Hien
Phan
*a,
Vincenzo
Taresco
b,
Jacques
Penelle
a and
Benoit
Couturaud
*a
aUniv Paris Est Creteil, CNRS, Institut de Chimie et des Matériaux Paris-Est (ICMPE), UMR 7182, 2 rue Henri Dunant, 94320 Thiais, France. E-mail: Couturaud@icmpe.cnrs.fr
bSchool of Chemistry, University of Nottingham, University Park, Nottingham, NG7 2RD, UK
First published on 9th October 2020
Stimuli-responsive amphiphilic block copolymers have emerged as promising nanocarriers for enhancing site-specific and on-demand drug release in response to a range of stimuli such as pH, the presence of redox agents, and temperature. The formulation of amphiphilic block copolymers into polymeric drug-loaded nanoparticles is typically achieved by various methods (e.g. oil-in-water emulsion solvent evaporation, solid dispersion, microphase separation, dialysis or microfluidic separation). Despite much progress that has been made, there remain many challenges to overcome to produce reliable polymeric systems. The main drawbacks of the above methods are that they produce very low solid contents (<1 wt%) and involve multiple-step procedures, thus limiting their scope. Recently, a new self-assembly methodology, polymerisation-induced self-assembly (PISA), has shown great promise in the production of polymer-derived particles using a straightforward one-pot approach, whilst facilitating high yield, scalability, and cost-effectiveness for pharmaceutical industry protocols. We therefore focus this review primarily on the most recent studies involved in the design and preparation of PISA-generated nano-objects which are responsive to specific stimuli, thus providing insight into how PISA may become an effective formulation strategy for the preparation of precisely tailored drug delivery systems and biomaterials, while some of the current challenges and limitations are also critically discussed.
Typically, the room temperature self-assembly of amphiphilic block copolymers with a glassy hydrophobic block is usually performed using the co-solvent or solvent switch method.26 In this technique, the amphiphilic block copolymer is initially dissolved in an organic solvent such as DMF, acetone, or THF that is compatible with both blocks, followed by slow addition of a selective solvent, e.g. water, which is a good solvent for only one block, leading to the microphase separation and then to spontaneous self-assembly into nano-objects with a solvophobic core coated with a solvophilic layer.21 This allows a relatively low polymer concentration (<1 wt%) to be used.27–30 In addition, the self-assembly usually occurs via post-polymerisation of well-defined precursor copolymers, and hence is not economically ideal for scale-up and industrial environments. While the solution-phase self-assembly can still leverage a diverse set of particle morphologies such as micelles, nanofibers, and vesicles,31,32 which is primarily dictated by the packing parameter (p),26 in practice it is tedious to obtain higher order morphologies.
To meet such technical challenges, polymerisation-induced self-assembly (PISA) has recently been proposed. PISA has tremendously redefined the concept of block copolymer self-assembly by jointly facilitating polymerisation and self-assembly in a one-pot approach.33–40 Two types of polymerisation methods are used in PISA, dispersion and emulsion. The former requires both monomers and a macromolecular precursor to be solvophilic and compatible with each other in a continuous phase, either in an organic solvent or in water-based systems. Once a critical degree of polymerisation of the core-forming monomer is reached, this block becomes insoluble and tends to precipitate. The initial particles act as the loci for polymerisation on which monomers continue to be added covalently to the stabilised particles. Meanwhile, the second method is usually implemented in water, using a soluble homopolymer precursor that is chain-extended with an immiscible monomer to a critical degree of polymerisation. This leads to the formation of monomer droplets stabilised by the initial homopolymers. Both polymerisation methods generate in situ self-assembled nano-objects with various morphologies as depicted in Fig. 1, which are a very important and valuable asset compared to the typical nanoprecipitation technique through which mostly spherical particles are produced while the same copolymers produced by PISA can be shaped in several ways.41 The morphological evolution initially starts with the early formation of spherical micelles and provides a not-kinetically-frozen structure that evolves to cylindrical or worm-like nanoparticles, and finally to vesicles over the growing course of the core-forming block. The control over particle size and shape has been thoroughly discussed in recent reviews.42–49 Generally, the formation and transformation of NP morphologies are controlled and predictable. A higher order sphere-to-vesicle transition is often promoted by an increase in the degree of polymerisation (DP) and by modification of the reaction conditions. This feature facilitates the precise control of targeted particle shapes. In general, controlling the particle shape is of substantial importance to enhance the performance of drug delivery systems as each morphology has its own effects on the in vivo fate, biodistribution and cellular uptake after administration.50–52 More specifically, oblate particles exhibit a reduced uptake by macrophages and a prolonged blood circulation, strongly enhancing