THOC3 interacts with epithelial-to-mesenchymal transition to promote non-small cell lung cancer carcinoma progression through STAT3 signaling pathway.

Introduction
Lung cancer (LC) has remained the leading cause of cancer-related mortality worldwide since the mid-1900s, posing an ongoing public health challenge due to persistently high incidence and mortality rates [1]. Approximately 85% of LC cases are classified as non-small cell lung carcinoma (NSCLC), a subtype historically linked to poor survival outcomes, primarily due to late-stage diagnosis and limited therapeutic options [2]. Tumor progression and metastasis are significantly influenced by epithelial-to-mesenchymal transition (EMT), a cellular process associated with worse prognosis. Extensive evidence has shown that EMT not only enhances cancer cell motility and metastatic potential but also improves cellular survival, highlighting the need for targeting EMT pathways to enhance patient outcomes [[3], [4], [5], [6]].
Recent studies have highlighted the crucial role of mRNA nuclear export in tumor initiation and malignant progression across various cancer types [7]. As a key regulatory mechanism for eukaryotic cell growth, mRNA export significantly affects gene expression [8]. Integral to this process is the transcription-export (TREX) complex, within which the THO complex (THOC) plays a crucial role in mRNA splicing, transcript elongation of during transcription, and nuclear export of newly synthesized RNA molecules. In mammals, the THOC machinery consists of six specific protein: THOC1, THOC2, THOC3, THOC5, THOC6, and THOC7.
Dysregulation of THOC proteins disrupts the quality control of nuclear mRNA export, compromising transcriptional accuracy. Genetic loss of THOC family components leads to transcriptional instability and increased genomic aberrations. Growing evidence suggests that several THOC members significantly contribute to oncogenic progression in a variety of cancers [9,10]. For example, overexpression of THOC1 has been linked to increased proliferation and aggressive behavior in breast, prostate and lung cancers [11,12]. THOC2 and THOC5 promote cancer stemness and resistance to radiotherapy in triple-negative breast cancer (TNBC) [13].
However, the precise role of THOC3 in cancer remains poorly understood. While THOC3 is known to be crucial for early cellular differentiation [14], and to regulate multiple transcriptional factors in fish, its role in human cancers is less clear. Elevated THOC3 expression has been associated with poor clinical outcomes in LC [15]. In this study, we observed increased THOC3 expression in NSCLC and cell models, and found that its knockdown suppressed NSCLC cell proliferation. These data indicate that THOC3 contributes to the progression of NSCLC. Garcinone D is provided by Selleck Biotech (China). For more details, visit: https://www.selleck.cn/products/garcinone-d-stat-activator/. Garcinone D has been described as a STAT3 activator in previous literature [16]. Additionally, the role of Garcinone D as a STAT3 agonist has been supported by a recent study [17]. We will clarify its specificity in the manuscript and activate STAT3.
In our current investigation, we reported that overexpression of THOC3 in LC tissues and cell models correlated with advanced clinical stages and poor prognosis. Through loss- and gain-of-function studies, we demonstrated that THOC3 promoted to enhance tumor growth, facilitated cell proliferation, and inhibited apoptosis. On the molecular level, THOC3 exerted its tumor-promoting effects by activating the STAT3 signaling pathway, which in turn drove aggressive tumor phenotypes in LC.
STAT3 is a key regulator upstream of EMT, enhancing metastatic spread in various cancers, including gastrointestinal, brain, and thoracic tumors. Targeted inhibition of STAT3 has been shown to reduce metastatic dissemination and improve clinical outcomes. Collectively, our findings provide a detailed molecular mechanism by which THOC3 promotes LC progression and highlight its potential as a therapeutic target for LC treatment.

Methods

Cell lines
The human normal lung epithelial cell line BEAS-2B was obtained from the American Type Culture Collection. Four human-derived non-small cell lung cancer cell lines (H1975, A549, H1299 and PC9) were acquired from the China Academia Sinica Cell Repository (Shanghai, China). All cell lines were routinely verified in December 2022 through assays assessing cell viability, isozyme profiling, DNA fingerprinting, and mycoplasma contamination testing. Cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS) at 37 °C in 5% CO2.

