Kathleen L. Gould and Walter J. Chazin are leading the VUMC research. Their respective expertise in cell biology and structural biology made the discovery possible. (photo by Dana Johnson)
Protein may play dual role in preventing disease: VUMC study
Ribbon diagrams illustrating the structural similarity between the Prp19p U-box domain (a) and the c-CBL RING finger domain (b). The green spheres represent the networks of hydrogen bonds in U-box domains that are substituted for the zinc ions (red spheres) found in RING domains. (Illustration courtesy of Walter Chazin)
Coupling their expertise in structural biology and cell biology, a group of Vanderbilt scientists has made a novel discovery implicating a single yeast protein in two distinctly different cellular functions previously unknown to be related. The findings, which are expected to have a parallel in humans and possible implications associated with inflammatory diseases and cancer, shed light on how the protein facilitates normal pruning of a cell’s genetic instructions, and may play a role in disposal of cellular debris.
Researchers in the laboratories of Kathleen L. Gould, Ph.D., professor of Cell and Developmental Biology, and Walter J. Chazin, Ph.D., professor of Biochemistry and Physics, identified the molecular structure of a U-box domain in the essential mRNA splicing protein known as Prp19p, and found characteristics implicating it in the cellular degradation process known as ubiquitination. A report of the work currently appears online in Nature Structural Biology.
Without contributions from both fields, structural and cell biology, the discovery would not have been possible, according to Gould, who is also an Associate Investigator of the Howard Hughes Medical Institute. Knowing the structure of the protein makes answering questions about what Prp19p does in the cell much easier, she said.
“This is the first of what I hope will be many, many examples of the value of structural biology in the biological and medical sciences,” said Chazin, who established and directs the Center for Structural Biology.
Prp19p is a splicing factor, one of a number of proteins that make up the “molecular machine” in the cell’s nucleus called the spliceosome. There the earliest products of DNA transcription, called pre-mRNAs, are trimmed before being shipped out to the cytoplasm, where the modified mRNAs spell out the recipe for protein synthesis.
Gould’s lab was studying Prp19p’s role within the spliceosome when she stumbled on a publication identifying it as one of a number of proteins having genetic sequences identical to those encoding a U-box in a protein known to have E4 ubiquitin ligase activity. (The ubiquitination process requires the sequential action of up to four enzymes — E1, E2, E3, and E4 — to add activated ubiquitin to the substrate marked for destruction.)
“We immediately got very excited about the possibility of Prp19 functioning as a ubiquitin ligase,” said Gould.
As a first test, Gould said, they asked how closely the 3D structure of the U-box resembled that of the RING finger, another domain known to be associated with protein ubiquitination.
“And that’s what this work shows,” she said, “that this domain does look like a RING finger, and in vitro, at least, Prp19p acts like a ubiquitin ligase, providing the possibility that this whole subcomplex in the spliceosome is involved in ubiquitination.
“Ubiquitination has never been linked to splicing before,” she added, “so we have no idea of what the targets might be — whether it targets proteins for destruction or modifies the function of something — but a function is now implicated for this set of proteins.”
Chazin’s part in the project grew from his interest in multi-protein complexes, or “molecular machines,” as they are sometimes called, that execute sophisticated biological tasks.
“As this project defined itself, it really caught fire in my mind,” he said. “Knowing that there is this big, interconnecting network of protein-protein interactions has captured my interest.”
Using a structural biology tool called nuclear magnetic resonance, or NMR, the researchers found the atomic structure of Prp19p to be remarkably similar to the RING finger proteins, with a few interesting differences. At its core, the RING finger domain requires two zinc atoms to maintain its stability and function. The U-box domain, on the other hand, has the same stabilizing zone, but instead of zinc atoms, there is a network of hydrogen bonds.
The hydrogen bond network in the U-box domain makes it a more flexible structure, Chazin says, which will likely allow Prp19p to be more dynamic in its interactions with other proteins than a zinc-based RING finger protein.
“It’s sort of like using glue vs. using a magnet,” he said. “The zinc acts like a nice epoxy to hold things together. But just like the epoxy, it’s kind of rigid. The hydrogen bond network is much more dynamic, because hydrogen bonds are made and broken easily.”
This kind of flexibility should make the protein better at binding or changing conformation, which would come in handy in light of Prp19p’s need to interact with various proteins during the multiple steps involved in RNA processing, as well as during ubiquitination.
Comparing the U-box with the RING finger domain, it’s clear that the two share similar residues at the E2/E3 binding interface critical to ubiquitin ligase activity in the RING domain. When those residues are mutated in the U-box, it doesn’t perturb the structure, but function of the protein is lost.
“This loss suggests that the interface is, in fact, used by Prp19p to bind to something that is critical for its function,” said Gould. “These mutations will allow us, now, to go genetically in search for the E2 enzyme that’s interacting with Prp19p. Making the mutations at random would not have been useful in identifying interacting partners.”
“Rather than having to mutate every residue in the whole domain,” added Chazin, “it’s a good example of how using structure can help to focus the biology.”
“If we are able to identify the E2 enzyme genetically,” Gould continued, “then we can come back and ask structurally, does this (interaction) change the U-box?”
The ubiquitination process, it turns out, may be involved in more than just removal of cellular flotsam and jetsam. In the last few years, a number of non-traditional roles of ubiquitination have been discovered, from involvement in inflammatory processes to DNA repair, and it has become a hot field of study, one that Gould and Chazin expect will heat up a little more as a result of their findings.
“Everyone who was looking at the RING finger domain as the one for ubiquitination now has the U-box to consider,” said Gould. “This will add a whole other complexity to the situation.”
The current work is the result of the equal efforts of Melanie D. Ohi, Ph.D., while she was a graduate student in Gould’s lab, and Craig W. Vander Kooi, M.Sc., who is currently a third-year graduate student in Chazin’s lab. Graduate student Joshua A. Rosenberg was instrumental in completing experiments in Gould’s lab required after Ohi left for a post-doctoral position at Harvard.
“It was Melanie’s dream to learn NMR, and it was wonderful that she was allowed that opportunity,” said Gould. “There are not many other institutions where it’s possible to spend half a day doing cell biology and genetic and molecular analysis of a protein, and the other part of the day…linking the backbone structure up.”
“It’s very much my philosophy of how I think structural biology will need to be done in the future,” Chazin added. “It will be done more by people like Melanie and Craig because the real power is using the structural perspective to address the biological problem.
“This project demonstrates the potential I am seeking to exploit in the developing structural biology program at Vanderbilt. We will be ahead of the curve, nationally and internationally, because this is something unique we can do here.”
The studies were funded by grants from the National Institutes of Health and the Howard Hughes Medical Institute.