Steps towards de-novo life
Monreal Santiago, Guillermo
DOI:10.33612/diss.121581426
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Monreal Santiago, G. (2020). Steps towards de-novo life: compartmentalization and feedback mechanisms in synthetic self-replicating systems. University of Groningen. https://doi.org/10.33612/diss.121581426
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Complex coacervation as a
compartmentalization strategy
for synthetic self-replicators
In this chapter, complex coacervates are used to compartmentalize the peptide-based self-replicator shown in Chapter 1, and its precursors. We discuss both the possibility of using self-replicating fibres as part of coacervates, and their partitioning into preformed coacervate droplets. Next, we analyse the spatial distribution, selective diffusion, and self-replication of encapsulated peptide-based macrocycles, by using liquid chromatography and confocal microscopy. Our observations show that the crowded, charge-rich environ-ment inside of the coacervates has a strong effect on the library composition, preventing the growth of hexamer (the main component of the library in bulk aqueous solution) and allowing for the emergence of another replicator, based on trimers of the same building block. We finish the Chapter by providing an outlook on some strategies to continue this research, by coupling self-replication to coacervate formation.
3.1
Introduction
Compartmentalization is one of the key problems in developing de-novo life [1– 3], and it must have been a crucial step in the development of prebiotic organ-isms [4–6]. In order to synthesize biomolecules and ensure their activity, it is critical to mantain a local high concentration, and compartmentalization is a sim-ple way to achieve that [7–9]. In the early stages of life, before organisms had the necessary machinery to synthesize molecules such as phospholipids, simpler compartments must have performed that function. A common hypothesis (unap-pealingly named the garbage-bag world) in the prebiotic life community states that coacervates, charged droplets made from relatively simple molecules, could have played that role [10–12]. Whether complex coacervation took place in the first life forms or not, it is an important process in the current ones, as it is involved in the concentration, segregation, and transport of biomolecules [13, 14].
Complex coacervates are produced by the interaction between positively and
negatively charged molecules (typically polymers) in water1. In the right
con-ditions and with the adequate pair of molecules [17], phase separation can take place, leading to a new liquid phase instead of a precipitate [15, 18, 19]. This
phase contains both ions and water, and is referred to as the coacervate phase.2
This phase behaves like a viscous liquid that is immiscible with water (despite being water-based itself): it can flow, remain as a layer in contact with the di-luted aqueous phase, or be dispersed into a colloidal suspension of droplets that eventually coalesce into a single layer again. Coacervates typically have both a higher concentration of charges and a lower electric permeability than the aque-ous solution in equilibrium with them. As a consequence, both hydrophilic and hydrophobic molecules tend to be more soluble in the coacervate phase than in the surrounding solution, and partition into it [18, 20] with distribution constants
higher than 103[21].
The properties described above, together with the similarity between the coa-cervate phase and the crowded, protein-rich environment of cells, have led to a growing interest in their use as protocell models [18, 21–26]. Proteins [27, 28] and nucleic acids [20, 28, 29] have been succesfully compartmentalised in coacervates without loss of 3D structure, and it has been shown that it is possible to have transcription and enzymatic catalysis in them [23, 28]. Their responsiveness to salt concentration and other stimuli [22, 30] has been used to design complex be-haviours, and to synthesise them in out-of-equilibrium conditions. Despite their potential as protocells, self-replication or autocatalysis has still not been shown inside of them.
1There are exceptions to this: Coacervates formed from a single type of molecule [15] (known as
simple coacervates), or even from molecules with the same charge [16] have also been reported.
2The terminology in this field is not completely established and varies from publication to
publi-cation. In this Thesis, we refer to the dense phase containing the charged molecules as coacervate phase, and to the droplets formed upon its dispersion as coacervate droplets, or simply coacervates.
We considered coacervates to be a very attractive model for the compartmen-talization of our self-replicating fibres (formed from hexamers of the building block XGLKFK, 1). The peptide sequence of this building block includes two lysines, that together with the -COOH terminus add up to one positive charge
per 1 unit. This amounts to six per hexamer, and 600 in a 50 nm fibre3, which
should be enough charge density to allow for their partition into coacervates. Pre-vious reports of coacervates mantaining the secondary structure of encapsulated proteins, and even enhancing their self-assembly in some cases [32], encouraged us even more to study the replication of these fibres in them.
N H O H N O N H O NH3 H N O N H O NH3 O O SH HS Partition and selective difussion Self-replication 1 13 16 13/14
3.2
Preliminary results
As shown in the introduction, the chains of 1 in 1n macrocycles are positively
charged at pHs close to 8. Due to this, fibres of 16 can be considered
supramo-lecular polycations with six charges per macrocycle, while smaller macrocycles have considerably less charges. In general, molecules with a higher number of charges lead to more stable coacervates [20, 30]. Therefore, with the right oligo- or polyanion, this difference could lead to the formation of coacervates only in
pres-ence of the replicator 16 but not in presence of its precursors. The initial goal of
this project was to find such anion, and obtain a system where the self-replication
of 16would be directly coupled to the formation of coacervates.
Initially, ATP was used as a candidate for this purpose. Upon mixing stock
solutions of ATP and 16at the ratio of charge neutralization the mixture became
turbid, indicating polyelectrolyte complexation. However, it quickly became evi-dent that the formed structures were not coacervates, but fibrous aggregates that could be observed by optical microscopy. (Figure 3.1, left). As a control, poly-lysine was mixed with ATP in the same conditions, leading to the formation of coacervates (Figure 3.1, right). We attribute this difference in behaviour to the
rigidity of the 16 fibres, which facilitates their packing into hydrogel-like
struc-tures upon shielding of their surface charges. On the other hand, poly-lysine is much more flexible and less prone to packing into an ordered microstructure. For this reason, we assume that coacervate formation becomes the most stable option for this polycation.
(a) Fibrous structures formed from ATP + 16 (b) Coacervates formed from ATP + pLys
Figure 3.1: Optical microscopy images of the structures formed upon mixing ATP and either 16fibres (a) or poly-lysine (b) [21] in borate buffer (pH = 8.2, 50 mM, See Methods).
The concentration of ATP was 1.0 mM, and the concentration of either 16 or poly-lysine
Following this, in order to obtain coacervates using 16and anions, we needed
to increase the disorder of their microstructures. Different approaches were tested for this purpose: using shorter replicator fibres, screening charge interactions with NaCl, changing ATP for larger and more disordered polyanions, and adding chaotropes [33] to destabilize the formed structures (Table S3.1). All these strate-gies led to either amorphous aggregates or the complete solubilization of all struc-tures, without us finding any set of conditions where complex coacervation was achieved.
After some exploration of the parameter space, without obtaining encourag-ing results, we decided to change the goal of the project to the compartmental-ization of replicators into coacervate droplets composed of other polyions. Our
aim was to establish whether the 16system was compatible with coacervates, and
to study the effect that compartmentalization had on it. Our long-term goal, to couple replication and compartment formation, could be achieved in a later stage through a different strategy (see Further experiments).
