The human normal lung epithelial cell line BEAS-2B, and the human non-small cell lung cancer cell lines A549 and HCI-H1975 used in this study were all purchased from The Cell Bank of Type Culture Collection of The Chinese Academy of Sciences. The cells were cultured in DMEM (Wuhan Servicebio Technology Co., Ltd.) supplemented with 10% fetal bovine serum (FBS; Nanjing BioChannel Biotechnology Co., Ltd.) and 1% penicillin-streptomycin (Wuhan Servicebio Technology Co., Ltd.), at 5% CO₂ and 37°C. Small interfering (si)RNA-TBX20 (si-TBX20) and siRNA-negative control (NC) were transfected into A549 cells using Lipofectamine^® 3000 transfection reagent (Invitrogen; Thermo Fisher Scientific, Inc.). TBX20 overexpression plasmid (oe-TBX20) was also transfected into A549 cells using pRP[Exp]-EGFP/Puro-CAG>hTBX20 [Yunzhou Biosciences (Guangzhou) Co., Ltd.); at the same time, the empty vector pRP[Exp]-EGFP/Puro-CAG (without the TBX20 insertion fragment) was used as the negative control. A549 cells were added to six-well plates, with 2×10⁶ cells per well, and 2 µl DMEM medium was added. A total of 5 µg of DNA or siRNA, was added at 25°C for 6 h. Culture medium was removed, and 10% serum-containing DMEM medium was added, and the cells were further cultured at 37°C for 24 h. Finally, the transfection efficiency was observed under a fluorescence microscope. The sequences of siRNA-TBX20, siRNA-NC and oe-TBX20 were designed and synthesized by Sangon Biotech (Shanghai) Co., Ltd. The siRNA sequences used are as follows: siRNA-TBX20, forward 5′-CCGAGAUGAUCAUCACCAAGU-3′, reverse 5′-UUGGUGAUGAUCAUCUCGGUG-3′; siRNA-NC, forward 5′-UUCUCCGAACGUGUCACGUUU-3′ and reverse 5′-ACGUGACACGUUCGGAGAAUU-3′.

Total cellular protein was extracted from A549 and HCI-H1975 using a protein extraction kit (cat. no. BC3710; Beijing Solarbio Science & Technology Co., Ltd.) according to the manufacturer's protocol. After determining the protein concentration using BCA Protein Quantification kit, loading buffer was added and the samples were incubated at 95°C for denaturation. SDS-PAGE was performed on 12% gels with 30 µg protein/lane. Proteins were transferred to a PVDF membrane, which was blocked with 5% skim milk solution for 2 h at 25°C. The PVDF membrane was incubated with diluted primary antibodies at 4°C for 12 h, washed and incubated with diluted secondary antibodies on ice in the dark for 2 h. An ECL reagent (cat. no. MA0186; Dalian Meilun Biology Technology Co., Ltd.) was used for visualization and a gel imaging system (SH-523; Hangzhou Shenhua Technology Co., Ltd.) was used for development. ImageJ (National Institutes of Health, V1.8.0.112) was used to calculate the gray values of proteins. The following antibodies were used in the present study: TBX20 (cat. no. 83414-5-RR; 1:1,000; Wuhan Sanying Biotechnology), E-cadherin (cat. no. 60335-1-Ig; 1:1,000; Wuhan Sanying Biotechnology), N-cadherin (cat. no. 66219-1-Ig; 1:2,000; Wuhan Sanying Biotechnology), vimentin (cat. no. A19607; 1:1,000; ABclonal Biotech Co., Ltd.), CD44 (cat. no. 84854-5-RR; 1:500; Wuhan Sanying Biotechnology), Sox2 (cat. no. 66411-1-Ig; 1:500; Wuhan Sanying Biotechnology), Nanog (cat. no. 14295-1-AP; 1:500; Wuhan Sanying Biotechnology), PD-L1 (cat. no. 66248-1-Ig; 1:100; Wuhan Sanying Biotechnology), IL-6 (cat. no. A0286; 1:200; ABclonal Biotech Co., Ltd.), CCL2 (cat. no. MA5-17040; 1:500; Thermo Fisher Scientific, Inc.) and β-actin (cat. no. 81115-1-RR; 1:2,000; Wuhan Sanying Biotechnology) primary antibodies, and HRP-conjugated rabbit anti-mouse and goat anti-rabbit secondary antibodies (cat. nos. ab6728 and ab6721; 1:5,000; Abcam).

