Prebiotic Protocell Model Based on Dynamic ... - ACS Publications

Article Cite This: Langmuir 2019, 35, 9593−9610

pubs.acs.org/Langmuir

Prebiotic Protocell Model Based on Dynamic Protein Membranes Accommodating Anabolic Reactions Andreas Schreiber,*,†,‡ Matthias C. Huber,*,†,‡ and Stefan M. Schiller*,†,‡,∥,⊥,# †

Zentrum für Biosystemanalyse (ZBSA), Albert-Ludwigs-Universität Freiburg, 7 Habsburgerstrasse 49, D-79104 Freiburg, Germany Faculty of Biology, University of Freiburg, Schänzlestrasse 1, D-79104 Freiburg, Germany ∥ BIOSS Centre for Biological Signalling Studies, University of Freiburg, Schänzlestrasse 18, D-79104 Freiburg, Germany ⊥ IMTEK Department of Microsystems Engineering, University of Freiburg, Georges-Köhler-Allee 103, D-79110 Freiburg, Germany # Cluster of Excellence livMatS @ FITFreiburg Center for Interactive Materials and Bioinspired Technologies, University of Freiburg, Georges-Köhler-Allee 105, D-79110 Freiburg, Germany

Downloaded via UNIV OF SOUTHERN INDIANA on July 24, 2019 at 07:01:18 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: The nature of the first prebiotic compartments and their possible minimal molecular composition is of great importance in the origin of life scenarios. Current protocell model membranes are proposed to be lipid-based. This paradigm has several shortcomings such as limited membrane stability of monoacyl lipid-based membranes (e.g., fatty acids), missing pathways to synthesize protocell membrane components (e.g., phospholipids) under early earth conditions, and the requirement for different classes of molecules for the formation of compartments and the catalysis of reactions. Amino acids on the other hand are known to arise and persist with remarkable abundance under early earth conditions since the fundamental Miller−Urey experiments. They were also postulated early to form protocellular structures, for example, proteinoid capsules. Here, we present a protocell model constituted by membranes assembled from amphiphilic proteins based on prebiotic amino acids. Self-assembled dynamic protein membrane-based compartments (PMBCs) are impressively stable and compatible with prevalent cellular membrane constituents forming protein-only or protein−lipid hybrid membranes. They can embed processes essential for extant living cells, such as enclosure of molecules, membrane fusion, phase separation, and complex biosynthetic elements from modern cells demonstrating “upward” compatibility. Our findings suggest that prebiotic PMBCs represent a new type of protocell as a possible ancestor of current lipid-based cells. The presented prebiotic PMBC model can be used to design artificial cells, important for the study of structural, catalytic, and evolutionary pathways related to the emergence of life.

■

INTRODUCTION The study of the emergence of life and underlying fundamental principles such as compartmentalization and self-replication attracts great attention since decades. For self-sustaining protocells, self-assembly, compartmentalization, catalytic chemical conversions, storage of information, and replication are essential requirements. Therefore, the prebiotic protocell requires three functional molecular species: informationstoring molecules capable of replication such as RNA, catalysts, for example, ribozymes structurally encoded by that information, and compartment-forming molecules.1 To keep the replicase and template in proximity some form of compartmentalization is essential. In addition, replication on the compartment level is required for the selection and evolution of more efficient compartments and replicases. The question arises which molecular entities are capable of constituting early cell-like phenotypes demonstrating compartmentalization and catalytic function. It would be favorable if one class of molecules could enable more than one function. © 2019 American Chemical Society

Furthermore, the synthetic access and functional assembly of prebiotically available molecular building blocks allowing to constitute protocells are required. Urey and Miller evidenced the accessibility of amino acids successfully in 1953,2 whereas Fox demonstrated capsular compartment formation of proteinoid macromolecules derived from amino acids accessible under prebiotic conditions in the late 1950s.3,4 As we have shown before, amino acids could constitute amphiphilic proteins assembling to dynamic protein membranes and form novel compartments.5 Reducing their synthetic and molecular complexity by utilizing only prebiotic amino acids and shorter peptide length we present and investigate a new class of prebiotic compartments. However, the first reaction compartments and protocells are widely assumed to be surrounded by amphiphilic monoacyl- or diacyl lipid-based membranes,1,6−8 Received: February 14, 2019 Revised: June 24, 2019 Published: June 24, 2019 9593

DOI: 10.1021/acs.langmuir.9b00445 Langmuir 2019, 35, 9593−9610

Article

Langmuir partly because most cellular compartments and cells known today are based on phospholipid membranes. Nevertheless, there are limitations of monoacyl- and diacyl lipid bilayer membranes in protocell models: the stability of vesicles based on monoacyl lipid bilayer systems such as the promising fatty acids protocell model9−12 is compromised as they are stable only in a narrow pH range and low ionic strength.13 Regarding the limitation of the diacyl lipid protocell models, so far there is no plausible way of phospholipid synthesis14−16 described under prebiotic conditions in recent literature. In addition, even though lipid vesicles are prominent models in synthetic protocell research,17−22 the conceptual link between catalytically active entities and compartmentalization strategies based on lipid bilayer membranes is still under debate. Alternative models of prebiotic compartments are based on proteinoid microspheres,23 nanofibers,24 and membrane-free compartments such as peptide−nucleotide microdroplets25 and polyamine−DNA aggregates.26 Here, we propose a protocell model of dynamic protein membrane-based compartments (PMBCs) containing simple pentapeptide repeats constituting amphiphilic peptides solely composed of prebiotic amino acids. Studies of abiotic chemical synthesis in early earth scenarios as well as analysis of comets and meteorites lead to a consensus set of α-amino acids widely accepted as prebiotics.27 Accordingly, pentapeptide repeats consisting of only four different prebiotic amino acids (aa) for each repeating unit can form amphiphilic block domain proteins such as (VPGHG)5(VPGLG)4 or (VPGHG)20(VPGIG)30 assembling into PMBCs in vitro. These extremely stable PMBC protocells withstand temperatures up to 100 °C, pH 2−11, 5 M salt, and at least 8 months of storage at room temperature (RT). They are compatible with membrane constituents of current protocell models such as fatty acids, lipid precursors,12 and contemporary phospholipid membranes. These cell-sized PMBCs can obtain cellular functions by encapsulating small molecules up to intact proteins and complex anabolic reactions. The exchange of vesicular content can be observed by dynamic membrane fusion. Investigated PMBC protocells accomodate anabolic ligation reactions of short DNA fragments (15 bp) to ligation products of up to 1200 bp. In addition and in contrast to the complex phospholipid synthesis, PMBCs embed the synthesis of their own membrane constituents through in vitro transcription and translation and incorporate these amphiphilic protein building blocks into the membrane. Although not representing origin of life situation, this emphasizes the simplicity and abilities of the presented protocell model solely based on amphiphilic components which could self-assemble into functional PMBCs. The emergence of PMBCs, their self-assembly as possible primary prebiotic protocells, and associated catalytic processes are illustrated in Figure 1.

■

Figure 1. PMBCs and their self-assembly as possible prime prebiotic protocells. (a) Gray panel, illustration of the abiotic chain elongation and amphiphilic peptide formation based on prebiotic hydrophobic (violet symbols) and hydrophilic (blue symbols) aa.33 Oligopeptides in the prebiotic era have been shown to be synthesized on different inorganic substrates,36,37 as well as through their catalytic43 or autocatalytic44 activity (a). The respective peptides or (auto-)catalyzed interconnection of hydrophilic and hydrophobic proteins (1a, asterisk) will then lead to amphiphilic protein block-copolymers that are able to self-assemble into PMBCs. Our results indicate two prebiotic PMBC assembly scenarios. One is solely based on swelling of rehydrated amphiphilic peptides in water and a second is based on a mixture of amphiphilic proteins and fatty alcohols (e.g., 5−20% v/v n-butanol) or fatty acids respectively (65 mmol/L) (see Figure 2a,b, right column and Figure S5). Fatty acids and their corresponding alcohols can be synthesized under prebiotic conditions45,46 and were found on meteorites,47,48 making fatty alcohol-triggered PMBC assembly possible. White panel, (left) PMBC as first entity encapsulating other components necessary on the path toward the emergence of living evolvable systems. Once simple prebiotic amphiphilic peptides are self-assembled into PMBCs, they can enclose short peptides, inorganic molecules, RNA-precursors (brown symbols) as information carriers and the primordial RNA replicases or ribozymes, allowing them to evolve into more efficient biological catalysts favoring corresponding PMBCs. The fluorescence image on the right shows artificially assembled PMBCs based on a prebiotic amphiphilic protein encapsulating other PMBCs. The enclosure of different-sized molecules into our PMBC protocell model is shown in detail in Figure 3 (right column), illustrating the PMBC functionality as a potential dynamic reaction chamber. The dynamic nature of PMBC membranes and fusion potential for the exchange of PMBC content is illustrated in Figure 4 (right column). Finally, depicted is the evolvement toward phospholipid membranebased vesicles employed by the cellular system of extant cells. The

EXPERIMENTAL SECTION

Methods and Materials. Protein Expression and Purification. Escherichia coli strains BLR (DE3) and BL21 (DE3) were used for the expression of the desired protein constructs. Bacteria were grown while shaking at 250 rpm in LB medium at 37 °C until they reached an OD600 of 0.7. At this point, protein expression was induced with f.c.1 mM IPTG. After shaking for 6−7 h at 20 °C and 180 rpm, the cells were harvested by centrifugation at 4000g for 30 min. They were resuspended in lysis buffer (50 mM Tris-HCl pH 8.0, 500 mM NaCl, 20 mM imidazole), 1 mM TCEP (f.c.), and 10 μg/mL lysozyme per liter of culture volume were added and the suspension was incubated on ice for 30 min, frozen in liquid nitrogen, and thawed. 9594


Article

Langmuir

polyacrylamide gel electrophoresis (SDS-PAGE) and liquid chromatography−mass spectrometry (LC−MS)/MS were carried out. LC−MS/MS Analysis of Prebiotic Amphiphilic Proteins. Samples for LC−MS/MS analysis have been prepared by trypsin in-gel digestion using a standard protein digestion protocol.28 Extracted tryptic peptides were separated by LC chromatography (Agilent 1200 SL G1312B system) on C18 column AdvanceBio Peptide Mapping, 2.1 × 150 mm, 2.7 μm, LC column, kept at 45 °C. The peptides were separated over a linear gradient from 10 to 45% acetonitrile in 0.1% formic acid at a flow rate of 200 μL/min. The total time for an LC− MS/MS run was 50 min including loading and washing steps. Online MS analysis was performed on an Impact HD UHR time-of-flight mass spectrometer system ion trap (Bruker Daltonics) equipped with an Apollo II ion funnel electrospray source. The voltage applied to the capillary corresponded to 4.5 kV, end plate offset was 500 V, nebulizer of 1.8 bar dry gas 8 L/min, and dry temperature 200 °C. Peptides were analyzed in positive MS ion mode using an enhanced resolution scan in a mass range of 150−2800 m/z, MS−MS/MS cycle time was set to 3 s with 0.5 s for MS, and 0.25 s summation time for MS/MS spectra. Singly charged ions were excluded from fragmentation. MS/ MS spectra were obtained using CID fragmentation, recorded in mass range initiating at 150−2800 m/z. The collision energy was adjusted between 23 and 65 eV as a function of the m/z value. Absolute threshold for MS/MS fragmentation was set to 1755 counts and active exclusion was set to two spectra, which were released again after 1 min or if peak intensity increased threefold. All data were analyzed using Biotools 3.2 Peptide Editor 3.2 (Bruker Daltonics, Bremen) and Mascot 2.5 (Matrix Science, London, UK) search engine. Mascot 2.5 scored peptides for identification based on a search with an initial allowed mass deviation of the precursor ion of up to 15 ppm and the allowed fragment mass deviation was 0.05 Da. The search engine was used for the MS/MS spectra search against the UniProt E.coli database (downloaded 21 Dec, 2015, containing 4305 entries, 116 contaminants and the added target proteins). The contaminants list used was downloaded from ftp.the pgm.org as of Jan 21, 2016. To determine the number of false positive peptide hits, data were

Figure 1. continued principal compatibility of these two distinctive dynamic membrane systems is demonstrated in Figure 5 and presents the plausible transition of a prebiotic PMBC protocell model to current phospholipid membrane-based cells. (b) Illustrated are the different PMBC membrane compositions and encapsulated molecules investigated in this study. Pictograms of the membrane constituents and enclosed molecules or amphiphilic peptides and proteins are denominated in Table 1. Different composition schemes are depicted in six segments surrounding the PMBC: there are three states with protein-only membrane assembly His-H40I30, mEGFP-H40I30, and ComboRhod-K20I30 (or BDP-K40I30) or BDP-H5L4) (10−16 o’clock) and three membrane states containing phospholipids or fatty acids with His-mEGFP-H40I30 or His-H40I30 (16−10 o’clock). Further membrane states are shown in Figure 2c−f,h. Two PMBCs are depicted on the right, illustrating their functionalization through the encapsulation of catalytic reactions as a step toward selfreplicating systems. First, we demonstrate the ability of anabolic biosynthesis processes by performing ligation reactions of small DNA oligonucleotides (15 bp) to large DNA (1200 bp) fragments within the PMBCs described in Figure 6. Finally, the implementation of IVTT into the PMBCs demonstrate the intrinsic ability of the PMBC protocells to synthesize their own structural constituents as basis for a self-replicable entity as demonstrated in Figure 7.

