Role of PAX6, TRPA1, BCL11B, MCOLN2, CUX1, EMX1 in colorectal cancer and osteosarcoma.

1. Introduction

Colorectal cancer is one of the most common malignant tumors of digestive system in the world, with a high mortality rate. [ 1 ] In its initial stages, colorectal cancer may manifest asymptomatically, but as the tumor progresses, patients often encounter symptoms such as abdominal pain, alterations in stool patterns, bloody stools, anemia, and weight loss. [ 2 ] China is a low incidence region of colorectal cancer in the world, but the incidence of colorectal cancer has an increasing trend in many regions. [ 3 ] Commonly employed screening methods for colorectal cancer encompass colonoscopy, fecal occult blood tests, virtual colonoscopy, and others. [ 4 ] Colorectal cancer treatment depends on the stage of cancer, the location of the tumor and the patient overall health. Treatment may include surgery, radiation therapy, chemotherapy and targeted therapy [ 5 ]

Bone tumors can manifest in bone or its associated tissues and may be categorized as either benign or malignant. Although the incidence of malignant bone tumor is not high, accounting for <0.2% of the total tumor, it develops rapidly and has a high mortality. [ 6 ] OS stands out as the most prevalent primary malignant bone tumor, particularly impacting children and adolescents. [ 7 ] Osteosarcoma (OS) significantly impacts patients’ life expectancy and quality of life. Individuals diagnosed with primary malignant bone tumors typically present with pain and a mass at the primary site. X-rays often reveal soft tissue masses and tumors characterized by mixed osteoblastic and osteolytic destruction. [ 8 ] OS occurs mainly in the long bones of the limbs covering the femur, tibia, and humerus, and less frequently in the skull, jaw, or pelvis with different molecular and biological behaviors. [ 9 ] Given the high morbidity and mortality associated with OS, comprehensive testing, risk assessment, and survival prognosis analysis are critical to improving the diagnosis and treatment of OS.

As an important part of the development of life science, bioinformatics has always been in the forefront of life science and technology research. In recent years, the rapid development of biotechnology, biological information resources also appeared explosive growth. Bioinformatics serves as a conduit connecting data and clinical insights, unveiling the biological significance inherent in vast biological datasets. Particularly in the realm of tumor therapy, bioinformatics plays a crucial role, notably in the analysis and reporting of genetic testing data. [ 10 , 11 ]

This paper intends to use bioinformatics technology to excavate core genes between colorectal cancer, OS and normal tissues, and conduct enrichment analysis and pathway analysis. Public data sets were used to verify the significant role of molecule in colorectal cancer and OS.

2. Method

2.1. Colorectal cancer and osteosarcoma datasets

In this study, the configuration files for the colorectal cancer dataset GSE142279 and OS datasets GSE197158 and GSE206448 were downloaded from the Gene Expression Omnibus database ( http://www.ncbi.nlm.nih.gov/geo/ ) and generated using the platforms GPL20795 , GPL20301 , and GPL24676 . GSE142279 includes 20 colorectal cancer and 20 normal tissue samples. GSE197158 includes 11 OS and 3 normal tissue samples. GSE206448 includes 5 OS and 5 normal tissue samples.

3. The remove batch effect

For the merging and de-batching of multiple datasets, the initial step involved merging the GSE197158 and GSE206448 datasets using the R package. The merging process utilized the Silico Merging package [DOD:10.1186/14712105-13-335], resulting in a merged matrix. Subsequently, the R package limma (version 3.4.2.2) was employed to eliminate batch effects. The “removeBatchEffect” function was employed to eliminate batch effects, and finally obtain the matrix after removing the batch effect. This batch effect-corrected matrix was subsequently utilized in subsequent analyses.

4. Screening of differentially expressed genes (DEGs)

R package “limma” is used for probe aggregation and background correction of the merge matrix of GSE197158 and GSE206448 . The Benjamini-Hochberg method is used to adjust the original P value. The fold change was calculated using the false discovery rate (FDR). The cutoff standard of DEGs was set at P < .05. A volcano plot was generated to visually represent the DEGs.

