box 1 (HMGB1), NLR family pyrin domain containing 3 (NLRP3), and phosphorylation of nuclear factor-$\kappa$B (p-NF-$\kappa$B) were detected by human enzyme-linked immunosorbent assay (ELISA) kits. Specifically, the human CD3D ELISA kit (Cat. no. XY-EH1234) was purchased from Xinyu Biotechnology Co., Ltd. (Shanghai, China). ELISA kits for HMGB1 (Cat. no. JL13693-48T), p-NF-$\kappa$B (Cat. no. JL42948-48T), and NLRP3 (Cat. no. JL10272-48T) were obtained from Jianglai Biological Technology (Shanghai, China). The MyD88 ELISA kit (Cat. no. ab171341) was sourced from Abcam (Shanghai, China). The manufacturers’ protocols applied to all assays. Utilizing a microplate reader (Cat. no. DNM-9606; Perlong, Beijing, China), absorbance was measured at 450 nm.

hPMECs (Cat. no. CP-H001, Procell, Wuhan, China) were cultured in complete hPMEC medium (Cat. no. CM-H001, Procell, Wuhan, China) under standard conditions of 37 °C with 5% CO₂. According to the supplier’s quality control statement, hPMECs were authenticated by CD31 immunofluorescence and tested to be over 90% pure. Additionally, the supplier confirmed that the cells were free from human immunodeficiency virus (HIV)-1, hepatitis B virus (HBV), hepatitis C virus (HCV), Mycoplasma, bacteria, fungi, and yeast contamination. To establish a sepsis-induced ALI model, cells were exposed to LPS (10 µg/mL; Cat. no. L8880, Solarbio, Beijing, China) for 24 h. For evaluating the protective role of calcitonin, LPS-treated hPMECs were further incubated with calcitonin (Cat. no. T3535, Sigma-Aldrich, St. Louis, MO, USA) at concentrations of 1, 5, and 10 nM for 24 h.

After being seeded at a density of 2 $\times$ 10⁵ cells per well in 24-well plates, hPMECs were cultivated in full growth media for the whole night until they attained a confluency of around 70–80%. As directed by the manufacturer, Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA, USA) was used for transfection. In particular, hPMECs were transfected with either CD3D-targeting small interfering RNAs (siRNAs; si-CD3D-1 and si-CD3D-2) or a negative control siRNA (si-NC). After 48 hours of transfection, cells were harvested for further experimentation.

RNA purity and concentration were determined with a NanoDrop spectrophotometer (Thermo Fisher Scientific, Shanghai, China). First-strand cDNA was synthesized from 1 µg of RNA using the PrimeScript RT Kit (Takara, Shiga, Japan). qRT-PCR was performed on a StepOnePlus Real-Time PCR System (Applied Biosystems, Shanghai, China) using SYBR Green PCR Master Mix (Cat. no. 4309155; Thermo Fisher Scientific, Waltham, MA, USA). All reactions were performed in triplicate. Relative gene expression levels were calculated using the 2^{-Δ⁢Δ⁢Ct} method, with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) serving as the internal control. Primer sequences are provided in Table 1.

hPMEC proteins were isolated using radioimmunoprecipitation assay (RIPA) buffer (Cat. no. 89900; Thermo Fisher Scientific, Waltham, MA, USA) supplemented with protease and phosphatase inhibitors. Protein levels were measured using a bicinchoninic acid (BCA) assay (Cat. no. P0011; Beyotime, Shanghai, China). Equal quantities of protein were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently blotted onto poly-vinylidene difluoride (PVDF) membranes (Cat. no. FFP39; Beyotime, Shanghai, China). The membranes were blocked and subsequently incubated with primary antibodies against CD3D (1:1000, ab109531), p21 (1:2000, ab109520), Cyclin D1 (1:1000, ab109531), cyclin dependent kinase 2 (CDK2) (1:1000, ab32147), cyclooxygenase-2 (COX-2) (1:2000, ab179800), inducible Nitric Oxide Synthase (iNOS) (1:2000, ab178945), HMGB1 (1:1000, ab92310), MyD88 (1:1000, ab133739), p-NF-$\kappa$B (1:1000, ab207297), NF-$\kappa$B (1:1000, ab283716), NLRP3 (1:1000, ab263899), Bcl-2 (1:2000, ab182858), Bax (1:2000, ab32503), and Caspase-3 (1:5000, ab32351), with GAPDH (1:2500, ab9485) serving as a loading control. All antibodies were purchased from Abcam (Shanghai, China). After secondary antibody incubation with Goat Anti-Rabbit IgG H&L (horseradish peroxidase (HRP), 1:10,000, ab6721), proteins were visualized by enhanced chemiluminescence (ECL) (Cat. no. P0018FS; Beyotime, Shanghai, China), and band intensities were quantified through ImageJ (version 1.54; National Institutes of Health, Bethesda, MD, USA) analysis.

