Bottom‐up Strategies for Generating Polymer Protocells That Mimic Cellular Communication
Gloria Saorin, Xinan Huang, Voichita Mihali, Cornelia G. Palivan
TL;DR
This review discusses bottom-up methods to create polymer-based protocells that mimic how cells communicate, aiming to better understand biological processes and develop future medical applications.
Contribution
The paper introduces polymer-based protocells as a novel platform for studying both intra- and intercellular communication in a controlled and programmable manner.
Findings
Protocells can be designed to communicate via chemical signals, mimicking natural signaling pathways.
Interconnected protocell systems allow for the study of single and cascade enzymatic reactions.
Polymer-based protocells offer stability and design flexibility for artificial cell and tissue-like systems.
Abstract
In the last decade, the design of artificial organelles and cells has emerged as an area with far‐reaching implications due to the potential that such systems have for both understanding bioprocesses in a simple and controlled manner, and for developing advanced solutions for medical applications. Significant efforts have been devoted to developing artificial organelles and cells as single compartments or compartments‐in‐compartments to enable the study of internal reactions. However, a major challenge in approaching natural processes is to be able to mimic complex signaling and communication pathways. In this review, we present bottom‐up strategies that introduce polymer‐based protocells for studying intra‐ and intercellular communication. Whereas intracellular communication involves in situ reactions that are triggered by an external stimulus, intercellular communication is achieved…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
FIGURE 1
FIGURE 2
FIGURE 3
FIGURE 4
FIGURE 5
FIGURE 6
FIGURE 7
FIGURE 8
FIGURE 9
FIGURE 10
FIGURE 11
FIGURE 12| Polymer type | Polymers | Polymerization method | Assembly | Reference |
|---|---|---|---|---|
| Homopolymer | Poly(allylamine hydrochloride) (PAH) | Free‐radical polymerization (FRP) of allylamine monomers; conventional radical chain‐growth process |
Coacervates LbL capsules |
|
| Poly(acrylic acid) (PAA) | FRP of acrylic acid monomers |
Coacervates LbL capsules |
| |
| Poly(L‐lysine) (PLL) | Ring‐opening polymerization (ROP) of N‐carboxyanhydride (NCA) of L‐lysine; nucleophilic chain‐growth mechanism | LbL assemblies |
| |
| Amphiphilic diblock copolymer | Poly(butadiene)‐ | Living anionic polymerization; sequential addition of butadiene and ethylene oxide | Vesicles |
|
| Poly(ethylene oxide)‐b‐poly[2‐(diisopropylamino)ethyl methacrylate] (PEO‐ | Controlled radical polymerization via RAFT or ATRP; pH‐responsive tertiary amine block | pH‐responsive vesicles |
| |
| Poly(acrylic acid)‐ | Anionic polymerization of butadiene followed by hydrolysis (for PAA) or coupling with hydrophilic block | Ion‐ and pH‐responsive cylindrical micelles and vesicles |
| |
| Poly(ethylene glycol)‐ | ROP of lactide initiated from PEG; catalyzed by Sn(Oct)2 | Vesicles |
| |
| Poly(ethylene glycol)‐ | Atom transfer radical polymerization (ATRP) of styrene from PEG‐Br initiator; controlled radical growth | Rigid, stable vesicles |
| |
| Conjugated diblock copolymer | Polystyrene‐ | ATRP or anionic polymerization (PS block) + Kumada catalyst‐transfer polymerization (KCTP) for conjugated block | Vesicle‐in‐vesicle architectures |
|
| Poly(glycerol monomethacrylate)‐ | RAFT‐mediated polymerization‐induced self‐assembly (PISA); controlled radical polymerization coupled with in situ self‐assembly | Vesicles |
| |
| Poly(ethylene glycol)‐ | RAFT‐PISA; dispersion polymerization in aqueous media leading to vesicle formation | Vesicles |
| |
| Poly(ethylene glycol) methacrylate‐based block copolymers (Mb‐initiated) | Enzyme‐catalyzed ATRP (biocatalytic polymerization‐induced self‐assembly) using myoglobin (Mb) as catalyst in water | BioPISA vesicles encapsulating active enzymes; cascade‐reactive protocells |
| |
| Poly(N, N‐dimethylacrylamide)‐ | RAFT polymerization of acrylamide monomers followed by fluorophore conjugation | Thermo‐responsive vesicles |
| |
| Amphiphilic triblock copolymer | Poly(2‐methyl‐2‐oxazoline)‐ | Cationic ROP (CROP) of 2‐methyl‐2‐oxazoline from PDMS macroinitiators | Vesicles |
|
| Poly(ethylene oxide)‐ | Anionic polymerization of propylene oxide followed by ethylene oxide extension | Semi‐permeable GUVs, coacervate‐in‐GUVs |
| |
| Amphiphilic terpolymer | Poly(ethylene glycol)‐ | Sequential ROP of cyclic monomers (ε‐caprolactone, trimethylene carbonate, glycolide) | Coacervate |
|
|
Polyelectrolyte | PAH/Poly(styrene sulfonate) (PSS)/Poly(L‐lysine) (PLL)/Poly(methacrylic acid) (PMA) | FRP or ATRP for vinyl monomers; ROP (NCA) for polypeptides; post‐sulfonation for PSS | Layer‐by‐layer (LbL) capsules, coacervate droplets |
|
| Poly(diallyldimethylammonium chloride) (PDDA) | FRP of diallyldimethylammonium chloride; quaternary ammonium cationic polymer | Coacervates |
| |
| Diethylaminoethyl‐dextran (DEAE‐dextran)/Carboxymethyl‐dextran (CM‐dextran) | Postfunctionalization of dextran via etherification or carboxymethylation | Tunable charge coacervates |
| |
| Quaternized amylose/ Carboxymethylated amylose/ Amylose– modified with nitrilotriacetic (NTA) |
Post‐polymer modification: quaternization, carboxymethylation, or NTA coupling | Metal‐binding or charged coacervates |
| |
| Hybrid system | PMOXA‐ | Cationic ROP of triblock + blending with bioactive PDMS‐heparin | Biocompatible hybrid membranes |
|
| Spiropyran‐functionalized poly(butadiene)‐ | Anionic polymerization followed by photochromic spiropyran conjugation | Light‐responsive vesicles |
| |
| Poly(ethylene glycol) (PEG)/Dextran | PEG: anionic ROP of ethylene oxide; Dextran: enzymatic biosynthesis by glycosyltransferase | Aqueous two‐phase coacervate compartments |
|
- —NCCR‐MSE
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsSupramolecular Self-Assembly in Materials · Origins and Evolution of Life · Supramolecular Chemistry and Complexes
Introduction
1
In nature, communication is essential for life, and involves exchange of signals and molecules both inside cells, and with the environment, in order to activate key molecular processes in neighboring cells [1, 2]. Intracellular communication takes place by an incredible variety of reactions that represents the basis for metabolic pathways, while intercellular communication enables collective information‐processing functions that cannot be achieved by cells in isolation [3]. Synthetic cell–cell communication systems based on genetically engineered living cells allow various communication scenarios, ranging from bidirectional [4, 5] or synchronized [6] communication up to complex sender–receiver functions in multicellular colonies [7, 8]. However, to avoid the large number of context‐dependent effects, related to genetically engineered living cells, the bottom‐up strategy for constructing synthetic protocell compartments by combining and hierarchically assembling various building blocks is in focus today as a powerful alternative approach. Its advantages lie in the existence of a large variety of the building blocks in terms of chemical nature (proteins, lipids, peptides, small synthetic molecules, polymers) and type of nano‐ and micro‐assemblies (membrane‐based and membrane‐less compartments) together with a high degree of control and reduced design–build–test cycles [9]. Due to their minimalistic design compared with genetically engineered cells, synthetic protocells serve as powerful models for understanding specific functionalities of living cells in a significantly simpler manner. They enable the study of the conditions necessary for cascade reactions that mimic communication between segregated species that represent artificial organelles (AOs) and/or protocells, and propose advanced materials for further applications in medicine [10, 11].
As the interior of cells is compartmentalized by a membrane that protects their elements from the environment and allows them to perform central functions [12], the first biomimicry step was to develop micrometer‐sized unicompartment protocells and load them with various active biomolecules for specific in situ reactions [13]. To reach an architecture that is closer to that of natural cells, multicompartment protocells were then developed as compartments‐in‐compartment systems by encapsulating inside microcompartments various assemblies with specific functions that serve as AOs [14, 15].
Protocells and their communication have been largely studied in compartments with membranes based on lipids (liposomes and giant unilamellar vesicles, respectively) as their phospholipid composition compares to that of natural cells [16, 17, 18]. However, as lipid‐based compartments have limited mechanical stability and their membranes have intrinsic defects, their use for advanced and long‐time applications has motivated researchers to search for more robust compartments.
In this review, our focus is on polymer‐based protocells generated by bottom‐up strategies, and which mimic cellular communication in a controlled manner over long periods of time. The large variety of amphiphilic copolymers allows control of molecular weight [19], block length [20], and functionalization with specific groups [21] as key features for advanced compartments to serve as cell mimics. In addition, the chemical nature of the copolymers can be selected to generate membranes/assemblies with improved properties, including stimuli‐responsiveness, specific permeability, and shape plasticity [18, 22, 23], or to support self‐organization into clusters to mimic organelles or cellular interactions [24]. Although cell mimics have been variously called artificial cells, synthetic cells, or protocells, in this review, we use only the term protocells to present intra‐ and intercellular communication between these synthetic systems. First, we introduce protocells with various architectures (as uni‐ or multicompartments) and their generation based on synthetic polymers as major constituents of the assemblies. Then, we present various communication scenarios that protocells support to mimic natural communication (Figure 1). The most explored communication is that with the environment, where specific signals induce a change in the shape and properties of the protocells, or diffuse and trigger reactions of specific biomolecules (enzymes, proteins, or mimics thereof) inside them. Intracellular communication that involves cascade reactions between proteins or enzyme‐loaded nano‐assemblies serving as AOs is then presented in various protocells. Finally, intercellular communication, as the most complex molecular signaling pathway, is introduced as an emerging topic in the field of synthetic protocells and prototissues. Although the majority of protocells are models in an early research stage, they indicate how their different architectures allow communication, and open new avenues for further medical applications.
Representative communication models for synthetic protocells: (a) communication in unicompartment protocells, and (b) communication in multicompartment protocells.
Protocells Containing Polymer Building Blocks: Fabrication Methods
2
Unicompartment Protocells
2.1
The simplest way to mimic cells is to produce micrometer‐sized compartments with a membrane mimicking the natural cell membrane, and to enclose them within an aqueous cavity or a medium containing specific components (proteins, enzymes, mimics of thereof) that intrinsically induce the functionality required to mimic natural behavior [25]. This spherical architecture offers a well‐defined reaction space for encapsulated compounds that are free to move inside or are entrapped in a matrix, so that they perform when there is a molecular flow of substrates and products through the membrane. Similar to natural cells, the membrane serves as a protective barrier and, when specifically engineered, allows molecular diffusion that triggers in situ reactions. In addition, a variety of molecules or nano‐assemblies can be integrated inside the protocell, in its membrane, and/or attached to its interface. As the microcompartments have sizes that match cell dimensions, the reactions or interactions that can be investigated are relevant for understanding, in a simpler manner, cellular processes, and communication. There are different types of polymer‐based microcompartments, intended to serve as mimics of cells in terms of size. Here, we present: (i) polymer giant unilamellar vesicles (p‐GUVs), (ii) layer‐by‐layer microcapsules (LbL‐capsules), and (iii) polymer‐coated coacervates, as being the most representative examples of polymer compartments used for the development of protocells (Figure 2).
Schematic overview of polymer‐based unicompartment protocells: p‐GUVs, LbL capsules, and polymer‐coated coacervates.
