Latest News


Computer-Designed Rocker Protein World's First to Biomimic Ion Transport

Published on April 13, 2015 by Jorge Salazar

Rocker protein world's first engineered to transport ions (pink). Credit: Andrei N. Lupas Science 2014;346:1455-1456.

For the first time, scientists recreated the biological function of substrate transportation across the cell membranes by computationally designing a transporter protein. The designed protein, dubbed Rocker, was shown to transport ions across the membrane, a process crucial to cell and organismal survival in various functions, such as nutrient intake, efflux of waste or drug, and cell signaling, for instance, between nerve cells in the brain and spinal cord.

Michael Grabe, Associate Professor, Department of Pharmaceutical Chemistry and the Cardiovascular Research Institute, University of California, San Francisco.
"To our knowledge, this is the first transport protein designed from scratch - that is, it didn't exist in nature beforehand," said study co-author Michael Grabe, an associate professor in the Department of Pharmaceutical Chemistry and the Cardiovascular Research Institute at the University of California, San Francisco.

This research has wide potential application, such as targeting medicines more specifically into cancer cells and driving charge separation potentially for harvesting energy for batteries.

The researchers published their results in the journal Science in December 2014.

"To our knowledge, this is the first designed transport protein from scratch, that is it didn't exist in nature before hand."
Michael Grabe, U.C. San Francisco.
The engineered Rocker protein acts like a tiny gate, designed so that zinc ions and protons can flow in a controlled way across the lipid-membrane barrier around the cell-like vesicle.

A helically-shaped short chain of amino acids was designed to assemble into a four-helix bundle, like four springs bundled around a central axis, and to form two special pockets for binding zinc ions and protons along the cavity within the bundle. The individual helices embed into the lipid membrane where they form the bundle, and the engineered system works like natural transport proteins found throughout the human body.

William DeGrado, Professor, Department of Pharmaceutical Chemistry, University of California, San Francisco.
Rocker was designed to perform this function by doing one thing — 'rocking' between two different shapes, or conformations. One conformation opens up the pocket near one side of the membrane to grab zinc ions or protons. Once the zinc ion binds to the pocket, Rocker changes its shape to close off the pocket, while opening up the second pocket near the other side of the membrane. This allows the ions from the closed pocket to travel to the second pocket before being released to the outside of the membrane.

The catch is that Rocker can't have both pockets bind the ions at the same time, nor permit the cavity to open all the way through the membrane at one time because this would leak down the ion concentration levels important for keeping cells intact and healthy. Like a gatekeeper, the engineered transport protein controls what moves in and out of the cell from one side of the membrane to the other.

Nate Joh, Postdoctoral Researcher, Department of Pharmaceutical Chemistry, University of California, San Francisco.
The experiment began with designing of the protein from scratch.

"Bill Degrado and his postdoc, Nathan Joh, hypothesized that the complex transportation function could be generated through engineering of this simple ‘rocking' motion, and he spearheaded in testing the idea," Grabe said. The design was all done computationally by Joh and the study co-author Gevorg Grigoryan, said Grabe. "They came up with amino acid sequences computed to form the desired structure. Then, the peptide was chemically synthesized, put into the membrane, and tested whether the functional bundle was formed."

"What we did use TACC for was to take the predicted model and carry out molecular dynamics simulations on it."
Michael Grabe, U.C. San Francisco.
The protein was crystallized in lipid-mimicking substance and diffracted by x-ray to verify, at the atomic level, that the desired structure was adopted. The scientists further probed the structure and dynamic nature of Rocker embedded in lipid and in detergent by using nuclear magnetic resonance (NMR). Also, Rocker reconstituted in membrane vesicles was tested to show that it really pushed zinc ions from one side of the membrane to the other, while protons traveled the other direction.

"Designing this protein is an amazing accomplishment made possible by bringing together scientists with complementary areas of expertise."
Jean Chin, National Institute of General Medical Sciences, National Institutes of Health.
"Designing this protein is an amazing accomplishment made possible by bringing together scientists with complementary areas of expertise," said Jean Chin of the National Institutes of Health's National Institute of General Medical Sciences, which funded the work through grants to several of the paper's co-authors. "The highly collaborative team used its deep knowledge of the structures, mechanisms and interactions of known transporters to overcome several barriers in protein engineering, including membrane protein folding."