Production of lentiviruses and cell transfection
A lentiviral construct (H145 pLenti-EF1a-EGFP-F2A-Puro-CMV-MCS) encoding the full human THOC3 sequence was generated. The empty plasmid vector (pcDNA 3.1) was served as the negative control (NC). To stably knockdown THOC3, two short hairpin RNA (shRNA) constructs targeting human THOC3 (designated shTHOC3#1 and shTHOC3#2) were inserted into lentiviral expression plasmids (pLKD-CMV-G&PR-U6-shRNA). Human NSCLC cell lines H1299 and A549 were seeded in 6-well plates and cultured until cell confluence reached approximately 30–60%. Lentiviral vectors containing either THOC3-targeting shRNAs or NC shRNA (purchased from Genechem Co., Ltd., Shanghai, China) were then transduced into these cells.

Tissue samples
Paired lung carcinoma and adjacent non-tumor tissues samples were obtained from patients undergoing surgery at the Oncology Department, First Affiliated Hospital of Wannan Medical College. All tumor specimens underwent independent pathological evaluation by two experienced pathologists based on diagnostic criteria. Tissue specimens were immediately frozen in liquid nitrogen upon collection and stored at −80 °C until further analysis.

Analysis of differential expression
Total RNA was extracted from five paired lung carcinoma and adjacent non-tumorous tissue specimens was using the TRIzol extraction method (Thermo Fisher Scientific, USA). RNA quality and purity assessments were assessed using the Agilent Bioanalyzer 2100 (Agilent Technologies, USA). Purified RNA samples were subsequently hybridized onto Affymetrix Human Gene Expression Arrays, and scanning was performed according to the manufacturer’s protocols. Public databases, including The Cancer Genome Atlas (TCGA), Chinese Lung Cancer Genome Atlas (CGGA), and Gene Expression Omnibus (GEO), were integrated with R-based bioinformatics tools to identify differentially expressed genes (DEGs).

RT-qPCR
Following transfection, total RNA was extracted using TRIzol reagent (Invitrogen, Canada). The NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA) was utilized to ascertain RNA concentration and purity. Subsequently, 0.4 µg of the purified RNA per specimen underwent reverse transcription into complementary DNA (cDNA) employing a first-strand synthesis kit (Thermo Fisher Scientific). Quantitative PCR was performed on the synthesized cDNA using a CFX-96 PCR detection system (Bio-Rad, USA). Gene expression were normalized to GAPDH, and relative expression levels were calculated using the 2⁻ΔΔCt method.

IHC and tissue microarray analyses
In this study, a total of 152 paraffin-embedded lung carcinoma specimens were selected for IHC analysis. The slides were independently reviewed by two experienced pathologists, who were blinded to the clinical data of the patients. Tissue sections were initially dewaxed in xylene and then rehydrated through a graded series of ethanol solutions. Antigen retrieval was performed using microwave irradiation in EDTA buffer (pH 8.0). Endogenous peroxidase activity was blocked by incubation with 3% hydrogen peroxide in methanol, followed by incubation with 1% goat serum albumin to prevent non-specific antibody binding. The sections were then incubated overnight at 4 °C with primary antibodies. After thorough washing, sections were treated with secondary antibodies for 20 min, and antigen detection was visualized using 3,3′-diaminobenzidine (DAB) chromogen.
Tissue microarrays (TMAs) were constructed from the same 152 paraffin-embedded lung carcinoma specimens, following established protocols. Using a tissue array device (Beecher Instruments, USA), cylindrical cores (0.6 mm in diameter) were extracted from donor blocks and inserted into recipient blocks, which were subsequently sectioned into 5 µm-thick slices. Immunostaining was evaluated based on staining intensity: absent (0), mild (1), moderate (2), or intense (3), and the percentage of positively stained tumor cells was categorized as follows: 0 (no positive cells), 1 (up to 5%), 2 (5–25%), 3 (26–50%), 4 (51–75%), and 5 (above 75%). The final histochemical score (H-score) was calculated by multiplying intensity scores by the proportion of positive cells. Expression levels of THOC3 were classified as negative (H-score: 0–4), low (H-score: 5–6), or high (H-score: 8–12).