Total RNA was extracted from A549 and HCI-H1975 cells using a TRIzol kit (Invitrogen), and the RNA concentration and purity were measured with a NanoDrop microplate spectrophotometer (Thermo Fisher Scientific, Inc.). cDNA was subsequently synthesized by RT using a PrimeScript RT-PCR kit (Takara Bio, Inc.), performed according to manufacturer's protocol. Finally, qPCR was performed on cDNA using the eQ9600 real-time fluorescent qPCR detection system (Suzhou Dongsheng Xingye Scientific Instruments Co., Ltd.), Thermocycling conditions were as follows: Pre-denaturation at 95°C for 3 min; Denaturation at 95°C for 5 sec, annealing at 60°C for 30 sec, extension at 72°C for 20 sec, a total of 40 cycles. The mRNA expression levels of TBX20, vimentin, N-cadherin, E-cadherin, CD44, Sox2, Nanog, PD-L1, IL-6 and CCL2 were analyzed using the 2^−ΔΔCq method (15), with β-actin as the internal reference gene. The primer sequences are shown in Table I.

After 48 h of transfection, A549 cells from each group were seeded at 2,000 cells/well in a 96-well plate. After 24 h of incubation, 20 µl CCK-8 reagent (GlpBio, GK10001) was added per well, followed by a 2-h, 37°C incubation. The absorbance was measured at 450 nm using a microplate reader to assess cell viability, as follows: Cell viability (%)=(OD_{experimental group}/OD_{positive drug group}/OD_{control group}) ×100.

After the A549 and HCI-H1975 cells were resuspended in serum-free DMEM, 1×10^4 cells were inoculated into each well of a 24-well low-adhesion plate and total volume is 500 µl. A 200-µl cell suspension was added to the upper chamber of the Transwell system. The lower chamber was supplemented with 500 µl DMEM containing 20% fetal bovine serum. After incubation at 37°C for 24 h, the supernatant and cell debris were removed. At 25°C, the cells were fixed with 100% methanol for 30 min, stained with 0.1% crystal violet for 20 min and washed with PBS. Finally, three randomly selected fields were observed under a ×200 inverted light microscope to assess cell migration. For the invasion assay, the same protocol was performed; however, the upper chamber of the Transwell system was coated with 8% Matrigel at 37°C for 3 h before cell seeding.

A549 cells were seeded at 1×10⁴ cells/500 µl volume per well into low-adhesion 24-well plates. DMEM/F12 (Wuhan Servicebio Technology Co., Ltd, containing 20 ng/ml epidermal growth factor, 10 ng/ml basic fibroblast growth factor and 10 µg/ml B27) was added and the cells were incubated at 37°C for 24, 48 and 72 h. Images were captured using an light microscope at 72 h post-seeding, after which the spheroids were counted.

A total of 12 male BALB/c nude mice (age, 5–6 weeks; body weight, ~16 g) were provided by Hunan Sileke Jingda Experimental Animal Co., Ltd. [license no. SCXK (Xiang) 2025-0004] and the experiment was approved by the Animal Ethics Committee of Fuzhou University Affiliated Provincial Hospital [approval no. IACUC-FPH-SL-20250527[1092]). Human peripheral blood mononuclear cells (PBMCs) were purchased from the American Type Culture Collection (cat. no. PCS-800-011). The use of commercially available anonymous primary cells was exempted for ethics review by the Ethics Committee of Fuzhou University Affiliated Provincial Hospital. The mice were randomly divided into four groups: A549, A549 + sh-NC, A549 + short hairpin (sh)RNA-TBX20 (sh-TBX20; A549 cell line stably transduced with sh-TBX20 lentiviral plasmid) and A549 + oe-TBX20 (A549 cell line stably transfected with pcDNA3.1-oe-TBX20). Each group contained 3 mice.