Subsequently, the cells were sonified and centrifuged at 10 000g for 40 min. Finally, the supernatant was loaded onto a Ni-NTA column (Macherey-Nagel), washed, and eluted with buffer containing 50 mM Tris pH 8.0, 500 mM NaCl, 4 M urea, and 500 mM imidazole. If amphiphilic protein still remained within the pellet, another sonification cycle in lysis buffer with 4−6 M urea was performed followed by centrifugation as described before. For purity analysis and identification standard 10% Tris/glycine sodium dodecyl sulfate-

Table 1. Amphiphilic Protein Block-Copolymers for PMBC Assemblya

The first and second columns specify the pictograms and abbreviations of the amphiphilic proteins used in this study. In the third column, the amphiphilic protein blocks are listed. Gray letters mark the domains primarily responsible for purification (His) or visualization (mEGFP, EYFP, mCherry, pAzF) aspects. The blue (hydrophilic) and violet highlighted (hydrophobic) letters indicate the protein domains responsible for the selfassembly property of the amphiphilic proteins. The repeat numbers of the pentameric peptide repetitions are indicated. The fourth column lists the number of amino acids (aa), first of the self-assembling amphiphilic block-domains (BD) and second of the overall protein (total). The fifth column presents the molecular weight (Mw in kDa) of the overall amphiphilic protein. The sixth column specifies the aa components present in the amphiphilic protein domains. The correctness of the aa sequences of all applied “prebiotic” amphiphilic proteins was confirmed by SDS PAGE and/ or LC−MS/MS (see Data S1). Asterisk: PMBCs assembled from these short amphiphilic proteins dissociate within 10−30 min when assembled with n-butanol and after a few hours after assembly with 1-octanol. PMBCs assembled from the longer aa (>91 aa) sequences preserve their vesicular integrity for up to 9 months (Figure S3, Video S4). a

9595


Article

Langmuir searched against the full-length decoy database using the Mascot 2.5 percolator function. The false discovery rate was set below 1% to exclude false-positive hits. Enzyme specificity was set as C-terminal to Arg and Lys, also allowing cleavage at proline bonds and a maximum of two missed cleavages. Deamidations of asparagine and glutamine and methionine oxidation were set as variable modifications. Target modifications [e.g., p-azido-phenylalanine (pAzF)] were calculated with respect to phenylalanine (F) and searched as variable modification. Protocol for the Assembly of PMBCs. PMBC Assembly Protocol Using Fatty Alcohols (n-Butanol and 1-Octanol). The desired amphiphilic proteins composed of prebiotic amino acids (see Table 1) were dissolved at a concentration of 1−50 μM in Milli-Q H2O or buffer (typically 10 mM Tris-HCl pH 8, but may contain up to 5 M salt). Subsequently, this solution was mixed with 5−20% v/v nbutanol using a pipette to yield PMBCs with medium PMBC concentrations. Alternatively, to yield solutions with high PMBC concentrations, the amphiphilic proteins together with 10−20% v/v nbutanol can be mixed five times using a Hamilton syringe equipped with a 0.25 × 25 mm needle. In order to achieve a narrow size distribution, vesicles can be extruded using an Avanti Mini Extruder with 15 passages through a membrane with 1 μm pore size at RT. Standard dialysis was performed for removal of organic solvent. Extruded PMBCs (50 μL) containing 15% v/v n-butanol (1.6 mol/L) were dialyzed against 200 mL of buffer (10 mM Tris pH 8) using a 12−14 kDa molecular weight cut-off (MWCO) cellulose mixed ester membrane (ROTH). After dialysis was performed twice, residual nbutanol concentration should theoretically be below 1 nmol/L, which is far below the amphiphilic protein concentration of 1−50 μM. Alternatively (e.g., Figure 6, ligase reaction), amphiphilic protein solutions as mentioned above can be mixed with 10% v/v 1-octanol using a syringe with gauge 17 needle carrying out 5−10 passages into a 1.5 mL reaction tube to receive PMBCs with high yield. PMBC Assembly Protocol Using Fatty Acids. His-mEGFP-H40I30 [40 μL (14 μM)] in 10 mM Tris-HCL pH 8 and lauric acid [10 μL (66 mg/mL)] in ethanol were mixed five times with a Hamilton syringe equipped with a 0.25 × 25 mm needle. The size of the imaged PMBCs ranged from submicrometer species up to species of multiple micrometers in diameter (Figure S3). PMBC Assembly Protocol Omitting Any Organic Solvent. HismEGFP-H40I30 [50 μL (14 μM)] in 10 mM Tris-HCL pH 8 was added to a 0.25 mL RNA-free polymerase chain reaction (PCR) tube containing small pieces of a broken coverslip. Subsequently, the protein film on the coverslip pieces was dried in a speedvac (Eppendorf concentrator 5301) for 30 min and dried in high vacuum overnight. The addition of 50 μL of 10 mM Tris-HCL pH 8 resulted in swelling of PMBCs within 2−20 min. The size range of the imaged PMBCs ranged from submicrometer species up to species of multiple micrometers in diameter (Video S1). Epifluorescence Imaging and Bright Field Microscopy of PMBCs. Protein-based vesicles in buffer (50 μL) were transferred onto a microscope slide equipped with a spacer to protect the vesicles. The coverslip was positioned on top and fixed with nail polish, and immersion oil was applied. The vesicles were imaged by using appropriate fluorescence filters on a Nikon Eclipse TS100-F fluorescence microscope with DS-Qi1 camera and oil immersion objective CFI Achromat 1.25 × 100. Filter setup F26-513 (AHF Analysentechnik AG) was used for BDP-FL-PEG4-DBCO (BDP)labeled and mEGFP-labeled objects. The triple band filter F66-417 (AHF Analysentechnik AG) was applied for the yellow and red channel fluorescence detection of EYFP, mCherry, and combo rhodamine respectively. Images were processed using ImageJ 1.52c. Multiplate Reader Fluorescence Measurements. After a 5 h period of IVTT (in vitro transcription and translation) for mCherryH40I30, 10 μL of the sample volume for condition II (IVTT inside PMBC) and 10 μL of sample volume for condition III (no IVTT) were diluted into 40 μL of 10 mM Tris-HCL pH 8(final volume of 50 μL). These two sample solutions and pure 10 mM Tris-HCL pH 8 as control were applied into a 96 well plate and red fluorescence was measured at 635 ± 35 nm upon excitation at 560 ± 20 nm with a

Tecan Infinite 200 multiplate reader with the following settings: mode: fluorescence top reading; multiple reads per well (square) 3 × 3; multiple reads per well (Border) 1000 μm; gain 100, number of flashes 25, integration time 20 μs. Transmission Electron Microscopy of PMBCs. After extrusion in 10 mM Tris pH 8 with 10% v/v n-butanol protein membrane-based vesicles were dialyzed against 10 mM Tris-HCL pH 8. Further, they were transferred and incubated on a glow-discharged carbon-coated copper grid for 2 min and negatively stained with 1% uranyl formate for 20 s. Micrographs were taken with a 1024 × 1024-pixel chargecoupled device detector (Proscan CCD HSS 512/1024; Proscan Electronic Systems) using an electron microscope (EM Philips CM 100 TEM). Incorporation of pAzF. Each of the plasmids pET28-(TAG)NMBXL-His-TEVrec-K20I30 and pET28-(TAG)NMBXL-His-TEVrec-K40I30 and pEVOLpAzF (kindly provided by Schultz)29 containing the orthogonal pair of t-RNA/t-RNA synthetase (t-RNARS/AzPheRS for pAzF incorporation) were co-transformed into the E. coli strain BLR (DE3). For expression of the desired protein construct, for example His-pAzF-K40I30, bacteria were grown through shaking at 250 rpm in LB medium containing 34 μg/mL chloramphenicol and 40 μg/mL kanamycin (Kan) at 37 °C until they reached an OD600 of 0.7. At this point, 2 mM (f.c.) pAzF was added. After 10 min of incubation the t-RNA synthetase expression under control of an arabinose promoter was induced with 0.2% arabinose and target protein expression was triggered by adding 1 mM (f.c.) IPTG. Expression was allowed while shaking for 20 h at 25 °C and 180 rpm; then, the cells were harvested by centrifugation at 10 000g for 40 min. Purification was conducted with dimmed light on ice and no reduction agent was applied during cell lysis to prevent reduction of azide to amine. Labeling of Amphiphilic Proteins with Fluorescent Dyes via SPAAC. Strain-promoted azide alkyne cycloaddition (SPAAC) was conducted with 2−3 equiv BDP-FL-PEG4-DBCO (purchased from Jena Biosciences) or Combo-Rhod (synthesized in the lab of Kele30) at 10 °C overnight. To remove excess dye, the reaction mixture was purified via dialysis, repeated three times against 3 L of buffer using a 12−14 kDa MWCO dialysis membrane as described above. Determination of Amphiphilic Protein Concentration. The concentration of amphiphilic proteins was calculated by measuring the extinction of 1.8 μL protein solution at the appropriate wavelength (Table S8) using the spectrophotometer Nanodrop 1000 from PEQLAB Biotechnology GmbH. Lambert−Beer law was applied to calculate the protein concentration in μmol/L using the extinction coefficients of the respective fluorophores (Table S8). Encapsulation of Small- and Macromolecules into PMBCs. mCherry Encapsulation. His-mEGFP-H40I30 [30 μL (10 μM)] together with 1 μL [5.25 mg/mL] mCherry and 6 μL n-butanol were mixed five times with a Hamilton 250 μL syringe. PMBC samples were directly applied to microscope slides and imaged via epifluorescence microscopy. To evaluate the (co-)localization of the red fluorescent dye or protein and the green fluorescent amphiphilic protein constituting the PMBCs, images of the red and green channel were recorded and both images were superimposed and analyzed with ImageJ. Texas Red Dextran 3000 Encapsulation. His-mEGFP-H40I30 [50 μL (10 μM)] together with dextran [2 μL (0.1 mg/mL)] labeled with Texas Red MW3000 (Thermo Fisher Scientific) and 10 μL n-butanol were mixed 5−10 times in a 1 mL syringe with a gauge 17 needle. PMBC samples were directly applied to the microscope slide and imaged via epifluorescence microscopy as described above. For further analysis see “mCherry encapsulation”. FM4-64 Encapsulation. His-mEGFP-H40I30 and His-mEGFPS40I30 solutions were diluted to f.c. 5 μM with 10 mM Tris-HCl pH 8 and 5% glycerol. Of this protein solution, 100 μL was mixed with 5% 1-octanol for up to 10 times in a 1 mL syringe with a gauge 17 needle. Of each PMBC sample, 5 μL was directly applied to the microscope slide and imaged via epifluorescence microscopy as described above. For further analysis see “mCherry encapsulation”. 9596