Cell proliferation and viability were determined by CCK-8 assay (Cat. no. C0042; Beyotime, China). Briefly, hPMECs were seeded in 96-well plates at a density of 5 $\times$ 10³ cells per well with 100 µL of complete culture medium. Untreated or treated cells were incubated for 24, 36, 48, 72, and 96 hours. CCK-8 solution (10 µL) was added at each time point, and optical density at 450 nm was measured after 2 h using a microplate reader (Model DNM-9602; Perlong, Beijing, China).

Apoptosis and cell-cycle profiles of hPMECs were determined by flow cytometry. After 24 h culture in 24-well plates (1 $\times$ 10⁵ cells/well), cells for apoptosis assays were digested with trypsin-ethylenediaminetetraacetic acid (EDTA), rinsed in phosphate-buffered saline (PBS), and incubated with Annexin V-FITC (5 µL) and PI (5 µL) in the dark for 15 min. Stained cells were examined on a flow cytometer (ACEA Biosciences, China), and results were analyzed with FlowJo (Version 10.8.1, FlowJo, LLC, Ashland, OR, USA). To analyze the cell cycle, cells were fixed with 75% ethanol (4 h, 4 °C), stained with PI (50 µg/mL) plus RNase A (10 mg/L) for 1 h at 37 °C, and subjected to flow cytometry.

Tumor necrosis factor (TNF)-$\alpha$ (Cat. no. MM-0122H2), interleukin (IL)-6 (Cat. no. MM-0049H2), and IL-1$\beta$ (Cat. no. MM-0181H2) levels in culture supernatants were measured using ELISA kits (Meimian, Jiangsu, China) according to the manufacturer’s instructions. After adding samples and standards to capture antibody-coated wells, plates were incubated for 2 h at 37 °C. Detection was performed using HRP-conjugated antibodies and 3,3^′,5,5-tetramethylbenzidine (TMB) substrate, and absorbance was measured at 450 nm with a microplate reader (Model DNM-9602; Perlong, Beijing, China). The concentrations of cytokines were measured using standard curves as a guide.

To determine cellular metabolic and oxidative stress markers, hPMECs were cultured in 24-well plates (1 $\times$ 10⁵ cells/well), washed with PBS, and lysed. After centrifugation at 10,000 $\times$g for 5 min at 4 °C, supernatants were collected. Biochemical parameters were analyzed with commercial kits (Nanjing Jiancheng Bioengineering Institute, China), including lactate dehydrogenase (LDH, Cat. no. A020-1-2) activity (440 nm), malondialdehyde (MDA, Cat. no. A003-4-1) content (532 nm), superoxide dismutase (SOD, Cat. no. A001-3-2) activity (450 nm), and glutathione (GSH, Cat. no. A006-1-1) levels (420 nm).

At least three independent experiments were carried out for each dataset. Statistical evaluation was conducted in R. Differences between the two groups were determined using Student’s t-test. In contrast, multiple-group comparisons were analyzed by one-way analysis of variance (ANOVA) with Tukey’s post-hoc test. Data are shown as mean $\pm$ SD, with statistical significance defined at p 0.05.