3.3
Results and discussion
As shown above, when 16 macrocycles were the main polycation in a mixture,
their interaction with polyanions led to the precipitation of the resulting com-plex. However, we expected that if a second polycation was present in a much higher concentration, the phase behaviour of the resulting polyelectrolyte com-plex would be dictated by this more concentrated polymer, and the final
struc-tures would not be affected by the presence of 16. We studied this by mixing 16
with a pair of polyions known in literature to form coacervates, sodium poly-acrylate (pAA) and poly-diallyldimethylammoniumchloride (pDADMAC) [34] (Figure 3.2a). Coacervates were succesfully prepared in these conditions, in
pres-ence of 16(Figure 3.2b) or other 1nmacrocycles (Figure S3.1), and remained stable
for hours until they coalesced into a separate phase. In order to decrease the ex-tent of this coalescence, we tried to minimise the ionic strength of the prepared dispersions. To do this, we first minimized the concentration of salts and buffers used during their preparation. Second, we used centrifugation and redispersion to remove the solution in equilibrium with the coacervate phase and substitute
it for ion-free ddH2O (See Methods). The redispersed coacervates had a longer
lifetime as droplets (in the order of days instead of hours), although there was a high variability from sample to sample.
We initially studied the compartmentalization of 1nmacrocycles in the formed
coacervates by fluorescence microscopy, using thioflavin T (ThT) as a fluorophore. Thioflavin T is a β-sheet - sensitive probe [35] that becomes fluorescent in presence
of amyloid-like structures. It has been shown that 16replicators, and to a smaller
extent other 1n macrocycles, are also able to trigger its fluorescence [36]. When
coacervates were prepared in presence of ThT and either 16or 13/14, microscopy
showed a much higher fluorescence inside of the coacervate droplets than in the background (Figures 3.2b and S3.1). As expected, droplets prepared in presence
of ThT but without 1nmacrocycles showed almost no fluorescence (Figure S3.3a).
These observations suggested that the macrocycles partitioned into the droplets, consistent with a high affinity between the charged macrocycles and the charge-rich environment of the coacervate phase.
Interestingly, while the fluorescence in the case of 13/14 was homogeneous
throughout the droplets, the 16fibres seemed to aggregate inside of them,
form-ing patterns of different fluorescence intensities (Figure 3.2b). Besides that, the distribution of the macrocycles appeared to be similar from one droplet to an-other, as long as the macrocycles were present in solution before coacervate
for-mation. However, when 16was added to a preformed dispersion of coacervates,
its distribution was irregular, suggesting slow diffusion in, out and through the coacervate phase (Figure S3.2).
To study the partition of 1nmacrocycles quantitatively, samples of coacervates
cen-N Cl n O O Na S N N Cl n pDADMAC pAA ThT
(a) Compounds used for coacervate formation and imaging.
(b) ThT fluorescence of coacervates prepared in presence of 16. T h T 1 1 6 1 3/1 4 0 , 0 0 , 5 1 , 0 N o rm a liz e d C s u p e rn a ta n t
(c) Depletion of the supernatant in presence of coacervates.
Component KD(ccoa/csup)
ThT (2 ± 7)*100 1 (2.2 ± 0.6)*102 16 (2.8 ± 0.8)*105 13/14 (2.5 ± 0.5)*105
(d) Partition constants into coacervates.
Figure 3.2: (a) Structures of the compounds used in this section. (b) Fluorescence mi-croscopy image of pAA/pDADMC coacervates (50 mM each, in monomer units) prepared in presence of ThT (80 µM) and 16(40 µM). (c) Concentration of different species in the
su-pernatant in presence of coacervates, normalized against controls without them. The total concentrations were 50 mM for pAA and pDADMAC (in monomer units), 80 µM for ThT, 200 µM for 1 and 400 µM for either 16or 13/14. (d) Distribution constants calculated from
the data in (c). See Methods for the calculation of errors.
trifuged to separate the coacervate phase from the bulk solution. The concentra-tion of the different species in the supernatant was then studied by UPLC, com-paring it with control samples that did not contain coacervates (Figure 3.2c). We
observed that, for 16and 13/14, the supernatant was completely depleted in
pres-ence of coacervates. This was extremely remarkable, specially considering the small volume of the coacervate phase (1.5 ± 0.2 % of the total solution). A dis-tribution constant was then calculated for each of the components (Figure 3.2d),
showing particularly high values for 16 and 13/14.4 These results, as reported
previously for smaller molecules [23], seem to follow a trend where entities with more charges are partitioned to a larger extent into the coacervate phase. As an estimate of the magnitude of this compartmentalization: in a typical sample
con-taining 16with a total concentration of 80 µM, its concentration in the coacervate
phase would be higher than 5 mM.
The fact that the partition of ThT into the coacervates was minimal encour-aged us to continue using this dye as a probe. Since ThT was present in similar concentrations inside and outside of the coacervates, the intensity of its fluores-cence should depend only on the concentration of 1 derivatives. We confirmed this by confocal microscopy, studying images of coacervates with increasing
con-centrations of either 16or 13/14. As shown in Figure S3.3, in the range of
concen-trations studied the fluorescence intensity was directly proportional to the
con-centration of 1nin the droplets. It should be noted that the average fluorescence
intensity did not change upon increasing the concentration of ThT (Figure S3.3d),
confirming that its concentration was high enough to bind all 1nmacrocycles.
Next, we studied the exchange of 1nmacrocycles between different droplets.
The partition of solutes into the coacervate phase is a well-established pheno-menon, but there is still some controversy in the literature about whether the diffusion of such solutes in and out of the droplets is a slow (longer than days) or fast (shorter than minutes) process, with a large variation between very similar systems [37, 38]. The rate at which this exchange takes place is critical for the syn-thesis of de-novo life: if it happens too fast, the molecules from different droplets can diffuse together and spatial separation is impossible. On the other hand, if the process is too slow, the system cannot show behaviours that require interac-tions between different droplets (such as cooperation, parasitism or competition) and is "only" a collection of independent microreactors running in parallel. In order to study exchange in our system, we prepared populations of coacervates
containing ThT and either 16, 13/14, or nothing else. We then mixed the "empty"
coacervates with the droplets containing macrocycles, resulting in a dispersion where each coacervate contained either a high concentration of macrocycles or no macrocycles at all. In this case, instead of redispersing the droplets as in pre-vious experiments, we deposited a layer of the dispersion on a glass slide. We made this layer thin and diluted so, practically, no droplets would be touching each other. Therefore, the only possible mechanism for exchange of material be-tween them was diffusion.
This sample was then studied by confocal microscopy, measuring the average fluorescence of a number of droplets and studying the change of the resulting
dis-4In order to confirm that these values were not artificially high due to a kinetic trap, we prepared
controls adding 13/14or 16to the supernatant after the coacervates had been formed and centrifuged.
These samples reached the same final composition as the ones in which the macrocycles had been
added first. Therefore, we can conclude that the two phases were equilibrated, and the KDvalues
0 1 0 2 0 n = 1 9 1 N u m b e r o f d ro p le ts 0 h a v g = 1 0 2 1 0 1 0 2 0 2 h n = 1 7 3 n = 1 8 2 a v g = 1 2 4 0 0 1 0 2 0 3 h n = 1 5 5 a v g = 1 0 5 8 0 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 5 0 0 0 0 1 0 2 0 A v e r a g e f l u o r e s c e n c e i n t e n s i t y ( c o u n t s / p x ) 2 4 h a v g = 1 2 1 9 (a) Diffusion of 16 0 1 0 2 0 0 h n = 1 2 1 a v g = 9 1 8 n = 1 5 6 a v g = 5 2 1 n = 5 8 a v g = 8 2 3 n = 7 4 a v g = 3 1 6 0 2 5 5 0 2 h 0 7 1 4 3 h 0 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 5 0 0 0 0 1 5 3 0 A v e r a g e f l u o r e s c e n c e i n t e n s i t y ( c o u n t s / p x ) 2 4 h (b) Diffusion of 13/14
Figure 3.3: Exchange of 1nmacrocycles between coacervates. The histograms represent the
distribution of average fluorescence intensities in coacervates containing ThT and either
16(left) or 13/14(right) after mixing them with coacervates containing only ThT. The
con-centrations used for the preparation of the coacervates were 40 µM for 1nmacrocycles and
ThT, and 25 mM for pAA and pDADAMAC. Immediately after mixing the coacervates containing 1nand ThT with the ones containing only ThT, an aliquot of each sample was
diluted 4 times, placed in a covered glass slide at room temperature and imaged over time using confocal microscopy. Only the coacervates with an average fluorescence intensity higher than a threshold of 50 counts/px are shown.