the odds of reaching the targeted sites. Due to their high surface areas, elongated NPs can also form stronger ligand–receptor bindings compared to nanospheres, and thus improve the selectivity and specificity during endocytosis.53–55 On the other hand, NPs with sharp edges can disrupt the endosomal membranes, inducing less exocytosis, thereby staying longer in the cytoplasm than the spherical particles.56 By producing a wide scope of well-controlled nanostructures, PISA can provide more possibilities for achieving better cellular uptake and tracking,57 which ultimately results in an enhanced bioavailability of the NPs. NP uptakes at target cells are also directly governed by their sizes. NPs should ideally be designed with sizes between 5 and 100 nm in diameter. Sizes lower than 5 nm are known to generate particles that will be eliminated by renal filtration,53 whereas higher diameters (over 200 nm) trigger excessive accumulation in the liver and spleen.58 Particles with diameters lower than 100 nm exhibit less adsorption of serum proteins while circulating in the blood, thus enhancing half-lives in vivo.59,60 Moreover, PISA can be performed at relatively high solid contents up to 50 wt% and usually imparts very high monomer conversions within short reaction times. Hence, PISA is of high benefit to the formulation of NPs, whose shapes and sizes are tuneable even during large-scale productions. Besides, simultaneous self-assembly and drug encapsulation may enable relatively high loading efficiencies (67% for doxorubicin-encapsulated NPs,61 54% for cisplatin-loaded micelles and 24% for BSA-loaded vesicles,62 under simple working conditions), likely owing to high polymer concentrations. PISA synthesis has been carried out using several types of living polymerisation techniques, but reversible addition–fragmentation chain transfer (RAFT) is probably the most prevalent synthesis method currently available to prepare NPs by PISA approach due to its adaptability to a broad range of reaction conditions and monomer families.63–68 RAFT dispersion or emulsion PISA processes can be initiated via various approaches, most commonly by thermal initiation but also by other new alternative approaches such as photochemical, ultrasonic and microwave activations, or using enzymes.49 Moreover, a few types of polymerisation techniques that have recently emerged can be of interest to PISA for drug delivery formulations, e.g. ring opening polymerisation (ROP)69 or ring-opening metathesis polymerisation PISA (ROMPISA).70 In conclusion, with superior cellular uptakes resulting from diverse and tuneable morphologies as well as relatively high drug loading under industrially relevant approaches, PISA appears to be a versatile synthesis and formulation technique for polymer-based nanomaterials in drug delivery systems.
Fig. 1 Illustrative mechanism of NP morphological transformation by RAFT-PISA. Reproduced from ref. 77 with permission from the Royal Society of Chemistry, copyright 2015. |
Based on simple and accessible formulations of NPs from PISA, an extensive amount of literature has been focused on producing stimuli-sensitive NPs, using mostly RAFT and some ROP/ROMP combinations, thereby conferring an efficient programmable drug release through a timely and direct nano-formulation.
The present review therefore aims to highlight a number of the latest examples regarding the design and preparation of polymeric NPs sensitive to a variety of triggering stimuli by using mainly RAFT-PISA and a few ROP/ROMPISA methods that show remarkable applications in many precise drug delivery systems. The authors will conclude with several critical remarks on the challenges and opportunities underlying the role of PISA as a formulation strategy for polymer-based nanomedicines in the context of clinical translatability.
For example, Karagoz et al. described a one-pot PISA synthetic route to asymmetric-block-copolymer NPs: poly[oligo(ethyleneglycol)methacrylate]-block-[poly(styrene)-co-poly(vinyl benzaldehyde)] (POEGMA-b-P(ST-co-VBA)) was obtained using a RAFT-mediated dispersion polymerisation in methanol and enabling self-assembly into various morphologies from spheres, worm-like and rod-like to vesicles.73 After the particle synthesis, the aldehyde groups in the core-forming blocks were conjugated with doxorubicin (DOX) via its amine functional groups to form the pH-responsive imine moieties (–CN–) on Schiff bases. A release study demonstrated that more than 80% DOX was liberated from the NPs at pH 5 due to the acidic pH-breakable bonds, whereas only 10% were released at pH 7.4. Furthermore, while particle morphologies did not affect the DOX release, they had a significant impact on the cell internalisation in MCF-7 breast cancer cell line, whereby worm- and rod-like particles displayed an improved cell internalisation compared to spherical micelles and vesicles. The authors reasoned that cylindrical morphologies imparted greater aspect ratios or surface areas, allowing for higher adhesions to cell membranes and thus favouring cellular uptake. In addition, rod-like and worm-like DOX-particles exhibited higher cytotoxicity than vesicles and micelles, as observed from IC50 variations on MCF-7 cell line (Fig. 2). In summary, this system, among all the possible obtained morphologies, has demonstrated an effective combination of responsiveness, cellular internalisation, and cytotoxicity, a feature that is very difficult to achieve from conventional nanoprecipitation.