WB
For WB analysis, protein-transferred membranes were blocked with 5% non-fat milk for 1 hour, followed by overnight incubation at 4 °C with primary antibodies (1:1000 dilution), including anti-THOC3 (Cat#PA5–112,761), anti-N-cadherin (Cat#13,116), anti-E-cadherin (Cat#14,472), anti-Vimentin (Cat#5741), anti-Bcl-2 (Cat#15,071), anti-BAX (Cat#5023), anti-Bcl-xl (Cat#2764), anti-p-STAT3 (Cat#4176), anti-STAT3 (Cat#8480), and anti-β-actin (Cat#8198). After thorough washing, membranes were probed with corresponding secondary antibodies. Immunoreactive protein bands were visualized using the Tanon 5200 imaging system (Tanon, China). Following protein visualization, membranes were stripped and re-probed for further assessments. Band intensities were quantified using Quantity One software, with expression levels normalized to the internal control β-actin. Each experiment was repeated independently at least twice.

CCK-8 assay
Cells were seeded onto 96-well plates and cultured overnight prior to exposure to the CCK-8 reagent. Following incubation within a humidified environment at 37 °C containing 5% CO2, absorbance measurements at a wavelength of 450 nm were captured using a microplate reader (Infinite M2009PR, Switzerland). Each experiment was independently repeated three times.

Colony formation assay
Cells were seeded into 6-well plates containing DMEM supplemented with 10% FBS and cultured for two weeks to allow colony formation. Colonies were fixed with methanol, stained with crystal violet, and rinsed twice with PBS. The number of colonies per well was counted under a microscope (Nikon Corporation, Japan).

Cell apoptosis assay
Apoptosis was assessed using an Apoptosis Detection Kit (Invitrogen, Canada). Cells were harvested, washed twice with ice-cold PBS, and centrifuged (1000 rpm, 7 min). Following another wash with d-Hanks buffer (pH 7.3), cells were resuspended in 1 × binding solution containing apoptosis detection reagents and incubated in the dark at room temperature for 15 min. The proportion of apoptotic cells was determined by flow cytometry using a FACScan instrument (Beckman Coulter, USA).

Cell migration assays
Cell migration was assessed using scratch assays and transwell migration assays. For scratch assays, cells were seeded into 6-well plates (4 × 10^5 cells per well) and allowed to reach confluence. A linear wound was made by scraping the cell monolayer with a sterile 10-µl pipette tip. Microscopic images of randomly selected areas were taken every 48 h. Cellular migration was quantified as follows: migration (%) = (remaining average scratch width / initial scratch width) × 100%. In transwell assays, cells (5 × 10^4) were seeded into the upper compartments of polycarbonate membranes (8 µm pore size, Millipore) in serum-free medium. The lower compartments contained medium supplemented with 20% fetal bovine serum as a chemoattractant. After 24 h, migrated cells attached to the underside of the membrane were fixed with 95% ethanol, stained with 0.1% crystal violet for 15 min, and quantified by counting cells in five random fields under a microscope.

Tumor xenograft models
BALB/c nude mice were obtained from the Shanghai Institute of Materia Medica and housed in pathogen-free conditions. Mice were randomly assigned to experimental groups. All animal procedures were conducted in accordance with the ethical guidelines of the Animal Experimental Ethics Committee of Wannan Medical College. A549 cells (2 × 10^6 cells/mouse) expressing either shCtrl or shTHOC3 were injected subcutaneously into the right dorsal region of each mouse. Tumor growth was monitored daily using Vernier calipers to measure tumor dimensions. Mice were euthanized on day 35, and tumors were excised and weighed. Mice with severe tumor ulceration or significant weight loss that compromised tumor measurement were excluded from the study.

Statistical analysis
Data were analyzed using SPSS software (version 13.0). Results are presented as means ± standard deviation (SD). Comparisons between two groups were performed using Student's t-test, while multiple group comparisons were made using one-way ANOVA. Survival curves for overall survival (OS), progression-free survival (PFS), and disease-free survival (DFS) were generated using the Kaplan–Meier method and analyzed using log-rank tests. A p-value < 0.05 was considered statistically significant.

