1 Introduction
Lung cancer has the highest mortality rate of all cancers worldwide [
1]. Every year, approximately 156 000 people die from lung cancer, even more than the next three most incident cancers combined. More women have died from lung cancer than breast cancer since 1987 [
2]. Although smoking is one of the most important risk factors contributing to lung cancer, the mortality of non-smoking lung cancer patients has already risen to the seventh position [
3]. Among several histological subtypes of lung cancer, non-small cell lung cancer (NSCLC) accounts for the majority of new diagnosis at 85%, while small cell carcinoma only accounts for the remaining 15% [
4]. The histological subtypes of NSCLC includes: large cell carcinoma, squamous cell carcinoma and adenocarcinoma accounting for 10, 25 and 40% separately [
5]. The rest are mainly carcinoid tumors and bronchalveolar carcinoma. Although some clinicopathological parameters including tumor stage, size, and lymph node metastasis have been proven to be reproducible prognostic determinants in lung cancer, they are far from sufficient to explain the individual variability [
6]. Therefore, there is an urgent need for reliable cancer markers for diagnosis and prognosis of patients with lung cancer.
T-box family members play important role in regulation of embryonic development, and particularly in morphogenesis and the assignment of cell fate [
7].
TBX2 subfamily is a member of T-box transcription factors, which includes the closely related genes such as
TBX2,
TBX3,
TBX4 and
TBX5 [
8]. T-box proteins contain a highly conserved T-domain which can recognize the core sequence GGTGTGA (T-element) and affect dimerization and DNA binding [
9]. T-box proteins function as putative transcription factors which regulate the expression of downstream targeted genes. Recently, it has been reported that
TBX2 subfamily members might be involved in generation and development of tumors [
10]. Recent studies indicated that
TBX2 was amplified in 8.6%-21.6% of sporadic human breast carcinomas, where the protein was overexpressed [
11]. Besides, ectopic expression of
TBX2/3 contributed to chemotherapy resistance, DNA polyploidy and tumor proliferation [
12]. It was also reported that the overexpression of
TBX2 in prostate cancer was correlated with pathological grade and tumor stage and might act as a potential cancer marker in prostate cancer [
13]. On the other hand, low expression of
TBX4 in patients with pancreatic ductal adenocarcinoma (PDAC) predicted a poor prognosis [
14]. Taken together, these evidences suggest that
TBX2 subfamily genes may function as tumor promoting or inhibiting factors and play vital role in cancer progression.TBX2 subfamily controls lung growth by repressing the cell cycle inhibitor genes Cdkn1a and Cdkn1b [
15]. TBX2 subfamily function downstream of Shh by directly repressing the Wnt antagonists Frzb and Shisa3, thereby maintaining the proliferative mesenchymal Wnt signaling [
16]. TBX2 subfamily also maintains the proliferation of lung mesenchyme through at least two molecular mechanisms: regulating cell cycle control and integrating the activity of multiple signaling pathways. However, TBX2 subfamily regulates the proliferation of lung interstitial cells through various mechanisms, yet its regulatory mechanism as a transcription factor in lung cancer remains unclear. Previous studies have reported low expression of the TBX2 subfamily genes in lung cancer tissue. Previous studies have reported low expression of the TBX2 subfamily genes in lung cancer tissue, the relationship between low expression of
TBX2 subfamily and prognosis of lung cancer still remains elusive [
17]. Here on this basis, we further sought to survey the relationship between
TBX2 subfamily expression and prognosis of lung cancer by bioinformatics and verified it in human lung cancer specimens. Finally, we used two databases to verify the epigenetic mechanism of low expression of
TBX2 subfamily.
3 Discussion
In previous researches,
TBX2 subfamily of transcription factors disorder has been reported to be involved in various human diseases [
18]. Until recently, the role of
TBX2 subfamily in the process of carcinogenesis comes into the eyesight of researchers. Although there have been certain reports which clarify the contributions of
TBX2 subfamily members in cancer pathogenesis, the effect of
TBX2 subfamily in carcinogenesis and prognosis of lung cancer still remains elusive. We report for the first time that low expression of Tbx2 subfamily predicts poor prognosis in patients with lung cancer, and further verify the low expression of Tbx2 in pathological specimen of lung cancer. The low expression of
TBX2 subfamily is a poor prognostic factor for NSCLC, which may be related to the increased methylation of promoter region in NSCLC patients. This study is promising to shed light on discovering a reliable prognostic marker for patients after lung cancer resection.
It has been reported that
TBX2, member of Tbx2 subfamily of transcription factors, was overexpressed in breast and bladder cancer [
19,
20].
TBX2 started mechanisms of normal growth control by recruiting a series of inhibitory compounds to combine with reactive promoter of
EGR1, which led to uncontrolled proliferation of breast cancer cells [
21]. Nevertheless, our research suggested that the expression of
TBX2 was significantly downregulated in lung cancer, which was supported by Khalil’s study [
22]. It is worth noting that
TBX2 represented a higher expression in normal lung tissues compared with other specific normal tissues, which suggesting an important role of
TBX2 in embryonic development and morphogenesis of lung [
15].
