Desolvating is the main preparation strategy of zein NPs. It refers to adding a reagent that can reduce zein solubility in the solution, so that zein molecules can self-assemble into nanoparticles. Many variations of the desolvating technique exist, includes solvent/nonsolvent method, coacervation method, pH-driven method and dialysis method.

For conventional desolvating methods, zein NPs are formed from the bulk solution. The formed zein NPs are easy to aggregate to form the flocculates or precipitates during the drying process, which limit to some extent their usages in pharmaceutical industries. A general strategy to break this limitation is dispersing the stock solution to small droplets before the desolvating or drying process. The dispersing methods include but not limited to liquid-liquid dispersion, spray drying, electrospraying, supercritical antisolvent (SAS) process and atomizing/antisolvent precipitation (AAP) process. There are mainly two procedures regarding the particle formation: the breakup of stock solution into droplets, and the self-assembly and solidification of these droplets into nanoparticles during solvent attrition.

Different from the above methods, micromixing strategy prepares zein NPs in a confined volume, which allows a continuous operation and a better control on the fabrication of zein NPs. Thus, micromixing strategy is a continuous alternative to fabricate zein NPs. The representative methods are flash nanoprecipitation (FNP) and microfluidic technique.

Particle morphology can be observed using techniques such as scanning electron microscopy (SEM), atomic force microscopy (AFM) and transmission electron microscope (TEM). As an amphiphilic molecule, zein can self-assemble into spheres, micro-capsule, films, rods and particles with other morphologies (Wang et al., 2008; Wang and Padua, 2010; Wang and Padua, 2012a). In most cases, the obtained zein NPs are nanospheres with smooth surface. A possible mechanism for zein self-assembly from single molecules to nanospheres was proposed (Wang and Padua, 2012b). It mainly refers to four steps: α-helix to β-sheet transformation, packing of β-sheet into stripes, curling of stripes into rings, growing and rounding of rings into nanospheres. In some cases, hollow zein NPs are fabricated for their high loading capability, sustained release and low density. For example, hollow zein NPs were fabricated in a two-step process (Xu et al., 2011): preparation of sacrificial cores from sodium carbonate and precipitation of zein onto cores. The obtained hollow zein NPs had average diameters as small as 65 nm, and had 30% higher loading of metformin compared to the solid zein NPs. In another study, crocin loaded hollow dextran sulfate/chitosan-coated zein NPs were fabricated using sodium carbonate as a sacrificing template by a layer-by-layer self-assembly technique (Song et al., 2022).

Particle size is the most important characteristic of nanoparticles in pharmaceutical industry. In general, nanoparticles are applicable to a wide range of biological targets due to their high intracellular absorption. But particles that are ‘too small’ exhibit poor encapsulation efficiency, may migrate from the site of injection, and may exhibit undesirably rapid release of their payload (Gaumet et al., 2008). Particle size is often determined using dynamic light scattering (DLS). Samples analyzed by DLS devices consist of particles that are well-dispersed in liquid media. DLS theory is based on the Brownian motion of particles in suspension, i.e., larger particles move slower than smaller particles at the same temperature. DLS can be applied over a wide size range, from nanometers to micrometers. In the case of polydisperse systems, one can simply determine the mean size and PDI from a cumulant analysis or, by a more complicated method (Glatter, 2018). However, when the particles are irregular and have a strong polydispersity, direct observation and measurement from their SEM or TEM images is suggested.

