Prebiotic gas flow environment enables isothermal nucleic acid replication

Nucleic acid replication is a central process at the origin of life. On early Earth, replication is challenged by the dilution of molecular building blocks and the difficulty of separating daughter from parent strands, a necessity for exponential replication. While thermal gradient systems have been shown to address these problems, elevated temperatures lead to degradation. Also, compared to constant temperature environments, such systems are rare. The isothermal system studied here models an abundant geological environment of the prebiotic Earth, in which water is continuously evaporated at the point of contact with the gas flows, inducing up-concentration and circular flow patterns at the gas-water interface through momentum transfer. We show experimentally that this setting drives a 30-fold accumulation of nucleic acids and their periodic separation by a 3-fold reduction in salt and product concentration. Fluid dynamic simulations agree with observations from tracking fluorescent beads. In this isothermal system, we were able to drive exponential DNA replication with Taq polymerase. The results provide a model for a ubiquitous non-equilibrium system to host early Darwinian molecular evolution at constant temperature.


Introduction
The emergence of life on Earth is still an unsolved puzzle to contemporary research.It is estimated that this event dates back approximately 3.7 -4.5 billion years, with fossil carbon isotope signatures being the oldest evidence for life around 3.7 billion years ago [1,2].In order to reconstruct how early molecular life began before this time, it is crucial to identify and understand plausible geological environments, which support early prebiotic reaction networks that could have lead to the life we know today [3].
The common theory is that the Darwinian evolution of informational polymers was at the core of the origin of life [3].Among these, nucleic acids, like RNA, stand out for their capability to both store genetic information and catalyze their own replication through transient formation of double-stranded helices [4].These abilities allow them to mutate and evolve, enabling them to adapt to diverse environments and eventually encode, build and utilize proteins as the catalysts used in modern life.Dilution, however, poses a significant obstacle, since such prebiotic reactions require sufficiently high concentrations of their reagents to work [5].Large reservoirs, such as the ocean, cannot compensate for diffusion, because they lack local sources of energy to drive reaction pathways out of equilibrium [6].The resulting homogeneity renders these environments unlikely to have harbored early molecular life [7].
However, the accumulation of salts and molecules comes * dieter.braun@lmu.de at a cost.Single-stranded nucleic acids replicate into double-42 stranded forms.These strands must separate again to complete 43 a full replication cycle.But strand separation becomes increas-44 ingly difficult after accumulation, because the melting temper-45 ature of oligonucleotides is strongly dependent on the local salt 46 concentration [18].Despite high Mg 2+ concentrations being 47 required for replication and catalytic activity [19], they can 48 elevate the melting temperature of nucleic acid duplex struc-49 tures to levels surpassing even the boiling point of water [20].

50
Oligonucleotides readily hydrolyze into nucleotide fragments 51 under these conditions, rendering high temperature spikes as 52 a primary strand separation mechanism more detrimental than beneficial [21].

64
The above scenarios require temperature gradients or thermal 65 66 cycling.This creates degradation stress for nucleic acids and 67 limits the scenarios to geological settings with a thermal gra-68 dient.Here, we investigated a simple and ubiquitous scenario 69 in which a water flux through a rock pore was dried by a gas 70 flux at constant temperature (Fig. 1).This can be found in 71 the vicinity of underwater degassing events, where gases per-   1. Replication at the gas-water interface.We considered a geological scenario in which water, containing biomolecules, is evaporated by a gas flow at the scale of millimeters.In volcanic porous rock, many of such settings can be imagined.The gas flow induces convective water currents and causes it to evaporate.Dissolved nucleic acids and salts accumulate at the gas-water interface due to the interfacial currents, even if the influx from below is pure water.Through the induced vortex, nucleic acids pass through different concentrations of salt, promoting strand separation and allowing them to replicate exponentially.Our experiments replicate this environment on the microscale, subjecting a defined sample volume to a continuous influx of pure water with an airflux brushing across.
We created an experimental model of such an evaporation pore, shown in Fig. 1, and studied how combined gas and water 77 fluxes can lead to early replication of nucleic acids.We first 78 analyzed accumulation flow speeds at the interface in Fig. 2, 79 then monitored cyclic strand separation dynamics in Fig. 3, and of an influx of pure water, simulating a continuous dilution scenario.

