The analysis of nonsteroidal antiinflammatory drug selectivity in prostaglandin G/H synthase (PGHS)-null cells

Nonsteroidal antiinflammatory drugs (NSAIDs) that are more selective for PGHS-2 maintain their antiinflammatory properties but exhibit fewer unfavorable gastrointestinal side effects. To evaluate the usefulness of the murine PGHS-null cell system in analyzing PGHS-2 selective inhibitors, we tested PGHS-2 non-selective NSAIDs and two PGHS-2 specific compounds using either endogenous or exogenous sources of substrate, arachidonic acid.

A whole-cell assay system for testing the efficacy of PGHS isozyme-specific inhibitors using murine PGHS-1 or PGHS-2-null fibroblast cell lines derived from lung tissues of PGHS-2(-/-) and PGHS-(-/-) mice, respectively, was employed. PGHS-1 and PGHS-2 null cell lines were exposed to three widely used NSAIDs, ibuprofen, indomethacin and aspirin, and two PGHS-2 specific inhibitors, MK-966 (rofecoxib) and NS-398. Excess arachidonic acid (AA) was provided externally and internally via calcium ionophore A23187. PGHS-1 and PGHS-2 activity were determined by measuring the prostaglandin E2 production by radioimmunoassay. IC50 and IC80 values of each compound were determined from the reduction of PGE2 levels as a measure of inhibition of existing PGHS isozyme in the PGHS-null cells.

In our murine PGHS-null cell systems, both PGHS-1 and PGHS-2 null cells can use both externally provided AA and endogenous, A23187-derived AA. Both NS-398 and MK-966 were potent inhibitors and demonstrated strong selectivity for PGHS-2. Among the non-selective NSAIDs, based on the PGHS-2 IC50/PGHS-1 IC50 ratio ranking, ibuprofen is more selective for PGHS-2 than indomethacin while aspirin is the least selective inhibitor regardless of the arachidonic acid source. Indomethacin and MK-966 IC50 values for PGHS-2 were in the range of 10(-9)-10(-8) M while the IC50 values for aspirin were in the range of 10(-5) M. There were differences in the ranking of indomethacin and ibuprofen when the IC80 ratios were used.

Here, we report on a whole cell assay system for testing the efficacy of PGHS isozyme-specific inhibitors using murine PGHS-1 or PGHS-2-null cell lines. This system, using cells that express either PGHS-1 or PGHS-2, offers a convenient and reliable method to determine IC50 and IC80 values of the two PGHS isoforms, entirely independent of each other, in the same cell type. The results of our evaluation using a panel of NSAIDs, both PGHS-2 selective and non-selective inhibitors, correlate well with previously published clinical and laboratory data, demonstrating the usefulness of the whole-cell assay system described here.