Multiomic analysis of the anoikis genes in colon cancer based on bioinformatics

doi:10.21203/rs.3.rs-2727557/v1

Download PDF

Research Article

Multiomic analysis of the anoikis genes in colon cancer based on bioinformatics

https://doi.org/10.21203/rs.3.rs-2727557/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Colon cancer is one of the most common malignant tumors at present. It was relatively common in western developed countries, and has gradually become one of the common causes of cancer death in developing countries in recent years. With the development of imaging, endoscopic technology and early screening, the mortality rate of colon cancer has decreased significantly, but the prognosis of patients with colon cancer is still poor due to certain recurrence and metastasis risks. The latest research shows that the anoikis gene also has a certain regulatory role in colon cancer. Therefore, this paper studies the screening of related genes and their interaction with immune cells and drugs based on bioinformatics.

Anoikis

Colon Cancer

Immune Cell

As one of the most common malignant tumors, colon cancer is the second and third most common cancer in women and men, respectively. [1] And the morbidity and mortality rate of women is about 25% lower than that of men. [2] Colon cancer was more common in western developed countries in the early days, and it has gradually become one of the frequent causes of cancer death in developing countries in recent years. [3] With the popularization of imaging, laparoscopic technology and early screening, [4] the mortality rate of colon cancer has been greatly reduced. [5] But the prognosis of colon cancer patients is still poor due to the risk of recurrence and metastasis. [6–7] The current treatment options for colon cancer include surgical treatment, immune targeting and chemotherapy, [8–10] but these treatments often have serious adverse reactions [8–10] and do not significantly improve the 5-year survival rate. [11] Therefore, in view of the regulatory role of anoikis in cancer, this research can be used as a guide or to develop new drugs that are safer and more effective.

Anoikis is a specific apoptotic death due to loss of cells or inappropriate adhesion, [12] which is anchored by the loss of extracellular matrix (EMC). In this process, there are both intrinsic pathways caused by mitochondrial perturbation and extrinsic pathways triggered by cell surface death receptors. Studies have shown that anoikis can cause tumor cell migration, invasion and survival by affecting the extracellular matrix. [13] Therefore, it has been reported that anoikis of tumor cells is a potential cancer therapy. [14]

1.1 Collection and processing of colon cancer gene data

Colon cancer-related transcriptome, clinical, mutation, and copy number data were obtained from the TCGA database (https://portal.gdc.cancer.gov/) and sorted out. We searched “colon cancer survival” in the GEO database (https://www.ncbi.nlm.nih.gov/geo/) and defined the data type (expression profiling by array) and species (homo sapiens) to obtain gene expression and clinical data. The probe matrix was converted into a gene expression matrix by R language, and the expression matrix of the GEO database was obtained. Then the data was annotated with gene symbols, and the TCGA data and GEO data were merged. We used R language (“maftools” package) and Perl language for differential analysis to find genes with differential expression between tumor samples and normal samples, and finally used R language for visual analysis.

1.2 Analysis of anoikis gene mutation, copy gene variation, differential survival and prognosis in colon cancer

We downloaded the data of anoikis by searching in the genecards database, and screened out the most related genes with anoikis through EXCEL. We also downloaded the anoikis gene set in the Harmonizone database, and combined the two data sets for the gene expression level. Then we searched the UCSC Xena (https://xena.ucus.edu/) database and download the data related to the copy number of colon cancer. The Perl language was used to extract the copy number of anoikis-related genes in the sample, and R language was utilized to extract the sample cell number, thus we could have the results showing the copy number of anoikis-related genes in the sample cells and draft a copy-number frequency map and a circle map. The sample expression data in the TCGA and GEO databases were merged using the R language “limma” package to extract the expression of anoikis genes. Then the data of survival time and survival status in the two databases were merged, with the R language (“survival” and “survminer” packages), and we performed and visualized survival and prognosis analysis.

1.3 Typing of colon cancer cell anoikis and genotype survival analysis, gene set variation analysis (GSVA), immune cell difference analysis (single sample gene enrichment analysis, ssGSEA), principal component analysis (pPCA) and difference analysis

According to the expression of anoikis genes, the “consensuclusterplus” of R language was used to cluster and type the anoikis genes of colon cancer cells, in which the clustering algorithm was K-means, and the distance was Euclidean distance. Then we used “RcolorBrewer”, “GSVA”, and “ARGcluster” in R language to perform survival analysis, anoikis gene expression analysis, GSVA and PCA and visualized the anoikis genes of the two groups of cells after typing. Finally, we conducted differential analysis of the two groups in terms of their gene expression and obtained the differential gene between them.

