Papex

Open problem · Biology · Open

Nature of the last universal common ancestor

What was the last universal common ancestor of all living things like?

Summary

All living organisms appear to descend from one ancestral population, but no fossil of it is known. Its biochemistry, habitat and age have to be inferred from genes shared by modern life, and estimates still differ.

Why it is open

The near-universal genetic code, shared protein-making machinery and ATP-based energy metabolism support common ancestry of cellular life [14]. LUCA is the node on the tree of life where the two prokaryotic domains, Archaea and Bacteria, diverge [14]; eukaryotes appeared much later [7], and how is a separate problem, the origin of eukaryotes. LUCA is inferred from living organisms; fossils do not directly reveal its genome [7, 14].

Genes move between lineages by horizontal gene transfer, so a gene shared by archaea and bacteria need not have been in LUCA, and later transfers complicate reconstruction [14]. Carl Woese argued in 1998 that early transfer was so pervasive that the universal ancestor "is not a discrete entity" but "a diverse community of cells" [1]. Methods also disagree: published estimates range from 80 proteins to 1,529 gene families [14], and of eight reconstructions compared by Andrew Crapitto and colleagues in 2022, none shared "a high or even moderate degree of similarity" with any other [11].

The sources ask for a gene content robust to transfer and to uncertainty in the species tree [14], ages that do not hinge on the calibration of the tree's root [12], and, for Aaron Goldman and Arturo Becerra, a characterisation focused "on traits under strong natural selection", which they consider "more robust and testable" than a detailed list of metabolic genes [15].

What is known

A core of shared genes (2003–2016). In 2003 J. Kirk Harris, Norman Pace and colleagues found a small set of genes whose trees follow the pattern of ribosomal RNA back to the universal ancestor; almost all of them handle genetic information, and most interact directly with the ribosome [2]. In 2016 Madeline Weiss, William Martin and colleagues screened 6.1 million prokaryotic protein-coding genes and found 355 protein families, about 0.1% of the clusters, that trace to LUCA by phylogenetic criteria [6]. Their functions depict LUCA as anaerobic, thermophilic and dependent on hydrogen, fixing carbon dioxide with the Wood–Ljungdahl pathway and fixing nitrogen, in "a geochemically active environment rich in H2, CO2 and iron", which the authors read as support for an origin of life in a hydrothermal setting [6].

Challenges to that picture (2008–2024). A 2016 reply by Johann Peter Gogarten and David Deamer, as Luis Delaye summarises it in a 2024 review, holds that the method favours genes shared by transfer between prokaryotes adapted to the same environment, hot environments being one case, and filters out many genes LUCA had [13]. Rebuilding the trees with more homologous sequences, Sarah Berkemer and Shawn McGlynn found in 2020 that about 82% of the families no longer fitted a simple LUCA origin [9]. Reverse gyrase, an enzyme present in all hyperthermophiles and absent from mesophiles, was among Weiss and colleagues' LUCA families [13]; in Ryan Catchpole and Patrick Forterre's 2019 trees of 376 sequences its archaeal and bacterial forms did not form separate groups, and they suggested that LUCA was not a hyperthermophile [8]. The consensus of eight studies describes a LUCA that at minimum made proteins, metabolised amino acids and nucleotides and used common nucleotide-derived cofactors [11].

Age estimates (2018–2024). Molecular clocks convert genetic distance into time using calibrations from fossils and geochemistry [14]. Combining a reappraisal of early fossils with clock analyses of 29 universal genes, Betts and colleagues placed LUCA in 2018 before the end of the proposed late heavy bombardment, more than 3.9 billion years ago [7]. In 2023 Tara Mahendrarajah and colleagues used cross-bracing, which forces nodes recording the same speciation event in duplicated genes or in host and endosymbiont lineages to share one age, and obtained 4.52–4.32 billion years [12]. In 2024 Edmund Moody and colleagues used five gene families duplicated before LUCA, with fossil and isotope calibrations, to estimate 4.09–4.33 billion years [14].