5. Weighted gene co-expression network analysis (WGCNA)

Firstly, the de-batched and post-merged matrix of GSE197158 and GSE206448 was utilized to compute the median absolute deviation for each gene. Subsequently, the top 50% of genes with the smallest median absolute deviation were excluded. The “goodSamplesGenes” method from the R package WGCNA was then employed to eliminate outlier genes and samples. Following this, WGCNA was applied to construct a scale-free co-expression network. In order to classify the genes with similar expression profiles into gene modules, the average linkage hierarchical clustering was carried out according to the dissimilarity measure based on Topological Overlap Matrix, and the minimum size (genome) was set at 30, with a sensitivity of 3. For further module analysis, the feature genes’ differences within the module were calculated, a cut line was chosen for the module tree, and some modules were merged. In addition, we also merged the modules with a distance <0.25, and finally obtained 66 co-expression modules. Notably, the gray module comprises genes considered unassignable to any specific module.

6. Construction and analysis of protein-protein interaction (PPI) network

The Search Tool for the Retrieval of Interacting Genes (STRING) database ( http://string-db.org/ ) is designed to aggregate, score, and integrate protein-protein interaction information from publicly available sources. It also supplements these sources by generating predictions. In this study, the list of DEGs was input into STRING database to construct a PPI network for predicting core genes (confidence > 0.4). Cytoscape software can provide biologists with biological network analysis and 2-dimensional visualization. In this study, the PPI network formed by STRING database was visualized and core genes were predicted by Cytoscape software. First of all, we imported the PPI network into the Cytoscape software, found the module with the best correlation through MCODE, and calculated the genes with the best correlation through 2 algorithms (MCC and MNC). After visualization, a list of core genes was derived.

7. Functional enrichment analysis

Gene Ontology (GO) and Kyoto Encyclopedia of Gene and Genome (KEGG) analysis are computational methods for evaluating the function and biological pathways of genetics. In this study, the list of differential genes screened by Wayne map was input into KEGG API ( https://www.kegg.jp/kegg/rest/keggapi.html ) to obtain the latest KEGG Pathway gene annotation. This annotation served as the background against which genes were mapped, and the R software package cluster Profiler (version 3.14.3) was used for enrichment analysis to obtain the results of gene set enrichment. The GO annotation of genes in R software package (version 3.1.0) was also used as a background for gene mapping. The minimum gene set was set to 5, and the maximum gene set was set to 5000, with a significance threshold of P < .05 and a FDR of < 0.25. It was considered to be a statistically significant measure.

In addition, Metascape database can provide comprehensive gene list annotation and analysis resources, and can be visually exported. Metascape ( http://metascape.org/gp/index.html ) was employed for functional enrichment analysis of the aforementioned list of differentially expressed genes, with the results being visually exportable.

8. GSEA

For Gene set enrichment analysis (GSEA), we obtained the GSEA software from GSEA. In order to evaluate the related pathways and molecular mechanisms, based on gene expression profile and phenotypic grouping, the minimum gene set was 5, the maximum gene set was 5000, and a thousand resampling times, with a significance threshold of P < .05 and a FDR of < 0.25 considered statistically significant. GO and KEGG analyses were performed on the entire genome using GSEA.

9. Gene expression heat map

To visualize the expression levels of core genes identified in the Protein-Protein Interaction (PPI) network, heatmaps were generated using the R package “heatmap.” These heatmaps were constructed using the merged matrix of GSE197158 and GSE206448 , as well as the gene expression matrix from GSE142279 . The visualizations aimed to represent the expression differences of core genes across colorectal cancer, OS, and normal tissue samples.

10. CTD analysis

Comparative Toxicogenomics Database (CTD) database integrates a large number of chemical substances, genes, functional phenotypes and disease interaction data, which provides great convenience for the study of disease-related environmental exposure factors and drug potential mechanism. Core genes were input into the CTD website to identify the diseases most closely associated with them. Excel was then used to create a radar map illustrating the differential expression of each gene.

11. miRNA

TargetScan ( www.targetscan.org ) is an online database for predicting and analyzing miRNA and target genes. In our study, TargetScan was used to screen the miRNAs that regulates central DEG.

12. Results

12.1. Analysis of differentially expressed genes

In this study, differentially expressed genes analysis was conducted on the OS datasets GSE197158 and GSE206448 , utilizing a batch-corrected merged matrix with predefined cutoff values (Fig. 1 A). Similarly, for the colorectal cancer dataset GSE142279 (Fig. 1 B), differential gene expression was identified from the gene expression matrix using the specified threshold.