1.4 Construction of prognosis model, prognosis analysis and its visualization

(1) We randomly divided the set of single-factor significant gene expression data and clinical survival data into training group (train) test group (test) group, and constructed the regression model of lasso algorithm of gene expression, survival time and state (least absolute shrinking and selection operator, Lasso), with which we conducted cross-validation. The gene corresponding to the point with the smallest error was selected as the significant gene of the Lasso regression model. Combined with the significant gene and its expression, we constructed a COX regression model.

(2) The model calculation formula was constructed, that is, to multiply the gene expression amount by the coefficient and then sum. Then we calculated the risk score of subtypes A and B according to the formula, and distinguished high- and low-risk groups according to the median score. In other words, the risk score was calculated by the formula, and the high- and low-risk groups of the verification group were distinguished according to the median of the training group. Survival difference analysis was performed, and P value of difference, receiver operating characteristic curve (ROC), risk curve, independent prognostic analysis and nomogram were obtained.

1.5 Analysis of immune cell infiltration and immune cell content

We used the R language CIBERSORT to obtain the relative content of immune cells with a total of 1 by combining the data obtained from the TCGA and GEO databases in Chap. 1.2. Then we filtered the data and results of the relative content of immune cells according to P < 0.05, and compared them with the results of the “risk score” in Chap. 1.5 to get the intersection. Finally, all the immune cells were calculated in a loop to obtain the correlation results of risk score, genes and immune cells.

1.6 Analysis on drug sensitivity and tumor microenvironment

(1) With the “limma”, “oncopredict” and “parallel” packages of the R language, we analyzed the data obtained from the TCGA and GEO databases obtained in Chap. 1.2, so as to get the scores of drug sensitivity by the “limma”, “ggplot2” and “ggpubr” packages of the R language. That was how the drug sensitivity of high- and low-risk groups and presented by boxplots.

(2) After getting the union of two datasets in Chap. 1.2, which were obtained from TCGA and GEO databases, we used the “estimate” package of the R language to analyze the purity of stromal cells, immune cells, and tumors and to score the tumor microenvironment. The differences in the tumor microenvironment were analyzed by the score results and the high- and low-risk groups obtained from Chap. 1.5. These results were presented by violin plots.

1.7 Single cell analysis

We defined the data type (CRC) and species (human) in the TISCH2 database (http://tisch.comp-genomics.org/) to find the database related to colon cancer, and then according to the colon-cancer anoikis found in Chap. 1.5 for death-related genes, data analysis of individual genes was performed for differences in immune cells.

We use Flow Path figure (Figure.8) to present the idea of the article.

2.1 Data collection and processing results

By searching TCGA and GEO databases, we got the TCGA-colon cancer samples included 39 normal samples and 398 tumor samples; and GEO data (GSE39582) included 566 tumor samples. The data of the two databases were merged and differentially analyzed, and the genes related to colon cancer were analyzed with the “limma” and “pheatmap” packages of R language. A total of 156 genes related to colon cancer were obtained and visualized (Fig. 1), among which up-regulated genes accounted for 69% and down-regulated genes accounted for 31%.

2.2 Processing, copy number variation analysis, differential analysis, survival analysis and prognosis analysis of anoikis gene data of colon cancer cell

Anoikis-related genes were obtained by searching “anoikis” in the genecards database and Harmonizone database, and the two data were combined to obtain a total of 516 genes related to anoikis, which were related to colon cancer obtained in Chap. 2.1. A total of 12 related genes were got through combined screening of genes. Figure 2-A shows that most anoikis genes have changes in copy number, and the frequency of copy number gain is slightly greater than that of deletion. The TCGA and GEO database data were combined for survival analysis, as shown in Table 1. We represent it in Figrue 2-B with a boxplot, and blue means low risk genes and red means high risk genes. Among them, hazard ratio (HR) > 1 is labeled as high-risk gene, and P < 0.05 indicates that the gene has a significant difference in prognosis (Fig. 2-C). The results of survival analysis in Table 1 show that the genes related to prognosis include NAT1, CDKN2A, IGF1, TIMP1, HOTAIR, SNAI1, UCHL1, MUC4, NOTCH3, PAK3, INHBB, and LTB4R2, most of which are high-risk genes. .Then we constructed the prognosis network map, as shown in Fig. 2-C, the purple nodes represent high-risk genes, the green nodes represent low-risk genes, and the size of the nodes represents the P value. The larger the node, the more likely it is a prognostic gene. The red line represents the positive regulatory relationship between the two genes, and the blue line represents the negative regulatory relationship. It can be seen from the connection that the two interacts closely with each other for expression. In addition, according to immunohistochemical staining test, in Human Protein Atlas (HPA), tumor is significantly more stained in COAD(Fig. 2-D).