Polymer Giant Unilamellar Vesicles (p‐GUVs)
2.1.1
p‐GUVs are membrane‐based microcompartments (dimensions 1–100 µm) resulting from the self‐assembly of amphiphilic copolymers (diblock: hydrophobic‐ hydrophilic domains or triblock: hydrophilic‐hydrophobic‐hydrophilic domains). The molecular weights (MWs) of block copolymers are typically orders of magnitude larger than those of lipids and, consequently, the thicknesses of p‐GUV membranes are significantly greater than those of lipid bilayers, ranging on average from 5 to 50 nm for block copolymers compared to 3–5 nm for lipid bilayers [26, 27, 28]. This leads to mechanically stiffer membranes with increased bending rigidity and reduced defect formation, which improves resistance to osmotic and shear stresses in cell‐sized and larger vesicles with relatively low curvature and allows reactions and interactions to be followed over extended periods. At very small, highly curved dimensions, however, the higher bending modulus increases the energetic cost of vesicle closure, so copolymer MW and architecture must be tuned to the targeted size regime [27, 28]. Although polymer membranes are less bioresembling than phospholipid bilayers in terms of thickness, lateral dynamics, and sometimes protein mobility making the integration of membrane proteins more challenging and their superior mechanical and chemical stability enables application scenarios that are difficult to realize with lipids alone, such as long‐term or multistep enzymatic nanoreactors and AOs, or polymer‐based prototissues that must preserve structural integrity and controlled communication under flow or in nonphysiological environments [14, 29, 30, 31]. The use of polymer GUVs as functional platforms for developing protocells is increasing compared to historically used lipid GUVs [32]. There are various methods for producing p‐GUVs, including electroformation [33, 34], film rehydration [35], polymerization‐induced self‐assembly (PISA) [36] and microfluidics (Figure 3a) [37].
Schematic illustration of selected production methods for unicompartment protocells: (a) Polymeric GUVs (P‐GUVs) prepared via electroformation, film rehydration, polymerization‐induced self‐assembly (PISA), and microfluidic techniques. (b) Layer‐by‐layer assembly of polymer capsules. (c) Polymer‐coated coacervates.
Electroformation was one of the first approaches used to generate p‐GUVs. Polymer films are deposited on indium tin oxide (ITO) coated glass slides, and self‐assembly is induced by applying alternating electric currents while rehydrating (Figure 3a) [34, 38]. The electric field induces a periodic electroosmotic movement of water between the individual bilayer lamellae in the film, and drives detachment of a vesicle from the surface [39]. This method generates relatively uniform GUVs, but with limited control over the size and morphology as multilamellar and multivesicular GUVs can also form [33, 40]. In the case of the film rehydration method, amphiphilic copolymers dried as a thin film are slowly hydrated with buffer (e.g., phosphate‐buffered saline, PBS) containing sucrose, which induces formation of micro‐assemblies [33, 35]. When the copolymers have a specific ratio of hydrophilic and hydrophobic blocks, the resulting micro‐assemblies are GUVs (Figure 3a). An advantage of the film rehydration method is the absence of organic solvents, which might affect the sensitive biomolecules used to supplement the GUVs in the development of protocells. Moreover, this approach induces formation of p‐GUVs with significant size‐distributions and moderate encapsulation efficiency, especially when there is an aim to encapsulate more than one type of molecule. To narrow the size distribution of protocells prepared by electroformation or film rehydration methods, p‐GUVs suspensions are subjected to additional processes, such as freeze‐thawing, sonication, or extrusion through a polycarbonate membrane. Alternatively, dialysis and size‐exclusion chromatography may also be used to obtain relatively monodisperse populations. PISA produces nano‐ and microscale objects by a one‐pot synthesis and self‐assembly‐induced process [41]. PISA is based on chain extension of a water‐soluble macroinitiator using water‐miscible monomers that become water‐insoluble at a certain critical degree of polymerization (Figure 3a). This change induces self‐assembly into various supramolecular assemblies [21, 42], including p‐GUVs [43]. For example, PISA of GUVs obtained from poly(glycerol monomethacrylate)‐b‐poly(2‐hydroxypropyl methacrylate) (PGMA19‐b‐PHPMA25) [44] and poly(ethylene glycol)‐block‐poly(2‐hydroxypropylmethacrylate) (PEG‐b‐PHPMA) [45] produced vesicles in a straightforward manner, with controlled size, desired surface chemistry, and morphology. However, these vesicles have relatively low encapsulation efficiency, and their membranes are intrinsically permeable. In addition, the size dispersity of p‐GUVs is significant and induces differences in encapsulation efficiency [41, 46].
Microfluidic techniques generate p‐GUVs with a narrow size distribution [37] and achieve almost 100% encapsulation efficiency of several types of compounds, including proteins and enzymes, which are essential components of protocells (Figure 3a) [47]. There are various microfluidic devices with specific geometry and based on various materials, including glass capillary devices [47], 3D‐printed devices [48], and PDMS systems based on soft lithography processes [49]. The size of p‐GUVs is controlled by the channel geometry and the flow rate of the outer aqueous (OA) and inner aqueous (IA) streams [49]. The amount of encapsulated fluid inside p‐GUV cavities depends on the flow rate of the IA stream, while the frequency of the pinch‐off is determined by the flow rate of the OA stream. However, to obtain GUVs with a homogeneous polymeric membrane necessary for the development of protocells, a higher level of precision and reproducibility is required for generating double emulsions with thin organic shells. For example, homogeneous polymer GUVs have been formed by using double‐emulsion templates generated in a six‐way junction microfluidic device with controlled geometry [47, 50]. The IA phase was pinched and enveloped by polymer organic (PO) and OA phases, resulting in the formation of double‐emulsion droplets. At the two liquid‐liquid interfaces of the droplets, the copolymer chains self‐assembled into two monolayers, which resulted in the formation of a bilayer upon polymer solvent dewetting. A drawback still not solved is the complete removal of organic solvents, which might denaturate sensitive biomolecules, such as enzymes or induced toxicity for cells when the p‐GUVs are incubated with. While the relatively large size of p‐GUVs matches the sizes of cells, obtaining smaller sized‐compartments as AOs still requires optimization of the microfluidics set‐up geometries and conditions of the streams.
Based on the chemical nature of the copolymers, p‐GUVs can exhibit various properties, including stimuli‐responsiveness, flexibility/rigidity, permeability/impermeability, and viscosity. In order to mimic the ability of living cells to initiate metabolic reactions at a specific time and at a desired location, it is important to induce temporal control of the protocell systems. In this respect, various stimuli‐responsive p‐GUVs have been generated, and their rupture has been explored to enable the delivery of desired compounds. For example, synthetic GUVs based on poly(butadiene)‐block‐poly(ethylene oxide) (PBut_2.5_‐b‐PEO_1.3_) loaded with different cargos released their contents upon irradiation, which induced an increase in osmotic pressure due to internal accumulation of photogenerated species [51]. In a different approach, poly(ethylene oxide)‐b‐poly[2‐(diisopropylamino)ethyl methacrylate] (PEO‐b‐PDPA) diblock copolymer used in combination with commercial poly(ethylene oxide)‐block‐poly(1,2‐butadiene) (PEO_34_‐b‐PBD_46_) resulted in homogeneous pH‐responsive p‐GUVs, which ruptured in response to external changes in pH under biological conditions [49]. However, although responsive GUVs allow understanding how cells initiate changes with temporal precision, a key aspect of protocell development is their ability to preserve their contents and remain functional over long periods of time. In this respect, the use of amphiphilic copolymers instead of lipids is essential for significantly increasing their stability. Similar to cell membranes, domain formation in p‐GUVs can be generated by addition of cations within assembled mixtures of neutral and anionic polymer amphiphiles [52]. Mixtures of poly(acrylic acid)_poly(butadiene) (PAA_PBD) and nonionic poly(ethylene oxide)_PBD (PEO_PBD) copolymers in various proportions self‐assembled into p‐GUVs with calcium‐induced domains or rafts [52], which were stable for several days.
While the synthetic membranes of p‐GUVs are designed to mimic the bilayer morphology of cell membranes, other properties, including permeability, viscosity and rigidity, are different from those of lipid‐GUVs. A key property required by protocells is the permeability of their membrane, which, depending on the chemical nature of the polymers, ranges from impermeable up to highly permeable (porous). For example, the thicker membrane of p‐GUVs normally has low permeability [27, 53] unless copolymers are specifically selected to assemble in an intrinsically porous membrane [54]. When the polymer membrane of p‐GUVs is impermeable, it limits the initiation of in situ reactions by preventing the molecular transport of substrates and reaction products across the membrane. To tackle this challenge, molecular flow was achieved by inserting into the membrane ion channels [55], ionophores [56], or DNA‐origami nanopores [50]. Although ionophores such as ionomycin were inserted during membrane formation [56], a more complex scenario of conditions is required to reconstitute membrane proteins, because of a significant mismatch between the size of the biomolecules and the thickness of the membranes [13]. However, flexibility is another key property of membranes, which, when appropriately selected, allows the reconstitution of membrane proteins inside. In addition, it supports the insertion of biomolecules even in membranes that are thicker than the cell membrane [57].
In order to develop protocells based on p‐GUVs, the first step is to equip them with desired molecules (proteins, enzymes, biopores) that will induce cell‐like functionality. With the film rehydration method, the biomolecules should be present in the solution used for rehydration [13]. Then, hydrophilic molecules are encapsulated inside the cavity of GUVs, while hydrophobic molecules are inserted in the membrane. When the biomolecules are chemically incompatible, another option is a step‐wise approach, in which the hydrophilic molecules are first encapsulated inside the cavity of GUVs, and hydrophobic molecules in detergent solution are added to the preformed GUVs before finally removing the detergent [58]. With PISA, the biomolecules have to be present from the beginning of the synthesis, and specifically selected to not be affected by the synthesis components [43]. In addition, as the resulting membrane obtained by PISA is permeable, fine‐tuning its composition is necessary to avoid uncontrolled release of the encapsulated compounds, especially when they are low MW molecules [59]. In the case of microfluidics approaches, the biomolecules are dissolved in the IA stream in concentrations depending on their intrinsic solubility. However, an alternative option is to first produce p‐GUVs simultaneously with encapsulation of hydrophilic molecules by microfluidics, and then to add from the environment biopores/membrane proteins to permeabilize the membrane to further support in situ reactions or communication with the environment [50].
LbL Capsules
2.1.2
LbL capsules are generated by sequential adsorption of oppositely charged polymers onto a substrate, followed by rinsing steps after each layer is formed. Thus, each adsorption step effectively reverses the surface charge, facilitating the deposition of the next layer, and resulting in a stepwise increase in the overall layer thickness (Figure 3b) [60]. The predominant driving forces for formation of LbL capsules are electrostatic interactions, although depending on the chemical nature of the polymers involved, the process is completed by nonelectrostatic interactions, including hydrogen bonds, covalent bonds, van der Waals forces, and hydrophobic interactions [61]. Two main approaches are used for generating LbL capsules: (i) By using sacrificial templates, and (ii) direct self‐assembly at fluid interfaces.
Sacrificial templating is based on depositing successive polymer layers onto a pre‐formed core particle, which is then selectively dissolved or degraded to form a hollow capsule. A variety of sacrificial templates have been used, each chosen for its specific properties and ease of removal; these include CaCO_3_ microparticles [62], melamine formaldehyde colloidal particles [63], micro and nanoparticles [64]. The choice and properties of the template are critical, in that they influence the size and stability of the final capsule, as well as the feasibility of the fabrication process. Removal of the sacrificial core is a critical step, which requires various methods that preserve the integrity of the LbL shell, including dissolution using specific chemical treatments, enzymatic degradation for protein‐based shells, or a pH change in the surrounding medium. A balance between efficient core removal and maintaining the structural integrity of the multilayered shell is essential for successful formation of LbL capsules.
Direct self‐assembly at interfaces is based on the accumulation of capsule‐forming components at a fluid–fluid interface, such as at an oil–water microdroplet interface [65, 66]. The fluid–fluid interface serves as a dynamic template for the formation of a shell that encapsulates the volume within a microdroplet. Charged copolymers are selectively directed to the microdroplet interface by a complementary charged surfactant based on electrostatic partitioning. Direct self‐assembly at interfaces allows dynamic formation of capsule shells at droplet boundaries, and is capable of producing hollow, solid, or core‐shell structures from a single solution [67].
The time‐consuming nature of LbL assembly represents a significant limitation of this approach, but methods, such as microfluidic approaches, have been developed to enhance speed, control, and scalability in LbL capsule production [68]. Microfluidic deposition controls precisely the flow and mixing of solutions within microchannels, enabling the formation of monodisperse droplets or templates for LbL coating. This method offers high control over droplet size and uniformity and facilitates continuous, high‐throughput production of capsules. The properties of LbL capsules can be precisely tuned by manipulating factors such as polymer charge, solution pH, ionic strength, and the chosen deposition method. These capsules are effective for encapsulating active agents and provide the protection, stability, and controlled release that are particularly important for challenging hydrophobic drugs.
