The TME includes hypoxia, acidic pH, angiogenesis, and immunosuppression. Due to the rapid growth of tumor tissues with active metabolism, it increases its oxygen demand. At the same time, the vascular system in and around the tumor is disorganized and irregular, resulting in inadequate oxygen diffusion (Jain, 2013). Therefore, the central area of the tumor presents a necrotic, hypoxic environment with an oxygen partial pressure as low as 7–28 mmHg, whereas the oxygen partial pressure of normal tissue is 40–60 mmHg (West and Slevin, 2019). Bacteria, as natural prokaryotic cells, can actively migrate into the hypoxic TME by their autonomous propulsion and anaerobic properties (Kasinskas and Forbes, 2006). Live facultative/obligate anaerobes such as E. coli, Salmonella,Bifidobacterium bifidum, and S. pyogenes can sense hypoxic environments by using their chemoreceptors, which are naturally targeted to hypoxic and necrotic areas of tumors (Kucerova and Cervinkova, 2016).

Due to insufficient oxygen supply, tumor cells are limited to metabolizing energy through anaerobic fermentation, which leads to lactic acid accumulation and the formation of an acidic tumor microenvironment (Kumari et al., 2021). Tumor acidic environment can also induce bacterial-specific gene activation. Flentie et al. (2012) co-cultured Salmonella with melanoma cells or colon cancer cells and identified 5 genes that can be activated specifically by cancer cells. Further studies revealed that the activation of these genes may be caused by the acidic microenvironment of the tumor cells, suggesting that the bacteria are capable of responding positively to the acidic TME. Tumor cells release a substantial quantity of vascular endothelial growth factor during growth, inducing the formation of chaotic and irregular tumor-specific blood vessels. These blood vessels can enclose the bacteria, not only promoting bacterial colonization but also providing an adequate supply of nutrients to the bacteria, which is conducive to bacterial proliferation (Khalaf et al., 2021). Hypoxia also leads to suppression of local immune cell function resulting in an immunosuppressive microenvironment (Chouaib et al., 2017). In tumor tissues, the ability of immune cells to recognize and kill bacteria is inhibited, which provides a favorable environment for bacterial survival. It has been shown that bacteria are more likely to survive in a tumor-immunosuppressive environment (Stritzker et al., 2010; Tian et al., 2022).

In conclusion, the characteristics of the TME and the interactions between bacteria and tumors are not only important for the natural targeting of bacteria to tumors but also provide a basis for modifying bacteria and improving their targeting ability.

Most bacteria used for tumor therapy have flagella, which mainly perform chemotaxis, motile, and invasive functions. These functions performed by the flagellum can help the bacteria to better penetrate the intratumor tissue, which is crucial for bacterial colonization in tumors (Min et al., 2008; Zhao et al., 2005; St Jean et al., 2008; Kasinskas and Forbes, 2007).

Bacteria and their secreted metabolites such as peptidoglycan, and lipopolysaccharide (LPS) augment the antigenicity of tumors and provide potent immunostimulatory signals. They combine with pattern recognition receptors (PRRs) expressed by immune cells such as dendritic cells (DCs) and macrophages, initiating an immune response that allows the immune system to recognize and kill tumor cells. Bacteria and their metabolites also initiate an adaptive immune response to kill tumor cells via natural killer cells, CD4⁺, and CD8⁺ T cells (Zitvogel et al., 2016).

Furthermore, certain components in bacteria can also cause phenotypic transformation of immune cells. Flagellin was found to convert pro-tumor macrophages into anti-tumor macrophages and to transform the immunosuppressive microenvironment into an environment of normal immune function (Zheng et al., 2017).

Genetic engineering of bacteria to develop safer and more effective cancer treatments is a major focus of bacterial anticancer therapies. Bacterial genetic modification can reduce bacterial cytotoxic effects to improve safety (Hill et al., 2011)and enhance bacterial targeting, invasiveness, and tumor cell killing (Yoon et al., 2017).

In addition, synthetic biology methods to implant sensing and responsive genetic circuits in bacteria enable bacteria to synthesize or release drugs in response to specific sensing stimuli. Moreover, combining multiple response circuits with bacteria through synthetic biology approaches could enable engineered bacteria to respond to the complex physiological environment of the organism. This allows for better targeting and control of the bacterial carrier delivery system, enabling precise drug delivery and reducing drug side effects (Feng et al., 2023).