2.3 Anoikis typing of colon cancer cells, survival analysis of typing genes, and differential analysis results of GSVA, ssGSEA, and PCA

By performing cluster analysis on the expression data of anoikis-related genes, the samples can be divided into subtypes A and B (Fig. 3-A). PCA analysis showed (Fig. 3-B) that A and B subtypes could be clearly distinguished according to gene expression. The survival analysis of the two subtypes, as shown in Fig. 3-C, shows that there is a statistical difference (P < = 0.05), and the B subtype has a better prognosis. As shown in Fig. 3-D, it can be seen that subtype A is mostly highly expressed, while subtype B is the opposite. Through GSVA analysis, subtype B mainly showed high expression of immune-related pathways, and type B mainly presented low expression (Fig. 3-E). According to the ssGSEA analysis (Fig. 3-F), there were 18 immune cells that were statistically different between the two subtypes, among which CD56bright.natural.killer.cell, CD56dim.natural.killer.cell, Eosinophil.immune.dendritic.cell, Macrophage, Mast.cell, Natural.killer.T.cell, Natural.killer.cel,l Regulatory.T.cell, T.follicular.helper.cell, and Type.1.T.helper.cell were high in the A subtype. Meanwhile, Activated.B.cell, Activated.CD4.T.cell, Activated.CD8.T.cell, Neutrophil, Plasmacytoid.dendritic.cell, Type.17.T.helper.cell, and Type.2.T.helper.cel were high in subtype B.

2.4 Construction of prognostic model, prognostic analysis and visualization results

(1) The prognostic model Lasso regression results and cross-validation results are shown in Fig. 4-A and B, and the COX regression model obtains the risk values of the two groups and the overall. The process of constructing prognostic model is shown in Fig. 4-C, that is: firstly according to the expression of anoikis-related genes, we divided the samples into subtypes A and B, from which we found out the differential genes through differential analysis; then we performed survival analysis on the differential genes to find out the prognosis related differential genes and built a prognosis-related model, according to which the patients were divided into high-risk and low-risk groups; inside type A is mostly a low-risk and last surviving sampleA small number of typed B is a low-risk and last surviving sample(Figure 4-C)and finally the survival period of the patients was predicted according to the prognostic model.

(2) The differential analysis results of the risk scores of anoikis subtypes showed that there was a statistically significant difference between subtypes A and B: the risk score of subtype B was higher, as shown in Fig. 4-D.

(3) According to the survival curve (Fig. 4E-G), regardless of the training group, verification group or the overall group (the sum of the training group and the verification group), there were significant differences in the survival time of the high- and low-risk groups. It shows that the constructed prognostic model can distinguish patients in high- and low-risk groups. And the ROC curve (Fig. 4H-J) shows that the area under the line of the training group is the largest, while the area of the verification group is the smallest. Therefore, the prognosis model bears with the greatest accuracy in terms of predicting survival-period training, while the verification the smallest. Regardless of the training group, the verification group or the overall group, the area below the line is greater than 0.5, so the accuracy of the prognosis model in predicting survival is high. Compared with other clinical traits, risk score as an independent analysis factor, P < 0.05 indicates that it can be used as an independent factor independent of others (Fig. 4-N). The nomogram can predict the survival of patients with different clinical characteristics of the prognostic model. From Fig. 4-O, it can be seen that women of 65 years old and with high risk scores have poor prognosis, and their 1-year, 3-year, and 5-year survival rates are respectively 94.3%%, 87.2%, 77.5%. Figure 4-P shows the accuracy of nomogram survival prediction. In Fig. 4-Q, We know red means high risk and blue means low risk, so, like TIMP1, SNAI1 and PAK3 are high risk genes. From the DCA curves (Fig. 4K-M), it was observed that the nomogram and risk score were superior to other clinical traits as an assessment of disease risk.

2.5 Analysis results of immune cell infiltration and immune cell content

By CIBERSORT calculation, the relative content of each type of immune cell in each sample was obtained as the result of immune cell infiltration. The results were used to further draw the correlation between risk score and immune cells, among which there was a significant negative correlation with T cell memory resting, B cell memory, plasma cells, and mast cells activated. There was a significant positive correlation with macrophages M0, macrophages, and mast cells resting, as shown in Fig. 5-A. The correlation between the genes involved in the model construction and immune cells is shown in Fig. 5-B. Most of the genes were correlated with immune cells (P < 0.05), and there were more negative correlations than positive correlations.

