Lipid divide
Why do archaea build their cell membranes from lipids of the opposite handedness to those of bacteria and eukaryotes?
Summary
Why it is open
All cell membranes are built from glycerol phosphate phospholipids [5]. In archaea these are generally isoprenoid chains joined by ether bonds to sn-glycerol-1-phosphate (G1P); in bacteria and eukaryotes they are fatty acids joined by ester bonds to sn-glycerol-3-phosphate (G3P) [2]. G1P and G3P are mirror-image forms (enantiomers), so the two kinds of lipid have backbones of opposite stereochemistry [7], and the dehydrogenase enzymes that make G1P and G3P are not homologous [2]. This split is the lipid divide. Philipp Rieche and colleagues write: "Despite decades of research, the lipid membrane composition of LUCA, as well as the cause of the lipid divide, remain unsolved" [14]. LUCA, the last universal common ancestor, has its own page: Last universal common ancestor.
The problem is several questions: how the divide arose from the common ancestor; why, if the two lipid types can form stable mixed membranes, most bacteria and all eukaryotes have only bacterial-type lipids and archaea only archaeal ones [10]; and why eukaryotes, thought to have evolved from archaea, have bacterial-type membranes, which would have required "a radical change in membrane composition" [7].
No body sets a bar for a resolution, and the divide itself is less sharp than the textbook picture: Rieche and colleagues find it "not as strict as previously thought" [14]. The deepest evidence lies in the histories of the genes that make the lipids, but these genes are ancient and "their evolutionary histories are poorly resolved" [9]. Experiments face their own limit: testing a hypothesis about early membranes in living cells is "inherently difficult", Antonella Caforio and colleagues note, "as the conditions of early evolution would need to be replicated" [8].
What is known
Gene histories (2004–2019). Juli Peretó, Purificación López-García and David Moreira found that the G1P and G3P dehydrogenases belong to two universally distributed superfamilies, suggesting that both existed in the common ancestor, and that archaea carry homologues of bacterial fatty acid genes; they concluded that the ancestor's lipid synthesis was "enzymatic but probably non-stereospecific" [2]. Jonathan Lombard and colleagues reviewed phylogenomic studies suggesting that the ancestor "did possess complex phospholipid membranes" [5]. Shin-ichi Yokobori and colleagues, from trees of the G1P and G3P dehydrogenases and glycerol kinase, proposed instead that the ancestor probably had a G3P-lipid membrane and that the archaeal lineage later acquired G1P dehydrogenase and replaced it [6]. Gareth Coleman, Richard Pancost and Tom Williams rooted the gene trees without outgroups and found extensive transfer of these genes between the domains, mainly from archaea to bacteria; their trees suggest that the capacity to make archaeal-type phospholipids was already present in LUCA [9].
Mixed membranes in the laboratory (2011–2018). Haruo Shimada and Akihiko Yamagishi made liposomes (lipid vesicles) from bacterial G3P and archaeal G1P polar lipids; the hybrids were all stable compared with homochiral liposomes of two common synthetic phospholipids, but stability "drastically changed with increasing carbon chain length", and they suggested that chain length, rather than chirality, is what matters [4]. Caforio and colleagues engineered the bacterium Escherichia coli to make archaeal ether lipids, reaching 20–30% of its membrane lipids [8]. The archaeal lipids had the G1P configuration even without an archaeal G1P dehydrogenase, implying that E. coli has an unknown way of making G1P [8]. After a lag of about 16 hours the cells grew at rates comparable with the parent strain, although strong induction of the pathway caused severe stress, and they tolerated heat, freezing and butanol slightly better [8]. The authors concluded that their findings "contrast the hypothesis of the thermodynamic instability of mixed membranes" [8]. This is a single engineered strain of a modern bacterium.
Mixed lipids in nature (2021–2026). Laura Villanueva and colleagues reported that bacteria of the Fibrobacteres–Chlorobi–Bacteroidetes (FCB) group encode a putative archaeal pathway for ether-bound isoprenoid lipids alongside the bacterial pathway, that key genes are expressed in the environment, and that expressing them in E. coli produced a mixed membrane [11]. Rieche and colleagues summarise reports of both glycerol phosphate enantiomers in the phospholipids of various Gram-positive bacteria, in the phosphatidylglycerol of four species in G3P to G1P ratios from 4:1 to 3:1; they showed that the bacterial enzymes PlsY and PlsC accept both forms, while PlsB is strictly specific for G3P [14].