The development of protocells involves equipping LbL capsules with active biological components, such as proteins or enzymes that confer specific functionality. Various strategies are used to load active compounds into LbL capsules, involving either pre‐loading before the LbL coating process, or post‐loading into hollow capsules. In the pre‐loading approach, the active compound is either incorporated into the sacrificial core particle before the LbL coating process begins [69], or adsorbed onto the template as one of the initial layers, such as DNA adsorbed onto positively charged silica particle templates, before polymer film formation [70]. Conversely, in the post‐loading approach, the active compound is loaded by a diffusion‐driven precipitation mechanism after the sacrificial core has been dissolved [71]. Diffusion‐driven precipitation occurs because hollow capsules retain abundant free ionic groups, which establish a polar gradient between the internal cavity and the environment. This then drives the precipitation of active compounds inside the capsules when their solvent is evaporated. The composition, thickness, and cross‐linking of the LbL shell can be precisely controlled to allow selective passage of small molecules while retaining larger encapsulated components (enzymes, genetic material, nanoassemblies). However, LbL encapsulation faces certain challenges: relatively low loading efficiency, particularly for large molecules/nanoassemblies, leakage prior to reaching the target site, and inhomogeneous distribution of the cargo within the capsule, especially during post‐loading. In addition, encapsulating and maintaining complex, multistep biochemical reaction networks within LbL capsules, especially those requiring precise spatial organization and energy supply, remains a significant challenge.
Polymer‐Coated Coacervates
2.1.3
Membrane‐less organelles, such as P bodies, Cajal bodies, and speckles, are phase‐separated condensates characterized by the absence of a lipid bilayer membrane, and hence are very dynamic in nature. They are involved in cellular functions ranging from endocytosis, gene expression, neurotransmission, and immune responses [72, 73]. Enzymes play a significant role in reactions controlling liquid−liquid phase separation assembly in cells, while coacervates are able to catalyze and provide a dynamic environment for spatiotemporal coordination of reactions. Thus, coacervates represent an appropriate in vitro model for membrane‐less organelles [74, 75]. Coacervate microdroplets form by liquid–liquid phase separation between oppositely charged polypeptides, polynucleotides, and polymers, and by segregating them through fine‐tuning their densities and critical salt concentrations, it is possible to confer selective partitioning ability to these membrane‐less compartments [76]. Assembly of coacervates can be tuned by physical factors (temperature [77] or light [78],) or by the activity of enzymes [79]. However, membrane‐less coacervates tend to coalesce easily and lack selectivity in sequestering surrounding substances due to the absence of a closed boundary [80]. Therefore, in order to build stable coacervates that can serve as protocells, it was essential to integrate a membrane shell to protect them from disassembly, and membranized‐coacervates have been generated by the addition of membrane‐building components based on amphiphilic (macro)molecules (Figure 3c) [81, 82, 83]. Protocells with a single‐compartment architecture were generated by the formation of micrometer‐sized coacervates covered with polymers through interfacial self‐assembly [75, 84, 85]. In addition to enhanced colloidal stability of these coacervate‐based protocells, this resulted in customizable membrane permeability, which allows diffusion of molecules with different MWs [86, 87].
Coacervates based on quaternized amylose (Q‐Am) and negatively charged carboxymethylated amylose (Cm‐Am) were indeed stabilized by the addition of a terpolymer (PEG−PCLgTMC‐PGA), which generated a membrane‐like shell around the coacervate droplets [88]. Use of membranized‐coacervates as protocell models has the advantages of strong incorporation of cargo into their core due to charge complementarity and/or hydrophobicity, and of being closer to the interior of a living cell by their inherently crowded environment, as ∼30% of the cytoplasmic volume is occupied by biomacromolecules [89]. Crowding is reported to influence macromolecular association, protein conformation, and diffusional processes [72, 90, 91]. In addition, the intrinsic limitation of coacervates in being able only to selectively recruit molecules has been overcome by the addition of amylose modified with nitrilotriacetic acid (NTA), which is well‐known to complex Ni^2+^ and reversibly bind polyhistidine‐tagged (His‐tagged) proteins. This modified coacervate was capable of efficient sequestration of different His‐tagged recombinant proteins, thus enabling control over their local concentrations [85]. However, there is still a limited understanding of the localization of molecules (amino acids, nucleotides, NADPH, etc.) and their accessibility in this type of protocell.
Another interesting aspect of such protocells is their de‐membranization for understanding the dynamics of biomimetic cell structures or for allowing responsive sequestration or uptake of guest species into the coacervate phase [92, 93]. De‐membranization can be achieved by altering or disrupting the noncovalent interactions (ionic, H‐bonds, etc.) between the coacervate matrix and membrane‐building components. For example, the addition of an anionic polysaccharide (carboxymethyl dextran with MW 40 kDa), induced an electrostatic competition with the membrane components and initiated complete membrane dissociation and structural reorganization of coacervate droplets [94]. This membranization/de‐membranization process enabled controlled structural deformation of coacervates, which exhibited varying uptake behavior toward different (macro)molecules and selective permeability to nano/micro‐scale objects.
Multicompartment Protocells
2.2
An essential step in mimicking the complex composition of cells is the encapsulation of nano‐assemblies (coacervates, hydrogels, polymersomes, nanoparticles) to serve as mimics of natural organelles, AOs and/or active compounds (enzymes, proteins, catalysts) inside microcompartments (Figure 4). The bottom‐up approach allows the generation of compartments‐in‐compartment architecture with high organizational complexity and multifunctionality. A large variety of multicompartments can be generated due to the incredible number of chemically available nano‐assemblies and combinations with biomolecules for mimicking the extremely large variety of natural cells.
Schematic of polymer‐based multicompartment protocells: p‐GUVs with synthetic membrane‐less AOs, coacervate in coacervate, p‐GUVs with membrane‐bound AOs, capsosomes, polymer‐coated coacervates with membrane‐bound AOs, and assembled prototissues.
Multicompartments Containing Synthetic Membrane‐less AOs
2.2.1
In order to mimic the membrane‐less organelles of natural cells, it is essential to encapsulate inside synthetic microcompartment smaller‐sized assemblies without a membrane and to equip them with active compounds, such as proteins or enzymes. For example, p‐GUVs based on poly(ethylene glycol)‐b‐poly(propylene glycol)‐b‐poly(ethylene glycol)(Pluronic L121) were generated by water‐in‐oil‐in‐water (W1/O/W2) double emulsion droplets and loaded with adenosine triphosphate (ATP) and poly(allyl amine hydrochloride)(PAH) for in situ formation of coacervates [95]. Pluronic‐based GUVs were stable for at least 1 week at room temperature and 37°C and were intrinsically permeable to molecules within MW up to 500 Da [96]. A change of external pH from 4.0 to 7.4 induced nucleation and formation of complex ATP‐PAH coacervate droplets, resulting in a compartment‐in‐compartment architecture [95, 96]. In addition, the formation of coacervates was triggered by in situ enzymatic reaction of encapsulated pyruvate kinase inside Pluronic‐based GUVs when its substrate, phosphoenol pyruvate (PEP), was added to the environment of the GUVs [95]. In the case of p‐GUVs with impermeable membranes, specific molecules are inserted into the membranes to induce stimuli‐responsive permeability [97], and allow molecular transport [50]. For example, light was an external stimulus for permeabilizing the membrane of p‐GUVs and supporting the formation of coacervates inside protocells. When spiropyran‐based permeability modulators were integrated into the membrane of impermeable p‐GUVs, their photo‐isomerization perturbed the polymer membrane and allowed ATP to pass through [97]. Once inside the p‐GUVs, ATP induced the positively charged polyelectrolyte (poly‐L‐lysine) to form coacervates. Such stable coacervates are able to sequestrate and localize encapsulated proteins, while their integrity can be reversed by encapsulating a counter enzyme (alkaline phosphatase), thus introducing transient behavior into the protocells [97].
PDMS‐PMOXA‐based p‐GUVs loaded with prior‐formed membrane‐less micron‐sized coacervates based on a mixture of polyethylene glycol (PEG) and dextran (DEX) polymers where generated by double‐emulsion microfluidics with different geometry (Figure 5a) [98]. The membrane of the resulting protocells was permeable to lactate and pyruvate, as an essential feature to support molecular diffusion and communication between protocells [98].
p‐GUVs (PMOXA10‐b‐PDMS25) hosting synthetic membrane‐less and membrane‐bound artificial organelles (AOs). (a) Schematic of a coacervate AO (ML‐AO) inside a p‐GUV and a CLSM z‐projection of a BODIPY 630/650–labeled membrane (red) with an FITC‐dextran–labeled ML‐AO (green). Reproduced with permission [98]. Copyright 2025, Wiley‐VCH. (b) p‐GUVs containing a polymersome AO (membrane‐bound) produced by double‐emulsion microfluidics, with 3D reconstruction of AO distribution (green) within a BODIPY 630/650–stained protocell (red). Reproduced with permission [104]. Copyright 2024, Wiley‐VCH.
In a different microfluidics approach, porous polymer microcompartments were generated in the presence of DNA and clay minerals. UV polymerization of acrylate monomers resulted in polymer membranes mechanically stable, rigid, and permeable to macromolecules up to 2 MDa [99]. By in situ condensation of clay minerals and DNA, the round and porous hydrogel structure resembled the nucleus of an eukaryotic cell and allowed the diffusion of macromolecules of up to 500 kDa MW, including cell‐free transcription and translation (TX‐TL) reagents. An advantage of the permeable membrane with large pores was to allow ribosomes, the largest components of TX‐TL reagents, to diffuse into the protocell membrane while retaining the internal DNA‐clay nuclei‐like system [99]. Increased control of molecular flow through synthetic membranes of protocells can be achieved by inserting membrane proteins (e.g., OmpF) into the membrane of p‐GUVs, either by the film‐rehydration method [13] or by microfluidic approaches [47, 50]. Another advantage of reconstituting membrane proteins and biopores in p‐GUV membranes is implementation of size specificity of the molecular flow in a closer‐to‐nature way that resembles cell membranes. The film‐rehydration method was also used to supplement p‐GUVs with stimuli‐responsive nanoparticles containing active molecules and acting as membrane‐less AOs [100]. p‐GUVs based on a mixture of amphiphilic block copolymers poly(2–methyl‐2‐oxazoline)5– *block–*poly(dimethylsiloxane)58– *block–*poly(2–methyl‐2‐oxazoline)5 (PMOXA_5_‐b‐PDMS_58_‐b‐PMOXA_5_) and PDMS_65_‐b‐heparin were loaded during self‐assembly with specific molecules and prior‐formed stimuli‐responsive nanoparticles. As p‐GUVs were intrinsically permeable to small molecules, such as dithiothreitol (DTT), their diffusion from the environment was intended to induce a change in the inner cavity and subsequent release of the molecules from the degraded AOs [100, 101].
Protocells can also be developed as coacervate‐in‐coacervates microdroplets by covering an inner coacervate with multilayers of polyelectrolytes that serve as the external coacervate [102]. Inner coacervates were formed by liquid–liquid phase separation of the strong polyelectrolytes, polydiallyldimethyl ammonium (PDDA) and deoxyribonucleic acid (DNA), while diethylaminoethyldextran (DEAE‐dextran) and carboxymethyl‐dextran (CMdextran) were deposited on the surface of as‐prepared PDDA/DNA compartments through LbL assembly [102]. Various enzymes were localized either inside the inner coacervate or inside both of them, while the intrinsic permeability of coacervates allowed substrates to penetrate and induce the specific functionality of the protocells.
Multicompartments Containing Synthetic Membrane‐bound AOs
2.2.2
Protocells based on p‐GUVs containing polymersomes equipped with various active compounds (enzymes, proteins, biopores) can mimic natural cells with their membrane‐bound organelles, such as nuclei, mitochondria, and peroxisomes. To obtain such protocells, a sequential approach is used that is different from coacervate‐containing microcompartments. First AOs with nanometer sizes and functionality are formed by encapsulating active compounds inside polymersomes. Then, these AOs are encapsulated inside p‐GUVs during their formation process. For example, a vesicle‐in‐vesicle structure was formed by a direct dissolution method when prior‐formed PDMS‐PMOXA‐based polymersomes were added to a solution of polystyrene‐b‐poly(3‐(isocyano‐lalanyl‐amino‐ethyl)‐thiophene) (PS‐b‐PIAT) [103]. Similarly, AOs initially generated by encapsulation of different enzymes in nanometer‐sized PS‐b‐PIAT polymersomes were mixed with cytosolic enzymes and reagents before encapsulation in polybutadiene‐bpoly(ethylene oxide) (PB‐b‐PEO) p‐GUVs by an emulsion‐centrifugation approach [14].