98
The micro scale gas-water evaporation interface consisted of a 1.5 mm wide and 250 µm thick channel that carried an upward   To confirm this, we simulated the accumulation of Mg 2+ ions in the chamber (Suppl.VII), since divalent ions have a large 188 effect on the melting temperature of nucleic acids [18].We    water) and mixed together to a final concentration of 5 µM in buffer (10 mM TRIS, 50 µM MgCl 2 , 3.9 mM NaCl, pH7).To promote annealing of the two complementary strands, the solution was heated and slowly cooled from 80°C to 4°C (ramp rate of -1°C per 5 s) in a standard thermocycler (Bio-Rad CFX96 Real-Time System) prior to each experiment.
Polymerase chain reaction (PCR) was performed using an AllTaq PCR Core Kit (QUIAGEN).Samples were mixed with 0.5 X AllTaq PCR Buffer, 5 nM template strand, 0.25 µM primers, 200 µM of each dNTP, 2 X SYBR Green I and AllTaq polymerase at 2.5 U/reaction.The reaction in the thermocycler was performed using a temperature protocol of 95°C for 2 minutes for heat activation of the enzyme, then annealing the primers to 52°C for 10 seconds, then 68°C for 10 seconds, and finally 10 seconds at 95°C.This cycle was repeated 40 times (See Suppl.Fig. IX.2b)).The reaction in the chamber was performed with 10 µl of the above mixture at 68°C.The solution was also heat activated at 95°C for 2 min followed by an annealing step to 52°C before loading into the chamber.The DNA sequences for the reaction were as follows Template (5'-3')-51bp DNA: TTAGCAGAGCGAGGTATGTAG-GCGGGACGCTCAGTGGAACGAAAACTCACG, Reverse primer (5'-3')-30bp DNA: AAAAACGTGAGTTTTCGTTCCACTGAGCGT, forward primer (5'-3')-30bp DNA: AAAAATTAGCAGAGCGAGGTATGTAG-GCGG.

335
For PAGE and gel imaging, a 15% denaturing (50% urea) polyacrylamide gel with an acrylamide:bis ratio of 29:1 was solidified with TEMED (tetram-336 ethylethylenediamine) and ammonium persulfate. 2 µl of sample was mixed 337 with 7 µl of 2X loading buffer (Orange G, formamide, EDTA), of which 5 µl 338 were loaded onto the gel.Staining was performed with 2X SYBR Gold in 1X 339 TBE buffer for 5 minutes and the gel was imaged using the ChemiDOC MP imaging station (Bio-Rad).340

53
Therefore, other mechanisms are required at the origin of life to separate nucleic acid strands with minimal thermal 54 stress, and at best combined with an environment where sup-55 plied biomolecules are accumulated from the environment and 56 trapped for long periods of time.Examples have used pH os-57 cillations to drive nucleic acid strand separation, which can be 58 caused either by differential thermophoresis of ionic species 59 or by periodic freeze-thaw cycles [22-24].Also, dew droplet 60 cycles in a rock pore subjected to a temperature gradient can pe-61 riodically melt strands by transiently lowering the salt concen-62 tration [25, 26].Heated gas-water interfaces were also shown 63 to promote many prebiotic synthesis reactions [14, 27, 28].

72
colate through rocks to reach the surface, or in porous rocks 73 at the surface exposed to atmospheric winds [29, 30].Such 74 a setting would be very common on volcanic islands on early 75 Earth which also offered the necessary dry conditions for RNA synthesis[31].

