Gene Signature Associated with Nervous System in an Experimental Radiation- and Estrogen-Induced Breast Cancer Model

Abstract

Breast cancer is frequently the most diagnosed female cancer in the world. The experimental studies on cancer seldom focus on the relationship between the central nervous system and cancer. Despite extensive research into the treatment of breast cancer, chemotherapy resistance is an important issue limiting the efficacy of treatment. Novel biomarkers to predict prognosis or sensitivity to chemotherapy are urgently needed. This study examined nervous-system-related genes. The profiling of differentially expressed genes indicated that high-LET radiation, such as that emitted by radon progeny, in the presence of estrogen, induced a cascade of events indicative of tumorigenicity in human breast epithelial cells. Bioinformatic tools allowed us to analyze the genes involved in breast cancer and associated with the nervous system. The results indicated that the gene expression of the Ephrin A1 gene (EFNA1), the roundabout guidance receptor 1 (ROBO1), and the kallikrein-related peptidase 6 (KLK6) was greater in T2 and A5 than in the A3 cell line; the LIM domain kinase 2 gene (LIMK2) was greater in T2 than A3 and A5; the kallikrein-related peptidase 7 (KLK7), the neuroligin 4 X-linked gene (NLGN4X), and myelin basic protein (MBP) were greater than A3 only in T2; and the neural precursor cell expressed, developmentally down-regulated 9 gene (NEDD9) was greater in A5 than in the A3 and E cell lines. Concerning the correlation, it was found a positive correlation between ESR1 and EFNA1 in BRCA-LumA patients; with ROBO1 in BRCA-Basal patients, but this correlation was negative with the kallikrein-related peptidase 6 (KLK6) in BRCA-LumA and -LumB, as well as with LIMK2 and ROBO1 in all BRCA. It was also positive with neuroligin 4 X-linked (NLGN4X) in BRCA-Her2 and BRCA-LumB, and with MBP in BRCA-LumA and -LumB, but negative with KLK7 in all BRCA and BRCA-LumA and NEDD9 in BRCA-Her2. The differential gene expression levels between the tumor and adjacent tissue indicated that the ROBO1, KLK6, LIMK2, KLK7, NLGN4X, MBP, and NEDD9 gene expression levels were higher in normal tissues than in tumors; however, EFNA1 was higher in the tumor than the normal ones. EFNA1, LIMK2, ROBO1, KLK6, KLK7, and MBP gene expression had a negative ER status, whereas NEDD9 and NLGN4X were not significant concerning ER status. In conclusion, important markers have been analyzed concerning genes related to the nervous system, opening up a new avenue of studies in breast cancer therapy.

Keywords: axon guidance; breast cancer; neural precursors; neuronal markers; neurotransmitter; radiation.

Figure 1

(A) Heatmap of Affymetrix array (U133A) data to allow the comparison of the gene expression of the cell lines derived from the model: MCF-10F/estrogen (Ct/E); control/Alpha3 (Ct/A3); estrogen/Alpha5 (E/A5); Alpha3/Alpha5 (A3/A5); Alpha5/Tumor2 (A5/T2); and Alpha3/Tumor2 (A3/T2) (reprinted/adapted with permission from Ref. [28], Spandidos Publications, 2013). The red color indicates a higher expression; blue indicates a lower expression; and yellow indicates an equal expression (left panel). (B) The gene box plot summarizes the range of differential gene expression in the same pairwise cell line comparisons (right panel).

Figure 2

The graphs show the profiling of differentially expressed genes obtained through an Affymetrix array (U133A), with data comparing genes such as (A) the ephrin A1 gene, EFNA1, (B) the roundabout guidance receptor 1 gene, ROBO1, (C) the kallikrein-related peptidase 6 gene, KLK6, (D) the LIM domain kinase 2 gene, LIMK2, (E) the kallikrein-related peptidase 7 gene, KLK7, (F) the neuroligin 4 X−linked gene, NLGN4X, (G) the myelin basic protein gene, MBP, and (H) the neural precursor cell expressed, developmentally down-regulated 9 gene, NEDD9 in the following cell lines: MCF-10F/Estrogen (Ct/E), Control/Alpha3 (Ct/A3), Estrogen/Alpha5 (E/A5), Alpha3/Alpha5 (A3/A5), Alpha5/Tumor2 (A5/T2), and Alpha3/Tumor2 (A3/T2). The graphs were obtained from a cluster dendrogram repository of gene expression from our laboratory for this article.