From the GSE89376 dataset, 182 genes were found to be upregulated and 178 downregulated (Fig. 1A). DEGs were significantly enriched in biological processes (BP)-related GO terms, mainly involving “Positive regulation of apoptotic process”, “Regulation of T cell proliferation”, and “Positive regulation of vascular endothelial growth factor production” (Fig. 1B). For cellular components (CC), enrichment was observed in “Coated vesicle membrane”, “ER-to-Golgi transport vesicle membrane”, and “Endocytic vesicle membrane” (Fig. 1C). Regarding molecular function (MF), DEGs were enriched in “Protein kinase activator activity”, “Exopeptidase activity”, and “Calcium channel regulator activity” (Fig. 1D). KEGG pathway analysis further indicated enrichment in “Human T-cell leukemia virus 1 infection”, “Cell adhesion molecules”, as well as “Hematopoietic cell lineage” (Fig. 1E).

Fig. 1.

Differential gene expression and functional enrichment analysis. (A) Volcano plot displaying identified DEGs from the GSE89376 dataset. Each point represents an individual gene. Up-regulated DEGs are depicted in pink, and down-regulated DEGs are depicted in blue. (B) GO enrichment analysis of BP terms associated with DEGs. The x-axis represents the GeneRatio, and the y-axis indicates the enriched terms. The size of the dots corresponds to the number of genes, and the color reflects the adjusted p value. (C) GO enrichment analysis of CC terms of DEGs. (D) GO enrichment analysis of MF terms of DEGs. (E) KEGG enrichment analysis predicted the pathways of DEGs involvement. DEGs, Differential Expressed Genes; GO, Gene Ontology; BP, Biological process; CC, Cell component; MF, Molecular function; KEGG, Kyoto Encyclopedia of Genes and Genomes.

Candidate genes associated with sepsis were identified using PPI network analysis. Three topological algorithms (MNC, MCC, and Degree) were applied to screen for hub genes, and the top 10 genes ranked by each algorithm were selected for comparison (Fig. 2A–C). Venn diagram analysis revealed four overlapping genes across all three algorithms: CR7, CD247, CD3D, and CD69 (Fig. 2D). Expression profiling using the GSE89376 dataset showed significant upregulation of all four genes in sepsis patients compared with controls (Fig. 2E). Validation in an independent dataset (GSE67652) confirmed their consistently elevated expression (Fig. 2F). In the GSE89376 dataset, the Cohen’s d for CD3D was 3.36, with a 95% confidence interval (CI) of 0.43–0.71 (p 0.0001); In the GSE67652 dataset, the Cohen’s d for CD3D was 2.10, with a 95% CI of 0.26–0.61 (p 0.0001). These results indicate that CD3D exhibited a significant effect size in both datasets, suggesting it may play an essential role in related biological processes. Notably, CD3D displayed the most important differential expression in both datasets, making it a candidate for further functional studies.

Fig. 2.

Identification of CD3D as a key hub gene in sepsis pathogenesis. (A) PPI network constructed using the MCC algorithm, showing the top ten highly interconnected genes (10 nodes and 45 edges). (B) PPI network constructed using the MNC algorithm, displaying the top ten hub genes (10 nodes and 34 edges). (C) PPI network constructed using the Degree algorithm, highlighting the top ten hub genes (10 nodes and 33 edges). (D) Venn diagram of the top ten genes identified by the MCC, MNC, and Degree algorithms, with four overlapping hub genes obtained. (E) Box plots of the expression levels of overlapping hub genes (CCR7, CD247, CD3D, and CD69) in sepsis and normal samples from the GSE89376 dataset. (F) Validation of the expression of overlapping hub genes (CCR7, CD247, CD3D, and CD69) in the GSE67652 dataset. **p 0.01 or ***p 0.001 or ****p 0.0001 vs. Normal group. PPI, Protein-Protein Interaction; MCC, Maximal Clique Centrality; MNC, Maximum Neighborhood Component.

