Skip to main content
Premium Trial:

Request an Annual Quote

Phosphoproteomic Screens Elucidate Feedback Loop Linked to mTOR Inhibitor Resistance


By Adam Bonislawski

This article was originally posted on June 14.

Two separate mass spec screens of the mTOR-regulated phosphoproteome – one led by researchers at Harvard University and the other by scientists at the Massachusetts Institute of Technology – have offered new insights into the workings of the far-reaching but poorly understood signaling network.

In particular, the studies, both of which were published this week in Science, elucidated feedback mechanisms linked to the development of resistance to cancer therapies and identified the protein Grb10 as a likely tumor suppressor.

While mTOR signaling has been implicated in a wide range of functions including cell growth and division, and diseases like cancer and diabetes, study of the protein has been difficult, in large part due to a lack of agents capable of completely inhibiting mTOR activity.

However, the introduction in recent years of complete mTOR inhibitors such as Torin1 – used in the MIT work – and Ku-0063794 – used in the Harvard research – "has really opened the door to a systematic investigation of mTOR substrates," Peggy Hsu, a researcher on the MIT study, told ProteoMonitor. "The field had rapamycin," she said, "but we know that inhibits only a subset of mTOR's functions," limiting its usefulness in broad mTOR screens.

It's not only as a research tool that rapamycin has shown limited utility. The compound is currently used to treat renal cell carcinoma, but, Hsu said, "overall it's been quite disappointing as an anti-cancer agent."

It has been proposed that "treatment with rapamycin leads to changes in feedback signaling that allows the cells to maintain proliferation and survival," she said. "And the idea is that if you were able to figure out how to perturb that feedback signaling, you could sensitize cells to the effects of rapamycin."

In both studies, efforts to explore this feedback signaling, as well as the wider mTOR signaling network, led the researchers to focus on the mTOR substrate Grb10.

In the Harvard-led screen, this protein showed up as a significant hit, exhibiting dramatically lower phosphorylation levels upon mTOR inhibition, said Yonghao Yu, one of the researchers on the project.

"From a mass spec perspective, we knew it was a target dramatically phosphorylated by mTOR," he told ProteoMonitor. At the same time, literature searches done by the Harvard team found that Grb10 had been linked to insulin signaling and growth inhibition, he noted.

"Regulating growth is one of the major functions of mTOR, so we knew there were some connections between mTOR's function and Grb10's function in that both seem to be involved in growth regulation," Yu said. "So that was another piece of evidence that suggested this was a particularly interesting molecule."

While both groups converged on Grb10 as a molecule of interest, Yu said that neither was aware the other was also investigating the protein until John Blenis – with Steven Gygi one of the leaders of the Harvard project – reached out to David Sabatini – head of the MIT effort – looking for reagents to use in the work.

"We just found out very coincidentally that we were working on similar [mTOR] screens using mass spec," Yu said. The particulars of the studies differed – the Harvard team used SILAC labeling on a Thermo Scientific LTQ Orbitrap Velos while the MIT researchers used iTRAQ labeling and an AB Sciex QSTAR Elite – but, he said, "it's a very nice coincidence that out of the screens … we both converged on the same point, the protein Grb10."

Both groups linked Grb10 to suppression of PI3K activity, an oncogenic kinase that has been implicated in a wide range of cancers. Using shRNA knockdown of Grb10, the two teams each discovered that a decrease in Grb10 led to an increase in PI3K activity. The Harvard researchers also found that expression levels of Grb10 were significantly lower in human cancer cells compared to adjacent normal cells.

These findings, both groups noted, indicate that Grb10 likely plays a key role in the feedback loop to PI3K and patients' responses to mTOR inhibition. Improved understanding of this role, Hsu said, "could lead to a better understanding of how to deploy mTOR inhibitors."

It's well known that PI3K activates mTOR, Yu said, and given that both are known to be up-regulated in a number of different cancers, the two proteins have emerged as important drug targets. Too much mTOR activation, however, spurs a feedback loop that then inhibits PI3K activity, he noted. This is a key factor to consider when treating patients with mTOR inhibitors given that a decrease in mTOR activity can eliminate this PI3K-suppressing feedback mechanism and lead to a worsening of the disease.

The researchers found, Yu said, "that mTOR phosphorylation of Grb10 leads to an accumulation of Grb10. And the functional consequence of Grb10 is to inhibit PI3K activation."

This discovery, he said, suggests that patients with high levels of Grb10 expression may respond well to mTOR inhibitors, because Grb10 will continue to suppress PI3K even in the absence of the PI3K suppression by mTOR.

From a practical standpoint, this means that therapies to stabilize or increase Grb10 expression could aid in cancer treatment, Yu said. It also suggests that Grb10 could be used as a biomarker to determine which cancer patients are most likely to respond to treatment with mTOR inhibitors.

The Harvard researchers collaborated with Millipore on an antibody to Grb10 and have filed a provisional patent covering the use of Grb10 for diagnostic and therapeutic purposes, Yu said.

The studies also open up a variety of additional avenues for research, he added, including investigations into mTOR's role in processes like autophagy and transcription.

"Even for functions where we've known mTOR is involved, we might not necessarily understand the entire mechanism," Hsu said. "So we hope this work will allow people to not only find new things regulated by mTOR, but even in the case of things we know mTOR is involved in, perhaps we can better understand the mechanism by which it does so."

"The screen has provided a lot of the missing bits of biological processes known to be regulated by mTOR," Yu said. "This really shows the power of the [phosphoproteomic] technique."

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