Results

Upregulation of THOC3 expression in lung cancer (LC) and its association with poor prognosis
We assessed THOC3 mRNA levels in lung squamous cell carcinoma (LUSC), lung adenocarcinoma (LUAD), and normal lung tissues using data from the Gepia2 platform (Fig. 1A and 1B). In addition, LC tissues from multiple patients were collected to evaluate THOC3 expression. RT-qPCR analysis revealed significantly elevated THOC3 mRNA levels in 50 non-small cell lung cancer (NSCLC) tumor samples compared to adjacent normal tissues (Fig. 1C). Western blotting further confirmed the upregulation of THOC3 protein expression in NSCLC tissues relative to matched non-cancerous samples (Fig. 1D). Increased THOC3 mRNA and protein levels were also observed in four NSCLC cell lines (PC9, H1299, A549, and H1975) by qRT-PCR and Western blot (Fig. 1E and 1F). Additionally, THOC3 expression was found to be higher in high-grade lung carcinomas (WHO III/IV) compared to low-grade carcinomas (WHO I/II) (Fig. 1G and 1H).
Kaplan-Meier survival analysis revealed a significant association between elevated THOC3 expression and poor overall survival (OS) and disease-free survival (DFS) (P < 0.05, Fig. 1I-L). The survival data shown in Figs. 1I-L are derived from the GEPIA2 database (http://gepia2.cancer-pku.cn), which primarily utilizes data from The Cancer Genome Atlas (TCGA) and the Genotype-Tissue Expression (GTEx) projects.Fig. 1I shows the overall OS data from 962 patients with lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC), where p > 0.05, indicating that there is no significant survival difference between the high and low THOC3 expression groups in non-small cell lung cancer (NSCLC) patients. Fig. 1J shows the OS data from 478 LUAD patients, where p 〈 0.05, indicating that there is a significant survival difference between the high and low THOC3 expression groups in LUAD patients. Fig. 1K shows the OS data from 482 LUSC patients, where p 〉 0.05, indicating that there is no significant survival difference between the high and low THOC3 expression groups in LUSC patients. Fig. 1I also shows the overall PFS data for NSCLC patients (combining LUAD and LUSC), where p < 0.05, suggesting that THOC3 expression significantly affects progression-free survival in NSCLC patients. In summary, our database analysis indicates that THOC3 expression significantly impacts PFS in NSCLC patients (p < 0.05) and OS in LUAD patients (p < 0.05), but has no significant effect on OS in LUSC patients (p > 0.05).

THOC3 silencing inhibits LC cell proliferation and promotes apoptosis in vitro
As shown in Fig. 2A–D, shTHOC3#1 exhibited the most effective silencing activity (P < 0.05) and was thus selected for further functional and animal experiments. Cell viability assays demonstrated that THOC3 knockdown significantly inhibited the proliferation of both A549 and H1299 cells compared to control cells (P < 0.05, Fig. 2E and 2F). Colony formation assays also showed a marked reduction in colony number following THOC3 silencing (P < 0.05, Fig. 2G and 2H). Additionally, cellular migration was significantly impaired in THOC3-depleted cells, as evidenced by transwell migration and wound healing assays (Fig. 2I and 2J; Fig. 3A and 3B).
To investigate the effects of THOC3 depletion on epithelial-mesenchymal transition (EMT) and apoptosis, we examined the expression of related markers. THOC3 knockdown in A549 and H1299 cells led to a significant reduction in N-cadherin, Vimentin, and Slug protein levels, with a concomitant increase in E-cadherin expression (P < 0.05, Fig. 3C and 3D). Flow cytometry analysis confirmed a significant increase in apoptosis in THOC3-depleted cells (P < 0.05, Fig. 3E and 3F). Moreover, THOC3 silencing resulted in elevated levels of the pro-apoptotic protein BAX, while anti-apoptotic proteins Bcl-xl and Bcl-2 were reduced (Fig. 3G and 3H).

THOC3 silencing suppresses the tumorigenicity of LC cells in vivo
To validate the inhibitory effect of THOC3 knockdown on tumorigenesis, nude BALB/c mice were subcutaneously implanted with A549 cells transduced with either THOC3-specific shRNA (shTHOC3) or control shRNA (shCtrl). Tumor growth was monitored weekly once tumors became detectable. Representative images of tumors harvested from both experimental groups are shown in Fig. 4A. Tumor weight and volume were significantly reduced in the THOC3-knockdown group compared to controls (P < 0.05, Fig. 4B and 4C). Immunohistochemistry revealed decreased Ki67 expression and increased Cleaved Caspase-3 levels in tumors from the THOC3 knockdown group (P < 0.05, Fig. 4D and 4E). WB analysis was performed to detect THOC3 levels in tissue samples, confirming the knockdown efficiency and comparing the expression of STAT3 and pSTAT3 (Fig. 4F).

