Researchers at Memorial Sloan Kettering Cancer Center have devised a method for selectively and continuously labeling different cell types grown in co-culture for subsequent proteomics analysis.
Called cell type–specific labeling using amino acid precursors, or CTAP, the method allows for the identification and quantification of proteins specific to a single cell type out of a mixed cell population. Detailed in a paper published last week in Nature Methods, the approach could prove particularly useful as a tool for proteomics investigations of tumor microenvironments, said Nicholas Gauthier, first author on the study and a graduate student in the lab of MSKCC researcher Chris Sander, one of the leaders of the effort.
Indeed, Gauthier told ProteoMonitor that motivation for developing the technique came from collaborative work on tumor microenvironments the researchers have undertaken with their MSKCC colleague Joanna Joyce, an expert in that field.
The researchers have filed a patent covering the method and are considering commercializing it in kit form, said Martin Miller, a postdoctoral fellow in Sander's lab and co-author on the study.
The CTAP technique relies on the fact that vertebrates are unable to synthesize essential amino acids, molecules that must be obtained through an organism's diet, or, in the case of cultured cells, through the growth medium.
Taking advantage of this phenomenon, the MSKCC researchers hit upon the idea of inserting into target cells genes coding for enzymes capable of producing these essential amino acids using precursors provided in the growth medium. Each cell type would contain a different transgenically expressed enzyme using a different supplemented precursor, and by isotopically labeling these precursors, the researchers could ultimately identify which proteins came from which population of cells.
The technique, Gauthier said, overcomes a variety of challenges that have faced previous approaches. For instance, researchers have in the past used traditional SILAC reagents to study proteins from different cell populations, labeling two distinct cell populations in isolation and then mixing them together. That method has proven effective for studying very early cellular events. However, as the cells grow and divide, the original labels quickly become diluted, making the approach poorly suited to use along longer time scales.
Another potential approach is to incorporate non-canonical amino acids into the cells of interest via tRNA synthetases. As the MSKCC authors noted, however, this could lead to structural changes in the proteins and potentially lead to altered functions.
CTAP, on the other hand, allows for continuous labeling of the cells of interest using canonical amino acids.
In the Nature Methods study, the researchers looked at two cell populations, the human embyronic kidney cell line HEK293T and the human breast cancer cell line MDA-MB-231. They engineered the HEK293T cells to express the protein DDC, which generates the essential amino acid L-lysine from the precursor 2,6-diaminopimelic acid, DAP; and engineered the MDA-MB-231 cells to express the LYR protein, which generates lysine from the precursor D-lysine. As in a conventional SILAC experiment, these precursors were either heavy- or light-isotope labeled, allowing the researchers to track and quantify them upon their incorporation into proteins.
Analyzing a sample from this labeled co-culture using LC-MS/MS on a Thermo Fisher Scientific LTQ Orbitrap XL and LTQ Orbitrap Elite, the MSKCC team identified 1,366 proteins and found that they were able, via the heavy and light labels, to determine the relative protein abundance between the two cell types.
They followed this up with an experiment to test the CTAP technique's usefulness in identifying the cell of origin of secreted proteins. Working under the assumption that the differential expression of a given protein expressed by the two cells in co-culture should roughly match the differential expression of that protein by the cells in monoculture, the researchers compared the relative abundances of proteins secreted by the two cells as determined by CTAP with the relative abundances as determined by traditional SILAC analysis of the cells grown separately. They found good concordance between the two, suggesting, they wrote, that the "method can be applied to determine the cell of origin of secreted factors in co-culture."
As Gauthier observed, this could prove useful for applications like the study of tumor microenvironments – allowing scientists, for instance, to better analyze interactions between tumor cells and the adjacent stromal cells.
"We're really excited to actually start applying this technique," Miller said, noting that the researchers were interested in following up recent investigations into how stroma cells affect tumor drug resistance.
"We have not set up a specific experiment yet, but we are looking at things like that – co-culturing stroma cells and cancer cells and maybe screening different drugs to see how stroma cells influence the drug response in these cancer cells," he said. "Then, once we find some interesting interactions, we will use our methods to pinpoint what the mechanisms are."
Miller added that the technique could potentially be expanded to study cultures containing as many as three or four different cell lines, though he noted that this would require identifying additional enzyme-precursor pairs suitable for the method. A significant portion of the original research, Gauthier said, involved reviewing the literature and identifying the DDC and LYR enzymes used in the Nature Methods study.
Beyond the challenges of identifying appropriate enzymes and precursors, the method has essentially the same limitations as conventional SILAC labeling, in that certain cells will grow more or less well on the labeled media, Miller said. "So far our experience with precursor enzyme pairs and growth on these precursors have been very positive," he added.
Ultimately, Gauthier said, the researchers would like to move the technique in vivo, developing a CTAP version of a SILAC-labeled mouse.
Such a tool could provide researchers with "the ability to look in the blood and say that a certain protein originates from a certain organ or cell type, like a tumor," he said, "And that might be of great utility when trying to develop a biomarker test."