13. Functional enrichment analysis

13.1. DEGs

We analyzed these differentially expressed genes by GO and KEGG. According to GO analysis, they were mainly enriched in acetylcholine binding donor-receptor activity involved in Wnt signal pathway, cell polarity pathway, PI3K-Akt signal pathway, receptor regulator activity, cytokine-cytokine receptor interaction, transcriptional disorders in cancer, and regulation of tryptophan channel by inflammatory mediators (Fig. 2 A, B, C, and D).

14. GSEA

In addition, we carried out GSEA enrichment analysis of the whole genome in order to find out the possible enrichment items in non-differentially expressed genes and verify the results of differentially expressed genes. The intersection of enrichment items and GOKEGG enrichment items of differentially expressed genes is mainly concentrated in acetylcholine binding donor-receptor activity involved in Wnt signal pathway, cell polarity pathway, cytokine-cytokine receptor interaction (Fig. 2 E, F, G, and H).

15. Metascape enrichment analysis

In the metascape enrichment project, the regulation of Wnt signal pathway, muscle system process and actin filament movement can be seen in the GO enrichment project (Fig. 3 A). At the same time, we also output the enrichment network colored by enrichment term and P value (Fig. 3 B and C), which visually represents the correlation and confidence of each enrichment project.

16. WGCNA

The selection of soft threshold power is an important step in WGCNA. The network topology is analyzed to determine the soft threshold power. The soft threshold power in the WGCNA is set to 10 (Fig. 4 A and B). The hierarchical clustering tree of all genes was constructed, and 66 important modules were generated (Fig. 4 C). The interaction between these modules was analyzed (Fig. 4 D). The module-phenotypic correlation heat map (Fig. 5 A) and the GS-MM correlation scatter map of related hub genes (Fig 5 B and C) were generated. The correlation between the module feature vector and gene expression was calculated to obtain MM values. Based on a cutoff criterion (| MM | > 0.8), 6 genes with high connectivity were identified as hub genes within clinically significant modules. A Venn diagram was also generated, illustrating the intersection of differentially expressed genes screened by WGCNA and DEGs (Fig. 5 D).

17. Construction and analysis of protein-protein interaction (PPI) network

The Protein-Protein Interaction (PPI) network for the Differentially Expressed Genes (DEGs) in OS was constructed using the STRING online database and analyzed using Cytoscape software (Fig. 6 A). Two core gene clusters were obtained through MCODE analysis (Fig. 6 B and C). Central genes were identified using 2 different algorithms (Fig. 6 D, E, G, and H), and their intersection was obtained using Excel (Fig. 6 F and I). Eight core genes were identified (CUX1, NES, BCL11B, PAX6, EMX1, MCOLN2, TRPA1, TRPC4).

Additionally, we established a Protein-Protein Interaction (PPI) network for colorectal cancer (Fig. 7 A). Core gene clusters were obtained through the MCODE algorithm (Fig. 7 B), and subsequently, central genes were identified using the MCC algorithm (Fig. 7 C). A total of 10 core genes were obtained. These core genes were then intersected with the core genes from OS (Fig. 7 D). Ultimately, we obtained the intersection core genes between OS and colorectal cancer (CUX1, EMX1, TRPA1, BCL11B).

18. CTD analysis

In this study, we entered the hub gene list into the CTD website to find diseases related to core genes, improving the understanding of the association between genes and diseases. We found that 4 genes (CUX1, EMX1, TRPA1, BCL11B) are associated with colorectal neoplasms, colorectal tumors, colonic diseases, multiple myeloma, OS, and inflammation (Fig. 8 ).

19. Survival analysis

Using the survival data downloaded from TCGA for colorectal cancer, we generated forest plots (Fig. 9 A) and Kaplan–Meier survival curves (Fig. 9 B) related to the core genes (CUX1, EMX1, TRPA1, BCL11B) in colorectal cancer.