2.6 Results of drug sensitivity analysis and tumor microenvironment analysis

(1) Drug sensitivity analysis: first we obtained the drug sensitivity score, and analyzed the sensitivity of each drug in the high- and low-risk groups. The results are presented in Fig. 6-A, which showed that among the 42 drugs, 12 of them are more sensitive in the high-risk group, namely Dasatinib, BMS-754807, AZD8055, AZD1332, AZ960, IGF1R-3801, Tozasewrtib, Sepantronium, NU7441, Luminespib, JQ1, and UMI-77, accounted for 28.57%; while the remaining 28 drugs were more sensitive in the low-risk group, accounted for 66.67%.

(2) Tumor microenvironment analysis: firstly, according to the scoring results of clinical samples in the two databases of TCGA and GEO, the tumor microenvironment was scored by three aspects: stromal cell score, immune cell score, and total score. Then the scores were analyzed according to the high- and low-risk groups. The Fig. 6-B shows that the scores of stromal cells, immune cells and total scores were statistically different between the high- and low-risk groups (P less than 0.05), and the tumor microenvironment scores were higher in the high-risk group.

2.7 Results of single cell analysis

After searching in the TISCH2 database, we drew Fig. 7 According to the content of genes related to anoikis in colon cancer in each immune cell. Figure 7-A shows the grouping of cells, Fig. 7-B annotates the groups corresponding to different immune cells, Fig. 7-C performs data analysis on the total expression of immune cells and represents it in pie charts and histograms, and Fig. 7-D shows the contents of genes with significant differences in expression in anoikis in different immune cells. It was found that PAK3 had the highest content in Treg, as HOTAIR in CD8Tex, MUC4 in CD4Tconv, and SNAI1 in CD8T.

Anoikis is a specific form of apoptosis associated with the loss of interaction or attachment of cells to their adjacent extracellular matrix (ECM). Anoikis plays a crucial role in preventing adhesion-independent cell growth and cell adhesion to inappropriate ECM proteins. [15] Anoikis is currently related to lung cancer, liver cancer, gastric cancer, prostate cancer and other diseases, [16–20] but there are few studies it’s the relationship with colon cancer and there is a lack of relevant informatics research involving drug sensitivity and prognosis. Therefore, this study conducts a research on the relationship between the three to some extent. Firstly, according to the comparative study between the tumor group and the normal group, the significantly different genes were found, and then the cross-comparison analysis was performed with the genes related to anoikis. It can be seen that TIMP1, LTB4R2, PAK3, HOTAIR, MUC4, and SNAI1 genes were high-risk genes, and the mutation rate was higher than other genes. Some studies believed that the relationship between anoikis and cancer was complex, and that anoikis in different tissues and genetic backgrounds had different effects on cancer. [16] Studies have shown [21] that in melanoma, TIMP1 was assembled in a supramolecular complex containing CD63 and β1-integrin, and confered resistance to anoikis by activating the PI3-K signaling pathway; the study by Jin-Wook Lee et al. [22] showed that LTB4R2 was observed in prostate cancer. Ther C Workman ectopic expression of LTB4R2 in normal prostate PWR-1E cells made the cells resistant to anoikis, and after detachment significantly reduced apoptotic cell death. The research by Chen S et al. [23] showed that p110C could bind to PAK family, and at the same time, the activity of PAK family is inhibited, thereby suppressed the activity during anoikis, which may be one of the reasons why p110C induced apoptosis. In the study of Dai ZY et al. [24], in ovarian cancer, Western blot analysis of anoikis-related markers showed that N-cadherin, ZEB1 and TWIST1 were down-regulated after inhibiting HOTAIR, while E-cadherin Protein and ErbB3 were upregulated to promote cancer cell migration, invasion and chemoresistance. In the study of MUC4 by Heather C Workman et al. [25], it was known that shRNA-mediated knockdown of MUC4 in breast cancer reduced the migration, proliferation and anoikis resistance of JIMT-1 cells, strongly indicating that MUC4 expression positively contributed to breast tumor metastasis-associated cellular properties. According to Yiju Lee’s study on melanoma, [26] it was found that the high expression of MIST1 and SNAI1 both helped melanoma cells to bypass anoikis, and MIST1 hijacked the PTEN/AKT signaling pathway by directly regulating SNAI1, affecting anoikis resistance of melanoma cells. In general, anoikis was not widely studied in colon cancer, but the same gene has a complex relationship with anoikis mechanism in other cancers and is related to prognosis.