Fig. 2 Schematic diagram of in situ self-assembly of POEGMA-b-P(ST-co-VBA) using PISA and its conjugation with DOX through a pH-cleavable bond. Reproduced from ref. 73 with permission from the Royal Society of Chemistry, copyright 2014. |
Similarly, (methacryloxyethoxy)benzaldehyde (MAEBA) was polymerised in the presence of poly((N,N′-dimethylamino)ethyl methacrylate) used as a macro-Chain Transfer Agent (macro-CTA) under RAFT dispersion PISA conditions, generating PDMAEMA block copolymer nano-objects as spheres, nanorods, nanowires, and vesicles.74 DOX was conjugated with the PMAEBA block via the acidic pH-breakable imine linkage (–CN–) in a similar way as in ref 73. A drug release study indicated that less than 1% of the DOX had been released at pH 7.4, while the fastest release was observed at pH 5 due to the cleavage of the aromatic imine under acidic conditions.
Lately, Armes and coworkers synthesised bio-inspired framboidal-like vesicles that simulated some Dengue virus features, with the vesicles exhibiting both moieties targeting the SR-B1 scavenger receptors that are available on triple-negative breast cancer cells, and pH-sensitive surface areas for local drug delivery.75 In particular, the two macro-CTAs poly(glycerol monomethacrylate) (PGMA) and poly(2-(methacryloyloxy)ethyl phosphorylcholine) (PMPC) were first chain-extended using 2-hydroxypropyl methacrylate (HPMA) and glycidyl methacrylate (GlyMA) that had been conjugated with rhodamine B piperazine (Rh). The obtained PGMA-P(HPMA-stat-GlyMARh) polymer was then chain-extended with the 2-(diisopropylamino)ethyl methacrylate (DPA) monomer via RAFT aqueous-seeded emulsion polymerisation, resulting in pH-responsive framboidal-like triblock copolymer vesicles. In this respect, the phosphorylcholine functionalities of PMPC endowed the vesicles with selective ligands to target the SR-B1 receptors on MDA-MB-231 cells, while the tertiary amines on DPA became protonated to liberate the plasmid DNA in the nucleus through intracellular acidification (Fig. 3).
Fig. 3 (a) Illustration of the intracellular release of a EGFP plasmid DNA within MDA-MB-231 cells using (97 PGMA58 + 3 PMPC60)–P(HPMA300-stat-GlyMARh1)–PDPA52 (97 G58 + 3 M60)–(H300-stat-ERh1)–D52 framboidal vesicles. (b) Schematic diagram of the loading of plasmid EGFP DNA in these triblock copolymer vesicles (the green colour indicates a successful EGFP expression while the red colour indicates the location of the rhodamine-labeled vesicles). Reproduced from ref. 75 with permission from the Royal Society of Chemistry, copyright 2019. |
Another recent example has been reported on the synthesis of cisplatin drug-encapsulated spherical micelles using ROMPISA.76 One of the most attractive features of ROMPISA is its tolerance to a wide range of initiators and functionalities, allowing the straightforward polymerisation of monomers in one pot. Amphiphilic block copolymers were synthesised in water, particularly a poly(oligo ethylene glycol) macro CTA, which was chain-extended by comonomers of a cisplatin norbornene analogue and pH-responsive 2-(diisopropylamino)ethyl norbornene dicarboximide.
The effect of sizes and surface charges of the resulting micelles on cellular uptake and cytotoxicity was investigated, indicating that smaller sizes exhibited improved cellular uptake and cytotoxicity, whilst the positively-charged NPs were better absorbed than the neutral or negatively-charged ones. A drug loading of 20% was achieved, with a selective release at pH 6.
In this regard, the anticancer drug camptothecin (CPT) was integrated with a prodrug monomer (CPTM). 2-(2-(Methoxyethoxy)ethyl methacrylate (MEO2MA) and the obtained CPTM comonomer were then polymerised in the presence of poly(ethylene glycol)-4-cyano-4-((phenylcarbono-thioyl)thio) pentanoic acid (PEG113-CPDB) used as macro-CTA, via a RAFT dispersion technique, leading to the formation of a diblock prodrug copolymer PEG-b-P(MEO2MA-co-CPTM).84 This synthetic technique resulted in prodrug NPs with a three-layered structure. In particular, the reduction-sensitive core was cross-linked via copolymerisation of benzyl methacrylate (BzMA) and N,N-cystaminebismethacrylamide (CBMA); the second layer contained the CPT-copolymer with the CPT conjugated to the polymer chains via reduction-sensitive disulfide linkages; and the outer layer included stabilising hydrophilic PEG chains. This prodrug NPs facilitated the cleavage of the conjugated CPTs upon exposure to a reductive microenvironment, e.g. at high levels of GSH in the cytosol. This synthetic strategy enabled a drug delivery system to achieve (1) simultaneous polymerisation and drug encapsulation in a one-pot approach at a relatively high concentration of 100 mg mL−1, (2) a high monomer conversion, (3) simultaneous core-cross-linking during prodrug particle preparation, (4) a CPT concentration that could be precisely tuned by altering the MEO2MA/CPT ratio, and (5) efficient drug unloading upon cell internalisation due to the high concentration of GSH in the cytoplasm.
