THEORY-1105: In Bernabeu, Manzano-Morales, Marcet-Houben and Gabaldón's reconstruction of the last eukaryotic common ancestor's proteome, two bacterial phyla besides the mitochondrial ancestor, Myxococcota and Planctomycetota, each account for a share of the proteome comparable to that of Alphaproteobacteria and Asgard archaea under thresholds calibrated on those accepted partners, and stem-length distributions order the acquisitions as waves, Asgard first, then Planctomycetota, then the mitochondrial ancestor and Myxococcota together; the authors read this as serial associations with several partners, while conceding that their data cannot say whether any donor was a symbiont

Source

Bernabeu, Manzano-Morales, Marcet-Houben & Gabaldón (2026), Nature 656: 399–405, LIT-1561, read in full in NOTE-1237 from the PubMed Central copy (PMC13468126). The sections “Determining the ancestries of LECA” and “Donor relative timing and metabolism”, the Discussion, Methods (“Stress test”, “Verticality test”, “Inferring the acquisition waves”), Figures 2–3 and Extended Data Figures 2–3, read from captions.

What was actually shown

• The proteome. Three 100-proteome eukaryotic datasets, LECA families defined by spread across supergroups, and maximum-likelihood trees with the closest prokaryotic and viral homologues from order-level pangenomes. About 53% of families are acquisitions from outside eukaryotes.
• Calibration. Support thresholds were chosen to maximize the share of trees assigned to Alphaproteobacteria and Asgard, “assuming that the selected thresholds properly removed noise and kept only the reliable signals”.
• Donors at those thresholds: Alphaproteobacteria 3.99%, Myxococcota 3.92%, Asgard 3.28%, Planctomycetota 2.24% of the proteome. Chloroflexota and Gammaproteobacteria failed in some datasets.
• Controls, reported in the main text, run in the supplement.
  • Trees for the new donors were not harder to resolve.
  • With Alphaproteobacteria removed from the database, alphaproteobacterial genes were assigned mainly to Gammaproteobacteria, then to more distant groups. The Myxococcota and Planctomycetota signals do not follow that pattern.
• Waves. Posterior modes of each donor’s normalized stem-length distribution give the order: Asgard, “the most ancestral signal”; then Planctomycetota; then “the mitochondrial ancestor and Myxococcota, with largely overlapping transfer waves”. The order was consistent across datasets and stringencies.
• The reading. “our data support premitochondrial waves of gene acquisitions from additional major prokaryotic donors as the most likely scenario”. The results “challenge models depicting eukaryogenesis as a one-off event resulting from a binary symbiotic interaction”. And: “our data do not provide information on potential types of interaction (endosymbiotic, ectosymbiotic, parasitic, and so on)”.

What could have failed: the extra donors could have failed the calibrated thresholds, or tracked the pattern expected of misplaced alphaproteobacterial genes, and they did neither. The waves could have had the same mode for every donor. They did not, though the method that measures them is the one in dispute.

What this does not say

• Not that there were more than two symbiotic partners. The donors are sources of genes. The authors say “different forms of interactions beyond symbiosis could explain this genetic exchange”.
• Not that the donors’ genes came to eukaryotes directly. The verticality test addresses misplacement of alphaproteobacterial genes. It does not address bacterial genes that reached the Asgard ancestor first, which Tobiasson et al. propose, and which would also leave a donor signal. The inference is mine.
• Not a timing result independent of stem length. If stem length is not a clock (THEORY-1101), the order of waves, and with it the claim that Planctomycetota preceded the mitochondrion, falls back to unknown.
• Not that Asgard contributed only 3% of LECA. That figure is the share passing strict thresholds. It is not comparable with Tobiasson et al.’s 50% of likelihood weight. The two measure different things.
• Not a reconstruction of the donors. Their metabolisms are prevalences in living relatives, “incomplete and indirect”.

Connections

• THEORY-1104 (Tobiasson et al.): rival on the donors, compatible on the mitochondrion. The two disagree on whether the non-alphaproteobacterial genes form a few major donors’ waves or diffuse transfer, and on the Asgard share. Tobiasson et al. attribute this paper’s larger bacterial signal to “more liberal criteria for core pangenome construction” (15% of genera per order here, against 50% of species per class there). Both papers have the mitochondrion arriving late and bacterial genes arriving before it. Neither has data on the kind of interaction.
• THEORY-1101: rival. It disputes the method that orders the waves.
• THEORY-1109 (Santana-Molina et al.): partly compatible. In central carbon metabolism, Planctomycetota–Verrucomicrobiota and Myxococcota– Desulfobacterota both appear as sisters, as this account would expect. But that paper finds “no additional dominant source of gene donations”.
• LIT-1595 (Pittis & Gabaldón 2016, read in NOTE-1240). This is that programme’s continuation, now with named donors.
• QUESTION-277. This is the “distributed” side of the record’s live disagreement. The distribution shown is of gene donors, so it bears on G. It bears on the question’s routes to dependence only if the donors were partners, and the paper leaves that open.

Note of 2026-10-30: promoted, and the wider thesis filed beside it

The twenty-second status review promoted this reading under ADR-065: it reports what Bernabeu and colleagues’ reconstruction finds and how the authors read it, and NOTE-1237 read the paper in full. Whether Myxococcota and Planctomycetota were major donors in waves is THEORY-1215’s question, Proposed, which keeps this entry’s old promote_when.