As, the encapsulation efficiency of the AOs is limited by the statistical character of the p‐GUV self‐assembly process by the film rehydration method, but a significant increase in AOs encapsulation inside GUVs protocells has been achieved by using a double‐emulsion microfluidics approach [104]. This involves adding different types of membrane‐bound prior‐generated AOs to the IA phase, for which GUVs with concentrations up to 3.4 × 10^11^ AOs mL^−1^ have been reported (Figure 5b) [104]. Use of such microfluidics approaches in the construction of hierarchical membrane‐bound protocells offers unparalleled precision and control over their content and the membrane composition, together with being able to fine‐tune the interorganelle distance, essential for replicating complex cellular structures and functions. The average nearest neighbor distance between AOs inside protocells lies between 1 and 3.3 µm, which is similar to the interorganelle distances found between lysosomes in natural cells [105]. However, co‐encapsulation of both membrane‐less and membrane‐bound AOs has not yet been achieved because of the complexity of combining the conditions for coacervate formation with those of maintaining the integrity of membrane‐bound AOs or that of the protocell itself.
Using a different microfluidic approach, aqueous‐core mesoporous‐shell silica nano‐capsules (AMSNs), loaded with enzymes to mimic the functions of natural eukaryotic organelles, have been encapsulated in (poly(butadiene)‐b‐poly(ethylene oxide) (PB‐b‐PEO)‐based p‐GUVs. Due to the semipermeable nature of the AMSNs, the encapsulated enzymes were able to perform enzymatic reactions, thus functioning as single or in tandem nano‐organelles (around 10^4^ randomly distributed AOs) [106, 107]. The unprecedented encapsulation efficiency and homogeneity of protocells generated by microfluidics approaches has not yet been achieved by other methods, as for example film rehydration.
LbL capsules have also been used to generate multicompartment protocells named capsosomes by incorporating enzyme‐loaded liposomes in their shell, in order to exploit the benefits of both systems while minimizing some of their drawbacks, including the mechanical instability of liposomes [108]. The polymer shell has the advantages of protecting the liposomes from degradation while being semipermeable to allow diffusion of molecules (substrates/products) that support the overall enzyme functionality. Capsosomes were generated by sequential deposition of interacting liposomes and polymer layers onto sacrificial SiO_2_ [109]. As the electrostatic interaction alone did not provide sufficient affinity between liposomes and the underlying polymer surface, modified polymers (e.g., cholesterol‐modified poly(L‐lysine) [109]) served as anchors for liposome attachment. A further increase in complexity of protocells has been achieved by integrating Au nanoclusters (AuNCs) inside capsosomes for system tracking purposes [110]. Several poly(l‐lysine) (PLL) polymer layers allowed the deposition of negatively charged BSA‐stabilized AuNCs, which provided efficient anchoring for the liposomes without rupture, and connection between the liposomes and the polymer shell (Figure 6a) [110, 111]. Capsosomes containing one type of enzyme inside liposomes, or two different enzymes in specific layers of liposomes supported enzymatic reactions for medical applications [110, 111].
Multicompartment protocells based on capsosomes and polymer‐coated coacervates with membrane‐bound AOs. (a) Enzyme‐loaded multilayer capsosomes enabling a GOx/HRP cascade reaction. Reproduced with permission [110]. Copyright 2017, American Chemical Society. (b) Coacervate‐based protocells encapsulating polymersome AOs and stabilized by a PEG–PCLgTMC membrane to form cell‐sized hierarchical structures. Reproduced with permission [88]. Copyright 2019, American Chemical Society.
Hierarchically organized protocells have been obtained with AOs entrapped inside micrometer‐sized coacervates that were stabilized by an external polymer coating [81]. Such protocells integrated distinct attributes of eukaryotic cells: a hierarchical structure formed by entrapment of protein‐loaded AOs, spatial organization of internal enzymes, and crowding due to the intrinsic nature of amylose‐based coacervates (Figure 6b) [88]. Enzyme‐loaded polymersomes comprising poly(ethylene glycol)‐b‐poly‐(caprolactone‐gradient‐trimethylene carbonate) (PEG− PCLgTMC) block copolymers served as AOs with a semipermeable membrane. The membranization of coacervate‐polymersome microdroplets was realized by subsequent addition of a PEG−PCLgTMC terpolymer once the coacervates attained cell size [88]. However, the sequestration efficiency of AOs was low (8.9 ± 0.4%), presumably due to the hydrophilic nature of surface PEG chains.
Prototissues
2.3
In order to generate prototissues, it is necessary to generate clusters or networks of protocells that are bound together by linkers to cells in close vicinity to permit exchange of signals or interactions. The protocells forming prototissues can have different inner architecture: (i) unicompartment and (ii) multicompartment (with only one type of AOs or two types of AOs). An important requirement is that the protocells preserve their integrity and functionality inside the prototissues. One of the most efficient methods of linking involves DNA hybridization between ssDNA exposed on one protocell and complementary ssDNA exposed on another protocell. Although the design of DNA‐linked vesicles has focused on lipid‐GUVs [112], the applicability of such multicompartment vesicles to the formation of prototissues is limited by the mechanical and chemical instabilities of lipid GUVs. Membrane instability can induce GUV fusion with the substrate or other adjacent GUVs, and uncontrolled rupture or growth of compartments, which compromises the segregated compartmentalization and composition of natural tissues. Therefore, p‐GUVs are more appropriate for generating stable clusters when zipped together by linkers. However, there are various molecular factors that increase the complexity of the approach: lower lateral diffusion of a synthetic membrane compared to a lipid membrane, possible blocking of the linker insertion due to a tightly entangled and dense synthetic membrane, and a larger hydrophilic volume of block copolymers compared to lipids. p‐GUVs of different composition have been generated by using glass capillary microfluidics with a co‐flow geometry and zipped through DNA hybridization between complementary cholesterol‐tagged ssDNA exposed on different p‐GUVs [113]. Three types of p‐GUVs consisting of PLA‐b‐PEG, poly(butadiene)‐b‐poly(ethylene glycol) (PBD‐b‐PEG), and poly(ethylene glycol)‐b‐poly(propylene glycol)‐b‐poly(ethylene glycol) (Pluronic L121) block copolymers allowed comparison of the interaction and anchoring of cholesterol‐tagged ssDNA for supporting prototissue formation. Although cholesterol‐tagged DNA rapidly accumulated at the membrane when added to Pluronic L121 polymersomes, the DNA was not incorporated into the other two copolymer p‐GUVs, thus indicating the importance of the copolymer properties in the anchoring process. Also, a smaller hydrophilic weight fraction was essential to facilitate incorporation of cholesterol‐tagged DNA into membranes of preformed p‐GUVs. Furthermore, various molecular factors, including copolymer properties, the absolute thickness of either hydrophobic or hydrophilic domains of the membrane, and the degree of compactness of the hydrophobic domain after the de‐wetting process contributed to successful anchoring of cholesterol‐tagged DNA to p‐GUVs, and the final formation of extensive clusters (Figure 7a) [113]. DNA hybridization was favored by lateral diffusion of ssDNA in the membrane of the p‐GUVs, thereby creating areas with higher DNA density [24].
Formation and assembly of polymeric prototissues. (a) DNA hybridization–driven p‐GUVs network formation Reproduced with permission [113]. Copyright 2020, Wiley‐VCH. (b) Stepwise fabrication of polymer‐coated protocells from CaCO3 templates and their assembly into crosslinked prototissue (PTMs). Reproduced with permission [114]. Copyright 2024, Wiley‐VCH. (c) DNA‐mediated clustering of complementary ssDNA‐functionalized p‐GUVs into prototissues. Reproduced with permission [115]. Copyright 2024, Wiley‐VCH.
Micrometer‐sized capsules were also combined with a matrix to generate prototissues with multiresponsiveness in aqueous medium [114]. A sacrificial approach based on CaCO_3_ microparticles incorporating cargos, membrane building blocks, and crosslinking agents has been used to produce protocells and support the interactions between their membranes and the surrounding gelatin matrix (Figure 7b) [114]. Cargos including enzymes and nanoparticles were loaded inside the protocells, while aminogelatin covalently crosslinked (using PEG‐bis‐NHS as crosslinker), served to generate the membrane. In addition, protocells containing entrapped tannic acid and carboxymethyl dextran supported a spherical architecture with sizes close to those of natural cells. Protocells population was then introduced in a gelatin matrix, which upon cooling, arrested them via physical crosslinking, resulting in a model for prototissue. Such prototissues retain their integrity over a wide pH range (3–10) for 7 days without breaking or dissolution and are able to respond to specific physical stimuli [114].
The complexity of prototissues can be further increased by incorporating multicompartment protocells to support intra‐ and intercellular communication in a closer‐to‐nature manner than unicompartment‐based prototissues. Furthermore, multicompartment‐based and unicompartment‐based protocells were clustered together in prototissues using DNA hybridization between complementary cholesterol‐tagged‐ssDNA exposed at their external interface (Figure 7c) [115]. Both types of protocell were produced by self‐assembly of poly(2‐methyl‐2‐oxazolineblock‐ poly(dimethylsiloxane)‐blockpoly(2‐methyl‐2‐oxazoline) (PMOXA‐b‐PDMS‐b‐PMOXA) amphiphilic copolymers in the presence of desired molecules and nano‐assemblies. Stimuli‐responsive membrane‐less AOs were produced by equipping the multicompartment protocells with ionophores in their membranes for diffusion of divalent ions and incorporating pH‐sensitive microgels for storing Mg^2+^. Unicompartment protocells consisted of p‐GUVs permeabilized by ionophores, and loaded with a Mg^2+^‐sensitive dye as a reporter molecule for the presence of Mg^2+^. By clustering together the two types of protocell through DNA hybridization, prototissues were produced with a stability of days [115]. Thus, these examples represent models that indicate the potential of such complex materials, and pave the way for developing more sophisticated systems.
Synthetic Polymers as Building Blocks for Protocells
3
The design of protocells relies on the availability of polymeric materials that are capable of forming stable, semipermeable, and functionally responsive compartments. Depending on their synthesis route, chemical composition, and self‐assembly behavior, polymers used in protocell fabrication can be broadly classified as: homopolymers; amphiphilic block copolymers and terpolymers; polyelectrolyte or hybrid systems. We present below relevant examples of polymers that served for formation of protocells together with their synthesis methods.
Homopolymers
3.1
Homopolymers, composed of a single type of repeating monomer, are generally synthesized through conventional radical, condensation, or ring‐opening polymerization (ROP). Their chemical simplicity allows reproducible preparation and straightforward functional modification, which are essential for constructing robust and semipermeable compartments. Homopolymers are typically synthesized through free‐radical polymerization (FRP) or ROP. FRP involves the initiation, propagation, and termination of radical species, enabling large‐scale production of simple linear polymers such as poly(allylamine hydrochloride) (PAH) [116], and poly(acrylic acid) (PAA) [117]. ROP, used for the synthesis of PLL, proceeds via nucleophilic opening of cyclic N‐carboxyanhydride (NCA) monomers, allowing the formation of well‐defined polypeptides with controlled MWs [118]. These homopolymers serve as polyelectrolytes in coacervate droplets and LbL assemblies, where their charge and hydrogen‐bonding capacity stabilize semipermeable membranes (Table 1) [76, 88, 119].