We also created a prognosis score plot illustrating the relationship between prognosis scores and the expression differences of core genes in colorectal cancer and normal tissue samples. It can be observed that with an increase in risk scores, the survival rate of patients significantly decreases, with the low-risk group exhibiting significantly higher survival times and rates compared to the high-risk group (Fig. 10 A). Visualizing the expression heatmap of core genes in colorectal cancer survival data reveals that TRPA1 acts as a protective factor, with its expression trending downward with increasing risk scores. On the other hand, core genes CUX1 and EMX1 act as risk factors, showing an upward expression trend with increasing risk scores (Fig. 10 B). We also plotted the ROC curve for the risk score, and the AUC values indicate a certain predictive value for the risk score (Fig. 10 C).

20. Gene expression heat map

We visualized the expression heatmap of core genes (CUX1, EMX1, TRPA1, BCL11B) in OS (Fig. 11 A) and colorectal cancer (Fig. 11 B) dataset samples. Core genes (CUX1, BCL11B) exhibited higher expression levels in both OS and colorectal cancer tissue samples, whereas their expression was lower in normal samples.

21. miRNA prediction and functional annotation related to hub gene

In this study, we input the hub gene list into targetsacan to find the relevant miRNA to improve the understanding of gene expression regulation (Table 1 ). We found that the related miRNA of TRPA1 was hsa-miR-489-3p; The related miRNA of BCL11B gene was hsa-miR-182-5p. The related miRNA of CUX1 gene were hsa-miR-520d-3p, hsa-miR-520e, and hsa-miR-520a-3p. The related miRNA of EMX1 gene were hsa-miR-15a-5p, hsa-miR-15b-5p and hsa-miR-16-5p.

22. Discussion

Globally, colorectal cancer is the third most common malignant tumor, is the second most common cause of death of malignant tumor. Colorectal cancer is associated with complications and is prone to recurrence and metastasis. [ 12 , 13 ] OS is the most common primary bone cancer, accounting for 19% of all bone malignancies. [ 14 ] Despite advances in cancer treatment over the past few decades, survival rates for patients with OS remain unsatisfactory. It is very important to explore the molecular mechanism of colorectal cancer and OS for the research of targeted drugs. The key findings of this study reveal elevated expression levels of PAX6, TRPA1, BCL11B, MCOLN2, CUX1, and EMX1 in both colorectal cancer and OS. Notably, higher expression of these genes correlates with poorer prognosis.

The PAX gene family encompasses 9 well-defined transcription factors that play pivotal roles in both development and disease. Although PAX gene is mainly expressed in embryo, constitutive expression promotes tissue hyperplasia. Rare tumor-specific mutations in PAX gene have been implicated in carcinogenesis, and persistent PAX expression is characteristic of several tumors. [ 15 ] PAX6 belongs to the highly conserved PAX family of transcription factors. This protein is important for the development of the central nervous system, eyes and endocrine pancreas, and is expressed in the precursor cells of the developing human cerebellum. [ 16 ] Dysregulation of PAX6 leads to developmental disorders and tumor formation. The expression of PAX6 has been reported in non-small cell lung cancer, breast cancer, bladder cancer, esophageal cancer, gastric cancer, and several cancer cells including melanoma and hematological malignancies. [ 17 ] PAX6 expression is highly tissue dependent, and in pancreatic cancer, it is upregulated and leads to oncogenic MET activation. In glioma cells, PAX6 acts as a tumor suppressor, and inhibition of PAX6 leads to increased proliferation and cell migration and higher cell expansion under stress conditions. [ 18 ] Studies have shown that up-regulated microRNA-19 negatively regulates the expression of PAX6 in OS cells, and Pax6 can promote the malignant phenotype of OS cells by activating extracellular signaling regulated kinase signaling pathways. [ 19 ] PAX6 is a new target of microRNA-7, which can promote the proliferation and invasion of colorectal cancer cells. [ 20 ] The above studies are consistent with our findings: PAX6 is highly expressed in colorectal cancer and OS, and the higher the PAX6, the worse the prognosis.

Transient receptor potential (TRP) channels are multimodal calcium-permeable cation channels that are widely used as cellular sensors. TRPA1 is the subject of various current research interests, and it has been recognized as a chemosensor activated by toxic electrophilic stimuli.TRPA1 is expressed in injurious neurons, where it acts directly on the onset of pain and pruritus, but is also found in the vascular endothelium and in the lung epithelium, among other places. [ 21 ] Notably, TRPA1 plays a prominent role in inflammation, both through its expression which can be regulated by inflammatory mediators and as a regulator of inflammatory signaling. [ 22 ] TNF, a well-defined pro-inflammatory cytokine, upregulates TRPA1 transport to the cell membrane and increases its membrane expression in peripheral nerves. [ 23 ] TRPA1 ion channels play a key role in metastasis, and is increasingly expressed in bone cancer. TRPA1 antagonists were found to downregulate OS cell migration and expression of pro-inflammatory cytokines. [ 24 ] The above studies are consistent with our findings: TRPA1 was highly expressed in colorectal cancer and OS, and the higher the TRPA1, the worse the prognosis.

