Edited by: Jennifer Fisher Wilson
Courtesy of John Kuriyan
Comments by Michael J. Eck, assistant professor of biological chemistry and molecular pharmacology, Harvard Medical School, and department of cancer biology, Dana-Farber Cancer Institute
F. Sicheri, I. Moarefi, J. Kuriyan, "Crystal structure of the Src family tyrosine kinase Hck," Nature, 385: 602-9, Feb. 13, 1997. (Cited in more than 160 papers since publication)
Comments by John Kuriyan, professor of molecular biophysics, Howard Hughes Medical Institute, The Rockefeller University
Decades of research have revolved around the Src-family kinases, a closely related group of nonreceptor tyrosine kinases including Src, Hck, and Lck that play a key role in eukaryotic signal transduction. The Src kinases are recognized as important targets for therapeutic interventions to treat cancer and immunological diseases, but a better understanding of Src's tyrosine kinase structure and functioning was needed to guide drug development.
"Studies of its function had told us how it had to work, but no one had ever seen how it works," recalls Harvard's Michael Eck.
Researchers had established the importance of Src's signaling proteins and the importance of the Src homology domains, SH2 and SH3, the molecular handshake devices that mediate the protein-protein interactions during signal transduction. But exactly how the domains regulated interactions was a mystery.
Serendipitously, two research groups, one at Children's Hospital and Harvard Medical School and one at Rockefeller University, illustrated the general principles of Src-family kinase regulation and signal transduction at the same time, publishing back-to-back papers in Nature in 1997.
Eck and colleague Wenqing Xu, working with Stephen Harrison at Children's Hospital, determined the structure of a large fragment of the c-Src tyrosine kinase, comprising the regulatory and kinase domains and the carboxy-terminal tail, at 1.7 angstroms in a closed, inactive state. John Kuriyan and his colleagues at Rockefeller University illustrated the crystal structure of Hck at 2.6/2.9 angstroms.
"Since the structure is so unexpected, the fact that the groups came up with the same architecture was very reassuring," Kuriyan says. Eck remarks, "We both have found the same answer to the question of how this kinase works. That another flavor of Src molecule shows exactly the same structural features strengthens both papers and confirms that the observed domain organization is relevant to all Src-family members."
Both studies showed that, as expected, although Src-family tyrosine kinases have catalytic machinery that is very similar to the hundreds of different tyrosine kinases, their regulatory mechanisms differ significantly from those of other tyrosine kinases. Instead of a simple mechanism in which the regulatory domains control activity by directly binding to and altering the catalytic machinery, they found that the Src homology domains, SH2 and SH3, affect activity by pulling on the catalytic region in a complicated manner.
The interactions among SH2 and SH3, stabilized by binding of the phosphorylated tail to the SH2 domain, lock the molecule in a conformation that disrupts the kinase active site and sequesters the binding surfaces of the SH2 and SH3 domains. By turning in on itself and inhibiting Src catalytic activity, the kinase locks into an inactive state.
Courtesy of Michael J. Eck
Both Kuriyan and Eck began studying the structure of the Src-family tyrosine kinase by using fragments of the Lck protein. But Kuriyan found the Hck to be easier to work with ultimately, and Eck chose to work on Src. Src family members share a common regulatory mechanism but differ in cellular expression and localization.
Both researchers say that when they were finally able to see the structure of the Src and Hck tyrosine kinases, they were amazed at its complexity and at the way the domains interacted to influence catalytic activity.
Eck and his colleagues did not expect to find the regulatory SH3 and SH2 domains on the "back" of the kinase domain, on the side opposite the enzyme's active site. They were also surprised to see that the so-called C helix was displaced from its active position near the catalytic cleft, an inactivating feature also seen in the cyclin-dependent kinase Cdk2.
Eck was most impressed to see the way that Src inactivated itself: "What is so remarkable about Src is that, in the absence of good binding partners for its SH2 and SH3 domains, it folds up and turns off, and when interacting with the right binding partners, it opens up again. Nature has devised a very clever way of coupling Src's catalytic activity to its appropriate subcellular location."
Kuriyan also reported that his group was surprised by some of their findings. They discovered that a tyrosine residue (Tyr 527), known to be crucial in regulation because it binds to the SH2 domain, is located far from the catalytic center (40 angstroms). It works by exerting its influence over a distance, according to Kuriyan.
The Rockefeller group was also surprised to discover that the SH3 domain binds to an internal structure known as a polyproline type II helix, which serves as a docking site for the SH3 domain. "This was well known to be the standard recognition motif for SH3 domain, but despite all of the years of staring at the sequences of Src kinases, none of us had guessed that the protein contains an internal polyproline helix," Kuriyan explains. "This shows how much we understand protein structure."
The papers by the Harvard and Rockefeller groups supported one another, even in such unexpected findings as these. Two years after the publication of these papers, both Eck and Kuriyan continue to work to reveal more about Src-family tyrosine kinases. Both research groups are now studying more deeply the structure of the Src-family kinases and how they are regulated. They believe that phosphorylation of Tyr 416 is crucial for activity.