By clustering and typing the different expressions of anoikis in colon cancer, it was found that the survival outcome of subtype B was better than that of subtype A, and the expression of anoikis genes in subtype B was low. GSVA and ssGSEA analysis showed that inflammation-related pathways and immune cells were enriched in subtype A, which was characterized by high expression of anoikis genes, immune activation and immune cell infiltration, and no survival advantage, while subtype B was the opposite. By constructing a prognostic model and performing a risk score, it was concluded that the high-risk group had a poor prognosis, and the low-risk group had low expression of anoikis genes, with high accuracy. Regardless of immune infiltration or tumor microenvironment analysis, the same conclusion was drawn, that is, the high-risk group had poorer prognosis, high expression of anoikis, shorter survival time, and less immune cell infiltration; the low-risk group was on the contrary, which also reflected the relationship between anoikis and colon cancer, its tumor microenvironment and immune response. Studies have shown that the tumor microenvironment was closely related to the development and progression of colon cancer. [27] And it was manifested by a large number of inflammatory cell infiltration. For example, the infiltration of B cells was associated with colon cancer, and the infiltration of CD4 + and CD8 + T cells was associated with the prognosis of colon cancer. [28] Thus, anoikis was significantly correlated with colon cancer. Therefore, mining drugs that regulated anoikis target genes can provide a certain reference for the development of colon cancer therapeutic drugs.

Through the analysis of the drug sensitivity of the high-risk group, we found twelve drugs that were more sensitive in the high-risk group, namely Dasatinib, BMS-754807, AZD8055, AZD1332, AZ960, IGF1R-3801, Tozasertib, Sepantronium bromine, NU7441, Luminespib, JQ1, and UMI-77. In the study by Aaron J Scott et al. [29], it was observed that Dasatinib-sensitive CRC cell lines exhibited G1 inhibition, and the anti-tumor effect resulted from G1 cell cycle arrest, which proved that Dasatinib did have a certain effect on colon cancer. Yi Li et al. mentioned in their study on IGF1R [30] that BMS-754807 had improved radiosensitivity and reversed radiation resistance in two colorectal cancer cell lines. Regarding AZD8055, currently only related literatures can be found in renal cell carcinoma, cholangiocarcinoma and breast cancer [31]. Hiroki Kawano [32] et al. found that Sepantronium bromine had anti-tumor effects on tumors in the study of mouse tumor models. Studies have shown that [33] lumepib interacted with BSA to form drug-loaded BSA NPs (DNPs) through non-covalent reversible interactions (MCF-7) treatment, but there was no research on colon cancer. The study of Wanjing Ding, Haojian Zhang et al. [34] found that exposure to JQ1 and DHA resulted in strong synergistic antiproliferative and proapoptotic effects associated with inhibition of c-Myc expression and activation of NF-κB in a colon cancer cell line. No relevant literature was found for AZD1332, AZ960, IGF1R-3801, Tozasertib, and NU7441. Generally speaking, there are few studies on drugs sensitive to anoikis-related genes in colon cancer at present, and relevant research can be conducted more in future studies.

Finally, in this study, data analysis was performed on single cells, and it was found that the significantly expressed genes of anoikis were not the same in different immune cells: PAK3 showed the highest content in Treg, as HOTAIR in CD8Tex, MUC4 in CD4Tconv, SNAI1 in CD8T. It also illustrated the complexity of various immune cells involved in anoikis, and provided some ideas for related research.

In this study, the biological information method was used to conduct multi-omics analysis on the colon cancer data of TCGA and GEO databases and the anoikis genes of genecards and Harmonizone databases, thus we concluded that there was a significant correlation between anoikis and colon cancer. It involves mechanisms related to immune cell infiltration in the tumor microenvironment, as well as single-cell data analysis. The drugs that regulate anoikis are mainly Dasatinib, BMS-754807, AZD8055, AZD1332, AZ960, IGF1R-3801, Tozaswrib, Sepantronium, NU7441, Luminespib, JQ1, UMI-77, and savolitinib. This study is only for reference for further clinical research and related statistical analysis.

Ethics approval and consent to participate

Not applicable.

Consent to publish

Not applicable.

Availability of data and materials

The raw data used in this experiment come from TCGA website (https://portal.gdc.cancer.gov/), starbase(https://starbase.sysu.edu.cn), GEO database (https://www.ncbi.nlm.nih.gov/geo/), the UCSC Xena (https://xena.ucus.edu/) and HPA online database (https://www.proteinatlas.org/). We would like to express our sincere thanks to the collectors and producers of the research data.

Competing interests

The authors claim that this study has no potential conflicts of interest.

Funding

No funding

Authors’ Contributions

YZ contributed to study design and supervision; LH contributed to data collection, HZ contributed to data analysis and article writing; YZ revised the manuscript. All authors read and approved the final manuscript.,and all authors reviewed the manuscript.

Acknowledgements

Not applicable.

