DNA Viral Size Fraction Metagenomics for Human Stool Samples

Our group studies viral communities, mostly through metagenomic approaches, and part of our work aims to help understand the healthy human virome. This protocol can provide a framework for any researchers that are interested in using viral metagenomics to study viruses that are in the human gut. To begin, obtain a human stool sample in a 50-milliliter conical tube, and add eight milliliters of PPBS buffer directly to the tube.

Vortex the tube to thoroughly resuspend the sample. Shake the tubes at 300 revolutions per minute for 10 minutes at four degrees Celsius. Then, centrifuge the samples at 4, 000g for 10 minutes at four degrees Celsius.

In a fume hood, carefully decant the supernatant into a new sterile 50-milliliter conical tube. Store the tube at four degrees Celsius. Retain the original conical tube containing the pellet and add eight milliliters of PPBS.

Repeat the resuspension, shaking, and centrifugation steps two more times. After pooling the supernatant into the same tube, centrifuge the 50-milliliter conical tube containing the accumulated supernatant at 10, 000g for eight minutes at four degrees Celsius. Decant the supernatant into a new 50-milliliter conical tube and centrifuge the tube at 10, 000g at four degrees Celsius for eight minutes.

For filtration, stack a five-micrometer filter on top of a 0.2-micrometer filter and attach them to a single syringe one by one. Now, sequentially filter the supernatant through these filters and collect the filtrate directly into an ultracentrifuge tube. Check the mass of all ultracentrifuge tubes containing the samples and pair tubes based on equal mass for ultracentrifugation.

Ultracentrifuge the samples at 112, 000g under vacuum for two hours and 25 minutes at four degrees Celsius. Remove the supernatant, either by decanting or using a pipette, while avoiding disturbance of the pellet, which contains the virions. Then, resuspend the pellet gently in 100 microliters of nuclease-free, molecular biology-grade water.

If required, treat the concentrated virions with DNAs in the presence of a suitable buffer at 37 degrees Celsius for 30 minutes. Finally, perform DNA extraction using a commercial kit and quantify the extracted DNA using a fluorescence-based method. An exploratory analysis applying this protocol to one frozen stool sample and two soil samples demonstrated its ability to generate high-quality viromes across sample types.

Viral sequences were identified from quality filtered reads and de novo assembled contigs using an open-source virus identification tool. Reads mapped to viral contigs were used to estimate relative viral abundance in each sample. Viral reads comprised, on average, 74%of all sequenced reads in stool viromes, compared to only 4%in total metagenomes generated from the same sample.

High-confidence viral genomes were generated from stool samples with a mean virus confidence score of 0.97. Viral contigs assembled from stool virome replicates had significantly higher mean virus confidence scores than those from total metagenomes. Modified versions of the protocol using different buffer types, filter sizes, and virion concentration methods resulted in successful DNA extraction from the viral size fraction.

This protocol enriches for viruses in the sample before sequencing. It can produce high-quality viral metagenomes with limited non-viral contamination. Since certain steps, like filtration and DNAse treatment, can bias virus enrichment, it's important to consider any methodological choices when interpreting your downstream sequencing results.

Following the procedure, researchers can do shotgun metagenomics and bioinformatics to analyze the virome in their samples.