A study collaborates between three labs at UC San Francisco has resulted in an unprecedented look at a member of a vital and ubiquitous class of proteins called integrins (pronounced "INT-uh-grins"). Integrins are associated with fibrosis, scarring, and stiffening of tissues that is associated with nearly half of all deaths in developed countries, and yet researchers had no high-resolution structural model of the proteins in their active state.

Now, a combination of perseverance, technological achievement and insight have pinned down an elusive moving target. The study was published in Nature Structural and Molecular Biology includes Campbell's work, genetic manipulation, and protein engineering, purification.

Older techniques like X-ray crystallography require researchers to undergo laborious processes to pack proteins into crystals before they can make images to determine a protein's structure. This method works best on stationary, rigid, and symmetrical proteins: the opposite of integrins, which are quite flexible in their active form.

Integrins

Integrins are embedded on the surfaces of all animal cells, connecting each cell to its surroundings and allowing it to communicate and respond to external forces.  To encounter its targets, the new work suggests for the first time that an active integrin bends and sways at a flexible midpoint "like a sunflower seeking the sun."

Cryo-Electron Microscopy

To explore an integrin's structure, the team used cryo-electron microscopy, a technique that has recently benefited from major advancements in hardware and software at UCSF.  Melody Campbell, Ph.D., worked to visualize one type of integrin protein down to near-atomic precision. She imaged and analyzed the purified and frozen proteins in the lab of Yifan Cheng, Ph.D., a professor of biochemistry and biophysics at UCSF and the other senior author of the study.

But visualizing the protein was only part of the effort. Once the protein was visualized, the researchers validated their structural model by genetically engineering a related integrin that responded to biochemical cues exactly as the team's model predicted, suggesting that their findings extended to many, if not all, integrins.

Treatments

The authors have already developed several promising therapeutic antibodies, using the new structure as a template. Some companies are already working with those antibodies to develop treatments for conditions like cancer and fibrosis. 

But for Nishimura, who has been working with integrins for more than two decades, the detailed model is also personally satisfying: It's like seeking an old archnemesis, and finally freezing him in his tracks.