Amphiphilic Block Copolymers and Terpolymers
3.2
Amphiphilic block copolymers are the main structural components of p‐GUVs. These copolymers are usually synthesized through controlled/living polymerization methods such as anionic polymerization, atom transfer radical polymerization (ATRP), reversible addition‐fragmentation chain transfer (RAFT), and ROP. Anionic polymerization proceeds through highly reactive carbanion intermediates, and produces low‐dispersity polymers such as PB‐b‐PEO [34, 120]. ATRP, a reversible redox‐controlled radical technique, enables the fine MW control, necessary for example for preparing poly(ethylene glycol)‐b‐polystyrene (PEG‐b‐PS) [121]. RAFT polymerization, which uses thiocarbonylthio chain‐transfer agents, allows excellent control over radical polymerization and underpins PISA, a one‐pot strategy where polymer growth and self‐assembly occur simultaneously. PISA has also been extended to photo‐PISA and bio‐PISA, thus enabling the mild, scalable, and aqueous synthesis of vesicular architectures, such as poly(glycerol monomethacrylate)‐b‐poly(2‐hydroxypropyl methacrylate) (PGMA‐b‐PHPMA) and poly(ethylene glycol)‐b‐poly(2‐hydroxypropyl methacrylate) (PEG‐b‐PHPMA) (Table 1) [21, 41, 42, 43, 44, 45, 46, 59]. For example, myoglobin (Mb) has been employed in bio‐PISA as a biocatalyst for ATRP of hydrophilic monomers such as poly(ethylene glycol) methacrylates in aqueous media. As polymer chains extend, their increasing amphiphilicity induces in situ self‐assembly into vesicles, which spontaneously encapsulate the active enzyme catalyst. However, the relatively limited number of monomers that can be used in PISA and its extended versions, is an impediment to enlarging the properties of the resulting GUVs. ROP which proceeds via nucleophilic opening of cyclic NCA monomers is a versatile method for synthesizing polypeptides and related polymers with controlled MWs; a typical example is PLL, which has been widely employed in protocell fabrication [122]. This method yields well‐defined amphiphiles that are capable of forming p‐GUVs with tunable membrane properties and functional responsiveness.
Compared to diblocks, triblock copolymers and terpolymers provide enhanced stability and support multifunctionality. Examples include poly(2‐methyl‐2‐oxazoline)‐b‐poly(dimethylsiloxane)‐b‐poly(2‐methyl‐2‐oxazoline) (PMOXA‐b‐PDMS‐b‐PMOXA), synthesized via cationic ROP (CROP) of 2‐methyl‐2‐oxazoline from PDMS macro‐initiators [50, 100, 101, 103, 115], and poly(ethylene glycol)‐b‐poly(caprolactone‐gradient‐trimethylene carbonate)‐b‐poly(glycolic acid) (PEG‐b‐PCLgTMC‐b‐PGA), obtained through multistep ROP [81, 88]. The commercial Pluronic L121 [poly(ethylene oxide)‐b‐poly(propylene oxide)‐b‐poly(ethylene oxide)], prepared by anionic polymerization of propylene oxide followed by ethylene oxide, assembles into semipermeable membranes capable of hosting coacervate microdomains [95, 96].
Polyelectrolytes and Hybrid Systems
3.3
Polyelectrolytes and hybrid polymers are integral to the preparation of LbL capsules and coacervate‐based protocells. Their synthesis typically involves FRP, ATRP, or NCA‐ROP for preparing charged or functionalized polymers. Systems such as PAH/poly(styrene sulfonate) (PSS) and poly(L‐lysine) (PLL)/poly(methacrylic acid) (PMA) rely on electrostatic complexation to form multilayered structures with tunable permeability [76, 85, 86, 108, 109, 119].
Cationic polyelectrolytes such as poly(diallyldimethylammonium chloride) (PDDA) are obtained by FRP of diallyldimethylammonium chloride, and form coacervates with negatively charged biomolecules. [102] Modified natural polymers, including diethylaminoethyl‐dextran (DEAE‐dextran) and carboxymethyl‐dextran (CM‐dextran), are produced through postfunctionalization of dextran backbones to adjust charge density [102]. Similarly, amylose derivatives, such as quaternized amylose, carboxymethylated amylose, and amylose conjugates, are synthesized through quaternization, carboxymethylation, or carbodiimide‐mediated coupling to provide metal‐binding or pH‐responsive functionality [85, 88].
Hybrid systems combine synthetic and biological polymers to create dynamic and responsive protocell membranes. Examples include PMOXA‐b‐PDMS‐b‐PMOXA mixed with PDMS–heparin, which combined synthetic elasticity with bioactivity [100, 101, 103], and spiropyran‐functionalized PB‐b‐PEO, which was obtained via anionic polymerization followed by photo‐switch functionalization and exhibited light‐triggered membrane permeability changes [97].
Communication Between Protocells and Their Environments
4
Cells have a variety of channels for possible communication with the environment. They can sense physical signals (temperature, light, mechanical cues) or chemical signals (pH, presence of specific molecules) by a specific response, either in terms of membranes changes or exchange of molecules as the basis for inter‐ and intracellular communication. In order to mimic natural cells, protocells must be able to adapt to a dynamic environment and implement responses to chemical or physical signals. Therefore, in order to mimic the ability of natural cells to communicate with their extracellular matrix or other cells, it is essential for protocells to be able to communicate with their environment by uptake and/or exchange of molecules and signals.
Unicompartment Protocells: Communication With Their Environment
4.1
Generally, protocells have allowed the study of important biochemical processes, such as cell‐free protein synthesis [126, 127], DNA replication [128], metabolism [129], and cytoskeletal functions [130], when signals or substrates in the environment reach their inner compartments containing reconstituted functional molecules. There are two main directions of research that study how the communication of unicompartment protocells with the environment takes place. In response to external stimuli, protocells can: (i) change their shape/properties and (ii) initialize an internal reaction/interaction (outside‐in signaling).
The ability to change membrane morphology is based on the chemical nature of the copolymers, which support membrane manipulation in the presence of external stimuli (e.g., temperature, pH change). For example, by incorporating a temperature‐responsive polymer PDMA_30_‐b‐PNIPAM_200_‐BODIPY inside the membrane of nonresponsive poly(butadiene)‐block‐poly(ethylene oxide) (PBD‐b‐PEO) vesicles, it was possible to change the membrane shape without rupture or with self‐division when the environmental temperature changed [125]. At ambient temperature (28°C) and below the lower critical solution temperature (LCST), p‐GUVs had a spherical shape, but after heating to 34°C (i.e., slightly above the LCST), significant deformation was observed, and self‐division eventually resulted (Figure 8a) [125]. Mechanical fluctuations in the polymer membrane induced morphological changes that upon cooling further modulated the membrane shape. Similarly, protocell membranes based on the cross‐linked diblock copolymer PEO‐b‐P(NIPAAm‐r‐NAPMAm‐r‐NAPMAmRu(bpy)3‐r‐NAPMAmMA) exhibited unique size and shape changes upon variation of the environment temperature [131]. Increasing the temperature induced tangential stress in the p‐GUV membrane due to expansion caused by hydration of the NIPAAm thermo‐responsive segment. The membrane temporarily buckled inward to relax the tension and finally relaxed to an unbuckled equilibrium state with a larger diameter. Another approach to changing membrane properties in response to the presence of environmental signals was realized by swelling of the compartment due to pH‐response of the copolymers [132], or by introducing carboxylic acid terminated poly(N‐isopropylacrylamide) (PNIPAm) inside the membrane of poly(ethylene glycol)‐polystyrene (PEG‐b‐PS) vesicles and exploiting the co‐nonsolvency phenomenon [121]. However, as polymer membranes are significantly more stable than those based on lipids, it is more difficult to induce dynamic shape changes in p‐GUVs, and there are fewer reports on shape changes of p‐GUVs than for lipid‐based protocells [133].
*Environmental communication in unicompartment protocells. (a) Thermoresponsive polymersomes exhibiting reversible membrane deformation and self‐division upon temperature cycling. (top) Confocal image showing membrane deformation of PDMA30‐b‐PNIPAM200‐BODIPY : PDMA30‐b‐PNIPAM200 (1:100, 90 mg mL−
- polymersomes embedded in a PBD51‐b‐PEO27 membrane with 10 mol % cholesterol during heating (28–40°C). (bottom) Confocal image showing self‐division during cooling (39–28°C). Reproduced with permission [125]. Copyright 2022, Wiley‐VCH. (b) Enzyme‐loaded polymeric GUV microreactors prepared via PISA. (top) Single‐enzyme GUVs encapsulating myoglobin (Mb) catalyze peroxidase reactions to generate fluorescent resorufin. (bottom) Dual‐enzyme GUVs co‐encapsulating β‐galactosidase (β‐Gal) and Mb perform sequential hydrolysis and peroxidase reactions, producing fluorescein and resorufin. Scale bar: 5 µm. Reproduced with permission [43]. Copyright 2024, Springer Nature. (c) DNA‐based communication in GUVs containing DNA nanopores. (top) Schematic illustration of DNA duplex formation inside GUVs. (bottom) CLSM images showing GUVs encapsulating enzyme, Sybr Green (SG), and ssDNA template at different time points after addition of dNTPs and primers to the external solution. Reproduced with permission [50]. Copyright 2023, Wiley‐VCH.*
Outside‐in signaling triggers reactions/changes of properties inside unicompartment protocells based on the activity of encapsulated/entrapped biomolecules (proteins, enzymes, mimics thereof). Substrates and co‐factors present in the environment act as chemical signaling components, diffuse through the membrane to reach the biomolecules, and trigger a reaction. Molecular flow is achieved by producing protocells with intrinsically porous membranes [14], or by stimuli‐responsive membranes that became permeable on detecting the presence of a stimulus in the environment [132]. For example, p‐GUVs based on different molar ratios of triblock PEO_5_‐b‐PPO_67_‐b‐PEO_5_, PEO_4_‐b‐PPO_58_‐b‐PEO_4_ and PEO_2_‐b‐PPO_30_‐b‐PE_2_ showed a differential permeability to ions and small dye molecules, due to differences in membrane thickness [134]. A biomimicry strategy involves rendering synthetic membranes permeable by inserting membrane proteins/pores, and size‐selected molecular diffusion is achieved with ionophores (e.g., calcimycin [135], and N, N‐dicyclohexyl‐ N', N‘’‐dioctadecyl‐3‐oxapentane‐1, 5‐diamide [136]), biopores (e.g., gramicidin [56], melittin [58], DNA‐origami [50]), protein channels (OmpF [13] or bo3 oxidase [137, 138]) or α‐hemolysin [139, 140]. In order to overcome the hydrophobic mismatch between pore length and membrane thickness, the hydrophobic domain of the membrane should be highly flexible, as for example in the case of PMOXA‐PDMA membranes [56], or the curvature of the membrane should be decreased [58]. Furthermore, protein insertion influences the properties of the p‐GUV membrane, for example, by decreasing the bending rigidity and increasing the fluidity of the membrane [137]. In order to mimic the incredible variety of cellular reactions that start when a signal from the environment reaches the cell, several types of biomolecules have been used as models inside protocells, including horseradish peroxidase (HRP) [13], β‐Gal [97], GOx [47], and Mb [43].
Protocells with light‐activated membrane permeability allowed substrates present in the external medium to cross the membrane upon light irradiation and trigger an internal enzymatic reaction [97]. Spiropyran‐based permeability modulators perturbed the PB‐PEO polymer membrane by photo‐isomerization, allowing the hydrophilic ATP substrate to reach the GUV cavity, where condensation of PLL was induced, and internal coacervate droplets were formed. Also, when fluorescein di(β‐d‐galactopyranoside) as a substrate of β‐Gal diffused through the light‐permeabilized membrane of the p‐GUVs, the encapsulated enzyme was activated, and FITC was produced [97]. In a different approach, p‐GUV membranes based on PMOXA‐PDMS‐PMOXA copolymers were permeabilized by inserting the outer membrane protein F (OmpF), which allowed substrates (H_2_O_2_, Amplex UltraRed (AR)) as chemical signals to diffuse into the protocell and trigger the in situ reaction of HRP [13]. Protocells with GUV architecture have also been produced by bio‐ATRP using Mb as the active factor, and then performed an orthogonal activity with an encapsulated second enzyme, β‐Gal (Figure 8b) [43]. However, the efficiency of β‐Gal was lower in the p‐GUVs (β‐Gal retaining only 40% of its activity) than that of the free enzymes, due to deactivation during bio‐PISA and diffusion limitations imposed by the membrane [43].
In a microfluidic approach, p‐GUVs, consisting of a bilayer membrane of PEG‐b‐PLA diblock copolymers with an excess of PLA homopolymer in the hydrophobic region of the bilayer, allowed the ribosomal machinery to function and express in situ the prokaryotic cytoskeletal actin‐like protein, MreB [122]. The flux of water induced by an osmotic pressure difference between the inner core of the p‐GUVs and the environment caused swelling and formation of pores, which resulted in the release of expressed proteins analogous to signaling‐out ability of natural cells [122]. High reaction efficiency for different mixtures of biomolecules inside PMOXA‐b‐PDMS copolymers p‐GUVs has been achieved by using a different microfluidics approach together with postpermeabilization of the p‐GUV membrane by inserting DNA‐origami pores (Figure 8c) [50]. These protocells supported in situ production of double‐stranded DNA [50]. Mixtures of the DNA single strand (ss‐DNA) template and Klenow fragment of DNA polymerase I were co‐encapsulated during p‐GUV formation and served for DNA duplex formation when an externally added primer entered the protocell and dimerized with the ssDNA template.