HMGB1 has been reported to activate NF-$\kappa$B via a MyD88-dependent pathway, thereby providing transcriptional priming for NLRP3 inflammasome expression [20]. A previous study has highlighted the critical role of NF-$\kappa$B signaling in sepsis [21]. To investigate immune-inflammatory signaling changes in sepsis and sepsis-induced ALI, serum levels of CD3D, HMGB1, MyD88, NF-$\kappa$B, and NLRP3 were measured by ELISA. Compared with the healthy controls, sepsis patients exhibited significantly higher concentrations of CD3D (Cohen’s d = 2.63, 95% CI: 0.44–4.8), HMGB1 (Cohen’s d = 4.48, 95% CI: 1.48–7.48), MyD88 (Cohen’s d = 3.66, 95% CI: 1.04–6.28), NF-$\kappa$B (Cohen’s d = 4.86, 95% CI: 1.68–8.05), and NLRP3 (Cohen’s d = 4.69, 95% CI: 1.59–7.79). Moreover, serum levels of CD3D (Cohen’s d = 2.20, 95% CI: 0.17–4.22), HMGB1 (Cohen’s d = 5.55, 95% CI: 2.02–9.07), MyD88 (Cohen’s d = 3.04, 95% CI: 0.69–5.38), NF-$\kappa$B (Cohen’s d = 5.13, 95% CI: 1.81–8.44), NLRP3 (Cohen’s d = 10.98, 95% CI: 4.57–17.40) were further elevated in sepsis patients with ALI compared to those without pulmonary involvement (Fig. 3A–E). Collectively, these findings suggest that HMGB1 promotes NLRP3 inflammasome activation via a MyD88-dependent NF-$\kappa$B pathway, thereby contributing to inflammatory injury in sepsis and sepsis-associated ALI.

Fig. 3.

Sepsis and sepsis-induced ALI are associated with elevated serum levels of CD3D, HMGB1, MyD88, p-NF-$\kappa$B, and NLRP3. (A) Serum CD3D concentrations were measured by ELISA in the control, patients with sepsis (Sepsis), and patients with sepsis-associated acute lung injury (Sepsis-ALI). (B) Serum HMGB1 concentrations were measured by ELISA in the control, Spesis, and Sepsis-ALI groups. (C) Serum MyD88 concentrations were measured by ELISA in the control, Spesis, and Sepsis-ALI groups. (D) Serum p-NF-$\kappa$B concentrations were measured by ELISA in the control, Spesis, and Sepsis-ALI groups. (E) Serum NLRP3 concentrations were measured by ELISA in the control, Spesis, and Sepsis-ALI groups. Each assay was performed in triplicate to ensure accuracy and reproducibility. Statistical significance was determined by comparing the mean values of each group. *p 0.05 vs. Control group, ^#p 0.05 vs. Sepsis group. ALI, acute lung injury; HMGB1, high-mobility group box 1; p-NF-$\kappa$B, phosphorylation of nuclear factor-$\kappa$B; NLRP3, NLR family pyrin domain containing 3; ELISA, enzyme-linked immunosorbent assay.

The knockdown efficiency of two CD3D-targeting siRNAs in hPMECs was assessed by qRT-PCR and WB. Both siRNAs effectively reduced CD3D expression, with si-CD3D-2 showing the most substantial effect and therefore used for subsequent experiments (Fig. 4A–C). To mimic the septic inflammatory microenvironment, hPMECs were stimulated with LPS. LPS significantly increased CD3D expression compared with controls, whereas CD3D knockdown suppressed this induction (Fig. 4D–F). ELISA analysis revealed that LPS stimulation significantly promoted the secretion of TNF-$\alpha$, IL-6, and IL-1$\beta$, whereas CD3D knockdown reduced these pro-inflammatory effects (Fig. 4G). CCK-8 assays revealed that LPS inhibited hPMEC proliferation, whereas CD3D knockdown partially restored proliferation (Fig. 4H). These results indicate that CD3D knockdown mitigates LPS-induced inflammatory cytokine production and inhibition of proliferation in hPMECs.

Fig. 4.

