Eukaryogenesis: Questions in Questions in Questions

by Janie
 

Lynn Margulis's name is synonymous with symbiogenesis. Lesser known is her dabbling in fiction-writing, although some might argue that her more controversial theories could be called works of fiction… and in response to such critics in her time, she purportedly said, "I don't consider my ideas controversial. I consider them right."
 

It's interesting to see that even her single collection of short stories, Luminous Fish, is made of nested stories and intertwining histories of individuals, mirroring her life's work on endosymbioses. It's a montage of portraits – or rather, exposés – of scientist characters both fictional and real as they go about their days. The occasional snappy gibe seemed in-character, too.
 

With the recent spate of articles related to eukaryogenesis from the past handful of months, and in the spirit of such nested stories, here is a different montage of portraits: this time, of theories for eukaryogenesis.
 

Figure 1. Engraving print of Yggdrasil, the tree of life in Norse myth­ol­ogy, by Oluf Olufsen Bagge, 1847. Source. Frontispiece.

To begin with, the history is no linear path. For a very long time, symbiogenesis was scoffed at as a fringe science. This mindset plagued the field from the get-go in the 1880s, when the botanist Andreas Franz Wilhelm Schimper first put forth the idea that eukaryotic organelles might be bacterial, to the turn of the century, which saw the first key paper on symbiogenesis by Konstantin Mereschkowski (who, despite all his ugly ideologies and actions, did set the stage for further study on eukaryogenesis). The skepticism persisted well into the second half of the twentieth century. To finally turn the popular tide and garner more than a dismissive wave from scientific circles, it took more than Lynn Margulis's vocal advocacy alone. The watershed? The rise of molecular biology. In the 1970s, equipped with Carl Woese's newfangled rRNA cataloguing, Linda Bonen and W. Ford Doolittle turned their attention to the little red alga's ribosomes, to find that the rRNA sequences from the chloroplasts were far more similar to those of cyanobacteria than to those of the algal host cell itself. This was followed by similar comparisons for mitochondria in wheat germ, and in this way, the first pieces of genetic evidence for organellar kith and kin had been assembled. Once part of the bacterial family (strictly speaking, domain), always part of the bacterial family.
 

So, it was established that mitochondria and chloroplasts descend from bacteria. This would be further supported by later work with mitochondrial rRNA plucked from additional organisms and with comparisons of cytochrome C. But in the 100+ years since a botanist kickstarted the field, not much more beyond this has arrived at a consensus.
 

The traditional view of endosymbiosis goes like this: a couple billion years ago, a hungry prokaryotic cell engulfed another one. This swallowed-up "another one" was an early alphaproteobacterium that possessed the nifty ability to carry out aerobic respiration. In a twist of fateful indigestion, it persisted within its would-be predator, where it metabolically wrangled oxygen before the dangerous molecule could harm the host. Fast-forward through evolutionary time, and the situation had evolved from a would-be snack into the meat and potatoes of eukaryogenesis: the two nested cells became increasingly interdependent to the point that their metabolisms were mosaicked with each other.
 

But here is where skeptical eyebrows begin to raise. That very first phagocytosis step implies that the unspecified prokaryotic host cell was already equipped with an intricate cytoskeletal system. Alternative theories point instead towards a more gradual co-development of such complexity. These posit that the merging of the two cells was not as instantaneous as phagocytosis, and that rather, prokaryotic host and bacterium grew together in physical proximity, becoming increasingly co-dependent before finally merging. In some of these theories, the bacterial cell was not just a nice addition to a pre-existing collection of fancy intracellular baubles like cytoskeletons and nuclear compartments, but rather a key player in the evolutionary dynamics that would give rise to such novelties.
 

Figure 2. Left: Representative SEM image of 'Candi­datus Pro­meth­eoar­chae­um syn­troph­icum' MK-D1. White ar­rows indi­cate large mem­brane ves­icles. Source. Right: The Entangle-Engulf-Endogenize (E³) model prop­oses that euk­ary­otes evolved from a prog­res­sive intern­ali­za­tion of an alpha­prot­eobac­ter­ium inside an Asgard arch­aeon ini­tially assoc­iated to a sulfate-reducing delta­proteo­bac­ter­ium that would later dis­ap­pear from the con­sor­tium. Source

It's now widely agreed upon that the key to eukaryogenesis was the partnership between an archaeal cell – from within the Asgardarchaeota superphylum – and an alphaproteobacterial cell. In the last decade, genome and metagenome analyses showed that Lokiarchaeota in particular possess proteins previously thought to be unique to eukaryotes. But some theories also invoke a sulfate-reducing deltaproteobacterium player.
 

