Researchers at New York's Memorial Sloan-Kettering Cancer Center have identified undifferentiated prostate tumor-initiating cells that possess stem cell characteristics and multipotency while not expressing the commonly used prostate cancer protein biomarker prostate specific antigen.
Characterized by the co-expression of three surface marker proteins – TRA-1-60, CD151, and CD166 – the cells offer insights into the molecular basis of human prostate cancer and suggest potential new therapeutic and biomarker strategies for the disease, Rajasekhar Vinagolu, senior research scientist at Sloan-Kettering and one of the leaders of the project, told ProteoMonitor.
In many prostate cancer cases, stem-like tumor-inducing cells, or TICs, "have already escaped into other portions of the body, such as the bone and lung and brain," by the time patients come in for surgery, Vinagolu said, which makes identification of these cells potentially important for detecting and treating the disease.
In the study, detailed in a paper published this week in Nature Communications, the team, which also included Sloan-Kettering scientists Lorenz Studer and Howard Scher, disassociated prostate tumor samples and used fluorescence-activated cell sorting to identify TICs via novel surface protein markers that could be used for their isolation.
They began by testing established cancer stem cell markers like epithelial cellular adhesion molecule – EpCAM – and hyaluronic acid receptor – CD44. However, Vinagolu noted, because these proteins are associated with a variety of cell types, the researchers were concerned they wouldn't provide enough specificity to single out the cells of interest.
"In cancer stem cell research nowadays everyone is looking for [cell surface] markers, so we looked at some of these known markers," he said. "But these known markers are present in many other cell types. I wanted to have some unique markers."
With that aim, Vinagolu added TRA-1-60 – a surface protein typically expressed in human embryonic stem cells – to the group of proteins under study, finding that TICs expressing this marker were significantly more tumorigenic than cells positive for only the other known markers.
They then investigated the addition of the markers CD151 and CD166, which are associated with colon cancer, discovering that cells positive for all three proteins – TRA-1-60, CD151, and CD166 – were more tumorigenic than cells for any of the other known markers – suggesting that this triple-marker pattern could potentially be used to identify progenitor stem-like cells responsible for initiating prostate tumors.
With this triple-positive profile in hand, it might also be possible to discover additional markers that could enhance researchers' ability to isolate such TICs, Vinagolu said, suggesting that he might in the future undertake a broad proteomic analysis of these cells in hopes of identifying such proteins.
"What I personally want is to discover some new markers beyond these three," he said. "We have to isolate only these triple-positive cells and then look for additional markers within them."
Given that TRA-1-60 is a glycoprotein, focusing on the glycosylation patterns of cell surface markers within the triple-positive population is another possible approach, Vinagolu said. Mass spec would be the best technology for any such proteomic analysis, he suggested, noting it allowed for a "more sophisticated and more global" approach to analyzing protein expression patterns than immunoassay-based methods.
Improved isolation of prostate TICs might also enable the development of better biomarkers for prostate cancer, Vinagolu said. Given that these progenitor cells don't express PSA – currently the standard biomarker for the disease – detection of TICs might provide an earlier warning.
"You could take the tumor cells, analyze them by FACS so you can detect very limited numbers of these triple-positive cells, and that could be a better biomarker," he said.
The Sloan-Kettering team also investigated differences in mRNA and microRNA expression patterns for the various subsets of TICs, as well as differences in the activation of a number of commonly studied protein signaling networks, finding that the TRA-1-60 positive cells have enhanced activity in the NF-КB pathway, which is associated with anti-apoptotic functions.
Such increased NF-КB activity could explain why progenitor cells like the triple-positive TICs identified in the study are so resilient in the face of cancer therapies, Vinagolu suggested.
"Progenitor [cells] are very strong," he said. "If you treat a patient with radiation therapy, the normal cells will die because of the radiation, but the stem-like cells will be resistant. In the study, we demonstrated that the tumor-initiating cells have enhanced NF-КB activity. We don't know the mechanism yet, but the main function of this pathway is anti-apoptotic, so we have to make the [tumor-initiating] cells die by inhibiting this anti-apoptotic signaling pathway."
Given that normal cells also rely on the NF-КB pathway for survival, however, any inhibitors aimed at the pathway would need to be carefully targeted, Vinagolu said, noting that surface protein markers might aid such targeting.
"You can't kill all NF-КB activity, because you need that for cells to survive," he said. "So we need to design drugs such that the NF-КB inhibitors can only go into the cells that express unique surface markers like TRA-1-60 on the tumor-initiating cells."
Since the paper's release, the work has drawn interest from the pharma and biotech industries, Vinagolu said, although he declined to name any specific firms he had spoken to.
Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.