Recently, the development of dual-functional polymer NPs exhibiting a fluorescence-assisted tracking capability and reduction-responsive triggered drug release has been reported.85 Accordingly, aza-boron dipyrromethene (BODIPY), a near-infrared (NIR) fluorescence dye, was loaded in situ into particles via a RAFT aqueous dispersion PISA without specific purification steps. Firstly, a reduction-responsive polyprodrug camptothecin-monomer (pCPTM) was prepared via chain extension of a PPEGMA macro-CTA during the RAFT copolymerisation of PEGMA and CPTM. Secondly, PEGMA comonomers and pre-synthesised NIR fluorescent monomer (NIRM) were co-polymerised in the presence of the PPEGMA macro-CTA, yielding pNIRM. The last step was required to prepare core-cross-linked bifunctional NPs, wherein pCPTM and pNIRM were hydrophilic macro-CTA agents that were chain-extended with the HPMA monomer and dimethacrylate cross-linker (EGDMA) via one-pot RAFT aqueous dispersion polymerisation. The resulting spherical micelles had a three-layered structure, with a cross-linked core encapsulating the CPT drug, surrounded by the pNIRM fluorescent layer, and externally coated by the hydrophilic PPEGMA corona. The covalent attachment of the BODIPY imaging agent endowed distinctive optical properties to the spheres, enabling real-time monitoring and tracking of the loaded drugs and NPs via an NIR imaging technique. In addition, both cell internalisation and intracellular drug release induced by a high level of GSH (10 mM) was confirmed in HeLa cells in vitro. The in vivo imaging also clearly indicated a passive targeting of the spherical micelles in the tumour site over time. To summarise, PISA has successfully allowed the preparation of highly delicate spherical micelles with reduction-responsive drug release and stable optical characteristics via a direct and simple one-pot formulation strategy, thus paving a novel pathway for tracking dynamic processes and biological imaging therapies (Fig. 4).
Fig. 4 Synthetic routes for reduction-responsive, prodrug and NIR-fluorescence imaging dual-functional NPs by RAFT mediated dispersion polymerisation, producing P(CPTM-co-NIRM)-b-P(HPMA-co-EDGMA) spherical micelles in water. Reproduced from ref. 85 with permission from the Royal Society of Chemistry, copyright 2018. |
Couturaud et al. synthesised functionalised fluorine-based amphiphilic block copolymers using a photo-mediated RAFT dispersion polymerisation method, as promising scaffolds for drug delivery.86 As ester-activated pentafluorophenyl methacrylate (PFMA) can be readily functionalized with amino compounds, it was employed and polymerised with the poly(ethylene glycol)-4-cyano-4-(ethylthiocarbonothioylthio)pentanoic acid (PEG113-CEPA) macro-CTA via light-triggered PISA at 405 nm and 37 °C, yielding uniform spherical micelles independently of DP variations. A post-functionalisation of the PFMA-based polymer with primary diamines, e.g. ethylenediamine (EDA) or cystamine, generated cross-linking of the core-forming block, imparting a greater colloidal stability to the spherical particles. Upon exposure to increasing levels of GSH, the cystamine-bearing nanostructures dissociated due to the reduction in GSH-sensitive disulfide cross-links, whereas the diamine-cross-linked spherical micelles remained intact. The designed cross-linked polymeric NPs also had excellent biocompatibilities in in vitro human cells (Fig. 5).
Fig. 5 Schematic illustration of the preparation of PEG-b-PPFMA via photo-initiated PISA in DMSO at 37 °C (M1), followed by cross-linking reactions using either ethylenediamine or cystamine (P1 and P2, respectively) for transfer into aqueous media. Reproduced from ref. 86 with permission from Wiley Online Library, copyright 2018. |
A recent work by Hong and colleagues has demonstrated the use of PISA for dual reduction-triggered drug release and pH-responsive prodrug NP preparation.87 Poly(N-(2-hydroxypropyl)methacrylamide) (PHPMA) and poly(2-(diethylamino)ethyl methacrylate) (PDEAEMA) were used as macro CTAs and were chain-extended with CPTM to produce PHPMA/PDEAEMA-stabilized NPs. The PDEAEMA chains were protonated in acidic tumours and transitioned from a hydrophobic to a hydrophilic state, which in turn enabled an increased drug release and cancer cell uptake.
