There are two methods to synthesize the materials of polymersomes. One is mixing AB and BC diblock copolymers together, the other is directly using ABC triblocks to form asymmetrical polymersomes. Consisting of at least two homopolymer blocks, block copolymers are the ideal materials for forming self-assembling polymersomes. Normally, the homopolymer is designed to exhibit some specific physicochemical properties, and in consequence, the block copolymers will display versatile properties and utilization values (Lee and Feijen, 2012; Ge et al., 2014).

Sequential radical addition-fragmentation chain transfer (RAFT) polymerization is the most commonly used technique to synthesis triblock copolymers (Du et al., 2012). Ring-opening polymerization (ROP) is another feasible way to synthesize diblock or triblock copolymers (Kishimura et al., 2007). The properties of the diblock or triblock copolymers can determine the types and applications of the asymmetrical polymersomes by controlling the structures, compositions and molecular weights (Li et al., 2015). Therefore, we summarize some available diblock or triblock copolymers that constitute asymmetrical polymersomes, and the results are given in Table 1.

There are two major methods to form self-assembling asymmetrical polymersomes: solvent change method (phase inversion) and direct hydration method (Zhang and Eisenberg, 1995; Blanazs et al., 2009). The first method is more widely used. Interestingly, Asano et al. first used the co-assembly method to generate asymmetric polymersomes. They put two distinct diblock copolymers PS-b-PEO (SO) and PB-b-PEO (BO) in an oil-in-oil emulsion PS/PB/CHCl₃ (Asano et al., 2016).

Compared with UV stimuli for drug release, temperature stimuli are a more practicable way for building intelligent polymersomes, since natural temperature differences in tissues exist in human body. Tumor tissues have a higher temperature than normal tissues and the temperature can change easily in the external, such as hyperthermia (Xu et al., 2009). Furthermore, using the heating or cooling appliances for particular sections of the body can also achieve the effect of temperature differences (Christian et al., 2009).

The temperature-dependent mechanisms can be achieved by using the polymer poly(N-isopropylacrylamide) (PNIPAm) (Li et al., 2006; Qin et al., 2006). PNIPAm has a unique property that its conformation will change if the temperature is above the lower critical solution temperature (LCST) of PNIPAm (40°C). Above 40°C, the block will change its property from hydrophilic to hydrophobic. Thus, temperature-sensitive polymersomes are designed by using PNIPAm. Usually, PNIPAm chains are covalently combined with a hydrophilic block such as poly(N-(3-aminopropyl) methylacrylamide hydrochloride) (PAMPA) or PEO. The resulting copolymer will self-assemble to form polymersomes and have the capacity to load hydrophilic drugs. If the temperature decreases below 40°C, the polymersomes will disassemble to the former copolymer and release the encapsulated therapeutics. Cai et al. (2011) prepared asymmetric polymersomes based on poly(ethylene oxide)-b-poly(ethylene oxide-stat-butylene oxide)-b-poly(isoprene) (E-BE-I) ABC triblock copolymer, which presented the temperature dependent property and the LCST was 25°C. Compared to the usual temperature-responsive polymersomes using PNIPAm as responsive factors, these polymersomes had narrower molecular weight distribution and faster transformation rate because of the lacking of strong interchain hydrogen bonding (Figure 1).

As a matter of fact, in the tumor microenvironment there is an innate pH difference (pH 6.5−7.2) and the cancer cells' endosomes and lysosomes are acidic (pH 4.0−6.5) (Grabe and Oster, 2001; Rofstad et al., 2006; Meng et al., 2014). Ahmed et al. found that a number of polymersomes approached cellular endolysosomes by pinocytosis passageway (Ahmed et al., 2006b). Hence, this intracellular cue can be applied for pH-dependent asymmetric polymersomes. This release method has an advantage over external cues (temperature, UV) owing to the accessibility of such intracellular cues. Using polymersomes whose structures are susceptibility to this pH difference, targeted intracellular release of the specific encapsulated cargo can be achieved. Du et al. investigated endosomal pH-sensitive degradable asymmetric polymersomes constructed by ABC triblock poly(ethylene glycol)-b-poly(trimethoxybenzylidene tris (hydroxymethyl)ethane methacrylate)-b-poly(acrylic acid) (PEG-PTTMA-PAA). In this study, they demonstrated that PTTMA block could quickly destabilize due to the degradation of acetal groups in lower pH in the endo/lysosomal compartments of cancer cells, achieving targeted intracellular release of DOX·HCl and high anti-tumor therapeutic effects (Du et al., 2012).

