Steroids Trigger a 'Domino Effect' on Genes at Different Points in Time, UB Studies Find

BUFFALO, N.Y. -- Corticosteroids, drugs that simultaneously deliver powerful therapeutic effects and potentially severe adverse effects, cause a remarkably complex "domino effect" of genomic changes, according to a landmark paper by University at Buffalo pharmaceutical scientists.

The paper was published June 13 online in the Journal of Pharmacology and Experimental Therapeutics (JPET).

By helping researchers unravel the complex ways in which steroids act in the body, the research will begin to provide insights into ways to develop more individualized dosing for patients.

The UB research is the first to reveal the huge spectrum of genomic changes triggered in vivo by corticosteroids, or for that matter, any class of drugs.

"By giving us a more complete understanding of the changes that drugs cause in genes, this work provides us with the tools we need to better control those changes," said William J. Jusko, Ph.D., professor and chair of the Department of Pharmaceutical Sciences in the UB School of Pharmacy and Pharmaceutical Sciences and co-author on the paper.

"In the future, researchers will be able to use this kind of information to tailor dosing regimens for patients so as to maximize positive effects and reduce adverse ones," he said.

What sets the work of the UB researchers apart from others is that they study pharmacodynamics, the time course of drug effects, on multiple genes and gene products.

"In our research, we try to look at the time frame of both positive and negative effects," said Richard R. Almon, Ph.D., professor of biological sciences in the UB College of Arts and Sciences, co-author and Jusko's partner in the work for more than a decade. "If we can find ways to increase the former and reduce the latter, then we can design better dosing regimens."

Because of the severity of the side effects caused by steroids, more individualized dosing regimens are sorely needed, particularly for transplant patients, who usually have to take steroids for their lifetime to suppress their immune systems so that the transplanted organ is not rejected. Corticosteroids also are prescribed for patients suffering from rheumatoid arthritis, asthma, lupus and inflammatory bowel disease.

"When corticosteroids are prescribed for patients, many of them develop type 2 diabetes, muscle atrophy, atherosclerosis and osteoporosis," said Almon.

"We don't yet know how to turn up the drugs' positive effects and turn down the others," he said, adding that right now, corticosteroids are prescribed -- as are most drugs -- for an individual who is assumed to be average in all respects.

"No one is average in all variables," he said.

The researchers focused on genomic changes in the livers of lab rats that had been administered methylprednisolone.

"We looked first at the liver because that's the big factory of metabolic functions," said Almon. "It probably expresses more genes than any other tissue."

Until now, researchers studying genomic effects of pharmaceuticals have only looked at a single point in time. In fact, the UB paper is a follow-up to a pilot study the UB team published last year in the Journal of Pharmacokinetics and Pharmacodynamics in which the researchers did exactly that.

"In that paper, we studied effects at only one time value, focusing on what we thought was an optimum time following dosing of the drug," said Jusko.

But the UB team's newest results on genomic changes in the liver caused by steroids demonstrate that these changes are highly time-dependent.

"In order to understand what's happening after administration of any drug, it turns out that you have to look at the full genomic time profile," said Jusko. "Looking at only one point in time, we've found, can be severely misleading."

In examining 8,000 genes at 17 different time points lasting 72 hours, the UB researchers discovered that 200 genes undergo significant changes as a result of what they call the "domino effect" triggered by steroids.

Of those changes, the researchers identified six distinct patterns, or profiles, with clusters of genes showing similar increases, decreases and mixed changes in the production of messenger RNA.

"These detailed studies provide more extensive information on what's happening," said Jusko. "To do a proper assessment of the importance of gene changes, you have to study messenger RNA that each gene controls and the proteins or enzymes that those proteins, in turn, control. The genes we looked at took 72 hours to undergo the full range of genomic changes in response to steroids."

He added that the amount of time needed to see a complete genomic profile probably will differ with each drug.

"We think this is the first publication showing such complicated time-dependent genomic changes," said Jusko, "yet the genes have consistency in obeying some basic biological rules."

To do the experiments, Almon uses gene microarray technology to acquire massive amounts of data on changes in gene expression that result from dosing rats. He then looks for ways to "connect the dots," as he puts it, in all the biochemical pathways in which these genomic changes are involved. Jusko then puts these data into rational mathematical models to predict the cascade of biological events produced by the drug.

"We are looking for rational, temporal patterns of gene expression on a grand scale, to see how they all fit together," said Almon.

That kind of information may turn out to be useful for drug companies, he explained, who could use it to identify potential drug targets.

"If we can identify all the genetic variations and understand how things cascade together, it will allow us to move toward pharmacogenomics, where dosing and eventually, medicines, are developed based on genetic variations," said Almon.

Jusko has been continuously funded for more than 25 years by the National Institutes of Health to study pharmacodynamics, the time-course of drug effects on the body.

Jusko and his colleagues at UB recently received additional NIH funding for what will be a complicated extension of this research -- the bioinformatics piece -- in which they will expand the number of tissues examined and begin to connect the diverse genomic responses to what is already known about various biochemical pathways in the body.

Other co-authors on the JPET paper include Debra C. DuBois, Ph.D., UB research associate professor of biological sciences, Jin Y. Jin, a doctoral candidate in the UB Department of Pharmaceutical Sciences, and Keri E. Peterson and Dietrich A. Stephan of the Research Center for Genetic Medicine at Children's National Medical Research Center, Washington, D.C.