Versatility of LNPs across different administration routes for targeted RNA delivery

Abstract

The advent of lipid nanoparticle (LNP) technology has marked a significant breakthrough in the field of drug delivery, offering unprecedented opportunities in gene therapy, vaccine delivery, and personalized medicine. The use and effectiveness of LNPs greatly depend on their optimization, often tailored to specific routes of administration. Different routes can significantly influence organ distribution, expression kinetics and therapeutic outcomes of LNPs, with the choice of the route dependent on LNP properties, target sites, and therapeutic indications. In this review, we summarize recent studies that highlight the versatility of LNPs, through optimization for delivery across different routes of administration, while scrutinizing the route-dependent formulation strategies. We then outline key challenges facing LNP optimization for site-specific RNA administration and propose future prospects for employing appropriate administration routes to develop LNP-based RNA medicines.

---

Min Qiu

Min Qiu, PhD, is a Principal Investigator at the Human Phenome Institute, Fudan University, and a member of the mRNA Innovation and Translation Center. He earned his doctoral degree from Soochow University under the supervision of Professors Zhiyuan Zhong and Chao Deng, and subsequently completed his postdoctoral training in the laboratory of Professor Qiaobing Xu in the Department of Biomedical Engineering at Tufts University. In 2021, Dr Qiu joined the Human Phenome Institute, where he leads an interdisciplinary research group dedicated to the development of tissue- and cell-targeted lipid nanoparticles for applications in vaccines, gene editing, and immuno- and cell-based therapies.

---

1. Introduction

Lipid nanoparticles (LNPs) have emerged as a cornerstone technology in modern pharmaceuticals, offering a versatile platform for the delivery of a wide range of therapeutic agents, including RNA, DNA, proteins, and small molecules.^1,2 This system represents a significant breakthrough in drug delivery, especially for RNA-based therapeutics, as it protects the therapeutic payload (like mRNA or siRNA) from degradation, facilitates cellular uptake, and ensures targeted delivery.³ The design and optimization of LNPs are crucial, as they directly influence efficacy, safety, and overall therapeutic outcomes.⁴

The LNP/RNA system can be administered through various routes, including systemic, local, oral, or inhalable delivery. Due to the unique barriers associated with each route of administration, the physicochemical properties of LNPs must be carefully optimized to enhance efficacy while minimizing toxicity.^5–7 For instance, the oral administration route has been shown to have the lowest bioavailability of drugs among all routes, although it is the most convenient and least invasive route.⁸ This is due to its characteristic gastrointestinal effect, as well as the first-pass effect in the liver. The super-protective blood–brain barrier which mainly protects the brain from foreign materials is unique to the challenges of systemic delivery to the brain.⁹ Other routes are also characterized by their own unique physiological barriers that greatly influence the therapeutic efficacy of the drug cargo. The barriers to inhalation mRNA delivery include anatomical complexity and physiological clearance, enzymatic degradation, and fluid dynamics.⁶ Hence, it is important to understand the properties of each route, so that LNP formulations are designed to achieve maximum bioavailability, minimum toxicity, and thus maximum efficacy.

Before delving into specific administration routes, it is crucial to understand the general principles of LNP design.¹⁰ LNPs typically comprise of a lipid bilayer enclosing the therapeutic payload, and mainly consist of four components: ionizable lipids, neutral phospholipids, cholesterol, and PEGylated lipids.^4,11 However, composition can vary depending on the targeted route, that is, some LNPs can be five-, three-, or two-component systems.^7,12 Ionizable lipids are typically the core components of LNPs, and exhibit a positive charge at acidic pH, thus allowing nucleic acids to aggregate into LNPs during the preparation process.¹³ After cellular internalization, the ionizable lipids become protonated in the acidic endosomes and interact with anionic endophospholipids to form conical ion pairs incompatible with the bilayer, resulting in membrane fusion, endosomal escape, and finally, cargo release.⁷ Phospholipids, with their strong bilayer formation characteristics and high phase transition temperature, and cholesterol, with its excellent membrane fusion ability, ensure the structural stability of LNPs, regulate transfection efficiency, and promote endosomal escape and intracellular uptake of mRNA.^4,5 PEG-lipids play important roles in improving hydrophilicity, determining LNP size, preventing LNP aggregation to maintain stability and rapid removal, and improving the circulation half-life of LNPs in the blood.^14–16 In parallel, emerging insights into structure–activity relationships—particularly how variations in lipid head groups, tail length, and linker chemistry influence endosomal escape, immune activation, and cellular uptake—have provided valuable guidance for rational LNP design.¹⁷ For example, shorter lipid tails may enhance the potency of LNPs encapsulating larger mRNAs, while longer lipid tails can improve delivery efficacy for smaller mRNAs.¹⁷ Additionally, unsaturated tails may facilitate membrane fusion and promote endosomal escape.¹⁸ Each of the LNP components can be designed and optimized to meet the challenges provided by each route of administration in order to achieve maximum bioavailability in target sites, and thus maximum therapeutic efficacy.^16,19

