When pathogens invade, macrophages, as one of the major immune cells, fight against infection by expressing a series of receptors to trigger innate immunity. This surveillance recognition is achieved through sensors called pattern recognition receptors (PRRs) that detect pathogen-associated molecular patterns (PAMPs) and damage-related molecular patterns (DAMPs) (Plüddemann et al., 2011; Broz and Monack, 2013). After the detection of pathogens, phagocytosis plays an essential role in anti-bacterial host defense, which is mainly manifested by the effective internalization of pathogens. In such a conversion, the engulfment of pathogens by macrophages and a series of sequent membrane remodeling leads to the formation of a membrane-bound vesicle named phagosome. Following the internalization, it is a clearance process in which the phagosome acquires bactericidal and degradative functions termed phagosomal maturation. At the terminal stage of its maturation, the microenvironment of the phagosome becomes highly acidic, degradative, and oxidative, which all contribute to clearing the invaded pathogens. The low pH in phagosomes is related to the progressive acidification within the vacuole, which is realized by the V-ATPase-mediated proton pump (Flannagan et al., 2009). Meanwhile, this acidic condition favors the subsequent formation of phagolysosomes and optimal enzymatic activity of hydrolases in lysosomes.

Another mechanism to kill bacteria is the delivery of molecules with degradative functions, such as defensins, cathelicidins, lysozyme and hydrolases, into phagosomes. Defensins are able to permeabilize bacteria membranes due to the formation of ion transport channels. Cathelicidins, on the other hand, induce permeabilization acting on the cell wall as well as both the outer and inner membranes of bacteria. In addition, hydrolases targeting carbohydrates and lipids also exist in phagosomes, which degrade a wide range of invading bacterial components (Flannagan et al., 2009).

Phagocytes could also act through a large amount of reactive oxygen species (ROS) and reactive nitrogen species (RNS) produced by the NOX2 NADPH oxidase (Quinn and Gauss, 2004; Minakami and Sumimoto, 2006) and NOS2 nitric oxide synthase (Fang, 2004). The transfer of electrons from NADPH to molecular oxygen results in the formation of superoxide radicals (O₂·⁻), which are released into phagosomes (Quinn and Gauss, 2004; Mizushima et al., 2011). After that, O₂·⁻ in phagosomes react with H₂O₂ to produce hydroxyl radicals (·OH) and singlet oxygen (¹O₂) (Minakami and Sumimoto, 2006). It is worth noting that the fusion of the lysosomes and vacuoles leads to partial disruption of the membrane connection, releasing the contents, such as myeloperoxidase, into the phagosomes. Myeloperoxidase within the phagosomes can catalyze abundant H₂O₂ and halogen ions to highly bactericide hypochlorous acid (HClO) and chloramines (Thomas, 1979; Shepherd, 1986). RNS is also essential antimicrobial effectors, which reacts with ROS to destroy pathogens, causing nitrosative stress. The production of RNS starts with NOS catalyzing L-arginine and citrulline to produce nitric oxide (NO), which begins with superoxide to form peroxynitrite (ONOO⁻), which is a highly reactive species that can directly act with several biological targets and cell components, including lipids, amino acid residues, and DNA bases (Webb et al., 2001). Besides, peroxynitrite also has the ability to get across cell membranes to some subcompartments like phagosomes through anion channel (Fang, 2004).

Accordingly, those produced free radicals and oxidation-state components contribute to protein denaturation and lipid peroxidation by oxidative damage, which leads to irreversible damage to the invading bacteria (Boyle and Randow, 2013). This natural defense process provides ideas for biomimetic strategies to eliminate intracellular bacteria.

Unlike extracellular bacteria, intracellular bacteria that can survive and replicate in host cells, especially macrophages, adapt to challenging intracellular microenvironment and evolve intelligent mechanisms to evade host clearance. Those intracellular bacteria are typically divided into two categories: phagosomal and cytosolic bacteria. Most intracellular bacteria studied to date are stored in phagosomes. In detail, the phagosomes provide a safe haven for bacteria, shielding them from immune system detection, and facilitating their replication using the components within the phagosomes (Cullinane et al., 2008; Lamkanfi and Dixit, 2010). Alternatively, cytosolic bacteria benefit from rich nutrient conditions and a relatively spacious microenvironment, enabling them to survive in the host cell cytoplasm despite the presence of immune defenses. In general, treatments that can eliminate bacteria in phagosomes can also act on cytosolic bacteria. It is much thornier to treat phagosome bacteria than cytosolic ones, so our review focus on phagosome bacteria.

Survival of intracellular bacteria presents three main challenges: evading immune system surveillance, resisting the microenvironment, and evading the phagolysosomal pathway. Numerous strategies, including actin-based cell-to-cell spread, low expression of flagellin, avoidance, blockage and adaptation to the phagolysosomal pathway, are utilized by intracellular pathogens.

