Computers Help Decipher Blood Pressure Control Circuits in Mammals

Release Date:
Alison Davis, NIGMS

At first glance, NIGMS grantee Dr. Oliver Smithies' laboratory at the University of North Carolina, Chapel Hill resembles a typical molecular biology workspace, with mice getting their blood drawn to measure levels of blood pressure-controlling hormones, petri dishes growing recombinant bacteria, and the usual assortment of tubes and beakers. But there is something distinctly different about the way Dr. Smithies, a world-renowned geneticist whose pioneering "knockout mouse" technology (which selectively removes genes from mouse chromosomes) has propelled studies of mammalian genetics, is conducting his experiments. He has begun work with a different kind of model "organism" altogether: the computer. Dr. Smithies, who studies high blood pressure using mice as model organisms, is turning to computers to discern the effects of both nature and nurture (genes and drugs, in this case) on two specific components of the blood pressure control circuit: the proteins angiotensinogen (AGT) and angiotensin converting enzyme (ACE).

In previous studies, Dr. Smithies has skillfully wielded a genetic knife--either removing or adding multiple copies of genes to mouse DNA--to learn more about the roles these proteins play in controlling how much and how fast blood traverses vessels throughout the body. Such "gene titration" experiments, as he calls them, have helped Dr. Smithies evolve a model to describe essential hypertension, a complex disease with multiple genetic and environmental causes. These prior studies have revealed that while increasing the number of AGT genes in a mouse elevates blood pressure, similar manipulations to the number of ACE genes have no effect whatsoever on a mouse's blood pressure. Plugging such data into his customized computer program allows Dr. Smithies to ask--and answer--further questions about the effects of genetic and pharmacologic manipulations. The approach is helping Dr. Smithies and his colleagues to plan experiments better as well as to continually refine their model of hypertension, a widespread and often deadly disease in humans. Already, the model seems to be producing valid data: The computer has "replicated" many experiments performed by Dr. Smithies or other researchers in the field, yielding nearly identical results to studies done in animals.

At present, Dr. Smithies' modeling work--like all modeling work--hinges upon many assumptions garnered from the results of prior animal studies, and it will not be complete until many more such studies provide additional data to input into the model. But the power of the technique rests in the ability to orchestrate multiple, competing research variables--say, the actions of multiple proteins in different organs and feedback loops--simultaneously. The new work takes a step forward toward the day when computer models might be able to make life-saving predictions about an individual's susceptibility to disease, or even about how people might react to drugs used to treat disease.


Smithies O. A mouse view of hypertension. Hypertension 1997;30:1318-24.

Reporters may call the NIGMS Office of Communications and Public Liaison at (301) 496-7301 to obtain the name of a scientist in the NIGMS Division of Genetics and Developmental Biology who can comment on this work.