Deep Proteome Characterization Highlights Breast Cancer Cell Functionality

Pozniak et al. (2016) recently completed a deep proteome characterization in estrogen receptor–positive (ER+) breast cancer.¹ The researchers found that, compared to normal breast epithelium, the cancer cells showed alterations in metabolic profiles in addition to dysregulated protein homeostasis.

Although transcriptomic and genomic profiles exist for breast cancer tumors, there is less known about the cellular proteome. Pozniak et al. maintain that characterizing the breast cancer proteome gives a deeper understanding of the pathology; determining differential abundances in protein levels provides insight into changes in cellular functionality. For this reason, they used a super–stable isotope labeling with amino acids in cell culture (super-SILAC) approach to conduct quantitative proteomic analysis of formalin-fixed paraffin-embedded (FFPE) tissue blocks from breast cancer patients obtained before treatment. Comparing the results from the primary tumor sections with those obtained from lymph node metastases and normal breast epithelium, the researchers provide a comprehensive library of ER+ breast cancer.

The team obtained FFPE tissue blocks from patients with confirmed ER+, human epidermal growth factor receptor 2–negative (HER2−) breast cancer. They analyzed sections from 21 patients with no lymph metastases (LNN) and from 25 with metastatic spread (LNP; n = 20 lymph node sections). As normal controls, the researchers chose normal breast epithelium resected during surgery and confirmed as cancer-free margins to the primary excision. They confirmed morphology and pathology using microscopy and immunohistochemistry to select areas for proteomic evaluation. The team extracted proteins with standard protocols according to a FFPE–filter-aided sample preparation (FFPE-FASP) workflow.

In preparation for quantitative proteomic evaluation, Pozniak et al. prepared a super-SILAC mix for spike-in studies during mass spectrometric analysis. They labeled breast cancer cultures, selecting well-established cell lines that show various receptor status, location and pathological type characteristics (HC1599, MCF7, HCC1937 and HCC2218), in addition to a normal breast epithelial control (HMEC). The researchers added the protein mix to the proteins extracted from the tissue sections before digesting them with trypsin. They then cleaned the peptide preparations using strong cation exchange fractionation and FASP before mass spectrometry on a Q Exactive or Q Exactive Plus mass spectrometer in combination with a nano-ultra-high-performance liquid chromatography (nUHPLC) EASY-nLC 1000 system (all Thermo Scientific). The team analyzed the mass spectrometry raw files with MaxQuant v.1.5.0.36 and searched protein identities against the UniProt human database November 2014.

From the FFPE sections, the team identified 150,471 peptides corresponding to 10,124 proteins. They managed to quantify 10,043 of these. On analysis, the team found no significant differences in protein numbers between the lymph node, breast cancer and normal breast epithelial samples. However, they did note functional changes associated with cancer following network analysis of the results. The team also found differences between normal tissue and cancer tissue from the breast and lymph node metastatic spread, with differential expression in 969 proteins (563 upregulated and 406 downregulated in tumor tissues, with 95% of these showing a more than fivefold difference).

Pozniak et al. examined the differential data and found that most proteins grouped into two categories:

1. Protein homeostasis, including increases in ribosomal, professional and lysosomal proteins, and a reduction in those involved in the unfolded protein response pathways
2. Central metabolism, including reduced glycolytic and fatty acid pathway proteins, and increased oxidative phosphorylation factors

To further characterize these findings, the research team performed functional assays in culture. Pulsed SILAC studies of HMEC and MCF7 cells showed increased protein turnover in the cancer cell lines. Pozniak et al. also found increased mitochondrial activity in the tumor cells, confirming results using immunohistochemistry to show increased presence of ACOT1, GLUL and SLC25A11 enzymes. The altered functionality shows that the cells reduce glycolysis, fatty acid breakdown and synthesis, and cell respiration. Moreover, they also reduce glutamine uptake while increasing production.

Comparing LNN with LNP primary tumor sections, the researchers found that 78 proteins increased in abundance and eight reduced with cancer. However, they found very little difference in tumor proteomes comparing the lymph node metastases with the matched LNP primary tumor sections (four proteins increased in abundance; six decreased).

Overall, Pozniak et al. suggest that the deep proteomic analysis performed on the tumor sections shows that cancer cells alter metabolic activity as a regional adaption to surrounding tissue. They are confident that the results give valuable functional insights into tumor biology and development that genomic profiling may not reveal.

Reference
1. Pozniak, Y., et al., (2016) “System-wide clinical proteomics of breast cancer reveals global remodeling of tissue homeostasis,” Cell Systems, 2(3) (pp. 172–184), doi: 10.1016/j.cels.2016.02.001.