Figure 3

(A) Heatmap table and (B) box plots representing the correlation between ESR1 expression with purity adjustment (left) and the ephrin A1 gene, EFNA1; the roundabout guidance receptor 1 gene, ROBO1; the kallikrein-related peptidase 6 gene, KLK6, and the LIM domain kinase 2 gene, LIMK2 expression levels (right) in breast invasive carcinoma (BRCA) with tumor purity. Heatmap table: the red color indicates a statistically significant positive correlation, the blue color indicates a statistically significant negative correlation, and the gray denotes a non-significant result. Box plots: the correlation values for each analysis are stated in red in the top right of the box. The statistical significance (Spearman’s, p < 0.05) was estimated using TIMER2.0, accessed on 17 June 2022 [30].

Figure 4

(A) Heatmap table and (B) box plots representing the correlation between ESR1 expression with purity adjustment (left) and the kallikrein-related peptidase 7 gene, KLK7; the neuroligin 4 X−linked gene, NLGN4X; the myelin basic protein gene, MBP, and the neural precursor cell expressed, developmentally down-regulated 9 gene, NEDD9 expression levels (right) in breast invasive carcinoma (BRCA) with tumor purity. Heatmap table: the red color indicates a statistically significant positive correlation, the blue color indicates a statistically significant negative correlation, and the gray denotes a non-significant result. Box plots: the correlation values for each analysis are stated in red in the top right of the box. The statistical significance (Spearman’s, p < 0.05) was estimated using TIMER2.0, accessed on 17 June 2022 [30].

Figure 5

The differential gene expression levels between tumor and normal tissues across various cancer types. The box plots show the distribution of gene expression levels of (A) the ephrin A1 gene, EFNA1, (B) the roundabout guidance receptor 1 gene, ROBO1 (C), the kallikrein-related peptidase 6 gene, KLK6, (D) the LIM domain kinase 2 gene, LIMK2, (E) the kallikrein-related peptidase 7 gene, KLK7, (F) the neuroligin 4 X-linked gene, NLGN4X, (G) the myelin basic protein gene, MBP, and (H) the neural precursor cell expressed, developmentally down-regulated 9 gene, NEDD9 in tumors versus normal tissues (Wilcoxon test, ***: p < 0.001) estimated using TIMER2.0 in breast invasive carcinoma, accessed on 17 June 2022 [30]. (1) BRCA.Tumor (n = 1093), (2) BRCA.Normal (n = 112), (3) BRCA—Basal.Tumor (n = 190), (4) BRCA—Her2.Tumor (n = 82), (5) BRCA—LumA.Tumor (n = 564), (6) BRCA—LumB.Tumor (n = 217).

Figure 6

Box plot transcript expression of (A) the ephrin A1 gene, EFNA1; (B) the roundabout guidance receptor 1 gene, ROBO1; (C) the kallikrein-related peptidase 6 gene, KLK6; (D) the LIM domain kinase 2 gene, LIMK2; (E) the kallikrein-related peptidase 7 gene, KLK7; (F) the neuroligin 4 X-linked gene, NLGN4X; (G) the myelin basic protein gene, MBP; and (H) the neural precursor cell expressed, developmentally down-regulated 9 gene, NEDD9 in breast cancer (TCGA, n = 782) stratified by nature2012 for estrogen receptor status (one-way ANOVA, p < 0.05). Raw data were extracted from the University of California, Santa Cruz (ena.ucsc.edu). UCSC Xena functional genomics explorer (https://xenabrowser.net) was accessed on 17 June 2022 [31].

Scheme 1

Summary of the alpha model and main findings observed in this study. (1) EFNA1, ROBO1, KLK6, KLK7, LIMK2, NLGN4X, MBP, and NEDD9 gene expression in the estrogen (E), the Alpha3 (A3), the Alpha5, and the Tumor2 (T2) cell lines; (2) estrogen receptor α gene ESR1 expression in BRCA patients; (3) differential gene expression between tumor and normal tissues; (4) estrogen receptor status.