Thermally Driven Membrane Phase Transitions Enable Content Reshuffling in Primitive Cells

Self-assembling single-chain amphiphiles available in the prebiotic environment likely played a fundamental role in the advent of primitive cell cycles. However, the instability of prebiotic fatty acid-based membranes to temperature and pH seems to suggest that primitive cells could only host prebiotically relevant processes in a narrow range of nonfluctuating environmental conditions. Here we propose that membrane phase transitions, driven by environmental fluctuations, enabled the generation of daughter protocells with reshuffled content. A reversible membrane-to-oil phase transition accounts for the dissolution of fatty acid-based vesicles at high temperatures and the concomitant release of protocellular content. At low temperatures, fatty acid bilayers reassemble and encapsulate reshuffled material in a new cohort of protocells. Notably, we find that our disassembly/reassembly cycle drives the emergence of functional RNA-containing primitive cells from parent nonfunctional compartments. Thus, by exploiting the intrinsic instability of prebiotic fatty acid vesicles, our results point at an environmentally driven tunable prebiotic process, which supports the release and reshuffling of oligonucleotides and membrane components, potentially leading to a new generation of protocells with superior traits. In the absence of protocellular transport machinery, the environmentally driven disassembly/assembly cycle proposed herein would have plausibly supported protocellular content reshuffling transmitted to primitive cell progeny, hinting at a potential mechanism important to initiate Darwinian evolution of early life forms.

Phase transition temperature extrapolation. Phase transition temperatures were calculated as previously reported. 8 Specifically, the absorbance versus temperature profiles were converted to normalised absorbance (A n ) versus temperature (T) curves. Assuming a two-state model, the lower (BLT) and upper (BUT) baselines (corresponding to the vesicle phase and the droplet phase, respectively) were calculated by performing a linear regression on the vesicle and droplet components of the normalised melting curves. The equations of each baseline were used to transform the normalised absorbance versus temperature plots into a "vesicle fraction" (fv) versus temperature (T) plots using the following equation: Finally, the phase transition temperature was extracted by reading the temperature at fv = 0.5. An average of three temperature values were taken as the final phase transition temperature.
Thermal cycling. For turbidity and fluorescence analysis, fatty acid vesicles were heated up to 95°C (at a rate of 0.1°C·s -1 ) and kept at 95°C for 15 minutes, then cooled down to 25°C at the same rate, while recording absorbance values at 420 nm or fluorescence intensity at 425 nm and 495 nm. The same thermal settings were used for hot stage epifluorescence microscopy. All experiments were performed in a sealed quartz cuvette. For confocal microscopy and cryo-EM, samples were visualised before and after thermal cycles (in the latter case, after waiting 1 hour for re-equilibration of the sample at room temperature).
Leakage studies. Vesicles containing 1 mM FITC-dextran were prepared, extruded and purified as previously described. The purified vesicles were incubated at a constant temperature (20-95°C) for 1 h. Vesicles were then purified, and fractions were analysed by fluorescence spectrophotometry (λexc = 454 nm). The retention percentage of FITC-dextran was reported as a function of temperature.