A prokaryote-grade ancestor (2020–2024). Mart Krupovic, Valerian Dolja and Eugene Koonin projected today's viruses back to LUCA and reconstructed "a remarkably complex virome", implying a complex LUCA genome [10]; where viruses came from is the separate problem of the origin of viruses. Moody and colleagues modelled gene duplication, transfer and loss for 9,365 gene families across 350 archaea and 350 bacteria, assigned 399 families to LUCA under their stringent probability and taxonomic-sampling criteria and, integrating over all families, estimated a genome of at least 2.5 megabases encoding around 2,600 proteins, comparable to modern prokaryotes [14]. Their LUCA lacks terminal oxidases and so was probably anaerobic; it has almost all of the archaeal Wood–Ljungdahl pathway, suggesting an acetogen (an organism that makes acetate and can grow on hydrogen and carbon dioxide), and Cas proteins that they interpret as an early immune system [14]. They found no strong evidence for nitrogen fixation, the reconstruction "was not definitive with respect to the optimal growth environment", and they consider two main habitats, the deep ocean near hydrothermal vents and the ocean surface [14].

Recent claims. Moody and colleagues' reconstruction appeared in Nature Ecology & Evolution in July 2024 [14]. In an August 2024 commentary, Goldman and Becerra welcomed the methodological advance but questioned how to interpret its estimated proteome [15]. Noting the gap between 399 high-confidence families and 2,451–2,855 estimated proteins, they suggested reading the reconstruction as the core of a pangenome shared by a population of cells [15].

Main approaches

Strict phylogenetic filters. Keep only gene families whose trees separate archaea from bacteria [2, 6]. Strongest evidence: the genes that pass Harris and colleagues' test follow the ribosomal RNA tree, and almost all handle genetic information [2]. Main difficulty: filters admit genes transferred between the domains and exclude many LUCA genes transferred later [13]; a re-analysis rejected most of one such set [9].

Probabilistic reconciliation. Model each gene family's history against a species tree, transfers included, and give it a probability of being in LUCA [14]. Strongest evidence: modelling transfer admits many more families than filters do, and the result agrees with earlier studies on core features such as ATP synthase [14]. Main difficulty: not all genes LUCA must have had, in the ribosome and the aminoacyl-tRNA synthetases, are recovered, so the authors call the result "probably incomplete" [14].

Temperature markers. Infer LUCA's temperature from the composition of ancestral sequences or from marker genes [3, 8]. Strongest evidence: in 2008 Bastien Boussau and colleagues' models of rRNA and protein evolution favoured a mesophilic LUCA whose two descendant lineages adapted to heat independently [3]. Main difficulty: the markers conflict: Moody and colleagues assign reverse gyrase to LUCA with probability 0.97 while noting that its evolution is complex [14], whereas Catchpole and Forterre read its trees as an origin after the domains split, followed by transfer [8].

Cross-braced clocks. Use gene duplications older than LUCA, so that LUCA appears twice in each tree and each calibration counts twice [14]. Strongest evidence: single genes, and concatenations leaving out each gene in turn, give compatible timescales [14]. Main difficulty: in Mahendrarajah and colleagues' analysis LUCA's interval still ran into the maximum set by the Moon-forming impact; the estimated age depends on that root calibration [12].

Energy and membranes. Víctor Sojo, Andrew Pomiankowski and Nick Lane modelled a LUCA powered by natural proton gradients; this works only in leaky cells, so ion-tight phospholipid membranes would have arisen later and independently in archaea and bacteria [5]. Strongest evidence: the model accounts for why the two domains share ATP synthase yet differ in membrane lipids and DNA replication [5]. Main difficulty: phylogenomic studies suggest instead that the common ancestor had complex phospholipid membranes [4], and Moody and colleagues recovered enzymes for making both ether and ester lipids without establishing the membrane's composition [14]. See also the lipid divide.

Open sub-questions

  1. Reverse gyrase. Does reverse gyrase still map to LUCA when Catchpole and Forterre's 376 sequences are reconciled with an explicit model of transfer [8, 14]?
  2. Root calibration. How far does the 4.2-billion-year estimate move when the root maximum and fossil or isotope calibrations are varied separately [12, 14]? See also Origin of life.
  3. Agreement between methods. Which of the 399 high-probability families are in the eight-study consensus, and do families that fail Berkemer and McGlynn's test cluster in metabolism [9, 11, 14]?
  4. LUCA's membrane. Does reconciliation give LUCA enzymes for both glycerol-1-phosphate and glycerol-3-phosphate lipids with high probability, or only one [4, 5, 14]?
  5. LUCA's viruses. Do the key virus genes placed in LUCA's virome reach LUCA in a reconciliation that models transfer [10, 14]?
  6. Contemporaries. Can genes passed into living lineages by extinct contemporaries of LUCA be detected, and how much of the estimated proteome would they explain [1, 14]?

Work on this in Papex

If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.