Article

Langmuir ATTO Dye Rho13 Encapsulation. His-mEGFP-H40I30 [100 μL (10 μM)] in 10 mM Tris pH 8 together with 3 μL ATTO dye Rho13 [50 μM, f.c. 1.3 μM] from ATTO Tec Inc. and 15 μL n-butanol were extruded using an Avanti mini extruder 15 times through a membrane with a pore size of 1 μm. Excess ATTO dye Rho13 was removed subsequently through dialysis against 3 L of 10 mM Tris pH 8 buffer using a 12−14 kDa MWCO dialysis membrane three times. PMBC Fusion. His-BDP-K40I30 [40 μL (2 μM)] in 10 mM TrisHCL pH 8 buffer and 10 μL n-butanol were mixed five times using a Hamilton 250 μL syringe. This solution was applied to a microscope slide and fusion events were recorded with a Nikon Eclipse TS100-F fluorescence microscope in the green channel. To demonstrate the fusion of PMBCs in the solution, PMBCs assembled from His-mEGFP-H40I30 and His-mCherry-H40I30 as well as His-EYFP-S20I30 and His-mCherry-S20I30 were applied. For PMBC preparation, 100 μL of each protein solution (5−11 μM) was mixed with 10% of 1-octanol and extruded with a 1 mL syringe 10 times. Five minutes after extrusion, 5 μL of the PMBC solutions with the same amphiphilic protein domain and different fluorescent domains were placed into one droplet on a microscope slide without further mixing. The droplet was protected by a spacer and covered with a coverslip for subsequent epifluorescence microscopy analysis. Images of the green channel and red channel were taken and superimposed with ImageJ software to localize green and red fluorescent PMBCs. Co-localization of green and red fluorescent PMBCs demonstrate fusion events. Preparation of Multilamellar Liposomes. The preparation of multilamellar liposomes was done according to a standard procedure.31 ATTO 647N 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE, 100 μL)[1 mg/mL, 0.2 μmol, 1 equiv] in CHCL3 and 6.6 mg [0.025 mmol, 125 equiv] lyophilized DOPE from Avanti Polar Lipids were dissolved and mixed in 232 μL of CHCl3. Using the rotary evaporator KFN RC 900, the lipid film was dried in a 5 mL flask at 400 mbar for 2 h. Milli-Q water (1.5 mL) was used to dissolve the lipid film through vortexing. The film was homogenized with a Branson Sonifier Microtip for 1 min with 6 pulses for 10 s, at 10% amplitude and a 10 s pause between pulses. Subsequently, the homogenized lipid solution was extruded through an Avanti Mini Extruder using a 1 μm pore-sized Whatman membrane. Finally, the multilamellar liposomes were split in 40 μL aliquots and lyophilized overnight using the freeze-dryer ALPHA 2-4 LSC equipped with a vacuum pump (Vacuubrand RZ 2.5). Fusion of Multilamellar Liposomes and PMBCs. To assemble the PMBCs, 60 μL [10 μM] His-mEGFP-H40I30 in 10 mM Tris-HCL pH 8 together with 15 μL of n-butanol were mixed in a Hamilton 250 μL syringe (aspirating the solution five times up and down). Of the PMBC-containing solution, 5 μL was mixed with 3 μL of multilamellar liposomes of a fivefold diluted 40 μL aliquot (see “Preparation of Multilamellar Liposomes”) [f.c. 0.9 mg/mL DOPEbased liposomes]. Epifluorescence images were recorded in the green and red channels for 5 min after mixing of the solutions on the microscope slide. Internalization of ATTO 647N DOPE into the PMBC Membrane. His-mEGFP-H40I30 [330 μL (10 μM)] in 10 mM Tris-HCL pH 8 and 70 μL of n-butanol were extruded using an Avanti Mini Extruder with 15 passages through a membrane of 1 μm pore size. An aliquot of the assembled PMBC solution (40 μL) was lyophilized overnight. ATTO 647N DOPE (1.5 μL, 1 mg/mL) in CHCl3 was diluted in 998.5 μL of 10 mM Tris pH 8 to get a final concentration of 1 μmol/ L. Of this solution, 30 μL was added to a lyophilized aliquot of the pre-assembled PMBCs. Epifluorescence images were recorded after 7 min of incubation on ice. T4 DNA Ligase Reaction Inside the Assembled PMBCs. All ligation reactants were added and encapsulated during PMBC assembly. Ligation was allowed to proceed for 20 h after PMBC assembly and reactant encapsulation. Subsequently, DNA fragments from the ligation outside the PMBCs were captured via QIAXII silica particles (Qiagen) and loaded onto an agarose gel (detailed description below). In order to characterize the ligation reaction in the presence of PMBCs and to elucidate if the ligation occurs within

the assembled PMBCs, we designed and executed three different experimental setups (Figure 6c). First, DNA ligation in the presence or absence of PMBCs was performed. We used His-mEGFP-H40I30 as the amphiphilic protein constituting PMBCs and mEGFP as a control protein which cannot assemble into PMBCs (see Figure S8a3). PMBCs were assembled as stated above. The second setup was used to investigate if DNA ligation successfully occurs within PMBCs. We conducted an experimental approach where the ligase was added to the PMBC assembly reaction to be in- and outside the PMBCs. This was compared to a PMBC solution where the ligase was only outside the PMBCs because it was supplemented after the assembly reaction. In a third experimental setup, we executed a ligation with ligase added to the in- and outside of the PMBCs and additionally applied proteinase K (PK) to the in- and outside of the PMBCs (or only outside of the PMBCs if supplemented after the assembly reaction). His-EYFP-H20I30 amphiphilic protein [100 μL (7 μM) (f.c.)] was used for PMBC assembly. For the experimental setups with T4 DNA ligase in- and outside the PMBCs versus only outside addition as well as PK addition in- and outside versus only outside addition, these experiments were performed with amphiphilic HisEYFP-H20I30 because of the determined stability and very efficient assembly into PMBCs (see Figures S8b−d and 6a,b,d). For a standard ligation reaction inside PMBCs, 10 μg of phosphorylated and annealed oligonucleotides P0375 (5′GGTGTTCCGGGTCTG3′) and P0376 (5′ACCCAGACCCGGAAC3′) with three base pair 5′ overhangs (GGT, upper strand and CCA, lower strand) were used. These short DNA oligonucleotides were ligated with T4 DNA ligase yielding longer, multimerized DNA strands. These exemplary oligonucleotides encoding for the pentapeptide motif GVPGL are supplemented with a T7 promoter and set in the correct frame. Multiple ligation reactions result in (VPGLG)n multimers to be potentially expressed. The standard ligation reaction volume was 100 μL in 1× ligase buffer. T4 DNA ligase (1 μL f.c., 200 U/μL; NEB) and 11 μM (f.c.) of His-EYFPH40I30 amphiphilic protein for PMBC assembly were used. After mixing the components the reaction was supplemented with 10% v/v of 1-octanol (T4 DNA ligase activity is completely inhibited in the presence of n-butanol, but it remains active in the presence of 1octanol, data not shown) and extruded with an Avanti Mini Extruder as described above. As a negative control, we replaced the amphiphilic protein His-mEGFP-H40I30 with the same concentration of the nonamphiphilic mEGFP protein (no assembly to PMBC via extrusion). For the “only outside” scenarios, T4 DNA ligase or PK was supplemented immediately after the assembly of the PMBCs and incubated as described before. The ligation reaction was performed for 20 h at 20 °C. The amount and length distribution of the resulting ligated DNA strands were analyzed using 1.5% agarose gels in a 1× TAE buffer. Following the PMBC assembly and ligase/proteinase supplementation, we dialyzed the samples for 24 h at 8 °C against 10 mM Tris-HCL pH 8 buffer. Epifluorescence images were taken from these samples to demonstrate the stable formation of PMBCs. The volume of the dialyzed samples was measured and the concentration of DNA within these samples was calculated. DNA capture with QIAEX II particles was performed without a previous dialysis step. A total amount of 30 μL of QIAEX II silica particles (10 μL QIAEX II suspension theoretically binds up to 5 μg DNA) was added to each sample and the captured DNA was incubated for 2 h at 8 °C. The QIAEX II particles with the captured DNA were then precipitated via centrifugation (for 30 s at 300 rcf). This captured DNA was eluted with 20 μL of 10 mM Tris·Cl, pH 8.5, and analyzed on a 1.5% agarose gel (SN2 fraction). The amount of residual uncaptured DNA in the supernatant and the ligated DNA sheltered by PMBC was analyzed using a 1.5% agarose-gel (SN-fraction). The sample volume for the agarose gel was adjusted with respect to the calculated DNA concentrations within each sample. The quantification of the DNA in each lane was performed with gel analysis tools of the ImageJ Analysis Software. To compare the amount of DNA and the fragment length distribution of ligated DNA fragments, the gray values in the respective lanes of each experimental setup and fraction were 9597


Article

Langmuir determined from the original gel images. We categorized the DNA fragment lengths in groups corresponding to the 1 kb Plus DNA Ladder (NEB), normalized them against empty background lanes of the same agarose-gel, and calculated the relative amount and distribution of DNA fragments for each reaction fraction. For each experimental setup at least three independent experiments (biological replicates) were performed and the results are displayed as a column chart or line charts (Figures 6 and S8). For the first experimental setup, the average of the single experiments was calculated and the results are displayed as a column chart (Figure S8b). For the second and third experimental setups, we showed the results as line plots of the independent experiments in Figures 6 or S8, which all support the same conclusions. Furthermore, for the PK experiment we executed a T-test to validate the statistical relevance of our statement (Figure 6d). In Vitro Transcription and Translation inside PMBCs. For IVTT, 50 μL [14 μM] His-mEGFP-H40I30 in 10 mM Tris-HCL pH 8 was added into a 0.25 mL RNA-free PCR tube containing small pieces of a broken coverslip. After applying 15 μL of n-butanol and mixing five times with a Hamilton syringe equipped with a 0.25 × 25 mm needle PMBCs were yielded. Subsequently, the PMBCs were dried in a speedvac (Eppendorf Concentrator 5301) for 30 min and dried in high vacuum overnight to remove all residual n-butanol, which is detrimental for the IVTT. Milli-Q water (7.75 μL), 10 μL of buffer, and 3.5 μL of template DNA pQE-HA-mCherry-H40I30 [127 ng/ μL] were mixed (see Table S7). To this solution, 8.75 μL of crude E. coli extract containing ribosomes (70S), tRNA, translation initiation, elongation, and termination factors, the T7 RNA polymerase, the 20 amino acids, the 4 ribonucleotides ATP, GTP, UTP, CTP, and magnesium salt were added to yield a final volume of 30 μL IVTT mix. This 30 μL of IVTT mix was added to the dried PMBCs attached to the glass slides. After 20 min of incubation on ice and consequent encapsulation of the IVTT mix through swelling, 30 μL of the PMBC solution containing the IVTT mix was added to 20 μL of 50 mM TrisHCL pH 8100 mM NaCl and 4 mg/mL Kan (for condition II, IVTT inside) (Table S7). No Kan addition corresponds to the first condition (IVTT outside and insideno suppression). Kan addition after 20 min of swelling on ice corresponds to condition II (IVTT insidesuppression outside). Kan addition to the initial swelling solution corresponds to condition III (no IVTTsuppression in- and outside and IVTT). After expression for 5 h, images were recorded as described in the epifluorescence imaging Methods and Materials section. The in vitro translation inside the vesicular PMBC membranes was first imaged in the bright field (to avoid fluorophore bleaching) and subsequently imaged using the red fluorescent channel and consistently 10 s of exposure time for conditions II and III. Finally, vesicles were imaged in the green channel applying 100−400 ms of exposure time. IVTT-T-Test Analysis. Line plots of red fluorescent PMBCs were generated using ImageJ win 64 analysis software. All line plot maxima (maximal red membrane fluorescence signal), minima (minimal red background fluorescence signal), and the minimal red lumen fluorescence signal were collected. The average membrane signal to noise ratio (SmN ratio) and membrane fluorescence signal to lumen ratio (SmL ratio) were calculated. The p-value was computed using Students t-test by defining the following three null hypotheses: first, average SmN ratios for condition I (IVTT inside and outside), II (IVTT inside), and III (no IVTT) are not different from each other. Second, average SmL ratios for conditions I, II, and III are not different from each other. Third, average total fluorescence of 50 μL of PMBC solution (diluted fivefold) for conditions II and III and 10 mM Tris-HCL pH 8 buffer are not different from each other. Parameters set for Students t-test were set as unpaired t-test, unequal variances, and the null hypothesis was rejected if p < 0.01. All three null hypotheses were rejected for all cases. Thus, the differences between SmN ratios, SmL ratios, and total red fluorescence signals for conditions I, II, and III are statistically significant with a probability greater 99%. Plasmids and Constructs for the Assembly of PMBCs. The plasmids used in this study are either derived from the pET-vector

system and cloned as described on the basis of the OVTP cloning system32 or derived from the Qiagen pQE vector (kindly provided by AG Roth) with a T7 promoter used for the IVTT experiments described above. All vectors are expressed in E. coli strains BLR (DE3) or BL21 (DE3) cells and listed in Table S1.

