Thread by @Ge_BoRo

Our paper on microbiota in #lungtransplants was out last week in @NatureComms!
This immense amount of work was a collaboration between @CHUVLausanne and @DMF_UNIL led by the amazing @sudu87 and Eric Bernasconi, about characterization and ecology of #lungmicrobiota 1/14

The preprint was out some time ago, but now you can read the new and improved version here! Sudip and Eric did an amazing work incorporating reviews and restructuring it. 2/14
nature.com/articles/s4146…

You can read @sudu87 thread here, but I did not want to miss the chance to tweet my own (focusing on the microbial ecology), because I think is the product of a very nice multidisciplinary collaboration. 3/14

When @sudu87 first came to talk about it, it wasn't clear if the lower respiratory tract had a core resident microbiota, or if the microbial loads were simply organisms in transit and residuals from the upper respiratory tract. 4/14

So he followed a very unique cohort of 64 #lungtransplant patients from the Organ Transplant Centre at @CHUVLausanne with bacterial cultures, 16SrRNA ampliseq, host gene expression, viral loads and transplant acceptance. 5/14

First surprising result (to me) was the high variability in bacterial biomass across samples, up to six orders of magnitude in some cases. This suggested some samples were very poor and hence pointed towards only transient microbes (and random assemblages). 6/14

But with 16SrRNA amplicon sequence we found >7000 bacterial OTUs. Of these, 22 OTUs were prevalent across samples, and 19 were very dominant in some samples, also found in healthy lungs, suggesting also some kind of shared "core". 7/14

@sudu87 also created the Lung Microbiota Culture Collection #LuMiCol with 300 isolates, more than half matching to 47 OTUs and 17 matching the most dominant and abundant OTUs. So, a good representation of what was found in the ampliseq. 8/14

The community analysis revealed four distinct compositional and ecological states or "pneumotypes":
1. one microbiota depleted,
2. one dominated by Pseudomonas aeruginosa
3. one dominated by Staphylococcus aureus
4. one cohesive and diverse or "balanced"
9/14

These pneumotypes also differ in host gene expression patterns and Anellovirus loads...
10/14

but most importantly, they are also linked with transplant health!

The staph- and depleted- pneumotypes were associated with allograft injury, staph- and pseudomonas- were associated with reduced lung function, and the balanced- and depleted- ones preserved lung function!

11/14

Finally, @sudu87 made clever use of the time series to analyse transitions between pneumotypes, and showed that the balanced communities are the most stable. 12/14

In summary, we found four (community) compositional states in the post-transplant lung microbiota associated with distinct (ecosystem) functioning, in what for me is an awesome example of how microbial ecology can inform clinical microbiology. 13/14

This is a nice conclusion to a series of intense, pleasant and enriching discussions with @sudu87, who has taught me more immunology and human physiology than I ever planned to learn! 14/14