Luis Vidali, left, and Erkan Tüzel examine a sample of Physcomitrella patens, a moss they use as a model organism.
by Michael I. Cohen
By design, WPI approaches life sciences research in a collaborative way, driven by a desire to solve important problems that often cross departmental boundaries. That model also applies to recruiting new faculty. As the university’s life sciences program has grown significantly in recent years, department heads have thought strategically, targeting emerging, multifaceted disciplines and reaching out as a group to recruit the right talent. One such success story at WPI is the growing concentration of work in biophysics.
The convergence of lab-bench experimentalists and mathematical theorists, biophysics explores the systems contained within the cell, where a complex web of physical interactions drives biological processes. Working across traditional academic borders, biophysicists apply molecular modeling to the search for new cancer therapies, study electric fields that may herald the onset of Alzheimer’s disease, and create mathematical models to understand the basic process of cell growth.
Structural Dynamics in Plant Cells
It didn’t take long for Erkan Tüzel and Luis Vidali to find each other. Both were recruited to WPI in 2009, and even before they had set up their respective offices, they were talking about how to collaborate.
With support from the Eppley Foundation for Research, Tüzel, assistant professor of physics, studies the dynamics of microtubules — strong filaments that give cells their structure — and molecular motors that transport proteins along microtubules like trains traversing an intercellular railway. Vidali, assistant professor of biology and biotechnology, uses molecular genetics to explore the basic processes of plant growth, which also involve filaments and motors. “We were interested in similar problems, so we decided to try to solve some together,” says Vidali.
Their first collaboration focused on a plant gene called myosin XI, which Vidali studies in moss. The gene encodes a protein that works as a molecular motor that can move material along the filaments that form a plant cell’s cytoskeleton. Experiments in Vidali’s lab had shown that when both copies of the gene are turned off, the cells fail to grow properly. Instead of growing from their tips, with new cells extending in a linear fashion, moss with silenced myosin XI genes produced small rounded cells that left the plants severely stunted.
"Bringing physicists and biologists together teaches us all how to communicate with each other. That improves all of our understanding and helps build more effective collaborations.”
— Erkan Tüzel
The next step was to test whether the lack of myosin XI affects the dynamics of the cytoskeleton, which is made of filaments of a protein called actin. While different from the microtubules Tüzel studies, the dynamics of actin filaments are similar enough so he could quickly adapt some of his mathematical models to fit the problem. Vidali tagged the actin filaments with a fluorescent marker that could be seen through a confocal microscope. He took images of the movement of actin over time, which became data points for a model Tüzel developed to analyze the movement of actin and correlate it with the loss of myosin. The model showed that the missing protein did not affect the movement of the cytoskeleton, even in the stunted plants, suggesting that myosin XI may play a more complex role in plant cells than was previously realized.
Based on that initial project, Tüzel and Vidali have moved on to extended questions about the functioning of microtubules and molecular motors. “Through this collaboration, Erkan can invest his resources in the computational part, and my lab can invest resources in the experimental part, so we can approach the problem from both directions and potentially see better results faster,” Vidali says.
Beyond working together as colleagues, Vidali and Tüzel regularly bring their students and postdoctoral researchers together to discuss their progress and to look for new avenues to pursue. “Bringing physicists and biologists together teaches us all how to communicate with each other,” Tüzel says. “That improves all of our understanding and helps build more effective collaborations.”
The Physics of Alzheimer’s
Characterized by progressive dementia, memory loss, and cognitive impairment, Alzheimer’s is a fatal disease that claims its victims bit by bit. “To me, this is the cruelest disease,” said Izabela Stroe, assistant professor of physics, who came to WPI in 2008. “If you lose an arm or a leg, that’s terrible, but you can go on. But if you lose your mind and your memories and your thoughts, what is left?”
Stroe doesn’t approach this subject in the abstract. Growing up she learned mathematics from her uncle, a high school teacher who developed Alzheimer’s. “He was a great teacher, and I watched him slowly lose everything — even the ability to recognize his own family,” she says. “I greatly admired him for all of the things he knew, and in the end everything was gone. I do this work because I hope that one day I can help people like my uncle who are afflicted by this terrible disease.”
