CPSA Digest 2001

Proceedings -Tuesday, October 9, 2001

TuOB2

Structure Profiling:
Integration of Software-Based Strategies in the Drug Discovery Metabolite Identification Process

Background:
One of the components of drug discovery research is the assessment of the metabolic fate of lead compounds. A substantial number of new chemical entities are terminated late in the development stage due to problems with drug metabolism or pharmacokinetics. This traditionally occurs after a substantial amount of time and money has already been invested. In an effort to prevent these late-stage failures, it is useful to know the metabolic fate of a promising lead compound early in the discovery phase. If there are toxic metabolites identified, then structural analogs of early drug lead candidates may be designed to block portions of the molecules that are particularly susceptible to metabolism. Note that due to the large number of chemical entities in development, this metabolic characterization process must be amenable to high throughput in order to be a broadly useful technique.

Premise:
Mass spectrometry coupled with liquid chromatography is an effective method for metabolite profiling. The integration of data collected from the ion trap, triple quadrupole and quadrupole/-time-of-flight instruments allow for a comprehensive evaluation of biotransformation products. This approach is routinely used to evaluate metabolites generated from in vitro and in vivo systems.

Traditional metabolite identification experiments involve iterative evaluation of one potential metabolite after another in a serial manner, requiring large amounts of user intervention and time. The approach at Schering is a parallel method whereby multiple experiments are completed within one LC cycle. A systematic approach to data acquisition can be employed in which MS/MS experiments are run on all 'expected' metabolites. These include common oxidative metabolic alterations as well as any alterations common to a structural series. While this approach improves sample throughput, manual evaluation of the resulting MS/MS spectra is still required.

Metabolite identification software is utilized to thoroughly process MS data, numerically evaluating the MS spectra, looking for expected metabolites or characteristic isotopic patterns, intelligently providing lists of potential metabolites and setting up further experiments to confirm the identity of the metabolites. While operator intervention is important and data interpretation continues to be a bottleneck, current software packages provide automated ways to 'mine' the large amounts of data generated. The metabolite characterization software programs are continuing to evolve, developing 'smarter' ways of interpreting MS and MS/MS data. Data-dependent software is available to maximize the amount of information generated from a single analysis.

Value of the Technology
The utilization of sophisticated software programs early in the discovery process has dramatically increased the throughput of metabolite identification studies. Early identification of active or toxic metabolites results in significant time, money and resource savings to pharmaceutical companies. Characterization of metabolic liabilities for particular structural series provides important information to early drug design.

Links
Advanced Chemistry Development

Press Releases: www.acdlabs.co.uk/publish/press_release.html

Nigel J. Clarke, Diane Rindgen, Walter A. Korfmacher and Kathleen A.
Cox, "Systematic LC/MS Metabolite Identification in Drug Discovery"
Anal. Chem. 73(15), 430A-439A (August 1, 2001).

Return to Proceedings »