Knockdown of CD3D alleviates LPS-induced ALI in hPMECs. (A) qRT-PCR analysis of CD3D mRNA expression in hPMECs following transfection with CD3D-specific siRNA. *p 0.05 vs. si-NC group. (B) WB analysis of CD3D protein expression after siRNA transfection in hPMECs. (C) Quantification of CD3D protein levels following knockdown. *p 0.05 vs. si-NC group. (D) qRT-PCR analysis of CD3D mRNA levels in hPMECs treated with LPS alone or in combination with CD3D knockdown. (E) WB analysis of CD3D protein expression in LPS-treated hPMECs with or without CD3D knockdown. (F) Quantification of CD3D protein levels under LPS stimulation with or without CD3D knockdown, normalized with GAPDH. (G) ELISA-based quantification of TNF-$\alpha$, IL-6, and IL-1$\beta$ levels in the supernatants of hPMECs following LPS stimulation with or without CD3D knockdown. (H) Cell viability assessment by CCK-8 assay in hPMECs exposed to LPS with or without CD3D knockdown. hPMECs, human pulmonary microvascular endothelial cells; qRT-PCR, Quantitative Real-Time Polymerase Chain Reaction; WB, Western Blotting; CCK-8, Cell Counting Kit-8; LPS, Lipopolysaccharide. Each assay was performed in triplicate to ensure accuracy and reproducibility. Statistical significance was determined by comparing the mean values of each group. *p 0.05.

To investigate the effect of CD3D knockdown on cell cycle progression, flow cytometry was performed. LPS exposure reduced the S-phase population and increased the proportion of cells in the G1 phase, indicating G1 arrest. CD3D knockdown partially reversed this effect, reducing G1 accumulation and increasing the S-phase fraction (Fig. 5A). Consistently, qRT-PCR showed that LPS treatment upregulated p21 expression while downregulating Cyclin D1 and CDK2. These transcriptional changes were reversed by CD3D knockdown and further confirmed at the protein level by WB (Fig. 5B–D). These findings suggest that CD3D silencing alleviates LPS-induced G1 arrest by modulating cell cycle regulators.

Fig. 5.

Effect of CD3D knockdown on sepsis-related LPS-induced hPMECs cell cycle. (A) Flow cytometry analysis depicts the cell cycle distribution in hPMECs stimulated with LPS alone or in combination with CD3D knockdown. (B) qRT-PCR analysis of p21, Cyclin D1, and CDK2 mRNA expression in hPMECs treated with 10 µg/mL LPS alone or in combination with CD3D knockdown. (C) WB analysis of p21, Cyclin D1, and CDK2 protein levels in hPMECs following 24-hour LPS stimulation with or without CD3D knockdown. (D) Quantification of p21, Cyclin D1, and CDK2 protein expression levels based on WB results. Each assay was performed in triplicate to ensure accuracy and reproducibility. Statistical significance was determined by comparing the mean values of each group. *p 0.05.

To further evaluate the role of CD3D in sepsis-induced ALI, particularly in relation to oxidative stress, ELISA was used to measure LDH, MDA, SOD, and GSH levels. LPS stimulation significantly increased LDH and MDA, while decreasing SOD and GSH, indicating enhanced oxidative stress and damage. CD3D knockdown reversed these effects by lowering LDH and MDA and restoring GSH and SOD (Fig. 6A–D). In addition, LPS strongly upregulated COX-2 and iNOS, two key inflammatory mediators, at both mRNA and protein levels, whereas CD3D knockdown suppressed this induction (Fig. 6E–I). These results suggest that CD3D plays a role in LPS-induced oxidative stress and inflammatory responses in hPMECs.

Fig. 6.

Effects of CD3D knockdown on oxidative stress and inflammatory markers in LPS-treated hPMECs. (A) LDH release in hPMECs after treatment with 10 µg/mL LPS alone or in combination with CD3D knockdown for 24 hours. (B) SOD activity in hPMECs under the indicated treatments. (C) MDA levels measured in hPMECs under the indicated treatments. (D) GSH levels in hPMECs under the indicated treatments. (E) qRT-PCR analysis of COX-2 mRNA expression in hPMECs treated with LPS alone or combined with CD3D knockdown. (F) qRT-PCR analysis of iNOS mRNA expression in hPMECs under the same conditions. (G) Representative WB images showing COX-2 and iNOS protein levels in hPMECs stimulated with LPS with or without CD3D knockdown. GAPDH served as the loading control. (H) Quantification of COX-2 protein expression from WB analysis. (I) Quantification of iNOS protein expression from WB analysis. LDH, Lactate Dehydrogenase; SOD, Superoxide Dismutase; MDA, Malondialdehyde; GSH, Glutathione. Each assay was performed in triplicate to ensure accuracy and reproducibility. Statistical significance was determined by comparing the mean values of each group. *p 0.05.