Key factors in LNP design include lipid composition, particle size, surface charge, and PEGylation, with each playing a crucial role in determining absorption, adsorption, biodistribution, cellular uptake, and release kinetics.⁵ Each of these factors can be optimized to address the challenges posed by the selected route of administration.^10,19,20 For example, in intravenous administration where plasma proteins in the bloodstream play a key role in the biodistribution of the therapeutic cargo, adsorption efficiency is a key process.¹⁰ The LNP surface may be modified to achieve active targeting, such as by modifying the surface with antibodies, ligands, or other adjuvants to achieve specific targeting.¹² That is, in systemic administration, advanced strategies in LNP design include the incorporation of modified PEGylation to extend circulation time, the use and modification of ionizable lipids for improved endosomal escape, and the incorporation of targeting ligands for cell-specific delivery.²¹ These strategies are often employed complementarily to achieve the desired biodistribution and therapeutic outcomes.

Technological advancements have played a pivotal role in improving LNP production. Among these, microfluidic mixing has become the standard for laboratory- and clinical-scale synthesis due to its reproducibility and ability to precisely control nanoparticle characteristics. This method relies on rapid mixing of an ethanol–lipid solution with an aqueous RNA-containing buffer, yielding monodispersed LNPs with high encapsulation efficiency and tunable sizes.^22,23 Furthermore, scalable microfluidic systems using T-shaped mixers and herringbone structures have enabled continuous production at industrial volumes.²⁴ Alternative approaches like membrane micromixing—where ethanol is diffused through ceramic membranes into an aqueous phase—offer another scalable, low-shear method to maintain consistency in size and loading characteristics, even at large volumes.²⁵ Complementary characterization techniques, including dynamic light scattering (DLS), zeta potential analysis, cryo-electron microscopy (cryo-EM), and RNA-specific fluorescence assays like RiboGreen, are used to assess particle size, surface charge, morphology, and encapsulation efficiency.²⁶ These tools are essential not only for ensuring formulation reproducibility but also for predicting biological performance. Smaller particles with narrow polydispersity indices typically exhibit enhanced tissue penetration and lymphatic drainage, while surface charge influences interactions with cell membranes and extracellular matrices. Cryo-EM provides high-resolution structural validation, enabling researchers to visualize lamellarity, core architecture, and lipid layer integrity, which can correlate with stability and release kinetics. Furthermore, accurate determination of encapsulation efficiency using RiboGreen or SYBR-based dyes ensures that the therapeutic RNA is sufficiently protected within the nanoparticle, reducing susceptibility to nucleases during systemic circulation. Together, these formulation and design insights as well as the analytical techniques form the foundation for a quality-by-design approach, which is critical when adapting LNPs to specific administration routes. Thus, accordingly, rational LNP design must consider route-specific challenges to ensure sufficient bioavailability and therapeutic impact.¹⁰

While there have been many reviews on LNP optimization for RNA delivery in relevant literature, few published reviews have focused primarily on the versatility of LNPs across the different delivery routes. The versatility of LNPs lies in their ability to be systematically modified to achieve precise biodistribution profiles, enabling therapeutic delivery to specific organs or tissues. This adaptability underscores the importance of tailoring LNP composition to meet the demands of different therapeutic targets while maintaining safety and efficacy. This review article elaborates on LNP optimization specific to different routes of administration for RNA delivery (Scheme 1), highlighting the key modifications of physicochemical properties, challenges to be addressed, and examples of case studies for each route. The article then details a comparative analysis of the different optimization strategies for various routes and their future perspectives.

Scheme 1 Versatile LNPs across various routes of administration. Created with https://BioRender.com.

2. LNP optimization across different delivery routes

2.1. Intravenous (i.v.) delivery

Intravenous injection is one of the most frequent routes of administration.²⁷ LNPs designed for intravenous administration must be optimized for stability in the bloodstream, minimal clearance by the mononuclear phagocyte system, and efficient targeting of specific tissues or cells. Particle size and surface characteristics, particularly PEGylation, are critical factors in achieving these objectives.^10,16 The ideal size range for i.v.-administered LNPs is typically between 50 and 200 nm, balancing the need for prolonged circulation with efficient tissue penetration and cellular uptake.^10,19 In i.v. delivery, adjusting the proportions of the lipids and the types of phospholipids and sterols in LNPs enhances efficacy, especially for extra-hepatic RNA delivery.

2.1.1 Hepatic delivery. The liver is the most reported organ for LNP-mediated RNA delivery. Hepatic targeting leverages the liver's natural affinity for lipid-based carriers.²⁸ A study by Kim et al. is a case in point.²⁹ In this study, LNPs were modified for systemic delivery and designed to specifically target the liver by modifying their composition. The liver-targeting mechanism was achieved using a 14-C lipid of DMG-PEG_2k, which plays a crucial role in stabilizing the particle in circulation and facilitating selective interaction with apolipoprotein E (ApoE). The ApoE protein binds to the surface of LNPs and interacts with low-density lipoprotein receptors (LDLR) on hepatocytes, ensuring targeted delivery to liver cells. This modification enhanced the biodistribution of the LNPs, achieving systemic delivery to the liver without significant accumulation in other organs. By incorporating DMG-PEG_2k, the researchers reduced rapid clearance by the reticuloendothelial system (RES), allowing the particles to remain in circulation long enough to reach hepatocytes effectively. Proof of the critical role of this modification was furnished by comparative studies conducted by researchers and by literature reviews.^10,16 LNP formulations lacking the DMG-PEG_2k component exhibited significantly higher clearance rates and poor hepatic accumulation. Such outcomes rendered the unmodified LNPs unsuitable for systemic administration, emphasizing the necessity of the DMG-PEG_2k modification.