tribution over time (Figure 3.3). Our aim was to conclude from this observation whether diffusion between droplets was taking place: if the macrocycles were being exchanged, the distribution would shift to lower values as their
concentra-tion would average out between the "full" and "empty" droplets. In the case of 16,
this was not the case, as the distribution did not change significantly over time
(Figure 3.3a). However, for 13/14 we observed that the distribution was slowly
changing, as both the number of droplets with high concentration and the aver-age fluorescence were decreasing (Figure 3.3b). A Spearman rank-order
correla-tion was run to confirm these relacorrela-tionships. In this statistical analysis, rs values
values indicate a correlation of the same sign (stronger the closer that rs is to +1
or -1). We found no correlation between time and fluorescence for 16(rs= 0.060,
p = 0.11), and a negative correlation between time and fluorescence for 13/14(rs
= -0.476, p < 0.0001). This is a clear indication that, while 16fibres remain trapped
in coacervate droplets, small macrocycles can be exchanged between them. This opens the door to the preparation of systems with different "species" of replica-tors, which could coexist in independent droplets while having an influence on each other through the exchange of small macrocycles - in example competing for them.
The results so far focus on the individual components of the 1nsystem.
How-ever, this study would not be complete without addressing the dynamics of the system and how different macrocycles convert into one another. We used UPLC analysis for this purpose, measuring over time the composition of libraries of 1 prepared in presence of coacervates. This allowed us to study the relative concen-trations of the different components of the system, however it had a limitation: In the conditions of this experiment, the coacervates slowly sediment towards the bottom of the vial, which results in a constant loss of peak area in UPLC that can be partially recovered upon sonication. Since oxidized species partition com-pletely into the coacervate phase, the peak area of all of them decrease to the same extent through this process, and we can expect that their relative amounts, as observed by UPLC, correspond to the overall ones. However, 1 is significantly present both inside and outside of coacervates (Figure 3.2d), so the observed ratio
between it and 1nchanges as sedimentation progresses (For a clear example, see
Figure S3.4a). For this reason, the percentage of 1 has been omitted for most of the calculations in this study. The graphs where it is plotted must only be interpreted qualitatively.
Two unexpected behaviours were observed in 1n libraries inside of
coacer-vates: An accelerated oxidation of thiols to disulfides, and a change in the final
li-brary composition, going from almost exclusively 16to almost exclusively 13
(Fig-ures 3.4a and 3.4b). Controls in presence of both polyions (Fig(Fig-ures 3.4d and 3.4c) showed that the accelerated oxidation was also taking place in presence of pDAD-MAC alone. To investigate the nature of this process, we prepared a sample in presence of coacervates and, after it had been completely oxidized, we reduced it partially using TCEP. After the reduction was completed, the rate of oxidation was not enhanced anymore (Figure S3.4a). These observations would fit a sto-chiometric reaction between 1 and an impurity in the pDADMAC stock solution. We initially suspected of tert-butylperoxide (a common initiator for pDADMAC polymerization [39]), but peroxides were not detected in the pDADMAC stock solution using a standard test. Further research will need to be done in order to identify the impurity or phenomenon responsible of this behaviour, and discern if other polycations will have the same effect too.
0 2 0 4 0 6 0 0 5 0 1 0 0 1 1 4 1 6 1 3 % o f 1n m a c ro c y c le s T i m e ( h )
(a) Library in presence of coacervates. 0 2 0 4 0 6 0 0 5 0 1 0 0 % o f 1n m a c ro c y c le s 1 1 4 1 6 1 3 T i m e ( h )
(b) Control without polyions.
0 2 0 4 0 6 0 0 5 0 1 0 0 1 1 4 1 6 1 3 % o f 1n m a c ro c y c le s T i m e ( h )
(c) Library in presence of pAA.
0 2 0 4 0 6 0 0 5 0 1 0 0 1 1 4 1 6 1 3 % o f 1n m a c ro c y c le s T i m e ( h ) (d) Library in presence of pDADMAC.
Figure 3.4: Effect of coacervates and their components in 1 libraries. The libraries were set up by mixing stock solutions of 1, pAA and/or pDADMAC and diluting them with ddH2O to final concentrations of 200 µM for 1 and 25 mM for pAA and pDADMAC.
The libraries remained unstirred at 40 °C, and were periodically sonicated before UPLC analysis. See above for an important note on the calculation of percentages.
The emergence of 13 as the main component of the library made us suspect
that this macrocycle was also behaving as a self-replicator. Previous studies in
our group have shown that, in conditions where 16 is not stable (such as high
concentrations of denaturants), 13can emerge as a self-assembled self-replicating
macrocycle [40]. Interestingly, we observed that the environment of the
coacer-vate phase was having a similar effect in our system: preformed seeds of 16were
not growing inside of the coacervates (Figures 3.5a and S3.5), but 13 was, and
its growth was autocatalytic (Figure 3.5b). Negative stain electron microscopy of
samples of 13 formed in this way showed fibrous structures (Figure 3.5c),
build-ing blocks. Since 13 was also growing after partial reduction of the library
(Fig-ure S3.4b), where disulfide exchange was possible for longer times, we conclude
that this replicator is not a kinetic trap, but the thermodynamic product of 1n
libraries in the coacervate phase.
0 1 2 3 4 5 0 2 0 4 0 % o f 1n m a c ro c y c le s T i m e ( h ) 1 4 1 6 1 3
(a) 1 and 13/14seeded with 16
0 2 0 0 4 0 0 0 5 0 1 0 0 (2.9 ± 0.7) %/min % o f 13 Time (min) Seed No seed (6 ± 1) %/min
(b) 1 seeded with 13 (c) Electron micrograph of 13
Figure 3.5: Effects of the coacervate phase in 1n replicators. (a) Evolution of 1 libraries,
(200 µM, 50% oxidized using NaBO3), seeded with 16(5 % of the total concentration). The
data shown is the average of three repeats, and the error bars show the standard deviation. (b) Growth of 13in libraries of 1 (200 µM), seeded with coacervates containing preformed 13(33% of the total volume), and without them. In both cases, the libraries were kept at 40 °C without stirring and sonicated before each injection. (c) Supramolecular assemblies of 13, observed by negative stain TEM upon destruction of the coacervates by addition of
3.4
Conclusions
The goal of this project was to determine whether self-replication could be com-bined with compartmentalization in coacervate droplets. This encapsulation strat-egy appears promising: coacervates are easy to prepare, chemically compatible with our system, and due to the high charge density of both coacervates and macrocycles, encapsulation is simple and almost complete.
However, the compartmentalization of our system comes with a significant number of challenges too. As we have seen in this chapter, the behaviour of libraries in presence of coacervates is completely different than without them, leading to the formation of another self-replicating species. This is perhaps un-surprising, due to the charge-rich and crowded nature of the coacervate phase. Since the behaviour of our system relies so much on mechanical processes and interactions between library members (See Chapter 2), it is understandable that a change in environment of this magnitude would greatly affect it. We have shown that it is possible to achieve replication inside of coacervates, but the self-replicating species is not the one that we had predicted. This is an interesting process by itself (self-replicating species that change in different environments are not common [41]), but an obstacle for the design of systems that use coacer-vates as a compartment. As the results of this Chapter show, the properties of replicators in bulk solution cannot be directly translated to the coacervate phase, and therefore any studies performed in bulk would need to be repeated in coac-ervates before designing any further experiments.
