Clinicopathological Significances of Cancer Stem Cell-Associated HHEX Expression in Breast Cancer

Abstract

Aberrant expression of the transcription factor hematopoietic ally expressed homeobox/proline-rich homeodomain (HHEX/PRH) is implicated in numerous cancers. However, the association of HHEX with breast cancer (BC) remains unclear. In this study, HHEX mRNA and protein expression were analyzed using the Oncomine, UALCAN, GEPIA, TCGAportal, and HPA databases. We evaluated the effect of HHEX on clinicopathological parameters using Kaplan-Meier plotter, OncoLnc, TCGAportal, PROGgeneV2, and BC-GenExMiner. Western blotting was performed to compare the level of HHEX in breast samples of Tientsin Albino 2 mice, human breast precancerous lesions, benign breast tumors, and BC. The correlation between HHEX and cancer stem cells was investigated using the GEO (GSE52327 and GSE94865) and GEPIA datasets. Networks between HHEX and survival-related gene marker sets and microRNAs were analyzed using GEPIA, StarBase, and Cytoscape. Results of this study showed that HHEX expression in BC was significantly lower than those in breast precancerous lesions and benign breast tumors at both mRNA and protein levels. BC patients with lower HHEX expression had significantly worse overall survival and disease-free survival. Moreover, HHEX significantly affected the clinicopathology of BC. Specifically, low HHEX expression was correlated with the following groups of patients: age ≤51 years, ER-negative or PR-negative patients, HER-2 positive, triple-negative breast cancer, and basal-like BC. Immunohistochemical analysis of the breast samples showed significant differences of HHEX staining index (P < 0.001) among the three groups. To further investigate the mechanism, we determined the intersection of differentially expressed genes related to BC stem cells and those genes after HHEX expression was altered. This led to the identification of four potentially regulated genes-CXL12, BLNK, PAG1, and LPXN. Using StarBase and km-plotter, the negative regulation of HHEX expression and survival trends, including miR-130b, miR-30e, and miR-301b were joined into miRNA-HHEX-mRNA potential regulatory network. The abilities of proliferation, migration and invasion increased in MDA-MB-231 and BT-549 breast cancer cell lines after HHEX down expression and decreased after HHEX overexpression compared them in the control cells. In conclusion, these data suggest that HHEX expression is downregulated in BC and HHEX may regulate the development of BC through the stem cell-related genes.

Keywords: HHEX; breast cancer; cancer stem cells; miRNA-HHEX-mRNA network; prognosis.

Figure 1

Expression levels of HHEX in pan-cancers and breast cancer (BC) patients. (A,B) Transcriptional levels of HHEX in different types of cancers determined using the Oncomine and UALCAN databases. Cell color was determined using the best gene rank percentile for the analyses within the cell, and the gene rank was analyzed using percentile of the target gene in the top of all genes measured in each study. (C–F) mRNA levels of HHEX in BC determined using Oncomine.

Figure 2

HHEX expression in human breast cancer (BC) tissues. (A) HHEX protein expression in BC tissues determined using HPA. (B) Expression levels of HHEX. (a) Expression levels of HHEX by western blot analysis in TA2 mice with different numbers of pregnancies and SBC. (b) Histogram shows the quantitative results of the HHEX expression in breast tissue TA2 mice with different numbers of pregnancies and SBC. (C) Immunohistochemical (IHC) staining of HHEX expression in human breast tumors (200×). (a–c) HHEX expression in group I–III, respectively. (D) Interquartile spacing diagram showing the quantitative results of the HHEX expression in the three groups. a: precancerous lesions; b: breast cancer; c: benign breast tumors. n, breast tissues of virgin TA2 mice; 0, breast tissues of TA2 mice without pregnancy; 2, breast tissues of TA2 mice with 2 pregnancies; 4, breast tissues of TA2 mice with 4 pregnancies; 6, breast tissues of TA2 mice with 6 pregnancies;. SBC, spontaneous breast cancer. Statistically significant differences are indicated: **P < 0.001.

