Strategies for Bioanalysis of an Oligonucleotide Class Macromolecule from Rat Plasma Using Liquid Chromatography−Tandem Mass Spectrometry

Abstract
Electrospray ionization (ESI) liquid chromatography−tandem mass spectrometry (LC/MS/MS) assays provide high-throughput and selective methods for quantitation of small molecules. Use of LC/MS/MS assays for macromolecules, like oligonucleotides, is challenging due to lack of sensitivity and low analyte recovery from biomatrixes. Due to this fact, the method of choice for oligonucleotides quantitation remains hybridization-based ligand-binding assays. These biological assays usually possess high sensitivity but low selectivity and narrow dynamic range. They also require optimizing suitable “capture and detection” probes, which can be prohibitively time-consuming and expensive in a drug discovery lead−optimization scenario. In this paper, we present a unique LC/MS/MS assay for a model phosphorothioate backbone oligodeoxynucleotide (ODN) drug (7692 amu) from rat plasma. Multiple analytical challenges were encountered. The strategies used to solve these challenges should prove useful to scientists pursuing mass spectrometry (MS) to quantitate oligonucleotides. The challenges include analyte multiple charging and cation adduction (reduced sensitivity), oxidation of analyte on drying and high protein binding (low recovery), ODN affinity to exposed silica (low chromatographic reproducibility and high carryover), nonspecific binding of analyte to containers (low storage stability), and optimization/synthesis of an appropriate internal standard (interference and cross-talk). A buffer (7 mM triethylamine and 3 mM ammonium formate)/methanol, 50:50 (v/v), was used as an ESI-MS infusion solvent and produced a sharp multiple charge-state distribution. The sample extraction method combined a phenol/chloroform liquid−liquid extraction and solid-phase extraction steps, which improved the absolute recovery to >70%. The method was validated in the range of 5−2000 ng/mL and had precision (percent relative standard deviation) <10.1% and accuracy (percent relative error) <11.4%.

This publication has 23 references indexed in Scilit: