Making It Last: Storage Time and Temperature Have Differential Impacts on Metabolite Profiles of Airway Samples from Cystic Fibrosis Patients

Metabolites of human or microbial origin have the potential to be important biomarkers of the disease state in cystic fibrosis (CF). Clinical sample collection and storage conditions may impact metabolite abundances with clinical relevance. We measured the change in metabolite composition based on untargeted gas chromatography-mass spectrometry (GC-MS) when CF sputum samples were stored at 4 degrees C, -20 degrees C, or -80 degrees C with one or two freeze-thaw cycles. Daily measurements were taken for 1 week and then weekly for 4 weeks (4 degrees C) and 8 weeks (-20 degrees C). The metabolites in samples stored at -20 degrees C maintained abundances similar to those found at -80 degrees C over the course of 8 weeks (average change in Bray-Curtis distance, 0.06 +/- 0.04) and were also stable after one or two freeze-thaw cycles. However, the metabolite profiles of samples stored at 4 degrees C shifted after 1 day and continued to change over the course of 4 weeks (average change in Bray-Curtis distance, 0.31 +/- 0.12). The abundances of several amino acids and other metabolites increased with time of storage at 4 degrees C but remained constant at -20 degrees C. Storage temperature was a significant factor driving the metabolite composition (permutational multivariate analysis of variance: r(2) = 0.32 to 0.49, P -0.001). CF sputum samples stored at -20 degrees C at the time of sampling maintain a relatively stable untargeted GC-MS profile. Samples should be frozen on the day of collection, as more than 1 day at 4 degrees C impacts the global composition of the metabolites in the sample. IMPORTANCE Metabolomics has great potential for uncovering biomarkers of the disease state in CF and many other contexts. However, sample storage timing and temperature may alter the abundance of clinically relevant metabolites. To assess whether existing samples are stable and to direct future study design, we conducted untargeted GC-MS metabolomic analysis of CF sputum samples after one or two freeze-thaw cycles and storage at 4 degrees C and -20 degrees C for 4 to 8 weeks. Overall, storage at -20 degrees C and freeze-thaw cycles had little impact on metabolite profiles; however, storage at 4 degrees C shifted metabolite abundances significantly. GC-MS profiling will aid in our understanding of the CF lung, but care should be taken in studies using sputum samples to ensure that samples are properly stored.