The analysis of coacervates is also far from trivial. Confocal microscopy has been a useful analytical tool for these systems: it allows for the analysis of each droplet separately, and it can be performed in a thin film of solution, which pre-vents coalescence. It is also a fast technique, and the analysis of its output is easy to automatize, allowing for a fast analysis of large populations of coacervates with minimal input. However, it is limited by our current fluorescent probe, as it only provides one single channel of information: its fluorescence intensity. There-fore, it does not allow us to distinguish between the different species in a library, or between building blocks with a different chemical composition.
UPLC has the opposite strengths and weaknesses. It does give information about the exact composition of the library at each moment, and it can distinguish between replicators with different chemical structures, but it does not distinguish between droplets. Furthermore, it requires preparing samples in a larger volume, and as the droplets sediment and coalesce the analysis loses reliability. A com-bination of both techniques was enough to study the single-component system described in this Chapter, but for more advanced studies we will need to devise new analytical solutions.
A promising analytical technique to study this system could be confocal mi-croscopy combined with more elaborate fluorescent probes, that can discriminate
between different macrocycles and building blocks through the combination of several fluorophores. These probes have been described in the literature for amy-loid proteins [42], and our group is working towards their implementation in our systems.
Despite the challenges, significant progress has been made towards our final goal. We have confirmed the high efficiency of the encapsulation of replicators inside of coacervate droplets, and the fact that they are still able to bind and ac-tivate ThT is an indication that their structure is, to some extent, conserved. We have also described how these coacervates can discriminate between large self-assembled macrocycles, and smaller ones, retaining the first but allowing for the diffusion of the others. We foresee that this will be critical in future experiments, where replicators in different droplets would be able to interact with each other through the exchange of material, but not directly mixing and forming interme-diate species.
Lastly, we have described for the first time self-replication inside of coacer-vates. Besides the milestone that this represents in the de-novo synthesis of life, the self-replicator discovered in this work is also worthy of further investigation by itself: Can it also be obtained at room temperature? If its formation is triggered by the coacervate environment, can it be formed in other coacervate systems or simply in crowded solutions with high ionic strength? In the seeding experiment, coacervates with seed were mixed with coacervates without it and an effect was observed: is this due to the diffusion of the replicator, the diffusion of the pre-cursors, the diffusion of both, or simply to coalescence? The work in this chapter lays the groundwork for the combination of self-replication and compartmen-talization, but more characterization and mechanistic studies will be required to understand this phenomenon completely and bring it to the next level.
3.5
Further experiments
As we said in the introduction, the ultimate goal of this project is to couple self-replication and coacervate formation. Since this could not be done directly (See Preliminary results), another connection between these two properties needed to be developed. With this goal in mind, we protected the tetraamine spermine with 2-sulfo-Fmoc groups ("Smoc", see Figure 3.6a). Past research in our group
has shown that self-replicators of the type of 16are able to deprotect amines that
have been protected with this motif [43, 44]. Coacervate formation with spermine has also been described [29], and we were able to prepare coacervates in our lab by mixing spermine and pAA (not shown).
Based on these observations, we envisioned a system where replicator - medi-ated deprotection of Smoc-spermine, followed by polyelectrolyte complexation with a suitable polyanion, would lead to the formation of coacervates as self-replication progressed.
Our results showed that 16was indeed able to deprotect spermine. However,
coacervate formation could not be achieved in these conditions, as the deprotec-tion was inhibited by the presence of pAA (Figure 3.6b). Nonetheless, this does not necessarily mean that the reaction is incompatible with polyanions (Smoc-spermine is an oligoanion itself), so a different anion could show the desired be-haviour. H3N H2 N N H2 H3N SmocN SmocN N Smoc SmocN Smoc = O O O3S 16 O3S + 4 x
(a) Proposed reaction for the
16-mediated synthesis of spermine
0 4 5 0 9 0 0 0 , 0 0 , 5 1 , 0 A b s 3 0 8 n m ( a .u .) T i m e ( m i n ) b l a n k 16 + p A A 1 6 (b) Deprotection of spermine in presence and absence of pAA
Figure 3.6: Strategy for a replicator-mediated synthesis of coacervates. (a) Reaction catal-ysed by the replicator and compounds involved. (b) Spermine deprotection in presence and absence of pAA, monitored by UV-Vis. The absorbance at 308 nm corresponds to 2-sulfo-dibenzofulvene, a side-product of the deprotection. The concentrations used were 100 µM for Smoc-spermine and 40 µM for both 16and pAA.
3.6
Materials and methods
All reagents, solvents, and buffer salts were purchased from commercial sources and used without further purification. Building block 1 (XGLKFK) was obtained from Cambridge Peptides Ltd (Birmingham, UK). pDADMAC and pAA were obtained from Sigma-Aldrich, pDADMAC as a 20% (w/w) solution in water (molecular weight and purity not analysed), and pAA as a sodium salt (8-14 %
of water as an impurity, average Mw ∼ 5100). For both polymers, the
concen-trations used in the text refer to their monomer residues. ThT was also obtained from Sigma Aldrich (dye content ≥ 65%). Image analysis was performed with ImageJ (Using the Fiji distribution [45]). Statistical analysis was performed with
OriginPro 8 (OriginLabs, USA).1H-NMR was performed using a 400 MHz
spec-trometer.
UPLC analysis
UPLC analysis was performed on a Waters Acquity UPLC H-class, equipped with a PDA detector. All analyses were performed using a reversed-phase UPLC col-umn (Aeris Peptide 1.7 µm XB-C18 x 2.10 mm, Phenomenex). The colcol-umn tem-perature was kept at 35 °C, and the sample plate was kept at 25 °C, unless other-wise specified. UV absorbance was monitored at 254 nm. For each injection, 10 µL of sample was injected.
The following solvents and gradient were used for UPLC analysis. The flow rate was kept at 0.3 ml/h.
Time(min) % MeCN / 0.1 % TFA %H2O/ 0.1 % TFA 0 10 90 1 10 90 1.3 25 75 3 28 72 11 40 60 11.5 95 5 12 95 5 12.5 10 90 17 10 90
Buffer preparation
Two different buffers were used in this project, both at pH 8.2 and based on borate ions, but each of them with different ionic strengths. For the Preliminary results, we used borate buffer with a total concentration of 50 mM in boron atoms. For the rest of experiments, borate buffer with a total concentration of 200 mM in boron atoms was used. The low concentration borate buffer (50 mM) was prepared
from B2O3, and the high concentration borate buffer (200 mM) was prepared from
Na2B4O7· 10 H2O, both purchased from Sigma-Aldrich (Purity ≥ 99.5%)
Preparation of 1
6Stock solutions of 16 were typically prepared by dissolving building block 1 in
borate buffer to a final concentration of 4.0 mM, and stirring the library at 40 °C. The library was monitored by UPLC analysis until completion.
Preparation of 40 nm 1
6fibres
16 fibres were mechanically sheared using a modified protocol previously
pub-lished by our group [46]. A 150 µL aliquot of a previously prepared solution of 16
(4.0 mM) was placed in a Couette cell (Rcup= 20.25 mm, Rbob= 20.00 mm,
aver-age radius (R) = 20.125 mm). The sample was mechanically sheared by rotation of the inner cylinder, with a frequency of 4000 rpm (corresponding to a shear rate
of 33702 s-1). The resulting fibres were used within 48 h of preparation.