Whole bacteria can be categorized into inactivated and live bacteria, which can be improved through genetic engineering or other technological methods to enhance their ability to deliver drugs to treat cancer.

Inactivation of bacteria is a special inactivation technique that prevents the bacteria from growing and reproducing while retaining their original structure and characteristics. Records of inactivated bacteria for the treatment of cancer date as far back as the advent of “Coley’s toxin” in the 19th century, when William Coley injected inactivated S. pyogenes and Serratia marcescens into sarcoma patients (Hoption Cann et al., 2003). Following Coley’s work, Teoh’s laboratory demonstrated that the inactivation of Clostridium sporogene, a naturally tumorigenic bacterium, significantly reduced the proliferation of CT26 and HCT116 colorectal cancer cells to 57.3% and 26.2%, respectively (Figure 2) (Bhave et al., 2015). However, inactivated bacteria often suffer from insufficient anti-tumor effects. To address this problem, Yang and Xuan incorporated 125I/131I into inactivated bacterial carriers, which can prolong the presence of radioactive iodine at the site of the lesion, thus enabling internal radioisotope therapy (IRT) of the tumor (Pei et al., 2022).

In contrast to inactivated bacteria, live bacteria can actively target most tumors, including metastases (Forbes, 2010), and most live bacteria can better penetrate and colonize intratumoral tissues depending on the flagellum (Felgner et al., 2018). However, the toxicity and uncontrolled immunogenicity of the bacteria greatly limit the use of live bacterial therapies in practice. Besides, bacteria can suppress macrophage migration and disrupt the integration of the barrier established by epithelial cells, thereby allowing tumors to diffuse and invade surrounding healthy tissues (Gagnaire et al., 2017). Therefore, improving the safety of bacterial therapy has become a key issue in bacterial cancer therapy.

The attenuated modification of bacteria is a common method to improve the safety of bacterial therapies. The BCG vaccine is a live bacterial preparation made from attenuated suspensions of Mycobacterium bovis by Charles Calmette and Camille Guérin in 1900 (Kamat et al., 2015). When they cultured M. bovis in bile-containing media, they found that the bacterial virulence gradually decreased and after more than 1,000 passages the attenuated strain that we use today was obtained. Bladder instillation of BCG is a standard treatment option for patients with high-risk non-muscle invasive bladder cancer in the clinic. Compared with traditional means of inactivation and attenuation, obtaining attenuated bacteria by knocking out virulence factors through genetic engineering techniques not only reduces the pathogenicity of the bacteria but also preserves their tumor-targeting properties. For example, an attenuated Salmonella typhimurium strain (named VNP20009) obtained by deletion of the purI and msbB genes not only dramatically improved safety, but also promoted their accumulation in the tumor tissue (Toso et al., 2002).

Subsequently, a large number of researchers have concentrated on attenuating the virulence of various bacteria, including Salmonella, Clostridium, E. coli, and Listeria, through genetic and chemical modifications (Felgner et al., 2018; Mercado-Lubo et al., 2016; Fritz et al., 2016; Dang et al., 2001; Chowdhury et al., 2019; Zhou et al., 2018). In addition, engineered bacteria are a promising carrier for drug delivery, and today substantial studies are using engineered bacteria for targeted delivery of interfering RNA (Jiang et al., 2007), DNA vaccines (Kong et al., 2012), proteins (Binder et al., 2013), and small molecule therapeutics (Xie et al., 2017) for the treatment of malignant tumors. Protein therapeutic molecules, including antibodies (Daassi et al., 2020), inflammatory factors (Schmitt and Greten, 2021), metabolism-blocking proteins (Liu et al., 2017), enzymes (Yu et al., 2024), and apoptosis-inducing agents (Hu and Kavanagh, 2003) have also been successfully combined with engineered bacteria for targeted therapy of tumors (Table 1).

Bacterial skeleton includes bacterial ghosts (BGs), and cell wall skeleton (CWs). BGs are cavities formed by lysing Gram-negative bacteria using φX174 phage lysin gene E. BGs have large inner space and intact outer walls, and can be applied as a drug delivery carrier to load cargoes of drugs, antigens, and nucleic acids (Rabea et al., 2020; Xie et al., 2020; Tabrizi et al., 2004). It has been shown that encapsulation of doxorubicin (DOX) in BG derived from Mannheimia haemolytica not only targets human colorectal adenocarcinoma cells (Caco-2) releasing large quantities of DOX within tumor cells but also enhances cytotoxicity against tumor cells (Paukner et al., 2004). In addition, BG retains various antigenic components of the bacteria, allowing them to also be used as a biologic agent and immune adjuvant for immunotherapy (Kudela et al., 2005). For example, when BG carries a plasmid, it is recognized and phagocytosed by antigen-presenting cells and efficiently activates CD4⁺ and CD8⁺ T cells to generate an anti-tumor immune response (Warrier et al., 2019).