Optimization by changing the ratio of lipid components in the LNP could potentially confer different therapeutic efficacies in the liver. Han et al. created an LNP loaded with Cas9 mRNA and mouse AT-targeted sgRNA to edit the AT gene and inhibit the activity of antithrombin.³⁰ While the typical molar ratio of the components of the LNP formulation is 50 : 38.5 : 10 : 1.5, this LNP carrier was composed of ionizable lipid, phospholipid, cholesterol, and PEG lipid at a molar ratio of 26.5 : 20 : 52 : 1.5. Intravenous administration of this LNP resulted in accumulation in the liver and long-term therapeutic effects on haemophilia A and B in mouse models, without causing hepatotoxicity or off-target toxicity. These results demonstrated the safety and effectiveness of this LNP for haemophilia treatment.

2.1.2 Non-hepatic delivery. In contrast, extra-hepatic tissue delivery requires innovative strategies to avoid the natural hepatic sequestration of LNPs.³¹ This is typically achieved by altering the surface characteristics of LNPs, including incorporating non-hepatic tissue-targeting ligands, adding a fifth component, discovering new ionizable lipids, and optimization of the LNP formulation ratios.

Modification of LNPs with targeting ligands is the most straightforward method for non-hepatic RNA delivery. For example, Huang et al. developed tumor-targeted LNPs by integrating a bispecific T-cell engaging (BiTE) mechanism, modifying lipid composition to achieve selective tumor localization while minimizing hepatic uptake.³² The LNP they used was composed of DMG-PEG, DSPC, cholesterol, and a different ionizable cationic lipid, IC8, with a size of 118 nm and a surface charge of 10 mV. After intravenous injection, the BiTE mRNA-loaded LNP primarily accumulated in the liver and spleen, inducing the activation of T cells specific to B7H3-positive tumor cells. Beyond general strategies for reducing hepatic uptake, other approaches for extra-hepatic delivery via intravenous routes involve engineering LNPs for cell-specific targeting. This is often achieved by functionalizing the LNP surface with ligands or antibodies that selectively bind to receptors on target cells, in addition to optimizing the internal lipid composition. To realize a cell-specific mRNA therapy for inflammatory bowel disease (IBD), Veiga et al. prepared an antibody-modified LNP for targeted delivery of IL-10 mRNA to Ly6c+ inflammatory leukocytes.²¹ They used an ASSET (anchored secondary scFv enabling targeting) micelle, an original modular targeting platform that bridges LNPs with targeting antibodies under mild conditions. After intravenous injection into dextran sodium sulfate-induced colitis mice, this surface-modified LNP actively targeted Ly6c+ leukocytes and induced the production of IL-10, significantly inhibiting inflammation in the colon. Similarly, the hydrophobic properties of cell membranes can be exploited to enhance cellular uptake and regulate Kupffer cell immune responses by increasing the size of LNPs and modifying their surfaces with hydrophobic molecules.³³ In another study, researchers developed liver sinusoidal endothelial cell (LSEC)-targeted LNPs to treat peanut-induced food allergies.³⁴ This targeting capability was achieved by modifying LNPs with mannose, incorporated into the formulation as DSPE-PEG_2k-Mannose. The resulting LNPs, approximately 150 nm in size with a neutral surface charge, showed high accumulation in the liver following intravenous injection. Cellular uptake by LSECs was significantly greater than with unmodified LNPs. These mannose-modified LNPs offer a promising platform for treating allergic disorders and autoimmune diseases.

Apart from modifications of targeting ligands, the addition of a fifth component to the LNP formulations has also been shown to redirect LNPs to tissues such as the lungs, spleen, or lymph nodes. In one study, the Siegwart group developed passive targeting LNPs called selective organ targeting (SORT) nanoparticles.³⁵ These LNPs include a fifth lipid added to the traditional four-component LNP composition, enabling selective targeting of organs such as the liver, lungs, and spleen after i.v. injection (Fig. 1). To target the liver, the ionizable cationic lipid DODAP was incorporated into the LNP formulation as the fifth lipid, enhancing encapsulation and intracellular release of the nucleic cargo. The anionic 18PA lipid was incorporated to ensure targeting to the spleen, modulating the charge and stability of the LNP and enhancing membrane fusion. To target the lungs, the cationic DOTAP lipid was incorporated as the fifth lipid to further enhance encapsulation efficacy. They found that the incorporation of the fifth lipid altered the biodistribution to target organs, and this effect was related to the proportion of the fifth lipid and the categories of serum proteins that bind with LNPs after intravenous injection.^35,36

Fig. 1 Design of selective organ targeting (SORT) nanoparticles and their organ-specific targeting delivery after intravenous injection. Typical four-component LNPs with fixed ratios mainly deliver to the liver, which is caused by the interactions between adsorbed ApoE on LNPs and LDL receptors on hepatocytes. To realize organ-specific drug delivery, the fifth component SORT lipid (DOTAP: red, 18PA: green, DODAP: blue) was added to the prescription of LNPs. By changing the SORT lipid with different charges or adjusting the percentage of the SORT lipid, organ-specific systemic delivery of LNPs can be realized. Briefly, cationic lipids (DOTAP), anionic lipids (18PA), and ionizable lipids (DODAP) help LNPs target to the lungs, spleen, and liver, respectively.³⁶ ©2021. Published under the PNAS license.

