Using Shellac to Build Nanoparticles for Delivering mRNA

Protein synthesis via messenger RNA (mRNA) translation highlights the central role of mRNA in the transfer of genetic information, positioning it as a promising platform for disease treatment and prevention [1]. The success of mRNA‐based COVID‐19 vaccines developed by Pfizer and Moderna further demonstrates the translational potential of mRNA technology in vaccine development and healthcare [2]. However, mRNA is susceptible to RNase‐mediated degradation, while its large molecular weight and highly anionic nature pose significant barriers to efficient intracellular delivery. These challenges have inspired the development of a wide range of nanoparticle (NP)‐based delivery platforms. Most NP platforms involve the incorporation of cationic, amine‐containing components (e.g., ionizable lipids [3], polyethyleneimine (PEI) [4], and poly‐l‐lysine (PLys) [5]) for enhancing mRNA delivery efficacy [3], mainly by promoting electrostatic interactions with mRNA and facilitating endosomal escape upon endocytosis. However, the molecular weight of cationic components can influence delivery efficacy and cytotoxicity [6]. High‐molecular‐weight cationic polymers typically offer good transfection efficiency, but often exhibit cytotoxicity and poor biocompatibility. In contrast, low‐molecular‐weight polymers show lower cytotoxicity, but typically lack sufficient transfection capability because of weaker mRNA binding or cellular uptake. For example, PEI with a high molecular weight (e.g., 25 kDa) can induce modest apoptosis, strong necrosis, and DNA damage [7, 8], whereas PEI with a low molecular weight (e.g., 1.8 kDa) exhibits substantially lower cytotoxicity but a compromised transfection efficiency (< 5%) [9, 10]. Therefore, NPs need to be appropriately engineered to balance the trade‐off between transfection efficiency and toxicity. A common strategy to improve the transfection efficiency of low‐molecular‐weight PEIs involves conjugating them with highly biocompatible or hydrophobic moieties to enhance subsequent mRNA complexation and facilitate membrane interactions [11, 12]. For example, a cyclodextrin‐modified 2 kDa PEI platform demonstrated mRNA delivery efficacy. Specifically, the modification reduced toxicity and resulted in a higher transfection efficiency [13]. Similarly, a vitamin E succinate‐modified PEI copolymer was developed to deliver mRNA vaccines, eliciting strong humoral and cellular immune responses following intramuscular injection [9]. However, despite advances, current approaches are limited by the need for sophisticated synthetic procedures. Thus, developing a simple and effective method for preparing mRNA NPs that can achieve both high transfection efficiency and low cytotoxicity is still lacking.

Shellac (SL), a natural resin excreted by the insect Laccifer lacca, is generally recognized as safe by the U.S. Food and Drug Administration [14, 15]. It is composed of aleuritic acid and cyclic terpene acids connected by ester bonds, which act as the hydrophobic and hydrophilic segments, respectively [16, 17]. Owing to its amphiphilic property [18], SL is expected to facilitate the formation of compact NPs with mRNA and low‐molecular‐weight cationic polymers, improving NP stability and facilitating membrane interactions, thereby increasing transfection efficiency. We hypothesized that the versatile supramolecular binding affinity of SL, its neutral‐to‐negative charge that enables cationic shielding, and its ability to protect labile cargo from enzymatic degradation may provide a generalizable pathway for mRNA assembly and delivery.

In the present study, we report a facile SL‐mediated approach to assemble and stabilize mRNA within NPs under ambient conditions. In this approach, mRNA is first complexed with select cationic molecules (CMs), encompassing low‐molecular‐weight polymers (e.g., PEI), polysaccharides (e.g., chitosan), lipids (e.g., SM‐102), and PLys, followed by capping of the complexes with SL to yield mRNA‐loaded‐cationic molecule‐shellac NPs (i.e., mRNA‐CM‐SL NPs; Scheme 1). The electrostatic interactions conferred by the cationic motifs, combined with the versatile binding affinity of SL, afford over 90% mRNA encapsulation efficiency and mRNA transfection both in vitro and in mice. Notably, the pH‐dependent solubility of SL confers stability to the mRNA‐CM‐SL NPs in gastric fluids, allowing for deposition throughout the gastrointestinal (GI) tract and gene editing in the colon post‐oral gavage. Furthermore, cation shielding by SL preserves the viability of cells treated with these NPs. This study is expected to underpin the rational design of therapeutic delivery platforms using naturally occurring building blocks and, in particular, provides a versatile platform for mRNA delivery via both intravenous and oral administration, thus expanding its potential clinical applications.

