Modeling of Corticosteroid Pharmacogenomics in Rat Liver Using Gene Microarrays

Abstract

Corticosteroid (CS) pharmacogenomics was studied using gene microarrays in rat liver. Methylprednisolone (MPL) was administered intravenously at 50 mg/kg. Rats were sacrificed and liver excised at 17 time points over 72 h. RNAs from individual livers were used to query Affymetrix GeneChips that contain sequences for 8000 genes. Cluster analysis revealed six temporal patterns consisting of 197 CS-responsive probes representing 143 genes. Based on our fifth-generation model of CS pharmacokinetics/pharmacodynamics (PK/PD), mechanistic models were developed to describe the time pattern for each CS-responsive gene. Two clusters showed increased expression with different effect duration. PK/PD models assuming CS stimulation of mRNA synthesis were applied. Another two clusters showed an initial decline followed by delayed increase, suggesting two mechanisms might be involved jointly. The initial suppression was captured by CS inhibition of mRNA synthesis or stimulation of degradation. CS may also stimulate the production of a biosignal (transcription factors or other hormones), which can cause secondary induction of the target mRNA. One cluster showed a very abrupt increase in message followed by rapid decrease. These genes were lymphocytic in origin and were modeled combining the fast gene induction effect of CS in lymphoid cells and its direct lymphocyte trafficking effect. Another cluster showed reduction persisting for 18 h, which was described by CS inhibition of mRNA synthesis. Our results reveal the marked diversity of genes regulated by CS via a limited array of mechanisms. These PK/PD models provide quantitation of CS pharmacogenomics and new hypotheses regarding understanding of diverse mechanisms of CS receptor-gene mediated action.

Assessing the Dynamics of Nuclear Glucocorticoid-Receptor Complex: Adding Flexibility to Gene Expression Modeling

“…Thus in previous reports parameters associated with these two model compartments were based on the data available for the drug, its receptor, and the GR mRNA. The transduction compartment, DR(N), was used to model many CS driven effects, such as changes in hepatic TAT mRNA and activity (12,14,20,21), hepatic PEPCK mRNA and activity (15), and many other genomic effects of CS (18). Since these effects are generally of slow onset (i.e.…”

“…Animals (3-4 per time point) were sacrificed over a period of 72 hr for the single dose groups (0.25, 0.5, 0.75, 1, 2, 4, 5, 5. 5,6,7,8,12,18,24,30,36, 48 and 72 hr) and over a period of 120 hr (0. 25, 1, 3, 5, 6, 7, 8, 12, 18, 24, 24.25, 25, 27, 29, 30, 31, 32, 36, 42, 54, 72, 96 and 120 hr) for the dual dose group.…”

“…The details of our microarray studies are published (17,18). Briefly, total RNA from rat liver was extracted using Trizol reagent according to manufacturer's protocols.…”

“…The mRNA time profiles for hepatic TAT, CEBP-β, ornithine decarboxylase (ODC), argininosuccinate synthetase (ASS), aldehyde dehydrogenase (ADH), low density lipoprotein receptor (LDL-R), carbonic anhydrase (CA) and glutaminase mRNA from our microarray study (17,18) from the high-dose group (50 mg/kg single dose) of MPL treated animals are examples of rapid and slow responding genes. Equation (12) was used to describe the increase in transcription of TAT, CEBP-β, ODC and ASS mRNA.…”

“…In these earlier studies, we were technically limited to studying relatively few target genes at a time, such as the up-regulation of hepatic TAT mRNA and activity, hepatic phosphoenol-pyruvate carboxykinase (PEPCK) mRNA and activity (15), glutamine synthetase (GS) mRNA and activity in the muscle (16), and down-regulation of GR mRNA. However, in microarray experiments (17,18), where the effects of CS treatment were simultaneously examined for several thousands of genes, we encountered classes of rapidly up-or down-regulated genes which could not be described by our previous model. The nuclear drug-receptor complex (DR(N) in Fig.…”