BCL11B can act as both a transcriptional repressor and an activator. Previous studies have shown that BCL11B is an important transcription factor for T cell lineage commitment and differentiation, and its dysregulation has been shown to be associated with T cell tumorigenesis. [ 25 ] BCL11B has been shown to play a crucial role in the regulation and inhibition of T cells by regulating the expression of FOXP3, IL-10 and pro-inflammatory cytokines. [ 26 ] Studies have shown that BCL11B is involved in and regulates key processes such as progenitor cells, effector T cells, Tregulation and iNKT cell development, function generation and survival. [ 27 ] These reports highlight the potential importance of BCL11B in T cell maturation and the development of T cell malignancies. The above studies are consistent with our findings: BCL11B is highly expressed in colorectal cancer and OS, and the higher the BCL11B, the worse the prognosis.

TRPML proteins belong to the TRP superfamily of nonselective cation channels. TRPML1, TRPML2, and TRPML3 proteins are encoded by mucin MCOLN-1, MCOLN-2 and MCOLN-3 genes, respectively. [ 28 ] Dysregulation of TRP channels may malfunction Ca 2+ signaling, which is a pathophysiological marker of cancer, such as cancer cell survival, proliferation, and invasion. [ 29 ] Previous studies have implicated MCOLN2 in the progression of prostate cancer, with overexpression promoting prostate cancer cell proliferation, migration, and invasion through the IL-1β/NF-κB pathway. [ 30 ] Similarly, another study demonstrated that TRPM2-AS promotes malignant tumors in OS cells by targeting the miR-15b-5p/PPM1D axis. [ 31 ] The above studies are consistent with our findings: MCOLN2 is highly expressed in colorectal cancer and OS, and the higher the MCOLN2, the worse the prognosis.

CUX1, also known as CULT1 or CDP, belongs to the homologous domain transcription factor family. It features 4 similar DNA binding domains, including a CUT homology domain and 3 CUT repeats (CR1, CR2, CR3), which are evolutionarily and functionally conserved from metazoans to humans. [ 32 ] As a major member of the CUX family, CUX1 plays a crucial role in various biological processes, including cell differentiation, proliferation, cell cycle regulation, tissue development, and double-chain break repair responses. In addition, CUX1 expression is abnormally elevated in many malignant tumors, associating it with tumorigenesis. Many evidence has shown that CUX1 overexpression exists in various types of cancer, such as melanoma, pancreatic cancer, multiple myeloma and breast cancer. [ 33 , 34 ] Studies have shown that CUX1 promotes the development of carcinogenic properties by activating the Wnt/ beta-catenin signaling pathway in gliomas. [ 35 ] The above studies are consistent with our findings: CUX1 is highly expressed in colorectal cancer and OS, and the higher CUX1, the worse the prognosis.

EMX homologous domain proteins are members of the EMX transcription factor family and are associated not only with sarcomas but also with melanoma. [ 36 ] EMX1/2 proteins have been reported to be involved in regulating various biological processes, such as cell proliferation, migration, and differentiation during brain and neural crest development. Many sarcoma cells and tumors maintain high levels of Emx1/Emx2 protein expression. [ 37 ] A study demonstrated that EMX2 inhibits WNT signaling and activation of the WNT pathway bypassing EMX-dependent dry inhibition and inducing sarcomas. [ 38 ] The above studies are consistent with our findings: EMX is highly expressed in colorectal cancer and OS, and the higher the EMX, the worse the prognosis.

Despite the rigorous bioinformatics analysis, there are still some shortcomings in this paper. In this study, no animal experiments on gene overexpression or knockout were conducted to further verify its function. Future research endeavors should prioritize conducting such experimental validations to enhance the robustness of the findings.

