Abstract
BACKGROUND: There is an increasing interest in investigating the human gut virome for its influence on the gut bacterial community and its putative influence on the trajectory towards health or disease. Most gut virome studies are based on sequencing of stored fecal samples. However, relatively little is known about how conventional storage buffers and storage conditions affect the infectivity of bacteriophages and influence the downstream metavirome sequencing. RESULTS: We demonstrate that the infectivity and genome recovery rate of different spiked bacteriophages (T4, c2 and Phi X174) are variable and highly dependent on storage buffers. Regardless of the storage temperature and timespan, all tested phages immediately lost 100% (DNA/RNA Shield) or more than 90% (StayRNA and RNAlater) of their infectivity. Generally, in SM buffer at 4 °C phage infectivity was preserved for up to 30 days and phage DNA integrity was maintained for up to 100 days. While in CANVAX, the most effective buffer, all spiked phage genomes were preserved for at least 100 days. Prolonged storage time (500 days) at - 80 °C impacted viral diversity differently in the different buffers. Samples stored in CANVAX or DNA/RNA Shield buffer had the least shifts in metavirome composition, after prolonged storage, but they yielded more contigs classified as "uncharacterised". Moreover, in contrast to the SM buffer, these storage buffers yielded a higher fraction of bacterial DNA in metavirome-sequencing libraries. We demonstrated that the latter was due to inactivation of the DNases employed to remove extra-cellular DNA during virome extraction. The latter could be partly avoided by employing additional washing steps prior to virome extraction. CONCLUSION: Fecal sample storage buffers and storage conditions (time and temperature) strongly influence bacteriophage infectivity and viral composition as determined by plaque assay and metavirome sequencing. The choice of buffer had a larger effect than storage temperature and storage time on the quality of the viral sequences and analyses. Based on these results, we recommend storage of fecal virome samples at in SM buffer at 4 °C for the isolation of viruses and at - 80 °C for metagenomic applications if practically feasible (i.e., access to cold storage). For fecal samples stored in other buffers, samples should be cleared of these buffers before viral extraction and sequencing. Video Abstract.