CWS is prepared from bacteria by hydrothermal treatment and is an agonist of TLR2 and TLR4 and can be used as an immune adjuvant. CWS from M. bovis BCG has Immune activation ability (Ni et al., 2017). Researchers have found that the shape of CWS affects DC internalization (Yoo et al., 2002). The results indicated that the CWS of Lactobacillus rod-shaped exhibited the greatest amount of DC internalization compared to other shapes. Also, rod-shaped Lactobacillus CWS upregulated immunoreactive cytokines, which can be used for TME regulation or loading stimulatory signals in next-generation delivery systems (Seya et al., 2001).

Bacterial components include magnetosomes, spore microcells, and bacterial outer membrane vesicles. Magnetotactic bacteria (MTB) were found by American scholars in 1975 (Makela et al., 2022). This group of bacteria can be driven by flagella to move in the direction of magnetic lines of force and are well targeted in cancer therapy. Magnetosomes are intracellular organelles of MTB containing magnetic minerals such as magnetite Fe3O4 or greigite (Fe3S4) encapsulated by proteolipid membranes (Grünberg et al., 2004; Schüler, 2008). Magnetosome membranes are abundant in primary amine groups, which can bind certain chemotherapeutic drugs to the amine groups on the surface of the magnetosome for cancer therapy. Polymers wrapped with small molecule drugs can be attached to the magnetic vesicles to improve their drug-carrying efficiency and stability. Alternatively, chemical drugs with amine groups can be directly coupled to magnetosomes by crosslinking agents (Sun et al., 2011). Some researchers used polyethyleneimine (PEI) as a cross-linking agent to enable the magnetosomes to bind the antitumor drug DOX (loading rate of 57.7%) and small interfering RNAs (siRNAs) via a hydrazone bond (Long et al., 2018). In addition, genipin and polyglutamic acid can be selected as double cross-linking agents to load the anticancer drugs acitretin (Ara-c) and doxycycline (DNR) on the surface of the magnetic vesicles (68.4% encapsulation and 32.4% drug loading rate for Ara-c and 36.1% encapsulation and 17.9% drug loading rate for DNR) (Long et al., 2016). It was found that this magnetic vesicle drug-carrying system exhibited good stability and long-term sustained drug release, and significantly reduced the nonspecific toxicity of the drug. Meanwhile, as magnetic nanoparticles with good magnetic responsiveness, magnetosomes are able to target tumors under the action of the external magnetic field. The targeting ability can be further enhanced by combining specific targeting molecules with chemical groups on the membrane of the magnetosome (Zhang et al., 2018a). Moreover, according to the properties of magnetosomes, it is also possible to combine the delivered drugs with various therapeutic regimens, including chemodynamic therapy (CDT) (Ye et al., 2022), magnetically induced thermotherapy (Liu et al., 2012), magnetically induced photothermal therapy (PTT)and magnetically induced immunotherapy (Zhang F. et al., 2019). For example, the researchers extracted magnetosomes from the Magnetospirillum magneticum AMB-1 and used them as a “magnetosome chassis” (MSC) to generate gold nanoparticles in situ. The resulting hybrid (MSC-Gold) is capable of targeting tumors in a tumor-centric magnetic field and enables combined chemodynamic and photothermal therapies (Figure 3) (Ye et al., 2022).

Spores are formed in extreme environments and have functional components that can be used in cancer therapy. Clostridium spp. as a spore-producing bacterium can survive in hypoxic environments, and the spores it produces can germinate in hypoxic and necrotic areas of tumors and release carrier substances to shrink tumors (Mengesha et al., 2006; Mengesha et al., 2007). Genetically modified C. novyi-NT spores lack lethal toxins and have anticancer activity along with improved safety. Intratumor injection of Clostridium histolyticum spores and intravenous injection of Clostridium perfringens spores lead to lysis of tumor tissues without adverse effects on normal tissue (Kubiak and Minton, 2015). In addition, bacterial spores serve as carriers for anticancer drugs, therapeutic proteins, and cytotoxic peptides.