Using pcDNA3.1 as the vector, restriction endonucleases NheI and XbaI were used as the cutting sites. Mutant primers were designed and synthesized by Shanghai Sangon Biotech Co., Ltd., and the target gene was amplified by PCR to obtain the mutant TBX20 cDNA sequence. The forward primer sequence was 5′-GCTAGCATGGCGGCGGCGGCGGCGGCGGC-3′ (containing NheI site) and the reverse primer sequence was 5′-TCTAGATCAGTCATCATCATCGTCATCAT-3′ (containing XbaI sit). For the T262M point mutation, the mutagenic primer sequences were as follows: Forward 5′-GCTAGCATGGCGGCGGCGGCGGCGGCGGC-3′ and reverse 5′-TCTAGATCAGTCATCATCATCGTCATCAT-3′, with the mutation introduced by site-directed mutagenesis using PrimeSTAR high-fidelity DNA polymerase (Takara Bio, Inc.; cat. no. R045A). PCR amplification was performed with the following conditions: Initial denaturation at 98°C for 2 min; 35 cycles of 98°C for 10 sec, 60°C for 15 sec, and 72°C for 2 min; final extension at 72°C for 5 min. The target gene was recombined and transformed with pcDNA3.1, and the obtained sequence was confirmed by Sanger sequencing (Shanghai Sangon Biotech Co., Ltd.) to ensure its accuracy. Wild-type TBX20-pcDNA3.1 and mutant TBX20-T262M-pcDNA3.1 overexpression plasmids were constructed by Shanghai Sangon Biotech Co., Ltd.

The second-generation lentiviral packaging system was used for lentivirus production. 293T cells (human embryonic kidney cells; American Type Culture Collection, Cat. No. CRL-3216) were seeded at 4×10⁶ cells per 10-cm dish 24 h prior to transfection. At 70–80% confluence, cells were co-transfected with 7.5 µg shRNA-TBX20 plasmid (forward, 5′-CCGGGCAGAAGATCAAGATCAATTCTCGAGAATTGATCTTGATCTTCTGCTTTTTG-3′ and reverse, 5′-AATTCAAAAAGCAGAAGATCAAGATCAATTCTCGAGAATTGATCTTGATCTTCTGC-3′), plasmid backbone (pL KO.1-puro Lentiviral shRNA expression vector (purinomycin resistance); Sigma-Aldrich, Cat. No. SHC001), 5.6 µg psPAX2 packaging plasmid (Addgene, 12260), and 1.9 µg of pMD2.G envelope plasmid (Addgene, Inc.; cat. no. 12259) using Lipofectamine 3000 (Thermo Fisher Scientific, Inc., L3000015) according to the manufacturer's instructions. The plasmid ratio was 4:3:1 (shRNA:psPAX2:pMD2.G). Transfection was performed in Opti-MEM reduced serum medium at 37°C, 5% CO₂ for 12 h, after which the medium was replaced with fresh DMEM containing 10% FBS. At 48 and 72 h after transfection, supernatants and centrifuge at 500 × g for 5 min at 4°C. Then filter through a 0.45 µm PVDF membrane. After aliquoting, store at −80°C. Viral titer was determined by RT-qPCR.

For lentiviral transduction, A549 cells were seeded at 2×10⁵ cells per well in 6-well plates. When cell density reached 50% confluence, lentivirus at MOI of 8 were added. The cells were gently mixed and cultured at 37°C for 24 h. Cells were selected using puromycin (2 µg/ml). Continuous screening was conducted for 5 days. Cells were cultured for 24 h before subsequent experiments. Infection efficiency of the lentivirus was verified by RT-qPCR and Western blot. Culture medium containing puromycin (1 µg/ml) was used to maintain the cells.

Stable overexpression and interference cell lines (oe-TBX20 and shTBX20) of the TBX20 gene were constructed. A549 cells that stably expressed oe-TBX20 and shTBX20 were resuspended in PBS (2×10⁷ cells/ml) and gently mixed with an equal volume of Matrigel to obtain a cell-scaffold mixture with a final concentration of 1.0×10⁷ cells/ml. Subsequently, 100 µl mixture was injected subcutaneously into the right axilla of each nude mouse. PBMCs were expanded by culturing in a complete medium containing 100 ng/ml IL-2. The cells were resuspended in sterile PBS to a concentration of 1.0×10⁸ cells/ml and stored on ice for later use. The PBMCs were resuspended in PBS at a concentration of 1.0×10⁸ cells/ml, and 100 µl PBMC suspension was injected subcutaneously at the edge of the tumor nodule (not directly injected into the tumor center). At 30 days after the injection of tumor cells, the mice were euthanized by intraperitoneal injection with 1% pentobarbital sodium (200 mg/kg) and decapitation, and the tumors were removed and weighed. Humane endpoints were as follows: Maximum tumor diameter exceeded 20 mm; maximum tumor volume exceeded 2,000 mm³; obvious signs of cachexia. No mice were sacrificed prematurely.