Supplementary Information

No supplementary information

Biller LH, Schrag D. Diagnosis and Treatment of Metastatic Colorectal Cancer: A Review. JAMA. 2021 Feb 16;325(7):669–685. doi: 10.1001/jama.2021.0106. PMID: 33591350.
Dekker E, Tanis PJ, Vleugels JLA, Kasi PM, Wallace MB. Colorectal cancer. Lancet. 2019 Oct 19;394(10207):1467–1480. doi: 10.1016/S0140-6736(19)32319-0. PMID: 31631858.
Baidoun F, Elshiwy K, Elkeraie Y, Merjaneh Z, Khoudari G, Sarmini MT, Gad M, Al-Husseini M, Saad A. Colorectal Cancer Epidemiology: Recent Trends and Impact on Outcomes. Curr Drug Targets. 2021;22(9):998–1009. doi: 10.2174/1389450121999201117115717. PMID: 33208072.
Nelson RS, Thorson AG. Colorectal cancer screening. Curr Oncol Rep. 2009 Nov;11(6):482-9. doi: 10.1007/s11912-009-0065-8. PMID: 19840526.
Kijima S, Sasaki T, Nagata K, Utano K, Lefor AT, Sugimoto H. Preoperative evaluation of colorectal cancer using CT colonography, MRI, and PET/CT. World J Gastroenterol. 2014 Dec 7;20(45):16964-75. doi: 10.3748/wjg.v20.i45.16964. PMID: 25493009; PMCID: PMC4258565.
Lawrence BJ, Alexander E, Grant H, O'Connor M. Colorectal cancer and absolute risks. Gut. 2021 Nov;70(11):2215–6. 10.1136/gutjnl-2020-323505. Epub 2020 Dec 24. PMID: 33361346.
Xu W, He Y, Wang Y, Li X, Young J, Ioannidis JPA, Dunlop MG, Theodoratou E. Risk factors and risk prediction models for colorectal cancer metastasis and recurrence: an umbrella review of systematic reviews and meta-analyses of observational studies. BMC Med. 2020 Jun;26(1):172. 10.1186/s12916-020-01618-6. PMID: 32586325; PMCID: PMC7318747.
Fan A, Wang B, Wang X, Nie Y, Fan D, Zhao X, Lu Y. Immunotherapy in colorectal cancer: current achievements and future perspective. Int J Biol Sci. 2021 Sep 3;17(14):3837–3849. doi: 10.7150/ijbs.64077. PMID: 34671202; PMCID: PMC8495390.
Ma YS, Li W, Liu Y, Shi Y, Lin QL, Fu D. Targeting Colorectal Cancer Stem Cells as an Effective Treatment for Colorectal Cancer.Technol Cancer Res Treat. 2020 Jan-Dec;19:1533033819892261. doi: 10.1177/1533033819892261. PMID: 32748700; PMCID: PMC7785997.
Pérez-Escalante E, Cariño-Cortés R, Fernández-Martínez E, Ortiz MI, Muñoz-Pérez VM, Sánchez-Crisóstomo I, Jiménez-Ángeles L. Colorectal Cancer: Causes and Evidence of Chemopreventive Treatments. Curr Pharm Biotechnol. 2018;19(14):1135–1155. doi: 10.2174/1389201020666181226112712. PMID: 30585544.
Voutsadakis IA. The pluripotency network in colorectal cancer pathogenesis and prognosis: an update. Biomark Med. 2018 Jun;12(6):653–665. doi: 10.2217/bmm-2017-0369. PMID: 299440
Taddei ML, Giannoni E, Fiaschi T, Chiarugi P. Anoikis: an emerging hallmark in health and diseases. J Pathol. 2012 Jan;226(2):380 – 93. doi: 10.1002/path.3000. PMID: 21953325.
Simpson CD, Anyiwe K, Schimmer AD. Anoikis resistance and tumor metastasis. Cancer Lett. 2008 Dec 18;272(2):177 – 85. doi: 10.1016/j.canlet.2008.05.029. Epub 2008 Jun 24. PMID: 18579285.
Sattari Fard F, Jalilzadeh N, Mehdizadeh A, Sajjadian F, Velaei K. Understanding and targeting anoikis in metastasis for cancer therapies. Cell Biol Int. 2022 Dec 1. doi: 10.1002/cbin.11970. Epub ahead of print. PMID: 36453448.
Santagostino SF, Assenmacher CA, Tarrant JC, Adedeji AO, Radaelli E. Mechanisms of Regulated Cell Death: Current Perspectives. Vet Pathol. 2021 Jul;58(4):596–623. doi: 10.1177/03009858211005537. Epub 2021 May 27. PMID: 34039100.