LbL capsules with encapsulated enzymes that function as unicompartment systems are able to communicate with their environment. For example, systems for the detection of lactate, oxygen, and glucose levels [141] have been produced by encapsulating lactate oxidase, peroxidase, or glucose oxidase and the respective sensitive dyes inside LbL capsules based on (PSS/PAH). In another (PSS/PAH) microcapsule containing malate dehydrogenase, diffusion of external NADH induced a decrease in local proton concentration caused by the in situ enzymatic reaction and a change in the fluorescence of the pH‐sensitive fluorescent dye (seminaphtharhodafluor (SNARF‐1)‐dextran) [142]. While LbL capsules enable convenient internal encapsulation of enzymes, their semi‐permeability for unspecific transport of small molecules can be a limiting factor when the protocells should allow only specific molecules to pass through their membranes.
Protocells based on coacervates stabilized by a terpolymer shell are also used to communicate with their environment. An increased variety of proteins sequestrated inside the coacervates can be obtained when coacervate droplets are modified by the addition of amylose coupled with nitrilotriacetic (NTA), which is well‐known to complex Ni^2+^ and reversibly bind polyhistidine‐tagged (His‐tagged) proteins [85]. Such modified coacervates were capable of efficient sequestration of different His‐tagged recombinant proteins and improved enzyme activity because the substrates, as chemical signals from the environment, diffused through the semipermeable polymer shell and reached the enzymes [85].
Multicompartment Protocells: Communication With Their Environment
4.2
In the case of protocells containing AOs, chemical signals from the environment either have to pass through two successive types of membrane in the case of protocells containing membrane‐bound organelles, or through the protocell membrane and the crowded space of their membrane‐less organelles. Diffusion through such boundaries and different milieu decreases enzyme activity compared to that of free enzymes. However, multicompartment protocells have a significant advantage in protecting the encapsulated biomolecules from harmful environment agents and therefore prolonging their in situ activity compared to that of the free biomolecules in bulk. Another key advantage results from the segregated reaction spaces of AOs inside the protocells, which allows to have inside a protocell different sensitive biomolecules acting without affecting each‐other.
Protocells With Membrane‐bound AOs: Environmental Communication
4.2.1
Light‐responsive protocells were produced by loading p‐GUVs with stimuli‐responsive polymersome‐based AOs in order to mimic the light‐triggered response of cells below the pigment epithelium of the eye [104]. Polymersome membranes containing light‐driven synthetic rotary motors [143], which destabilize the membrane upon illumination with light at 430 nm, induced the release of cargo from the AOs. It has been demonstrated that such AOs preserve their integrity upon encapsulation by the microfluidics approach inside the formed protocells, even at a very high encapsulation efficiency, and then release their cargo only upon illumination [104]. However, to create such light‐responsive protocells, it is essential to ensure that the AOs are encapsulated in the p‐GUVs without having their functionality compromised, and with the integrity of the protocells being preserved upon illumination.
Protocells based on catalytic capsosomes were constructed using the LbL method by integrating β‐lactamase‐loaded liposomes between PLL precursor layers and capped with a poly(methacrylic acid)‐co‐(cholesteryl methacrylate) (PMAc) layer. They demonstrate activity inside the liposomes, as substrates could diffuse through the permeable polymer layers [109]. Furthermore, encapsulation of tyrosinase in liposomal subunits of capsosomes inhibited melanoma cell growth as the protocells maintained their catalytic activity in the vicinity of melanoma cells and under conditions of intratumor shear stress [10]. Having enzymes inside organelles mimics the architecture of cells and preserves their long‐term activity by protecting them from harmful environmental factors.
Protocells With Membrane‐less AOs: Environmental Communication
4.2.2
Enzymatic reaction cycles dynamically regulated transient formation and dissolution of coacervates inside protocells, as a model of out‐of‐equilibrium systems [96]. Adding PEP as a chemical signal from the environment triggered an in situ PyK enzymatic reaction, which induced formation of ATP‐coacervates inside Pluronic protocells, while externally added glucose started the HK enzymatic reaction that dissolved the coacervates (Figure 9a) [96]. A pH change in the environment can also serve as a signal, as demonstrated by the release from ion‐permeable GUV protocells of Mg^2+^ ions entrapped inside membrane‐less AOs composed of stimuli‐responsive microgels [115]. Similarly, protocells containing membrane‐less AOs based on reduction‐sensitive nanoparticles loaded with cargo inside p‐GUVs were activated when a stimulus entered from the environment [100]; protocells were formed from a mixture of poly(2–methyl‐2‐oxazoline)5– *block–*poly(dimethylsiloxane)58– *block–*poly(2–methyl‐2‐oxazoline)5 (PMOXA_5_‐b‐PDMS_58_‐b‐PMOXA_5_) and PDMS_65_‐b‐heparin in the presence of specific molecules (enzymes, reporter compounds) and stimuli‐responsive nanoparticles containing enzyme substrates or ion channels. When the external signaling molecule dithiothreitol, (DTT) diffused across the p‐GUV membrane, it induced disassembly of the nanoparticles and subsequent release of their cargo, thereby triggering an enzymatic reaction, ion channel recruitment, or cytoskeleton polymerization [100, 101].
Environmental communication in multicompartment protocells. Signal‐driven formation and dissolution of nucleotide‐based coacervates within polymersomes. (a, b) ATP‐coacervates: formation by ADP + PEP and dissolution by ATP + glucose. (c, d) NADPH‐coacervates: formation by NADP⁺ + G6P and dissolution by NADPH + pyruvate. Scale bars: 100 µm. Reproduced with permission [96]. Copyright 2024, Wiley‐VCH.
Use of the microfluidics approach allows controlled production of p‐GUVs with micron‐sized coacervates containing sequestrated enzymes, which act as membrane‐less AOs [98]. For example, coacervates based on a mixture of polyethylene glycol (PEG) and dextran (DEX) polymers with the entrapped enzyme lactate dehydrogenase (LDH) efficiently produced lactate when pyruvate penetrated the p‐GUVs due to the permeability of their membranes [98].
An important approach in mimicking natural cells is to study how cell‐free proteins synthesis can take place in protocells. While cell‐free protein synthesis has been intensively studied in various lipid‐based protocells [144, 145, 146], it is still underexplored in polymer‐based protocells. An interesting example consists of porous protocells containing a nucleus‐like DNA‐hydrogel that successfully expressed and displayed proteins encoded by the DNA on diffusive supply of cell‐free transcription and translation (TX‐TL) reagents [99]. Their large pores even allowed ribosomes to diffuse into these protocells and support the DNA‐like nuclei for the synthesis of proteins. In this manner, synthetic protocells were able to communicate with their environment both through signals entering (the necessary transcription factors), and by production and release of diffusive protein signals. The protocells were highly stable, preserving their integrity and retaining full expression capabilities after 2 years of storage. [99] In addition, the role of the protocells membrane composition in membrane protein folding and production it has also been studied [147].
In order to mimic the behavior of different organelles, protocells have been produced based on capsosomes equipped with both liposomes and hydrogel polymeric capsules [148]. Separation of these different AOs by a poly(N‐vinyl pyrrolidone‐b‐(cholesteryl acrylate) (PVPc) layer enabled the generation of “free‐floating” liposomes when destabilized by a change in pH. Control over the spatial positioning of AOs simulated the arrangement of natural organelles inside eukaryotic cells. When reached by glutathione, hydrogel subcompartments disassembled, whereas liposome AOs loaded with β‐lactamase maintained their catalytic activity and converted nitrocefin into its red hydrolyzed product [148].
Intracellular Communication
5
Intracellular communication is essential to support the complex metabolic pathways that require exchange of molecules between different organelles initiated by detection of a specific stimulus or molecule. While unicompartment‐based protocells allow simple internal cascade reactions that are triggered by an environmental stimulus, multicompartment protocells support distinct biochemical reactions in a more natural cell‐like manner by employing spatial separation of biomolecules and reactions in different domains [149]. In addition, the presence of internal AOs helps protect incompatible components from degradation or cross‐talk inside the protocells [150]. There are different intracellular communication paths depending on the inner architecture of the protocells. In the case of unicompartment protocells, only reactions between biomolecules co‐encapsulated inside can be studied. Multicompartment protocells allow a more complex scenario of reactions by segregating the biomolecules inside different AOs. We present first how intracellular communication takes place inside unicompartment protocells and then how it take place inside multicompartment protocells.
Intracellular Communication in Unicompartment Protocells
5.1
The simplest manner to mimic intracellular communication involving unicompartment‐based protocells is to explore internal cascade enzymatic reactions that start when a signal from the environment reaches the enzyme involved in the first step of the reactions. By co‐encapsulating two/three types of enzyme inside p‐GUVs, a cascade reaction can be observed upon detection of a specific chemical stimulus, which serves as an out/in signal. For example, a cascade involving tandem of glucose oxidase (GOx) and Mb was successfully established in p‐GUVs generated by bio‐PISA [43]. The first step of the reaction involved glucose entering the protocell and triggering the GOx‐catalyzed production of glucono‐1, 5‐lactone and H_2_O_2_, while the second step was a peroxidase reaction that was catalyzed by Mb in the presence of that H_2_O_2_ produced in the first reaction step. (Figure 10a) [43]. However, this cascade reaction efficiency was lower than with free enzymes because of limitations related to enzyme deactivation in bio‐PISA conditions, and limited diffusion of glucose through the p‐GUV membrane.
Intracellular communication of unicompartment and multicompartment protocells. (a) Multicompartment catalysis in polymersome‐in‐GUV artificial cells. Different enzymes encapsulated in PS‐b‐PIAT polymersomes and cytosolic components were co‐encapsulated within PB‐b‐PEO GUVs, enabling cascade reactions monitored by fluorescence spectroscopy. Reproduced with permission [43]. Copyright 2024, Springer Nature. (b) Capsosomes containing trypsin (LTRP) and horseradish peroxidase (LHRP) catalyze the conversion of BA‐Rho‐110 and Amplex Red into fluorescent products MA‐Rho‐110 and resorufin, respectively, with reaction kinetics monitored by fluorescence spectroscopy. Reproduced with permission [111]. Copyright 2017, Wiley‐VCH.
A significant increase in activity of protocells has been obtained by using a double emulsion microfluidic approach to produce p‐GUVs loaded with three different enzymes [47]. The first enzyme, β‐Gal, hydrolyzed fluorescein‐di‐β‐D‐galactopyranoside (FDG) into fluorescein and galactose, then the second enzyme, GOx, oxidized the galactose and produced H_2_O_2_, which was used by HRP to oxidize Amplex Red (AR) to resorufin. This cascade has been evaluated in different topologies: encapsulating just one enzyme at the time inside the GUVs, co‐encapsulating two enzymes inside the GUVs with the third free in solution, and co‐encapsulating all three enzymes in the same GUV; these allowed the limiting factors to be determined by having environmental signals in different steps of a cascade reaction taking place partially inside unicompartment protocells [47].
A different model for intracellular communication has been achieved by sequestering enzymes inside membranized‐coacervates [88]. For example, with a PEG‐PLCgTMC‐Pglu membrane enclosing cell‐sized amylose‐based coacervate microdroplets through the electrostatic interaction, GOx and HRP, prior incorporated in the coacervates, were able to act in tandem when glucose from the environment diffused through the permeable polymer membrane. The coacervates serve as the crowded environments of model cytosols, while the polymer shell played the role of the cell membrane boundary [88]. Even when enzymes do not have a natural affinity for each other, their recruitment inside coacervates demonstrates that they perform better when brought together in a confined environment [85]. Protocells containing coacervates with internal tryptophan anhydrase (TnaA) and flavin‐containing monooxygenase (FMO) have been shown to perform a two‐step cascade reaction converting L‐tryptophan (L‐Trp) to indigo [85]. When L‐Trp and the pyridoxal‐5‐ phosphate (PLP) cofactor diffused through the terpolymer membrane of the protocell and reached the TnaA inside the coacervate, this was converted to indole, which was then further oxidized by FMO with the consumption of nicotinamide adenine dinucleotide phosphate (NADPH) added in the process [85]. Importantly, co‐sequestration of enzymes in the same coacervate supports cascade reactions by the close proximity of enzymes, and the only limiting factor is the semipermeability of the polymer border of the protocells.