Reconstitution of a split Broccoli aptamer inside vesicles.
A minimal version of the Broccoli aptamer 9 was split into two oligonucleotides (a 23-nt oligonucleotide (5′-GCGGAGACGGUCGGGUCCAGAUA-3′) and a 28-nt oligonucleotide (5′-UAUCUGUCGAGUAGAGUGUGGGCUCCGC-3′)). Two different batches of vesicles, made of myristoleic acid, were prepared by direct dispersion in an aqueous buffered solution, containing 100 μM of each oligonucleotide. Vesicles were extruded and purified as previously described. Purified vesicles, containing different oligonucleotides, were mixed either before or after heat exposure and purified again, all fractions being collected in 96 well plates. A solution containing DFHBI (stock solution: 20 mg/mL in DMSO) and KCl was prepared and added to each well (final concentrations:          (7, 7.5, 8 and 8.5). Absorbance was monitored at 420 nm. The increase in turbidity is observed at higher temperatures for higher initial pH values, as the buffer reaches a pH value lower than 6.5 (when fatty acid vesicles become unstable) at higher temperatures. No destabilisation can be observed when myristoleic acid vesicles are prepared in 200 mM Tris-HCl buffer at pH 8.5, as the buffer does not reach pH values low enough to destabilise vesicles. b) Turbidity profile for 100 nm-radius vesicles made of 50 mM myristoleic acid in 200 mM Tris-HCl, at pH = 9. Absorbance was monitored at 420 nm. When myristoleic acid is hydrated with 200 mM Tris-HCl buffer, pH 9, it self-assembles into myristoleate micelles 11,12the increase observed in the turbidity profile is due to the micelle-to-vesicle phase transition. c) Temperature vs pH graph with vesicle-to-oil droplet phase transition values extrapolated from the curves in a). d) Turbidity profile for 100 nm-radius vesicles made of 50 mM decanoic acid:decanol (2:1 ratio) in 200 mM Tris-HCl, at different initial pH values (7, 7.5, 8, 8.5, 9). Absorbance was monitored at 420 nm. The increase in turbidity follows the same trend observed for myristoleic acid vesicles. As decanoic acid-based vesicles are more stable at higher pH values in the presence of decanol, vesicles are formed even at an initial pH value of 9. e) Temperature vs pH graph with vesicle-to-oil droplet phase transition values extrapolated from the curves in d). n = 3 independent experiments.

Figure S10 -Effect of additives on vesicle-to-oil droplet phase transition temperatures for fatty acid vesicles. a) Phase transition temperatures observed for vesicles made of 50 mM myristoleic acid in 200 mM Tris-HCl, pH 8 in the presence of 5 mM charged peptides (GlyGly is reported for comparison). Values have been extrapolated from turbidity curves. b) Phase transition temperatures observed for vesicles made of 50 mM myristoleic acid in 200 mM Tris-HCl, pH 8 in the presence of 5 mM hydrophobic peptides. Values have been extrapolated from turbidity curves. c) Phase transition temperatures observed for vesicles made of 50 mM myristoleic acid in 200 mM Tris-HCl, pH 8 in the presence of KCl (different concentrations). Values have been extrapolated from turbidity curves. d) Phase transition temperatures observed for vesicles made of 50 mM myristoleic acid in 200 mM Tris-HCl, pH 8 in the presence of 5 mM nucleotides. Values have been extrapolated from turbidity curves. e) Phase transition temperatures observed for vesicles made of 50 mM decanoic acid:decanol (2:1 ratio) in 200 mM Tris-HCl, pH 8 in the presence of 5 mM additive (peptides). Values have been extrapolated from turbidity curves. The observed trend suggests that an increase in ionic strength, as well as additional electrostatic interactions can dramatically decrease the phase transition temperature for both fatty acid systems. f) Confocal microscopy images taken for 100 nm-radius vesicles made of 50 mM decanoic acid:decanol (2:1 ratio) in 200 mM Tris-HCl, pH 8 in the presence or absence of arginylarginine (ArgArg), after heating them up to 45°C, and then cooling them down to 25°C and re-equilibrating them for 1 h. In the presence of the peptide, the vesicle-to-oil droplet transition occurs at low temperatures (~30°C); in the absence of the peptide, the phase transition occurs at higher temperatures (~60°C). When the samples are heated up to 45°C, only in the presence of the peptide vesicles undergo phase transition to oil droplets and reassemble, upon cooling, in larger multilamellar vesicles. n = 3 independent experiments.
Figure S11 -Hot stage epifluorescence microscopy images collected for decanoic acid-based vesicles. Once the phase transition temperature is reached (~60°C), decanoic acid:decanol (2:1 ratio) vesicles (50 mM) collapse into small oil droplets, which merge while kept at high temperature. Faceted structures can be observed at high temperatures during oil droplets merging. b) Electron cryo-microscopy images collected before and after thermal cycles for decanoic acid:decanol (2:1 ratio) vesicles (50 mM).