■

RESULTS AND DISCUSSION Prebiotic Peptides for PMBC Assembly. There is agreement about a set of very abundant (Gly, Ala, Asp, Val) and less-abundant prebiotic aa such as (Glu, Ser, Ile, Leu, Pro, His, Lys). The most abundant prebiotic aa Gly, Ala, Asp, Val are spontaneously formed under Stanley Miller or similar prebiotic atmosphere conditions (endogenous sources) and were frequently found on meteorites or deep sea black smoker (exogenous sources).27,33 The less-abundant prebiotic aa are structurally more complex and mainly evidenced from endogenous sources.33 In our work, peptides composed of rather hydrophobic prebiotic aa such as Gly, Ile, Val, Pro, and Leu and hydrophilic aa such as Ser, His,34,35 and Lys have been applied (Table 1). Oligopeptides have been shown to be synthetically accessible under prebiotic conditions on clays,36 minerals,37 metals, and under hydrothermal conditions38 and even at air−water interfaces.39 Rode and Schwendinger showed the simplest mechanism of salt-induced peptide synthesis from prebiotic aa.40 The consequent peptide chain elongation on clay minerals within the same environment and reaction conditions yields oligopeptides in reasonable amount and is applicable to all aa investigated so far (Gly, Ala, His, Val, Lys, Leu, Glu, Asp, and Pro).41 Recently, high yields up to 50% and up to 20 aa could be synthetized applying “one-pot” reaction conditions.42 They demonstrated the co-condensation of glycine with eight other aa (Ala, Asp, Glu, His, Lys, Pro, Thr, and Val), incorporating a range of side-chain functionality. Thus, all the constituent aa and potentially the oligopeptides of the amphiphilic peptides employed in this work (Table 1) are accessible under prebiotic conditions. The minimal aa and hence genetic sequence, which allow to constitute amphiphilic proteins forming dynamic membranes, require an efficient sequence motive providing the structural requirements for dynamic membrane formation. The pentapeptide repeat motif applied as a basic structural element in this work corresponds with the VPGVG aa sequence repeat denominated as elastin-like protein (ELP) basic repeat motive. This ß-spiral repetitive sequence motif was applied because of its possible cylindrical geometry and dynamic analogy to phospholipids49 and their membrane-forming capacities.5 Hence, the repetitive sequence motif applied in this work is exemplary for other simple prebiotic amphipathic peptides with structure-forming capability. In view of prebiotic peptide synthesis, repetitive peptide motifs comprise synthetic plausibility as first membrane constituents for several reasons. In the presence of one abundant or dominant protein building block motif such as an individual prebiotic amino acid (e.g., glycine)42 or one prebiotic peptide motif (e.g., VPGHG) the multimerization of the same motive (aa or peptide motive) has a high propensity. In addition, idealized computational model simulations predict a favored appearance of highly ordered and repetitive sequence patterns starting from monomer pools for autocatalytically active binary polymers,50 supporting the emergence of simple repetitive protein motifs comparable to those we used in this study. The role of peptides under early earth conditions is currently under debate in the context of the 9598


Article

Langmuir

Figure 2. Epifluorescence images and TEM images demonstrate the formation of PMBCs from amphiphilic protein building blocks composed of prebiotic aa. All mentioned constructs in this figure (except in c,d) have a (VPGIG)30 domain as the hydrophobic block of the amphiphilic protein. (a) Epifluorescence images of PMBCs supplemented with an mEGFP domain for visualization. On the left, PMBCs assembled of amphiphilic proteins contain a hydrophilic histidine protein domain (VPGHG)40. In the middle, PMBCs composed of an amphiphilic protein contain a hydrophilic serine protein domain (VPGSG)40 instead of histidine. On the right, a TEM image of negatively stained PMBCs is assembled of the same construct as on the left. At the right end of the row the membrane composition of PMBCs constituted from mEGFP-supplemented amphiphilic proteins and n-butanol before and after dialysis is illustrated. (b) Images of PMBCs assembled of amphiphilic proteins are supplemented with an EYFP domain for visualization in epifluorescence microscopy. The left image depicts PMBCs assembled of amphiphilic proteins containing a hydrophilic histidine domain (VPGHG)20. The middle image shows a correspondent amphiphilic protein with a hydrophilic serine protein domain (VPGSG)20 instead of histidine. At the right, a TEM image of negatively stained PMBCs of the same construct as in the middle is shown. At the right end of the row the membrane composition of the PMBCs constituted from EYFP-supplemented amphiphilic proteins and n-butanol before and after dialysis is illustrated. (c,d) Preserving the PMBC structure, reduction of prebiotic aa composition and length is proven by PMBCs assembled of amphiphilic proteins composed of (c) hydrophilic (VPGHG)10 and hydrophobic (VPGLG)4 domains and (d) hydrophilic (VPGHG)5 and hydrophobic (VPGLG)4 domains. For visualization proteins are supplemented with an unnatural amino acid pAzF, which is labeled with a chemical dye indicated as DBCO-bodipy (BDP) (see methods and Figure S2). The scale bar corresponds to 2 μm. The stronger background fluorescence in (c,d) (compared to a,b,e,f) indicates a reduced self-assembly efficiency because of weaker noncovalent interaction forces of the very short amphiphilic peptides. (e) Image of PMBCs assembled of amphiphilic proteins composed of hydrophilic (VPGKG)40 and hydrophobic (VPGIG)30 supplemented with the unnatural amino acid pAzF, which is labeled with BDP. The structural formula of BDP, used in c, d and e, is depicted aside with the respective pictogram. (f) Image of PMBCs assembled of amphiphilic proteins composed of hydrophilic (VPGKG)20 and hydrophobic (VPGIG)30 supplemented with the unnatural amino acid pAzF, which is labeled with red dye (ComboRhod) (see methods and Figure S2). The formula of the chemical dye used in (f) is presented aside with the respective pictogram. (g) Image of PMBCs assembled of amphiphilic proteins composed of hydrophilic (VPGHG)40 and hydrophobic (VPGIG)30 with no fluorescent domain or dye. The formula of the chemical dye FM4-64, which is enclosed by the PMBCs and indirectly used to visualize them in fluorescence microscopy, is presented below with the respective pictogram. (h) The illustration represents the different membrane (or compartmental) compositions of the PMBCs constituted from amphiphilic proteins as indicated in (c/d/e,f,g) (indicated by gray arrows from the respective adjacent panels). The scale bars correspond to 5 μm (except for c,d) in epifluorescence images and 100 nm in TEM images. The gray background of the left column (a−d) emphasizes the fact of the successively decreasing length of the amphiphilic PMBC constituents from panel (a−d): His-mEGFP-H40I30 > HisEYFP-H20I30 > BDP-H10L4 > BDP-H5L4 (also see Table 1).

information might have settled before the “RNA-world era”.51 In this context, it was currently demonstrated that short

amyloid world hypothesis. Accordingly, peptide amyloids as evolvable molecular entities able to self-replicate and transmit 9599


Article

Langmuir

pentapeptides, Table 1). Successful PMBC assembly from different “prebiotic” amphiphilic peptides is accomplished applying a very simple assembly method32 (see the Methods and Materials section) (Figure 2). The fluorescent images depict vesicles of 1 μm up to 15 μm enclosed by a homogenous fluorescent membrane boundary. Corresponding negatively stained TEM images show vesicular structures up to 800 nm in diameter. Amphiphilic peptides equipped with mEGFP or EYFP, different hydrophilic block lengths (n = 20 vs n = 40), and serine versus histidine as additional prebiotic aa were assembled (Figure 2a,b). Successful PMBC assembly is demonstrated with two different hydrophilic block lengths (n = 20 vs n = 40) and prebiotic guest aa (H, S) (Figure 2a,b), emphasizing the promiscuity of the composition of the amphiphilic peptides. In order to prove that PMBC assembly does not rely on “non-prebiotic” fluorescent proteins we introduced two different small fluorescent dyes via click chemistry shown in Figure 2c−f (for reaction details see Figure S2). Fluorescent images in red illustrate Combo-Rhod-K20I30 and in green BDP-K40I30 assembled into PMBCs of 0.7−6 μm in diameter. BDP-H10L4 and BDP-H5L4 assembled into PMBC of 0.3−4 μm in diameter. This illustrates that successful PMBC formation does not rely on fluorescent fusion proteins. To also exclude the necessity of organic dyes for vesicular assembly, exclusively prebiotic His-H40I30 PMBCs and shorter amphiphiles (see Figure S1) were assembled and imaged via an encapsulated dye (Figure 2g,h). This demonstrates the ability of PMBC assembly solely based on prebiotic aa. Two prebiotic PMBC assembly scenarios are described below, which yield vesicles with different efficiencies and compositions. One method completely omitting any additional organic component is based on rehydration, whereas the second method applies alkanols yielding PMBCs in very high abundance. Simple dehydration of the amphiphilic protein film on glass slides and rehydration with subsequent swelling of amphiphilic protein film yields vesicles up to 2 μm (Video S3). This very simple vesicular self-assembly only requires prebiotic amphiphilic proteins, water, and one cycle of de- and rehydration to yield PMBCs, illustrating a plausible prebiotic assembly scenario. In the second method, “prebiotic” amphiphilic proteins are extruded with 5−20% v/v alkanols such as n-butanol or 1-octanol to induce the PMBC selfassembly with high yield of PMBCs up to 15 μm in diameter. Rapid mixing of amphiphilic proteins and alkanols during extrusion is essential to achieve a high yield of vesicle formation. Illustrating the robustness and prebiotic relevance, the salt concentration during assembly may vary from zero up to 5 M and the temperature during vesicle formation can range from 4 °C up to 70 °C. Further, PMBCs assembled from, for example, His-EYFP-H20I30 with 5% v/v octanol preserve their vesicular integrity and size distribution for at least 9 months (Figure S3, Video S4). This alkanol-assembly method is applied throughout all experiments if not stated otherwise and details can be found in the methods section. After dialysis against buffer and removal of n-butanol the PMBCs are stable up to 100 °C, pH 2−11 (Figure S4). However, it cannot be excluded that after dialysis residual amounts of n-butanol could be present within the dynamic protein membranes. When compared to the alkanol-free swelling protocol, the number of vesicles is roughly increased a 1000-fold. The very simple protocol of mixing amphiphilic proteins with fatty acid derivatives (fatty alcohols) is also of prebiotic relevance, as fatty acids and fatty alcohols have been synthesized at prebiotic

repetitive peptide sequences may act as information-coding molecules that catalytically allow the template-directed synthesis of defined repetitive peptide motives.51 Whereas this was shown for repetitive amyloid ß-sheet motives, this could be a mechanism to receive and maintain sequence motives necessary for the formation of PMBCs. This demonstrates the bridging ability of replicating first information-coding molecules on the basis of prebiotic aa with compartmentforming capability. On the basis of previous results, where nonprebiotic PMBCs were assembled in vivo5 and in vitro52 here we designed prebiotic amphiphatic ELP sequences exclusively based on prebiotic aa (Table S1 and Data S2). The used aa repeat motif (VPGVG) tolerates mutations at the fourth position (valine; V) while maintaining its structural and functional properties.53 This variability permits the adjustment of the protein domain properties by changing the number of repeat units or the aa at the fourth position. Therefore, the hydrophilic and hydrophobic domains, respectively, were modified with hydrophilic prebiotic aa (S, K, H) at the fourth position of each pentapeptide motif and with hydrophobic prebiotic aa (I, L, V) at the same position to yield “prebiotic” amphiphilic peptides (Table 1). N-terminal fusions of fluorescent proteins mEGFP and mCherry were added for direct characterization via fluorescence microscopy. The ratio of the amphiphilic block-domain proteins (Table 1) consisting of 20 or 40 repeats of (VPGXG)n (X = S, H, K) and thirty units of VPGIG were chosen based on geometric properties of ELPs,53 hydropathic indices of the domains, and stability results of the previously assembled “non-prebiotic” PMBCs.5,52 Previously, we could show that different assembly conditions, ratios of hydrophilic and hydrophobic domains, and the nature of the guest residue determine the shape, physiochemical properties, and stability of the assembled supramolecular structures in vitro.52 Even though the constituting aa and potentially the oligopeptides of the amphiphilic peptides employed in this work (Table 1) are amenable under prebiotic conditions, considerably shorter amphipathic peptides preserving the structure-forming capacity would improve the feasibility of a prebiotic assembly scenario. The reduction of aa composition, complexity, and length down to nine pentapeptides allows to preserve the vesicle-forming capacity even though PMBC assembly efficiency and stability is attenuated (Figure 2, Tables 1 and S2, Figure S1, Videos S1 and S2). Thus, PMBCs assembled from these short amphiphilic proteins dissociate within 10−30 min when assembled with n-butanol and after a few hours when assembled with 1-octanol. The overall assembly efficiency of the short amphiphilic proteins (9−10 pentapeptides) is only a few percent when compared to the very high PMBC assembly yield of longer amphiphilic proteins (70 pentapeptides, Table 1). Our observation that even very short amphipathic peptides assemble into dynamic PMBCs, stable for a few hours, underlines the probability of a prebiotic peptide or protein membrane-based compartmentalization and supports our model. It also opens a path of evolutionary pressure and development toward longer and more stable amphipathic proteins up to the point of being superseded by lipid membrane constituents. Nevertheless, the limited stability of these short peptides quests for other more efficient assembly motifs. Therefore, to comply with requirements regarding stability and assembly efficiency, further PMBC assembly, characterization, and application experiments in this work were conducted using longer amphiphilic peptides (50−70 9600