Intracellular Communication in Protocells Containing AOs
5.2
Various protocells containing AOs and/or different biomolecules have been reported for investigating a specific feature/process that occurs inside natural cells, or to indicate how protocell architecture affects the efficiency of the signaling process. To mimic the natural diversity of cells, protocells were developed by encapsulation of: (i) AOs with single functionality, and (ii) AOs with different functionalities that are working in tandem through cascade reactions. p‐GUVs loaded with different types of membrane‐bound AO (nanocapsules [106, 107], polymersomes [14, 104]) have been proposed as protocells that communicate with their environment and integrate intracellular communication by in situ cascade reactions that are triggered by internal diffusion of specific substrates or signals. The first polymersome‐loaded protocells involved a three‐step cascade reaction involving encapsulated enzyme‐loaded PS‐b‐PIAT polymersomes within PB‐b‐PEO p‐GUVs [14]. The enzymes alcohol dehydrogenase and CalB or alcalase were either separately encapsulated in permeable polymersomes to serve as AOs, or free inside the cavity of the protocells, together with phenylacetone monooxygenase (PAMO) which was used as a supplementary cytosolic enzyme. The efficiency of the cascade reaction in different spatial arrangements of the enzymes (inside polymersomes or free in the cavity of the protocells) depended strongly on their specific location because of the decrease in diffusion of the molecule when it has to pass through the polymersome membranes [14]. Similarly, porous p‐GUVs based on PS‐b‐PIAT supported communication between GOx, which was able to move freely in the protocell medium, and AOs‐containing HRP encapsulated in PMOXA‐b‐PDMS‐b‐PMOXA polymersomes equipped with OmpF [151]. The same cascade reaction has also been studied in protocells equipped with AOs based on AMSNs loaded with the enzymes [106]. HRP‐AMSNs and GOx‐AMSNs were co‐encapsulated in (PB‐b‐PEO) PB‐b‐PEO p‐GUVs resulting in multicompartment protocells. The semipermeable silica nanocapsules then allowed rapid diffusion of substrates and products for each step of the cascade reaction, and finally, resorufin, the final product of the cascade reaction, diffused from the p‐GUVs, thus mimicking an “out” communication of the protocells [106, 107].
The complexity of reactions inside capsosomes can be increased by loading liposomal subcompartments with different enzymes to support cascade reactions. While the subcompartments have the same chemical nature (liposomes), they have different functionality induced by the specificity of the encapsulated enzymes, including two single‐enzyme conversions in for example, parallel using HRP and trypsin [110], or tandem with GOx and HRP (Figure 10b) [111]. Due to the straightforward manner of combining liposomes loaded with different enzymes inside the same capsosome system, two different cascade reactions involving a total of five enzymes have been demonstrated in a system that is closer to mimicking metabolic pathways [152]. β‐Gal‐GOx‐CAT enzymes supported the first cascade reaction by using a lactose substrate to trigger the first reaction step, while glutamate dehydrogenase (GDH) and glutathione reductase (GTR) used NADP^+^/NADPH as a common self‐renewable cofactor in cyclic reactions [152].
In order to generate a closer‐to‐nature protocell the compartments were loaded with cytoplasm mimicking internal milieu. For example, amylose‐based coacervate microdroplets loaded with polymersomes as AOs were further stabilized with an external PEG‐PLCgTMC‐Pglu polymer membrane [88]. Then the same tandem of enzymes, GOx and HRP, located either in separate polymersomes or co‐encapsulated in one polymersome, performed the cascade reaction in the crowded coacervate environment [88]. However, segregation of enzymes in separate polymersomes reduces the cascade reaction efficiency as a result of various factors, including slower diffusion in the crowded coacervate, increased distance that H_2_O_2_ has to travel from the GOx‐AOS to the HRP‐AOs, and the presence of two polymer membranes of polymersomes that H_2_O_2_ has to pass through.
The requirement of exchange of molecules between different types of organelles that is required for intracellular communication can be achieved by co‐encapsulating/entrapping two types of AO with different functionality that work in tandem. For example, a double‐emulsion microfluidic approach can be used to generate protocells co‐loaded with two types of AO based on polymersomes, each of which is specifically enriched with components to provide different key functionalities [104]. In an example, one type of AO was photoresponsive, resulting from PDMS‐b‐PMOXA‐based polymersomes equipped with membrane‐embedded synthetic rotary molecular motors for light‐activated cargo release, and the second type had catalytic activity by encapsulating an enzyme (β‐Gal) in polymersomes and embedding biopores to render the membrane permeable for diffusion of substrates and products [104]. Thus, by co‐encapsulating the photoresponsive and catalytic AOs inside a single artificial cell, a simple intracellular signaling pathway was established upon illumination with a specific wavelength. Overall, this cell functions by light‐inducing rotation of the molecular‐motors inside the membrane of the responsive AOs, resulting in release of the substrate FDG that diffuses through melittin pores into a second catalytic AO with encapsulated β‐Gal. The β‐Gal then hydrolyzes FDG to the fluorescent fluorescein product that is released from the second AO, thus providing a simple intracellular signaling pathway that is triggered by light illumination.
In a more complex protocell composition scenario, coacervate‐in‐coacervate multicompartment protocells have been developed by loading different enzymes inside two immiscible coacervate phases with distinct physical and chemical properties [102]. GOx, HRP, and catalase (CAT) were simultaneously immobilized either in the inner coacervate microdroplets composed of polydiallyldimethyl ammonium/deoxyribonucleic acid or the outer coacervate formed by deposition of LbL layers of carboxymethyl‐DEX and diethylaminoethyl‐DEX. The enzymes then performed competitive cascade enzymatic reactions involving GOx and (HRP) or CAT. Notably, different efficiencies of the cascade reaction were observed as a result of the spatial location of the enzymes, the associated competitive reaction (HRP/CAT), and the different fluxes of substrates and products between the coacervate phases.
Intercellular Communication
6
Intercellular communication represents one of the essential mechanisms for transmission/exchange of signals between cells for supporting life in tissues or organs. This requires cells to be in close proximity or in contact so that the intercellular matrix does not block the signal transfer. In addition, intercellular communication is affected by the presence of specific molecules in the cellular environment or/and by their topological arrangement. Protocells developed to mimic intercellular communication are constructed and arranged to allow signals/molecules originating from a “sender” protocell to be transmitted to and processed by the “receiver” cell (Figure 1). Then they are brought together either by mixing without contact in the same solution or by linking together to mimic tissues, and support cascade reactions that mimic intercellular signaling/molecular transfer. Molecular transfer between sender and receiver cells involves diffusion through both protocell borders favored by a semipermeable membrane [9, 153], or through biopores integrated into the membrane [26, 98]. Protocells based on polymers, which are significantly more stable, have very recently started to be used for advanced intercellular communication.
Segregating enzymes in independent protocells induces a complex scenario at the molecular level because: (i) it involves molecular passage through multiple boundaries (protocell membranes/boundaries), (ii) it affects subsequent reactions from the cascade by a dilution effect, which occurs when products leave the sender and enter the bulk environment, only a fraction of which eventually reach the receiver, (iii) it is related to the probability that molecules enter in the “right” protocell (i.e., that substrates meet the appropriate enzyme, which allows the sequence of the cascade to operate), and (iv) the creation of a nonhomogeneous concentration of substrates for subsequent reactions inside receivers.
Unicompartment Protocells: Intercellular Communication
6.1
Providing conditions in which the relative amounts of enzymes inside specific protocells are tightly controlled is vital for maintaining the stoichiometry between enzyme concentrations and mimicking complex cell‐cell communication. In this respect, GUVs formed via the double emulsion microfluidic approach overcome the issues of low and distributed concentrations of enzymes [154]. A key example of a three‐enzyme cascade reaction studied in different combinations of enzyme‐loaded GUVs established how intercellular communication can be affected by segregation of enzymes in different spatial organizations of GUVs (Figure 11a) [47]. Full segregation of the enzymes (β‐Gal, GOx, and HRP) reduced significantly the amount of the products from each step of the reaction compared to the arrangement in which all three enzymes were co‐encapsulated inside one GUV. This difference is due to diffusion through the p‐GUV membranes, which results in all involved molecules having a 66% chance of diffusing into a wrong p‐GUV. By using other arrangements, involving enzyme couples being co‐encapsulated in one protocell and the third in another, it was possible to determine configurations to avoid because of competing reactions. Notably, the optimum reaction conditions were obtained when the first two enzymes were co‐encapsulated, because H_2_O_2_ rapidly passes through the membranes, and production of the final reaction product (resorufin) is maximized [47].
Intercellular communication between unicompartment protocells. (a) Three‐enzyme cascade reactions within polymeric GUVs (PDMS26‐b‐PMOXA9) prepared by microfluidic double emulsion. (ai) Schematic of the three‐step cascade reaction. (aii) Control GUVs containing individually segregated enzymes (β‐Gal, GOx, HRP). (aiii, aiv) Time‐dependent fluorescence of fluorescein and resorufin in single‐enzyme GUVs. Scale bar: 50 µm. Reproduced with permission [47]. Copyright 2017, Wiley‐VCH. (b) Schematic (bi), confocal images (bii), and time‐dependent quantification of Cy5 fluorescence (biii) showing DNA signal exchange between coacervate protocells containing distinct supramolecular nanoscaffolds. The redistribution of fluorescence signals (Cy3, FAM, and Cy5) after addition of a DNA fuel strand demonstrates successful transfer and activation of a DNA‐based signal between neighboring protocells. Reproduced with permission [155]. Copyright 2020, American Chemical Society.
A complementary approach involves intercellular communication between membranized‐coacervate protocells, which, when loaded with supramolecular nanoscaffolds, allow single‐stranded DNA to be trafficked between neighboring protocells (Figure 11b) [155]. Due to the semipermeable terpolymer membrane, the protocells received and transduced external signals, and effected changes in the local spatial organization and transmission of a DNA‐based reporter. In another example, membranized‐coacervates supported an intercellular cascade reaction involving GOx and HRP sequestrated in sender and receiver protocells, respectively [81]. H_2_O_2_ was transferred between sender and receiver and activated AR peroxidation by HRP inside the receiver protocells with 50% efficiency of the cascade reaction compared to that of the free enzymes. Thus, these examples indicate that diffusion of molecules/signals through the sender and receiver membranes are key factors that influence the efficiency of the intercellular communication between unicompartment protocells.
Multicompartment Protocell: Intercellular Communication
6.2
Intercellular communication between protocells with a multicompartment architecture represents a more‐complex scenario, because the molecular flow must pass through various barriers (organelle membranes, boundaries of different protocells and the milieu in which the protocells are situated). Various arrangements of sender and receiver protocells have been introduced: (i) involving one protocell type with a multicompartment architecture and the second as a unicompartment, and (ii) both protocells as multicompartments. An example involves two protocell populations, one as a unicompartment comprising Pluronic GUVs containing ADP, PyK and FITC‐DEX and a second based on PB‐PEO GUVs containing pre‐formed NADPH‐coacervate, LDH, and RITC‐DEX, which were mixed together to study intercellular communication [96]. External injection of PEP into the media only affected the Pluronic‐based sender cells and resulted in the formation of ATP within their interior via the in situ PyK enzymatic reaction. Pyruvate, as a byproduct of this reaction, then diffused out from the sender into the adjacent receiver, and triggered the dissolution of the NADPH‐coacervate within the PB‐PEO protocell (Figure 12a) [96]. When both the sender and receiver are multicompartments, it is necessary to have precise control of the subcompartmentalization architecture and timing of a specific sequence to promote overall functionality in response to stimuli. Such control can be achieved by using microfluidic approaches due to their capacity to produce homogeneous sized‐protocells with desired encapsulation efficiency of the AOs and the biomolecules involved in the communication reactions [156].