Extra-hepatic selective organ targeting can also be achieved by adjusting new lipid structures. A study by Qiu et al. aimed to systemically deliver LNPs co-loaded with mRNA to the lungs to treat pulmonary lymphangioleiomyomatosis (LAM).³⁷ Lung delivery via systemic administration requires smaller LNPs with highly stable PEG coatings to navigate through pulmonary capillaries without aggregation. Using a library screening approach, they identified LNPs with cationic lipids containing tails with amide bonds (referred to as N-series LNPs) that selectively targeted the lungs, unlike O-series LNPs, which mainly targeted the liver.³⁸ Targetability to the lungs could be explained by specific interactions with unique plasma proteins that adsorbed onto the surface of the N-series LNPs. Additionally, by tuning the head group structure of the N-series LNPs, they were able to selectively target different pulmonary cell types. Their study demonstrated highly efficient LNP–mRNA delivery to the lungs, reducing the tumor burden of LAM. The study also highlighted how modifying the lipid components of the delivery system according to the route of administration can result in more ideal therapeutic effects.

Finally, optimization of LNP formulations with analogs or ratios can also lead to non-hepatic RNA delivery. In a study by Paunovska et al., researchers developed LNPs specifically targeting Kupffer cells by incorporating various types of cholesterol into their formulations.³⁹ They found that the structure of cholesterol played a critical role in the targeting efficiency. Using the fast identification of nanoparticle delivery (FIND) system, which employs a DNA barcode-based platform to evaluate how over 100 LNPs deliver mRNA to target cells within a single mouse, the researchers demonstrated that oxidized cholesterol enhanced mRNA delivery to liver microenvironment cells, including Kupffer cells and LSECs.⁴⁰ This differed from traditional LNPs, which primarily target hepatocytes.³⁸ Another study demonstrated that the proportion of PEG lipids in LNPs significantly influences their cell-specific targeting ability in the liver after i.v. injection.⁴¹ When the PEG lipid content was increased from 1.0% to 3.0%, LNPs preferentially targeted hepatocytes while reducing delivery to Kupffer cells and LSECs. However, substituting a portion of the PEG lipid with mannose-modified lipids redirected LNPs specifically to LSECs. These findings highlight how fine-tuning the proportion and composition of PEG lipids can enable efficient navigation within the bloodstream to achieve cell-specific RNA delivery, expanding the potential applications of LNP technology.

2.2. Intramuscular (i.m.), intradermal (i.d.), and subcutaneous (s.c.) delivery

2.2.1 Vaccine development. Intramuscular (i.m.) injection is a widely used route for vaccine delivery, requiring LNPs that are specifically optimized for stability within muscle tissue and efficient uptake by immune cells such as dendritic cells and antigen-presenting cells (Fig. 2), allowing for a strong immune response at the injection site and subsequent lymphatic drainage to stimulate wider immune system activation.^6,42–44 Adjuvants and targeting ligands can be incorporated into the LNP formulations to enhance the immune response, making i.m. delivery particularly effective for vaccination purposes.⁴⁵ For example, the mRNA-based COVID-19 vaccines exemplify this optimization, demonstrating robust immunogenicity and efficacy through i.m. administration.^43,44,46 In an earlier study, LNPs formulated with either 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP) or dimethyldioctadecylammonium (DDA) co-delivered with saRNA encoding HIV-1 Env gp140 elicited strong IgG antibody responses in mice when delivered intramuscularly.⁴⁷ In that study, Blakney et al. compared LNP formulations based on cationic and ionizable lipids with self-amplifying RNA (saRNA) on either the interior or exterior surface of the particle. Notably, they demonstrated that by intramuscular injection, these formulations protected the self-amplifying RNA (saRNA) from RNase degradation even when it was adsorbed onto the particle surface rather than encapsulated. This finding suggests that for i.m. delivery, the RNA payload does not always need to be encapsulated for effective delivery, expanding the design possibilities for RNA-based therapies.

Fig. 2 Schematic representation of i.m., i.d., and s.c. administration routes showing the characteristic immune cell-rich layers ideal for vaccination purposes. Created with https://Biorender.com.