Figure 3

Prognostic value of HHEX in breast cancer (BC) and relationship between HHEX and clinicopathological parameters of BC patients. (A) Kaplan–Meier plotter, GEPIA, TCGAportal, and OncoLnc reveal the prognostic value of HHEX in BC. (a) Kaplan–Meier plotter reveals the overall survival curves based on the mRNA level of HHEX in BC patients. (b) Kaplan–Meier plotter reveals relapse-free survival curves based on the mRNA level of HHEX in BC patients. (c) Overall survival curves of HHEX in BC patients determined using GEPIA. (d) Overall survival curves of HHEX in BC patients determined using TCGAportal. (e) Overall survival curves of HHEX in BC patients determined using OncoLnc. (B) Survival curves based on protein level of HHEX in BC patients determined using HPA. (C) Relationship between mRNA levels of HHEX with subclasses of BC determined using UALCAN. (D) Correlation between HHEX and the pathological stage of BC patients determined using GEPIA. (E) Relationship between mRNA levels of HHEX and clinicopathological parameters of BC patients determined using bc-GenExMiner. **P < 0.05, ***P < 0.001.

Figure 4

Promoter methylation and genetic alterations of HHEX and potential regulatory mechanism of HHEX in regulating breast cancer (BC). (A) Promoter methylation and genetic alterations of HHEX in BC patients. (a) Promoter methylation levels of HHEX in BC determined using UALCAN. (b) Relationship between nodal metastasis status of BC with the promoter methylation of HHEX determined using UALCAN. (c) OS for HHEX alterations was analyzed using cBioportal. (B) Genetic alterations of HHEX in BC patients determined using cBioportal. (C) Venn of the DEGs related to BC stem cells and the DEGs after HHEX changed. (D) KEGG pathway enrichment of the 15 common genes. (E) The neighbor gene network of 15 genes was constructed using STRING. ***P < 0.001.

Figure 5

Potential regulatory mechanism of HHEX in breast cancer (BC). (A,B) Kaplan–Meier plotter reveals the overall survival and Relapse-free survival curves based on the mRNA level of CXL12, BLNK, PAG1, and LPXN in BC patients. (C) The positive correlation between HHEX expression and CXL12, BLNK, PAG1, and LPXN by GEPIA.

Figure 6

miRNA-HHEX-mRNA potential regulatory mechanism of HHEX in breast cancer (BC). (A) Overall survival curves of HHEX, miR-130b, miR-30e, and miR-301b in BC patients determined using Kaplan–Meier plotter. (B) The negative correlation between HHEX expression and miR-130b, miR-30e, and miR-301b by StarBase. (C) The miRNA-HHEX-mRNA potential regulatory network was constructed using Cytoscape. (D) The potential regulatory mechanism of HHEX in BC.

Figure 7

miRNAs negatively regulated HHEX and influenced the migration, invasion, and proliferation ability in MDA-MB-231 and BT-549. (A) HHEX expression of MDA-MB-231 and BT-549 transfected with and without miRNA inhibitors, mimics. (a) HHEX expression in MDA-MB-231. (b) HHEX expression in BT-549. (c) Histogram showed the quantitative result of the HHEX expression in MDA-MB-231 with different treatments. (d) Histogram showed the quantitative result of the HHEX expression in BT-549 with different treatments. (B) Plate clone formation experiment was used to compare the proliferation ability of MDA-MB-231 transfected with and without miRNA inhibitors, mimics. (C) Plate clone formation experiment was used to compare the proliferation ability of BT-549 transfected with and without miRNA inhibitors, mimics. (D) Transwell migration assay showed the cells migration capacity of MDA-MB-231 transfected with and without miRNA inhibitors, mimics (100×). (E) Transwell migration assay showed the cells migration capacity of BT-549 transfected with and without miRNA inhibitors, mimics (100×). (F) Transwell invasion assay of MDA-MB-231 cells transfected with and without miRNA inhibitors, mimics (100×). (G) Transwell invasion assay of BT-549 cells transfected with and without miRNA inhibitors, mimics (100×). NC, normal control; M1, miR-130b mimics; M2, miR-30e mimics; M3, miR-301b mimics; I1, miR-130b inhibitor; I2, miR-30e inhibitors; I3, miR-301b inhibitor.