Introduction

In recent decades, synthetic nanoparticles (NPs) including liposomes [1], micelles [2], dendrimers [3], nanocapsules [4], nanodiamonds [28], which are cell-derived proteolipid membrane vesicles, are emerging in nanomedicine-related fields [29]. Major types of EVs include exosomes, microvesicles, and apoptotic bodies [30]. Our understanding of the between-cell communication has been elevating in the last decade due to EVs, particularly exosomes, which are nano-sized (30–150 nm) subtype of EVs. Exosomes are enriched with various biological components, including proteins, nucleic acids and lipids from their parental cells [31]. Exosome-mediated cell-to-cell communication plays an important role in multiple physiological and pathological processes like tumor metastasis, drug resistance, immune responses and microenvironment homeostasis [32].

Exosomes are also competent candidates for targeted drug delivery [33,

Fig. 3
figure 3

(Reprinted with permission from Ref. [3).

Table 3 Summary of artificial exosomes generated by biohybrid approaches

Biohybrid by freeze-thawing

Despite tropism of exosomes, the membrane engineering may be necessary for advanced drug delivery [96]. The hybrid NPs are slightly large in size but have similar morphology to exosomes and retained most protein markers of exosomes. Importantly, the hybrid NPs showed improved circulation and preferential accumulation in tumors and released drugs in response to temperatures. Genetically engineered thermosensitive liposome-exosome hybrid can escape mononuclear phagocytic system and target the tumor site, in which hyperthermia therapy stimulates the hybrid and induces combined immune-chemotherapy (Fig. 4).

Fig. 4
figure 4

(Reprinted with permission from Ref [96]. Copyright WILEY–VCH 2020)

A typical biohybrid strategy for generating hybrid exosomes by freeze-thawing induced fusion for combined tumor chemo-immunotherapy. Genetically engineered exosomes (gExos) were fused with thermosensitive liposomes (TLs) to form hybrid nanoparticles (gETL NPs), which can escape from clearance of mononuclear phagocytic system (MPS) and be activated by hyperthermic intraperitoneal chemotherapy (HIPEC) at the tumor site to release drugs for chemotherapy and polarize tumor-associated macrophages to M2 type to activate T cells for tumor immunotherapy.

Biohybrid by incubation

Incubation is a mild and commonly used method for various cellular processes and reactions. Fusion of liposomes and exosomes may occur during incubation as their membranes both have lipid bilayer structure. Incubation-induced spontaneous fusion may result in wide size distribution of hybrid particles with a high polydispersity index (PDI). The impact of incubation conditions on the fusion process requires further investigation.

Exosomes are promising targeted delivery nanocarrier, but have limited efficiency, for their small size, in encapsulating exogenous large nucleic acids, such as CRISPR/Cas9 System. The fusion of exosomes and liposomes may achieve efficient loading of large plasmids into the hybrid. Lin et al. developed a hybrid NP by simple incubation of HEK293FT cell-derived exosomes with CRISPR/Cas9 expressing liposomes at 37℃ for 12 h [97]. The hybrid NPs delivered the CRISPR/Cas9 system to mesenchymal stem cells (MSCs) and achieved gene editing by expressing the encapsulated genes in the MSCs, which cannot be transfected by the liposome alone. Taken together, the exosome–liposome hybrid NPs can deliver CRISPR–Cas9 system in MSCs, providing a promising tool for targeted gene editing.

The incubation-induced fusion of exosomes and liposomes can be enhanced by mediators on the surface, such as PEG. Piffoux et al. modified the EVs by incubation with liposomes via PEG-mediated fusion [98]. PEG-mediated fusion was proved to be an efficient approach to engineer EVs with exogenous compounds while preserving their inherent contents. Moreover, this fusion method enabled bioengineering of liposomal particles with biogenic molecules; importantly, the PEG-mediated fusion strategy allows efficient loading of cargoes and the feasible engineering of EV membranes with adaptable functions.

Biohybrid by co-extruding

Another reported method for generating biohybrid of exosomes and liposomes is co-extrusion. Under physical stress, membranes of exosomes and liposomes would break and re-assemble to form hybrid vesicles when passing through the membrane pore with controlled size. Exosomes have been explored as a drug delivery candidate owing to their natural functionalities. However, it has been reported that exosomes obtained by different method may differ in yield and purity [101]. The HEs efficiently accumulated in the fibrotic lesion and exhibited significant penetration of pulmonary fibrotic tissue for the improved affinity for fibroblasts by homologous exosome.