Preparation of 1
3/1
4Stock solutions of oxidized 1 libraries, named "13/14" in the text by their two main
components, were typically prepared by dissolving building block 1 in borate
buffer, adding 1 equivalent of NaBO4, and diluting the library to a final
concen-tration of 4.0 mM in residues of 1. The libraries were monitored by UPLC until completion, which was typically done in less than 1 hour.
Polyelectrolyte complexes using 1
6and polyanions
Stock solutions of 16or 13/14(3.8 mM) were directly mixed with stock solutions
of the different polyanions and chaotropes in the concentrations indicated in the text. Stock solutions were prepared in borate buffer (pH = 8.2, 50 mM in boron), and doubly distilled water was used for further dilutions. To ensure that the observed structures were the thermodynamic products in the samples containing additives, either the additive was added first or the sample was submerged in an ultrasound bath for at least 1 minute before observation. All images were recorded using a Nikon Coolpix 5000 camera attached to a Nikon eclipse TS100
microscope, with a 40x magnification. The scalebar was set by comparison with a reference slide.
General procedure for the preparation of pAA/pDADMAC
coacervates
Stock solutions of pDADMAC, pAA, 1nand ThT were mixed in the desired ratios
for coacervate formation. Except for the 1nstock (which was prepared in borate
buffer, pH = 8.2, 200 mM in boron), ddH2O was used as a solvent. The volumes
added were calculated for concentrations of 50 mM for pAA and pDADMAC, 80
µM for ThT and 0-80 µM for 1n macrocycles. Unless otherwise specified in the
text, the addition order was ddH2O, pDADMAC, ThT, 1n, and pAA,
homogeniz-ing the solution before addhomogeniz-ing pAA.
Typically, samples were centrifugued (RCF = 3969×g, 5 minutes) after
coa-cervate formation, the supernatant was carefully decanted, ddH2O was added to
the desired concentrations, and the samples were redispersed using a standard vortex mixer for 1 minute.
Fluorescence and confocal microscopy
Fluorescence and confocal microscopy experiments were performed using a Mi-croTime 200 setup (PicoQuant, Germany), using an Olympus IX73 microscope with a 100x oil immersion objective. For the fluorescence images, a standard 120 W mercury lamp (X-Cite®series 120 Q) was used as an irradiation source, to-gether with a fluorescence cube with exciter and emitter wavelengths of 466 ± 20 and 525 ± 25 nm, respectively (Semrock GFP-4050B). For confocal images, a 440 nm laser was used together with a 10%R/90%T Beamsplitter, an interference fil-ter with wavelengths of 482 ± 35 nm, and a standard Single-Photon Counting Module (Excelitas Technologies, USA). The laser power remained under 10 µW, and was kept constant for each series of experiments.
Glass slides were coated with PEG to prevent interactions with the coacer-vate phase, using a protocol modified from [47]: 3-[Methoxy(polyethyleneoxy)-propyl]trimethoxysilane (6-9 PE-units, ACBR GmbH) (0.5% v/v) and acetic acid (1% v/v) were dissolved in ethanol to make the coating reagent. 50 µL of this solution was placed on top of each microscope slide, placing another slide on top of it to spread the solution through both surfaces. More slides and droplets of coating reagent were put on top of each other in this way, and then heated in an
oven to 100 °C for at least 30 minutes. The slides were then submerged in ddH2O,
Partition constants of 1
nmacrocycles in pAA/pDADMAC
coacervates
To calculate the volume of the coacervate phase, solutions of pDADMAC and pAA were mixed to a final concentration of 50 mM and volume of 15 mL. These samples were then centrifuged, and the volume of the supernatant was then mea-sured using a standard micropipette (error = 1 µL). The volume of the coacervate phase was then calculated by difference with the original volume, and the data reported is the average and standard deviation of four independent repeats.
pAA/pDADMAC coacervates were then prepared following the general pro-cedure, but after centrifugation the supernatant was collected and analysed by UPLC. The total concentrations used were 80 µM for ThT, 200 µM for 1, and 400
µM for 1noxidized macrocycles. Three repeats and three controls were prepared
for each of the components studied (in the controls, ddH2O was added instead of
polyions). Aliquots of these controls were taken close to the surface immediately after centrifugation, and analysed by UPLC.
The concentration in the supernatant of each component was calculated by comparison of its average peak area with the average peak area of its control samples, and the average concentration in the coacervates was calculated from the difference between this value and the total concentration. Finally, the distri-bution constant was calculated for each component as the ratio between its con-centration in the coacervate phase and its concon-centration in the supernatant. An error value was calculated for each of the measured magnitudes (peak area in the supernatant, peak area in the controls, coacervate volume) as the standard
devia-tion of the three repeats, and the error reported for KDwas calculated combining
those values.
In order to determine whether the 13/14 and 16 samples had reached
equi-librium, two controls were prepared following the same procedure, but adding each macrocycle after the coacervate phase had been formed and centrifuged. These samples were left to equilibrate at room temperature for 20 days, followed by UPLC analysis of their supernatant. For both controls, the peak area values
corresponding to 1n were in the same range as for the the samples shown in the
text.
Validation of ThT as a fluorescence probe for 1
nCoacervates were prepared following the general procedure, containing
increas-ing amounts of either 16or 13/14. After removal of the supernatant, ddH2O was
added so the final concentrations of pAA and pDADMAC were 25 mM and the
fi-nal concentration of 1nwas 0-40 µM. Aliquotes of each sample were deposited on
pretreated glass slides, and measured by confocal microscopy. The droplets were selected manually and their average fluorescence (counts/px) was measured us-ing ImageJ.
Diffusion of 1
nbetween coacervate droplets.
Coacervates containing ThT (40 µM) and either 16, 13/14, or nothing else, were
prepared following the general procedure, excluding centrifugation and redisper-sion. The coacervate dispersion containing only ThT was mixed with each of the other two separately, and an aliquote of the mixture was diluted 4 times, placed
in a pretreated glass slide and covered using a SecureSealTMimaging spacer (9 x
0.12 mm, Grace Bio-Labs, USA) and a cover slip. Each sample was kept at room temperature and studied periodically using confocal microscopy, recording three different pictures in different spatial regions for each time point in each sample.
Three micrographs in total (one corresponding to 2h of diffusion of 16, the other
two corresponding to 3h of diffusion of 13/14) showed unusually high noise, and
were discarded before analysis.
The remaining micrographs were then analysed automatically using an Im-ageJ script (Appendix A) to detect each droplet and measure its average fluores-cence. The process was initially optimized, changing the circularity and thresh-old parameters, to ensure that no artifacts were measured. Then, once that a working set of parameters was found, all images were analysed using the same script. The coacervates not containing macrocycles (average fluorescence < 50 counts/px) were not reliably detected neither by the program nor manually, so they were removed before analysis.
Dynamic combinatorial libraries of 1
nin coacervates analysed by
UPLC
Libraries of 1 and its derivatives were prepared in presence of coacervates fol-lowing the general procedure, excluding centrifugation and redispersion. The libraries were kept at 40 °C inside of the UPLC sample holder, and they were sonicated for 30 seconds before injection. 10 µL of sample were injected directly from the dispersion for analysis. The controls with individual polymers and with-out polymers were kept in exactly the same conditions and sonicated before each analysis as well.
The percentages shown in the figures represent the fraction of each of the main
species of the library (1, 13, 14and 16) over the total peak area (integrated between
6 and 9 minutes). For the figures where 1 is not plotted, the percentages represent the fraction of each component over the total peak area, minus the peak of 1 (7 to 9 minutes). All the peaks shown were assigned by comparison of their retention times with previous data.