Having realised that the current stimuli-sensitive nanocarriers usually require tedious preparations, Sobotta et al. developed a one-pot synthetic pathway using RAFT-mediated dispersion PISA to produce well-defined spherical micelles of tuneable sizes that display ROS-triggered drug release.95 A series of block copolymers with varying chain ratios was prepared, whereby poly(N-acryloylmorpholine) (PNAM) was utilised as a macro-CTA for chain extension with N-acryloylthiomorpholine (NAT), a monomer carrying an oxidation-sensitive thioether moiety. This procedure readily afforded P(NAM-b-NAT) spherical particles of 25–73 nm in diameter (DLS) via the PISA process. The sphere size-range could be tailored by tuning the size or composition of the diblock copolymers. Nile Red dye was chosen as a model for drugs and was encapsulated into the spheres. In response to a H2O2 concentration of 10 mM, the hydrophobic thioether was oxidised to the hydrophilic sulfone, triggering a solubility transition of the core-forming block. As a result, the micellar structures dissociated, and the drug was released as confirmed by the fluorescence quenching of Nile Red proportional to the increased H2O2 levels. In addition to oxidation-triggered drug release, the formulated nano-vehicle showed excellent biocompatibility and no cytotoxicity, therefore providing a versatile, selective, and controlled drug delivery to the sites of interest (Fig. 6).
Fig. 6 Schematic diagram of the one-pot synthesis of P(NAM-b-NAT) micelles of tuneable sizes displaying ROS-triggered drug release. Reproduced from ref. 95 with permission from the Royal Society of Chemistry, copyright 2018. |
Additionally, CO2 exhibits excellent biocompatibility and membrane permeability.96–100 Even though CO2 responsiveness has not been largely adopted in the context of therapeutic treatments, some of the designs available for CO2-sensitive polymeric materials have been considered as proofs-of-concepts for uses in drug delivery.
Recently, silica nanoparticles (SiO2 NPs) have been utilised as effective adhesives for hydrogels and biological tissues.101,102 It was reported that SiO2 NPs (20 nm) can be encapsulated into PISA-produced hybrid vesicles displaying an on-demand release of the internal contents due to the CO2-responsive characteristic of the core-forming block. Here, mPEG113 was used as a macro-CTA for the photo-initiated RAFT aqueous dispersion copolymerisation of 2-hydroxypropyl methacrylate (HPMA) and 2-(dimethylamino)ethyl methacrylate (DMAEMA) at room temperature. The obtained results indicated that complete monomer conversion was rapidly achieved within 15 min upon exposure to 405 nm visible-light, affording a well-defined CO2-sensitive diblock copolymer mPEG-b-P(HPMA-co-DMAEMA). The SiO2 NPs were concurrently encapsulated into the lumen of the polymeric vesicles via PISA, in a convenient one-pot synthetic strategy. The loaded cargos were released from the NPs in response to CO2 bubbling under mild conditions, due to the presence of the DMAEMA monomer in the hydrophobic block. After treatment with CO2 for 2 minutes, DMAEMA was protonated, enhancing the hydrophilicity of the core-forming block, and the vesicles therefore disintegrated to release the SiO2 NPs (Fig. 7).103 The same research group used the same concept to encapsulate fragile biomacromolecules such as bovine serum albumin (BSA) (5 wt%) into the polymeric vesicles prepared from an aqueous photo-initiated PISA under mild conditions. BSA was encapsulated in situ into the hybrid vesicles made of PEGMA-b-P(HPMA-co-DMAEMA). The addition of increased amounts of DMAEMA led to higher order morphologies during the block copolymer self-assembly, due to increased hydrogen bonding interactions. In the presence of CO2, BSA was unloaded from the vesicles due to the disassembly of copolymer chains induced by the protonation of DMAEMA.104
Fig. 7 CO2-triggered release of SiO2 NPs from mPEG-b-P(HPMA-co-DMAEMA) hybrid vesicles. Reproduced from ref. 104 with permission from Wiley Online Library, copyright 2017. |
Nevertheless, several challenges remain to produce CO2-sensitive NPs via PISA. First, the polymerisation solution has to be degassed by purging inert gas or freeze–pump–thaw cycles, thus hampering large-scale manufacturing of those NPs. Second, the obtained particles are likely to disintegrate upon exposure to surfactants or organic solvents. Third, the switchability of CO2 with inert gas (N2 or Argon) is limited. To overcome these challenges, Yu et al. described a robust in situ cross-linking strategy that provides (1) a mild PISA process using photo-induced initiation in water; (2) an open-air polymerisation using enzyme-assisted degassing; and (3) CO2-responsive vesicles to precisely release the internal cargos. Particularly, HPMA and DMAEMA were polymerised by RAFT aqueous dispersion copolymerisation, with a 4-cyano-4-(ethylthiocarbonothioylthio)pentanoic acid (CDPA) macro-CTA, to directly produce PGMA-b-P(HPMA-co-DMAEMA-co-AMA) vesicles via photo-induced PISA.105 The reaction was deoxygenated by adding glucose oxidase and glucose. In situ crosslinking was achieved by including allyl methacrylate (AMA), an asymmetric cross-linker, during the PISA process. Under CO2 treatment, the core-contained DMAEMA was protonated and became hydrophilic, in a similar way to the aforementioned studies. However, due to the AMA cross-linked core, the vesicles were not disrupted but became prone to swelling or expansion in volume and size. On the other hand, when purging with N2, the swelling vesicles shrunk because of the deprotonation of DMAEMA under these inert conditions. This swelling–shrinking process was reversible over four CO2/N2 cycles. Methylene Blue (MB) dye was used as a drug model in a proof-of-concept release experiment and revealed enhanced release when subjected to a CO2 treatment.