23. Conclusion

In summary, PAX6, TRPA1, BCL11B, MCOLN2, CUX1, and EMX1 are highly expressed in colorectal cancer and OS, and may play a significant role in the development of colorectal cancer and OS through inflammatory effects and immune cell regulation. PAX6, TRPA1, BCL11B, MCOLN2, CUX1, and EMX1 may serve as molecular targets for the early diagnosis and precise treatment of colorectal cancer and OS. These findings offer valuable insights into the underlying mechanisms, providing a foundation for future mechanistic research in this field.

Author contributions

Conceptualization: Jie Zhang, Qiang Gao, Meiliang Zheng.

Data curation: Jie Zhang, Qiang Gao, Shiyang Hou, Xiaoqian Chi, Meiliang Zheng, Qijun Zhang, Haifeng Shan, Xiaoyu Zhang, Chunbo Kang.

Formal analysis: Jie Zhang, Qiang Gao, Shiyang Hou, Xiaoqian Chi, Meiliang Zheng, Qijun Zhang, Haifeng Shan, Xiaoyu Zhang, Chunbo Kang.

Methodology: Jie Zhang, Qiang Gao, Shiyang Hou, Xiaoqian Chi, Meiliang Zheng, Qijun Zhang, Haifeng Shan, Xiaoyu Zhang, Chunbo Kang.

Resources: Haifeng Shan.

Software: Xiaoqian Chi, Xiaoyu Zhang, Chunbo Kang.

Visualization: Haifeng Shan.

Writing – original draft: Jie Zhang, Qiang Gao, Meiliang Zheng, Xiaoyu Zhang, Chunbo Kang.

Writing – review & editing: Jie Zhang, Qiang Gao, Meiliang Zheng, Xiaoyu Zhang, Chunbo Kang.

Abbreviations: CTD Comparative Toxicogenomics Database DEGs differentially expressed genes FDR false discovery rate GO Gene Ontology GSEA gene set enrichment analysis KEGG Kyoto Encyclopedia of Gene and Genome OS Osteosarcoma PPI protein-protein interactions STRING Search Tool for the Retrieval of Interacting Genes TOM topological overlap matrix WGCNA weighted gene co-expression network analysis

JZ and QG contributed equally to this work.

The authors have no funding and conflicts of interest to disclose.

The data for this study were obtained from the GEO database and were exempt from ethical review.

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

How to cite this article: Zhang J, Gao Q, Hou S, Chi X, Zheng M, Zhang Q, Shan H, Zhang X, Kang C. Role of PAX6, TRPA1, BCL11B, MCOLN2, CUX1, EMX1 in colorectal cancer and osteosarcoma. Medicine 2024;103:5(e37056).

References

[1] Li J Ma X Chakravarti D . Genetic and biological hallmarks of colorectal cancer. Genes Dev . 2021 ; 35 : 787 – 820 . 34074695 10.1101/gad.348226.120 PMC8168558

[2] Baidoun F Elshiwy K Elkeraie Y . Colorectal cancer epidemiology: recent trends and impact on outcomes. Curr Drug Targets . 2021 ; 22 : 998 – 1009 . 33208072 10.2174/1389450121999201117115717

[3] Thanikachalam K Khan G . Colorectal cancer and nutrition. Nutrients . 2019 ; 11 : 164 . 30646512 10.3390/nu11010164 PMC6357054

[4] Dekker E Tanis PJ Vleugels J . Colorectal cancer. Lancet . 2019 ; 394 : 1467 – 80 . 31631858 10.1016/S0140-6736(19)32319-0

[5] Fan A Wang B Wang X . Immunotherapy in colorectal cancer: current achievements and future perspective. Int J Biol Sci . 2021 ; 17 : 3837 – 49 . 34671202 10.7150/ijbs.64077 PMC8495390

[6] Ni M . Update and interpretation of 2021 National Comprehensive Cancer Network (NCCN) “Clinical Practice Guidelines for Bone Tumors”]. Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi . 2021 ; 35 : 1186 – 91 . 34523287 10.7507/1002-1892.202103073 PMC8444131

[7] Chen C Xie L Ren T . Immunotherapy for osteosarcoma: Fundamental mechanism, rationale, and recent breakthroughs. Cancer Lett . 2021 ; 500 : 1 – 10 . 33359211 10.1016/j.canlet.2020.12.024