Article

Langmuir

Figure 3. PMBCs encapsulate small dyes, macromolecular dyes, functional proteins, and small PMBCs emphasizing their capacity as a functional reaction chamber. (a) Fluorescence images in green (left), red (middle), and merged channels (right) demonstrate that a PMBC assembled from His-mEGFP-H40I30 and 13% v/v n-butanol encapsulates small ATTO Rho 13 dye during assembly. Subsequent dialysis after assembly against 10 mM Tris-HCl buffer pH 8 leads to removal of n-butanol and excess of ATTO Rho 13. (b) Fluorescence images (upper and lower rows) in green (left), red (middle), and merged channels (right) demonstrate that a PMBC assembled from His-mEGFP-H40I30 and 16% v/v n-butanol encapsulates macromolecular Texas Red Dextran 3000 dye during assembly. The encapsulated dye concentration between PMBCs differs (marked with white arrows in the left and right upper images). The lower images demonstrate the encapsulation of smaller PMBCs (containing the red dye) and protein membrane fragments within the PMBC assembly process. (c) Fluorescence images in green (left), red (middle), and merged channels (right) demonstrate that a PMBCs assembled from His-mEGFP-H40I30 and 16% v/v n-butanol encapsulates correctly folded mCherry during assembly. On the right side of each row (a−c), the formula and pictogram of the enclosed molecules are illustrated together with their mass or dimension. Pictograms in the left upper edge of each image illustrate the visual constituents of the reaction. All scale bars correspond to 5 μm. (d) Pictograms illustrate the process of assembly and encapsulation of chemical dyes and mCherry into the PMBCs. In case of extrusion without subsequent dialysis residual amounts of n-butanol could be present within the dynamic protein membranes as illustrated.

conditions45,46 and were found on meteorites.47,48 Fatty acids and their corresponding alcohols are attractive candidates for protocell constituents itself.9 They form bilayer membrane vesicles47 that are capable of growth and division54,55 and such membranes are highly dynamic.12 So far, all proposed prebiotic protocell boundaries are based on pure and single constituents,6 even though prebiotic reaction conditions and meteorites contain always a mixture of molecules of different classes (aa, fatty acids, sugars). In addition, recent studies have shown that mixtures of amphiphiles allow for an efficient and more stable formation of membranes.56 Therefore, we further tested the compatibility of our PMBCs by extruding simple fatty acids (e.g., lauric acid) together with our prebiotic

amphiphilic proteins also yielding PMBCs in medium concentrations (Figure S5). The compatibility of alkanols and fatty acids (Figure S5) with our PMBCs increases the plausibility of a mixed assembly scenario supposing both amphiphilic components are prebiotically available in spatial proximity (as indicated by meteorite analysis). Nevertheless, PMBC assembly with lower yield solely necessitates prebiotic amphiphilic proteins as described above (Video S3). Comparing the different assembly scenarios with respect to the vesicular yield the following order can be observed: the highest PMBC yield is observed for amphiphilic protein/ alkanols > amphiphilic protein/fatty acid > pure amphiphilic protein, whereas the long-time stability (>2 weeks) is observed 9601


Article

Langmuir

Figure 4. Epifluorescence images demonstrate the fusion of cell-sized PMBCs assembled from different amphiphilic proteins. (a) Time series of the fusion of two 10 μm sized PMBCs consisting of BDP-K40I30 in 10 mM Tris-HCl pH 8 and 10% v/v n-butanol recorded with four frames per second in the green channel. (b) Time series of the fusion of 3 and 14 μm sized PMBCs consisting of BDP-K40I30 in 10 mM Tris-HCL pH 8 and 10% v/v n-butanol recorded with four frames per second in the green channel. (c) Negatively stained TEM images of the fusion of 150 nm (left) up to 700 nm sized PMBCs (right) consisting of His-mEGFP-H40I30 in 10 mM Tris-HCL pH 8. (d) Schematic illustration of the fusion of PMBCs assembled from amphiphilic proteins (a−c). (e) Epifluorescence images of post assembly mixtures of two variants (His-mEGFP-H40I30 and HismCherry-H40I30) of PMBCs in 10 mM Tris-HCl pH 8 and 10% v/v 1-octanol. The left image in the green channel visualizes His-mEGFP-H40I30 proteins and PMBC assembled thereof. In the middle image the red channel visualizes the His-mCherry-H40I30 proteins and PMBCs assembled thereof. The right epifluorescence image displays an overlay of the green and red channel images and reveals disjoined PMBCs and also fused PMBCs consisting of amphiphilic proteins with both fluorophores (mEGFP and mCherry). The gray rectangle marks a fused PMBC, which is illustrated with pictograms of the present amphiphilic proteins at the right of this image. (f) Epifluorescence images of two PMBC variants (one composed of His-EYFP-S20I30 and the other His-mCherry-S20I30) in 10 mM Tris-HCl pH 8 and 10% v/v 1-octanol. In the left image the green channel visualizes the His-EYFP-S20I30 proteins and PMBCs assembled thereof. In the middle image the red channel visualizes the His-mCherryS20I30 proteins and PMBCs assembled thereof. The right image displays an overlay of the green and red channel images and reveals a disjoined PMBC and also a fused PMBC consisting of amphiphilic proteins with both fluorophores (EYFP and mCherry). The gray rectangle marks a fused PMBC, which is illustrated with pictograms of the present amphiphilic proteins at the right of this image. (g) Schematic illustration of the fusion process of PMBCs consisting of variants of amphiphilic proteins (here His-EYFP-S20I30 and His-mCherry-S20I30). The quadratic rectangle in the right image indicates the homogeneously distributed amphiphilic constituents within the resulting PMBC membrane after fusing. Pictograms in the upper left edge of each image illustrate the PMBCs constituting amphiphilic proteins. Scale bars of the epifluorescence images correspond to a 5 μm and scale bars in TEM images correspond to 200 nm.

only for the first and latter scenario. This demonstrates the possibility of different PMBC membrane compositions but does not allow for any prioritized assembly scenario. In addition, the alkanol assembly protocol and yielded PMBCs

are more robust under different environmental conditions when compared to the very promising fatty acids protocell model9,12 which is only stable in a narrow pH range and under low ionic strength.13 In this regard, vesicular assembly of 9602


Article

Langmuir

Figure 5. Epifluorescence images demonstrate compatibility with and internalization of contemporary phospholipids (red) into prebiotic PMBCs (green). Epifluorescence image of (a) His-mEGFP-H40I30-based PMBCs in 10 mM Tris-HCL pH 8 and 10% v/v n-butanol in the green channel (left), multilamellar DOPE-based liposomes in the red channel (middle) and merged channels (right) demonstrate successful incorporation of DOPE-based liposomes into PMBCs (green). (b) Line scans of left and right images illustrate the successful fusion of multilamellar DOPE-based liposomes (red) with PMBCs. (c) Schematic illustration of the fusion of DOPE-based liposomes and PMBCs based on His-mEGFP-H40I30. (d) Whitefield (left) and epifluorescence images demonstrate the homogeneous incorporation of DOPE ATTO red dye 647N DOPE (red channel) into protein-based vesicles membrane (green) upon mixing with lyophilized PMBCs. The merged green and red channels on the right demonstrate the homogenous incorporation. (e) Schematic illustration of the internalization of DOPE ATTO red dye 647N into the protein membrane. The scale bars correspond to 5 μm in (a) and 2 μm in (d).

themselves (Figure 3 a−c). This illustrates a complete or partial membrane tightness and diffusion barrier toward molecules of different nature such as sugars (dextran), positively charged small molecules (ATTO Rho 13), and completely folded proteins, essential under harsh environmental conditions. Further, these results demonstrate that this assembly protocol does not compromise the functionality of the encapsulated dyes or proteins. In addition, the internalization of a whole PMBC inside another PMBC (Figure 3b) illustrates the possibility of sub-compartmentalization of spatially separated reactions within one prebiotic reaction compartment (Video S5). Another requirement for prebiotic protocells, apart from molecular encapsulation, is membrane fluidity, fusion, and fission potential which is described below (Figure 4). PMBC Membrane Dynamics. Dynamic membrane properties are required for membrane growth, exchange of vesicular contents, and nutrient intake, for example, exo- and endocytosis.11 Membrane fusion is a process important to many cellular events, especially for transport of nutrients or signaling. The dynamic properties and fluidity of the protein-

PMBC omitting alkanols requires an amphiphilic protein concentration 4 orders in magnitude less compared to lipid vesicles (10 mM fatty acids12 vs 16 μM of amphiphilic protein) and tolerates up to 5 M urea, which is in contrast to fatty acid vesicles. Hence, supposing the successful synthesis of amphiphilic protein building blocks, the formation of protocell-like compartments at such a low concentration may have interesting advantages in a prebiotic world. In the light of environmental conditions on the prebiotic earth, the stability of PMBCs up to 100 °C and pH ranges from 2 to 11 supports the emergence of prebiotic PMBC as first protocell models. Encapsulation of Molecules into PMBCs. The encapsulation of intact molecules into PMBCs is one important step toward the design of artificial cells and characterizes the PMBC properties and membrane permeability. The addition of molecules to be encapsulated during PMBC assembly leads to the inclusion of these molecules into the vesicular lumen (Figure 3). During assembly, these PMBCs can encapsulate molecules of different length scales, ranging from small dyes (800 Da) to macromolecular dyes (3 kDa), to functional completely folded proteins (28 kDa) and PMBCs (8 μm) 9603


Article

Langmuir

Figure 6. DNA ligation inside prebiotic PMBCs emphasizes their compatibility with important anabolic reactions. (a) Interpolated line plots display the size distribution of DNA length after ligation for the experimental setups illustrated in (c): ligase is present in- and outside the PMBCs or only outside (Table S3b). The blue lines indicate the reactions (SN, SN2) with ligase in- and outside the PMBCs, the red line with ligase only outside (SN, SN2). (b) Interpolated line plots display the size distribution of DNA length after ligation for the experimental setups: T4 DNA ligase present in- and outside PMBCs and PK also applied [see (c)] in- and outside (gray line) or only outside (red line) the PMBCs or PK not present at all (blue line) (for details see Table S3c). (c) The flowchart depicts the experimental setups of the described approaches and the respective ingredients. From left to the right the approaches with ligase inside and outside PMBCs (A1/L1), without PMBC formation (C1), with ligase only outside the PMBCs (L3), with additionally PK applied to in- and outside of the PMBCs when ligase is also present in- and outside (L4) and finally the PK is added only outside the PMBC (L5) whereas ligase is present in- and outside. The light red and blue fields at the bottom of each reaction process mark the reaction products/DNA fragments used for quantitative analysis via agarose gel-electrophoresis. The red fields and the red arrow indicate the DNA captured from outside the PMBC by DNA binding beads. The blue field and the blue arrow indicate the ligation products inside PMBCs or in excess with respect to binding capacity of the DNA binding beads. These fractions (SN/blue and SN2/red) were used for further quantitative analysis. (d) The chart depicts the average length of the DNA fragments for the three outlined experimental setups. The average length of the DNA fragments for no PK addition or PK outside vs PK in- and outside are significantly larger (with * for p < 0.05 and ** for p < 0.01). Ligation results are based on at least three independent experiments (for details see methods and Figure S8d and Tables S3 and S4).

flattening, and deformation57 which can be observed in Figure 4a,b. The resulting increase in interfacial adhesion energy leads to membrane stretching and the exposure of hydrophobic areas and pockets and provokes the ultimate breakthrough fusion event.49 In addition, PMBCs composed of His-mEGFPH40I30 in buffer lacking alkanols undergo fusion events observed on the TEM grid surface (Figure 4c), illustrating that alkanols are not required for fusion. In order to demonstrate

based membrane is shown in Figure 4. The fusion of cell-sized PMBCs composed of BDP-K40I30 is illustrated in Figure 4a,b and Video S6. PMBC fusion events are observed on distinct areas of the microscope cover slide, where PMBCs are adherent on the glass slide. In Figure 4 two fusing, adherent 10 μm sized PMBCs can be observed, whereas in Figure 4b only one of the two fusing PMBCs is adherent. Surface adhesion of vesicles often leads to tensile membrane stress, 9604


