Cracking the cases of the wonky antenna and the vanishing chaperone

On a Wednesday afternoon in July, Jeremy Hollis defended his PhD thesis at Fred Hutch Cancer Center before an audience of fellow graduate students, professors and mentors.

The presentation capped six years of molecular detective work plus a full other life spent hiking, running marathons, designing sets for local theaters and teaching hip hop.

“He seems to have more hours in a day than the rest of us, which allows him to get a phenomenal amount of things done, both inside and outside of the lab,” his primary mentor, Melody Campbell, PhD, said in her introduction.

A roomful of scientists knew very well that Hollis gets 24 hours in his day just like everyone else, but they laughed because it sure seems like Hollis has super-temporal powers. How else could he have solved not one, but two evolutionary cold case mysteries like Sherlock Holmes in a white lab coat?

How he did it all says something important about Hollis as a scientist, but also something important about Fred Hutch as an institution that values making good scientists as highly as it values making good science.

The Case of the Vanishing Chaperone Hiding in Plain Sight

Hollis first heard about the mystery of the vanishing chaperone in 2020 while interviewing for graduate programs during the COVID-19 pandemic.

During a Zoom interview, Fred Hutch evolutionary biologist Harmit Malik, PhD, told him the story, which traces back to a 1991 discovery by Fred Hutch researchers.

They identified CENP-A, a molecule that helps mark the centromere, the spot on a chromosome where a large molecular machine called the kinetochore is built during cell division. The kinetochore pulls duplicated chromosomes apart, so each daughter cell receives an identical set.

The mystery centers on a helper molecule called a chaperone that delivers CENP-A to the correct location. This chaperone is ancient. Versions of it exist in fungi, meaning it dates back at least to the last common ancestor of humans and mushrooms more than a billion years ago.

Yet it appeared to be missing from fish, insects and worms, even though these organisms should also possess such a fundamental component of cell division.

It could be that those species evolved in a different way to do the job, which had not yet been discovered. Or it’s possible the chaperone had changed so much over time that standard search methods could no longer recognize it.

The chance to solve mysteries like this sold Hollis on Fred Hutch's Molecular and Cellular Biology Graduate Program, run jointly with the University of Washington.

During his first-year rotation in Malik's lab, he persuaded Malik to let him investigate the vanishing chaperone.

"The seed was kind of planted before I even knew I was moving to Seattle," Hollis said. "This was my first rotation project in grad school. Usually, those projects help you learn techniques and get to know a lab. They're not necessarily something you continue."

Malik likened the mystery to a river of time. On one bank sat an ancient ancestor with a CENP-A chaperone. Modern humans and mushrooms sit on the other side, both still using the same type of chaperone. To trace the chaperone’s lineage and connect the past with the present, Hollis and his colleagues needed to find evolutionary steppingstones across the river.

Hollis searched gene-sequence databases like a historian tracing a family tree. Gene sequences can change over time, just as a family name might evolve from Smith to Smythe and eventually become almost unrecognizable.

Using an iterative search method, Hollis identified increasingly distant relatives of the chaperone. His first searches among ray-finned fish such as salmon came up empty.

Then he found a version in the coelacanth, a lobe-finned fish often called a living fossil.

When he used the coelacanth sequence to search again, the chaperone began turning up in species where it had seemed absent, including sea stars, sea urchins and ray-finned fish.

Malik had previously found evidence of the chaperone in fish, but Hollis independently reproduced and extended those results.

“I didn't tell him any of this stuff, but he essentially recreated my findings, Malik said. “He actually went even further.”

Maybe the vanishing chaperone hadn’t vanished after all.

“We had very preliminary biological evidence that we were on the right track, but we were not at the point where this would actually convince anybody except Jeremy and me.”

Choosing a lab

Hollis had taken the first step in cracking the case, but it was time to move on to other rotation labs. He learned structural biology in Campbell’s lab and did his third rotation with Sue Biggins, PhD, director of the Basic Sciences Division who is an expert on kinetochores.