[8] Siclari VA Qin L . Targeting the osteosarcoma cancer stem cell. J Orthop Surg Res . 2010 ; 5 : 78 . 20979639 10.1186/1749-799X-5-78 PMC2988747

[9] Corre I Verrecchia F Crenn V . The osteosarcoma microenvironment: a complex but targetable ecosystem. Cells . 2020 ; 9 : 976 . 32326444 10.3390/cells9040976 PMC7226971

[10] Goh J Goh C Lim QW . Transcriptomics indicate nuclear division and cell adhesion not recapitulated in MCF7 and MCF10A compared to luminal A breast tumours. Sci Rep . 2022 ; 12 : 20902 . 36463288 10.1038/s41598-022-24511-z PMC9719475

[11] Hartsough EJ Weiss MB Heilman SA . CADM1 is a TWIST1-regulated suppressor of invasion and survival. Cell Death Dis . 2019 ; 10 : 281 . 30911007 10.1038/s41419-019-1515-3 PMC6433918

[12] Zhang C Stampfl-Mattersberger M Ruckser R . Colorectal cancer. Wien Med Wochenschr . 2023 ; 173 : 216 – 20 . 36348129 10.1007/s10354-022-00975-6

[13] Ionescu VA Gheorghe G Bacalbasa N . Colorectal cancer: from risk factors to oncogenesis. Medicina (Kaunas) . 2023 ; 59 : 1646 . 37763765 10.3390/medicina59091646 PMC10537191

[14] Cersosimo F Lonardi S Bernardini G . Tumor-associated macrophages in osteosarcoma: from mechanisms to therapy. Int J Mol Sci . 2020 ; 21 : 5207 . 32717819 10.3390/ijms21155207 PMC7432207

[15] Muratovska A Zhou C He S . Paired-Box genes are frequently expressed in cancer and often required for cancer cell survival. Oncogene . 2003 ; 22 : 7989 – 97 . 12970747 10.1038/sj.onc.1206766

[16] Simpson TI Price DJ . Pax6; a pleiotropic player in development. Bioessays . 2002 ; 24 : 1041 – 51 . 12386935 10.1002/bies.10174

[17] Tabasaran J Schuhmann M Ebinger M . PAX6 is frequently expressed in ependymal tumours and associated with prognostic relevant subgroups. J Clin Pathol . 2022 ; 75 : 759 – 65 . 34183436 10.1136/jclinpath-2021-207526 PMC9606524

[18] Kiselev Y Andersen S Johannessen C . Transcription factor PAX6 as a novel prognostic factor and putative tumour suppressor in non-small cell lung cancer. Sci Rep . 2018 ; 8 : 5059 . 29568088 10.1038/s41598-018-23417-z PMC5864921

[19] Meng Q Dai M Nie X . MicroRNA-19 contributes to the malignant phenotypes of osteosarcoma in vitro by targeting Pax6. Tumour Biol . 2018 ; 40 : 1010428317744704 . 29345189 10.1177/1010428317744704

[20] Li Y Li Y Liu Y . PAX6, a novel target of microRNA-7, promotes cellular proliferation and invasion in human colorectal cancer cells. Dig Dis Sci . 2014 ; 59 : 598 – 606 . 24185687 10.1007/s10620-013-2929-x

[21] Habgood M Seiferth D Zaki AM . Atomistic mechanisms of human TRPA1 activation by electrophile irritants through molecular dynamics simulation and mutual information analysis. Sci Rep . 2022 ; 12 : 4929 . 35322090 10.1038/s41598-022-08824-7 PMC8943162

[22] Meng J Wang J Steinhoff M . TNFα induces co-trafficking of TRPV1/TRPA1 in VAMP1-containing vesicles to the plasmalemma via Munc18-1/syntaxin1/SNAP-25 mediated fusion. Sci Rep . 2016 ; 6 : 21226 . 26888187 10.1038/srep21226 PMC4758037

[23] Silverman HA Chen A Kravatz NL . Involvement of neural transient receptor potential channels in peripheral inflammation. Front Immunol . 2020 ; 11 : 590261 . 33193423 10.3389/fimmu.2020.590261 PMC7645044

