Sometimes it takes a little magic to solve a molecular puzzle.

Not so much magic as MAGI1, a nondescript scaffolding molecule that turns out to play a big role in explaining why a certain drug works in a rare liver cancer, and how it could work better, according to a new study from the Kugel Lab at Fred Hutch Cancer Center.

Researchers describe how they discovered that MAGI1 is key to understanding how to exploit a vulnerability in a mutated form of an under-researched cancer that occurs in the slender tubes that deliver digestive fluids from the liver to the small intestine.

The study, by liver and pancreatic cancer researcher Sita Kugel, PhD, and lead co-authors Iris Luk, PhD, and Caroline Bridgwater, was recently published in the journal Science Translational Medicine.

Much of the work described in the study — conducted so far in the lab with human cell lines and with genetically engineered mice — was initiated at Fred Hutch by Kugel’s late husband, Supriya “Shoop” Saha, MD, PhD, who died of complications from a bone marrow transplant in 2020 at the age of 40. Kugel continued and completed the work after his death.

Saha had discovered that the disease, intrahepatic cholangiocarcinoma, or ICC, has become more common over the last 40 years and made the first mouse model of the disease.

Half of patients die within a year and only 9% survive five years. Doctors have few options to treat it, but Saha found that a genetic mutation occurring in up to 37% of patients causes their tumors to grow differently than ICC tumors without the mutation.

Somehow this difference makes those mutant cells vulnerable to a drug called dasatinib, a weakness in the cancer’s defenses that could lead to a new therapy for a community of patients and their families who sorely need more options.

Toppling dominoes

Before discovering MAGI1 and its pivotal role in orchestrating dasatinib’s effect in killing mutant cells, the lab had to discover other key molecules in the cellular signaling pathway.

Important events in the busy life of a cell — growth, division and even death — are governed by enzymes that pass chemical messages to one another in a signaling pathway like a row of dominoes toppling. The first domino sets it off and the last domino accomplishes the task.

Some enzymes act like on-switches that topple a row of dominoes and some enzymes act like off-switches that stop the toppling when the action is no longer needed, creating feedback loops that amplify some signals and dampen others. Some pathways are well known, but many await discovery.

Cancer often creeps in when the wrong dominoes topple and mix up the signals.

The first domino in the pathway the researchers discovered is an enzyme called SRC (pronounced sarc as in sarcoma), a notorious on-switch involved in many cancers. In these mutant cells, the drug turned SRC off and the cells died, but the researchers had zero idea how that happened.

They cast a wide net to see what other enzymes the drug affected and discovered the final domino in the pathway, an enzyme called S6K that regulates cell size and protein synthesis.

Kugel was surprised that S6K popped out so unambiguously in their analysis.

“I’ve been around signaling pathways and cancer biology for a long time. I was struck, truly, by how specific this effect was in these cells,” Kugel said. “When we saw that, we knew we had to follow this.”

Finding magic 

Now they had the first domino (SRC) and the last domino (S6K) in the pathway, but they also had a new mystery to solve.

There was no way that SRC could activate S6K by itself. They didn’t physically connect, at least not directly.

Somewhere in the mix was a mystery domino that bridged the gap between SRC and S6K.

They scratched their heads for a long time trying to unmask that go-between domino until they devised a screening technique to tease out even the slightest connection to SRC and discovered magic.

Well, technically they discovered MAGI1, a scaffolding molecule that provides a platform for enzymes to message each other. MAGI1 linked to SRC, and now they had another domino.

Kugel marveled at the technique that plucked MAGI1 out of the bubbling cellular soup.

“It pulled out the same modified protein in every cell we tried. Again, I was struck by the specificity.” Kugel said.

But despite the wizardry its nickname implied, MAGI1 couldn’t turn anything on or off by itself because it lacked enzyme activity.

It couldn’t bridge the gap between SRC and S6K.

“We were still stuck,” Kugel said. “I remember pacing, and trying to think, what exactly this was doing.”

Then they had an insight.

They were thinking that dasatinib turned off an on-switch that triggered the tumors in these mutant cells, and they’d come up empty in their search for that mystery on-switch.

But they wondered if maybe this drug turned on an off-switch instead, which would achieve the same result.

The off-switch they envisioned turned out to be an enzyme called PP2A.

“Dasatinib turned on the off-switch,” said Luk, a post-doctoral research fellow at Fred Hutch.

Somehow in the toppling of dominoes, the drug activated PP2A, which then turned off S6K, which then killed the cells.

They had found all the dominoes in the signaling pathway, starting with SRC and MAGI1 and ending with PP2A, the off-switch that toppled S6K and caused cell death.

But they didn’t know precisely how the dominoes lined up and toppled down.

Dr. Sita Kugel runs the Kugel Lab at Fred Hutch Cancer Center, October 18, 2018, in Seattle, Washington. Fred Hutch file photo

Bringing order to chaos 

Somehow MAGI1 was bringing together the other key dominoes in a way that orchestrated the signaling pathway from drug to cell death.

They ran several tests to figure out when, where and how the other dominoes linked to MAGI1, both with and without the drug. The tests confirmed their hypothesis that MAGI1 was the linchpin of the whole signaling pathway.

MAGI1 formed a complex — a magic complex — that lined the dominoes and imposed order on chaos.

“There is a degree of randomness in how an enzyme finds its target,” Kugel said. “Dominoes sound very orderly. The way you can bring order to that is to create these signaling complexes.”

Once they understood how the magic complex came together, they understood how the drug killed the mutant cells.

When it was untreated, SRC toppled the magic complex out of commission and the cancer ran wild. But when the drug switched SRC off, the magic complex stopped the runaway growth that causes tumors.

Landing a one-two punch

Once they discovered how the magic complex worked, they could figure out how the cancer would likely fight back.

If a drug takes away something that cancer depends on, then cancer often finds a workaround and becomes resistant.

They accelerated the process artificially in the lab to see what they were up against so they could nip it in the bud.

The experiments revealed a likely source of resistance — the buildup of a certain protein that would inhibit the magic complex’s tumor-suppressing work.

They tested several drugs and found the one that best prevented the buildup.

By pairing this drug with dasatinib, they could deliver a one-two punch to knock out the cancer.

They tested the drug combo on tumor cells that had been removed from patients and directly implanted into genetically engineered mice — a key step before moving on to clinical trials.

“It was remarkably effective,” Kugel said.

Kugel said the study honors her late husband.

“I really wanted to finish this for him and for the patients he so desperately wanted to help,” Kugel said. “This will now be a part of his lasting legacy.”

The work was supported by the National Institutes of Health, the Evening for Maria Fund, a Cholangiocarcinoma Foundation Postdoctoral Fellowship, the Flow Cytometry Core and the Proteomics & Metabolomics Shared Resources at Fred Hutch Cancer Center and grants from the Spanish Ministry of Science and Innovation.

John Higgins, a staff writer at Fred Hutch Cancer Center, was an education reporter at The Seattle Times and the Akron Beacon Journal. He was a Knight Science Journalism Fellow at MIT, where he studied the emerging science of teaching. Reach him at


This article was originally published June 4, 2024, by Fred Hutch News Service. It is republished with permission.