The particle size of obtained zein particles can vary from nanometers to micrometers under different conditions or preparation methods. The control of particle size is of prime importance in encapsulation and controlled delivery. The size increase of zein NPs during evaporation-induced self-assembly was measured by TEM, and modeled from the hydrophobic and hydrophilic contributions to the interfacial free energy from zein and solvent (Wang and Padua, 2012a). R ² of 0.92 showed their model might allow the prediction of evaporation time and thus control over microsphere size. For self-assembly of zein NPs using BUDP, a semi-empirical model was developed for predicting particle size (Dp ₅₀ ) according to the analysis of influence mechanisms of operating parameters, i.e., D p 50 = 13904 ⋅ 1 + 1220 C − 1 / 3 ⋅ exp 0.049 C P − 0.4 (P is ultrasound power, and C is zein concentration) (Liu et al., 2016b). This model could predict Dp ₅₀ at known experimental conditions or setting the process variables to achieve a given size, which was useful for a system capable of precise particle formation and scaling up BUDP. Besides, the modelling relationship between electrospraying product diameter and key experiment parameters were established by machine learning (Wang et al., 2022b), which could expedite the optimization process by accurately predicting particle diameter for both nano- and micron-sized particles.

Surface charge is a critical element in evaluating the dispersity and stability of nanoparticles in vitro and in vivo. It is often measured by the zeta potential measurement. In general, the absolute zeta potential values >30 mV can prevent the particles from clotting effectively, resulting in the good dispersity and stability of nanoparticles. At the isoelectric point, the zeta potential of particles is zero. Individual zein NPs are prone to destabilization and aggregation in aqueous and physiological fluids, because their neutral isoelectric point (pH 6.2) (Pascoli et al., 2018). The surface charge of zein NPs can be modified by adding surfactants, coating and/or incorporating with other biopolymers. For example, sodium caseinate was used to stabilize zein colloidal particles, which showed a shift of isoelectric point from 6.0 to around pH 5.0 (Patel et al., 2010b). In another study, the complex nanoparticles based on zein and fucoidan were developed to encapsulate resveratrol (Liu et al., 2022b). The obtained particles showed stable colloidal state and no visual aggregation at pH 2.0–8.0. These results are attributed to that fucoidan contains many sulfated groups and can be ionized at pH > 2.0, which provide electrostatic repulsion among particles.

The intermolecular interactions among different components of zein NPs are often characterized by circular dichroism spectroscopy, Fourier transform infrared (FT-IR) spectrophotometer and fluorescence spectrum. Circular dichroism spectroscopy is applied to characterize the conformational transitions of zein. When zein is combined with other components, the changes in the secondary structure of zein NPs can be observed (Liu et al., 2022b; Liu et al., 2022c). FTIR spectrum of zein shows peaks at around 3,310 cm⁻¹ (O-H stretching), 1,660 cm⁻¹ (amide I band) and 1,535 cm⁻¹ (amide II band) (Sessa et al., 2013; Yin et al., 2015). Shift of these peaks provide evidence for the formation of hydrogen bonds, hydrophobic interactions, or crosslinking among different components of zein NPs. Fluorescence spectrum also can be applied to investigate intermolecular interactions. Zein contains a high proportion of tyrosine residues, which has a typical emission maximum around 304 nm after being excited at 280 nm. The interaction between zein and other compounds may lead to the micro-environmental alteration, which can be reflected through the fluorescence spectrum (Wei et al., 2019; Zhu et al., 2021).

In some cases, molecular dynamics (MD) simulation and molecular docking are used to analyze the intermolecular interactions. For example, the binding mechanism of zein with epigallocatechin-3-gallate (EGCG) was investigated via multi-spectroscopy and MD simulation (Liu et al., 2021a). MD simulation clarified that the van der Waals and electrostatic interactions were involved in binding of EGCG to zein, and the EGCG preferred to bind to the pocket of zein generated by residues Y171, Q174, L176 and L205. In another study, molecular docking was used to predict the precise binding sites and binding modes of ferulic acid to zein (Wang et al., 2022c), and the results revealed the binding stability under alkaline condition was stronger than that under acidic and neutral conditions.