According to the National Institutes of Health, some five million Americans have Alzheimer’s. The prevalence of the disease doubles every five years beyond the age of 65, so the incidence is rising as the population ages — it is expected to afflict between 11 and 16 million Americans by 2050. The disease begins when small toxic chains of amino acids (the building blocks of proteins) start to associate into larger and larger structures. They coalesce into fibrous bundles that attach to the neurons, eventually covering them in plaque that blocks the transmission of signals. The precise cause of Alzheimer’s is unknown, and there currently is no way to diagnose the disease until the brain is seriously damaged. Stroe, in collaboration with researchers at the University of California, Davis, is developing technology that she hopes will lead to a method for early detection. “If we can do this,” she says, “then perhaps that can lead to better therapies.” The plaque in the Alzheimer’s brain is made primarily of a large protein called amyloid-beta. The building blocks of amyloid-beta are small molecules called amyloid precursors. Stroe is working to measure the precursors in the bloodstream and to detect when they first begin to aggregate.
As the precursors form bundles, some of the water molecules that were attached to them get squeezed out. Stroe believes she can detect the movement of those molecules with a technique called dielectric relaxation spectroscopy. The samples are placed in a capacitor and exposed to an electric field, which produces a shift in the sample’s dielectric relaxation spectra. The shifts seem to correlate with the aggregation of precursors into bundles.
“This is not a trivial problem,” she says, “but we have early data that is supportive of our approach, and this is encouraging.”
Fighting Cancer with Molecular Modeling
George Kaminski’s research is aimed at finding new ways to fight cancer. An associate professor of chemistry and biochemistry who came to WPI in 2008, Kaminski applies the laws of physics to the movement of molecules in human cells. Using advanced mathematics and a lab filled with high-powered computers, he seeks to supercharge the process now used to develop new pharmaceuticals by creating more accurate simulations of how new drugs might bind to proteins.
Proteins, the building blocks of all cells and tissues, also regulate most physiologic processes. The over- or underproduction of proteins, or the creation of toxic proteins by mutated genes, are behind many diseases — including cancer. Often, the physical structure and chemical make-up of toxic proteins are known. The challenge for drug development is finding compounds that will bind to harmful proteins in highly specific ways that will prevent them from causing damage.
Compounds that bind to proteins are called ligands. Pharmaceutical companies have libraries of millions of compounds that could be therapeutic ligands against particular proteins. Through a complicated experimental process, they add them to the target protein to see if they bind. “And just binding isn’t enough,” Kaminski says. “If a ligand binds to everything, then it’s toxic. You want a ligand that binds only to the bad protein — like a lock and key.”
Kaminksi’s approach to matching lock and key is based on robust physics implemented in computer code and is aimed at significantly cutting the need for more expensive and slower wet chemistry research. In theory, one can characterize the energy fields of the protein and the ligand and predict if they will bind — and how tightly — using quantum mechanics and other principles of physics. Electrons play a critical role in molecular binding, but accounting for the movement of every electron in just one protein-ligand combination is beyond the reach of current computational technology. So Kaminksi and his team must develop mathematical models that simplify the system, yet retain the essential physics that will yield an accurate answer. “It’s all about approximation and figuring out what you need to retain in the model,” he says.
Even this simplified approach entails serious number crunching; some of the calculations run for several days on fast computers. Over the next several years, Kaminski, who is funded by the National Institutes of Health, hopes to be able to predict protein-ligand binding based on the chemical and structural parameters of specific compounds. If successful, he says his approach will speed up the drug development process by narrowing the field of likely cancer-killing compounds worth testing.
“In recent years we’ve seen a great increase in computational power that could be applied to biologic systems, but creation of an adequate toolset has been lagging,” Kaminski says. “We are trying to build a bridge by combining the right physics with this new computational power.”