Article

Langmuir

PMBC protocell model to current phospholipid membranebased cells. Anabolic Reaction Pathways toward Evolvable PMBC Protocells. The simple encapsulation of functional molecules into the PMBCs is shown in Figure 3 and is an important requirement in building artificial cells but is essential to define PMBCs as evolvable protocells. Beyond that, a major step toward a self-sustaining protocell model is the capability to assemble and to propagate system properties or genetic information and the ability to accommodate anabolic reactions as for example the polymerization of RNA, DNA molecules, or aa. Therefore, the ability of PMBCs to embed the polymerization of genetic information that subsequently could serve as an evolvable template for the repetitive protein membrane constituents of the PMBC was investigated. In Figure 6 we present the ligation of 15 base pair (bp) oligonucleotides (encoding for one pentapeptide ELP motive) to high molecular DNA strands with up to 1200 bp within prebiotic PMBCs. The length of the ligated DNA presented here potentially encodes for proteins of up to 80 pentapeptide repeats, which fits well to the DNA template size of our investigated amphiphilic PMBC constituents. To prove the ability of PMBCs to enclose successful anabolic DNA ligation processes we performed three complementing experimental setups (Figure 6c). First, we conducted the DNA ligation in the presence or absence of PMBCs. Therefore, amphipathic His-mEGFP-H40I30 proteins were used as PMBC constituents or alternatively mEGFP was applied as a control protein not capable of PMBC formation (details in the methods section). In the presence of PMBCs, the DNA fragment length distribution is shifted toward longer fragment length when compared with the control (Figure S8b) at least for fragment lengths larger than 400 bp. This result implies that the confined PMBC space increases the ligation efficiency. Second, an experimental approach was executed where T4-DNA ligase was present in- and outside PMBCs or only outside the PMBCs (details in the Methods and Materials section). Accordingly, captured DNA from the inner fraction (SN) yields at least twofold higher amounts of ligated DNA than in the outer fraction (SN2) when ligase is present inside and outside the PMBCs (for fragment lengths between 200 and 800 bp, compare Figures 6a and S8c). For the inverse case the relations are rather inverted or equal. This indicates successful DNA ligation within the PMBCs. To independently prove the ligase action inside the PMBCs a third experimental setup was conducted (Figures 6b and S8d). Therefore, amphipathic His-EYFP-H20I30 proteins were used as PMBC constituents because of their proven long-term stability and great assembly efficiency. Here, T4 DNA ligase was present in- and outside the PMBCs and additionally PK was supplied in- and outside the PMBCs or only outside (for details see Methods and Materials section). PK degrades and inactivates the T4 DNA ligase. The results (Figures 6a,b and S8d) demonstrate the efficient inactivation of T4 DNA ligase within this experimental setup observing a statistically significant reduction of ligated fragment length when PK is present. The average length of the DNA fragments for no PK addition or PK outside versus PK in- and outside (Figure 6d) are significantly larger (Figure 6d, Table S4). Thus, it can be assumed that after several hours (approximately 8−10 h) PK also degrades the protein membrane of the PMBCs and abrogates the ligase-protecting barrier function against PK. For a limited period of time the PMBC membrane protects the

fusion events and phase separation of assembled PMBCs in solution, two variants of “prebiotic” amphiphilic proteins were equipped with two different fluorescent fusion proteins each. The right images in Figure 4e,f illustrate PMBC membrane patches of different color (red and yellow or green) belonging to a single PMBC. These patches indicate a previous fusion event of PMBCs, equipped with different fluorophores (mCherry vs EYFP or mEGFP), but composed of the same amphiphilic domains H40I30 and S20I30, respectively. This emphasizes that PMBC fusion events upon mixing in solution at RT do occur and that fusing membranes stay phaseseparated for at least 20 min. However, mixing of PMBCs constituted of different “prebiotic” amphiphilic proteins (HismEGFP-S40I30 and His-mCherry-S20I30) here with a substantial length difference in the hydrophilic domain does not lead to visible fusion events (Figure S6). This observation indicates a specific or selective type of fusion where protein membranes composed of the very same constituents selectively fuse with each other. Such fusion events are frequent in contemporary phospholipid-bound cells where membrane proteins, for example, at synapses usually initiate specific fusion events.58 In the protocellular context, the dynamic PMBC membrane permits specific fusion and therefore has the potential for directed exchange of vesicular content for, for example, more complex multistep reactions or exchange of evolved ribozymes. Another important property of prebiotic PMBCas a new type of protocell modelis the compatibility with contemporary membranes and their constituents. Liposome and Phospholipid Compatibility with PMBCs. The main building block of extant cellular membranes are phospholipid bilayers. In Figure 5 we show the compatibility of PMBCs with one of the main contemporary cell membrane constituents. Multilamellar liposomes (red) based on DOPE can be internalized into PMBCs by mixing preassembled PMBCs and liposomes (Figures 5a,b and S7). Solubilized DOPE ATTO red dye 647N (DOPE) is homogeneously incorporated into protein-based vesicle membranes (green) upon swelling together with lyophilized PMBC (Figure 5d,e). Solubilized DOPE lipids integrate into the protein membrane. In contrast, preassembled uni- and multilamellar DOPE-based liposomes are internalized into the PMBC lumen and phospholipids are rearranged to dense homogenously distributed spheres (Figure S7). Internalized preassembled phospholipids seem to favor the intermolecular interaction with their analogues instead of protein amphiphiles (Figure 5a). This process can be described as a sort of phase separation in analogy to lipid−lipid phases separation. Such phase separations are due to collective properties of macromolecules and particularly relevant for the spatial organization of metabolic reactions in living organisms.59 In our previous work,5 we observed a strong interaction of non “prebiotic” PMBCs with the lipid-based cellular membrane of E. coli, in vivo. The close association and invagination of the cellular membrane associated with PMBCs was observed using TIRF structured illumination microscopy. Those PMBCs were assembled from peptides containing some nonprebiotic aa (e.g. Glu). However, the association of the E. coli membrane and PMBCs also demonstrates a strong interaction and fusion tendency of lipid bilayers and dynamic protein membranes in vivo.5 The compatibility of phospholipids and lipid membranes with PMBCs might have facilitated a transition of a prebiotic 9605


Article

Langmuir

Figure 7. Synthesis and incorporation of the PMBC membrane building block mCherry-H40I30 (red) inside PMBCs (assembled of His-mEGFPH40I30) via IVTT. (a) Schematic illustration of a PMBC assembled from His-mEGFP-H40I30 (green) after encapsulation of the IVTT mix, (see the methods section) and plasmid DNA encoding for mCherry-H40I30. After encapsulation of the IVTT mix into the PMBCs, mCherry-H40I30 expression and subsequent membrane incorporation leads to red fluorescent PMBCs. (b) Red fluorescence images of PMBC after 5 h of IVTT with three different IVTT conditions. (I) In the absence of Kan the IVTT of mCherry-H40I30 inside and outside the PMBCs leads to strong red fluorescent membrane signal and red protein agglomerates outside. (II) Addition of Kan to the outer PMBC solution inhibits outer IVTT (no red agglomerates) and leads to a red fluorescent membrane signal of the PMBC via IVTT and incorporation of mCherry-H40I30 from inside the PMBCs. The different red membrane intensities demonstrate the variable IVTT mix encapsulation efficiencies upon swelling. (III) The addition of Kan to the outer and inner PMBC solution entirely inhibits IVTT outside and inside the PMBCs. The weak red signal results from red auto fluorescence of the PMBC upon excitation at 587 nm. (c) Green and red fluorescence images of PMBC after 5 h of IVTT of mCherry-H40I30. (I) Line plots of the red fluorescent PMBC upon IVTT inside and outside the PMBC show a 5.4 ± 2.5-fold membrane signal to noise ratio (SmN ratio) and 2.2 ± 0.4-fold membrane signal to lumen ratio (SmL) [also see d(ii,iii)]. (II) Line plots of the red fluorescent PMBC upon IVTT inside the PMBC show a 2.7 ± 0.5-fold SmN ratio and about 1.5 ± 0.2-fold SmL ratio {also see [d(ii,iii)]}. This indicates that about half (SmN ratio 2.7 vs 5.4) of the expressed mCherry-H40I30 was synthesized and incorporated into the PMBC membrane inside the PMBC when compared to condition (I). (III) Line plots of the control PMBC (no IVTT) where mCherry-H40I30 IVTT is inhibited inside and outside leads to 1.4 ± 0.2fold SmN ratio and a SmL ratio of 1.1 ± 0.1 {also see [d(ii,iii)]}. [d(i)] The left plot illustrates the total red fluorescence emission of the 50 μL PMBC solution (fivefold diluted in buffer) after 5 h mCherry-H40I30 IVTT inside PMBC quantified using photospectrometric quantification see Table S6). The red fluorescence emission (at 610 nm) of the PMBC solution is significantly different (p < 0.01) when compared to 10 mM TrisHCL buffer solution (blue) and the control (no IVTT) where IVTT is inhibited outside and inside (red) (ii). The middle plot illustrates the mean SmL ratio of line scans through 10 PMBCs for the three conditions IVTT outside and inside, IVTT inside, and no IVTT (see Table S5). The SmN 9606


Article

Langmuir Figure 7. continued

ratio of IVTT inside is significantly different (p < 0.01) when compared to both the other IVTT conditions. (iii) The right plot illustrates the mean SmL ratio of line scans through 10 PMBCs for IVTT outside and inside, IVTT inside, and no IVTT. The SmL ratio of IVTT inside is significantly different (p < 0.01) when compared to both other IVTT conditions. Error bares in d represent the s.d. of four measurements of one replicate (i) and the s.d. of 10 independent vesicle line scans (ii,iii). The scale bars in (b,c) correspond to 5 μm. All assays were independently replicated at least twice.

confirms a significantly higher signal of IVTT inside compared to the background signal for the “no IVTT” condition (Figure 7d(i) and Table S6). Vogele et al.62 recently obtained very similar results for IVTT inside ELP-based vesicles constituted from nonprebiotic (VPGEG)n and (VPGFG)n ELP domain derivatives previously described to assemble and to be modified chemically in vivo and in vitro by our group.5,32 This demonstrates the principal feasibility of IVTT inside dynamic PMBCs independent of the exact molecular composition of the membrane building blocks. The DNAencoded synthesis of protein-based amphiphilic membrane constituents and their subsequent insertion into the prebiotic PMBC membrane demonstrates the top-down compatibility of the extant protein synthesis machinery to the proposed prebiotic protocell model. Regarding the PMBC permeability, it can be concluded that small molecules such as Kan and chloramphenicol and PK do not permeate the membrane to a significant amount. Further permeability tests characterizing the protein membrane are envisioned.

enclosed ligase from PK degradation and allows the DNA ligation inside the PMBCs until the membrane becomes permeable for the PK. These results prove our statement that anabolic DNA ligation reactions function inside PMBCs. Therefore, we can conclude that anabolic reactions such as the polymerization of genetic information can be conducted within PMBCs realizing an important step toward an evolvable protocell model. Taking one definition of life as a self sustaining chemical system capable of Darwinian evolution60 the ability for self-replication is crucial for valid protocellular models. In Figure 7 we demonstrate the DNA-encoded synthesis and subsequent incorporation of protein-based membrane constituents through IVTT within the PMBCs as an essential step toward an evolvable system. For this experiment, preassembled and vacuum-dried PMBCs were swollen on glass slides together with the IVTT mix in buffer. The PMBC swelling leads to partial encapsulation of IVTT mix containing the DNA, which encodes the membrane constituent mCherry-H40I30 and leads to its subsequent expression and membrane incorporation. The resulting red membrane fluorescence signal was imaged (Figure 7b) after 5 h of expression and quantified through line plots across the PMBC profile (Figure 7c,d). In order to distinguish between out- and inside expression, three IVTT conditions depending on suppression location (for details see Methods and Materials section) where defined: IVTT outside and inside (no suppression) (I), IVTT inside (suppression outside) (II), and no IVTT (suppression in- and outside) (III). Here, Kan as a strong prokaryotic ribosomal translation inhibitor was applied for IVTT inhibition and was comparable to chloramphenicol-, RNAase-, or DNAse-induced inhibition (data not shown). For condition I (IVTT outside and inside) the strongest red membrane fluorescence, greatest signal to noise- (SmN), and signal to lumen (SmL) ratio (defined as in Figure 7c right panel) can be observed in Figure 7c,d (I) (see Table S5). For the desired scenario (IVTT inside and outer suppression) a significantly smaller SmN- and SmL ratio is measured. This significantly lower red membrane fluorescence (compared to condition I) indicates that mCherry-H40I30 is exclusively expressed inside the PMBCs and gets incorporated into the PMBC membrane from inside. A difference in red membrane fluorescence signal intensity of a single PMBC for condition II is observed in Figure 7b II (multiple PMBCs within the red image). This variation can be associated with variable IVTT mix encapsulation efficiency. High variations of encapsulation efficiencies for the applied swelling method are described in the literature for lipid-based vesicles.17,31,61 The suppression of IVTT outside and inside the PMBCs leads to a weak red auto fluorescent background signal where the red signal intensity of lumen and vesicle membrane are indistinguishable (SmL ratio = 1.1) (line plot of Figure 7c III). This can be attributed to Kan-induced complete suppression of mCherry-H40I30 expression and the lacking membrane incorporation. The photospectrometric quantification of the red fluorescence of the entire sample solution