Tumor tissues were fixed with 4% paraformaldehyde solution for 24 h at 4°C. The tissue blocks were sectioned into 4 µm slices, baked at 60°C for 30 min. The sections underwent routine xylene dewaxing, followed by alcohol hydration with a gradient, and three washes with distilled water for 3 min each. Subsequently, they were rinsed three times with PBS for 3 min each. The antigen retrieval solution containing EDTA, pH 9.0, Wuhan Servicebio Technology Co., Ltd.) was boiled in a high-pressure cooker for 2 min. The sections were then placed in an incubation chamber, and 3% hydrogen peroxide solution was added. After incubation at 25°C for 12 min, the sections were rinsed three times with PBS solution for 3 min each. Then, 5% goat serum (Nanjing SenBeiJia Biological Technology Co., Ltd.) was added, and the sections were incubated at 25°C for 20 min. TBX20 primary antibody (Thermo Fisher Scientific, Inc., cat. no. PA5-110464; 1:500) was added overnight at 4°C. Sections were rinsed three times with PBS buffer for 2 min each to remove excess primary antibody. The secondary antibody (HRP-labeled rabbit anti-IgG, 1:500; cat. no. TBAG0030, Wuhan Servicebio Technology Co., Ltd.) was added, and the sections were incubated at 25°C for 30 min. After rinsing three times with PBS buffer for 2 min each, the prepared DAB chromogenic solution (Wuhan Servicebio Technology Co., Ltd.) was added. The staining was observed under an light microscope. Sections were stained with hematoxylin (Beijing Biosharp Technology Co., Ltd) for 5 min, followed by rinsing with distilled water. Staining was evaluated independently by two pathologists who were blinded to the clinical data. Positive staining was defined as brown nuclear staining. Images of five representative high-power fields (magnification, ×400) per tumor were photographed. The percentage of positive tumor cells was recorded.

Flow cytometry was used to detect immune cells. Fresh tumor tissues were placed in cold PBS and minced with ophthalmic scissors. The tissue was then transferred to a 15-ml centrifuge tube and digested with 5 ml digestion buffer (RPMI-1640 + 2 mg/ml collagenase IV + 20 µg/ml DNase I) in a 37°C water bath with oscillation. After the digestion was complete, 5 ml RPMI-1640 containing 10% FBS was added to terminate the reaction. The suspension was filtered through a 70-µm cell sieve to collect the single-cell suspension and was then centrifuged at 400 × g for 5 min at 4°C, after which the supernatant was discarded, and the cells were resuspended in PBS. Next, 1 ml ACK lysis buffer was added, the mixture was incubated for 2 min and the reaction was terminated with PBS. The cell suspension was adjusted to 1×10⁶ cells/100 µl, and was transferred to a flow cytometry tube. Fc blocking agent (Rat Anti-Mouse CD16/CD32, BD Biosciences, cat. No. 553142) was added, and the sample was incubated at 4°C for 10 min. A surface antibody mixture (CD3, CD4, CD8, Foxp3, CD86 and CD206) was then added and incubated at 4°C in the dark for 30 min. The labelled antibodies for CD4+ T cells are: CD3 (FITC anti-human CD3, BioLegend, Inc., cat. No. 317306), CD4 (PE-anti-human CD4, BioLegend, Inc., Cat. No. 300508). The labelled antibodies for CD8+ T cells are: CD3, CD8 (PE anti-human CD8 Antibody, BioLegend, Inc., Cat. No. 344705). The labelled antibodies for Treg cells are: CD3, CD4, Foxp3 (APC Anti-FOXP3Antibody[3G3], Abcam, Cat. No. ab200568). The labelled antibody for M1 cells is: CD86 (FITC anti-human CD86 Antibody, BioLegend, Inc., Cat. No. 374203). The labelled antibody for M1 cells is: CD206 (FITC anti-human CD206 (MMR) Antibody, BioLegend, Inc., Cat. No. 321103). The suspension was subsequently washed twice with PBS (4°C, 300 × g, 5 min each), the supernatant was discarded, and a FoxP3 fixation/permeabilization kit (eBioscience™ Foxp3/Transcription Factor Staining Buffer Set; Cat. no. 00-5523-00, Thermo Fisher Scientific, Inc.) was used. Briefly, 100 µl fixation/membrane disruption solution was added and the sample was incubated at 25°C in the dark for 30–60 min. The sample was then washed once with 1X membrane disruption buffer and then centrifuged (4°C and 400 × g for 5 min) to remove the supernatant. The sample was then resuspended in 50 µl membrane disruption buffer, PE-labelled anti-FoxP3 antibody was added and the sample was incubated at 4°C in the dark for 30 min. The cells were then washed once with membrane disruption buffer, resuspended in 200–300 µl PBS and prepared for flow cytometry. A BD LSRFortessa flow cytometer (BD Biosciences) with appropriate laser and filter settings was used. At least 50,000 CD45⁺ leukocyte events per sample were collected, isotype control (Mouse IgG1, κ isotype Ctrl antibody, BioLegend, Inc., Cat. no. 400105) and single-stain compensation tubes were used for normalization, and the data were analyzed using FlowJo v10 software (BD Biosciences).