The pH of all samples was measured upon preparation and after equilibrium had been reached, by adding a drop of sample to pH-paper. Despite the low concentration of buffer, the pH did not change significantly in any experiment, remaining close to 8.
For the seeding experiments with 16, the seed and a solution of 1 and 13/14
(that had been previously oxidized using NaBO3to the desired ratio) were mixed
immediately before coacervate formation. For the seeding experiments with 13,
a coacervate dispersion containing only 1 was mixed with one containing only
13 in a proportion 2:1 (v/v). The control samples consisted only of coacervates
containing 1.
Quantofix®peroxides test strips (Sigma-Aldrich) were used to determine the
presence of peroxides in the stock solution of pDADMAC. The concentration of peroxides was determined to be lower than the sensitivity of the scale (1 mg/mL) in the 20% w/w stock solution. Lower concentrations than that are not enough to oxidize 1 in the concentrations used in the experiments of this chapter.
Negative staining transmission electron microscopy of 1
3An aliquot of a coacervate dispersion containing 13was mixed in a 1:1 (v/v)
ra-tio with a 0.1% v/v aqueous solura-tion of TFA. Immediately after mixing, a small drop (5 µL) of the resulting solution was deposited on a 400 mesh copper grid covered with a thin carbon film (Ted Pella Inc., USA). After 30 s, the droplet was blotted on filter paper. The sample was then stained twice (5 µL each time) with a solution of 2 % uranyl acetate, by depositing a drop of it on the grid and blotting it on the filter paper after 30 s. The grids were observed in a Philips CM12 cryo electron microscope operating at 120 kV. Images were recorded on a slow scan CCD camera.
Synthesis of Smoc-Spermine
9-(2-Sulfo)fluorenylmethyloxycarbonylchloride was synthesized following a pub-lished procedure [48] (Figure S3.6).
In a 25 mL round bottom flask equipped with a magnetic stirring bar, 9-(2-sulfo)fluorenylmethyloxycarbonylchloride (303 mg, 0.9 mmol) was dissolved in 1.6 mL of 1,4-dioxane. Then, spermine (37 mg, 0.18 mmol) was dissolved in 1.3
mL of a 10% aqueous Na2CO3solution and transferred dropwise to the reaction
flask, under cooling with a water bath. The reaction was stirred at room temper-ature and followed by TLC, using a 1:4 mixture of MeOH:DCM. After 4 hours of stirring the mixture was then acidified to a pH of 2-3 with HCl (1 M), resulting in a yellow solution which was purified by reverse-phase flash chromatography
using a H2O:CH3CN gradient. 1H-NMR in D2O showed only the pure product
(Figure S3.7).1H NMR (400 MHz, D
2O) δ 8.09 - 7.51 (m, 8H), 7.46 – 6.50 (m, 20H),
Deprotection of Smoc-spermine
Stock solutions of Smoc-spermine and pAA were mixed in a quartz cuvette (Hellma
GmbH & Co. KG, germany), diluted using ddH2O, and equilibrated for 20-30
minutes. Right before the measurement started, 16 was added. The final
con-centrations were 100 µM of Smoc-spermine and 40 µM of 16 and pAA. The
ab-sorbance at 308 nm was monitored over time by using a UV-spectrophotometer V-650 (Jasco Benelux B.V., The Netherlands).
3.7
Supplementary material
Macrocycle Anion Additive Result
16(3.8 mM) ATP (1.0 mM) Fibrous structure
13/14(3.8 mM) ATP (1.0 mM) Amorphous aggregate
16(40 nm, 3.8 mM) ATP (1.0 mM) Amorphous aggregate
16(40 nm, 0.38 mM) ATP (0.10 mM) GdmCl (1.0 M) No structures
16(40 nm, 0.38 mM) ATP (0.10 mM) Ca(OAc)2(1.0 M) No structure
16(40 nm, 0.76 mM) ATP (0.20 mM) KBr (1.0 M) Amorphous aggregate
16(40 nm, 0.76 mM) ATP (0.20 mM) Urea (1.0 M) Amorphous aggregate
13/14(3.4 mM) ATP (0.97 mM) GdmCl (0.90 M) No structures
16(2.2 mM) ATP (0.62 mM) GdmCl (3.6 M) Amorphous aggregate
16(1.9 mM) ATP (0.55 mM) GdmCl (4.2 M) No structures
16(1.9 mM) ATP (0.5 mM) NaCl (0.50 M) Amorphous aggregate
16(1.9 mM) ATP (0.5 mM) NaCl (1.0 M) No structures
16(1.9 mM) BSA (2.0 g/L) Fibrous structure
16(3.8 mM) DNA 1 * Gel formation
16(3.8 mM) DNA 2 * Fibrous structure
16(3.8 mM) Κ-casein * Fibrous structure
16(3.8 mM) α-casein * Fibrous structure
16(3.8 mM) Hyaluronic acid * Gel formation
Table S3.1: Screening of conditions for the formation of coacervates in presence of 16and
its precursors. *Instead of calculating the point of charge neutralization for these anions, aliquotes of stock solutions containing them were mixed with 16until the solution became
Figure S3.1: Optical (left) and fluorescence (right) microscopy images of pAA/pDADMAC (25 mM) coacervates prepared in presence of ThT (40 µM) and 13/14(40 µM). The contrast
was increased to facilitate observation.
Figure S3.2: Optical (left) and fluorescence (right) microscopy images of pAA/pDADMAC (50 mM) coacervates prepared in presence of ThT (40 µM), to which 16(80 µM) was added.
The fluorescence signal is still localized inside of the droplets, but it is much less homoge-neous that in the droplets of Figure 3.2b
(a) Confocal images of coacervates in presence of increasing concentrations of 1n 0 2 0 4 0 0 7 0 0 1 4 0 0 A v e ra g e f lu o re s c e n c e i n d ro p le ts ( c o u n ts /p x ) [16] (µM ) (b) Linear relationship
between 16concentration and
ThT fluorescence. 0 2 0 4 0 0 5 0 1 0 0 1 5 0 [1 3/1 4] (µM ) (c) Linear relationship between 13/14concentration and ThT fluorescence. 4 0 8 0 0 , 0 0 , 5 1 , 0 [ T h T ] (µM ) N o rm a liz e d fl u o re s c e n c e i n d ro p le ts
(d) An excess of ThT does not lead to higher fluorescence.
Figure S3.3: Validation of ThT fluorescence as a method for measuring the concentration of 1nin coacervates. (a) Confocal micrographs of pAA/pDADMAC coacervates (25 mM
in monomer units for each polymer) in presence of ThT (40 µM) and growing amounts of 16(top images) or 13/14(bottom images). The black-white scale in the bottom pictures
was increased by a factor of 10 to aid visualization. The average fluorescence intensity inside of the droplets is plotted against the initial concentration of 16or 13/14 in (b) and
(c), respectively. Panel (d) Compares the average fluorescence of two coacervate samples containing 16(40 µM) and different initial concentrations of ThT. The data shown in panels
(b) to (d) is the average and standard error of the mean of 15-20 droplets for each concen-tration. We calculate the concentrations of 1nassuming that they partition completely to
0 2 4 6 0 5 0 1 0 0 T i m e ( h ) 1 1 4 1 6 1 3 % o f 1n m a c ro c y c le s
(a) Oxidation of 1 in coacervates after reduction with TCEP
0 2 4 6 0 2 5 5 0 1 4 1 6 1 3 T i m e ( h ) % o f 1n m a c ro c y c le s (b) Evolution of 1nmacrocycles
in coacervates after reduction
Figure S3.4: Behaviour of a library in presence of coacervates after reduction. The library shown in both (a) and (b) was prepared by mixing 16(50 µM), a 50% oxidized mixture of 1and 13/14(200 µM), pAA and pDADMAC (25 mM each). Right after the first
measure-ment, TCEP (40 µM) was added to the library to reduce it partially. The library is the same in both panels, but (a) includes 1 in the graph and takes it into account for calculations, and (b) only includes oxidized macrocycles.