PEI‐SL NPs were assembled by dissolving SL in ethanol and subsequent mixing with low‐molecular‐weight PEI (e.g., M_w = 800 Da; PEI₈₀₀) in water. The color transition of the original yellow solution to a pink suspension upon mixing, accompanied by a shift in the UV–vis absorbance band from 279 to 308 nm, indicates potential interactions between SL and PEI (Figure 1a). Furthermore, the absence of the band at 1599 cm⁻¹ (ascribed to amide II in PEI) and the emergence of a sharp peak at 1639 cm⁻¹ (ascribed to amide linkage (─CONH─)) in the Fourier transform infrared spectrum of PEI‐SL NPs indicate interactions between the carboxyl groups in SL and amine groups in PEI, suggesting the complexation of SL and PEI (Figure 1b).

Three types of mRNA complexes were prepared: mRNA–PEI, mRNA–SL, and mRNA complexed with PEI and then capping with SL, to explore their potential for biomedical applications. Among the three complexes formed, the mRNA–PEI complexes were the largest and featured the highest polydispersity (i.e., size = 825 nm ± 33 nm, polydispersity index (PDI) = 0.40) (Figure S1). The average size of the mRNA–SL complexes was 187 nm ± 3 nm, and the PDI was 0.15. Complexing mRNA with both PEI and SL resulted in the formation of solid NPs (i.e., mRNA‐PEI‐SL NPs) with an average size of 372 ± 10 nm and PDI of 0.07. The reduced size and lower PDI suggest that the incorporation of SL in the mRNA–PEI complex facilitates the assembly of the NPs and improves their monodispersity. The incorporation of SL also led to a reduction in zeta‐potential (i.e., from 31 to 7 mV when the mass ratio of PEI‐to‐mRNA was 10), indicating cation shielding by SL (Figure S2). Furthermore, from the encapsulation efficiency results, we found that mRNA–SL complexes exhibited a relatively low encapsulation (15%), likely due to the combined effect of electrostatic repulsion, hydrogen bonding, and hydrophobic interactions. In contrast, the mRNA–PEI complexes and mRNA‐PEI‐SL NPs showed a higher encapsulation efficiency (98%; Figure S3), which was likely due to the strong electrostatic interactions between PEI and mRNA. Collectively, these results demonstrate that complexation with cationic moieties facilitates mRNA encapsulation, while the incorporation of SL enables the formation of stable, mRNA‐loaded supramolecular NPs. To further explore the versatility of the mRNA‐CM‐SL NPs, other CMs (i.e., chitosan, PLys, and SM‐102) were assembled with mRNA to form mRNA–CM complexes. These complexes displayed a broad size distribution with an average diameter of >200 nm (Figure S4). Upon incorporation of SL, the resultant mRNA‐CM‐SL NPs showed a relatively narrow size distribution and less positive zeta‐potential, indicating cationic shielding from SL (Figure 1c,d; Figures S5–S7). Similarly, the SL‐mediated NPs also showed high mRNA encapsulation efficiency (Figure 1e).

To evaluate the stability of the NPs, mRNA‐PEI‐SL NPs were incubated in RNase‐free water and Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS). As shown in Figure S8, the mRNA‐PEI‐SL NPs remained relatively stable in biological media for 1 week, which may be attributed to the formation of a protein corona on the NP surface. The mRNA release experiments revealed that less than 20% of mRNA was released from the mRNA‐PEI‐SL NPs in Dulbecco's phosphate‐buffered saline (DPBS) within 24 h, whereas 67% of mRNA was released in biological media, which was more than twice that released from the mRNA–PEI complexes (Figure 1f; Figure S9). To examine the interactions for the stability of the mRNA‐PEI‐SL NPs, the size of the mRNA‐PEI‐SL NPs was measured upon incubation with urea (hydrogen bond competitor), Tween 20 (hydrophobic competitor), or sodium chloride (NaCl; electrostatic competitor). Exposure to either urea or NaCl led to larger particles, indicating particle aggregation. However, the size of the NPs rapidly decreased to ∼10 nm within 10 min of incubation with Tween 20 (Figure S10), suggesting that hydrophobic interactions play a role in stabilizing the mRNA‐PEI‐SL NPs. Collectively, these results suggest that multiple interactions stabilize the mRNA‐PEI‐SL NPs.

We next assessed the stability of the mRNA‐PEI‐SL NPs under GI‐mimetic conditions. The mRNA‐PEI‐SL NPs remained stable in simulated gastric fluid (SGF), as deduced by negligible changes in their size and morphology (Figure 1g; Figure S11). This finding suggests the potential of the mRNA‐PEI‐SL NPs to protect encapsulated bioactive payloads from degradation by gastric fluids and enzymes during transit through the GI tract. In contrast, distinct changes in NP size were observed in simulated intestinal fluid (SIF), likely due to aggregation caused by electrostatic interactions between deprotonated SL and PEI. The mucus‐penetration ability of the NPs was evaluated using an agarose gel [19]. The results demonstrated that approximately 20% of the NPs penetrated through the mucus layer (Figure 1h). Dynamic light scattering revealed an increase in NP size after incubation with mucin, indicating interactions between the NPs and mucin (Figure S12).