Key references

  1. Woese, C. (1998). The universal ancestor. Proceedings of the National Academy of Sciences 95(12), 6854–6859. 10.1073/pnas.95.12.6854 · Checked 15 Oct 2026
  2. Harris, J. K., Kelley, S. T., Spiegelman, G. B. & Pace, N. R. (2003). The genetic core of the universal ancestor. Genome Research 13(3), 407–412. 10.1101/gr.652803 · Checked 15 Oct 2026
  3. Boussau, B., Blanquart, S., Necsulea, A., Lartillot, N. & Gouy, M. (2008). Parallel adaptations to high temperatures in the Archaean eon. Nature 456(7224), 942–945. 10.1038/nature07393 · Checked 15 Oct 2026
  4. Lombard, J., López-García, P. & Moreira, D. (2012). The early evolution of lipid membranes and the three domains of life. Nature Reviews Microbiology 10(7), 507–515. 10.1038/nrmicro2815 · Checked 15 Oct 2026
  5. Sojo, V., Pomiankowski, A. & Lane, N. (2014). A bioenergetic basis for membrane divergence in archaea and bacteria. PLoS Biology 12(8), e1001926; correction 13(3), e1002102 (2015), which fixes a typographical error and a small code error without changing the conclusions. 10.1371/journal.pbio.1001926 · Checked 15 Oct 2026
  6. Weiss, M. C., Sousa, F. L., Mrnjavac, N., Neukirchen, S., Roettger, M., Nelson-Sathi, S. & Martin, W. F. (2016). The physiology and habitat of the last universal common ancestor. Nature Microbiology 1(9), 16116. 10.1038/nmicrobiol.2016.116 · Checked 15 Oct 2026
  7. Betts, H. C., Puttick, M. N., Clark, J. W., Williams, T. A., Donoghue, P. C. J. & Pisani, D. (2018). Integrated genomic and fossil evidence illuminates life's early evolution and eukaryote origin. Nature Ecology & Evolution 2(10), 1556–1562. 10.1038/s41559-018-0644-x · Checked 15 Oct 2026
  8. Catchpole, R. J. & Forterre, P. (2019). The evolution of reverse gyrase suggests a nonhyperthermophilic last universal common ancestor. Molecular Biology and Evolution 36(12), 2737–2747. 10.1093/molbev/msz180 · Checked 15 Oct 2026
  9. Berkemer, S. J. & McGlynn, S. E. (2020). A new analysis of archaea–bacteria domain separation: variable phylogenetic distance and the tempo of early evolution. Molecular Biology and Evolution 37(8), 2332–2340. 10.1093/molbev/msaa089 · Checked 15 Oct 2026
  10. Krupovic, M., Dolja, V. V. & Koonin, E. V. (2020). The LUCA and its complex virome. Nature Reviews Microbiology 18(11), 661–670. 10.1038/s41579-020-0408-x · Checked 15 Oct 2026
  11. Crapitto, A. J., Campbell, A., Harris, A. J. & Goldman, A. D. (2022). A consensus view of the proteome of the last universal common ancestor. Ecology and Evolution 12(6), e8930. 10.1002/ece3.8930 · Checked 15 Oct 2026
  12. Mahendrarajah, T. A., Moody, E. R. R., Schrempf, D., Szánthó, L. L., Dombrowski, N., Davín, A. A. et al. (2023). ATP synthase evolution on a cross-braced dated tree of life. Nature Communications 14, 7456. 10.1038/s41467-023-42924-w · Checked 15 Oct 2026
  13. Delaye, L. (2024). The unfinished reconstructed nature of the last universal common ancestor. Journal of Molecular Evolution 92(5), 584–592. 10.1007/s00239-024-10187-8 · Checked 15 Oct 2026
  14. Moody, E. R. R., Álvarez-Carretero, S., Mahendrarajah, T. A., Clark, J. W., Betts, H. C., Dombrowski, N. et al. (2024). The nature of the last universal common ancestor and its impact on the early Earth system. Nature Ecology & Evolution 8(9), 1654–1666. 10.1038/s41559-024-02461-1 · Checked 15 Oct 2026
  15. Goldman, A. D. & Becerra, A. (2024). A new view of the last universal common ancestor. Journal of Molecular Evolution 92(5), 659–661. 10.1007/s00239-024-10193-w · Checked 15 Oct 2026

Page reviewed 15 Oct 2026

Source

Open, as listed by Wikipedia · Unsolved problems in biology. Checked 08 Oct 2026.

The status is the source’s. Papex does not decide whether a problem is solved.

Records on this problem

No public record addresses this problem yet.

More in Life sciences