Calcitonin, secreted by thyroid parafollicular C cells, is known to reduce serum calcium levels [22]. To explore its protective role, hPMECs were treated with calcitonin at concentrations of 1, 5, and 10 nM under LPS stimulation. Calcitonin alone did not significantly alter cell viability. However, LPS markedly reduced viability, which was restored by calcitonin in a dose-dependent manner (Fig. 7A). WB further revealed that calcitonin had no significant effect on CD3D protein expression in control cells; however, in LPS-stimulated hPMECs, calcitonin suppressed CD3D expression in a concentration-dependent manner (Fig. 7B,C). Flow cytometry analysis further showed that calcitonin attenuated LPS-induced apoptosis in a concentration-dependent manner (Fig. 7D,E). Consistently, ELISA results indicated that calcitonin suppressed LPS-induced secretion of TNF-$\alpha$, IL-6, and IL-1$\beta$, while not affecting their basal levels (Fig. 7F–H).

Fig. 7.

Effect of calcitonin on cell viability, apoptosis, and inflammatory cytokine production in LPS-treated hPMECs. (A) Cell viability was assessed using CCK-8 assay after 24 hours of stimulation with LPS and treatment with different concentrations of calcitonin (1, 5, 10 nM). The y-axis represents the cell viability, and the x-axis represents different treatment conditions. (B) WB analysis of CD3D protein levels in hPMECs after 24 h of LPS stimulation and treatment with different concentrations of calcitonin (1, 5, 10 nM). (C) Quantification of CD3D protein expression levels from (B), normalized to GAPDH. (D) Flow cytometry plots showing apoptosis in hPMECs after 24 hours of stimulation with LPS and treatment with different concentrations of calcitonin (1, 5, 10 nM). (E) Quantification of the apoptosis rate in hPMECs under the indicated treatment conditions. (F) ELISA analysis of TNF-$\alpha$ expression levels in hPMECs following 24 hours of stimulation with LPS and treatment with different concentrations of calcitonin (1, 5, 10 nM). (G) ELISA analysis of IL-6 expression levels in hPMECs under the same treatment conditions. (H) ELISA analysis of IL-1$\beta$ expression levels in hPMECs under the same treatment conditions. Each assay was performed in triplicate to ensure accuracy and reproducibility. Statistical significance was determined by comparing the mean values of each group. *p 0.05 vs. Control group, ^#p 0.05 vs. LPS group, ^#⁢#p 0.01 vs. LPS group. ns vs. Control group.

To evaluate the effects of calcitonin on oxidative stress, ELISA was used to measure SOD, MDA, and GSH levels. LPS stimulation significantly increased MDA, while decreasing SOD and GSH, indicating oxidative stress and injury. Co-treatment with calcitonin reversed these changes in a dose-dependent manner, reducing MDA levels and restoring SOD and GSH activity (Fig. 8A–C).

Fig. 8.

Calcitonin modulates oxidative stress markers in LPS-stimulated hPMECs. (A) SOD activity in hPMECs was measured using a commercial SOD assay kit after 24 h of stimulation with 10 µg/mL LPS, with or without calcitonin treatment at concentrations of 1, 5, or 10 nM. The y-axis represents SOD activity (U/mL). (B) MDA levels in hPMECs were determined using a lipid peroxidation assay kit under the same treatment conditions. The y-axis represents MDA content (nmol/mL). (C) GSH levels in hPMECs were quantified using a GSH detection kit under the same treatment conditions. The y-axis represents GSH concentration (µg/mL). Each assay was performed in triplicate to ensure accuracy and reproducibility. Statistical significance was determined by comparing the mean values of each group. *p 0.05 vs. Control group, ^#p 0.05 vs. LPS group. ns vs. Control group.