In the meantime, these CO2-sensitive vesicles could carry an increased amount of SiO2 NPs after the CO2 treatment. This was explained by the interactions between SiO2 NPs and tertiary amines on DMAEMA in the hydrophobic block and the higher surface area of the vesicles favouring more SiO2 NPs to be loaded (Fig. 8).
Fig. 8 Schematic illustration of an enzyme-assisted photo-PISA for preparing cross-linked CO2-responsive vesicles. Reproduced from ref. 105 with permission from the Royal Society of Chemistry, copyright 2019. |
Lately, Mable et al. prepared a series of in situ SiO2-encapsulated amphiphilic block copolymer vesicles via a one-pot RAFT aqueous dispersion polymerisation method, in which a poly(glycerol monomethacrylate) (PGMA) macro-CTA was chain-extended with the HPMA monomer while simultaneously adding the hydrophilic 18 nm SiO2 NPs.62 SiO2 NPs were utilised as a bio-related active species due to their intrinsic self-healing effects on synthetic hydrogels or biological tissues.106
The targeted degree of polymerization was maintained relatively low to explore the thermo-responsiveness of the PHPMA core-forming block. SAXS confirmed that the release of SiO2 NPs occurs within the first 12 min upon cooling to 0 °C, as the PGMA58-b-PHPMA250 diblock copolymer underwent a morphological transformation from vesicles to a mixture of spherical and short-worm micelles, thereby triggering the release of internal cargos. Furthermore, an onset of vesicle dissociation was observed at about 10 °C, and promptly increased as the temperature decreased. The authors reported that this smart formulation and effective on-demand release of SiO2 NPs would be applied to self-healing adhesives of synthetic hydrogels and living tissues, thus enabling a possible model for encapsulating globular proteins (BSA), enzymes, and antibodies (Fig. 9).
Fig. 9 (a) Synthesis of PGMA58-b-PHPMA250 (G58H250) diblock copolymer via RAFT aqueous dispersion; (b) schematic illustration of the in situ encapsulation of SiO2 NPs via PISA and the subsequent release of SiO2 NPs upon cooling to 0 °C due to vesicle dissociation. Reproduced from ref. 62 with permission from ACS Publications, copyright 2015. |
On the other hand, worm-like gels have been widely adopted as potential materials in a range of biomedical applications. The group of Armes reported the preparation of biocompatible poly(glycerol monomethacrylate)-poly(2-hydroxypropyl methacrylate) (PGMA-b-PHPMA) diblock copolymer nanoworm gel that was responsive to temperature changes.107 These PGMA-b-PHPMA diblock copolymer worm-like micelles display a reversible thermal-responsive behaviour. Here, a PGMA-b-PHPMA diblock copolymer was dispersed into fluorophore-labelled Staphylococcus aureus at 4 °C, the temperature at which nanoworm degelation occurred, and led to worm-to-sphere transition, thus lowering the solution viscosity. This solution was then filtered through a 0.45 μm membrane, allowing for the smaller diblock copolymer particles to cross whereas the micrometer-sized bacteria stayed on the membrane. Brought back to ambient temperature, the obtained solution reversibly shifted from spheres to worms, favouring the gelation, which in turn served as an injectable scaffold for stem cell growth. Later, the same research group exploited the use of PGMA-b-PHPMA diblock copolymer nanoworm gel in human Pluripotent Stem Cells (hPSCs) and human embryos.108 When cooled down from 37 to 5 °C, PGMA-b-PHPMA worm gels switched from worms to spheres, allowing for a reversible gel formation-destruction via an order-order transition, on which cells can be embedded and recovered. These worm gels showed a good aptitude for mimicking natural mammalian mucins and for forming an excellent 3D matrix to maintain the viability and pluripotency of hPSCs without any passaging for 14 days, as well as for possible short-term storage of hPSCs and human embryos without a need for cryopreservation.