Jin L, Chun J, Pan C, Kumar A, Zhang G, Ha Y, Li D, Alesi GN, Kang Y, Zhou L, Yu WM, Magliocca KR, Khuri FR, Qu CK, Metallo C, Owonikoko TK, Kang S. The PLAG1-GDH1 Axis Promotes Anoikis Resistance and Tumor Metastasis through CamKK2-AMPK Signaling in LKB1-Deficient Lung Cancer.Mol Cell. 2018 Jan4;69(1):87–99.e7. doi: 10.1016/j.molcel.2017.11.025. Epub 2017 Dec 14. PMID: 29249655; PMCID: PMC5777230.
Song J, Liu Y, Liu F, Zhang L, Li G, Yuan C, Yu C, Lu X, Liu Q, Chen X, Liang H, Ding Z, Zhang B. The 14-3-3σ protein promotes HCC anoikis resistance by inhibiting EGFR degradation and thereby activating the EGFR-dependent ERK1/2 signaling pathway. Theranostics. 2021 Jan 1;11(3):996–1015. doi: 10.7150/thno.51646. PMID: 33391517; PMCID: PMC7738881.
Du S, Yang Z, Lu X, Yousuf S, Zhao M, Li W, Miao J, Wang X, Yu H, Zhu X, Chen H, Shi L, Xu E, Xia X, Guan W. Anoikis resistant gastric cancer cells promote angiogenesis and peritoneal metastasis through C/EBPβ-mediated PDGFB autocrine and paracrine signaling. Oncogene. 2021 Sep;40(38):5764–5779. doi: 10.1038/s41388-021-01988-y. Epub 2021 Aug 2. Erratum in: Oncogene. 2022 Feb;41(9):1384–1385. PMID: 34341514.
Yu Y, Song Y, Cheng L, Chen L, Liu B, Lu D, Li X, Li Y, Lv F, Xing Y. CircCEMIP promotes anoikis-resistance by enhancing protective autophagy in prostate cancer cells. J Exp Clin Cancer Res. 2022 Jun 2;41(1):188. doi: 10.1186/s13046-022-02381-7. PMID: 35655258; PMCID: PMC9161511.
Yu Y, Liu B, Li X, Lu D, Yang L, Chen L, Li Y, Cheng L, Lv F, Zhang P, Song Y, Xing Y. ATF4/CEMIP/PKCα promotes anoikis resistance by enhancing protective autophagy in prostate cancer cells.Cell Death Dis. 2022 Jan10;13(1):46. doi: 10.1038/s41419-021-04494-x. PMID: 35013120; PMCID: PMC8748688.
Toricelli M, Melo FH, Peres GB, Silva DC, Jasiulionis MG. Timp1 interacts with beta-1 integrin and CD63 along melanoma genesis and confers anoikis resistance by activating PI3-K signaling pathway independently of Akt phosphorylation. Mol Cancer. 2013 Mar 25;12:22. doi: 10.1186/1476-4598-12-22. Erratum in: Mol Cancer. 2015;14:161. PMID: 23522389; PMCID: PMC3635912.
Lee JW, Kim JH. Activation of the leukotriene B4 receptor 2-reactive oxygen species (BLT2-ROS) cascade following detachment confers anoikis resistance in prostate cancer cells.J Biol Chem. 2013 Oct18;288(42):30054–30063. doi: 10.1074/jbc.M113.481283. Epub 2013 Aug 28. PMID: 23986446; PMCID: PMC3798474.
Chen S, Yin X, Zhu X, Yan J, Ji S, Chen C, Cai M, Zhang S, Zong H, Hu Y, Yuan Z, Shen Z, Gu J. The C-terminal kinase domain of the p34cdc2-related PITSLRE protein kinase (p110C) associates with p21-activated kinase 1 and inhibits its activity during anoikis. J Biol Chem. 2003 May 30;278(22):20029-36. doi: 10.1074/jbc.M300818200. Epub 2003 Mar 6. PMID: 12624090.
Dai ZY, Jin SM, Luo HQ, Leng HL, Fang JD. LncRNA HOTAIR regulates anoikis-resistance capacity and spheroid formation of ovarian cancer cells by recruiting EZH2 and influencing H3K27 methylation. Neoplasma. 2021 May;68(3):509–18. 10.4149/neo_2021_201112N1212. Epub 2021 Jan 28. PMID: 33502891.
Workman HC, Miller JK, Ingalla EQ, Kaur RP, Yamamoto DI, Beckett LA, Young LJ, Cardiff RD, Borowsky AD, Carraway KL, Sweeney C, Carraway KL 3. The membrane mucin MUC4 is elevated in breast tumor lymph node metastases relative to matched primary tumors and confers aggressive properties to breast cancer cells. Breast Cancer Res. 2009;11(5):R70. 10.1186/bcr2364. PMID: 19761616; PMCID: PMC2790847.
