Skip to main content
Premium Trial:

Request an Annual Quote

Dundee Scientists Characterize Phosphoproteome of Cytotoxic T Lymphocytes


By Adam Bonislawski

Researchers at Scotland's University of Dundee have completed a phosphoproteomic analysis of cytotoxic T lymphocytes, identifying roughly 2,000 phosphorylation sites on more than 900 proteins, including 450 phosphosites controlled by T-cell antigen receptor signaling.

In particular, the study, which was published this month in Nature Immunology, examined signaling networks tied to CTL chromatin regulation in hopes of better understanding the systems of gene regulation characteristic of these cells.

"It's kind of the burning question of all biology," said Doreen Cantrell, professor of cellular immunology at the University of Dundee and one of the authors of the paper. "What makes a T-cell a T-cell? What makes a B-cell a B-cell? Because if you understand enough about signaling and how these processes are controlled, then you could make the cell do what you want theoretically."

"We've had a long history of knowing that serine kinases are important in T-cells, and what we're now at the point of doing is trying to understand what serine kinases actually do – what their substrates are," she said. "We've got serine kinases interacting in T-cells and genes being regulated, but what's in the black box in between?"

One of the study's key findings in this regard was the identification of a constitutive signaling pathway that maintains phosphorylation of the histone deacetylase HDAC7, moving it out of the nucleus to allow high expression of the protein CD25, a key to CTL proliferation.

Previously, Cantrell noted, HDAC7 was thought to be phosphorylated in CTLs only in response to T-cell antigen receptor activation. The pathway identified by her team's analysis, however, suggests that it's phosphorylated even in the absence of such activation. Because phosphorylation causes export of HDAC7 out of the nucleus allowing for high expression of proteins tied to CTL growth and proliferation, this discovery might offer insights into CTL response to antigens like viruses, Cantrell told ProteoMonitor.

According to conventional theory, HCAC7 "is supposed to be sitting in [CTLs] non-phosphorylated, and only be phosphorylated when you engage the [antigen] receptor," she said. "What we were seeing is that [HDAC7 phosphorylation] has become a constitutive pathway in the cell… no longer under the control of the T-cell receptor. That uncoupling process was a surprise."

This constitutive phosphorylation pathway may allow CTLs to react more quickly to infections by maintaining their chromatin in such a way that allows rapid transcription of proteins involved in such response, Cantrell suggested.

"If you've got to get a chromatin regulator [like HDAC7] out of the nucleus and then regulate the chromatin by the binding of a transcription factor, that may be fine during development when you've got all day to do these things," she said. "But if you get a viral infection, you've got to deal with it right now. You probably want to have your genes ready to go. So the uncoupling makes sense because it would make the T-cell receptor response work much more efficiently."

In addition to HDAC7, the researchers also identified a T-cell receptor-controlled pathway involving the transcriptional receptor TRIM28, which, Cantrell said, is known to be important in stem cell biology but is little-studied in CTLs. The team also identified a large number of proteins important in methylation pathways, which, she noted, could help elucidate how DNA methylation is tied to CTL state changes.

More generally, she added, the fundamental breadth of the phosphorylation activity observed was surprising.

"I think the immunological community tends to think of [CTLs] just sitting there doing nothing unless they see the T-cell receptor [activated]," she said. "What our paper is saying is, 'No, they're not just sitting there doing nothing. They're actually doing quite a lot to actively maintain themselves in the correct state.'"

For the study, the researchers used SILAC-labeled cultures analyzed on Thermo Fisher Scientific LTQ Orbitrap XL and LTQ Orbitrap Velos machines. Protein identifications were made using the MaxQuant software, which, Cantrell said, has put mass spec data analysis "more in the realms of what a non mass spec expert can do."

The study, she said, is one of the broadest phosphoproteomic profiles done to date in primary T-cells, a fact due in part to the difficulty of growing enough CTLs in such a way as to allow SILAC labeling.

"The cell [culturing] bit is the green finger part," she said. "It's not that easy to grow normal [CTLs] in sufficient numbers and get them growing enough to use the SILAC approach. That's why the study is interesting to immunologists. It's not just a phosphoproteomics analysis of some leukemic cell line. This is a real T-cell responding to a viral peptide."

Cantrell's team now plans to perform similar analyses for a variety of T-cell types with the ultimate aim of identifying therapeutic targets to manipulate the function of various lymphocyte subsets.

"If you're a clinician and your patient has multiple sclerosis and I can tell you the kinase pathways that are on in a T-cell that's mediating that particular autoimmune disease, these are kinases you might want to hit therapeutically with inhibitors," she said. "That's kind of the long-term goal."

The researchers are also planning global phosphoproteomics studies looking at the impact of the drugs rapamycin and cyclosporin A – commonly used in organ transplant procedures – on lymphocyte phosphorylation, with the aim of better understanding how they work as immunosuppressants.

Cantrell plans to use mass spec-based methods for this work, as well, she said, noting that years of struggling with bioinformatic and in vitro peptide array approaches to kinase work have left her convinced that mass spec analysis of in vivo systems is the best technique.

"I spent the last five years of my program grant [trying these approaches]. We've tried exhaustively to look at that stuff," she said. "It cost us hundreds of thousands of dollars and everything we got was an in vitro [kinase] target but not an in vivo target."

"You can't use bioinformatics because what the sequence information can't tell you about is the position of proteins in cells," she added. "In vitro, kinases can phosphorylate a lot of things, but in vivo they may never see that protein, so that's why I think you have to do wet lab work to get this [information], and mass spec is the way to do it."

Steven Pelech, president and chief scientific officer of the biotech firm Kinexus, which offers bioinformatic and in vitro array-based analyses of protein kinase substrates, suggested to ProteoMonitor that Cantrell's past troubles with these approaches might be due in part to the software used.

Calling the Nature Immunology study "a very nice piece of work," he said that some established kinase specificity software examines only around one-third of the protein kinases that could actually phosphorylate a given substrate.

Cantrell, however, said she's now committed to in vivo approaches.

"I've been doing this for 20 years. Every time I make a best guess and say, 'OK, that protein has a perfect substrate for my kinase,' well, it turns out that it might have [the perfect substrate] but [the kinase] doesn't use it," she said. "Now it's going to be just hard biochemical data using mass spectrometry."

Have topics you'd like to see covered in ProteoMonitor? Contact the editor at abonislawski [at] genomeweb [.] com.