Different species of bacteria have their own intracellular lifestyles. The majority of phagosome bacteria have abilities to prevent phagosomes maturation, the terminal stage before lysosomes fusion (Ray et al., 2009). For instance, some bacteria have evolved metabolic pathways to prevent acidification in phagosomes, or express specific proteins to withstand low pH microenvironment (Park et al., 1996; Vandal et al., 2008; Huang et al., 2009; Martinez et al., 2011). In addition, some bacteria are able to express detoxifying enzymes (Schmidtchen et al., 2002) like catalase or superoxide dismutase to balance ROS/RNS levels within phagosomes (John et al., 2001; Ng et al., 2004), or interfere with the biological function of enzymes that catalyze ROS/RNS production (Mott et al., 2002). As a result, bacteria protect themselves from being destroyed and eliminated by ROS/RNS (Rudel et al., 2010; Ashida et al., 2011).

Other bacteria escape into the cytoplasm, which constitutes a wild and favorable microenvironment, through sophisticated mechanisms, such as escaping from vesicles, and permeabilizing phagosomes. For those bacteria, it is essential to escape the phagosome subcompartments as early as possible after internalization to avoid fusion with lysosomes. In this process, protein secretion plays a key role in allowing bacteria to cross the cytoplasmic membranes, cell walls, vacuole and host cell membranes, to achieve both intracellular vacuole spread and cell-to-cell spread. Among them, the four major protein secretion ssystems are the type III secretion system (T3SS) (Cornelis, 2006), type IV secretion systems (T4SSs) (Vogel et al., 1998; Christie and Cascales, 2005), type VI secretion systems (T6SSs) (Coulthurst, 2013; Kudryashev et al., 2015) and type VII secretion system (T7SS) (Abdallah et al., 2007), which hold the potential to serve as inhibitory targets for treatment design.

With these findings in mind, a comprehensive understanding of the mechanisms by which intracellular pathogens evade host cells and respond to the innate immune systems are critical to elucidate the pathogenesis of intracellular infections. Moreover, systematically deciphering the interactions between host cells and intracellular pathogens may provide clues for designing advanced nanotherapeutics against intracellular infections.

Unlike non-phagocytes, which tend to engulf spherical nanoparticles in the 20–50 nm range, phagocytes preferentially take up micro scale particles (González et al., 1996; Champion et al., 2008; Jiang et al., 2008). Experimental results for silver nanoparticles have shown that well-dispersed 20–200 nm particles were internalized better by non-phagocytes than phagocytes (Lankoff et al., 2012), whereas aggregated silver particles were more likely to be internalized by phagocytes (Wang et al., 2012). The same phenomenon has been found for smaller nanoparticles like iron oxide particles, where small iron oxide particles exhibited a higher level of phagocytic accumulation than ultra-small particle size particles (Raynal et al., 2004).

This difference in internalization may be attributed to the influence of size on internalization pathways. Normally ultra-small particles are not recognized as exogenous agents by macrophages and can enter directly into cells based on pores in the cell membranes, whereas microscale particles are more likely to be absorbed by the reticuloendothelial system. Besides, smaller sizes possess larger surface areas, which contributes to particles diffuse into cells. When the diameter of the nanospheres is less than 200 nm, their penetration is mainly regulated by the clathrin pathway, but when the size increases to 500 nm, their internalization is mainly mediated by caveolae pathway (Rejman et al., 2004). The effect of size is also significant in the uptake of different sized anionic polystyrene particles, with smaller particles being taken up mainly through clathrin-independent cavelae-independent pathways, while larger particles are uptaken via clathrin-mediated endocytosis (Lai et al., 2007). This distinction can be explained by the fact that the clathrin-mediated pathway has a higher uptake rate than clathrin-independent cavelae-independent pathways, resulting in particles internalized through this pathway exhibiting faster accumulation within cells.

Surface charges also play a key role in cellular uptake. It is mainly manifested in the fact that the neutral surface charge nanomaterials have a lower plasma protein adsorption rate, accompanied by a longer blood circulation time, which results in a higher cell uptake due to a longer margin from the phagocytes.

Positively charged nanomaterials bind to negatively charged cell membranes through electrostatic interactions (Tahara et al., 2009; Duceppe and Tabrizian, 2010). A mass of works has demonstrated that positively charged particles were internalized into cells at a higher degree than their respective anionic particles, such as gold and silver particles, iron oxide particles, silicon dioxide, chitosan, liposome, and polymers. Besides, for negatively charged nanomaterials, the internalization decreases with increasing surface charge, while positively ones, on the contrary, performs a positive correlation with a certain range of surface charge. Surprisingly, a representative example is antibacterial silver nanoparticles coated with chitosan (CS-AgNPs) with enhanced antibacterial effect (Jena et al., 2012). The designed CS-AgNPs exhibited a significantly improved therapeutic effect on intracellular bacteria mainly due to their strong cell internalization.

