Quantitative Proteomic Profiling Identifies a Potential Novel Chaperone Marker in Resistant Breast Cancer

Abstract

Development of aromatase inhibitor resistant breast cancer among postmenopausal women continues to be a major clinical obstacle. Previously, our group demonstrated that as breast cancer cells transition from hormone-dependent to hormone-independent, they are associated with increased growth factor signaling, enhanced cellular motility, and the epithelial to mesenchymal transition (EMT). Given the complexity of cancer stem cells (CSC) and their implications on endocrine resistance and EMT, we sought to understand their contribution towards the development of aromatase inhibitor resistant breast cancer. Cells cultured three dimensionally as mammospheres are enriched for CSCs and more accurately recapitulates tumors in vivo. Therefore, a global proteomic analysis was conducted using letrozole resistant breast cancer cells (LTLT-Ca) mammospheres and compared to their adherent counterparts. Results demonstrated over 1000 proteins with quantitative abundance ratios were identified. Among the quantified proteins, 359 were significantly altered (p < 0.05), where 173 were upregulated and 186 downregulated (p < 0.05, fold change >1.20). Notably, midasin, a chaperone protein required for maturation and nuclear export of the pre-60S ribosome was increased 35-fold. Protein expression analyses confirmed midasin is ubiquitously expressed in normal tissue but is overexpressed in lobular and ductal breast carcinoma tissue as well as ER+ and ER- breast cancer cell lines. Functional enrichment analyses indicated that 19 gene ontology terms and one KEGG pathway were over-represented by the down-regulated proteins and both were associated with protein synthesis. Increased midasin was strongly correlated with decreased relapse free survival in hormone independent breast cancer. For the first time, we characterized the global proteomic signature of CSC-enriched letrozole-resistant cells associated with protein synthesis, which may implicate a role for midasin in endocrine resistance.

Keywords: aromatase inhibitors; breast cancer; cancer stem cells; letrozole resistance; midasin; translation.

Figure 1

Western Blot analysis of midasin expression. Midasin expression in letrozole-sensitive (AC-1) and letrozole-resistant (LTLT-Ca) breast cancer cells cultured adherently (2D) or as mammospheres (3D). All cells were evaluated by immunoblot to examine the expression of midasin and GAPDH (loading control). (A) Graphs depict normalized percentages of protein expression intensities relative to 2D cell counterparts. (B) Representative immunoblot depicts the protein expression of midasin and GAPDH.

Figure 2

Functional enrichment analysis of LTLT-Ca proteome. Functional enrichment analysis were performed using the DAVID tool (32), resulting in 19 gene ontology (GO) terms.

Figure 3

KEGG pathway (hsa03010:Ribosome) Graphs. KEGG pathway were over-represented (BH adjusted p value < 0.01) by the top 65 down-regulated (BH adjusted p-value < 0.01) proteins. The pathway hsa03010:Ribosome was graphically represented and the ribosomal proteins were marked with red stars. The graphics were retrieved from the output of the functional enrichment analysis using the DAVID tool. The bar plot was generated by our lab-owned R codes.

Figure 4

Molecular pathway and network analysis of proteomics analysis of differentially expressed proteins in LTLT-Ca mammospheres and LTLT-Ca adherent cells. The proteome data was uploaded into IPA software that determined a network view of predicted associations for midasin. The network nodes represent proteins, and solid lines indicated a direct relationship while dashed lines represent an indirect relationship between proteins.

Figure 5

Increased Midasin Expression Profile in Normal Breast and Breast Cancer Samples. Representative protein staining of midasin protein expression in normal mammary tissue (A, B), lobular carcinoma tissue (C–F), and ductal carcinoma tissue (G–J). Image credit: Human Protein Atlas, www.proteinatlas.org, (Uhlén et al, 2015). Image available at the following URL: v19.proteinatlas.org/humancell.

Figure 6

Kaplan Meier plots of RFS based on midasin expression in ER+ and ER- breast cancer patients. Using Kaplan Meier Plotter, publicly available microarray repositories for breast cancer were interrogated to determine whether midasin expression was associated with different survival rates among (A) Estrogen Receptor negative (ER-) and (B) Estrogen Receptor positive (ER+) breast cancer patients. Hazard ratio (HR) and Logrank P values are shown. Low midasin expression (below median) is noted in black, and the high midasin expression (above median) is noted in red.