Cut to 2020. Imachi et al. were the first to cultivate an Asgard archaeon, through a decade-long effort. With this, Asgard archaea were no longer only metagenomic manifestations, but very real and very tangible living things! From the murk of deep-sea methane-seep sediment surfaced 'Candidatus Prometheoarchaeum syntrophicum,' a peculiar little round cell (~500 nm diameter) with long, tentacle-like membrane protrusions. This Asgard archaeon came closely associated with a deltaproteobacterium (Halodesulfovibrio) and a Methanogenium, and it possessed a suite of highly expressed eukaryotic-like proteins including cytoskeleton-related genes. The authors suggested a prehistoric scenario in which a H₂-producing Asgard archaeon was initially in cahoots with a sulfate-reducing deltaproteobacterium, a tie that would later be cut as the tentacled archaeon gradually enveloped an oxygen-respiring alphaproteobacterial progenitor of the mitochondrion. They call this "slow entanglement," in contrast to phagocytosis.
 

Figure 3. Representative SEM image of 'Candidatus Loki­arch­aeum os­si­ferum.' Source

Then in December 2022 came the first piece of visual evidence for archaeal actin-based cytoskeletons, after many years of speculation. Rodriguez-Oliveira and coworkers isolated an anaerobic Lokiarchaeon from a canal in Slovenia. This bug, like its 2020 kinfolk, is coccoid with tentacle-like membrane protrusions. Further, its interior is interlaced with actin-like cytoskeletal filaments composed of what the authors name "Lokiactin." These are actin homologs that are highly conserved throughout Asgard archaea, and are the reason for the authors' support for the model that the archaeal ancestor of eukaryotes already possessed a complex cytoskeletal system prior to its merge with bacteria.
 

There is a big theoretical problem with these eukaryogenesis models, however. Their hypothetical outermost membrane is archaeal. Today's eukaryotes have cell membranes that are much more like those of bacteria than those of archaea. Archaeal membranes, from a non-archaeal point of view, are plain weird: the chirality of the glycerol in their membrane lipids is L instead of D, the side chains connected to that glycerol are isoprene chains instead of fatty acids, and that connection itself is an ether linkage rather than an ester linkage. If the aforementioned models were true, that would mean that the protoeukaryotic cell would have had to do a serious overhaul of its membrane architecture and chemistry to arrive at the present-day phospholipid bilayer. Not exactly the kind of razor Occam would favor. (But maybe not an insurmountable hurdle after all, since it is apparently possible to Turn An E. coli  Into An Archaeon (Sort Of).)
 

So, yet another point of contention: Who was the original host for the alphaproteobacterium that became the mitochondrion? Was it really a Lokiarchaeal cell, or perhaps a bacterial cell?
 

One model for eukaryogenesis that sides with archaea-on-the-outside-bacteria-in-the-inside invokes an endomembrane system. This model addresses the membrane problem by suggesting that the engulfed alphaproteobacterium continuously shed outer membrane vesicles that merged with the archaeal plasma membrane, and that these bacterial lipids came to dominate the composition of the plasma membrane.
 

Figure 4. The revised Syntrophy Hy­poth­esis: euk­ary­otes ev­olved from the in­cor­por­ation of an Asgard arch­aeon (future nuc­leus) within a delta­proteo­bac­ter­ial host. This was fol­lowed by the ac­quis­ition of the alpha­proteo­bac­ter­ial an­ces­tor of mito­chon­dria. Source

Meanwhile, the syntrophic model of eukaryogenesis posits that a bacterial cell was the outermost nesting doll. This version of the story begins with a deltaproteobacterium engulfing an Asgard archaeal symbiont, followed next by the engulfing of an oxygen-respiring alphaproteobacterial symbiont. That first symbiont became the nucleus, and the second became the mitochondrion – the assumption here being the archaeal nuclear membrane was lost and/or replaced by a bacterial membrane.
 

A useful tack in this muddle of membranes among archaea, bacteria, and eukarya is the other very important "biological entity" that has not been mentioned yet. Viruses – which engage with their target cells' membranes whether archaeal, bacterial, or eukaryotic – could provide some insight. A couple of months ago, a paper from Mart Krupovic, Valerian Dolja, and Eugene Koonin presented a reconstruction of the virome of the last eukaryotic common ancestor (LECA). This was done by tallying groups of (known) viruses represented in three or more eukaryotic supergroups. Of the six realms of viruses – Varidnaviria, Monodnaviria, Riboviria, Ribozyviria, Duplodnaviria, and Adnaviria – the first four listed are primarily comprised of viruses that infect eukaryotes. These viruses are descendants of bacteriophages. This is interesting. With a LECA virome that has such ties to bacterial origins, that outermost cell of the protoeukaryote must, too, have been bacterial. So the authors reason.
 

Their theory is not the first to put viruses at center stage in eukaryogenesis. One that is more offbeat is the model of viral eukaryogenesis, which claims that the eukaryotic nucleus is the evolutionary descendant of a big DNA virus.
  

And so, from the 1880s to 2023, nearly all the big questions still remain big questions.
 

 

Do you want to comment on this post? We would be happy about it! Please comment on Mastodon or on Twitter.

Note: Thank you to Andrew Roger, who kindly pointed out a mix-up of names in the earlier post – Andreas Franz Wilhelm Schimper has now been given proper credit!