Several other researches have shown some contradictory results, which might contribute to negatively charged nanoparticles can promote cellular uptake via regulating the formation of aggregation and cluster after initial electrostatic repulsion. Taking the uptake of liposome with different surface charges as an example, a negatively charged PLGA-lipid hybrid system performed better uptake behavior compared to a positively charged system (Maghrebi et al., 2020). Another hypothesis is that bacterial surfaces also present a negative charge, and phagocytes may preferentially take up anionic particles (Fröhlich, 2012). Moreover, some surface groups with negative charges, such as citrate groups, can improve the stability of nanoparticles in culture media and increase their affinity for cell membranes (Kolosnjaj-Tabi et al., 2013).

Shape is another important factor influencing the internalization of nanomaterials, especially when endocytosis pathways of nanomaterials need to be mediated by receptors. Although the surface area of rod-shaped nanomaterials is smaller than that of spherical particles, the limited binding sites on its surface can more efficiently recognize and bind to target cell surface receptors due to its aspect ratio. This allows rod-shaped nanomaterials to exhibit higher cell adhesion efficiency than spherical ones (Kolosnjaj-Tabi et al., 2013; Shao et al., 2017). On the other hand, nanomaterials with sharp shapes such as spines are able to locate in the cytoplasm due to their better ability to penetrate membranes, which enables them to remain in cells benefiting from low exocytosis (Chu et al., 2014).

Surface functionalization of nanomaterials not only regulates surface charge, but also improves properties such as hydrophobicity and softness. The most common strategy is surface polyethylene glycolation (PEGylation), which is mainly used to reduce the hydrophobicity of nanomaterials and alter their biological characteristics. PEGylation forms a hydrated layer that reduces the adsorption of serum proteins, increases the hydrophilicity, and reduces macrophage uptake, which is often referred to the “stealth effect” (Figures 2A, B) (Yang et al., 2014; Sanchez et al., 2017). Additionally, the difference in uptake may also be related to the change in particle softness by functionalized modification, where softer particles are more likely to be taken up by macrophages. Surface functionalization also enables active targeting of nanomaterials. For example, nanoparticles modified with polysaccharides could interact with specific receptors on cell membranes, resulting in active targeting for more precise internalization. Surface arginine-modified mesoporous silica nanoparticle (MSN) was able to co-localize with intracellular Salmonella, leading to efficient antibiotic delivery (Figures 2C–E) (Mudakavi et al., 2017).

According to the above discussion, bacteria invade cells and locate in different sub-compartments, making it crucial for nanotherapeutics to deliver antibiotics in on-demand manner. This approach can improve treatment efficacy and reduce toxicity to normal tissues. Furthermore, bacterial invasion creates a unique infection microenvironment characterized by low pH, related enzyme secretion, and slight temperature changes. These conditions can be leveraged as stimulus to achieve drug targeting and release.

Beyond the application as drug delivery carriers, some nanomaterials with intrinsic antimicrobial bioactivities have recently shown promising aspects for potential clinical applications against intracellular bacteria.

From above mentioned findings, these bioactive nanomaterials can serve as alternative treatments to traditional antibiotics due to their multiple mechanisms involving intrinsic enzyme-mimic activities, metal toxicity, or physicochemical properties, making them less prone to cause antibiotic resistance. To achieve more efficient synergistic antibacterial effects, researchers are exploring the application of multifunctional nanomaterials integrating the functions of delivery with antibacterial activities (Chang et al., 2017a; Chang et al., 2017b; Lu et al., 2017; Lu et al., 2018; Zhang et al., 2020; Huo et al., 2021; Li et al., 2021; Liu et al., 2021).

Dube et al. presented nanoparticles with chitosan-functionalized shells and PLGA cores for the treatment of tuberculosis (Dube et al., 2014). This rationally designed nanoparticle showed a collaborative function between stimulation of ROS/RNS and delivery of rifampicin. In this case, the combined nanoparticles can not only regulate ROS/RNS levels but also modulate pro-inflammatory cytokine secretion. Meanwhile, it acted as a vehicle to transfer rifampicin inside alveolar macrophages. Similarly, Marwa et al. prepared Ag-coated PLGA particles loaded with pexiganan, which exhibited coordinated antibacterial functions, and were specifically uptaken by macrophages, but not by any non-phagocytic cells (Elnaggar et al., 2020). Additionally, Timothy et al. established composite materials containing Ag and ZnO for the targeted delivery of rifampicin (Figures 5E, F) (Ellis et al., 2018). The sophisticated nanoplatforms were developed to destabilize membranes, increase permeabilization of intracellular Mycobacterium tuberculosis and enhance penetration of rifampicin. As a result, multifunctional nanomaterials presented extraordinary bactericidal effects compared to free antibiotics.