0 1 2 3 0 2 5 5 0 % o f 1n m a c ro c y c le s 1 14 16 13 T i m e ( h )
(a) 40 nm 16fibres as a seed
0 1 2 3 0 2 5 5 0 % o f 1n m a c ro c y c le s 1 14 16 13 T i m e ( h ) (b) Coacervates prepared with an excess of pDADMAC 0 1 2 3 0 2 5 5 0 1 14 16 13 % o f 1n m a c ro c y c le s T i m e ( h ) (c) Coacervates prepared with an excess of pAA
Figure S3.5: Screening of different conditions for the growth of 16 seeds in presence of
coacervates. Libraries were prepared by mixing 16(30 µM), a 50% oxidized mixture of 1
and 13/14 (200 µM in total), and pAA and pDADMAC, with a total concentration of 50
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 (pp m ) 1.0 6 1.0 2 0.9 9 1.8 9 0.9 3 1.9 1 1.0 0 0.9 9 4.4 4 4.4 6 4.4 7 4.7 9 4.8 0 4.8 1 4.8 3 4.8 7 4.8 9 4.9 0 4.9 2 7.4 5 7.4 5 7.4 7 7.4 7 7.4 8 7.4 9 7.4 9 7.5 0 7.5 1 7.5 1 7.5 3 7.6 9 7.7 1 7.9 4 7.9 6 7.9 6 8.0 0 8.0 2 8.1 3 O Cl O -O 3 S
Figure S3.6: Structure and1H-NMR (400 MHz, CD3CN) of
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 (pp m ) 7.5 6 12. 00 4.2 8 8.2 7 20. 09 8.0 9 0.5 0 0.7 1 0.8 3 1.2 5 2.0 0 2.1 8 2.4 3 2.7 4 3.6 4 4.2 7 6.8 7 7.1 6 7.1 7 7.2 3 7.6 8 7.8 4 O O SO 3 -N H N N H N O O SO 3 -O O -O 3 S O O -O 3 S
Figure S3.7: Structure and1H-NMR (400 MHz, D
3.7.1
Appendix: ImageJ script used to analyse coacervate
pictures
run("Duplicate...", "title=bw");
run("Subtract Background...", "rolling=35"); run("Duplicate...", "title=gray");
selectWindow("bw");
setAutoThreshold("Default dark no-reset"); //run("Threshold...");
setThreshold(3.500, 1000000000000000000000000000000.0000); setOption("BlackBackground", false);
run("Convert to Mask"); run("Watershed");
run("Set Measurements...", "area mean standard median limit redirect=gray decimal=0");
run("Analyze Particles...", "size=10-Infinity
circularity=0.50-1.00 show=Outlines display clear"); String.copyResults(); selectWindow("bw"); run("Close"); selectWindow("gray"); run("Close");
3.8
Acknowledgements
Guillermo Monreal Santiago and Sijbren Otto conceived the project. Patricia Wolf performed the experiments related to spermine deprotection. Guillermo Monreal Santiago performed the rest of experiments and wrote the manuscript. Viktor V. Krasnikov performed fluorescence and confocal microscopy, and participated in discussions. Armin Kiani performed electron microscopy. Alba Algara López and Omer Markovitch are gratefully acknowledged for their help with the sta-tistical analysis of droplet populations. Xinkai Qiu is gratefully acknowledged for the preparation of a microfluidic device that could increase the stability of coacervates towards coalescence [37], even if the operation of that device was too complicated for its practical use.
3.9
References
[1] Shirt-Ediss, B.; Murillo-Sànchez, S.; Ruiz-Mirazo, K. Framing major prebiotic transitions as stages of protocell development: three challenges for origins-of-life research. Beilstein J. Org. Chem. 2017, 13, 1388–1395.
[2] Ruiz-Mirazo, K.; Briones, C.; De La Escosura, A. Prebiotic systems chem-istry: New perspectives for the origins of life. Chem. Rev. 2014, 114, 285–366. [3] Mann, S. Systems of creation: the emergence of life from nonliving matter.
Acc. Chem. Res. 2012, 45, 2131–2141.
[4] Koonin, E. V. The origins of cellular life. Antonie van Leeuwenhoek, Int. J. Gen. Mol. Microbiol. 2014, 106, 27–41.
[5] Chen, I. A.; Walde, P. From self-assembled vesicles to protocells. Cold Spring Harb. Perspect. Biol. 2010, 2.
[6] Schrum, J. P.; Zhu, T. F.; Szostak, J. W. The origins of cellular life. Cold Spring Harb. Perspect. Biol. 2010, 2, a002212.
[7] Paleos, C. M. Organization and compartmentalization by lipid membranes promote reactions related to the origin of cellular life. Astrobiology 2018, 19, 547–552.
[8] Kamimura, A.; Kaneko, K. Compartmentalization and cell division through molecular discreteness and crowding in a catalytic reaction network. Life
2014, 4, 586–597.
[9] Attwater, J.; Wochner, A.; Holliger, P. In-ice evolution of RNA polymerase ribozyme activity. Nat. Chem. 2013, 5, 1011–1018.
[10] Dyson, F. Origins of life, 2nd ed.; Cambridge University Press, 1999.
[11] Baeza, I.; Ibáñez, M.; Wong, C.; Chávez, P.; Gariglio, P.; Oró, J. Possible pre-biotic significance of polyamines in the condensation, protection, encapsu-lation, and biological properties of DNA. Orig. Life Evol. Biosph. 1992, 21, 225–242.
[12] Oparin, A. I. A. I. The origin of life, 2nd ed.; Dover Publications, 2003.
[13] Banani, S. F.; Lee, H. O.; Hyman, A. A.; Rosen, M. K. Biomolecular conden-sates: Organizers of cellular biochemistry. Nat. Rev. Mol. Cell Biol. 2017, 18, 285–298.
[14] Brangwynne, C. P.; Eckmann, C. R.; Courson, D. S.; Rybarska, A.; Hoege, C.; Gharakhani, J.; Jülicher, F.; Hyman, A. A. Germline P granules are liquid droplets that localize by controlled dissolution/condensation. Science 2009, 324, 1729–1732.
[15] Bungenberg de Jong, H. G.; Kruyt, H. R. Coacervation (partial miscibility in colloid systems). Proc K. Ned. Akad. Wet. 1929, 32, 849–856.
[16] Kim, S.; Huang, J.; Lee, Y.; Dutta, S.; Yoo, H. Y.; Jung, Y. M.; Jho, Y.; Zeng, H.; Hwang, D. S. Complexation and coacervation of like-charged polyelectrolytes inspired by mussels. Proc. Natl. Acad. Sci. 2016, 113, E847– E853.
[17] Kayitmazer, A. B. Thermodynamics of complex coacervation. Adv. Colloid Interface Sci. 2017, 239, 169–177.
[18] Nakashima, K. K.; Vibhute, M. A.; Spruijt, E. Biomolecular chemistry in liq-uid phase separated compartments. Front. Mol. Biosci. 2019, 6, 21.
