NEW YORK (GenomeWeb News) – Researchers at the University of California, San Francisco have identified a number of proteolytic peptides that could prove useful as markers of patient response to chemotherapy.
Detailed in a study published this week in Proceedings of the National Academy of Sciences, the work used a variety of mass spectrometry workflows to develop peptide signatures indicative of cell death by chemotherapy.
While significant validation work is still needed, the approach could offer a faster alternative to current chemotherapy monitoring techniques like imaging, UCSF researcher James Wells, lead author of the study, told ProteoMonitor.
Chemotherapy typically works by causing apoptosis in cancer cells, in the course of which a number of proteases are activated. Among the main proteases involved in apoptosis are caspases, which has led researchers to speculate that caspase cleavage products could be used as markers for measuring chemotherapeutic activity.
In fact, Wells noted, one such analyte, a caspase-cleaved peptide from the protein cytokeratin 18, has been used for this purpose in clinical trials. However, in general, researchers have had difficulty detecting caspase-cleaved peptides with the sensitivity and reproducibility required of a clinical biomarker.
To better detect such markers, Wells and his colleagues turned to a peptide enrichment technique developed in his laboratory. The method uses the enzyme subtiligase to biotin tag free N termini on peptides generated by protease activity. These labeled peptides can then be pulled down with streptavidin-coated beads, limiting interference from high-abundance proteins while increasing the likelihood of detecting low-abundance caspase cleavage products.
The researchers used this enrichment technique across several different mass spec experiments to identify peptides linked to chemotherapeutic-driven apoptosis. In the first, they used unbiased measurements on an AB Sciex QSTAR Elite QTOF to identify proteolytic fragments released in samples from five hematologic cancer patients and in cultured cells post-chemotherapy.
Using the results of this analysis, along with a list of proteolytic peptides previously found to be present during apoptosis, the researchers developed a targeted inclusion list for identification on a Thermo Fisher Scientific LTQ Orbitrap Velos. In this approach, peptides within mass windows around those on the inclusion list are preferentially selected for sequencing, increasing the sensitivity of detection for these analytes.
Combining the results of these two experiments, the UCSF researchers identified a total of 153 proteolytic peptides linked to cell death. They then followed up with quantitative SRM assays to 140 of these, measuring which were elevated in post-chemotherapeutic samples compared to pre-treatment samples. This analysis identified 90 peptides with increased abundance post-chemotherapy, including 77 that showed at least a two-fold increase.
Following this with SRM analysis of pre- and post-treatment samples from an additional 16 patients, the researchers identified 16 peptides that increased post-chemotherapy across multiple patients, making them, the authors noted, "the most promising targets for further exploration in clinical development as biomarkers of chemotherapeutic efficacy."
Moving forward, Wells said, the researchers plan to investigate these markers in larger numbers of patients with the aim of collecting patient treatment response and outcome data, as well.
More immediately, though, he and his colleagues have begun efforts to establish that the detected peptides are, in fact, coming from cancer cells as opposed to healthy cells also killed by the chemotherapy.
For this they are using mouse xenograft models — grafting human tumors into mice — which should allow them to determine which peptides are coming from which cells, Wells noted.
"Because the human proteins [from the grafted tumor] are different from the mouse proteins, we can then ask the question of how many of the products we see are derived from the mouse versus the human tumor we implanted," he said.
He added that the researchers also anticipate that "different types of cancers might give different apoptotic signatures." Therefore, he said, if "we saw that no matter what the cancer all the patients exhibited the same markers, we might then wonder if that was coming from [normal cells]."
"We'd like to get an answer to that [tumor-versus-healthy-cell] question before moving on to more clinical work, because once we have that then I think we'll know better how to focus our clinical experiments," Wells said.
The PNAS study examined hematological cancers, but Wells said he expected the technique would also be applicable to solid cancers. They decided to begin with blood-based cancers in the expectation that these would provide them the best chance of identifying the caspase-cleaved peptides, he said.
"We chose cancers where we though that we would have a good chance of seeing things," he said. "The proteins that would be released [from these cancer cells] would be released directly into the plasma."
While the UCSF team used mass spec for this initial discovery work, Wells said that as they move forward in the validation process they will probably transition at some point to immunoassays, which offer the high throughput needed for running large numbers of patient samples.
UCSF has recently developed a phage display-based platform for high-throughput antibody generation that, Wells said, he and his colleagues will likely use for this stage of the project.