Statistical analysis was performed using SPSS26.0 (IBM Corp.) and GraphPad Prism 8.0.2 (Dotmatics), with all experiments repeated three times. Quantitative data (conforming to normality and variance homogeneity) are presented as the mean ± standard deviation. One-way ANOVA was used for inter-group comparisons, after which, the SNK post hoc test was used to analyze datasets containing ≤3 groups, whereas multiple comparisons between >3 groups were analyzed using the Tukey post hoc test. Independent samples t-test was employed for comparisons between two groups. P<0.05 was considered to indicate a statistically significant difference.

The expression levels of TBX20 in lung cancer cells were assessed by western blotting, which revealed that TBX20 protein expression was significantly higher in in A549 and NCI-H1975 cancer cells compared with that in BEAS-2B cells (P<0.05; Fig. 1A). qPCR further confirmed the statistically significant increases in TBX20 mRNA expression levels in these cell lines (P<0.05; Fig. 1B). As the expression of TBX20 mRNA was highest in A549, A549 cells were used for subsequent experiments.

The effects of TBX20 overexpression/silencing were subsequently investigated in lung cancer cells. The successful transfection of si-TBX20 and oe-TBX20 is shown in Fig. 2A and B. Compared with the BEAS-2B group, the A549 group exhibited significantly greater cell viability (Fig. 2C), migration (Fig. 2D and G), invasion (Fig. 2E and G) and spheroid formation capacity (Fig. 2F and G) (P<0.05), whereas no significant differences were observed between the A549 and A549-siNC groups (P>0.05). Following TBX20 overexpression, compared with in the A549-NC group, cell viability, migration and invasion, and spheroid formation capacity were significantly increased (P<0.05). By contrast, after TBX20 silencing, cell viability, migration and invasion and spheroid formation capacity were significantly reduced compared with in the A549-siNC group (P<0.05).

To evaluate the effects of TBX20 overexpression/silencing on EMT, stemness and immune activity in lung cancer cells, western blotting and RT-qPCR were used to measure the expression levels of EMT markers (E-cadherin, N-cadherin and vimentin) (16), stemness-related markers (CD44 and Nanog) (17) and immune-related cytokines (PD-L1, IL-6 and CCL2) (18). The results of western blotting (Fig. 3A) revealed that, compared with in the BEAS-2B group, the A549 group exhibited significantly reduced E-cadherin protein expression; higher N-cadherin and vimentin levels; higher CD44 and Nanog expression; and higher expression levels of PD-L1, IL-6 and CCL2 (P<0.05; Fig. 3B-D and F-J). Compared with the A549 + NC group, the expression of Sox2 in the A549 + oe-TBX20 group showed no significant change (Fig. 3E). However, no significant differences were observed between the A549 and A549-NC groups (P>0.05). After TBX20 overexpression, the expression levels of E-cadherin were significantly lower, whereas those of N-cadherin, vimentin, CD44, Nanog, PD-L1 and IL-6 levels were higher than those in the A549-NC group (P<0.05). Conversely, compared with in the A549-NC group, TBX20 silencing led to increased E-cadherin expression, but significantly decreased N-cadherin, vimentin, CD44, PD-L1 and IL-6 levels (P<0.05).