■

CONCLUSIONS AND SUMMARY Taken together, we present a protocell model composed of an amphiphilic protein membrane based on prebiotic aa that we propose as the origin for the first evolvable entities preceding phospholipid-based compartmentalization. Considering the prevalent environmental conditions on the prebiotic earth, the higher stability of PMBCs compared to pure fatty acid protocell models and their straightforward building block synthesis and accessibility (compared to diacyl lipids) highlights their significance as a promising protocell model. The compatibility and integration of prevalent protocell membrane constituents (phospholipids, fatty acids, and fatty alcohols) into the investigated PMBCs might facilitate a possible transition of prebiotic protein membrane-based protocells toward current phospholipid membrane-based cells. The encapsulation of functional proteins or whole PMBCs into one another illustrates the possibility of subcompartmentalization and spatial separation of reactions within one PMBC (Figure 1). Limitations of the proposed protocell model include the statistical constraint for the probability of spontaneous co-oligomerization of individual blocks and their subsequent coupling of hydrophobic and hydrophilic blocks to yield the prebiotic membrane constituent. In addition, though the longer amphiphilic peptides employed in this work are potentially accessible under prebiotic conditions, the length of the aa sequence (>200 aa) attenuates the prebiotic synthesis propensity. The assembly of short amphipathic peptides into PMBCs was observed, but also the limited stability quest for other short and more efficient assembly motifs or additional stabilizing membrane constituents. The minimum length of amphiphilic proteins and stability properties of assembled PMBCs have to be investigated in detail. Characterizing the minimum length of prebiotic amphiphilic peptides, while preserving the dynamic three-dimensional structure-forming capacity as a starting 9607


Article

Langmuir

model enables new perspectives and provides new insights into the emergence of self-sustaining protocells on the pathway to living systems.

point for a path toward an evolvable entity will substantiate our hypothesis. We could show that prebiotic PMBCs accommodate not only anabolic ligation reactions but most notably DNAencoded synthesis and subsequent incorporation of their membrane constituents as a major step toward evolvable extant systems. Our findings support the hypothesis that prebiotic PMBCs represent a new type of protocell model, enabling the construction of simple artificial cells and reveal plausible structural and catalytic pathways to the emergence of lipidbased compartmentalization. The simplicity and multifunctionality of our amphipathic building units are important prerequisites for their emergence as constituents of first prebiotic protocells. Rather more speculative, prebiotic amphipathic protein membranes might not only play a role as a compartment boundary for selfreplicating RNA molecules or early chemical reaction networks but may also contribute to the emergence of protocells in multiple ways: in extant cells the dynamic multifunctionality of proteins, notably the integral membrane proteins, is indispensable, implying a significant role within the prebiotic protocellular membrane. The plasma membrane of extant cells is composed of roughly equal parts in weight percent of proteins and lipids whereas proteins are crucial for main tasks such as selective permeability, transport, recognition, regulation, and so forth. One scenario could have been the lipids as boundary molecules inserted into PMBCs as a subsequent transition event. A next step for the validation of the PMBC protocell model would be the demonstration of prebiotic PMBCs constituted of autocatalytic amphipathic peptides. Such a self-replicating activity of the compartment-forming unit is crucial for the evolution of two different replicating systems: the informational genome (RNA replicase) and the compartment in which it resides. A self-replicating protocell model based on peptides is considerably more conceivable than the lipid-based systems. Dipeptides such as diglycines,63 histidyl histidine,64 Ser-His65,66 can effectively catalyze their own synthesis or catalyze peptide bond formation. In addition, autocatalytic synthesis of glycine to oligo glycine in a simulated submarine hydrothermal system44 emphasizes the mulitfunctionality of peptides and therefore their potential as selfreplicating compartment constituents. Furthermore, the association and coevolution of an RNA precursor together with positively charged amphiphilic peptides, for example, K40I30 constituting one PMBC membrane is reasonable as Koga et al. have shown that poly(Lys) peptides associate with nucleotides to functional micro droplets.25 In addition, Bergstrom et al. reported delivery of chemical reactivity to RNA and DNA by a specific short peptide (AAKK)4.67 Accordingly, the proposed PMBC protocell model represents an important step toward an evolvable, self-replicating protocell compartment. In summary, a dynamic protocell membrane constituted of simple prebiotic amphipathic peptide molecules possesses major advantages when compared to a protocell enclosed by lipid or lipid-precursor membranes. First, the synthesis of peptide membrane constituents65,68 is simpler; second, the membrane itself can potentially exhibit autocatalytic and catalytic activity67,69−71 and potentially self-replicate under simple conditions. Third, an associated coevolution of the informational genome (e.g., RNA), functional catalytical pathways (proteins), and the structural entity of the PMBCs is conceivable. We believe that the presented PMBC protocell

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.9b00445. PMBCs assembled from short amphiphilic BDP-HisH10L4 in 10 mM Tris-HCL pH 8 containing 10% v/v 1octanol (AVI) PMBCs assembled from very short amphiphilic BDPHis-H5L4 in 10 mM Tris-HCL pH 8 containing 10% v/ v 1-octanol (AVI) PMBCs assembled from BDP-His-K40130 (f.c. of 16 μM) via swelling of a dried protein film in 10 mM TrisHCL pH 8 (AVI) Specialized cloning and expression vectors; amphiphilic protein library for PMBC assembly; ligation inside PMBCs; line plot data for at least 10 PMBCs under different IVTT conditions; photospectrometric fluorescence measurement of IVTT; stoichiometry for the in vitro translation and transcription inside PMBCs; extinction coefficient of fluorophores used for concentration determination of amphiphilic proteins; brightfield and fluorescence images of PMBCs assembled from short amphiphilic proteins; biorthogonal labeling of prebiotic pAzF-His-K40130 and pAzFHis-K20130 with green and red fluorescence dyes; PMBCs assembled from His-EYFPH20130 preserve their vesicular integrity and size distribution for at least 9 months; stability test of BDP-His-K40I30 PMBC; fluorescence and brightfield images for PMBCs using simple fatty acids; epifluorescence images demonstrating the incompatibility of PMBC membranes assembled from proteins with different amphipathic ratios; and epifluorescence images demonstrating the compatibility of and internalization of liposomes into prebiotic PMBCs; T4-DNA ligation inside assembled PMBCs; Mascot results for the LC-MS/MS analysis of the tryptic peptides from the corresponding ‘prebiotic’ amphiphilic proteins; Rational design and composition of amphiphilic protein domains (PDF) PMBCs assembled from His-EYFP-H20130 preserve vesicular integrity and size distribution for at least 9 months (AVI) PMBCs assembled from HGTV-H40130 constituents encapsulating various protein aggregates and smaller PMBCs of different sizes (AVI) Fusion of PMBCs assembled from BDP-His-K40130 in 10 mM Tris-HCL pH 8 containing 10% v/v n-butanol (AVI)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (A.S.). *E-mail: [email protected] (M.C.H.). *E-mail: [email protected]. Fax: +49761 203 8456 (S.M.S.). ORCID

Andreas Schreiber: 0000-0001-8046-1357 9608


Article

Langmuir Author Contributions

(14) McCollom, T. M.; Ritter, G.; Simoneit, B. R. T. Lipid Synthesis Under Hydrothermal Conditions by Fischer- Tropsch-Type Reactions. Origins Life Evol. Biospheres 1999, 29, 153−166. (15) Deamer, D. W.; Pashley, R. M. Amphiphilic Components of the Murchison Carbonaceous Chondrite: Surface Properties and Membrane Formation. Origins Life Evol. Biospheres 1989, 19, 21−38. (16) Hargreaves, W. R.; Mulvihill, S. J.; Deamer, D. W. Synthesis of Phospholipids and Membranes in Prebiotic Conditions. Nature 1977, 266, 78−80. (17) Walde, P.; Ichikawa, S. Enzymes inside Lipid Vesicles: Preparation, Reactivity and Applications. Biomol. Eng. 2001, 18, 143−177. (18) Walde, P.; Goto, A.; Monnard, P.-A.; Wessicken, M.; Luisi, P. L. Oparin’s Reactions Revisited: Enzymic Synthesis of Poly(Adenylic Acid) in Micelles and Self-Reproducing Vesicles. J. Am. Chem. Soc. 1994, 116, 7541−7547. (19) Oberholzer, T.; Wick, R.; Luisi, P. L.; Biebricher, C. K. Enzymatic RNA Replication in Self-Reproducing Vesicles: An Approach to a Minimal Cell. Biochem. Biophys. Res. Commun. 1995, 207, 250−257. (20) Nomura, S.-i. M.; Tsumoto, K.; Hamada, T.; Akiyoshi, K.; Nakatani, Y.; Yoshikawa, K. Gene Expression within Cell-Sized Lipid Vesicles. ChemBioChem 2003, 4, 1172−1175. (21) Oberholzer, T.; Albrizio, M.; Luisi, P. L. Polymerase Chain Reaction in Liposomes. Chem. Biol. 1995, 2, 677−682. (22) Noireaux, V.; Libchaber, A. A Vesicle Bioreactor as a Step toward an Artificial Cell Assembly. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 17669−17674. (23) Fox, S. W. The Evolutionary Significance of Phase-Separated Microsystems. Orig. Life 1976, 7, 49−68. (24) Santoso, S.; Hwang, W.; Hartman, H.; Zhang, S. Self-Assembly of Surfactant-like Peptides with Variable Glycine Tails to Form Nanotubes and Nanovesicles. Nano Lett. 2002, 2, 687−691. (25) Koga, S.; Williams, D. S.; Perriman, A. W.; Mann, S. Peptide− nucleotide Microdroplets as a Step towards a Membrane-Free Protocell Model. Nat. Chem. 2011, 3, 720−724. (26) Baeza, I.; Ibáñez, M.; Wong, C.; Chávez, P.; Gariglio, P.; Oró, J. Possible Prebiotic Significance of Polyamines in the Condensation, Protection, Encapsulation, and Biological Properties of DNA. Origins Life Evol. Biospheres 1992, 21, 225−242. (27) Longo, L. M.; Lee, J.; Blaber, M. Simplified Protein Design Biased for Prebiotic Amino Acids Yields a Foldable, Halophilic Protein. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, 2135−2139. (28) Shevchenko, A.; Tomas, H.; Havli, J.; Olsen, J. V.; Mann, M. In-Gel Digestion for Mass Spectrometric Characterization of Proteins and Proteomes. Nat. Protoc. 2006, 1, 2856−2860. (29) Wang, L.; Brock, A.; Herberich, B.; Schultz, P. G. Expanding the Genetic Code of Escherichia Coli. Science 2001, 292, 498−500. (30) Varga, B. R.; Kállay, M.; Hegyi, K.; Béni, S.; Kele, P. A NonFluorinated Monobenzocyclooctyne for Rapid Copper-Free Click Reactions. Chem.Eur. J. 2012, 18, 822−828. (31) Spencer, A. C.; Torre, P.; Mansy, S. S. The Encapsulation of Cell-Free Transcription and Translation Machinery in Vesicles for the Construction of Cellular Mimics. J. Vis. Exp. 2013, 80, No. e51304. (32) Huber, M. C.; Schreiber, A.; Wild, W.; Benz, K.; Schiller, S. M. Introducing a Combinatorial DNA-Toolbox Platform Constituting Defined Protein-Based Biohybrid-Materials. Biomaterials 2014, 35, 8767−8779. (33) Zaia, D. A. M.; Zaia, C. T. B. V.; De Santana, H. Which Amino Acids Should Be Used in Prebiotic Chemistry Studies? Origins Life Evol. Biospheres 2008, 38, 469−488. (34) Shen, C.; Yang, L.; Miller, S. L.; Oró, J. Prebiotic Synthesis of Histidine. J. Mol. Evol. 1990, 31, 167−174. (35) Shen, C.; Yang, L.; Miller, S. L.; Oró, J. Prebiotic Synthesis of Imidazole-4-Acetaldehyde and Histidine. Orig. Life Evol. Biosph. 1987, 17, 295−305. (36) Rode, B. M.; Son, H. L.; Suwannachot, Y.; Bujdak, J. The Combination of Salt Induced Peptide Formation Reaction and Clay