Simulations on the Stampede supercomputer of the Texas Advanced Computing Center (TACC) bridged the gap between the drawing board and the finished engineered protein. "What we used TACC for was to take the predicted model and carry out molecular dynamics simulations on it," Grabe said. The molecular dynamics simulations were about a microsecond of aggregate simulation time with classical force fields using the NAMD software program.

Schematic of the goal of obtaining conformational exchange between two oppositely oriented symmetry-frustrated states without being trapped in a symmetric state with both sites simultaneously occupied. The goal of protein design of Rocker was to get conformational exchange between two opposite states (left and right) without being trapped in a symmetric state with both binding sites occupied (middle). Credit: Joh et al.
The computer allocation was made through XSEDE, the Extreme Science and Engineering Discovery Environment, a single virtual system funded by the National Science Foundation (NSF) that scientists use to interactively share computing resources, data and expertise. "For my research lab, my XSEDE resources are absolutely essential," Grabe said.

Protein engineering lags far behind the genetic engineering of modifying DNA that has been around since the 1970s. Though his team's results have been described in Science as a "landmark study," Grabe views it as a first step in advancing the new discipline of protein engineering to the level of genetic engineering, and beyond.

"For my research lab, my XSEDE resources are absolutely essential."
Michael Grabe, U.C. San Francisco.
"Next, I'd like to use enhanced sampling techniques to more fully capture the entire transport process using Stampede resources," Grabe said, a task well-suited for the over 500,000 cores of the Stampede supercomputer.

Grabe was cautious about the potential application of this research, but he did say there could be benefits to society in areas such as energy and medicine.

"We envision that this protein can create an electrochemical gradient using things like pH, using protons. One can imagine in a totally non-cellular case that one could potentially harvest this kind of pumping to create things like batteries. This might not happen in the very near future, but things like this could be used to do some kind of energy harvesting," Grabe said.

The X-ray crystal structure of Rocker revealed a tight interface (axial view top left) and metal binding sites (right and lower left expanded). Red, green and cyan show different packing environments superimposed on the dimer subunit of the modeled design. Credit: Joh et al.
Moving things in and out of living cells would be even harder, Grabe said. "You have a very hard problem of spatial and temporal specificity, making sure that the transporter that you've made ends up in the right place at the right time. But that's a very interesting goal that you might have also in the future, that is, design your sequence to go to a particular membrane of a particular cell type, say, a cancerous cell with a stereotypical membrane environment, and load it full of zinc, or run down a chemical gradient that's already there. To an extent, bacteria do this already. Things like antimicrobial peptides cause cells to lyse or blow open cells by creating holes in their membranes. But if you could transport something into the cell such as a toxic ion or small molecule that could be quite interesting," Grabe said.

Among the science team from U.C. San Francisco were study lead Nathan Joh, Manasi Bhate, Yibing Wu, Michael Grabe, and William DeGrado. Researchers also included Mei Hong and Tuo Wang from the Massachusetts Institute of Technology; Rudresh Acharya of the National Institute of Science Education and Research, Bhubaneswar, Odisha, India; and Gevorg Grigoryan of Dartmouth College.

The National Institutes of Health, the Alfred P. Sloan Foundation and the National Science Foundation provided the bulk of the funding.

Story Highlights

Computationally-designed protein world's first to mimic biological function in cell membrane.

Engineered protein dubbed Rocker rocks between two conformations to gate ions across cell membrane in controlled fashion.

XSEDE resource Stampede supercomputer of TACC simulated molecular dynamics in protein design.

Potential applications include energy harvest for batteries and medicine delivery to cancer cells.


Faith Singer-Villalobos

Communications Manager | 512-232-5771

Aaron Dubrow

Science And Technology Writer

Jorge Salazar

Technical Writer/Editor | 512-475-9411