To explore the potential of the mRNA‐PEI‐SL NPs for biomedical applications, we first conducted cellular association studies. As shown in Figure 2a, the incorporation of SL enhanced association with HEK293T cells. For instance, in the absence of SL, the mRNA–PEI complexes displayed <20% cell association at 6 h post‐incubation, even when the PEI‐to‐mRNA mass ratio (W_PEI/W_mRNA) increased to 20:1. In contrast, incorporating SL enabled significantly higher cellular association, achieving approximately 70% at 6 h. In addition, the mRNA‐PEI‐SL NPs exhibited time‐dependent cellular association behavior; negligible cellular association was observed at 1 h, whereas nearly 100% association was achieved at 24 h (Figure S13). To investigate the mechanism underlying the cellular uptake of mRNA‐PEI‐SL NPs, different types of endocytic inhibitors, including pitstop2, EIPA, filipin from Streptomyces filipinensis, and cytochalasin D, were incubated with cells prior to NP treatment. As shown in Figure S14, incubation with EIPA, filipin, and cytochalasin D reduced cellular uptake to approximately 77%, 78%, and 61%, respectively, suggesting that the cellular entry of these NPs was mediated by a combination of micropinocytosis, caveolae‐mediated endocytosis, and clathrin‐mediated endocytosis. The cytotoxicity of the mRNA‐PEI‐SL NPs was then evaluated as a function of PEI‐to‐mRNA mass ratio. As shown in Figure 2b, the mRNA‐PEI‐SL NPs displayed negligible cytotoxicity to HEK293T cells following 24 or 48 h incubation. In contrast, the viabilities of the cells incubated with PEI_0.8k and PEI_25k were 100% and 50%, respectively, at 20 µg mL⁻¹ (Figure S15). To investigate the potential of the NPs for oral delivery, the viability of Caco‐2 cells (a human intestinal cell line, commonly used for oral drug delivery models [20]) and RAW264.7 cells (a murine macrophage cell line [21]) after treatment with mRNA‐PEI‐SL NPs were studied. For both cell lines, cell viabilities remained higher than 80% at a NP concentration of 80 µg mL⁻¹, suggesting the compatibility of the NPs for intracellular delivery (Figure S16). We next examined the secretion of tumor necrosis factor‐α (TNF‐α) in blood following intravenous administration in mice. No significant increase in TNF‐α was detected in the mRNA‐PEI‐SL NPs group compared with the DPBS control (Figure S17), indicating that the NPs exhibit a low inflammatory response. As most NPs are likely to undergo the endo/lysosomal pathway after cell internalization [22], we next investigated the colocalization of the mRNA‐PEI‐SL NPs with endo/lysosomal compartments. The endo/lysosomes were stained with LysoTracker Green, while mRNA was labeled with cyanine 5 (red). The minimal colocalization of green and red fluorescence signals observed in the confocal laser scanning microscopy images, along with a low Pearson's correlation coefficient value of 0.425, indicated some degree of endosomal escape of the mRNA‐PEI‐SL NPs (Figure 2c; Figure S18).

The transfection efficiency of the mRNA‐PEI‐SL NPs was explored using mRNA that encodes a reporter protein (i.e., mCherry). Similar to the pattern observed for cellular association, the mRNA–PEI complexes did not significantly interact with cells, resulting in a negligible transfection efficiency, regardless of the PEI‐to‐mRNA ratio. In contrast, the incorporation of SL enabled successful mRNA transfection, as indicated by both flow cytometry and confocal laser scanning microscopy images (Figure 2d,e; Figure S19). This is likely due to the increased hydrophobicity imparted by SL that further stabilized the NPs [12, 23] and facilitated cellular association (Figure 2a). Furthermore, the successful intracellular expression of mRNA suggests some degree of integrity of mRNA released from mRNA‐PEI‐SL NPs. Notably, the level of mRNA transfection was modulated by varying the PEI‐to‐mRNA mass ratio: increasing this ratio up to 10:1 gradually increased the transfection efficiency and mCherry mean fluorescence intensity (MFI; i.e., the level of mCherry expression within the cells) from approximately 6% to 41% and from 3454 to 19585, respectively. Therefore, a PEI‐to‐mRNA mass ratio of 10 was employed for further experiments.