Figure
Figure1.Replication at the gas-water interface.We considered a geological scenario in which water, containing biomolecules, is evaporated by a gas flow at the scale of millimeters.In volcanic porous rock, many of such settings can be imagined.The gas flow induces convective water currents and causes it to evaporate.Dissolved nucleic acids and salts accumulate at the gas-water interface due to the interfacial currents, even if the influx from below is pure water.Through the induced vortex, nucleic acids pass through different concentrations of salt, promoting strand separation and allowing them to replicate exponentially.Our experiments replicate this environment on the microscale, subjecting a defined sample volume to a continuous influx of pure water with an airflux brushing across.

99Figure 2 .
Figure 2. Flow and accumulation dynamics.(a) Imaging of fluorescent beads (0.5 µm) reveals a flow vortex right below the air-water interface, induced by the air flux across the interface (left panel).The bead movements were traced (middle panel) and the measured velocities were confirmed by a detailed finite element simulation (right panel).The color scale is equal for both simulation and experiment and the scale bar = 500 µm.(b) The accumulation of fluorescently labeled 63mer DNA was imaged and confirmed our understanding of the environment based on a diffusion model.Concentration reaches up to 30 times relative to the start c 0 .The accumulation profile of the experiment (middle panel) and simulation (right panel) match well, showcased by overlaying the simulated flowlines.Blue colorscale represents DNA accumulation for experiment and simulation, while grey color scale shows the relative vapor concentration in the simulation.Arrows (right panel) proportionally show the evaporation speed along the interface.(c) The simulated and experimentally measured distribution of flow velocities of dissolved beads plotted in a histogram, showing a similar profile.Color scale is equal to (a).(d) The maximum relative concentration of DNA increased within an hour to ≈ 30 X the initial concentration, with values following the simulation.

189Figure 3 .
Figure 3. Strand separation by salt cycling.Fluorescence resonance energy transfer measurements revealed cycles of strand separation.(a) Micrographs of 24bp DNA FRET pair in the chamber at 45°C. 1 µl sample (5 µM DNA, 10 mM TRIS pH7, 50 µM MgCl2, 3.9 mM NaCl) was subjected to a 3 nl/s diluting upflow of pure water and a gas flow of 230 ml/min across.The induced vortex, shown by the simulated flow lines (left panel), overlays with regions of high FRET indicative of double-stranded DNA.The vortex flow was expected to enable replication reactions by (1+2) strand replication in the high salt region and (3) strand separation of template and replicate in the low salt region.(b) FRET signals confirmed strand separation in low salt regions and strand annealing in high salt regions in (a).After about 10 minutes, DNA and salt accumulated at the interface forming stable and clearly separated regions of low -where the influx from below reaches the interface -and high -located at the vortex -FRET signals.(c) Comsol simulation of Mg 2+ ions (D = 705  2 / in the chamber agreed with the FRET signal and showed up to 9-fold salt accumulation at the interface.The path of a 61mer DNA molecule from a random walk model is shown by the green lines and the white flowlines are taken from the simulation.(d) Concentrations along the DNA molecule path in (c) show oscillations relative to the initial concentration of up to 3-fold for Mg 2+ and 4-fold for 61mer DNA.This could enable replication cycles, as the vortex provides high salt concentrations for replication, while drops in salt and template concentrations regularly trigger strand separation.

199VII. 2 )
. When plotting the simulated steady-state concentration 200 of other dissolved -complementary -61mer DNA molecules 201 along its path, we observed even stronger oscillations of up to 202 4X the initial concentration.Together with significant drops 203 in Mg 2+ concentration, this suggests the possibility of exponential replication by strand separation cycles.204 Isothermal Replication with PCR.We saw that nucleic acids and salts accumulated near the interface, but far from the in-205 terface, in the bulk below, the concentrations remained van-206 ishingly low due to the diluting inflow of pure water.The 207 air flux induced an accumulation pattern of vortices in which 208 molecules were trapped.The salt and DNA concentration 209 changed cyclically, resulting in periodic strand separation of 210 nucleic acids.Motivated by the above results, we used a model 211 system to test whether nucleic acid replication could actually be implemented in this environment.212 We chose to use Taq DNA Polymerase because it does not have a protein-based strand separating mechanism.Starting 213 with a 51mer template and two 30mer primer strands, each with 214 a 5'-AAAAA overhang for detection, the reaction is expected 215 to form a 61mer replicate (Suppl.IX), the same length as 216 the DNA used in the random walk model in Fig. 3(c)&(d).