Intercellular communication between multicompartment polymeric protocells. (a) Enzymatic communication between Pluronic and PB–PEO polymersomes. (ai) Schematic illustration of PEP‐triggered pyruvate transfer and NADPH consumption. (aii) Fluorescence micrographs showing time‐dependent signaling between protocell populations. Scale bar: 100 µm. Reproduced with permission [96]. Copyright 2024, Wiley‐VCH. (b) Light‐triggered signaling between spatially separated protocells via DNA nanopores. (bi) Schematic of FDG release from sender protocells and enzymatic conversion to fluorescein in receivers. (bii) Normalized fluorescence of receiver protocells with or without DNA nanopores after 1 h co‐incubation. (biii) Fluorescence images showing fluorescein formation only with DNA nanopores. Scale bar: 10 µm. Reproduced with permission [104]. Copyright 2024, Wiley‐VCH. (c) Protein diffusion–based communication between synthetic protocells. (ci) Schematic and time‐lapse images showing TetR‐sfGFP (green) transfer from rhodamine‐B–stained senders (magenta) to receivers. (cii) Wide‐field image after 180 min showing TetR‐sfGFP spreading in a protocell network. Scale bars: 25 µm. Reproduced with permission [99]. Copyright 2018, Springer Nature. (d) Lactate‐mediated signaling between protocells. (di) Schematic of lactate production, diffusion, and signal generation. (dii) Time‐resolved CLSM profiles of resorufin formation in mixed protocell populations after addition of 0, 5, and 10 mM pyruvate. Reproduced with permission [98]. Copyright 2025, Wiley‐VCH.
In order to closely replicate the complexity of natural signaling cascades in vitro, sender and receiver protocells with AOs require a fine control of conditions to allow them to work in tandem. For example, sender and receiver protocells, both equipped with different biomolecules‐loaded polymersomes as AOs, have progressed toward the creation of an artificial retinal synapse [104]. Light‐responsive AOs were encapsulated inside sender protocells, while catalytic AOs were located inside receiver protocells, and both types of protocell had permeable membranes formed by insertion of DNA nanopores. A signaling molecule (FDG) was released from the membrane‐bound light‐responsive AOs upon illumination and diffused out from the sender protocell. Then, a fraction of FDG molecules reached the enzyme inside the catalytic AOs after passing through the membranes of both the protocells and AOs, and triggered the in situ reaction of β‐Gal (Figure 12b) [104] A key aspect of this reaction was the optimization of each type of protocell to ensure that there are biologically relevant distances between the respective AOs and protocells, and to promote an efficient cascade reaction. By choosing a 6:1 ratio between sender and receiver protocells, it was possible to achieve signal integration of multiple sender protocells to a single receiver protocell, which resembles photoreceptor signal transduction. In addition, changes in environmental calcium levels enabled modulation of the downstream signaling response of the enzyme activity in the receiver protocells.
The communication capabilities of polymer‐based protocells has been expanded to large macromolecules such as RNAs and proteins by using the microfluidic method to produce porous protocells (pores diameters of 200–300 nm) that contain a clay‐DNA hydrogel as a mimic of the nucleus and have the capability of gene expression [99]. Communication with each other directly through genetic regulators has been realized by a two‐stage activation cascade distributed into two separate protocell types: activator and receptor. Both type of protocells were able to perform cell‐free protein synthesis supported by their complex composition and the porous membrane. Activator protocells contained the template for expression of T3 RNA polymerase (T3 RNAP), while the reporter protocells contained the template for the T3 RNAP‐driven synthesis of the TetR‐sfGFP reporter and the array plasmids to capture the reporter protein (Figure 12c) [99]. Porous membranes of the protocells were permeable to macromolecules up to 2 MDa MW, and therefore supported diffusion from/in of T3 RNAP, the signaling molecule transmitting the instruction to express the reporter gene from activators to the reporter protocells.
In nature, the spatial distribution of senders and receivers determines the specificity of communication, and signal propagation is significantly affected by the distance between protocells [157], which represents a limiting factor for efficient intercellular communication. An arrangement of specifically organized sender and receiver protocells has established the roles of intercellular distance and their ratio in mediating controllable lactate signaling [98]. Micron‐sized coacervates inside p‐GUVs sequestrating LDH serve as membrane‐less AOs of sender protocells, and produced lactate when pyruvate diffused from the environment (Figure 12d) [98]. Lactate released from the sender reached the receiver protocells, where membrane‐bound organelles based on polymersomes and equipped with lactate oxidase (LOx), converted it to a detectable signal [98]. Pyruvate acted as a communication modulator because its concentration triggered the enzymatic reaction cascade, even at distances in the range 20 to 30 µm, which aligns with the effective range of paracrine signaling [98, 158, 159].
These examples emphasize the importance of molecular diffusion through membranes when cascade reactions take place between different populations of protocells.
Prototissues: Intercellular Communication
6.3
A more complex scenario of conditions is required to promote intercellular communication in prototissues generated by linking together different types of protocell, because the signals from sender protocells should pass through each protocell and possible organelle borders, as well as the regions between cells which mimic the extracellular matrix. Therefore, there are only a few examples of polymer‐based prototissues that have been designed to exhibit intercellular communication. Such prototissues may be based on populations of protocells with similar architecture (e.g., unicompartment) or with different architecture (uni‐ and multicompartment). Unicompartment protocells based on enzyme‐loaded polymeric microcapsules self‐organized in prototissue spheroids, and maintained a stable microenvironment based on homeostatic regulation through collective behavior [160]. In the presence of urea as chemical signal, urease‐loaded protocells with a weak, pH‐responsive polycation, poly(2‐(diethylamino)ethyl methacrylate) (PDEAEMA), shell produce ammonia and increase the pH. Conversely, the second type of protocells, GOx‐loaded protocells with a weak, pH‐responsive polyanion, poly(methacrylic acid‐co‐N‐butylmethacrylamide) (PMAA), shell generated gluconic acid when glucose was present in the environment, thus lowering significantly the pH. This feedback‐regulated communication between two types of protocell through pH signals, triggered a self‐protection mechanism in the prototissue [160].
Prototissues based on senders with multicompartment architecture and receivers as unicompartments have been generated by DNA hybridization of p‐GUVs exposing complementary ssDNA when each of them possesses a specific functionality [115]. In an example, sender p‐GUVs had internal pH‐responsive microgels as artificial Mg^2+^ storage AOs, while a downstream linked set of GUV‐based protocells contained monomeric actin (G‐actin) and filamin, an actin‐binding protein, which stabilized actin filaments when they form. The controlled release of Mg^2+^ from microgels in AOs in the sender protocells in response to external stimuli served as a second messenger in communicating with the downstream protocells, where actin was polymerized into filaments [115].
Various types of protocell have been integrated into macroscopic bilayer prototissues based on the hierarchical assembly of gelatin matrix and polymeric microcompartments loaded with cargos [114]. The inner layer consisted of aminogelatin‐PNIPAAm‐PMAA protocells loaded with Ni nanoparticles that were able to convert a laser light stimulus into heat and induce the release of starch from the substrate prototissue. The outer layer, acting as a digestive layer, consisted of the surrounding matrix loaded with 𝛼‐amylase and two protocell populations loaded with GOx and HRP, respectively. In this way, the surrounding matrix and protocells in the outer layer functioned cooperatively to digest the starch released from the substrate prototissue through a three‐step cascade reaction (𝛼‐amylase, GOx, and HRP) [114], which enhanced the modularity and tunability of the prototissues.
Conclusion and Outlook
7
Cells have a variety of internal channels for possible communication with one another and the environment, and they support metabolic processes through the collective transmission of signals and molecules within populations. In this review, we have shown how bottom‐up approaches can generate synthetic protocells that are able to mimic the communication channels in a simple and controlled manner with at least one of their building blocks based on polymers. We present in a higher‐complexity manner unicompartment and multicompartment synthetic protocells and include relevant examples to indicate different communication scenarios (with the environment, intracellular and intercellular). While unicompartment protocells contain various biomolecules that provide their functionality, multicompartment protocells have a larger variety of inner components (biomolecules and nanoassemblies serving as membrane‐less and membrane‐bound AOs), which support intra‐ and intercellular communication in a closer‐to‐nature manner. In addition, we introduce the first examples of prototissues in which polymer protocells were linked together for more advanced communication. A key advantage of polymer‐based protocells is their increased stability compared to lipid‐based protocells, which allows the study of functionality and communication channels for longer periods of time. Increasing the time, the protocells are functioning and communicating is essential for development of advanced solutions in medicine or biosensing. An essential requirement for using protocells intercommunication (e.g. in prototissues) is to preserve their integrity as segregated reaction spaces. In this respect, polymer‐based protocells represent ideal candidates as they do not fuse, and even can be rendered more stable by chemical modifications, which support their use as sophisticated multifunctional systems. Significant progress has been realized in the design and development of a large variety of multicompartment protocells, but these have been mainly used to show their potential with a limited number of biomolecules, and their functionality remains relatively undeveloped. One of the “golden” models for cascade reactions involves the couple GOx‐HRP to demonstrate the biocatalytic potential of multicompartment systems. Such simple cascade reactions performed in protocells exhibit only a limited resemblance to biologically relevant reactions, which are yet to be mimicked in a more complex way. In addition, specific responses, either in terms of membrane changes or secretion of molecules, as the basis for inter‐ and intracellular communication, are usually studied in solutions with simple compositions, which are far from mimicking those of extracellular matrices. Besides, inside a majority of reported protocells the medium is not mimicking the cytoplasm but is rather based on the buffers in which the protocells were generated. Therefore, the intracellular reactions are studied in simpler conditions than inside natural cells, which might induce a bias in the efficiency of communication. While communication with the environment has been extensively studied, intercellular communication and prototissue behavior are by far less explored, especially because of the complexity of supporting communication through multiple signaling pathways and limitations related to molecular diffusion through various borders and membranes. As an early‐stage research topic, there are several open questions that need to be addressed involving increasing the range of composition of protocells and diversifying their communication channel. A necessary step forward requires testing these polymer protocells in terms of biocompatibility and biodegradability as crucial for medical applications. In this respect, the integration of polymer protocells with natural cells is still underexplored and will induce bio‐limitations that should be solved before advancing these systems for translational applications. However, the advantage of bottom‐up approaches is that they are able to support the variety of natural cells and cellular communication by a diversity of protocell architectures, which make the same cascade reaction to run differently in terms of efficiency, space, and time control, thus adapting it for a desired application. In addition, the capability of molecular transport of molecules/signals through protocell borders to trigger in situ changes and reactions represents the basis for developing advanced medical applications.
Conflicts of Interest
The authors declare no competing financial interest.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1C. M. Waters and B. L. Bassler , “Quorum Sensing: Cell‐to‐Cell Communication in Bacteria,” Annual Review of Cell and Developmental Biology 21 (2005): 319–346, 10.1146/annurev.cellbio.21.012704.131001.16212498 · doi ↗ · pubmed ↗
- 2B. A. Doğaner , L. K. Q. Yan , and H. Youk , “Autocrine Signaling and Quorum Sensing: Extreme Ends of a Common Spectrum,” Trends in Cell Biology 26 (2016): 262–271.26671200 10.1016/j.tcb.2015.11.002 · doi ↗ · pubmed ↗
- 3Y. Hart , Y. E. Antebi , A. E. Mayo , N. Friedman , and U. Alon , “Design Principles of Cell Circuits With Paradoxical Components,” Proceedings of the National Academy of Sciences 109 (2012): 8346–8351, 10.1073/pnas.1117475109.PMC 336138522562798 · doi ↗ · pubmed ↗
- 4F. K. Balagaddé , H. Song , J. Ozaki , et al., “A Synthetic Escherichia coli Predator–Prey Ecosystem,” Molecular Systems Biology 4 (2008): 187, 10.1038/msb.2008.24.18414488 PMC 2387235 · doi ↗ · pubmed ↗
- 5W. Bacchus , M. Lang , M. D. El‐Baba , W. Weber , J. Stelling , and M. Fussenegger , “Synthetic Two‐way Communication Between Mammalian Cells,” Nature Biotechnology 30 (2012): 991–996, 10.1038/nbt.2351.22983089 · doi ↗ · pubmed ↗
- 6T. Danino , O. Mondragón‐Palomino , L. Tsimring , and J. Hasty , “A Synchronized Quorum of Genetic Clocks,” Nature 463 (2010): 326–330, 10.1038/nature 08753.20090747 PMC 2838179 · doi ↗ · pubmed ↗
- 7L. You , R. S. Cox , R. Weiss , and F. H. Arnold , “Programmed Population Control by Cell–Cell Communication and Regulated Killing,” Nature 428 (2004): 868–871, 10.1038/nature 02491.15064770 · doi ↗ · pubmed ↗
- 8H. Youk and W. A. Lim , “Secreting and Sensing the Same Molecule Allows Cells to Achieve Versatile Social Behaviors,” Science 343 (2014): 1242782, 10.1126/science.1242782.24503857 PMC 4145839 · doi ↗ · pubmed ↗