Oxidation (Napoli et al., 2004b,a) or reduction-responsive (Cerritelli et al., 2007) reactions open up new spatial release mechanisms in the human body (Bodor, 1987; Corti et al., 2010). Oxidative environments exist at sites of inflammation, tumor tissues and in extracellular fluids, as well as intracellular compartments such as the endolysosomes (Grundl, 1994; Tew, 2007).

Wang et al. (2013) demonstrated reduction-responsive asymmetrical polymersomes based on PEG-SS-PCL diblock copolymer and asymmetric PEG-PCL-PDEA triblock copolymer. When exposed to reducing environments such as the nuclei and cytoplasm of cancer cells, this system will quickly rupture due to cleavage of the disulfide bonds between PEG and PCL blocks. This delivery system can protect encapsulated cargos in the extracellular environments (i.e., blood plasma) and induce efficient intracellular drug release, therefore it has the potential to be a versatile and multifunctional drug delivery platform by using intracellular stimuli (Krack et al., 2008).

Encapsulating drugs in asymmetric polymersomes can achieve efficient drug loading capacity, fast endocytosis rate and rapid endosomal escape ability. All of these advantages help to maintain the property of drugs in vivo or in vitro and control the release rate. With the development of smart delivery systems, many cancer-target chemotherapeutics drugs which have severe side effects now come to a renascence (Arrigo, 2000, 2005; Arrigo and Ducasse, 2002).

Liu et al. reported pH-sensitive asymmetric polymersomes based on PEO-b-PCL-b-PAA triblock copolymer. PEO chains constituted the outer corona of polymersomes because it showed well-behaved biocompatibility and was intrinsically stealthy to immune system. The pH-responsive PAA chains were designed in inner aqueous core since they could quickly destabilize at endosomal pH. Consequently, the release of encapsulated cargos in the cells was efficient and accurate (Liu et al., 2014).

Protein drugs have been recognized as one kind of the most potential leads for the growth of new therapeutics (Tan et al., 2010), and they are powerful antidotes toward many intractable diseases such as diabetes and cancers (Torchilin and Lukyanov, 2003; Futaki, 2006; Schrama et al., 2006). However, they are too sensitive to environmental conditions, which results in their short lives in vivo.

Polymersomes with huge aqueous compartments can encapsulate proteins and the membranes will protect proteins from degradation (Discher et al., 2007; Christian et al., 2009; Sun et al., 2009; Meng and Zhong, 2011; Lee and Feijen, 2012; Liu et al., 2012). So far, polymersomes have encapsulated a variety of proteins, including aquaporin Z (Kumar et al., 2007), cytochrome C (CC) (Hvasanov et al., 2011), hemoglobin (Rameez et al., 2008; Li et al., 2012), insulin (Xiong et al., 2007; Christian et al., 2009; Kim H. et al., 2012), immunoglobulin G (Fu et al., 2011), ovalbumin (Stano et al., 2013), and myoglobin (Kishimura et al., 2007). Recent investigations also have found that polymer vesicles with asymmetrical membranes exhibit much higher loading capacity of proteins than symmetrical polymersomes.

Li et al. reported self-aggregated vesicles with asymmetric membranes. These polymersomes can serve as hemoglobin (Hb)-based oxygen carriers with oxygen affinity and high Hb loading content and meanwhile no interference with blood cells. Further studies have shown that these vesicles have great stability and efficacy, which have the potential to be alternative blood substitutes (Li et al., 2014).

Magnetic resonance imaging (MRI) is a widespread diagnosis apparatus that is clinically used for taking pictures of organs and structures inside the body to find blood vessels, bleeding, tumors or infection. Using asymmetric polymersomes to deliver contrast agents has better sensitivity and lower toxicity. Liu et al. (2015) reported asymmetrical polymersomes based on two kinds of diblock copolymer FA-PGA75-b-PCL30 and DTPA-PGA22-b-PCL30, which had diagnostic and therapeutic effects simultaneously. The highest drug DLE can reach 52.6% and the T1 relaxivity can be increased by 8-fold, meanwhile maintaining a low toxicity and remarkable positive contrast enhancement for tumor imaging.

Polymersomes with non-spherical shapes play important roles in applications such as vaccine development. Non-spherical shape can cause different interaction to immune cells because the shapes determine the in vivo behavior of polymersomes. Actually, there are not comprehensive studies that investigated the influence of shape and topology on the in vivo behavior of polymersomes.

