Identification of a Potential PGK1 Inhibitor with the Suppression of Breast Cancer Cells Using Virtual Screening and Molecular Docking

Abstract

Background/objectives: Breast cancer is the second most common malignancy worldwide and poses a significant threat to women's health. However, the prognostic biomarkers and therapeutic targets of breast cancer are unclear. A prognostic model can help in identifying biomarkers and targets for breast cancer. In this study, a novel prognostic model was developed to optimize treatment, improve clinical prognosis, and screen potential phosphoglycerate kinase 1 (PGK1) inhibitors for breast cancer treatment.

Methods: Using data from the Gene Expression Omnibus (GEO) database, differentially expressed genes (DEGs) were identified in normal individuals and breast cancer patients. The biological functions of the DEGs were examined using bioinformatics analysis. A novel prognostic model was then constructed using the DEGs through LASSO and multivariate Cox regression analyses. The relationship between the prognostic model, survival, and immunity was also evaluated. In addition, virtual screening was conducted based on the risk genes to identify novel small molecule inhibitors of PGK1 from Chemdiv and Targetmol libraries. The effects of the potential inhibitors were confirmed through cell experiments.

Results: A total of 230 up- and 325 down-regulated DEGs were identified in HER2, LumA, LumB, and TN breast cancer subtypes. A new prognostic model was constructed using ten risk genes. The analysis from The Cancer Genome Atlas (TCGA) indicated that the prognosis was poorer in the high-risk group compared to the low-risk group. The accuracy of the model was confirmed using the ROC curve. Furthermore, functional enrichment analyses indicated that the DEGs between low- and high-risk groups were linked to the immune response. The risk score was also correlated with tumor immune infiltrates. Moreover, four compounds with the highest score and the lowest affinity energy were identified. Notably, D231-0058 showed better inhibitory activity against breast cancer cells.

Conclusions: Ten genes (ACSS2, C2CD2, CXCL9, KRT15, MRPL13, NR3C2, PGK1, PIGR, RBP4, and SORBS1) were identified as prognostic signatures for breast cancer. Additionally, results showed that D231-0058 (2-((((4-(2-methyl-1H-indol-3-yl)-1,3-thiazol-2-yl)carbamoyl)methyl)sulfanyl)acetic acid) may be a novel candidate for treating breast cancer.

Keywords: breast cancer; phosphoglycerate kinase 1 (PGK1); prognostic model; virtual screening.

Figure 1

Identification and functional enrichment analysis of DEGs. (A–D) Top 20 up-regulated and down-regulated genes in HER2 (A), LumA (B), LumB (C), and TN (D) subtype tumor samples from GSE29431, GSE38959, GSE45827, GSE65194, and GSE115275 datasets. The red color represents up-regulated genes, while green indicates down-regulated genes. The numbers shown in the figure represent the log fold change (logFC) of genes in each dataset. The cutoff criteria are p < 0.05 and |logFC| > 0.5. (E) The Venn diagram of DEGs of HER2, LumA, LumB, and TN subtype tumor samples from GSE29431, GSE38959, GSE45827, GSE65194, and GSE115275 datasets. (F) The bar plot of GO functional enrichment analysis. The top 10 terms of biological process, cellular component, and molecular function are shown. (G) The bar plot illustrates the results of KEGG functional enrichment analysis.

Figure 2

Analysis of the prognostic model in BC. (A) Forest plot of the signature risk model. (B) Lasso model for screening the key genes. (C) Multivariate Cox analysis confirming hub genes for risk model. (D) The expression levels of ten hub genes in breast cancer tissues compared to normal tissues. (E,G,I) Kaplan–Meier analysis of survival differences between high-risk and low-risk groups in training (E), test (G), and entire (I) sets. (F,H,J) Receiver operating characteristic (ROC) curve analysis on the ten model gene signatures in the training (F), test (H), and entire (J) sets. AUC, the area under the curve. These curves are performed by R package survival ROC. (K) Univariate Cox analysis of risk score and clinicopathological features in the entire set. (L) Multivariate Cox analysis of clinicopathological features and risk score in the entire set. (M) The ROC curve of the risk score and clinical characteristics. (N) The ROC curve and AUC values for the predictive signature at 1-year, 3-year, and 5-year survival rates.

Figure 3

Analysis of the relationships between risk score and clinical characteristics of breast cancer in the TCGA cohort. (A) Heat map of ten model genes and clinical characteristics in the high- and low-risk groups. *, p < 0.05; **, p < 0.01; and ***, p < 0.001. (B) Analysis of overall survival in TCGA-BC patients based on clinical stratification, focusing on high- and low-risk groups by age, clinical stage, N stage, and T stage.

Figure 4

The nomogram in predicting overall survival of breast cancer. (A) The nomogram predicts 1-, 3-, and 5-year overall survival. (B) Calibration maps were utilized to predict survival rates at 1, 3, and 5 years.

Figure 5

Analysis of functional enrichment across different risk groups. (A) Volcano chart of differentially expressed genes; (B) GO analysis explored the potential function in terms of biological process (BP), cellular component (CC), and molecular function (MF); (C) KEGG analysis showed the potential pathway enrichment; (D) GSEA analysis demonstrated the potential activated and suppressed pathway enrichment in the high-risk group compared with the low-risk group.

Figure 6

Immune features analysis in risk groups. (A,B) ssGSEA (single-sample gene set enrichment analysis) scores for immune cells (A) and immune function (B) in TCGA cohort. (C) The expression of immune checkpoint-related genes and the correlation between risk scores. aDCs, activated dendritic cells; APC, antigen-presenting cell; CCR, chemokine receptor; HLA, human leukocyte antigen; iDCs, immature dendritic cells; IFN, interferon; MHC, major histocompatibility complex; NK, natural killer; pDCs, plasmacytoid dendritic cells; Tfh, T follicular helper; Th, T helper cell; TIL, tumor-infiltrating lymphocyte; Treg, T regulatory cell. * p < 0.05; ** p < 0.01; *** p < 0.001; ns, non-significant.

Figure 7

Three-dimensional interaction between PGK1 (2X13) and D715-2871 (A), Y040-8304 (B), D715-0344 (C), and D231-0058 (D). Yellow dotted lines represent hydrogen bonds, pinkish-red dotted lines represent salt bridges, and green balls depict magnesium ions.

Figure 8

Inhibitory activity of D715-2871, Y040-8304, D715-0344, and D231-0058 against breast cancer cells T-47D and MCF-7. (A) CCK8 assay for cell viability. Cancer cells were treated with D715-2871, Y040-8304, D715-0344, or D231-0058 (0, 0.1, 1, 10, and 100 μg/mL) for 24 h. Data were presented as mean ± SD (n = 6). (B) CCK8 assay for cell viability. Cancer cells were treated with D231-0058 (0, 1, 3, 10, 30, and 100 μg/mL) for 24 and 48 h. Data were presented as mean ± SD (n = 6). (C) Microscopic observation of the cells treated with D231-0058 (0, 1, 3, 10, 30, and 100 μg/mL) for 24 h.