Predicting problems in one of the body's most complex organs soon may become easier because of work being done by Wayne State University researchers.
Howard Matthew and Yinlun Huang, professors of chemical engineering and materials science, recently received a $550,000 grant from the National Science Foundation (NSF) to develop a mathematical model of liver metabolism that can be used to analyze and more effectively predict responses to possible treatments for hepatic steatosis, more commonly known as fatty liver.
The condition affects between 15 and 20 percent of the U.S. population and often is a precursor to more serious problems. Accumulation of fat droplets, or lipids, inside liver cells is a key characteristic in many of the organ's failure modes. Increased lipid accumulation is usually the first symptom to appear before a measurable dysfunction occurs.
Identifying the causes of fat accumulation, however, is difficult because of the complex nature of the liver, which is involved in making and redistributing metabolites for most tissues in the body. Additionally, a variety of mechanisms trigger liver dysfunctions.
Matthew and Huang's goal is to develop a mathematical model to analyze and optimally compute possible interventions for treating fatty livers. Their main approach assumes that cellular control of fat metabolism acts as an optimal feedback-control system, and that the liver is trying to maintain certain levels of metabolites to satisfy the needs of other tissues.
A mathematical model based on that principle, Matthew believes, could predict liver cell responses to stimuli. Their model would allow more accurate predictions of metabolic responses than the models currently in use. Such methods, which assume that a cell's objective is primarily to grow, tend to work better with simpler organisms like bacteria or yeast, but not so well with animal cells.
"By focusing on cells' rates of response to disturbances, we can actually achieve a better model with narrower levels of error associated with our predictions," Matthew said.
A preliminary model using cultured liver cells already has achieved some reasonably good results, he said. The current NSF award is geared toward refining that model and collecting more accurate dynamic data.
The team will use perfusion systems that allow researchers to maintain precise levels of metabolites outside cells and to change them instantaneously, looking at response rates of cells to disturbances. Fluorescent lipids will be used so researchers can detect which ones the cells are taking up and how fast they are being redistributed. The team then can model the nature of metabolites being distributed.
"We're looking for changes that occur in minutes to hours in order to refine the model," Matthew said. "The liver typically begins responding within minutes."
Researchers must ascertain the accuracy of prediction at the cellular level before moving to animals and beyond, he said.
"We want our model to predict with a limited amount of data in order to identify which areas might be the key differences between individuals," Matthew said. "That allows biologists and physicians to narrow their focus to particular areas and possibly screen particular enzymes as drug targets.
"Our long-term goal is to use these kinds of models to find out why different people respond to treatments in different ways, so that we can develop a personalized medicine approach to treating the liver or any other tissues."
HMGB1 as a key mediator in liver disease pathogenesis