There were conflicting reports about the role of
TBX3 in different types of cancer. On the one hand, it was reported that
TBX3 was overexpressed in most human malignancies such as melanoma, breast, ovarian and bladder cancer [
23,
24]. In these studies, the overexpression always predicted metastasis and poor prognosis in patients, which suggested a promoting effect of
TBX3 on malignancy progression in cancer. Study on the mechanism has revealed that
TBX3 promoted invasion of tumors mainly through inhibition of
E-cadherin expression [
25]. On the other hand, our data suggested that the expression of
TBX3 in lung cancer was significantly downregulated, which was contrary to Wu’s study [
26]. Besides,
TBX3 correlated with survival of lung cancer patients. Low expression of
TBX3 in lung cancer predicted low survival and poor prognosis of patients, which indicated that
TBX3 was an independent prognosis factor and a potential therapeutic target for OS in lung cancer.
TBX3 might play an inhibitory role in malignancy progression of lung cancer.
A proteomics study revealed the relationship between
TBX4 and malignancy progression in cancer for the first time. This study found that
TBX4 was overexpressed in pancreatic ductal adenocarcinoma tissue (DAC). Thereafter, further research of Meijuan Zong found that
TBX4 was an independent prognostic factor for OS in stage II pancreatic ductal carcinoma (PDC). Low expression of
TBX4 predicted poor prognosis of stage II PDC [
14]. High-throughput sequencing technology analysis in lung fibroblasts suggested that
TBX4 in lung cancer associated fibroblasts (CAFs) was downregulated and highly methylated [
27]. It indicated that epigenetic silence of
TBX4 was involved in phenotypic alteration of CAFs from lung cancer. Similar to the result of Horie M, our data showed that the expression of
TBX4 in lung cancer tissues showed a significant decrease compared to normal tissues. Besides, low expression of
TBX4 predicted poor prognosis of lung cancer patients.
TBX4 was associated with tumor stage of lung cancer, similar to the result in PDC.
Earlier studies had shown that
TBX5 played an important role in cardiac and forelimb development in mammals [
28]. However, the role of
TBX5 in malignancy progression in cancer still remains elusive. Rosenbluh reported that in the development of colon cancer, the compounds formed by
β-catenin, yes-associated protein 1 (
YAP1) and
TBX5 were essential for survival of colon cancer, similar phenomena were observed in other tumors [
29]. High expression of
TBX5 had been verified to be related with low survival rate of stage I and II gastric cancer patients [
30,
31]. However, our research finding proved that the expression of
TBX5 was downregulated in lung cancer. Surprisingly, low expression of
TBX5 turned out to be associated with a decrease in OS, which predicted a poor prognosis in lung cancer. In conclusion, our research finds that the expression of all four
TBX2 subfamily members is downregulated in lung cancer tissues.
In our research, the TBX2 subfamily gene shows low expression in non-small cell lung cancer. Through survival analysis, we found that the low expression of the TBX2 subfamily gene in non-small cell lung cancer patients indicates poor prognosis. Finally, through mutual verification analysis of two methylation bioinformatics databases, it is suggested that the low expression of the TBX2 subfamily in non-small cell lung cancer patients may be highly correlated with methylation in its promoter region. Our study is promising to shed light on discovering a novel reliable cancer marker for prognosis of lung cancer patients.
4 Materials and methods
4.1 Ethics statement
This study was approved by the Academic Committee of Wuxi 9Th People's Hospital Affiliated to Soochow University and conducted according to the ethical standards formulated in the Helsinki Declaration. It was confirmed that all written informed consents were obtained as all data sets were retrieved from published literatures.
4.2 Classification and gene structure of TBX2 subfamily
The HUGO Gene Nomenclature Committee (HGNC) assigns unique symbols and names to human genes based on the European Bioinformatics Institute (EMBL-EBI). In current, the HGNC database not only integrates more than 40000 approved gene symbols, but also curates genes into family sets based on shared characteristics such as homology, function or phenotype. Herein, we retrieved basic information of Tbx2 subfamily from the HGNC database (
http://www.genenames.org/cgi-bin/genefamilies/set/766) [
32].
TBX2 subfamily contains four members, including
TBX2,
TBX3,
TBX4 and
TBX5. Then, we retrieved detailed information about TBX2 subfamily through the search terms of
TBX2/homo,
TBX3/homo,
TBX4/homo and
TBX5/homo from the NCBI (National Center for Biotechnology Information) database. Finally, graphic mode of structure of TBX2 subfamily transcription products including coding region, untranslated coding region (UTR) and T-BOX domain was drawn with software.