Minicells are the products of abnormal bacterial cell division, have a nanometer size, and are controlled by mutated minCDE genes. The capability of minicells to package multiple drugs by unidirectional diffusion, independent of their physicochemical properties, makes them ideally suited for drug delivery (MacDiarmid et al., 2009). In addition, minicells carry all the ingredients of the parental cells except chromosomal DNA and therefore cannot proliferate,which provides great safety for their use as carriers of drug delivery for the treatment of tumors (Di Ventura and Sourjik, 2011). For example, in animal studies, combining DOX with tumor-targeting minicells resulted in significant regression of in situ breast tumors, and no serious toxic side effects were observed.

In addition, minicells are easy to modify, and researchers can enhance their targeting and killing of tumors by modifying minicells with ligands or antibodies. For example, MacDiarmid et al. modified minicells using antibodies that enabled the minicells to target epidermal growth factor receptors (EGFR) or human epidermal growth factor receptor 2 (HER2) receptors that are specifically overexpressed on tumor cell membranes. These modified minicells carry drugs directly into target tumor cells after packaging DOX and paclitaxel (PTX) therapeutic drugs. In addition, the researchers modified the DOX-loaded minicells so that they could target drug delivery to solid tumors (Figure 4) (Zhang et al., 2018b). Minicell-based forms of drug delivery are more effective than free drugs and can effectively reduce the toxic side effects of drugs on normal tissues (MacDiarmid et al., 2007). Currently, EGFR-modified paclitaxel-loaded S. typhimurium minicells have been successfully tested in phase I trials, and no patients have died during the observation period due to adverse effects (Solomon et al., 2015).

To deliver drugs, minicells have an advantage in delivering RNAi and can be efficiently loaded with siRNA and shRNA (Jivrajani and Nivsarkar, 2019). Based on this feature, MacDiarmid et al. utilized microcells to deliver siRNA/shRNA for the treatment of cancer, and the loaded siRNA/shRNA induced apoptosis in tumor cells (Jivrajani and Nivsarkar, 2016). And because their cell walls contain varying levels of LPS, they are not only expected to be a drug delivery system (Giacalone et al., 2007) but also to act as immune adjuvants for antitumor effects (Lundin and Checkoway, 2009).

Bacteria secrete membrane vesicles called bacterial extracellular vesicles (bEVs), spherical structures with a lipid bilayer containing various biomolecules from the parent bacterium. bEVs derived from the outer membrane of Gram-negative bacteria are known as outer membrane vesicles (OMVs) (Jagannadham and Chattopadhyay, 2015). The OMVs carry periplasmic and cytoplasmic components, a large number of microbe-associated molecular patterns (MAMPs) (Ellis and Kuehn, 2010), and functional components. bEVs derived from the cytoplasmic membrane of Gram-positive bacteria are known as cytoplasmic membrane vesicles (CMVs), which contain material from the cytoplasm but do not contain components that can cause an acute toxic response (e.g., LPS and other cell wall components, etc.) (Silhavy et al., 2010).

bEVs have many characteristics such as good biocompatibility and easy genetic modification, so they can be used as delivery carriers for anticancer agents. The bEVs-mediated targeted delivery system of chemotherapeutic agents is a perspective method in oncology therapy to improve effects and reduce adverse effects caused by systemic therapy. Currently, bEVs have been successfully loaded with different kinds of chemotherapeutic drugs, such as PTX, DOX, gemcitabine, and tegafur (Figure 5) (Chen Q. et al., 2019; MacDiarmid et al., 2016; Sagnella et al., 2020; Sagnella et al., 2018; Alfaleh et al., 2017).

In addition, bEVs can carry antitumor substances such as antigens, TLR agonists, and photosensitizers (Qing et al., 2020; Chen Q. et al., 2020; Wang et al., 2020). Since bEVs inherit various immunostimulatory molecules from parental bacteria, they can generate immune responses against pathogens, which gives them the potential for immunotherapy of cancer (Nagakubo et al., 2020).

Combining the capacity of bEVs to deliver drugs and immune adjuvants could further enhance their anti-tumor therapeutic potential. This holds promise for complete tumor eradication and prevention of tumor recurrence and metastasis. Examples include loading bEVs with chemotherapeutic agents (Kuerban et al., 2020) or adding photothermal agents to bEVs. Chemotherapeutic agents or PTT can lead to immune cell death in tumor cells, which, in combination with the immune properties of bEVs, can enhance the efficacy of anti-tumor therapy (Wang et al., 2020).

