Shin et al. (2016) present data showing proteome characterization of secretomes from four breast cancer cell lines suggesting that concentrating on this circulating source might avoid the complexity of biomarker discovery among abundant serum proteins.1 They propose that concentrating on the proteins secreted by abnormally transformed tumor cells into the patient’s circulation could be a specific marker for disease, in addition to providing novel therapeutic anti-cancer targets.
The researchers grew breast cancer cell lines Hs578T, MCF-7, MDA-MB-231 and SK-BR-3 in culture media. They harvested secretomes from conditioned media and prepared cell lysates before digesting the samples with trypsin. Once the samples were digested, the team conducted a proteomic analysis, using high-performance liquid chromatography separation in conjunction with tandem mass spectrometry. They acquired spectral data by running samples through an LTQ XL ion trap mass spectrometer (Thermo Scientific) and then searched peptide matches with SEQUEST (v27 rev12) against the human UniProt Knowledgebase.
From the initial proteomic evaluation, Shin et al. identified 1,410 proteins from the cell culture preparations. The researchers compared proteomic data from the secretomes with that collected from the cell lysates, using α-tubulin as a marker to denote intracellular origin. Secretome preparations did not contain α-tubulin, thus establishing the validity of the team’s preparation workflow.
Once they confirmed the proteomic analysis, Shin et al. used immunohistochemistry to select candidate proteins for further evaluation. Fifty-three proteins showed stronger staining in sections of tumor tissue compared to normal control tissue. The team then used in silico modeling from various online algorithms and tools, including Ingenuity Pathway Analysis, to predict protein origin. Further filtering according to specific parameters, such as cell type and association with cancer, reduced the list of candidates from the 38 proteins predicted as secretory in origin to 12.
Shin et al. confirmed the presence of these 12 candidate proteins in conditioned media from all four breast cancer cell cultures by running Western immunoblots. However, they found only four of these biomarkers in plasma from patients: FBLN1, ATP6AP2, GM2A and IGFBP-5 (n = 104 breast cancer patients). Further examination looking at mRNA levels in tumor tissue sections showed increased expression for FBLN1, GM2A and IGFBP-5. The team found that increased expression of GM2A was most noticeable in tissue sections from estrogen receptor–negative (ER-ve) tumors.
The research team examined this finding more closely, comparing Western immunoblot results from patients with those obtained from control plasma (n = 40 healthy controls). They found no difference in GM2A levels between patients and controls, confirming these results with an ELISA that showed only a modest increase (1.25x, p = 0.0576). They did, however, note raised GM2A levels by immunoblot for ER-ve patients when compared to levels in estrogen receptor–positive (ER+ve) patients. They confirmed this result using the ELISA (p = 0.0276).
When Shin et al. examined the role of GM2A in vitro they saw that siRNA transfection of the four cell lines inhibited cell migration, whereas overexpression enhanced it. They found that the effect was more pronounced in the two ER-ve cell lines, MDA-MB-231 and SK-BR-3.
Although the results do not strongly indicate a candidate biomarker found in breast cancer secretomes, they do suggest the potential involvement of GM2A in ER-ve tumorigenesis. Acknowledging that the study looked at secretome characterization from only four breast cancer cell lines, Shin et al. suggest that the results are valuable but require further testing with greater numbers.
1. Shin, J., et al. (2016) “Identification of ganglioside GM2 activator playing a role in cancer cell migration through proteomic analysis of breast cancer secretomes,” Cancer Science, 107(6) (pp. 828–835).