Although the majority of reported examples have used the RAFT-PISA technique, other potential types of polymerisation techniques can be of interest for the production of biodegradable and responsive polymer NPs. To address the challenges of high polymerisation temperatures and strictly oxygen-free conditions, ROMPISA has been used to synthesise particles with high control over the polymerisation process under aerobic conditions.
In a recent report, Wright et al. generated peptide-based NPs sensitive to the proteolytic enzyme thermolysin in aqueous media, via a one-pot method in open air by ROMPISA at a high solid concentration (10 wt%).70 Such enzyme-responsive poly(norbornene) block copolymers are key to a series of future applications where more enzymatically sensitive biodegradable materials can be prepared via the versatile and robust ROMPISA method.
Recently, vesicles made from the ROP of N-carboxyanhydride (NCA)-induced self-assembly (NCA-PISA) have been reported. The polymerisation of L-phenylalanine NCA (L-Phe NCA) was initiated by methoxypolyethylene glycol amine(PEG45-NH2) at 10 °C without the need for degassing, using 20% solid contents.69 The obtained vesicles were then degraded by trypsin over 96 h.
In a similar work, Pan et al. described the RAFT dispersion copolymerisation of DIPEMA and CMA in the presence of PHPMA as a macro-RAFT agent to produce in situ vesicles.113 Upon irradiation with UV light at 365 nm, the coumarin moieties anchored in the vesicle membrane underwent dimerisation to crosslink together, thus bringing an improved structural stabilisation to the polymeric vesicles. Furthermore, the presence of tertiary amino groups on the DIPEMA segments of the membrane triggered a swelling–shrinking transition upon tuning the solution pH from acidic to basic. At pH 4, the DIPEMA was protonated and shifted to a hydrophilic state, inducing the swelling of the core-forming block, thus enabling the pores of the membrane vesicles to open and facilitating the transmembrane traffic of substances. In acidic solutions, both the membrane (DIPEMA) and DOX were protonated, and the escape of DOX was inhibited due to electrostatic repulsion, although the membrane pores were large enough for DOX to get through. In contrast to DOX, fluorescein was effectively released after loading through the vesicle membrane as the size of fluorescein is smaller than the pores and due to the absence of electrostatic repulsion.
Fig. 10 (A) In situ crosslinking in PISA to prepare crosslinked PEG-CPDB-b-P(DIPEMA90-co-CBMA10) membrane vesicles (D90C10); (B) reversible changes in the diameter of D90C10 vesicles upon switching solution pH between 4.0 and 8.0. (C) Cumulative release profiles of rhodamine B from D90C10 vesicles under different conditions. Reproduced from ref. 114 with permission from ACS Publications, copyright 2019. |
This transition resulted from the incorporation of ionisable groups that, at acidic pH values associated with the cell pathology, induced protonation and hydrophilicity of the internal fragments, leading to nano-objects disassembling and drugs being released.
The nano-structural dissociation arising from solubility transition was also observed under ROS-abundant conditions, whereby the hydrophobic thioether moieties in the core were oxidised to hydrophilic sulfones, disintegrating the core–shell structures. Micelles cross-linked via disulfide bonds displayed an enhanced colloidal stability under normal cell conditions; however, in the presence of reducing agents, they disintegrated to liberate the internal payloads owing to the cleavage of reduction-sensitive disulfide bonds. On the other hand, thermal responsiveness of spherical micelles triggered drug release in response to temperature changes because of the morphological evolutions. Dual triggers such as pH-and-reduction or pH-and-light stimuli combinations synergistically increased the sensitivity of particles for an enhanced drug release efficiency.
Although PISA research has vigorously been growing and showing great promise to overcome the drawbacks of traditional solution self-assembly techniques, many challenges remain to be addressed. One of the major bottlenecks preventing the expansion of PISA into more biomaterial applications is the use of relatively high temperatures and absolutely oxygen-free environments required for radical polymerisation methods, which can be challenging to achieve and hamper the final desired reaction yields. Most of the existing (meth)acrylate-based polymers are also difficult to degrade in biological media resulting in some increased accumulation and toxicity. Fortunately, some of these issues can be addressed using ROMP or ROP methods, involving the use of new and biodegradable monomers to synthesise self-assembled objects under mild conditions, or by using UV initiation to avoid the need for high reaction temperatures. These approaches are also of great benefit to the encapsulation of fragile therapeutic drugs or biomolecules such as peptides, antibodies, or DNA/RNA, thus avoiding their denaturation or degradation. Currently the majority of PISA formulations for drug delivery and biomaterials have been carried out based on RAFT polymerisation methods, some on ROP or ROMP. Hence there is a need to exploit other types of living radical polymerisation methods such as nitroxide-mediated polymerisation (NMP),115 iodine-mediated polymerisation (ITP),116 and sulfur-free RAFT117,118 for potentially preparing less toxic polymers. Though atom transfer radical polymerization (ATRP) has been extensively adopted to synthesise PISA-derived nano-objects,119,120 its application in drug delivery is very limited due to the use of cytotoxic copper catalysts for polymerisation.