THOC3 promotes LC progression by activating the STAT3 pathway
To further elucidate the mechanisms underlying THOC3′s role in LC progression, we performed rescue experiments. As shown in Fig. 5A and 5B, shTHOC3#1 demonstrated the most potent silencing effect and was selected for subsequent rescue studies. To restore STAT3 signaling, we used Garcinone D, a known activator of phospho-STAT3. Overexpression of THOC3 in THOC3-silenced A549 and H1299 cells reversed the effects of THOC3 knockdown, including restored cell proliferation. CCK-8 and colony formation assays confirmed that Garcinone D treatment effectively activated the STAT3 pathway and restored cell proliferation potential (Fig. 5C and 5D). Furthermore, Garcinone D suppressed proliferation and migration in THOC3-overexpressing cells, as shown by colony formation, wound healing, and transwell assays (Fig. 5E and 5F; Fig. 5G and 5H; Fig. 6A and 6B). Flow cytometry analysis revealed that Garcinone D treatment partially reversed the reduction in apoptotic cells caused by THOC3 overexpression (Fig. 6C and 6D).
To explore the underlying molecular mechanisms, we evaluated β-catenin signaling-related proteins by Western blot in THOC3-overexpressing A549 and H1299 cells treated with Garcinone D. Our results demonstrated that Garcinone D treatment increased the levels of BAX and E-cadherin, while suppressing the expression of Vimentin, p-STAT3, N-cadherin, Bcl-xl, and Bcl-2 (Fig. 6E and 6F). Collectively, these findings support the hypothesis that THOC3 promotes lung cancer progression through the activation of the STAT3 signaling pathway.