[24] Francesconi O Corzana F Kontogianni GI . Lipoyl-Based Antagonists of Transient Receptor Potential Cation A (TRPA1) downregulate osteosarcoma cell migration and expression of pro-inflammatory cytokines. ACS Pharmacol Transl Sci . 2022 ; 5 : 1119 – 27 . 36407953 10.1021/acsptsci.2c00114 PMC9667541

[25] Fang H Khoury JD Torres-Cabala CA . Expression pattern and diagnostic utility of BCL11B in mature T- and NK-cell neoplasms. Pathology (Phila) . 2022 ; 54 : 893 – 9 . 10.1016/j.pathol.2022.04.012 35864006

[26] Vanvalkenburgh J Albu DI Bapanpally C . Critical role of Bcl11b in suppressor function of T regulatory cells and prevention of inflammatory bowel disease. J Exp Med . 2011 ; 208 : 2069 – 81 . 21875956 10.1084/jem.20102683 PMC3182057

[27] Avram D Califano D . The multifaceted roles of Bcl11b in thymic and peripheral T cells: impact on immune diseases. J Immunol . 2014 ; 193 : 2059 – 65 . 25128552 10.4049/jimmunol.1400930 PMC4135393

[28] Alharbi AF Parrington J . Endolysosomal Ca(2+) Signaling in Cancer: the Role of TPC2, From Tumorigenesis to Metastasis. Front Cell Dev Biol . 2019 ; 7 : 302 . 31867325 10.3389/fcell.2019.00302 PMC6904370

[29] Li P Gu M Xu H . Lysosomal Ion channels as decoders of cellular signals. Trends Biochem Sci . 2019 ; 44 : 110 – 24 . 30424907 10.1016/j.tibs.2018.10.006 PMC6340733

[30] Yu H Xie M Meng Z . Endolysosomal ion channel MCOLN2 (Mucolipin-2) promotes prostate cancer progression via IL-1β/NF-κB pathway. Br J Cancer . 2021 ; 125 : 1420 – 31 . 34548638 10.1038/s41416-021-01537-0 PMC8575885

[31] Cai Y Yang Y Zhang X . TRPM2-AS promotes the malignancy of osteosarcoma cells by targeting miR-15b-5p/PPM1D axis. Cell Cycle . 2022 ; 21 : 835 – 50 . 35100080 10.1080/15384101.2022.2033414 PMC8973373

[32] Sacks D Baxter B Campbell B . Multisociety consensus quality improvement revised consensus statement for endovascular therapy of acute ischemic stroke. Int J Stroke . 2018 ; 13 : 612 – 32 . 29786478 10.1177/1747493018778713

[33] Krug S Weissbach J Blank A . CUX1-Transcriptional Master regulator of tumor progression in pancreatic neuroendocrine tumors. Cancers (Basel) . 2020 ; 12 : 1957 . 32707646 10.3390/cancers12071957 PMC7409270

[34] Vadnais C Shooshtarizadeh P Rajadurai CV . Autocrine Activation of the Wnt/β-Catenin Pathway by CUX1 and GLIS1 in Breast Cancers. Biol Open . 2014 ; 3 : 937 – 46 . 25217618 10.1242/bio.20148193 PMC4197442

[35] Feng F Zhao Z Zhou Y . CUX1 facilitates the development of oncogenic properties Via Activating Wnt/β-Catenin Signaling Pathway in Glioma. Front Mol Biosci . 2021 ; 8 : 705008 . 34422906 10.3389/fmolb.2021.705008 PMC8377541

[36] Bordogna W Hudson JD Buddle J . EMX homeobox genes regulate microphthalmia and alter melanocyte biology. Exp Cell Res . 2005 ; 311 : 27 – 38 . 16197942 10.1016/j.yexcr.2005.08.013

[37] Jimenez-García MP Lucena-Cacace A Otero-Albiol D . Regulation of sarcomagenesis by the empty spiracles homeobox genes EMX1 and EMX2. Cell Death Dis . 2021 ; 12 : 515 . 34016958 10.1038/s41419-021-03801-w PMC8137939

[38] Jimenez-García MP Lucena-Cacace A Otero-Albiol D . Empty spiracles homeobox genes EMX1 and EMX2 regulate WNT pathway activation in sarcomagenesis. J Exp Clin Cancer Res . 2021 ; 40 : 247 . 34364391 10.1186/s13046-021-02048-9 PMC8348834