Zein molecule has a very special bricklike structure that provides sufficient space for drug entrapment (Geraghty et al., 1981; Zhang et al., 2016). The loading capacity of zein NPs is often evaluated by DL and EE, calculated using Eqs 1, 2, respectively: D L = A m o u n t   o f   d r u g   l o a d e d   i n   p a r t i c l e s A m o u n t   o f   z e i n   N P s × 100 % (1) E E = A m o u n t   o f   d r u g   l o a d e d   i n   p a r t i c l e s A m o u n t   o f   d r u g   i n i t i a l   i n p u t × 100 % (2)

The drug content was measured by a high-performance liquid chromatography system or UV-spectrophotometer. Due to the high hydrophobicity of zein molecule, zein NPs possess high loading capacity for hydrophobic drugs than hydrophilic and amphiphilic drugs (Karthikeyan et al., 2014).

The solid form of drug in zein NPs can be further analyzed by X-ray diffractometer (XRD) and differential scanning calorimeter (DSC). The XRD pattern of zein shows peaks at the diffraction angles of 2θ = 9.2^o and 20^o, where 2θ = 9.2^o arises mainly from the α-helix structure of zein, whereas 2θ = 20^o is thought to come from lateral α-helix packing (Chen and Subirade, 2009). Drug can exist in diverse solid forms, such as polymorphs, pseudo-polymorphs, co-crystals and amorphous solids (Liu et al., 2021b). In most cases, the drug is dispersed in zein NPs in an amorphous state. As an interesting study, drug nanocrystals (NC) were successfully incorporated into zein particles by combining the SAS process with BUDP in our previous work (Liu et al., 2016a; Wang et al., 2017), where the co-precipitation of 10-hydroxycamptothecin (HCPT) and zein prepared using the SAS process was dispersed into ethanol-water as the dialysis solution for BUDP. As shown in Figure 5, the obtained HCPT NC-zein particles were microspheres with a mean particle size around 1.0 µm, DL of 5.98% and EE of 95.68%.

Storage stability is vital to the shelf-lives of pharmaceutical products. Zein NPs usually have a smooth surface without pores, strong hydrophobicity and low permeability to oxygen, which can effectively prevent the contact of drug with external environment (e.g., light, oxygen, heat and humidity), thereby improving the storage stability. For example, the storage stability of 7,8-dihydroxyflavone (7,8-DHF) was strengthened greatly after encapsulation in zein NPs, especially with the addition of glycosylated lactoferrin (Chen et al., 2022). Similarly, zein NPs has been used to improving the storage stability of cannabidiol (Wang et al., 2022a), resveratrol (Khan et al., 2021), curcumin (Yuan et al., 2020) and others.

The oral route is the preferred way for drug administration due to its advantages on patient compliance, safety, cost-effectiveness and feasibility. However, the complex nature of the gastrointestinal tract offers numerous obstacles, e.g., low pH of the gastric environment, presence of enzymes, intestinal flora and first-pass effect (Alqahtani et al., 2021). Due to their low oral bioavailability, many drugs have to formulate and administer as injections.

Zein NPs have already demonstrated an important potential to increase the oral bioavailability of both small and large molecules. Firstly, zein matrix is an excellent barrier for the protection of drugs (e.g., insulin (Reboredo et al., 2022), antigen (Hurtado-López and Murdan, 2006) and peptide (Khalid Danish et al., 2022)) in the gastrointestinal tract due to its water insoluble in a low pH environment (pH < 11.0). Secondly, the ability of zein in solid dispersion system can enhance drug dissolution rate of poorly water-soluble drug (Nguyen et al., 2017). In most cases, the drug is dispersed in zein NPs in an amorphous state. For the amorphous state, the drug molecules are not organized in a definite lattice pattern, and there is no crystal lattice energy to disrupt during dissolution, resulting in the highest level of dissolution (Liu et al., 2020). Thirdly, zein NPs possess a good bioadhesive nature with biological surfaces through hydrophobic interactions and hydrogen bonding. The mucoadhesive properties of zein NPs could lead to a slow drug release close to the cell membrane and a long retention time in the upper gastrointestinal tract for high drug absorption (Patel et al., 2010a). Moreover, their trans-mucus permeability and transmembrane transport can be further enhanced through the surface coating and modification (Bao et al., 2021).

