Abstract

Metabolite analysis is recognized as an important facet of systems biology, however complete metabolome characterization has not been realized due to challenges in sample preparation, inherent instrumental limitations and the labor intensive task of data interpretation. This work aims to compare several commonly used metabolite extraction strategies for their effect on the (1)H nuclear magnetic resonance (NMR) metabolic profile of extracts of the model plant Arabidopsis thaliana. Extractions were carried out on aliquots from a pool of homogenized plant tissue using CD(3)CN/D(2)O, buffered D(2)O, perchloric acid in D(2)O, CD(3)OD/D(2)O and CD(3)OD/D(2)O/CDCl(3) as the extraction solvents. The effects of lyophilization as a sample pretreatment, solvent evaporation and extract fractionation for removal of interfering species were studied. Representative spectra are presented for qualitative interpretation. Analytical reproducibility was evaluated by principal components analysis. Perchloric acid facilitated acid-catalyzed cleavage of sucrose, further complicating biological interpretation of the resulting metabolite profile. The solvent system CD(3)OD/D(2)O/CDCl(3) gave the least reproducible results in our hands. D(2)O extracts suffered from poor stability probably due to contamination by soluble enzymes, which were not denatured in this solvent. CD(3)CN/D(2)O extracts showed greater stability than D(2)O alone, but problems were encountered due to degradation of (1)H NMR spectral resolution during lengthy acquisitions due to partial phase separation. In addition, this solvent system produced spectra with significant contamination by lipids that obscured spectral regions containing the resonances of the aliphatic amino acids. These problems were solved by speedvacuuming the CD(3)CN/D(2)O extract and reconstituting in D(2)O solution.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call