The dermal layer of the skin is very rich in professional antigen-presenting cells, such as Langerhans cells and dermal dendritic cells, making it also an attractive target for vaccine delivery.⁴⁸ Intradermal administration can facilitate efficient antigen transport to lymph nodes, inducing potent immune responses while enabling dose-sparing strategies. This can reduce costs associated with vaccine production, storage, and transportation, thereby expanding the global vaccine supply chain.^49,50 The potential of LNPs for intradermal delivery of saRNA vaccines was demonstrated by studies highlighting their ability to elicit strong immune responses with lower doses compared to other routes.^50,51 Despite these advantages, published data on intradermal LNP delivery remain limited. Some studies suggest that optimizing LNP size, charge, and lipid composition could further enhance their efficacy.⁵² For example, LNPs modified with ionizable lipids tailored for dermal cell uptake may enable improved transfection efficiency. Additionally, micro needle-based i.d. delivery systems that combine physical and chemical methods for LNP deposition into the dermal layer are currently being explored as a next-generation approach to improve RNA vaccine delivery.⁵³

2.2.2 Local and systematic applications. Beyond vaccination, i.m. delivery has shown promise for gene-editing therapies. Kenjo et al. developed an LNP-based system to deliver Cas9 mRNA and sgRNA intramuscularly for the treatment of Duchenne muscular dystrophy (DMD), a condition caused by mutations in the dystrophin gene.⁵⁴ These LNPs were synthesized with ionizable lipids featuring triple hydrophobic alkyl tails and demonstrated significant local therapeutic effects after i.m. injection, as well as systemic effects following limb perfusion in a mouse model of DMD. These results highlight the adaptability of LNP modification in i.m. delivery for localized and systemic RNA-based gene editing, especially for diseases requiring tissue-specific targeting.

Subcutaneous injection delivers LNPs into the adipose tissue beneath the dermis, an area abundant in innate immune cells (Fig. 2) such as macrophages, dendritic cells, and antigen-presenting cells.^55,56 Subcutaneous administration is advantageous for achieving sustained release and targeted delivery, making it an attractive route for therapeutics requiring prolonged systemic exposure.^55,57 Optimizing LNP size, PEG-lipid type, and surface modifications is essential to maximize efficacy in s.c. delivery. The study by Sam Chen et al. investigated siRNA-loaded LNP formulations for s.c. administration, focusing on parameters such as size, PEG-lipid dissociation rate, and hepatocyte-specific targeting ligands.⁵⁵ LNPs with intermediate sizes (∼45 nm) and slow-dissociating PEG-lipids (PEG-DSG) exhibited optimal liver accumulation and achieved an 80% reduction in Factor VII (FVII) gene levels at 1 mg siRNA per kg body weight. By contrast, smaller LNPs (∼35 nm) with either rapidly dissociating PEG-DMG or PEG-DSG were less effective, but their activity improved significantly when 0.5 mol% of GalNAc-PEG lipid was added. This demonstrated the importance of balancing size and surface modifications for effective s.c. delivery.

2.3. Intraperitoneal delivery

Owing to the presence of the reticuloendothelial system, systemic LNP–RNA delivery to organs outside the liver, spleen, and lungs remains challenging. An alternate administration route, through the peritoneal cavity, offers a promising approach to address this challenge.⁵⁸ In contrast to intravenous delivery, intraperitoneal administration may reduce systemic toxicity, provide greater bioavailability, and prolong contact with peritoneal organ targets due to the high retention of nanoparticles within the peritoneal cavity.⁵⁹ Thus, for LNPs, altering the nanoparticle chemistry according to the physiology of this route allows a targeted RNA delivery to other organs such as the pancreas. In this context, a study by JR Melamed et al. sought to specifically deliver the mRNA cargo to the pancreas.⁶⁰ They developed an LNP that is strongly mediated by the macrophages within the peritoneal cavity, allowing the horizontal gene transfer specifically to the pancreas. They used three unique ionizable lipidoids (306O_i10, 200O_i10, or 514O_6,10) for the LNP formulation, and compared delivery across the intraperitoneal and intravenous route in mice. They found that for all three of these materials, intraperitoneal injection enhanced mRNA delivery to the pancreas in terms of efficacy and specificity compared to intravenous delivery (which mainly delivers to the liver), indicating how the choice of the delivery route affects therapeutic outcome. To further enhance the pancreatic specificity across this route, they optimized the LNP formulations using cationic helper lipids (40% of total lipids), and found that using the cationic helper lipid DOTAP most effectively reduced off-target luciferase expression in the liver and spleen for all lipidoids. Peritoneal macrophages found in the immune cell-rich peritoneum immediately bind or internalize the LNP–mRNA, thus allowing efficient transfer to the pancreas, either through excretory vesicles or through direct macrophage migration to the pancreatic tissue. Their study showed how adapting the LNP formulation according to the delivery route can enhance targeted efficacy. In another study by Chen et al., researchers synthesized novel biodegradable ionizable lipids for efficient delivery to tumor cells via the intraperitoneal route.⁶¹ The modifications in the ionizable lipid chemistry allowed enhanced endosomal escape and rapid mRNA release into the cytoplasm, addressing one of the key barriers to RNA delivery in the fluid-rich and enzymatically active intraperitoneal cavity. The researchers introduced disulfide bond-bridged ester linkers into the lipid structure, which conferred glutathione (GSH)-responsive degradation. This chemistry was particularly advantageous in the tumor microenvironment, where elevated GSH levels facilitated rapid lipid breakdown, ensuring timely and localized mRNA release. The modular design of these ionizable lipids incorporated cone-shaped architectures that optimized particle formation and stability in the aqueous intraperitoneal environment. By balancing hydrophobic and hydrophilic components, the LNPs achieved a uniform size distribution (∼100–120 nm), critical for efficient cellular uptake by tumor cells and minimized aggregation in peritoneal fluid. The study demonstrated that these modifications not only improved the efficiency of intraperitoneal mRNA delivery but also enhanced the biocompatibility of the LNPs, reducing inflammatory responses often associated with non-degradable lipids. Their work highlights how strategic modifications to LNP components can overcome the unique challenges posed by the intraperitoneal route, including stability in a dynamic environment, specific cellular targeting, and efficient intracellular RNA release, making it a promising platform for therapeutic RNA delivery in localized cancers.