Permeability (2023–2025). Urszula Łapińska and colleagues compared how 18 metabolites cross single-bilayer vesicles: archaeal-type G1P diether lipids with methyl branches were permeable to a wide range of compounds, including amino acids, sugars and nucleobases, and bacterial-type G3P diester lipids without branches significantly less so, a difference that depended on both the branches and the ether bond [12]. Archaea tend to have fewer families of transporter genes, "consistent with increased membrane permeation" [12]. In 2025 Olivia Goode and colleagues, from the same group, found that both an archaeal-type and a hybrid membrane mimic selected D-ribose and D-deoxyribose, while only the hybrid selected a reduced set of L-amino acids [13]. Why life uses one handedness of sugars and amino acids is the origin of homochirality.
Eukaryotes (2017–2026). In 2017 Villanueva, Stefan Schouten and Jaap Sinninghe Damsté found that Lokiarchaeota, described as descendants of the archaeal ancestor of eukaryotes, and marine euryarchaeota of groups II and III lack the gene for G1P synthesis, but could make G3P-based "chimeric lipids" with ether-bound isoprenoid chains or an ester-bound fatty acid [7]. These archaea "may reflect the 'archaea-to-eukaryote' membrane transition stage", they wrote [7]. Rieche and colleagues note that Asgard archaeal genomes carry the bacterial genes plsY and plsC but not plsB, and that initial lipid analysis of the only cultured Lokiarchaeota found only archaeal phospholipids, with a method that "might have overlooked bacterial ester-bonded phospholipids" [14]. The wider question is the origin of eukaryotes.
Main approaches
Segregation of a mixed ancestral membrane. Günter Wächtershäuser proposed pre-cells whose membrane held a racemic mixture (equal amounts of both mirror forms) of chiral lipids, which kept separating into domains enriched in one form; bacteria and archaea diverged as each evolved lipid synthesis selective for one form, and eukaryotes arose by symbiosis between bacteria and pre-cells dominated by bacteria-type lipids [1]. Yosuke Koga added that the hydrocarbon chains also drove the separation, G1P isoprenoid ethers segregating more easily from G3P fatty acid esters [3]. Strongest evidence: both dehydrogenase superfamilies are universally distributed [2], and mixed membranes exist in bacteria today; Villanueva and colleagues read the FCB bacteria as evidence of their "stability over a long period in evolutionary history" [11]. Main difficulty: the instability that would drive separation has not been found, since hybrid liposomes were stable [4] and the mixed-membrane E. coli was viable [8]; and Víctor Sojo argues that a LUCA with both lipid types "would imply a rather elaborate evolutionary scenario" [10].
One or no stereospecific pathway in the ancestor. The ancestor made lipids without strict stereospecificity [2], or made G3P lipids that archaea later replaced [6]. Strongest evidence: archaea carry homologues of bacterial fatty acid genes [2], several archaeal groups have enzymes that make G3P [14], and the bacterial enzyme PlsY "must have been a rather unselective enzyme for glycerol-phosphate already early in evolution" [14]. Main difficulty: the gene histories are ancient and poorly resolved [9], and the scenarios built on them differ: Yokobori and colleagues infer a G3P-lipid ancestor [6], whereas Coleman and colleagues place the capacity for archaeal-type phospholipids in LUCA [9].
Independent origins. LUCA may have had geochemically made lipids, with bacterial and archaeal phospholipids invented independently in the two lineages [14]; a bioenergetic version of this idea is discussed on the LUCA page. Strongest evidence: the dehydrogenases that make G1P and G3P are not homologous; Peretó and colleagues note that various models take them to have originated as the two domains separated, although they themselves argue for a non-stereospecific ancestor [2]. Main difficulty: phylogenomic studies suggest that the ancestor already had complex phospholipid membranes [5, 9].
Proteins and function maintain the divide. Sojo proposes that, since the two lipid types can mix, the divide may be kept by membrane proteins and their close adaptation to the lipids, and that with widespread gene transfer it "continues to occur actively to this day" [10]. Strongest evidence: the two lipid types differ in permeability, and archaea carry fewer transporter families [12]. Main difficulty: the engineered E. coli, whose membrane proteins evolved in ester membranes, grew at rates comparable with the parent strain [8], and bacteria with mixed lipids occur in nature [11, 14].
Open sub-questions
- G1P in E. coli. Which native enzyme makes the glycerol-1-phosphate in the engineered strain: glycerol kinase, or dihydroxyacetone kinase with a glycerol phosphate dehydrogenase, as Caforio and colleagues suggest [8]?
- Proteins in hybrid membranes. Do E. coli membrane transporters lose activity as the archaeal lipid fraction rises, as a protein-maintained divide predicts [8, 10]?
- Ester lipids in Asgard archaea. Does lipid analysis designed to detect ester bonds find bacterial-type phospholipids in cultured Asgard archaea that carry plsY and plsC [7, 14]?