Lee Y, Yao W, Yang C, Li Y, Ni H, Wang L, Ji B, Gu Y, Yang S. MIST1 regulates SNAI1 and acts through the PTEN/AKT signaling axis to promote anoikisresistance in human melanoma cells. Exp Ther Med. 2018 Aug;16(2):695–703. doi: 10.3892/etm.2018.6225. Epub 2018 May 29. Erratum in: Exp Ther Med. 2021 Jan;21(1):46. PMID: 30112032; PMCID: PMC6090440.
Li J, Ma X, Chakravarti D, Shalapour S, DePinho RA. Genetic and biological hallmarks of colorectal cancer. Genes Dev. 2021 Jun;35(11–12):787–820. 10.1101/gad.348226.120. PMID: 34074695; PMCID: PMC8168558.
Bai Z, Zhou Y, Ye Z, Xiong J, Lan H, Wang F. Tumor-Infiltrating Lymphocytes in Colorectal Cancer: The Fundamental Indication and Application on Immunotherapy.Front Immunol. 2022 Jan14;12:808964. doi: 10.3389/fimmu.2021.808964. PMID: 35095898; PMCID: PMC8795622.
Scott AJ, Song EK, Bagby S, Purkey A, McCarter M, Gajdos C, Quackenbush KS, Cross B, Pitts TM, Tan AC, Eckhardt SG, Fenton H, Arcaroli J, Messersmith WA. Evaluation of the efficacy of dasatinib, a Src/Abl inhibitor, in colorectal cancer cell lines and explant mouse model. PLoS One. 2017 Nov 1;12(11):e0187173. doi: 10.1371/journal.pone.0187173. PMID: 29091939; PMCID: PMC5665512.
Li Y, Lu K, Zhao B, Zeng X, Xu S, Ma X, Zhi Y. Depletion of insulin-like growth factor 1 receptor increases radiosensitivity in colorectal cancer. J Gastrointest Oncol. 2020 Dec;11(6):1135–45. 10.21037/jgo-20-210. PMID: 33456988; PMCID: PMC7807270.
Kauffman EC, Lang M, Rais-Bahrami S, Gupta GN, Wei D, Yang Y, Sourbier C, Srinivasan R. Preclinical efficacy of dual mTORC1/2 inhibitor AZD8055 in renal cell carcinoma harboring a TFE3 gene fusion.BMC Cancer. 2019 Sep13;19(1):917. doi: 10.1186/s12885-019-6096-0. PMID: 31519159; PMCID: PMC6743205.
Kawano H, Shakushiro K, Nakata M, Kita A, Maeda A, Watanabe S, Sako K, Oku N. Antitumor efficacy and biodistribution of liposomal sepantronium bromide (YM155), a novel small-molecule survivin suppressant. Eur J Pharm Biopharm. 2014 Sep;88(1):283–9. 10.1016/j.ejpb.2014.06.015. Epub 2014 Jun 30. PMID: 24993306.
K Rochani A, Balasubramanian S, Ravindran Girija A, Maekawa T, Kaushal G, Kumar DS. Heat Shock Protein 90 (Hsp90)-Inhibitor-Luminespib-Loaded-Protein-Based Nanoformulation for Cancer Therapy. Polymers (Basel). 2020 Aug 11;12(8):1798. doi: 10.3390/polym12081798. PMID: 32796651; PMCID: PMC7465148.
Ding W, Zhang H, Mei G. Synergistic antitumor activity of DHA and JQ1 in colorectal carcinoma.Eur J Pharmacol. 2020 Oct15;885:173500. doi: 10.1016/j.ejphar.2020.173500. Epub 2020 Aug 27. PMID: 32861663.

Table 1 is available in the Supplementary Files section.

No competing interests reported.

Table1.jpg

Download PDF

Version 1

posted

You are reading this latest preprint version

Multiomic analysis of the anoikis genes in colon cancer based on bioinformatics

Status:

Version 1

Abstract

Figures

Introduction

Methods

1.1 Collection and processing of colon cancer gene data

1.4 Construction of prognosis model, prognosis analysis and its visualization

1.5 Analysis of immune cell infiltration and immune cell content

1.6 Analysis on drug sensitivity and tumor microenvironment

1.7 Single cell analysis

Results

2.6 Results of drug sensitivity analysis and tumor microenvironment analysis

Discussion

Conclusion

Declarations

Ethics approval and consent to participate

Consent to publish

Funding

Supplementary Information

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1