A.S. and M.C.H. contributed equally to this work. A.S., M.C.H., and S.M.S. conceived the project. A.S. and M.C.H. designed and performed the experiments. M.C.H. designed and cloned the used constructs. M.C.H. and A.S. conceptualized the publication, A.S. wrote the initial draft and the final publication together with M.C.H. S.M.S commented on and discussed the paper. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank the BMBF for financial support and the Zentrum für Biosystemanalyse (ZBSA) for providing the research facility. The authors thank the BMBF for financial support: BMBF-Programm Ideenwettbewerb “Neue Produkte für die Bioökonomie”: “IBÖ -01: Entwicklung einer universellen/kontrollierbaren Produktionsorganimus-Platform”; Förderkennzeichen 031A490; BMBF Forschungspreis “Nächste Generation biotechnologischer Verfahren” 2014: “Universell modularer Produktionsorganismus”; Fö r derkennzeichen: 031A550. We are grateful to the EM facility of the Department of Neuroanatomy Freiburg and thank Barbara Joch and Prof. Andreas Vlachos for help with TEM imaging. We thank the lab of Peter Kele for synthesizing and providing Combo Rhodamin. We thank the lab of Günter Roth for providing the IVTT mix. We are grateful to P. G. Schultz, TSRI, La Jolla, California, USA, for providing the plasmid pEVOL pAzF.

■

REFERENCES

(1) Szostak, J. W.; Bartel, D. P.; Luisi, P. L. Synthesizing Life. Nature 2001, 409, 387−390. (2) Miller, S. L. A Production of Amino Acids Under Possible Primitive Earth Conditions. Science 1953, 117, 528−529. (3) Fox, S. W.; Harada, K.; Kendrick, J. Production of Spherules from Synthetic Proteinoid and Hot Water. Science 1959, 129, 1221− 1223. (4) Brooke, S.; Fox, S. W. Compartmentalization in Proteinoid Microspheres. Biosystems 1977, 9, 1−22. (5) Huber, M. C.; Schreiber, A.; von Olshausen, P.; Varga, B. R.; Kretz, O.; Joch, B.; Barnert, S.; Schubert, R.; Eimer, S.; Kele, P.; et al. Designer Amphiphilic Proteins as Building Blocks for the Intracellular Formation of Organelle-like Compartments. Nat. Mater. 2015, 14, 125−132. (6) Chen, I. A.; Walde, P. From Self-Assembled Vesicles to Protocells. Cold Spring Harbor Perspect. Biol. 2010, 2, a002170. (7) Luisi, P. L. The Emergence of Life: From Chemical Origins to Synthetic Biology; Cambridge University Press: Cambridge, 2006. (8) Deamer, D. The Role of Lipid Membranes in Life’s Origin. Life 2017, 7, 5. (9) Apel, C. L.; Deamer, D. W.; Mautner, M. N. Self-Assembled Vesicles of Monocarboxylic Acids and Alcohols: Conditions for Stability and for the Encapsulation of Biopolymers. Biochim. Biophys. Acta, Biomembr. 2002, 1559, 1−9. (10) Rasmussen, S.; Chen, L.; Nilsson, M.; Abe, S. Bridging Nonliving and Living Matter. Artif. Life 2003, 9, 269−316. (11) Meierhenrich, U. J.; Filippi, J.-J.; Meinert, C.; Vierling, P.; Dworkin, J. P. On the Origin of Primitive Cells: From Nutrient Intake to Elongation of Encapsulated Nucleotides. Angew. Chem., Int. Ed. 2010, 49, 3738−3750. (12) Mansy, S. S.; Schrum, J. P.; Krishnamurthy, M.; Tobé, S.; Treco, D. A.; Szostak, J. W. Template-Directed Synthesis of a Genetic Polymer in a Model Protocell. Nature 2008, 454, 122−125. (13) Namani, T.; Deamer, D. W. Stability of Model Membranes in Extreme Environments. Origins Life Evol. Biospheres 2008, 38, 329− 341. 9609


Article

Langmuir

(60) Joyce, G. F.; Deamer, D. W.; Fleischaker, G. Origins of Life: The Central Concepts. Foreword; Jones and Bartlett: Boston, 1994. (61) Altamura, E.; Carrara, P.; D’Angelo, F.; Mavelli, F.; Stano, P. Extrinsic Stochastic Factors (Solute Partition) in Gene Expression inside Lipid Vesicles and Lipid-Stabilized Water-in-Oil Droplets: A Review. Synth. Biol. 2018, 3, ysy011. (62) Vogele, K.; Frank, T.; Gasser, L.; Goetzfried, M. A.; Hackl, M. W.; Sieber, S. A.; Simmel, F. C.; Pirzer, T. Towards Synthetic Cells Using Peptide-Based Reaction Compartments. Nat. Commun. 2018, 9, 3862. (63) Plankensteiner, K.; Righi, A.; Rode, B. M. Glycine and Diglycine as Possible Catalytic Factors in the Prebiotic Evolution of Peptides. Origins Life Evol. Biospheres 2002, 32, 225−236. (64) White, D. H.; Erickson, J. C. Catalysis of Peptide Bond Formation by Histidyl-Histidine in a Fluctuating Clay Environment. J. Mol. Evol. 1980, 16, 279−290. (65) Gorlero, M.; Wieczorek, R.; Adamala, K.; Giorgi, A.; Schininà, M. E.; Stano, P.; Luisi, P. L. Ser-His Catalyses the Formation of Peptides and PNAs. FEBS Lett. 2009, 583, 153−156. (66) Adamala, K.; Szostak, J. W. Competition between Model Protocells Driven by an Encapsulated Catalyst. Nat. Chem. 2013, 5, 495−501. (67) Bergstrom, R. C.; Mayfield, L. D.; Corey, D. R. A Bridge between the RNA and Protein Worlds?: Accelerating Delivery of Chemical Reactivity to RNA and DNA by a Specific Short Peptide (AAKK)4. Chem. Biol. 2001, 8, 199−205. (68) Plankensteiner, K.; Reiner, H.; Rode, B. M. Prebiotic Chemistry: The Amino Acid and Peptide World. Curr. Org. Chem. 2005, 9, 1107. (69) Kochavi, E.; Bar-Nun, A.; Fleminger, G. Substrate-Directed Formation of Small Biocatalysts under Prebiotic Conditions. J. Mol. Evol. 1997, 45, 342−351. (70) Shibue, R.; Sasamoto, T.; Shimada, M.; Zhang, B.; Yamagishi, A.; Akanuma, S. Comprehensive Reduction of Amino Acid Set in a Protein Suggests the Importance of Prebiotic Amino Acids for Stable Proteins. Sci. Re p. 2018, 8, 1227. (71) Wieczorek, R.; Adamala, K.; Gasperi, T.; Polticelli, F.; Stano, P. Small and Random Peptides: An Unexplored Reservoir of Potentially Functional Primitive Organocatalysts. The Case of Seryl-Histidine. Life 2017, 7, 19.

Catalysis: A Way to Higher Peptides under Primitive Earth Conditions. Origins Life Evol. Biospheres 1999, 29, 273−286. (37) Georgelin, T.; Jaber, M.; Bazzi, H.; Lambert, J.-F. Formation of Activated Biomolecules by Condensation on Mineral SurfacesA Comparison of Peptide Bond Formation and Phosphate Condensation. Life Evol. Biospheres 2013, 43, 429−443. (38) Imai, E.; Honda, H.; Hatori, K.; Brack, A.; Matsuno, K. Elongation of Oligopeptides in a Simulated Submarine Hydrothermal System. Science 1999, 283, 831−833. (39) Griffith, E. C.; Vaida, V. In Situ Observation of Peptide Bond Formation at the Water−air Interface. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 15697−15701. (40) Schwendinger, M. G.; Rode, B. M. Possible Role of Copper and Sodium Chloride in Prebiotic Evolution of Peptides. Anal. Sci. 1989, 5, 411−414. (41) Bujdak, J.; Slosiarikova, H.; Texler, N.; Schwendinger, M.; Rode, B. M. On the Possible Role of Montmorillonites in Prebiotic Peptide Formation. Monatshefte Chem. 1994, 125, 1033−1039. (42) Rodriguez-Garcia, M.; Surman, A. J.; Cooper, G. J. T.; SuárezMarina, I.; Hosni, Z.; Lee, M. P.; Cronin, L. Formation of Oligopeptides in High Yield under Simple Programmable Conditions. Nat. Commun. 2015, 6, 8385. (43) Gorlero, M.; Wieczorek, R.; Adamala, K.; Giorgi, A.; Schininà, M. E.; Stano, P.; Luisi, P. L. Ser-His Catalyses the Formation of Peptides and PNAs. FEBS Lett. 2009, 583, 153−156. (44) Imai, E.-i.; Honda, H.; Hatori, K.; Matsuno, K. Autocatalytic Synthesis of Oligoglycine in a Simulated Submarine Hydrothermal System. Origins Life Evol. Biospheres 1999, 29, 249−259. (45) Rushdi, A. I.; Simoneit, B. R. T. Lipid Formation by Aqueous Fischer-Tropsch-Type Synthesis over a Temperature Range of 100 to 400 °C. Origins Life Evol. Biospheres 2001, 31, 103−118. (46) Walde, P. Surfactant Assemblies and Their Various Possible Roles for the Origin(S) of Life. Origins Life Evol. Biospheres 2006, 36, 109−150. (47) Deamer, D. W. Boundary Structures Are Formed by Organic Components of the Murchison Carbonaceous Chondrite. Nature 1985, 317, 792−794. (48) Studier, M. H.; Hayatsu, R.; Anders, E. Organic Compounds in Carbonaceous Chondrites. Science 1965, 149, 1455−1459. (49) Jacob, N. Intermolecular and Surface Forces, 3rd ed.; Elsevier Science: Waltham, MA, 2011. (50) Fellermann, H.; Tanaka, S.; Rasmussen, S. Sequence Selection by Dynamical Symmetry Breaking in an Autocatalytic Binary Polymer Model. Phys. Rev. E 2017, 96, 062407. (51) Maury, C. P. J. Amyloid and the Origin of Life: Self-Replicating Catalytic Amyloids as Prebiotic Informational and Protometabolic Entities. Cell. Mol. Life Sci. 2018, 75, 1499−1507. (52) Schreiber, A.; Stühn, L. G.; Huber, M. C.; Geissinger, S. E.; Rao, A.; Schiller, S. M. Self-Assembly Toolbox of Tailored Supramolecular Architectures Based on an Amphiphilic Protein Library. Small 2019, 1900163. (53) Urry, D. W.; Hugel, T.; Seitz, M.; Gaub, H. E.; Sheiba, L.; Dea, J.; Xu, J.; Parker, T. Elastin: A Representative Ideal Protein Elastomer. Philos. Trans. R. Soc., B 2002, 357, 169−184. (54) Walde, P.; Wick, R.; Fresta, M.; Mangone, A.; Luisi, P. L. Autopoietic Self-Reproduction of Fatty Acid Vesicles. J. Am. Chem. Soc. 1994, 116, 11649−11654. (55) Chen, I. A.; Szostak, J. W. A Kinetic Study of the Growth of Fatty Acid Vesicles. Biophys. J. 2004, 87, 988−998. (56) Szostak, J. W. An Optimal Degree of Physical and Chemical Heterogeneity for the Origin of Life? Philos. Trans. R. Soc., B 2011, 366, 2894−2901. (57) Kozlov, M. M.; Chernomordik, L. V. Membrane Tension and Membrane Fusion. Curr. Opin. Struct. Biol. 2015, 33, 61−67. (58) Sutton, R. B.; Fasshauer, D.; Jahn, R.; Brunger, A. T. Crystal Structure of a SNARE Complex Involved in Synaptic Exocytosis at 2.4 Å Resolution. Nature 1998, 395, 347−353. (59) Alberti, S. Phase Separation in Biology. Curr. Biol. 2017, 27, R1097−R1102. 9610


Prebiotic Protocell Model Based on Dynamic ... - ACS Publications

Recommend Documents