2.4. Pulmonary/inhalation and intranasal delivery

Due to the large absorption area of the lungs and rich pulmonary blood flow, drugs inhaled can be quickly transferred to blood circulation, increasing the bioavailability of drugs.^62,63 Thus, the inhalation route is also a preferred administration route, as it offers a non-invasive and localized approach for delivering RNA therapeutics directly to the lungs (Fig. 3(A)). However, this route presents challenges in dose control, largely due to airway clearance mechanisms, enzymatic degradation, immune surveillance, and fluid dynamics that reduce the efficiency of drug delivery.^64,65 Furthermore, the lung environment is enriched with alveolar macrophages, mucins, and surfactants, all of which can hinder the delivery and uptake of LNPs.^66,67 Therefore, it is imperative that great emphasis is put on LNPs’ physicochemical profiles, especially particle size and surface charge, during formulation. These characteristics greatly affect the behaviour of LNPs in the lungs, from deposition patterns to mucus penetration and immune evasion.⁶⁷ Particle size determines the depth of nanoparticle deposition within the airway.^67,68 Larger particles (>1 μm) are typically trapped in the upper respiratory tract, whereas smaller nanoparticles (<150 nm) have the capacity to diffuse more effectively into the lower respiratory tract, including the bronchioles and alveoli.^69–71 For therapeutic mRNA to reach alveolar epithelial cells and resident immune cells, the LNPs must be sufficiently small to bypass entrapment in upper airway mucus and navigate the tight surfactant layers of the alveoli. Surface charge, indicated by the zeta potential, also plays a crucial role.⁷² While positively charged LNPs may promote uptake via electrostatic interactions with negatively charged cell membranes, they are also more likely to be trapped by mucins and cleared by macrophages.^73–75 On the other hand, neutral or slightly anionic LNPs with PEGylated surfaces show better mucus penetration and prolonged residence time in the lungs, which can improve transfection efficiency and therapeutic outcomes.⁶⁷

Fig. 3 Schematic representation of LNP administration via the (A) inhalational route, showing aerosolized LNPs targeting the pulmonary organs, and (B) intranasal route, where optimized LNPs overcome the mucosal barrier to directly access the brain via three main pathways (a)–(c), and bypass the BBB. BBB = blood–brain barrier. Created with https://Biorender.com.

Nebulization is currently the go-to method for inhalation route formulations.^6,76 Firstly, the LNPs undergo nebulization to generate small droplets capable of flowing through the respiratory tract. However, nebulization often damages the original structure of the LNPs, leading to reduced mRNA encapsulation or transfection efficiency. Secondly, the nebulized LNPs must meet particle size and surface charge requirements to overcome mucosal barriers and enzymatic degradation within the nasal cavity.⁷⁷ To overcome the challenges of nebulization, Lokugamage et al. developed a screening method to identify optimal LNP compositions for mRNA delivery via nebulization.⁶ Their findings revealed that a higher molar ratio of PEG lipids improved the performance of cationic helper lipids, a critical factor for achieving low-dose mRNA delivery. Specifically, they formulated mRNA-loaded LNPs for lung delivery using a combination of modified polyethyleneimine compound 7C1, cholesterol, DMG-PEG_2k, and the cationic lipid DOTAP. The inclusion of a high proportion of DMG-PEG_2k (55%) enhanced lung delivery efficiency, outperforming LNPs used in clinical settings at the time. Nevertheless, aerosolization through nebulization poses unique challenges to LNP integrity.⁷⁷ Tam et al. reported that nebulized LNPs exhibited increased particle size and reduced encapsulation efficiency, likely due to the shear forces exerted during the nebulization process.⁶⁵ To address these limitations, the researchers developed LNPs specifically designed to target airway epithelial cells via the non-invasive intranasal route. Their study explored various helper lipids, including DSPC, DOPC, DOPE, DOPG, egg sphingomyelin (ESM), and DOPS, which influence the uptake efficiency by host cells. Dynamic light scattering measurements revealed that all LNP formulations maintained an average particle size of less than 100 nm and a polydispersity index below 0.2, indicating uniform and stable particles. In vivo experiments demonstrated that LNPs containing helper lipids such as DSPC, DOPC, ESM, or DOPS, loaded with luciferase mRNA, significantly increased luminescence expression in the nasal cavity and lungs, achieving levels at least 10 times higher than the baseline control. These results highlight the adaptability of LNP formulations for airway-targeted gene therapy. Other research studies addressed the nebulization bottleneck by completely substituting the PEG component with a zwitterionic polymer (ZIP)–lipid conjugate to greatly enhance the nebulizer stability.⁷⁸ LNPs formulated with ZIP–lipids (ZIP–LNPs) were stable to nebulization across a wide range of formulation parameters. The optimized ZIP–LNP formulation, containing reduced cholesterol content relative to traditional PEG-lipid LNPs, demonstrated improved inhaled mRNA delivery in both healthy and muco-obstructed mouse lungs. Repeat administration of the optimized ZIP–LNP formulation was well tolerated and did not result in pulmonary inflammation. Their study demonstrated the potential of zwitterionic polymer–lipid conjugates for improving the performance of inhaled mRNA-LNP formulations.

