Does endo-symbiosis explain the origin of the nucleus?

To the editor

Horiike et al.¹ found that yeast proteins involved in transcription, translation, DNA replication and the like are more similar to archaebacterial homologues, whereas those involved in metabolism are more similar to eubacterial homologues. They conclude that such findings “strongly support” the notion of an origin of eukaryotic nuclei through endosymbiosis of an archaebacterium in a eubacterial host¹. Their observations are valid but the inferences about cell evolution drawn from them are flawed.

First, their conclusions hinge upon the existence of a correlation between protein compartmentation and gene origin in eukaryotes. But each functional category of genes studied, particularly the ones specific to cell compartments, contained proteins of eubacterial and archaebacterial origins¹. At the genome-wide level, protein compartmentation is a poor indicator of gene origin⁶. Indeed, eukaryotic ribosomes are archaebacterial but are localized in the cytosol, whereby nuclear symbiotic models¹ would predict them to be nuclear.

Second, all nuclear symbiotic hypotheses³ derive a primitively amitochondriate, nucleus-bearing cell — an archaezoon — as the host for mitochondria. But all mitochondrion-lacking eukaryotes studied so far possessed a mitochondrion in their past^7,8, so they can hardly be descendants of that host. If all nucleus-bearing cells also possess(ed) mitochondria, how can we tell what came first³? Inferences that the origin of the nucleus has been revealed and that this involved an archaebacterial symbiont¹ are unsubstantiated.

Third, models for endosymbiotic nuclear origins draw upon the finding that the nucleus contains DNA (as do chloroplasts and mitochondria) and from the superficial similarity that can be construed (if the nuclear membrane is depicted incorrectly¹) between its membrane and the double membranes surrounding chloroplasts and mitochondria. The nucleus is bounded by a single, folded membrane (no free-living cells are bounded similarly)³, its pores are permeable to molecules of relative molecular mass 5,000 (not true for any prokaryote)³, and it disintegrates at open mitosis (no compartment of demonstrably endosymbiotic origin does anything vaguely similar)³.

The new report¹ underscores “the archaebacterial nature of the eukaryotic genetic apparatus and the eubacterial nature of eukaryotic energy metabolism”⁹ — the focus of an alternative model⁹ for the origin of eukaryotes that derives the nucleus in a mitochondrion-bearing cell³ and directly accounts for eubacterial lipids and importers¹ in the eukaryotic plasma membrane^3,9.

Not everything in a eukaryotic cell is a direct inheritance from prokaryotes — Darwin's principle of descent with modification includes the possibility of invention. After all, there must have been a time when the ancestor of eukaryotes did not possess a nucleus; the question is whether that cell possessed a mitochondrion or not.