In a microbial mat along the coast of Shark Bay, Australia, a research team has captured something biologists have long theorized but never directly witnessed in nature: an archaeon physically pressed against a bacterium in the kind of intimate cellular partnership thought to have launched complex life on Earth roughly two billion years ago.
The organism, a newly identified species called Nerearchaeum marumarumayae, belongs to the Asgard archaea, a group that genomic studies have placed as the closest known prokaryotic relatives of all eukaryotes, the domain that includes animals, plants, fungi, and humans. In a peer-reviewed study published in Current Biology in April 2026, a team led by Margaux Meslier and colleagues at the University of New South Wales used electron cryotomography to reconstruct three-dimensional images of the archaeon at near-molecular resolution, documenting direct surface contact between Nerearchaeum and neighboring bacteria within a living microbial community.
The result is not a computer simulation or a genetic inference. It is an observed association drawn from mats that scientists regard as modern analogs of ancient ecosystems, and it offers the first visual evidence from a natural setting for the type of archaea-bacteria alliance central to leading theories of eukaryotic origins.
What is verified so far
The imaging data form the core of the finding. Electron cryotomography freezes cells in a near-native state and generates detailed 3D reconstructions without the chemical fixation that can distort delicate structures. Applied to samples enriched from Shark Bay’s mats, the technique revealed Nerearchaeum marumarumayae cells in direct physical contact with bacteria, their membranes closely apposed in a way consistent with metabolic exchange.
That observation builds on a landmark 2020 study in Cell by Hiroyuki Imachi and colleagues at the Japan Agency for Marine-Earth Science and Technology (JAMSTEC). Over more than a decade of painstaking cultivation, that team grew the first Asgard archaeon in the laboratory: Prometheoarchaeum syntrophicum. The organism extended long, branching protrusions and engaged in syntrophic interactions with bacterial partners, exchanging metabolic byproducts in a mutually dependent relationship. The researchers argued this kind of partnership could explain how a simple archaeal cell eventually engulfed a bacterium that became the mitochondrion, the energy-producing organelle found in virtually all eukaryotic cells.
Where Prometheoarchaeum demonstrated that such contact was possible under controlled laboratory conditions, Nerearchaeum shows it happening in a natural environment with steep chemical gradients and tightly packed microbial communities.
Separate molecular evidence adds another dimension. Asgard archaeal genomes encode proteins strikingly similar to eukaryotic cellular machinery, including regulators involved in membrane trafficking and organelle dynamics. Phylogenomic analyses consistently place Asgard archaea as the closest prokaryotic relatives of eukaryotes, and studies of specific protein families have traced key regulators of membrane transport back to Asgard lineages. That body of work provides a mechanistic explanation for how an archaeal host could have developed the internal complexity needed to manage an endosymbiont.
Imaging, culture-based, and genomic data all point in the same direction: Asgard archaea are not merely distant relatives of eukaryotes but plausible stand-ins for the kind of host cell that once entered into a transformative alliance with bacteria. The Shark Bay mats offer a real-world stage on which such alliances can be seen rather than only inferred.
What remains uncertain
Physical contact is not the same as gene transfer. The Current Biology study does not document horizontal gene transfer at the moment of contact. The images show proximity and surface interaction, but whether these organisms are exchanging DNA, RNA, or metabolic signals at the observed interface has not been established. That gap matters because most models of eukaryogenesis require not just physical partnership but the wholesale transfer of genetic information from the bacterial endosymbiont to the host genome over evolutionary time.
Competing models of how eukaryotic cells arose also remain in play. A detailed survey of origin scenarios in Nature Reviews Microbiology by Purificacion Lopez-Garcia and David Moreira laid out several alternatives, including “inside-out” models in which the archaeal host gradually extended membranes around a bacterial partner, and “outside-in” models in which the bacterium was actively engulfed. The Shark Bay images are consistent with syntrophic partnership models but do not rule out other pathways.
There are also open questions about how representative Nerearchaeum marumarumayae is of ancestral Asgard lineages. The species was identified in one specific environment with particular salinity and nutrient conditions. Whether similar contact behaviors occur across diverse Asgard clades, or whether this is a specialized adaptation to Shark Bay’s mats, is not yet clear. Without a broader ecological survey, extrapolating from this single case to the universal ancestor of eukaryotes remains a stretch.
Timing is another sticking point. Molecular clock estimates for when the archaeal host and bacterial symbiont merged vary widely, and the fossil record offers only indirect constraints. The Shark Bay community is a modern system shaped by today’s oxygen levels, climate, and geochemistry. It may echo some conditions of early Earth, but it is not a replica. Any claim that what is seen there precisely recapitulates the steps of eukaryogenesis overstates what the data can support.
Strength of the evidence
The strongest piece of this puzzle is the direct imaging data: electron cryotomography of Nerearchaeum marumarumayae in physical contact with bacteria, captured from a natural environment. This is primary observational evidence. Future work may reveal what passes across that contact zone, but the existence of the contact itself is not in doubt.
The second layer is molecular and genomic. Asgard genomes contain genes that code for cellular machinery previously thought to exist only in complex organisms. Phylogenomic analyses consistently place Asgard archaea as the closest prokaryotic relatives of eukaryotes. These data do not show eukaryogenesis happening in real time, but they establish that the genetic toolkit for complex cell biology existed in archaeal lineages before eukaryotes appeared.
The third layer is interpretive. Syntrophic partnership models, inside-out hypotheses, and other eukaryogenesis frameworks are analytical constructs built on the first two layers. They generate testable predictions, such as expecting to find protrusion-forming archaea in environments rich in potential bacterial partners, but none has been confirmed to the exclusion of the others.
Readers should distinguish between what has been seen (physical contact in a modern mat), what has been inferred with high confidence (the close evolutionary relationship between Asgard archaea and eukaryotes), and what remains conjectural (the exact sequence of events that produced the first complex cell). The Shark Bay work is observational, it slots into models discussed for years, and its relevance to events two billion years ago is suggestive rather than definitive. Modern microbes are analogs, not time machines.
Independent lines of evidence, from cultured Asgard strains, from genomes, and now from natural microbial mats, all point toward a scenario in which an archaeal host with eukaryote-like features formed a tight metabolic alliance with bacteria. The Shark Bay finding does not close the debate over eukaryotic origins, but it makes that alliance visible in a way that genomic data alone never could. As additional Asgard species are characterized and more natural communities are imaged, biologists will be able to test whether Nerearchaeum marumarumayae is an outlier or a modern glimpse of the kind of partnership that ultimately made complex life possible.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
