In our recently published paper, Assessing saliva microbiome collection and processing methods¹, we wanted to understand ways to study the saliva microbiome in samples that were collected without that purpose in mind. Here is a little information about the process and what we found.

What did we do?

We had a lot of saliva banked here at our university that was collected using special kits to test for COVID. We wanted to study the microbiome of these samples but we were unsure if the samples were collected in a way that was appropriate for measuring the microbiome.

In short, we collected saliva from 22 people in two ways: (1) in a special saliva collection kit that contained a preservative and (2) in an empty tube. Six of these participants also came back the following day to give us a second set of samples so we could look at these methods over time. We then performed two types of sequencing on these samples to test how our collection methods compared.

Why are we interested in the saliva microbiome?

We believe that these saliva samples are important to study because the oral microbiome is a proxy for what’s going on in the lung². We aspirate a small amount of our saliva every day so it makes sense that some of the microbes that live in our mouths would end up in our lungs. By studying the mouth microbiome in these banked samples of people with or without COVID, we can get an idea of how our microbial environment in our mouth and lung might impact COVID disease. But first we had to see if we could even measure the microbiome in these saliva samples.

How did we do the study?

We started by looking through what had been done in the past. Several labs had addressed similar questions, but since our collection kits were so new, proprietary, and not designed for this purpose, we couldn’t be sure that they would work for accurately measuring the microbiome. As such, we recruited people from around our university who had taken part in the original study to give us some more samples for us to test. We collected samples using these kits and also in empty tubes. We had an eager population to draw from, and once we had approval to collect samples, it only took a few days to get what we needed.

With these samples, we performed a standard type of sequencing in the field (termed 16S rRNA gene sequencing) that examines a specific bacterial gene which can help us identify the bacteria that are in the sample. We used this type of sequencing to compare samples collected with the kits and those collected in empty tubes. We also performed a more robust type of sequencing, called shotgun metagenomics, on just the samples collected with preservative to test the effect of depleting human DNA in samples prior to sequencing.

What’s up with the human DNA stuff?

Most of the DNA that is present in saliva is from the human cells from whom the sample originated³. But since we are measuring the microbiome, we are only interested in the DNA that comes from the bacterial cells in the mouth. The second type of sequencing we performed, shotgun metagenomics, looks at all of the DNA in the sample not just the bacteria. We wanted to get rid of as much human DNA in the samples as possible so we could to focus our sequencing efforts on just the bacteria.

This isn’t a trivial task, however, so we turned to brilliant work already developed. There was an excellent paper that developed an inexpensive and effective method³. They showed that their method worked really well on fresh saliva and some frozen saliva but we weren’t certain whether this method would work for our samples with our specific collection method. So we sequenced some of our samples in the preservative, testing them with and without this human DNA depletion method prior to sequencing.

What did we find?

The preservative does actually allow for us to accurately measure the microbiome. Surprisingly, we also found that it gave us a more stable picture of the microbiome over time compared to samples collected without preservative. We can confidently say that these COVID collection kits can be used to study the saliva microbiome.

The human DNA depletion worked. While it didn’t work as efficiently as in fresh samples in the original paper, it was still effective in samples collected with our preservative kits that had been frozen. Most importantly, we didn’t see any large change in the microbiome composition between samples that had human DNA depleted before sequencing.

In all, we can feel confident that these methods are effective at adequately measuring the microbiome and that banked saliva samples collected with these special kits can be used for microbiome measurement.

What are some other takeaways from this paper?

We saw that people’s oral microbiomes were more similar to themselves over time that they were to others’ microbiomes. So while there is a lot of variability in the saliva microbiome, you still have a personal signature to the pattern of your mouth microbes

I am excited to apply these methods to the larger biobank samples we have and to study how the saliva microbiome might be related to COVID infection.

Many thanks to the authors of this paper and to all of the people at Rutgers who helped recruit the study participants. And a special thanks to our study participants, who were willing to help us extend the science. I hope you read the full paper here: https://doi.org/10.1038/s41522-021-00254-z

References:
1. Armstrong, A. J. S., Parmar, V. & Blaser, M. J. Assessing saliva microbiome collection and processing methods. npj Biofilms and Microbiomes 7, doi:10.1038/s41522-021-00254-z (2021).
2. Bassis, C. M. et al. Analysis of the upper respiratory tract microbiotas as the source of the lung and gastric microbiotas in healthy individuals. mBio 6, e00037, doi:10.1128/mBio.00037-15 (2015).
3. Marotz, C. A. et al. Improving saliva shotgun metagenomics by chemical host DNA depletion. Microbiome 6, 42, doi:10.1186/s40168-018-0426-3 (2018).