Discussion
Tumorigenesis is a complex and sequential process characterized by uncontrolled cell proliferation, primarily driven by cumulative genetic and epigenetic alterations in key cell-cycle regulatory genes [18]. Identifying critical proteins and genes involved in tumor growth is essential for developing advanced diagnostic and prognostic tools for cancer [19].
THOC3, which encodes a 351-amino-acid protein located on chromosome 5, is a crucial subunit of the THO complex, involved in transcriptional elongation, RNA splicing, and nuclear RNA export. Accumulating evidence has demonstrated elevated THOC3 expression across various cancer types, particularly in lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC), with higher expression levels correlating with poor clinical outcomes. Specifically, in LUSC, THOC3 upregulation contributes to accelerated tumor progression by enhancing cell proliferation, migration, and glycolytic metabolism. Conversely, THOC3 suppression markedly inhibits tumor growth and metastasis. Moreover, THOC3 may promote tumorigenesis by interacting with other proteins, such as YBX1, to regulate mRNA stability, as seen in alterations to PFKFB4 mRNA expression [20].
The epithelial-mesenchymal transition (EMT) is a dynamic and reversible process involved in various physiological and pathological conditions, including embryogenesis, wound healing, tissue regeneration, fibrosis, and cancer metastasis [21]. EMT enables epithelial cells to acquire mesenchymal features, which can be reversed through mesenchymal-to-epithelial transition (MET). Traditionally, EMT is characterized by reduced expression of epithelial markers and increased expression of mesenchymal markers, such as Vimentin and N-cadherin, while epithelial markers like E-cadherin and cytokeratins are elevated during MET [22]. In line with these definitions, our experiments revealed that THOC3 depletion in H1299 and A549 cell lines resulted in significantly increased E-cadherin expression and decreased levels of N-cadherin and Vimentin (P < 0.05), suggesting that THOC3 may influence EMT in LC.
In the cytoplasm, THOC3 interacts with YBX1, a known m5C reader protein involved in mRNA stability. Knockdown of either THOC3 or YBX1 led to the downregulation of 104 genes implicated in key biological processes, such as glycolysis and glucose metabolism, including a marked reduction in PFKFB4 levels [23]. PFKFB4 is a critical regulator of glycolysis and pentose phosphate pathway production, playing a vital role in tumorigenesis [24]. Despite these insights, the precise molecular mechanisms and prognostic implications of THOC3 in LC remain poorly understood. In the present study, clinical analysis revealed that elevated THOC3 expression was associated with poorer clinicopathological outcomes, particularly tumor grade, in a large cohort of LC patients. Furthermore, survival analyses showed that patients with lower THOC3 expression had significantly better overall survival, highlighting THOC3 as a potential prognostic biomarker for LC. Additionally, high THOC3 expression was consistently confirmed in human LC tissue samples in vivo.
Proteins within the THOC family are known to participate in diverse signaling pathways [[25], [26], [27], [28], [29]]. Our study identified a significant interaction between THOC3 and the STAT3 signaling pathway. By activating STAT3, THOC3 enhanced the malignant behaviors of LC cells. The aberrant activation of STAT3 has long been recognized as a crucial factor in tumor initiation, progression, and resistance to chemotherapy and radiotherapy. Therefore, understanding the molecular mechanisms of EMT, particularly in cancer, may lead to the development of more effective therapeutic interventions. Our findings demonstrate that THOC3 accelerates LC progression by modulating EMT through the activation of the STAT3 pathway. Elevated THOC3 expression may enable LC cells to survive chemotherapy, while THOC3 knockdown significantly promoted apoptosis. Thus, THOC3 plays a pivotal role in regulating cells during EMT and could serve as a promising target for novel anticancer therapies. Gene expression profiling using microarrays confirmed that THOC3 promotes LC cell proliferation, tumorigenesis, and survival while inhibiting apoptosis and modulating key cellular events relevant to tumor progression. Notably, STAT3 pathway activation by THOC3 strongly contributes to the aggressive growth of LC.
STAT3 is an oncogenic regulator implicated in various cellular processes, including apoptosis regulation, proliferation, metastasis, and immune evasion. Growing evidence supports that constitutive STAT3 activation significantly contributes to tumor progression in multiple malignancies [[30], [31], [32]]. Furthermore, high levels of phosphorylated STAT3 (p-STAT3) are strongly correlated with poor patient prognosis, making STAT3 a potential therapeutic target [[33], [34], [35]]. Our study suggests that THOC3 enhances the malignant traits of LC primarily through STAT3 activation.
However, some limitations of this study should be noted. The experiments conducted both in vitro and in vivo used low-passage LC cell lines, which may not fully replicate the complexity of clinical lung tumors. Therefore, further validation in primary cultures of LC cells is necessary. Future studies should also explore whether THOC3 overexpression influences other signaling pathways or impacts various LC cell subtypes. We are currently investigating the impact of THOC3 overexpression on LC cell behavior. A more comprehensive understanding of THOC3’s role in LC progression will help elucidate the mechanisms that govern LC cell survival and position THOC3 as a potential therapeutic target for this disease. Funding constraints limited the extent of molecular investigations into THOC3-mediated modifications in STAT3 signaling.

Conclusion
In conclusion, this study highlights THOC3 as a newly characterized functional gene implicated significantly in LC progression. Understanding of THOC3’s precise biological functions in LC could potentially guide the development of targeted treatments based on THOC3 inhibition. Further translational investigations are warranted to establish practical clinical approaches and diagnostic applications targeting THOC3. Additionally, while THOC3 expression may have utility as a biomarker for LC prognosis and disease progression, further clinical and experimental studies are required to validate its relevance in a therapeutic context.

Consent for publication
Not applicable.

Data availability
All the data used and analysed in this study are available from the corresponding author upon request.

Funding
The Young and Middle-aged Fund Science and Technology Project of Wannan Medical College (No. WK2024ZQNZ64; No.WK2022F28).

CRediT authorship contribution statement
Zhengzheng Ni: Software, Methodology, Conceptualization. Sheng Zhou: Writing – original draft, Data curation, Conceptualization. Yan Zhang: Writing – review & editing, Resources. Xinyang Wang: Resources, Formal analysis, Data curation. Hui Yang: Writing – review & editing, Validation, Investigation, Formal analysis. Hanyu Zhou: Writing – review & editing, Conceptualization. Zheng Tao: Funding acquisition, Conceptualization.

Declaration of competing interest
All authors declare that they have no known competing interest.