When the year was over, it was time to choose a lab to do his PhD thesis, which involved pitching potential projects to each of the labs he had spent time with during his first year.

“When the time came for Jeremy to decide which lab to join, I was floored because he came to my office with three separate innovative multi-year research projects,” Campbell said.

His proposals were as sophisticated and ambitious as the highly competitive R01 independent research grants from the National Institutes of Health that pay the bills in established scientists’ labs.

“These projects read more like R01s than a first-year grad student's ideas,” Campbell said. “I later learned that he came up with equally innovative and detailed projects that he had pitched to his two other rotation labs.”

It was a tough decision, but he chose Campbell’s lab for his PhD work. He also wanted to keep Malik as a co-mentor and collaborator.

Campbell was sold on his proposal.

“Harmit however, was a little bit more skeptical and gave Jeremy an ultimatum: put together a chalk talk of specific aims and convince us,” Campbell said. “Within a week, we met in Harmit's office and Jeremy gave his chalk talk and Harmit grilled him like a faculty candidate.”

Hollis successfully made his case to keep both mentors.

“He's a stubborn and very persuasive student,” Malik said.

The Case of The Wonky Antenna That Made Complex Immunity Possible.

The Campbell Lab focuses on how cells know what’s going on around them by sending and receiving signals from their external environments using all sorts of molecular structures that stud the surface of the cell.

Hollis proposed to solve a mystery involving a family of signaling structures called integrins.

Think of an integrin like a two-legged tower on the surface of the cell. Most of the tower rises from the cell surface to interact with stuff floating outside the cell and a shorter bit of the tower pokes through the membrane into the cell where it interacts with stuff floating around inside the cell. 

The stuff floating outside the cell attaches to a specific site on the tower, which initiates a signal that travels through the membrane and sets off more signaling inside the cell.

The signals tell cells how fast to grow and help them recognize threats in the outside environment so they can defend themselves from pathogens. Malfunctioning integrins lead to many diseases, including autoimmune disorders, infections and cancer.

About 500 million years ago, early in the evolution of animals with backbones, an integrin common in cells lining the gut acquired a new part that glommed onto the tower in the same spot where the molecule that normally binds there attaches, effectively blocking its function.

Disruptive mutations like that don’t tend to stick around long because the cells that harbor them die before they can reproduce. But somehow this wonky addition proved to be so beneficial that half of all integrins have it today.

Using  cryogenic electron microscopy, or cryo-EM, one of the most powerful tools scientists use to determine the structure of biological molecules, Hollis worked out what was going on.

Then he worked with Malik to figure out where this wonky antenna came from. They fed the genetic sequences from all the available examples in existing organisms into a computer to reconstruct what a wonky bit would have had to look like in our deep history to account for all the variation we see in the world today.

That ancestral protein doesn’t exist anymore, but its sequence can be inferred from all its descendants.

“What’s amazing to me is that computationally we get this sequence — and then Jeremy can express and purify this protein in the lab — and then we can literally visualize something that could have existed 500 million years ago,” Campbell said.

This new wonky part did indeed block the traditional binding site, but it also expanded the surface area where many kinds of molecules could now dock.

It was like replacing a radio antenna tuned only to the Boston symphony with a dish receiving a broad swath of American music: Grand Ol’ Opry from Nashville, rock and roll from Cleveland, Chicago blues, Miami salsa, Seattle grunge and New York hip hop.

This new dish on the integrin tower enabled the cell to interpret all kinds of new signals, an important step toward building a complex immune system that can recognize friend from foe among the many microbes and potential pathogens taking up residence in the gut.

By solving the case of the wonky antenna, Hollis opened new possibilities for drugs and therapies when integrins malfunction.

“Even though it wasn't the initial goal of the paper, the structure that Jeremy solved is at such high resolution that all of a sudden we can start trying to understand how those particular therapeutics work a little bit better,” Campbell said.