217
In contrast to standard PCR, which uses thermal cycling to 218 separate the strands, we operated the experiment at isothermal 219 conditions (68°C) and used 10 µl of the reaction mix (0.25 µM 220 primers, 5 nM template, 200 µM dNTPs, 0.5 X PCR buffer, 221 2.5 U Taq polymerase, 2 X SYBR Green I).This reaction 222 mixture was then exposed to a constant pure water influx of 223 5 nl/s towards the gas-water interface, matching the rate of evaporation at the interface.224 Through the oscillations in salts and DNA observed along the random walk, we expected the 61mer product strand to be 225 able to separate from its respective template strand, enabling 226 exponential replication.The progress was monitored using the 227 intercalating dye SYBR Green I, which binds preferentially to 228 double-stranded DNA [32].Fig.4(a) shows fluorescence mi-229 crographs of the reaction in the chamber.Initially, minimal 230 fluorescence is seen.This indicates that the replicated templates are below the detection limit of SYBR Green.231 The SYBR Green fluorescence increased after two hours, recording the increase of replicated DNA forming duplex 232 structures.In an identical setting when the gas-and water 233 flux were switched off, no fluorescence increase was found.234 Replication was confirmed under flux with the 61mer prod-235 uct being visible in gel electrophoresis with depleted primers 236 (Fig.4(b)).The fluorescence signal over time is shown in Fig.4(c), recorded from a rectangular region of interest.237 Fluorescence variations are caused by fluctuations in the position of the gas-water interface in addition to air bubbles 238 caused by degassing of the liquid at this temperature (Sup-239 plementary Movie 4&5).With both gas flow and water influx 240 turned off, no product band was found.We verified the replica-241 tion reaction by repeating the experiments without the addition 242 of the template, primer or DNA in the chamber as well as in 243 a test tube (Suppl.Fig. IX.2).As expected, no product was

Figure 4 .
Figure 4. Replication.(a) Fluorescence micrographs of the PCR reaction in the chamber.At isothermal 68°C, 10 µl of reaction sample was subjected to a constant 5 nl/s pure water flow towards the interface where a 250ml/min gas flowed perpendicularly.The initial state on the left shows the background fluorescence.Fluorescence increased under flux (middle, after 3:20h), while without flux the fluorescence signal remained minimal (right).The reaction sample consisted of 0.25 µM primers, 5 nM template, 200 µM dNTPs, 0.5 X PCR buffer, 2.5 U Taq polymerase, 2 X SYBR Green I. Scale bar is 250 µm.(b) 15% Polyacrylamide Gel Electrophoresis of the reactions and neg.controls.After 4 hours in the reaction chamber with air-and waterflux ON, the 61mer product was formed under primer consumption (2), unlike in the equivalent experiment with the fluxes turned OFF (3).At the beginning of the experiment (1) or in the absence of template (4), no replicated DNA was detected.The reaction mixture was tested by thermal cycling in a test tube (5-7).As expected, replicated DNA was detected only with the addition of template: (7) shows the sample after 11 replication cycles.The sample was also incubated for 4 hours at the chamber temperature (68°C) yielding no product (6).Primer band intensity variations are caused by material loss during extraction from the microfluidic chamber.(c) SYBR Green I fluorescence increased when gas and water flow were turned on, but remained at background levels without flow.Fluorescence was averaged over time from the green and red regions of interest shown in (a).SYBR Green I fluorescence indicates replication, as formed products are able to hybridize.