A bacteria has been found with many characteristics of eukaryote cells, especially something that looks like a nucleus:

https://blogs.scientificamerican.com/artful-amoeba/pores-inside-bacteria-are-eerily-fam...

Gemmata obscuriglobis excels at breaking rules. Like the platypus, to whom these bacteria have been compared, they possess a baffling arsenal of oddities.

"Although it has been controversial, they seem to contain membrane-bound compartments. One of those compartments surrounds their DNA. That would make it, apparently, a nucleus. But bacteria are thought to be devoid of nuclei – hence the terms prokaryote (“pre-kernel”) for bacteria and archaea, and eukaryote (“true kernel”) for all nucleated life (which includes all multicellular organisms).

"The eye-popping apparent commonalities don’t end there. Recently, scientists have also observed Gemmata obscuriglobis apparently engulfing its food in a process similar to endocytosis and  manufacturing its DNA, RNA, and protein inside special compartments. All of these abilities have long been considered the sole domain of eukaryotes. What had been hitherto unsuspected is that the roster of uncanny similarities extends right down to tiny, intricate holes in the "nucleus".

***

"Now scientists from Australia and New Zealand have discovered, in research published in February in PLOS ONE, that embedded in the putative nucleus of G. obscuriglobis are what appear to be gated pores of the same structure and type that stud our nuclei – right down to the way in which the pore proteins are folded.

"If that is the case, it means one of two equally astounding things must be true: either this humble bacterium, isolated from freshwater near the Maroon Dam in Queensland, Australia is the closest living relative of eukaryotes, and split from our last common ancestor with bacteria long ago. Or it has independently evolved, under similar pressures, shockingly similar solutions to the same cellular and biochemical problems, right down to the architecture of its nuclear pores.

***

"There are substantial differences, however. The bacterial pores are much smaller – just about a third of the size of our own. And the actual amino acid sequences of these proteins – proteins are long chains of amino acids -- did not bear a resemblance to eukaryotic versions, indicating they are not simply recent donations from a eukaryote to a bacterium. Whatever their origin, these pores evolved inside bacteria, and most likely have been there a long time.

"What does it mean that these bacteria have nuclear pores with architecture astoundingly similar to our own, but whose construction blueprints bear no obvious similarities? The authors of the paper suggest two possibilities. In one scenario, Gemmata shares a common ancestor with the first eukaryotes, and would be our closest living bacterial relative. The sequences are different because so much time has passed since we shared a common ancestor: at least a billion years and probably more. If this is the case, it would make Gemmata extremely important for understanding our evolution.

"On the other hand, these structures could have evolved independently in response to similar evolutionary pressures. Perhaps G. obscuriglobis has a strong need, for reasons not apparent right now, for extremely fine-tuned control of its protein production machinery. If so, the bacteria would be yet another example of convergent evolution – a potent force of physical similarity on Earth -- but a particularly striking one. The authors think this second possibility is more likely because of the radical differences in pore sizes and protein composition.

"The very existence of G. obscuriglobis also strikes a blow to endosymbiotic theory, the authors suggest. This idea, promulgated most famously by Lynn Margulis but also others, suggest that complex life arose thanks to a series of fusions between bacteria and archaea. Mitochondria – cellular engines -- and chloroplasts – solar panels -- almost certainly formed this way. Some have suggested the cell nucleus shares a similar origin, possibly when a bacterium engulfed an archaeon (the proteins we use to manufacture DNA, RNA and protein bear a striking resemblance to those of archaea, among other similarities). G. obscuriglobis at least suggests that it is possible to add a fundamental new piece of cellular furniture without the fusion of unrelated microbes.

Comment: An early eukaryote ancestor or an example of convergence? A puzzle. It does offer a possible challenge to Lynn Margulis endosymbiotic theory.