Comparison of natural exosomes and artificial exosomes

Along with the rapid growth in nanobiotechnology, the research of exosomes has been advancing from biology [32, 106] to biomarkers [107, 108] and nanomedicines [109] over the past decade. For drug delivery, exosomes have shown various advantages compared with conventional synthetic materials such as liposomes [110]. The therapeutic potential of exosomes-mediated drug delivery are still in tests of preliminary clinical trials (pancreatic cancer: NCT03608631; acute ischemic stroke: NCT03384433; colon cancer: NCT01294072), while the efficacy of cell-derived exosome-like vesicles has been evidenced in several pilot trials [111,112,113]. Nevertheless, clinical translation of natural exosomes has been challenging [114]. Major hurdles including large-scale production, standard purification protocols, characterization of complex composition, cargo loading, quality control and storage stability are in their way to products for therapeutic applications [115]. Mass production of exosomes may be achieved through the development of bioreactors [38]; however, as biological components, their standardized and reproducible production requires comprehensive control of genetic stability and culturing condition of producing cells; purification requires subtype identification and quantification of contaminants [116]; efficacy is dependent on drug loading efficiency and capacity [117]; storage of therapeutic exosomes is supposed to have high recovery without damage to exosome particles as well as their biological contents [118, 119]. From a current perspective, the development of exosomes for therapeutic drug delivery is still in infancy and the cost for translational research and clinical application would be very high.

In recent years, artificial exosomes have been developed with higher pharmaceutical acceptability to overcome the drawbacks of natural exosomes as new theranostic biomaterials for potential clinical applications [45]. However, there are some challenges for different strategies develo** artificial exosomes for the following aspects: yield, procedures, time–cost, manpower, sustainability, characterization and efficacy, which are summarized and compared to natural exosomes (Table 4).

Table 4 Comparison of natural exosomes and different types of artificial exosomes for translational nanomedicine

Top-down strategy is the most widely reported method for obtaining artificial exosomes. One major strength of the top-down strategy is the applicability because NVs are fully biological and have similar physiochemical and biological features to natural exosomes. Similar to natural exosomes, the yield of artificial exosomes by top-down strategies may vary from cell to cell. For the serial extruding method, most studies reported a nearly 100-fold higher yield than natural exosomes, but higher yields have also been reported and the maximum was 500-fold (Table 1). Compared to natural exosomes, preparation of artificial exosomes by top-down strategies could be cumbersome as UC-based purification procedures are still required following serial extruding or nitrogen cavitation. The development of specific devices may simplify the procedures and increase production efficiency and reduce time–cost and manpower [63]. However, for generating NVs, cells are sacrificed and broken into fragments, leading to limited production sustainability. Also, it has not been raised whether natural exosomes in cells should be considered as contaminants that may influence the characterization of NVs.

Bottom-up strategies are able to produce “clean” artificial exosomes with determined formulations and have the highest scalability because synthetic materials could be feasibly obtained and used for massive production. Besides, the cost of time and manpower could be remarkably reduced by using synthetic materials. Multiple modifications of liposomes are still challenging for complex procedures, uncertain conditions and instability. Another major drawback of bottom-up approaches for generating liposome-based artificial exosomes is that synthetic materials can still hardly mimic the complex composition of natural exosomes. Therefore, the functions of natural exosomes can hardly be fully reproduced by artificial exosomes based on bottom-up strategies. Publications that have compared artificial exosomes with natural exosomes are scarce. Most studies only evaluated physicochemical and biological properties of artificial exosomes (Table 2). A previous study reported a preliminary comparison concerning drug delivery efficiency [87]. Currently, preparing artificial exosomes that fully assembles components of natural exosomes may be pharmaceutically impossible [64], modification of liposomes with key proteins is dependent on the purpose and may not be consistent with natural exosomes. Therefore, the translational applicability of bottom-up strategies is limited.

Biohybrid approaches can only produce semi-artificial exosomes. The yield of artificial exosomes by biohybrid strategies would not be very high as natural exosomes are still required. In addition, the involvement of natural vesicles may face methodological challenges similar to natural exosomes. Preparation of biohybrid exosomes could be laborious as preparation of synthetic liposomes and isolation of natural vesicles are both required. Characterization of hybrid particles may be influenced by additional liposomes and exosomes and the purification would be rather difficult as liposomes, exosomes and the hybrid are very similar in multiple aspects. However, one strength of the biohybrid approach is that the hybrid possesses natural components from exosomes, despite dilution, those hybrid nanocarriers may have improved delivery efficiency than liposomes and higher stability than exosomes, leading to high applicability. Besides, the fusion method that is widely used in biohybrid strategies provided a feasible option for drug loading such as loading biological cargoes into liposomes and loading exogenous therapeutic agents into exosomes [98].

Concluding remarks

Biomimetic nanocarriers are the next generation of drug delivery systems in nanomedicine for improving health. Advancements in nanobiotechnology provide avenues for the development of artificial exosomes that may accelerate clinical translation for nanomedicine application. Currently, natural exosomes are just in their preliminary clinical trials and artificial exosomes are not yet ready for translation. Major challenges include the preparation protocols, characterization and biocompatibility concerns. Artificial exosomes have commercial advantages for their up-scale productivity. In the future, novel and multifunctional artificial exosomes will be developed, with contributions from multidiscipline efforts of biotechnology, nanotechnology, chemical engineering and pharmaceutical industry, to improve healthcare. We hold confidence for artificial exosomes’ potentials for personalized nanomedicine.