It is also of importance to be able to design multifunctional PISA-made particles that can load multiple therapeutic agents and/or diagnostic probes while being conjugated with targeting ligands for achieving enhanced site-specific drug delivery as well as precisely controlled drug release on demand. Additionally, more attempts are required to investigate the behaviour of these PISA-prepared NPs in complex 3D cell culture models as well as in vivo to rapidly accelerate translations to clinical studies. In principle, a formulation method of future interest should also display reasonable drug loading and increased cellular uptake or trafficking, and PISA has proven itself to be an outstanding candidate. Attempts have been made to apply external stimuli to impart higher control over sizes and morphologies of PISA-prepared NPs. Interestingly, light has been used to enable spatiotemporal control of initiation,121 cross-linking,112 or modification of the hydrophobic/hydrophilic block ratio.122
Nevertheless, while it is critical to carry out PISA in aqueous media for medical purposes, it can limit the ability of encapsulating hydrophobic drugs, especially via physical entrapment. Probably, the use of prodrug monomers as aforementioned could potentially provide a solution by which drugs can be conjugated to monomers or macro-CTAs prior to or during the PISA process via stimuli-responsive linkages, thus both increasing the loading efficiency and achieving local release. The question has also been raised about whether the particle shapes may have an impact on the resulting loading efficiency and cell internalisation since there have not been many reports addressing this issue. Hence more attention should be drawn to this critical aspect. Meanwhile, hydrophilic drugs can be easily compartmentalised into PISA-generated NPs, particularly in vesicles, imparting the protection of internal cargos from degradation by biological media. This aspect could be very useful for the preparation of less invasive oral formulations which can protect drugs from the extreme conditions of the digestive and intestinal tracks before reaching the disease areas. Finally, the majority of PISA-related work up to now has been focusing on anticancer drug delivery while it is very likely that the same methodology could be applied to other therapeutic treatments, such as diabetes, and cardiovascular, or infectious diseases (Table 1).
Block copolymer | Stimulus/stimuli | Drug/payload | Morphology | Ref. |
---|---|---|---|---|
POEGMA-b-P(ST-co-VBA) | Acidic pH | Doxorubicin (DOX) | Spheres, worms, rods, vesicles | 73 |
PDMAEMA | Acidic pH | DOX | Spheres, rods, wires, vesicles | 74 |
P(HPMA300-stat-GlyMARh1)–PDPA52 | Acidic pH | Rhodamine B | Vesicles | 75 |
PEG-b-P(MEO2MA-co-CPTM) | GSH 10 mM | Camptothecin (CPT) | Spheres | 84 |
P(CPTM-co-NIRM)-b-P(HPMA-co-EDGMA) | GSH 10 mM | BODIPY | Spheres | 85 |
PEG113-b-PPFMA | GSH 10 mM | None | Spheres | 86 |
P(NAM-b-NAT) | H2O2 10 mM | Nile red | Spheres | 95 |
mPEG-b-P(HPMA-co-DMAEMA) | CO2 | SiO2 NPs | Vesicles | 103 |
PEGMA-b-P(HPMA-co-DMAEMA) | CO2 | BSA | Vesicles | 104 |
PGMA-b-P(HPMA-co-DMAEMA-co-AMA) | CO2 | Methylene blue (MB)/SiO2 NPs | Vesicles | 105 |
PGMA-b-PHPMA | Temperature | SiO2 NPs | Vesicles | 62 |
PHPMA-b-P(NBMA-co-CMA) | UV and acidic pH | DOX | Spheres, nanorods, lamellae, vesicles | 112 |
PHPMA-b-P(DIPEMA-co-CMA) | UV and acidic pH | AuNPs/DOX/fluorescein | Vesicles | 113 |
PEG-CPDB-b-P(DIPEMA-co-CBMA) | Reductive (DTT) and acidic pH | Rhodamine B | Vesicles | 114 |
Beyond these challenges, it is widely believed that PISA will continue to thrive and rapidly expand, building on our growing understanding of synthetic parameters and formulation strategies. To date, PISA has successfully enabled the self-assembly, drug loading, and incorporation of stimuli-responsive functionalities in a one-pot synthetic approach. In the near future, one believes the rapid maturation of PISA can instil more important aspects of precise drug delivery into a self-assembled nano-object, thus furthering a range of therapeutic values to the ultimate translation from the bench to the bedside.
This journal is © The Royal Society of Chemistry 2021 |