4.3 UALCAN analysis
UALCAN gene expression array datasets is a cancer microarray database and online data-mining platform which aimed at facilitating discovery from genome-wide expression analysis [
33]. Herein, we use this database to analyze the transcriptional level of TBX2 subfamily genes in different types of cancers. We compared the expression level of TBX2 subfamily in clinical cancer specimens with that in normal controls and used Student’s t test for result analysis. The cutoff of P value and fold change were defined as 0.01 and 1.5, respectively. Gene rank: 10%, data type: mRNA. We also performed protein expression analysis of the TBX2 subfamily in lung squamous cell carcinoma and lung adenocarcinoma compared to normal lung tissue using the protein expression module of the UALCAN database.
4.4 Immunohistochemistry (IHC) staining
To evaluate differences in TBX2 subfamily expression at the protein level, IHC images of TBX2 subfamily protein expression in normal tissues and Lung Squamous Cell Carcinoma were downloaded from the The Human Protein Atlas (
http://www.proteinatlas.org/) and analyzed [
34]. The Human Protein Atlas serves as a comprehensive database housing information pertaining to the proteome of the human body, encompassing details regarding their distribution, expression levels, and functional roles within diverse tissues and cells. This resource furnishes detailed imagery, gene expression profiles, and other pertinent data, facilitating the comprehension of the pivotal roles played by proteins in the context of both health and disease.
4.5 Cell culture
H-226, H-1792, Beas-2B, and HepG2 cell lines are gifts donated by the Clinical Medical College of Suzhou University. All the above-mentioned cells were cultured in an environment with a temperature of 37 ℃ and a gas environment of 5% CO2. All cell lines were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum and 1% penicillin–streptomycin mixture (HyClone, USA). After being resuscitated from liquid nitrogen, each cell line was passaged 15 times before being used for subsequent experiments. Cell culture was performed using a 6-well plate with a cell density close to 80%, approximately 5*10^5 cells were used for protein extraction.
4.6 Western blotting analysis
To prepare protein lysate, cells were lysed with RIPA lysis buffer supplemented with protease inhibitors to reduce protein degradation. The lysate was maintained at an icy temperature during the entire protein extraction process to uphold sample integrity. The proteins were then extracted by centrifugation of the lysate at 1400 g for 15 min under 4 ℃. The protein concentration of the extracted proteins was quantified with the BCA protein assay kit. After protein extraction, SDS-PAGE gel electrophoresis was performed to separate the protein, which was then transferred onto a PVDF membrane. The PVDF membrane was shaken in 5% skim milk, and then blocked overnight at 4 ℃ in a wet box containing primary antibodies against TBX2(Abcam ab33298), TBX3(Abcam ab99302), TBX4(Santa Cruz sc-515196), TBX5 (Abcam ab137833)and GAPDH(Santa Cruz sc-47724).On the following day, after 1 h of incubation at room temperature with secondary antibodies, analysis was conducted using an ECL protein imprint imaging device. The dilution ratio for the secondary antibody is 1:2000.
4.7 Kaplan–Meier plotter
The Kaplan–Meier plotter (
http://kmplot.com/analysis/) enables to assess the effect of any gene or gene combination on survival in various types of tumors using over 50,000 samples measured with gene arrays, RNA-seq or next generation sequencing [
35]. The Kaplan–Meier plotter (K-M plotter) is capable of assessing the effect of up to 54 k genes on survival of 21 cancer types and the largest datasets contain breast (n = 6,234), lung (n = 3,452), ovarian (n = 2,190), and gastric (n = 1,440) cancer. Herein, we used the K-M plotter to evaluate the prognostic value of mRNA expression of distinct
TBX2 subfamily genes in lung cancers. In K-M plotter, we divided cancer participants into low and high expression group according to median values of
TBX2 subfamily mRNA expression. The prognostic value of
TBX2 expression was validated by K-M survival curves. Information about the number-at-risk cases, median values of mRNA expression levels, HRs, 95% CIs and p-values is all available on the K-M plotter website. P < 0.05 was considered statically significant.
4.8 DNA methylation database
MethHC was a database for human cancer gene expression and methylation [
36]. The correlation between TBX2 subfamily expression and its methylation were confirmed by the Pearson correlation coefficients. P < 0.05 has statistical significance. DiseaseMeth 2.0 is a database that aims to provide the most complete collection and annotation of abnormal DNA methylation in human diseases, especially in various cancers [
37]. The database includes 175 large-scale methylation datasets of 88 diseases and more than 14000 accurate experimental information extracted from PubMed. The DNA methylation index is calculated based on the mean methylated intensity (M) and unmethylated intensity (U) at each locus in every sample, using the formula (β = M/ [M + U]), and is presented as β values.
4.9 Statistical analysis
The Student's t-test was employed to identify variations in the expression of TBX2 subfamily mRNA between cancerous tissues and adjacent normal lung tissues. We utilized the log-rank test to evaluate the potential correlation between the expression of TBX2 subfamily proteins and the overall survival (OS) of lung cancer patients. A p < 0.05 is deemed to be statistically significant.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.