The transfection efficiency of the NPs after pre‐treatments with SGF and SIF was evaluated using Caco‐2 cells. As seen in Figure S20, the mRNA‐PEI‐SL NPs retained their mRNA transfection ability following exposure to simulated GI fluids, resulting in a 2.7‐fold higher MFI relative to untreated cells. Collectively, these results demonstrate the potential of the lead formulation (i.e., mRNA‐PEI‐SL NPs with a PEI‐to‐mRNA mass ratio of 10:1) for the oral delivery of mRNA. To improve reproducibility, reporting, and re‐analysis, this study conforms to the Minimum Information Reporting in Bio–Nano Experimental Literature (MIRIBEL) standard [24], and a companion checklist is provided in the Supporting Information.

Furthermore, SL exhibits a pK_a of approximately 5.6–7.0, which confers stability in an acidic gastric environment and is dissolved in environments with a pH greater than 7 (e.g., the small intestine and colon) [16]. These properties make SL a promising candidate for protecting mRNA from degradation in the GI tract following oral administration, as it can form a protective layer that shields bioactive cargo from degradation in gastric fluids [25, 26]. Given the previously demonstrated stability of the mRNA‐PEI‐SL NPs in SGF, mucin penetration ability, and effective transfection in vitro, we further examined the potential of the mRNA‐PEI‐SL NPs for oral delivery using mouse models (Figure 3a). For the in vivo oral delivery studies, the mRNA‐PEI‐SL NPs were labeled with Alexa Fluor 647 (AF647) to enable visualization of their biodistribution. As shown in Figure 3b, the mRNA‐PEI‐SL NPs were distributed throughout the GI tract, including the small intestine and colon, 6 h post‐oral administration. Furthermore, the presence of AF647 signal in the liver suggests that a trace amount of NPs may have crossed the epithelial barrier and entered systemic circulation. To assess the ability of the mRNA‐PEI‐SL NPs in mediating oral mRNA delivery in vivo, the genetically engineered tdTomato (tdTom) reporter mice and Cre recombinase (Cre) mRNA were employed. mRNA‐PEI‐SL NPs were injected on two consecutive days to enable cumulative protein expression. In these mice, tdTom expression was initially silenced by a loxP‐flanked stop cassette (Figure 3c). Successful transfection and expression of Cre mRNA would excise the cassette, thereby activating tdTom fluorescence. The tdTom signal observed in the colon implied that the mRNA‐PEI‐SL NPs preserved the activity of mRNA despite exposure to the harsh gastrointestinal environment, whereas mRNA–PEI complexes and SM‐102 lipid NPs (LNPs) led to minimal or no detectable tdTom activation under the same conditions (Figure 3c).

To further explore the in vivo transfection efficacy of the mRNA‐PEI‐SL NPs, a different administration route was examined and mFlame was encapsulated within the mRNA‐PEI‐SL NPs. mFlame is an engineered mRNA encoding brighter red fluorescence proteins (see Experimental Section). The mFlame‐encapsulated mRNA‐PEI‐SL NPs were injected intravenously into C57BL/6J mice. Major organs, including the liver, spleen, kidney, lung, and heart, were harvested after 24 h and imaged using an In Vivo Imaging System (IVIS) (Figure 3d). Unlike SM‐102 LNPs that typically enabled pronounced mRNA expression in the liver, mRNA‐PEI‐SL NPs mediated mRNA transfection primarily in the heart and lung (Figure 3e,f). Accordingly, the fluorescence intensity in these organs were 4.3‐ and 8.5‐fold higher, respectively, compared to those in the SM‐102 LNP‐treated mice. This variation in mRNA expression tropism is likely attributed to the protein corona formed on NPs, which is known to influence in vivo biodistribution and organ‐specific accumulation [27]. Although additional studies are required to determine the exact mechanism, this accumulation suggests potential of the SL NP platform for heart‐ and lung‐tropic mRNA delivery.

We developed a versatile, modular SL‐enabled NP platform for mRNA delivery, including oral delivery. The versatile binding affinity of SL, combined with its neutral‐to‐negative charge, facilitates the assembly of a compact structure and shields the charge of the cationic molecules assembled within the NPs. The SL‐based NPs exhibited high mRNA encapsulation efficiency (>90%) and demonstrated transfection efficacy both in vitro and in vivo. Notably, this platform protected mRNA from degradation in the harsh gastrointestinal environment, enabling transfection in the colon after oral administration. This work presents a promising alternative mRNA delivery and gene editing platform and provides a viable strategy for the oral administration of nucleic acid therapeutics to potentially treat intestinal and colonic diseases.

F.C. is a shareholder of Messenger Bio Pty Ltd. The remaining authors declare no conflicts of interest.