[19] van der Gucht, J.; Spruijt, E.; Lemmers, M.; Cohen Stuart, M. A. Polyelec-trolyte complexes: Bulk phases and colloidal systems. J. Colloid Interface Sci.
2011, 361, 407–422.
[20] Frankel, E. A.; Bevilacqua, P. C.; Keating, C. D. Polyamine/nucleotide coac-ervates provide strong compartmentalization of Mg2+, nucleotides, and RNA. Langmuir 2016, 32, 2041–2049.
[21] Koga, S.; Williams, D. S.; Perriman, A. W.; Mann, S. Peptide-nucleotide mi-crodroplets as a step towards a membrane-free protocell model. Nat. Chem.
2011, 3, 720–4.
[22] Aumiller, W. M.; Keating, C. D. Phosphorylation-mediated RNA/peptide complex coacervation as a model for intracellular liquid organelles. Nat. Chem. 2016, 8, 129–137.
[23] Yewdall, N. A.; Buddingh’, B. C.; Altenburg, W. J.; Timmermans, S. B. P. E.; Vervoort, D. F. M.; Abdelmohsen, L. K.; Mason, A. F.; van Hest, J. Physico-chemical characterization of polymer-stabilized coacervate protocells. Chem-BioChem 2019, cbic.201900195.
[24] Mason, A. F.; Buddingh, B. C.; Williams, D. S.; Van Hest, J. C. Hierarchical self-assembly of a copolymer-stabilized coacervate protocell. J. Am. Chem. Soc. 2017, 139, 17309–17312.
[25] Drobot, B.; Iglesias-Artola, J. M.; Le Vay, K.; Mayr, V.; Kar, M.; Kreysing, M.; Mutschler, H.; Tang, T. Y. Compartmentalised RNA catalysis in membrane-free coacervate protocells. Nat. Commun. 2018, 9.
[26] Poudyal, R. R.; Guth-Metzler, R. M.; Veenis, A. J.; Frankel, E. A.; Keat-ing, C. D.; Bevilacqua, P. C. Template-directed RNA polymerization and en-hanced ribozyme catalysis inside membraneless compartments formed by coacervates. Nat. Commun. 2019, 10, 490.
[27] Zhao, M.; Zacharia, N. S. Protein encapsulation via polyelectrolyte complex coacervation: protection against protein denaturation. J. Chem. Phys. 2018, 149, 163326.
[28] Sokolova, E.; Spruijt, E.; Hansen, M. M. K.; Dubuc, E.; Groen, J.; Chokkalingam, V.; Piruska, A.; Heus, H. A.; Huck, W. T. S. Enhanced tran-scription rates in membrane-free protocells formed by coacervation of cell lysate. Proc. Natl. Acad. Sci. 2013, 110, 11692–11697.
[29] Aumiller, W. M.; Pir Cakmak, F.; Davis, B. W.; Keating, C. D. RNA-based coacervates as a model for membraneless organelles: formation, properties, and interfacial liposome assembly. Langmuir 2016, 32, 10042–10053.
[30] Nakashima, K. K.; Baaij, J. F.; Spruijt, E. Reversible generation of coacervate droplets in an enzymatic network. Soft Matter 2018, 14, 361–367.
[31] Frederix, P. W.; Idé, J.; Altay, Y.; Schaeffer, G.; Surin, M.; Beljonne, D.; Bon-darenko, A. S.; Jansen, T. L.; Otto, S.; Marrink, S. J. Structural and spectro-scopic properties of assemblies of self-replicating peptide macrocycles. ACS Nano 2017, 11, 7858–7868.
[32] McCall, P. M.; Srivastava, S.; Perry, S. L.; Kovar, D. R.; Gardel, M. L.; Tir-rell, M. V. Partitioning and enhanced self-assembly of actin in polypeptide coacervates. Biophys. J. 2018, 114, 1636–1645.
[33] Kunz, W.; Henle, J.; Ninham, B. W. ’Zur lehre von der wirkung der salze’ (about the science of the effect of salts): Franz Hofmeister’s historical papers. Curr. Opin. Colloid Interface Sci. 2004, 9, 19–37.
[34] Jha, P. K.; Desai, P. S.; Li, J.; Larson, R. G. pH and salt effects on the as-sociative phase separation of oppositely charged polyelectrolytes. Polymers (Basel). 2014, 6, 1414–1436.
[35] Levine, H. Thioflavine T interaction with synthetic Alzheimer’s disease β-amyloid peptides: detection of β-amyloid aggregation in solution. Protein Sci.
1993, 2, 404–410.
[36] Malakoutikhah, M.; Peyralans, J. J.; Colomb-Delsuc, M.; Fanlo-Virgós, H.; Stuart, M. C.; Otto, S. Uncovering the selection criteria for the emergence of multi-building-block replicators from dynamic combinatorial libraries. J. Am. Chem. Soc. 2013, 135, 18406–18417.
[37] Vanswaay, D.; Tang, T. Y. D.; Mann, S.; DeMello, A. Microfluidic formation of membrane-free aqueous coacervate droplets in water. Angew. Chemie - Int. Ed. 2015, 54, 8398–8401.
[38] Jia, T. Z.; Hentrich, C.; Szostak, J. W. Rapid RNA exchange in aqueous two-phase system and coacervate droplets. Orig. Life Evol. Biosph. 2014, 44, 1–12. [39] Butler, G. B.; Angelo, R. J. Preparation and polymerization of
unsatu-rated quaternary ammonium compounds. VIII. A proposed alternating intramolecular-intermolecular chain propagation. J. Am. Chem. Soc. 1957, 79, 3128–3131.
[40] Yang, S.; Schaeffer, G.; Mattia, E.; Hussain, A. S.; Ottelé, J.; Mayer, C.; Otto, S. Chemical fueling enables complexification of assembly-driven self-replicators. Submitted
[41] Leonetti, G.; Otto, S. Solvent composition dictates emergence in dynamic molecular networks containing competing replicators. J. Am. Chem. Soc.
2015, 137, 2067–2072.
[42] Hatai, J.; Motiei, L.; Margulies, D. Analyzing amyloid beta aggregates with a combinatorial fluorescent molecular sensor. J. Am. Chem. Soc. 2017, 139, 2136–2139.
[43] Ottelé, J. O.; Hussain, A. S.; Otto, S. Chance emergence of catalytic activity and promiscuity in a self-replicator. Nat. Chem. 2020, Accepted. Preprint – doi: 10.26434/chemrxiv.10002032.v1.
[44] van Esterik, K. (Sul)Fmoc deprotection as a central strategy to incorporate proto-metabolism and compartmentalisation in a system of synthetic repli-cators. Master thesis - Rijksuniversiteit Groningen 2019,
[45] Schindelin, J. et al. Fiji: an open-source platform for biological-image analy-sis. Nat. Methods 2012, 9, 676–82.
[46] Pal, A.; Malakoutikhah, M.; Leonetti, G.; Tezcan, M.; Colomb-Delsuc, M.; Nguyen, V. D.; Van Der Gucht, J.; Otto, S. Controlling the structure and length of self-synthesizing supramolecular polymers through nucleated growth and disassembly. Angew. Chemie - Int. Ed. 2015, 54, 7852–7856. [47] Lau, A. W.; Prasad, A.; Dogic, Z. Condensation of isolated Semi-flexible
fila-ments driven by depletion interactions. EPL 2009, 87, 48006.
[48] Merrifield, R. B.; Bach, A. E. 9-(2-Sulfo)fluorenylmethyloxycarbonyl chlo-ride, a new reagent for the purification of synthetic peptides. J. Org. Chem.