That work alone would have distinguished any graduate student’s time at Fred Hutch, but Hollis had signed up to solve two cold cases.

During his first rotation in the Malik Lab, Hollis had found one stepping stone in The Case of the Vanishing Chaperone, but he wanted to find them all.

The Case of the Vanishing Chaperone revisited

Throughout the time Hollis was solving the case of the wonky antenna, there were times when research was on hold for various reasons and he could turn his attention back to the mystery that had enticed him when he was still applying to graduate programs.

When he was last in the Malik Lab, they had tried to find an elusive CENP-A chaperone helper that helps cells divide precisely so that each daughter cell gets an identical set of chromosomes.

It was fundamental to biology that both humans and mushrooms have it, but a lot of species in between seemed to lack a version of this chaperone and Hollis and Malik wanted to know why. Was it really missing because those species found another way to do the same job, or was it still there, doing what it had always done but hiding in the weeds because its genetic sequence had become unrecognizable?

They hypothesized that it was the latter — it still performed the same function but looked so different now that it had vanished. If they scoured the family tree, looking for bits and pieces of it among far-flung relatives with some new methods, they could identify the missing versions.

But Hollis would need more than genetic sequences to definitively make his case that this was the same chaperone that mushrooms and humans have and not some independently evolved molecular worker doing the same thing.

Their search had revealed a potential candidate for the chaperone in zebrafish, a ray-finned fish that a Fred Hutch colleague, Cecilia Moens, PhD, studies to understand the early development of the vertebrate brain. Moens holds the Raisbeck Endowed Chair for Basic Science.

The sequence for their candidate chaperone was way different than the human version, but it was possible that although it produces a much different string of amino acids, that string would fold into the same 3-D shape as the human version.

And if it was the same shape, it would perform the same function.

They used an AI tool to predict what shape the candidate chaperone would take if it did the same job as the human version, and the results gave them more confidence that they were on the right track. 

That got them closer, but they still needed to prove it with living zebrafish.

Hollis persuaded Moens and her lab technician, Jason Stonick, to assist in the hunt, and they conducted experiments that established that indeed, the chaperone functions like the human version and if you take it out, it causes the same problems in zebrafish that it does in humans when the chaperone malfunctions.

“The coelacanth ended up being this amazing stepping stone,” Malik said. “This is what he did in his rotation and then, in writing the paper, he basically revealed all of the tetrapods, essentially all of the fishes and sea urchins, everybody has it. And so that's one big chunk of the missing animal tree that was gone. Then he showed, using very similar approaches with tardigrades and horseshoe crabs, that he could trace how the arthropods started off with something that looked like it.”

That discovery alone would have made for a fine paper. They had found the missing chaperone in fish using several different methods.

But what about worms and fruit flies? Hollis thought they were close to solving the whole mystery and wouldn’t let it go.

“Jeremy convinced me — let's just do it all in one,” Malik said.

Looking for the vanishing chaperone in worms

They had found the chaperone in fish, but according to the published literature, there was just no trace of it in nematodes, which are tiny worms that have become highly informative model organisms for biologists to learn about fundamental processes such as cell division.

“The most amazing thing is Jeremy and I were both completely convinced that nematodes were in fact missing this,” Malik said.

But Hollis wanted that part of the mystery buttoned up, one way or the other.

He used the same approach he had used to look for the chaperone in fish, but couldn’t find anything in C. elegans, a common nematode that biologists study in the lab. This seemed to confirm that conventional wisdom that worms didn’t have it.

But when Hollis cast the net wider, he found a bit of something in a distant relative to C. elegans that looked like the vanishing chaperone.

“One nematode had something that looked maybe this could be it, but I was not convinced,” Malik said. “But to Jeremy's credit — I give him all the credit here — he persisted.”

