In every living cell, specialized proteins work to transport materials across the cell’s membrane, moving them in and out of the cell. These transport proteins are at the molecular-scale, so the motions are extremely small and, therefore, difficult to detect. However, understanding these molecular movements during the transport cycle holds the key to more effective drug design when fighting neurological and metabolic disorders, and even cancer.
Emad Tajkhorshid, a professor of biophysics and biochemistry, loves the challenge of “going after uncharted territories in the molecular world.” Visualizing the entire process of protein transportation across membranes has taken Tajkhorshid and colleagues a number of years, but they have met the challenge: They are the first in the world to be able to describe at an atomic level the entire motion of a transporter protein during its cycle.
“Up until this, all we knew was what the protein looked like at the beginning of the transport cycle, and what it looked like at the end,” said Tajkhorshid. “Based on our advanced simulation results, now we know, step by step, how it opens, grabs a substrate, closes, and opens the other end to let the substrate out.”
If we understand how this protein functions, maybe we can design an inhibitor that stops that mechanism and makes anti-cancer drugs more effective-Emad Tajkhorshid
One of the transporter proteins studied by Tajkhorshid is a protein found in a cell membrane that has evolved to pump drugs, usually those toxic to the cell, out of the cell as a protective mechanism. The toxic drugs, however, could be those that are targeting cancer cells, and the action of the protein causes the cells to develop resistance to those potentially life-saving drugs.
“If we understand how this protein functions, maybe we can design an inhibitor that stops that mechanism and makes anti-cancer drugs more effective,” said Tajkhorshid. “It’s a very important area of drug design and development—to know how and to optimize the way the drugs are absorbed, distributed, and metabolized in the body, the processes which are all affected by transporters.”
In an experimental lab, visualizing these protein movements is impossible—the transport cycle involves changes that are too small and too fast to observe naturally. Tajkhorshid specializes in computational approaches instead, using computers to solve the mystery of how proteins move and how these motions furnish efficient and selective transport of materials.
“We tell the computer what the rules are in terms of what is physically and biologically possible. The rules are based in theoretical physics, chemistry, spectroscopy, and other frameworks, and the computer can determine the movement based on those rules,” said Tajkhorshid. “However, weighing a program down with rules of behavior can cause it to become slow. The computation of such complex systems and complicated processes can become too much for the program to handle.”
In order to overcome the slow processing speed, Tajkhorshid uses a series of advanced computational algorithms, coupled with the power of supercomputers, including the Blue Waters supercomputer located at the National Center for Supercomputing Applications (NCSA) at the University of Illinois and Titan at the U.S. Department of Energy Oak Ridge National Laboratory.
To take advantage of the full power that the supercomputers offer, Tajkhorshid’s group—comprised of computer scientists, biophysicists, physical chemists, and others—developed a novel computational algorithm that helps them determine the route the transporter protein is most likely to take when moving its cargo across the cell membrane.
“We devised a new computational approach that is based on the idea of leveraging the large capacity of supercomputers to run many simulations simultaneously,” said Tajkhorshid. “Before we attempt to calculate the protein structural change and movement during the transport cycle, we have to explore a large number of initial potential transition pathways—a very efficient process on supercomputers—and then we develop theoretical frameworks that determine which one of these pathways has the greatest potential to be the pathway for the very long final calculation. Then we just run that big calculation once, saving us a lot of time and computing power.
“By design, the technology of computer simulation allows us to go beyond what can be done by experiments, but, in order to anchor our results in reality, we have made it a practice to make testable hypotheses that can be validated experimentally.”
Tajkhorshid’s options for what protein to study are “nearly limitless”—in this era of modern structural biology and biophysics there are endless simulations that will give scientists deep insight into the mechanism of protein nanomachines and other exciting biological systems.
“That’s the thing about membranes—everything that gets into the cell, or out of it, has to cross a membrane,” Tajkhorshid said. “If you take a drug that’s supposed to go from the stomach to the blood, it has to cross membranes several times. The drug’s ability to absorb into membranes directly impacts its effectiveness, so we need to look at its interactions with membranes and membrane proteins that are often actively involved in trafficking drugs across the membrane.”
As one of the leaders in computational imaging, Tajkhorshid’s team always has new problems to solve. “We’re really excited about the work we do—in all our projects, we stay relevant to the problems of society,” said Tajkhorshid.
Tajkhorshid has been at the Beckman Institute since he joined the Theoretical and Computational Biophysics Group as a postdoc. In 2007, Tajkhorshid earned a position as a tenure-track professor and started the Computational Structural Biology and Molecular Biophysics Lab. He was promoted to a full professor in 2013.
His current work focuses on computational simulations, but his early studies were focused on chemistry, math, and physics. In his native country of Iran, Tajkhorshid received his doctorate in pharmacology, and he earned a second Ph.D. in biophysics in Germany.
“It was clear to me that by focusing on computation and molecular simulation, I could take advantage of all those skills,” Tajkhorshid said. “That’s how I became interested in molecular simulation. When I came here as a postdoc, I got really excited by how much you can see with computational and simulation methods—how many details you can see as to how molecules may work.”
During his time as a postdoc at TCBG, he demonstrated how water molecules pass through a channel across a membrane, a result which was published in Science.
Leading his own lab has added to his list of accomplishments, with the opportunity to teach and mentor students being at the top of his list, Tajkhorshid said.
“I love to teach,” he said. “I spend a lot of time doing research, but I also love to transfer the acquired knowledge and to interact with my students. We have some of the brightest young minds on this campus, and it’s very exciting and rewarding to work with them.”
The combined expertise of his students has led to his group’s success in computationally understanding how protein transporters work, something he hopes will continue to expand and develop in the future.
“Creating a reliable model for membrane protein transport was the combination of having thought about a really important problem for an extended amount of time, several years even, and collaborating closely with people who are truly interested in and excited about understanding how transporters work,” he said. “I have excellent researchers in the group and access to unique computational resources. It all really came together.”
M. Moradi, G. Enkavi, and E. Tajkhorshid (2015) Atomic-Level Characterization of Transport Cycle Thermodynamics in the Glycerol-3-Phosphate:Phosphate Antiporter. Nature Communications, 6:8393.
M. Moradi and E. Tajkhorshid (2013) Mechanistic picture for conformational transition of a membrane transporter at atomic resolution. Proceedings of the National Academy of Sciences USA, 110: 18916–18921.