Why "CAT"? – A Question That Must Be Answered First
But before telling this story, I must address a more fundamental question: among the myriad microorganisms in global hydrothermal systems, why did we focus on just these three groups?
This was not a random choice. At the project's outset, we spent considerable time reviewing the literature and eventually settled on three phyla: Campylobacterota (formerly Epsilonproteobacteria), Aquificota, and Thermosulfidibacterota, collectively termed the CAT phyla. There are two reasons, both indispensable.
First, ecological fidelity. Hydrothermal vents are like "life stages" that switch on and off; most microbes are transient drifters carried by ocean currents, while only true residents thrive when the vent is active and vanish rapidly when it ceases. The CAT lineages exhibit exactly this extreme habitat specificity. Extensive field monitoring shows that their abundances are tightly coupled with hydrothermal activity; once a vent stops erupting and the sulfide chimney enters a dormant phase, the relative abundance of CAT taxa drops precipitously. This "live together, die together" dependency makes them an ideal model for studying biogeographic barriers.
Second, evolutionary time anchors. We constructed a high‑resolution phylogeny of the CAT lineages using 120 concatenated proteins and calibrated the divergence times for each family‑level clade. The results were striking: from the earliest families – Thermosulfidibacteraceae, Desulfurellaceae, and Hippeaceae (originating 2.85–2.08 billion years ago), to the intermediate Nautilaceae and Desulfurobacteriaceae (around 1.6 Ga), and finally to the aerobic families that emerged around 1.0 Ga (such as Sulfurimonadaceae, Aquificaceae, etc.) – all CAT family‑level clades arose long before the birth of the modern Pacific, Atlantic, and Indian Ocean basins.
This time lag is critical. If the CAT lineages had evolved in situ along with the modern ocean basins, we would only be seeing local adaptation. But since they already existed in Earth's ancient oceans long before these basins formed, the uneven modern distribution can only arise from a process of "inheritance" and "migration" from ancient to new oceans – where certain forces blocked their dispersal along the way. RDA analysis showed that the oceanic basin opening time accounted for the largest proportion of the variance in the uneven distribution of CAT families, followed by SO42-, H2S, water depth, NO3-, and tectonic settings.
Chasing "Age"
With this anchoring established, our question became more specific: if the Pacific, Mediterranean, and Arctic – the "old" oceans – inherited their CAT communities from ancient seas, why did the Atlantic and Indian – the "new" oceans – not receive the same legacy?
We systematically compared CAT diversity across ocean basins. The data on species richness and phylogenetic diversity clearly pointed in one direction: the Pacific is unequivocally the "species pool" – most OTUs found in other basins can also be found there; whereas the Atlantic and Indian Oceans not only have fewer total OTUs, but also lack any exclusive OTUs of their own.
But if CAT lineages were already widespread in the global ocean a billion years ago, why did the Atlantic and Indian Oceans fail to inherit these ancient groups?
Oxygen – The Invisible "Border Wall"
The turning point of the story came from a late‑night literature discussion within our team.
The predecessors of the Pacific, Mediterranean, and Arctic – whether the Panthalassa or the Paleo‑Tethys – all formed before the deep ocean became fully oxygenated. At that time, deep waters were rich in Fe²⁺, H₂S, H₂, and S⁰ – a reducing world. For anaerobic CAT lineages that rely on H₂ to reduce S⁰ or perform DNRA (dissimilatory nitrate reduction to ammonium), that was paradise.
In contrast, the Atlantic and Indian Oceans opened after the Phanerozoic deep‑ocean oxygenation (around 420 million years ago), following the breakup of the Pangea supercontinent. By then, seawater was saturated with O₂, SO₄²⁻, and NO₃⁻.
What does oxygen mean for anaerobes? On one hand, direct toxicity – reactive oxygen species damage biomolecules and oxidize low‑redox‑potential metalloenzymes. On the other hand, indirect energy starvation – in oxic waters, reducing substrates like H₂ and H₂S are rapidly depleted.
These two effects together constitute a dual dispersal barrier. Those strictly anaerobic CAT families, like travelers carrying ancient passports, were stopped at the "border" of the new oceans.
"Exceptions" That Prove the Rule
Of course, every scientific story has its unexpected exceptions.
We detected Nautiliaceae – a strictly anaerobic family – in the North and Central Atlantic. How did they get there?
Digging into their genomes, we found that Nautiliaceae is the only anaerobic CAT family that carries a cytochrome bd‑type oxygen reductase – an enzyme that helps detoxify oxygen in low‑oxygen environments. Coupled with the geological record: the North‑Central Atlantic opened first (around 120 million years ago), when the oxygen in the Atlantic deep ocean was assumed to be at a low level and it connected to the Pacific and Tethys Oceans simultaneously.
This explains why only Nautiliaceae managed to establish a foothold in the Atlantic – they had a "basic oxygen‑defense kit" and arrived before the "border closure".
Similarly, the trace amounts of Thermosulfidibacterota found in the Southwest Indian Ocean and the South Atlantic may also be linked to local tectonic histories – these regions might have evolved from parts of the proto‑Pacific, much like the Arctic Ocean.
The exceptions did not overturn the rule; instead, they refined our mechanistic understanding.
"Uniformitarianism" – Using Today's Bacteria to Reconstruct Yesterday's Oceans
What excites us most about this study is actually the methodological dimension.
Traditional Earth‑science research applying "uniformitarianism" (the present is the key to the past) often relies on rocks, minerals, and fossil isotopes. We discovered that microbial functional genes and distribution patterns are themselves living "geological recorders".
This is like using modern DNA to trace ancient human migrations – except our time scale is hundreds of millions of years, and our spatial scale is the global ocean.
Even more thought‑provoking is the evolutionary trajectory of energy metabolism: from the most ancient reaction S⁰ + H₂ → H₂S (simple reduction), to the later addition of DNRA (nitrate utilisation), and then to the emergence of aerobic respiration and sulfide oxidation around 1 billion years ago – this progression almost perfectly synchronises with Earth's atmospheric and oceanic oxygenation. Biological evolution is never an isolated event. It is the "log file" of the Earth system in operation.
Microbes May Be Small, but They Record the Pulse of a Planet
Looking back at the entire study, what I most want to convey is not any single scientific conclusion, but a shift in perspective:
Microbes are not passive chemical pawns merely "responding" to the environment. They are "planetary organisms" that have co‑evolved with plate motion, ocean chemistry, and atmospheric oxidation over billion‑year timescales.
The century‑old adage "everything is everywhere, but the environment selects" might need a geological footnote: "Everything is everywhere, but the deep history of the planet selects, too."
This paper is the evidence we found for that statement.