- Chain length and separation. At which chain lengths, if any, do G1P ether and G3P ester lipids separate into domains in hybrid membranes, as the segregation hypotheses require [1, 3, 4]?
- Backbone or tails. Does the L-amino acid selection of the hybrid membrane depend on backbone chirality, or on the methyl branches and ether bonds that set permeability [12, 13]?
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
- Wächtershäuser, G. (2003). From pre-cells to Eukarya – a tale of two lipids. Molecular Microbiology 47(1), 13–22 (published online 2002). 10.1046/j.1365-2958.2003.03267.x · Checked 16 Oct 2026
- Peretó, J., López-García, P. & Moreira, D. (2004). Ancestral lipid biosynthesis and early membrane evolution. Trends in Biochemical Sciences 29(9), 469–477. 10.1016/j.tibs.2004.07.002 · Checked 16 Oct 2026
- Koga, Y. (2011). Early evolution of membrane lipids: how did the lipid divide occur? Journal of Molecular Evolution 72(3), 274–282. 10.1007/s00239-011-9428-5 · Checked 16 Oct 2026
- Shimada, H. & Yamagishi, A. (2011). Stability of heterochiral hybrid membrane made of bacterial sn-G3P lipids and archaeal sn-G1P lipids. Biochemistry 50(19), 4114–4120. 10.1021/bi200172d · Checked 16 Oct 2026
- 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 16 Oct 2026
- Yokobori, S., Nakajima, Y., Akanuma, S. & Yamagishi, A. (2016). Birth of archaeal cells: molecular phylogenetic analyses of G1P dehydrogenase, G3P dehydrogenases, and glycerol kinase suggest derived features of archaeal membranes having G1P polar lipids. Archaea 2016, 1802675. 10.1155/2016/1802675 · Checked 16 Oct 2026
- Villanueva, L., Schouten, S. & Sinninghe Damsté, J. S. (2017). Phylogenomic analysis of lipid biosynthetic genes of Archaea shed light on the 'lipid divide'. Environmental Microbiology 19(1), 54–69 (published online 2016). 10.1111/1462-2920.13361 · Checked 16 Oct 2026
- Caforio, A., Siliakus, M. F., Exterkate, M., Jain, S., Jumde, V. R., Andringa, R. L. H. et al. (2018). Converting Escherichia coli into an archaebacterium with a hybrid heterochiral membrane. Proceedings of the National Academy of Sciences 115(14), 3704–3709. 10.1073/pnas.1721604115 · Checked 16 Oct 2026
- Coleman, G. A., Pancost, R. D. & Williams, T. A. (2019). Investigating the origins of membrane phospholipid biosynthesis genes using outgroup-free rooting. Genome Biology and Evolution 11(3), 883–898. 10.1093/gbe/evz034 · Checked 16 Oct 2026
- Sojo, V. (2019). Why the lipid divide? Membrane proteins as drivers of the split between the lipids of the three domains of life. BioEssays 41(5), e1800251. 10.1002/bies.201800251 · Checked 16 Oct 2026
- Villanueva, L., von Meijenfeldt, F. A. B., Westbye, A. B., Yadav, S., Hopmans, E. C., Dutilh, B. E. & Sinninghe Damsté, J. S. (2021). Bridging the membrane lipid divide: bacteria of the FCB group superphylum have the potential to synthesize archaeal ether lipids. The ISME Journal 15(1), 168–182 (published online 2020). 10.1038/s41396-020-00772-2 · Checked 16 Oct 2026
- Łapińska, U., Glover, G., Kahveci, Z., Irwin, N. A. T., Milner, D. S., Tourte, M. et al. (2023). Systematic comparison of unilamellar vesicles reveals that archaeal core lipid membranes are more permeable than bacterial membranes. PLOS Biology 21(4), e3002048. 10.1371/journal.pbio.3002048 · Checked 16 Oct 2026
- Goode, O., Łapińska, U., Morimoto, J., Glover, G., Milner, D. S., Santoro, A. E. et al. (2025). Permeability selection of biologically relevant membranes matches the stereochemistry of life on Earth. PLOS Biology 23(5), e3003155. 10.1371/journal.pbio.3003155 · Checked 16 Oct 2026
- Rieche, P., Hoekzema, M., Gan, S., Berrelkamp-Lahpor, G. A., Ezissi, A. W., Exterkate, M. & Driessen, A. J. M. (2026). Bacterial lipid synthesizing enzymes PlsY and PlsC utilize both stereo-forms of glycerol-phosphate. EMBO Reports 27(14), 3928–3943. 10.1038/s44319-026-00827-z · Checked 16 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.