To further explore the relationship between calcitonin and inflammatory signaling, the mRNA levels of key pathway components were assessed in hPMECs stimulated with LPS, with or without increasing concentrations of calcitonin. qRT-PCR analysis revealed that calcitonin alone did not alter HMGB1, MyD88, NF-$\kappa$B, or NLRP3 expression. LPS stimulation markedly increased HMGB1, MyD88, and NLRP3, whereas co-treatment with calcitonin suppressed this upregulation in a dose-dependent manner. NF-$\kappa$B mRNA levels remained unchanged (Fig. 9A–D). WB analysis confirmed that calcitonin reduced LPS-induced expression of HMGB1, MyD88, NLRP3, and p-NF-$\kappa$B (Fig. 9E,F).

Fig. 9.

Calcitonin attenuates LPS-induced inflammatory response via the NF-$\kappa$B signaling pathway in hPMECs. (A) qRT-PCR analysis of HMGB1 mRNA expression in hPMECs stimulated with 10 µg/mL LPS for 24 h, with or without calcitonin treatment (1, 5, 10 nM). (B) qRT-PCR analysis of MyD88 mRNA expression under the same conditions. (C) qRT-PCR analysis of NF-$\kappa$B mRNA expression under the same conditions. (D) qRT-PCR analysis of NLRP3 mRNA expression under the same conditions. (E) Representative WB images showing protein levels of HMGB1, MyD88, p-NF-$\kappa$B, NF-$\kappa$B, and NLRP3 in hPMECs after LPS stimulation with or without calcitonin treatment. GAPDH served as the loading control. (F) Quantification of WB band intensities for HMGB1, MyD88, p-NF-$\kappa$B/NF-$\kappa$B, and NLRP3, normalized to GAPDH. *p 0.05 vs. Control group, ^#p 0.05 vs. LPS group. ns vs. Control group.

Among the investigated doses, 10 nM calcitonin had the most antioxidative impact and was thus chosen for further investigations. To determine the combined effect of CD3D knockdown and calcitonin, hPMECs were exposed to LPS in the presence or absence of si-CD3D-2 and/or 10 nM calcitonin. CCK-8 assays showed that both CD3D knockdown and calcitonin treatment significantly restored LPS-impaired cell viability, while their combined application further enhanced viability, suggesting a synergistic effect (Fig. 10A). Flow cytometry confirmed that apoptosis was most effectively reduced in the co-treatment group (Fig. 10B). Both qRT-PCR and WB confirmed that combined therapy enhanced Bcl-2 levels while more effectively downregulating Bax and Caspase-3 compared to single-agent treatments (Fig. 10C–G). These data suggest that CD3D depletion and calcitonin act in concert to suppress LPS-triggered apoptosis and enhance endothelial resilience.

Fig. 10.

Combined effect of calcitonin and CD3D silencing on LPS-induced viability and apoptosis in hPMECs. (A) Cell viability assessed by CCK-8 assay in hPMECs stimulated with LPS for 24 hours, co-treated with 10 nM calcitonin and subjected to CD3D knockdown. The x-axis represents different treatment conditions, and the y-axis represents cell viability. (B) Apoptosis was detected by flow cytometry in hPMECs stimulated with LPS for 24 hours, co-treated with 10 nM calcitonin, and subjected to CD3D knockdown. (C) qRT-PCR analysis of Bax mRNA levels in hPMECs stimulated with 10 µg/mL LPS for 24 hours, with or without CD3D knockdown and/or 10 nM calcitonin treatment. (D) qRT-PCR analysis of Bcl-2 mRNA expression in hPMECs under the same conditions. (E) qRT-PCR analysis of Caspase-3 mRNA expression in hPMECs under the same conditions. (F) Representative WB images showing protein expression levels of Bax, Bcl-2, and Caspase-3 in hPMECs under the same treatment conditions. GAPDH was used as an internal loading control. (G) Quantitative analysis of protein expression levels of Bax, Bcl-2, and Caspase-3 from WB results in (F), normalized to GAPDH. *p 0.05 vs. LPS group, ^#p 0.05 vs. LPS+10 nM Calcitonin group.