Several researchers at Fred Hutch study C. elegans, including Nicolas Lehrbach, PhD, in the Basic Sciences Division.

Hollis proposed that they ask the Lehrbach Lab for some help to test it out the way they had with the zebrafish, and once again, he talked them into playing Watson to his Sherlock Holmes.

Hollis proposed a collaboration with Irini Topalidou, PhD, a senior staff scientist in the Lehrbach Lab.

“So, I was in my lab and I was doing my stuff and suddenly Harmit and Jeremy appear out of the blue and they tell me: ‘We would like to ask you if you are interested in a little story and if you have the time and the energy and the interest to join us in this little adventure’,” Topalidou said.

Hollis persuaded her to join in the hunt.

“I was extremely excited,” Topalidou said. “Then I found out that Cecilia Moens is also involved, and Jason from her lab, and of course Melody is involved. It was such a big collaboration of four labs, and it was very fruitful and it worked beautifully.”

They conducted experiments that established that indeed, the chaperone functions like the human version and if you take it out, the worms die.

With all they learned from finding the vanishing chaperone in fish and worms, they found a version of it in fruit flies.

“It's very satisfying because now there are no holes in the in the puzzle that we don't expect there to be,” Malik said.

How did Hollis do it all with the same 24 hours in a day that everyone else gets?

Like Sherlock Holmes, he solved his mysteries with keen intellectual work, astute observation and persistence.

Modern theatrical depictions of the fabled detective often portray Holmes as insufferably arrogant and blunt, even with his faithful investigative partner, Dr. Watson.

Hollis, however, also possesses a less acknowledged, but essential quality for collaborative scientific work — charm.

“Besides the fact that he's a great scientist and smart and ready to jump into anything that interests him, he's extremely charming,” Topalidou said.

Case Closed

By the time Hollis gave his PhD defense, he had not one but two significant publications to his name.

The Case of the Wonky Antenna he solved in the Campbell Lab was published in Science Advances under the more technical title: “Molecular exaptation by the integrin αI domain.”

“He said he was going to map the binding landscape of the Alpha I domain, and also that he was going to define the structural requirements of Alpha I Domain integrins,” Campbell said.  “That was five years ago that he said he was going to do this, and that is exactly what he has done. Suffice to say, these results have made significant contributions to the fields of both integrin and evolutionary biology, and Jeremy has been an invited speaker to several conferences to talk about this work in more depth.”

Meanwhile, The Case of the Vanishing Chaperone, a.k.a., “Remote homology and functional genetics unmask deeply preserved Scm3/HJURP orthologs in metazoans,” also was published in Science Advances.

“Huge chunks of the puzzle missing, yet the pieces were kind of there and I think it required somebody like Jeremy to come in and really fill in all the pieces,” Malik said.

It also required a four-lab collaboration with researchers studying different model organisms — the kind of intellectual adventure that defines the spirit of the Basic Sciences Division at Fred Hutch.

“I honestly believe it would be very difficult to imagine this occurring anywhere else,” Malik said.

Hollis concluded his PhD defense by thanking a long list of the Watsons throughout his life who had helped him solve his mysteries; their contributions to his success were anything but elementary.

Then everyone was asked to leave the auditorium except for the committee that would decide whether Hollis had done enough to earn his doctorate degree.

They didn’t meet long, and they soon joined the reception outside for Dr. Jeremy Hollis.

The work described in both Science Advances publications was supported by grants from the National Institutes of Health, the National Science Foundation, The Pew Charitable Trusts’ Biomedical Scholars Program, the Mahan Fellowship, Paul Neiman and the Translational Data Science Integrated Research Center awards from the Fred Hutch Cancer Center,  Fred Hutchinson Cancer Center Electron Microscopy shared resource, the Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium, which includes the Electron Microscopy, Genomics and Bioinformatics, Flow Cytometry, and Proteomics and Metabolomics shared resources, and the Howard Hughes Medical Institute.