The RT-qPCR results revealed that compared with in the BEAS-2B group, the A549 group exhibited significantly lower E-cadherin mRNA expression, and significantly higher mRNA expression levels of N-cadherin, vimentin, Sox2, CD44, Nanog, PD-L1, IL-6 and CCL2 (P<0.05; Fig. 4). However, no significant differences were observed between the A549 and A549-NC groups (P>0.05). After TBX20 was overexpressed, compared with the A549-NC group, E-cadherin expression was significantly decreased, whereas the levels of vimentin, Sox2, CD44, Nanog, PD-L1, IL-6 and CCL2 were significantly increased (P<0.05). Conversely, after TBX20 silencing, compared with in the A549-NC group, E-cadherin expression was significantly increased, and the levels of N-cadherin, vimentin, Sox2, CD44, Nanog, PD-L1, IL-6 and CCL2 were decreased (P<0.05).

In vivo experiments were conducted to evaluate the effects of TBX20 overexpression and silencing on tumor size. The expression levels of TBX20 in the A549 + oe-TBX20 group were significantly greater than those in the A549 + NC group (P<0.05), whereas the A549 + sh-TBX20 group exhibited significantly lower expression (P<0.05), confirming successful transfection (Fig. 5A). At 4 weeks after subcutaneous tumor transplantation, tumor volume measurements revealed that the tumors in the A549 + oe-TBX20 group were significantly larger than those in the A549-NC group (P<0.05); the maximum diameter and tumor volume in the A549 + oe-TBX20 group were 20 mm and 1,273 mm³, respectively, whereas those in the A549 + sh-TBX20 group were significantly smaller (P<0.05; Fig. 5B and C).

To assess the effects of TBX20 on tumor EMT and stemness markers in nude mice, IHC was performed (Fig. 6A) to detect the expression levels of E-cadherin, N-cadherin and vimentin (EMT markers), CD44 (cellular stemness marker) and PD-L1 (tumor immune escape indicator) in the subcutaneous tumor tissues of nude mice. Compared with in the A549 group, the A549 + oe-TBX20 group exhibited significantly lower E-cadherin levels (Fig. 6B), whereas N-cadherin (Fig. 6C), vimentin (Fig. 6D), CD44 (Fig. 6E) and PD-L1 levels (Fig. 6F) were significantly greater (P<0.05). Conversely, compared with in the control group, the A549 + sh-TBX20 group demonstrated significantly higher E-cadherin levels, with significant reductions in N-cadherin, vimentin, CD44 and PD-L1 levels (P<0.05).

To assess the effects of TBX20 on immune cell infiltration in nude mice, flow cytometry was performed to detect the infiltration of CD8⁺ T cells and CD4⁺ T cells in subcutaneous tumor tissues, along with the proportions of M1 and M2 macrophages, and regulatory T cells (Tregs). The inoculation of human PBMCs into nude mice (lacking functional T cells) provides a reconstructed host environment capable of eliciting an immune response for human-derived tumors. CD8⁺ T cells and CD4⁺ T cells are the core components of PBMCs, whereas Tregs are also a functional subgroup present in PBMCs. Compared with in the A549 + sh-TBX20 group, the A549 + oe-TBX20 group exhibited significantly lower infiltration levels of CD4⁺ T cells, CD8⁺ T cells and M1 cells, but significantly greater infiltration levels of M2 cells and Tregs (P<0.05; Fig. 7). Conversely, compared with in the A549 + NC group, the A549 + sh-TBX20 group exhibited significantly higher infiltration levels of CD4⁺ T cells, CD8⁺ T cells, and M1 macrophages, but significantly lower infiltration levels of M2 macrophages and Tregs (P<0.05).