Drug release is a significant factor of nanoparticles for successful drug delivery. The drug release behavior is mainly dependent on drug solubility, diffusion, and biodegradation. It can be evaluated by a number of kinetic models, including statistical methods (e.g., exploratory data analysis method, repeated measures design, multivariate approach and multivariate analysis of variance), model dependent methods (e.g., zero order, first order, Higuchi, Korsmeyer-Peppas model, Hixson Crowell, Baker-Lonsdale model and Weibull model) and other model independent methods (Costa and Lobo, 2001; Dash et al., 2010).

Zein has been demonstrated to be an effective excipient in extensive studies for controlled drug delivery due to its hydrophobicity, biodegradability and biocompatibility. Zein is more suitable for sustained release of hydrophobic drugs than hydrophilic and amphiphilic drugs, due to the better binding affinity and greater interactions (hydrophobic, electrostatic, and hydrogen bonding) of hydrophobic drug with zein (Karthikeyan et al., 2014). Recently, the ability of zein in the controlled release of poorly water-soluble drugs has been reviewed (Tran et al., 2019). Moreover, for the controlled release of drug in specific environment, pH-responsive zein NPs have be prepared by coating with tannic acid (Liang et al., 2018), PDA (Zha et al., 2020), sodium caseinate and poly ethylene imine (Pourhossein et al., 2020). Also, there are few studies focus on the controlled release of zein NPs by other specific stimuli (Marín et al., 2018; Zhong et al., 2022).

Targeted drug delivery that refers to predominant drug accumulation within a target zone, is an effective approach to decreasing drug toxicity and enhancing therapeutic effects. The targeting of drugs can be viewed on two levels: organ targeting and cellular targeting (Koo et al., 2005; Manzari et al., 2021). The organ targeting is mainly dependent on the particle size, morphology and material properties of the carrier employed. Zein NPs have capabilities of size-dependent organ targeting through leaky tumor capillary fenestrations into a tumor vasculature according to the enhanced permeability and retention (EPR) effect (Maeda, 2015). For example, the organ targeting of 5-fluorouracil loaded zein NPs was studied (Lai and Guo, 2011): the targeting efficiency of drug in liver increased from 22.67% to 54.00% after the encapsulation, and the relative uptake rate (zein NPs/plain drug solution) was in the order of liver (2.79) > spleen (1.94) > plasma (1.09) > lung (0.70) > heart (0.34) > kidney (0.24), indicating that zein NPs was specific to the liver tissue. The development of magnetic NPs can further enhance the delivery and controlled release of a drug to the targeting organ, especially the fabrication of temperature and pH-responsive magnetic NPs (Marín et al., 2018; Pang et al., 2018).

In spite of the preferred accumulation of zein NPs through organ targeting mechanisms, there is some degree of non-specific action in this method. It is of critical importance to develop the cellular targeting zein NPs through a more specific interaction at a molecular level between the carrier and the targeted cell (Byrne et al., 2008). There are abundant hydroxyl, amino and carboxyl groups on the side chains of zein molecular, which offer the possibility of conjugation and modification of zein molecules with targeting ligands or molecules (Allen, 2002; Xu et al., 2013). For example, hyaluronic acid (HA) cross-linked zein NPs were prepared (Seok et al., 2018; Zhang et al., 2020). Given the high targetability of HA to CD44 expressing cancer cells, the obtained HA-zein NPs could enter tumor cells effectively via a CD44 mediated endocytosis thereby enhancing the PK/PD. In our previous work (Liu et al., 2017), zein was decorated with folic acid (FA) for targeted delivery of HCPT. As shown in Figure 7, FA conjugation could not only facilitate the formation of small nanoparticles with good dispersity and stability, sustain the drug release, but more importantly improve the cellular up-take in folate receptor positive cells thereby enhancing the antitumor activity.