Besides delivering drugs to the lungs and epithelial cells of the airways, intranasal delivery can also bypass the systemic circulation and, unlike systemic delivery, which must overcome the highly restrictive blood–brain barrier (BBB), this route exploits direct pathways to the brain (Fig. 3(B)) via the olfactory and trigeminal nerves.^77,79 While traditional systemic administration often fails to achieve therapeutic concentrations in the central nervous system (CNS), intranasal delivery can directly access the brain.⁸⁰ The tunable composition of LNPs effectively addresses key challenges such as mucosal barriers (Fig. 3(B)), enzymatic degradation in the nasal cavity, and the need to circumvent the blood–brain barrier (BBB). For example, Sandbhor P et al. explored transferrin (Tf)-enhanced, nasal-targeted LNPs (Tf-PTX-LNPs) for the concomitant delivery of PTX (paclitaxel) and miltefosine (HePc), and assessed their anti-GBM efficacy in vitro and in vivo.⁸¹ Such targeted LNPs manifested significantly augmented cerebral concentrations compared with free drugs. In the in vivo GBM models, treatment with Tf-PTX-LNPs notably increased the survival rate, fortified the antitumor effectiveness, and diminished the toxicity relative to systemic and nasal PTX administration. These outcomes indicate that LNP modification based on the nasal route to the brain possesses considerable promise as an innovative, non-invasive therapeutic modality for GBM management. The potential of the intranasal route has also been explored for vaccine therapy. A study by Bowen Li et al. demonstrated that LNPs equipped with ionizable lipids modified with cyclic amine head groups and mRNA encoding antigens fused to a natural adjuvant derived from the C3 complement protein substantially enhanced immune responses when administered intranasally in mice.⁸² These LNP formulations induced robust adaptive and innate immune activation, suggesting their potential for improving the efficacy, safety, and convenience of mRNA-based immunization through intranasal administration. The results from Bowen Li et al. underscore the potential of intranasal delivery in overcoming limitations associated with traditional parenteral immunization methods. By bypassing systemic circulation and directly stimulating nasal-associated lymphoid tissue (NALT), intranasally delivered LNP-based vaccines efficiently trigger mucosal immunity.⁸³ This localized response can be particularly beneficial for combating respiratory pathogens, such as influenza, SARS-CoV-2, and respiratory syncytial virus (RSV), which often establish infection in the upper respiratory tract. Moreover, the ability to induce systemic immune responses in conjunction with mucosal immunity broadens the scope of intranasal LNP applications to include therapeutic vaccines targeting systemic diseases like cancer.

2.5. Local tissue delivery

2.5.1 Subretinal/intravitreal delivery. LNPs for ocular delivery, particularly for subretinal and intravitreal injections (Fig. 4), must be designed with utmost precision. The challenges include avoiding inflammation, ensuring compatibility with ocular tissues, and achieving targeted delivery to specific cells within the eye.⁸⁴ The use of specific lipids that can cross ocular barriers and provide sustained release of the therapeutic agent is key in these formulations. In a study led by Sahay G., the researchers discussed the use of LNPs with PEG-variant surface modifications for genome editing in the mouse retina.⁸⁴ They specifically modified the PEG on the LNP surface by generating LNP variants through the inclusion of positively charged amine-modified polyethylene glycol (PEG)-lipids (LNPa), negatively charged carboxyl-modified PEG-lipids (LNPz), and carboxy-ester modified PEG-lipids (LNPx). Each variant PEG modification showed a higher level of retina-specific transfection and gene editing capacity, suggesting their potential use for correcting genetic mutations that lead to blindness. A previous study by the same group used peptide-guided LNPs to deliver mRNA to the neural retina of rodents and nonhuman primates (Fig. 4(A)).⁸⁵ They used screening methods to select photoreceptor-targeting peptides from an M13 bacteriophage-based library, with the top-performing peptide candidates (MH42) conjugated with the LNPs. These peptides were conjugated to the LNP surface by chemically binding it to the functionalized polyethylene glycol (PEG) moiety, using maleimide-thiol coupling chemistry. Their approach enabled the stable attachment of targeting peptides at defined molar ratios to the PEG-lipid components of the LNPs. They demonstrated that intravitreally and subretinally injected peptide-conjugated LNPs delivered mRNA to not only the retinal pigment epithelium and Müller glia, but also to the photoreceptors in the retina (Fig. 4(B) and (C)). Similar results were observed in nonhuman primates, suggesting successful translation and potential mRNA-based therapy for inherited blindness.

