Thread by @m3thyl

It's published! The main study from my Ph.D. is finally out in @ISME_microbes! The Horizontal acquisition of a patchwork Calvin cycle by symbiotic and free-living Campylobacterota. Deep-sea symbiosis and genes hoping between symbiotic bacteria! (A thread) nature.com/articles/s4139…

First of all, nothing would have been possible without @NLeisch, shared first author, and @peterse47663885, @Chemosym, @SeepExplorer, @Manuel_Kleiner @GruberVodicka, Matthew Saxton, Tjorven Hinzke, Dimitri Meier, Halina E. Tegetmeyer and Anke Meyerdierks

So what is the paper about:
Deep-sea symbiosis, giant mussels, bacteria, and weird metabolic pathways. Let's go!
@MarineMicrobio

Bathymodiolus deep-sea mussels live close to hydrothermal vents and cold seeps worldwide. They harbor endosymbiotic chemotrophic bacteria in their gills: the bacteria use methane and sulfur, then the mussel eats the bacteria. Deep-sea Success! (Picture ©@Ifremer_en)

The first paper I published during my Ph.D. was describing a previously overlooked epibiont of Bathymodiolus deep-sea mussels. Long filamentous bacteria colonizing the surface of Bathy gills.

A shameless reshare:
onlinelibrary.wiley.com/doi/abs/10.111…

First thought to be present in low abundance, with some EM expertise (read: dark magic) from @NLeisch we soon realized they might be more abundant. On the pictures below you can see them nicely cushioned between the gills. Look at those pictures!

So we definitively wanted to know more. Early 16S rRNA phylogeny was placing this new family of Campylobacterota between Helico-/Campylo-bacter (mostly pathogenic) and the Thiovulgacea (mostly symbiotic). So what are those guys really capable of?

If you didn't follow the fascinating world of taxonomy, Epsilonproteobacteria changed named, they are now Campylobacterota.
ncbi.nlm.nih.gov/pmc/articles/P…

From metagenomic datasets, we assembled draft genomes and started digging for gold.

Hmmm...

Nothing pathogenic, oxidize sulfur, look like any other Campylobacterota.

Let's start by giving those epibionts a Candidatus name: "Ca. Thiobarba" A sulfur beard!

...wait is this a Rubisco sequence I see?

Yes, it is! Turns out that the Ca. Thiobarba genomes we reconstructed have all the genes for a fully functional Calvin cycle and are apparently missing key genes to have a reverse TCA cycle, which is characteristic of chemotrophic Campylobacterota!

In the realm of Inorganic carbon fixation pathways, you have the Calvin cycle king dominating the photosynthetic world, but down in the deep-sea (and other places), you have alternative pathways. In our case, the reverse TCA cycle is often dominant at seeps and vents.

Interestingly those pathways are distributed in different bacterial phylogenetic groups. In our system, the endosymbiont of Bathy mussels are gammaproteobacteria with the Calvin cycle and Campys have not been shown so far been associated with it! So where does it come from?

Looking into this deeper it turns out that the genes coding for 7/10 of the gene necessary for the CBB cycle come from different proteobacterial classes. There are two distinct operons in the genomes, coming from two different origins!

First one with the Rubisco related genes which aa sequences are closely related to the Sulfur-oxidizing endosymbionts of Bathymodiolus and the other cluster of genes are related to an unknown group of likely Betaproteobacteria.

One thing you could say is that from metagenomic assembly we simply assembled a hybrid monster and the Calvin Cycle genes might come from other bacteria present in the samples.

Metagenomic is a B****

But the beauty here we have a draft genome for two species of Ca. Thiobarba. Both have the same operons, but one is present within a host species that host the closely related sulfur-oxidizing endosymbiont but the other does not.

The cherry on top: Dimitri Meier (in @DOME_Vienna) found in a water column sample a third species of Campylobacter with the same operon structure for the Calvin Cycle. So this suggest that it is not a misassembly and we are looking at a real operon transfer.

Thanks to transcriptomics and proteomics we could confirm the expression of the cycle in one of the mussel species.

Many thanks to @Manuel_Kleiner who came up with an ingenious method to look at isotope signatures using proteomic data and show that the Thiobarba proteins have a Calvin cycle signature and not a rTCA one.
pnas.org/content/115/24…

So why all of this matter?
First, it shows the incredible plasticity of bacteria and their ability to patch up a metabolic pathway to colonize new grounds!
Second, if bacteria can assemble a cycle from different HGT in the wild, maybe we can do the same in the lab?

Third, this might be the tip of the iceberg as we also found a rubisco sequence in a free-living Campylobacterota! This might not only be a symbiotic unique feature but an overlooked widespread feature!

Fourth, isotopic profiles are often used in the deep sea world to infer which group of bacteria are present in sediment and rock, now Campylobacterota need to be considered when looking at CBB profiles!

One last time, thank you to everyone involved in this project. It's been a rollecoaster and I am very happy to see this finally out!
And Thank you, who made it up till here, I am happy to answer any questions!

Special thanks to @kbseah for helping us to get a proper Candidatus name!