To further assess the regulatory effects of CD3D knockdown and calcitonin on inflammatory and oxidative responses in hPMECs under septic conditions, ELISA analysis was performed. CD3D knockdown substantially decreased LPS-induced TNF-$\alpha$, IL-6, and IL-1$\beta$ production, while co-treatment with calcitonin further enhanced this inhibitory effect (Fig. 11A–C). Similarly, oxidative stress analysis demonstrated that CD3D knockdown reduced MDA levels while restoring SOD and GSH, and co-treatment with calcitonin amplified these protective effects (Fig. 11D–F). Together, these results indicate that CD3D silencing, particularly in combination with calcitonin, effectively reduces LPS-induced inflammation and oxidative stress in hPMECs.

Fig. 11.

Effects of calcitonin and CD3D knockdown on inflammatory cytokine production and oxidative stress in LPS-treated hPMECs. (A) ELISA analysis of TNF-$\alpha$ levels (pg/mL) in hPMECs exposed to 10 µg/mL LPS for 24 hours, with or without 10 nM calcitonin treatment and/or CD3D knockdown. (B) ELISA analysis of IL-6 levels (pg/mL) under the same treatment conditions. (C) ELISA analysis of IL-1$\beta$ levels (pg/mL) under the same treatment conditions. (D) Measurement of SOD activity (U/mL) using a commercial detection kit in hPMECs exposed to LPS with or without calcitonin and/or CD3D knockdown. (E) Quantification of MDA content (nmol/mL) using a lipid peroxidation assay kit in the indicated groups. (F) Determination of GSH levels (µg/mL) in hPMECs treated with LPS, calcitonin, and/or CD3D knockdown. Experimental groups included LPS+si-NC, LPS+si-CD3D-2, LPS+10 nM calcitonin, and combined treatment with LPS+si-CD3D-2+10 Nm calcitonin. *p 0.05 vs. LPS group, ^#p 0.05 vs. LPS+10 nM Calcitonin group.

To elucidate the underlying mechanisms, qRT-PCR analysis was performed. CD3D knockdown reduced LPS-induced expression of HMGB1, MyD88, and NLRP3, and co-treatment with calcitonin further enhanced this suppression, while NF-$\kappa$B mRNA levels remained largely unaffected (Fig. 12A–D). WB analysis confirmed these results, showing that combined treatment with calcitonin and CD3D knockdown more effectively suppressed MyD88, HMGB1, NLRP3, and p-NF-$\kappa$B compared with either therapy alone (Fig. 12E,F). These findings suggest that CD3D silencing and calcitonin synergistically protect against LPS-induced endothelial injury by inhibiting the HMGB1/MyD88/NF-$\kappa$B pathway.

Fig. 12.

CD3D knockdown combined with calcitonin suppresses LPS-induced activation of the HMGB1/MyD88/NF-$\kappa$B/NLRP3 signaling pathway in hPMECs. (A) HMGB1 mRNA expression in hPMECs measured by qRT-PCR after 24 h stimulation with 10 µg/mL LPS, with or without 10 nM calcitonin treatment and/or CD3D knockdown. (B) MyD88 mRNA expression was detected under the same treatment conditions. (C) NF-$\kappa$B mRNA expression was analyzed by qRT-PCR in the indicated groups. (D) NLRP3 mRNA expression was measured under identical conditions. (E) Representative WB images of HMGB1, MyD88, NF-$\kappa$B, p-NF-$\kappa$B, and NLRP3 in hPMECs under the same treatments. GAPDH served as a loading control. (F) Quantification of WB results showing relative protein expression of HMGB1, MyD88, NF-$\kappa$B/p-NF-$\kappa$B, and NLRP3. *p 0.05 vs. LPS group, ^#p 0.05 vs. LPS+10 nM Calcitonin group. ns vs. LPS group or LPS+10 nM calcitonin group.