March 03, 2017
By Kristi Birch
Imagine if doctors could diagnose and treat patients not only according to the standard protocols for particular conditions, but also by considering each patient's unique biology. For some diseases, this is already happening. For example, oncologists are using molecular tests to determine which treatment might work best for a cancer patient on the basis of whether that patient has a particular genetic mutation.
While there has been enormous progress in the diagnostic process, fueled largely by the completion of the Human Genome Project in 2003, which mapped out the order of the 3 billion bits of DNA that comprise genes and create a human being, this kind of personalized medicine - also called precision medicine - is in its infancy. Scientists still have much to untangle about what genetic variations or defects make people prone to particular diseases and why some drugs work for some people but not for others.
The field of personalized medicine stands to grow even more, thanks to a research team led by the University of Kansas Medical Center that received a $1 million grant from the W. M. Keck Foundation to develop a set of rules that will predict how specific amino acid mutations alter the functions of proteins in the body.
One of the largest philanthropic organizations in the United States, the Keck Foundation funds basic science, engineering and medical research and has a reputation for supporting cutting-edge projects.
Genes produce proteins, large molecules that perform many thousands of tiny tasks that make the human body function properly. Proteins matter because they do the work of the cell. Without them, muscles wouldn't contract, antibodies wouldn't fight infections and hair wouldn't grow. Understanding how mutations change the way proteins behave is crucial for the future of personalized medicine as well as for protein engineering.
Proteins are made of strands of amino acids, hooked together like beads on a string. The sequence of these acids determines the 3-D shape into which the protein folds. Each protein's folded shape determines what function that protein does. If a protein is mutated, or altered, then the shape and thus the function of the protein changes. The problem for scientists is figuring out exactly how the function is altered.
"Any two unrelated people can have up to 10,000 differences in their protein sequences," said Liskin Swint-Kruse, Ph.D., who co-leads the project with Aron Fenton, Ph.D., both associate professors in KU's Department of Biochemistry and Molecular Biology. "It's easy to figure out where the proteins are different [i.e., where there are mutations]; it's hard to figure out what the changes do."
Paul Smith, Ph.D., another researcher on the project and a professor in the Department of Chemistry at Kansas State University, says the key is for scientists to establish the rules of behavior for the various mutations. "Everyone's genome is a little different, and it can have big effects on how they respond to treatment, different drugs or diseases," he said. "For example, if one amino acid is substituted for another, what's going to happen? And how will that affect how the protein functions? If we can control and understand that, the long-term view is personalized medicine."
Making It Compute
Computer algorithms already exist that can find mutations and predict the functional outcomes, but in each given protein they are only correct about half the time. These programs can only accurately predict outcomes for amino acid positions that haven't been altered by evolution. These "conserved" positions abide by a set of mutation rules, gleaned over a half-century of research, whose findings have been programmed into the algorithms that predict mutation behavior.
KU researchers have found that for the nonconserved positions, those that have been altered by evolution, none of the rules apply. The research team coined the term "rheostat positions" to describe these positions, because their changes can dial the function of a protein up or down, like dials on a stereo.
During the course of the five-year grant, Swint-Kruse and her group will study nonconserved amino acid positions in proteins that represent three diverse structural classes to create a library of rheostat positions; gather biophysical data about them; and then use those data to develop a set of rules for these protein mutations. They hope their results can be used to improve mutation prediction computational algorithms, resulting in a more complete picture of what happens when proteins are mutated, which could help physicians design treatments for patients on an individual level.
"For example, some people experience side effects with certain drugs, while others don't," explained Swint-Kruse. "Now we could know that, if a patient has a certain mutation, not to give them a drug. Or, if a patient has a mutation, we also know what drug would help them."
"The highly interdisciplinary research to be carried out by the Swint-Kruse team will take personalized medicine many steps closer to reality," said Gerald Carlson, Ph.D., chair of the Department of Biochemistry and Molecular Biology.
In addition to Swint-Kruse and Fenton, researchers on this project include Alexey Ladokhin, Ph.D., and Joseph Fontes, Ph.D., both associate professors in the Department of Biochemistry and Molecular Biology; Antonio Artigues, Ph.D., research assistant professor in the Department of Biochemistry and Molecular Biology; Bruno Hagenbuch, Ph.D., professor in the Department of Pharmacology, Toxicology & Therapeutics; Audrey Lamb, Ph.D., professor in the Department of Molecular Biosciences, University of Kansas; Paul Smith, Ph.D., professor in the Department of Chemistry, Kansas State University, and John Karanicolas, Ph.D., associate professor in the Department of Molecular Therapeutics, Fox Chase Cancer Center.