In conclusion, using bacterial components as delivery carriers for anticancer therapy is an innovative and evolving study (Table 2).

Bacteria-nanoparticle biohybrid systems have good biocompatibility degradability, and high drug loading. In this part, we will conclude the use of bacterial-nanoparticle biohybrid systems in combination with different therapeutic regimens (Table 3).

Nanoparticles can be obtained by synthesizing them by various chemical methods, but chemical synthesis methods have disadvantages such as high toxicity, low purity, and poor biocompatibility. Synthesizing various nanoparticles by using bacteria as factories or base materials can solve the above problems, and thus has attracted attention. For example, metal nanoparticles can be synthesized naturally in bacteria or through genetic engineering, and most of the bacteria-mediated synthesis of metal nanoparticles has been shown to have antitumor activity in vitro (Kang et al., 2008; Park et al., 2010).

Almalki and Khalifa, (2020) synthesized silver nanoparticles (AgNPs) with diameters between 15 and 40 nm from Bacillus sphaericus KFU36. They found that the synthesized AgNPs could enter the TME and aggregate to induce significant tumor cell apoptosis. AgNPs synthesized by Yang et al. promoted intracellular ROS production in tumor cells, which increased lipid peroxidation and led to significant tumor shrinkage (Yang et al., 2020b). Vairavel uses Enterococcus faecalis to synthesize gold nanoparticles intracellularly, which inhibits tumor cell proliferation by generating ROS in tumor cells (Vairavel et al., 2020). Rajkumar et al. synthesized selenium nanoparticles (SeNPs) using Pseudomonas Schizosaccharomyces, which significantly inhibited tumor angiogenesis and killed tumor cells at low concentrations (Rajkumar et al., 2020).

Biomimetic nanoparticles (NPs) formed using cell membranes encapsulating NPs are a promising biomedical material. Among various types of cell membranes, OMV shows great potential in the biomedical field (Qing et al., 2019). Coating NPs with bacterial OMVs retains the intrinsic properties of synthetic NPs and also enhances the properties possessed by the OMVs (Figure 9). For example, fusion of drug-carrying liposomes with OMVs can further enhance the drug-carrying capacity (Piffoux et al., 2018). Chen Q. et al. (2019) used bacterial OMVs to encapsulate drug-loaded polymeric micelles. The OMVs activated the host immune response and combined with the chemotherapeutic drugs contained in the polymeric micelles, effectively inhibiting tumor growth. In a report by Wang et al. (2020), OMVs and B16-F10 cancer cell membranes were successfully coated on hollow polydopamine (HPDA) NPs. This material could combine immunotherapy and PTT to treat melanoma. Researchers genetically modified a bacterial strain by expressing a tyrosinase transgene and generated bacterial microvesicles loaded with the biopolymer melanin. These bioengineered bacterial microvesicles were used as bio-nano-heaters for not only in vitro and in vivo tumor imaging but also tumor growth inhibition under near-infrared light irradiation (Gujrati et al., 2019).

Chen et al. formed a hybrid cell membrane by fusing a tumor cell membrane with a bacterial cytoplasmic membrane. By encapsulating polylactic acid-glycolic acid copolymer (PLGA) nanoparticles within the hybridized membrane, antigens and adjuvants from both cell membrane sources could be delivered simultaneously, thereby enhancing the innate immune response. Improved anti-tumor effect while avoiding side effects (Chen et al., 2021). In addition, biofilm components extracted from bacteria can also be used as nanoparticles for drug delivery. Yi et al. assembled rhamnolipids extracted from Pseudomonas aeruginosa with the photosensitizer magnesium chlorophyllate (Pba), and the assembled nanomaterials showed a significant increase in accumulation in tumor tissues and demonstrated significant tumor inhibition (Yi et al., 2019). These encouraging studies suggest that nanoparticle-based bionic bacteria hold promise for cancer therapy.

Biological strategies allow for the fabrication of nanohybrids via biorecognition processes (e.g., antigen/antibody and biotin/streptavidin) and metabolic pathways (Mostaghaci et al., 2017). Among them, streptavidin and biotin are the most popular binding pairs used in biotechnology applications, which are highly selective and have strong interactions to withstand extreme temperatures (Tm of 112°C), pH, denaturants, and enzymes (Dundas et al., 2013). Suh et al. used this method to attach streptavidin-labeled drug-carrying poly (lactic-ethanolic acid) nanoparticles to Salmonella typhimurium VNP20009, which formed a biotin surface. They demonstrated a 100-fold increase in the ability to target tumors without any external driver (Suh et al., 2019).

