Saliva microbiome collection and processing methods

The oral microbiome has been connected with lung health and may be of significance in the progression of SARS-CoV2 infection. Saliva-based SARS-CoV2 tests provide the opportunity to leverage stored samples for measuring the oral microbiome. However, these collection kits have not been tested for accuracy of measuring the oral microbiome. Saliva is highly enriched with human DNA and reducing it prior to shotgun sequencing may increase the depth of bacterial reads. We examined both the effect of saliva collection method and sequence processing on measurement of microbiome depth and diversity by 16S and shotgun metagenomics. We collected 56 samples from 22 subjects. Each subject provided two saliva samples with and without preservative; 6 subjects provided a second set of samples the following day. 16S rRNA gene (V4) sequencing was performed on all of the samples, and shotgun metagenomics was performed on 8 of the samples collected with preservative with and without human DNA depletion before sequencing. We observed beta diversity distance within subjects over time was smaller than between unrelated subjects, and distances within subjects were smaller in samples collected with preservative. Samples collected with preservative had higher alpha diversity measuring both richness and evenness. Human DNA depletion before extraction and shotgun sequencing yielded higher total and relative reads mapping to bacterial sequencing. We conclude that collecting saliva with preservative may give more consistent measures of the oral microbiome and depleting human DNA increases yield.