Researchers have found that tubular structures are one of the most promising morphologies rather than spheres. Owing to the resembling with bacterial topologies, tubular structures offer larger contact areas between particles and cells. Van Hest's group has developed a series of methods that reshaped spherical vesicles to tubular structures. The ability of tunable membrane gives these spherical structures a chance to apply to the immunology field (Williams et al., 2017). As for polymersomes, not only the copolymer composition can influence the morphology but also the response to various stimuli such as temperature, pH, magnetic fields, or osmotic pressure can further change the structures. In a recent example by van Hest et al. (Abdelmohsen et al., 2016a), the formation of functional nanotubes from PEG-PDLLA was demonstrated. Biodegradable polymersomes made by PEG-PDLLA copolymer transformed their structures from spheres to nanotubes upon dialysis under hypertonic conditions. Owing to the dialysis with the increasing concentration of NaCl, the size can be osmotically controlled by the enrichment and elongation of the structures. In this way, such well-defined nanoparticles can be easily prepared and functionalized. This method established a novel platform for biomedical research where nanoscopic control over size and shape is highly valuable. Another example of non-spherical vesicles is the stomatocytes, which can also be prepared by dialysis. Through shape transformation of spherical polymersomes into bowl-shaped stomatocytes, polymersomes will form an extra nanocavity. Due to the direct contact with the outside environment, these vesicles present the ability to convert chemical energy into kinetic energy when encapsulating enzymes or catalytic nanoparticles in the nanocavity. Further investigations showed that stomatocytes have the chemotactic ability toward certain cell types (Williams et al., 2017). Stomatocytes are promising candidates for further useful applications such as immunoassays, protein and DNA isolation, detection and biosensing (Abdelmohsen et al., 2016b).

Much efforts have been made for targeting drugs and genes to specific sites. It is crucial for clinical therapeutics because drugs without targeting ability may cause low effect and high toxicity. Targeted delivery is categorized into two types: active targeting and passive targeting. Passive targeting aims at capturing nanoparticles that are smaller than the aperture gap of endothelial cells. These nanoparticles can go through the interstitium and therefore gather in tumor tissue (Danhier et al., 2009). Active targeting takes advantage of molecular recognition to transport drugs to specific sites. Attaching biological ligands or antibodies to the nanoparticles is the most widely applied method and it is usually carried out by chemically conjugating (Yang et al., 2002; Li et al., 2005; Rerat et al., 2010), mixing (Guo et al., 2009) and coating (Yang et al., 2002; Elloumi Hannachi et al., 2009). Targeting moieties are usually conjugated to the hydroxyl groups of their hydrophilic polymer blocks (e.g., PEO, PEG) (Torchilin et al., 2001; Velonia et al., 2002; Levine et al., 2008). In addition, polymersomes conjugated with targeting moieties may change their hydrophilic-block-to-total-mass ratio, resulting in the change of morphology. For example, from vesicles to micelles.

Lu et al. reported the modification of their triblock copolymer (PEG-PTTMA-PAA) polymersomes with anisamide (Anis) to target the cancer cell receptor sigma (Lu et al., 2015). Sigma receptor is an over-expressed membrane protein that appears in many human malignant diseases including lung cancer and prostate cancer (Lu et al., 2015). Anis ligands exhibit high affinity to sigma receptor and so far have been used as guides to deliver versatile drugs including proteins, siRNAs and doxorubicin (Della Rocca et al., 2011; Guo et al., 2012; Kim S. K. et al., 2012). Their results validated that Anis-chimaeric polymersomes (CPs) presented high targeting ability to H460 and when Anis contents increased, antitumor efficacy was improved. If competitive antagonist is used, the antitumor activity will be reduced drastically (Lu et al., 2015).

Many polymersomes are easily internalized by endocytosis of macrophages and induce inflammatory responses. Therefore, polymersomes are often designed to be biodegradable and biocompatible and with reduced in vivo inflammatory responses.

Leukopolymersomes are a kind of polymersomes that have adhesive properties to leukocytes. It can be easily made by just functionalizing the terminal groups on the membranes of the vesicles. There are two typically adhesive ligands on leukocytes, which are named as selectins and integrin. These two ligands can mimic the adhesive properties of activated leukocytes (Hammer et al., 2008). Engineered to express the adhesion molecules, leukopolymersomes can achieve binding to inflammatory substrates in the shear fluid flow in blood vessels (Hammer et al., 2008). Moreover, by altering the membrane materials and ligand ratio, the rate and type of adhesive interaction can be tuned. Hammer et al. demonstrated that the adhesion was specific because the adhesive rate was the same as that of leukocytes (Hammer et al., 2008).

The adhesiveness of leukopolymersomes can be coupled with other properties of polymersomes, such as their ability to encapsulate drugs and image contrast agents. These particles will ultimately be useful for creating theranostic particles that can detect, image and deliver drugs to inflammatory sites, cancer cells, and cardiovascular lesions (Robbins et al., 2010).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.