To attach nanoparticles to bacteria through streptavidin-biotin interactions, the utilization of genetically engineered bacteria is an alternative method of bioconjugation. For example, by increasing the binding affinity and introducing new binding sites, bacteria can be preferentially adsorbed to specific nanoparticles (Vizsnyiczai et al., 2020). By designing genetic circuits, it is possible to control the binding of bacteria to nanomaterials and maintain bacterial viability and proliferation. Bioconjugate gene modification with high loading capacity can solve the problem of mutual repulsion between negatively charged nanoparticles and negatively charged bacteria, making it a potential platform for a wide range of applications (Yang et al., 2020c).

Physical strategies involve the use of bacteria as carriers to load or anchor materials or drugs to the outer surface of the bacteria, including electroporation (Ding et al., 2021), electrostatic interactions (Xie et al., 2021), impregnation (Kuang et al., 2023), or membrane coating (Wu et al., 2019).

Electroporation techniques can facilitate the entry of therapeutic drugs into bacteria by applying an enhanced electric field to increase the permeability of the cell membrane. Electrostatic interactions, dip-coating processes, and cell membrane coating techniques can also be used to construct nano-biological mixtures. by anchoring nanomaterials to the outer surface of bacteria. Of these, electrostatic interactions are relatively flexible in the construction of nanohybrid materials, and using this method bacteria can be used with virtually any charged nanoparticle, polyelectrolyte, protein, or polysaccharide.

Although physical strategies have the advantages of simplicity and rapidity in constructing nanohybrid materials, they also have some drawbacks. For example, electroporation affects bacterial activity, while electrostatic interactions are less stable in vivo applications due to the prevailing competitive reactions (Zoaby et al., 2017).

Chemical groups on the surface of bacteria make it possible to bind bacteria and nanoparticles, which can provide a more stable bond between bacteria and nanomaterials through chemical bonding. To date, many chemical reactions have been employed to bind nanomaterials to bacterial surfaces, including carbodiimide chemistry (Taherkhani et al., 2014), Michael addition reactions (Xing et al., 2021), intermolecular disulfide bond cross-linking (Liu L. et al., 2020), host-guest chemistry (Chen QW. et al., 2020), Schiff base reactions (Chen QW. et al., 2020), and click chemistry (Xie et al., 2018). Of these, carbodiimide chemistry has received much attention due to the enrichment of bacterial surfaces with amino or carboxyl groups.

Biomineralization, the process by which organisms utilize mineral elements, is a promising approach to materials engineering. Unlike the method of combining nanoparticles with bacterial surfaces, this approach can assist bacteria form nanoparticles directly. For example, Chen et al. developed a self-mineralizing photothermal bacterium by biomineralizing palladium nanoparticles (PdNPs) on the surface of the parthenogenetic anaerobic bacterium Shewanella oneidensis MR-1 (188). Biomineralization consists of two main methods: biologically induced and biologically controlled. Mineralization during biologically induced processes usually occurs on the bacterial surface and is not directly regulated by genes. In turn, biologically controlled mineralization processes, such as nucleation, growth, and localization, can be regulated by genes and occur both intracellularly and extracellularly. Magnetic nanoparticles (magnetosomes) produced by MTB are a prime example of biologically controlled mineralization (Liu et al., 2024).

Currently, there are three main approaches to preparing bacterial OMV-coated nanomaterials. The first method, physical co-extrusion involves the repeated extrusion of a mixture of bacterial outer vesicles and nanoparticles through a thin porous membrane of nanoscale porous polycarbonate using an Avanti micro-extruder (Wu et al., 2020). Although this process is very efficient for the production of OMV-NPs, the deposition of raw materials on filters causes losses and makes it impractical for wide-scale production.

The second method, the ultrasonic fusion method, in which ultrasonic energy is used to promote the encapsulation of nanomaterials by extracellular membrane vesicles, seems to have solved the production scale problem (Zhang Y. et al., 2019). However, encapsulation inhomogeneity and dimensional variations are the shortcomings of this technique.

The third method is microfluidic electroporation (Rao et al., 2017). technique that involves mixing cell membranes with nanoparticles and then facilitating the entry of the nanoparticles into the cell membranes through microfluidic electroporation. It has the advantages of high accuracy and scalability, uniform product size, reproducibility, and stability.