Fig. 4 (A) Schematic of the LNP formulation and conjugation with the peptide via maleimide-thiol chemistry and Cre-mouse model depicting both routes of administration trialed. (B) Representative fundus images showing in vivo tdTomato expression after intravitreal delivery of unconjugated and conjugated LNPs. (C) Representative fundus images showing in vivo tdTomato expression combined with ×40 confocal images of retinal cross sections expressing tdTomato after subretinal delivery. Figures are included in the article's Creative Commons License (Creative Commons Attribution Noncommercial 4.0 International (CC BY-NC 4.0) License). Copyright©2023 The Authors.

2.5.2 Intracranial/intracerebral delivery. Lipid nanoparticles (LNPs) have proven to be highly adaptable for intracranial and/or intracerebral delivery of RNA therapeutics, offering precise targeting of neurons, microglia, and tumor cells in the central nervous system (CNS). Key to success in this route is the strategic modification of LNP components to overcome the unique challenges posed by the CNS environment. The study by Tuma et al. highlights the potential of lipid nanoparticles (LNPs) for delivering mRNA directly to the brain via intracerebral injection, demonstrating their adaptability for localized RNA delivery in the central nervous system (CNS).⁸⁶ This research explored the use of LNPs to deliver mRNA encoding Cre recombinase, achieving efficient protein expression in targeted regions of the brain with minimal off-target effects or immune activation. They employed MC3-based LNPs and successfully delivered Cre mRNA and Cas9 mRNA/Ai9 sgRNA to the adult Ai9 mouse brain. Greater than half of the entire striatum and hippocampus was found to be penetrated along the rostro-caudal axis by direct intracerebral injections of MC3 LNP mRNAs. The study underscores the versatility of LNPs for intracerebral delivery, showcasing their capacity to adapt to the challenges of RNA stability and delivery in the CNS. In another study, researchers devised a localized strategy to deliver RNA interference (RNAi) directly to the glioblastoma (GBM) site using hyaluronan (HA)-grafted lipid-based nanoparticles (LNPs).⁸⁷ The LNP surface was functionalized with hyaluronan (HA), a naturally occurring glycosaminoglycan that specifically binds the CD44 receptor expressed on GBM cells. The modified HA-LNPs were able to successfully bind to GBM cell lines and primary neurons of GBM patients after intracranial injection. The results suggest that RNAi therapeutics could effectively be delivered in a localized manner with HA-coated LNPs and ultimately may become a therapeutic modality for GBM, further underscoring the adaptability of LNPs.

2.6. Oral delivery

Oral administration is one of the most patient-friendly routes but poses significant challenges for LNP-based delivery.⁸ It is the ideal administration route for therapeutic target sites along the gastrointestinal (GI) tract. Due to its patient-friendly nature, it could also be an ideal route for other target sites beyond the GI tract. However, the harsh gastrointestinal environment (Fig. 5(A)), characterized by mucosal barriers, acidic pH, and enzymatic activity, can compromise the integrity of LNPs.^8,88 Additionally, the absorption of nanoparticles through the gastrointestinal mucosa is a complex process influenced by factors such as particle size, surface charge, and lipid composition.^89,90 Ideally, the drug-loaded LNP should be able to navigate the gastrointestinal tract and the intestinal transepithelial barrier into circulation with high bioavailability. Recent advancements in LNP formulations for oral delivery have focused on enhancing stability, bioavailability, and absorption, often through the use of protective coatings and mucoadhesive materials.⁸⁹

Fig. 5 (A) Schematic representation of LNP administration via the oral route, showing main therapeutic target sites as well as drug barriers along the GI tract. Created with https://Biorender.com. (B) LNPs remained in the GI tract for at least 8 hours after oral gavage in mice with buffer (PBS), Cy5.5-labeled siRNA, or LNPs with Cy5.5 labeled siRNA. The total siRNA dose was 0.5 mg kg⁻¹. This image (B)⁹¹ is included in the Creative Commons License (Creative Commons Attribution Noncommercial No Derivatives 4.0 International (CC BY-NC-ND 4.0) License). Copyright©2018 The authors.

To address the challenges associated with the oral route, several modifications have been made to the lipid components of LNPs. For example, the use of PEG in the LNP formulation provides protection against the acidic environment of the stomach, thereby enhancing stability.⁹¹ PEGylated LNPs also allow efficient traversal of the mucosal layer of the GI tract due to the hydrophilic nature of PEG. In a study conducted by Rebecca L. Ball et al., the percentage of PEG in the LNP was slightly increased to address the low potency caused by the presence of mucin in Caco-2 cells.⁹² Mouse biodistribution studies indicated that siRNA-loaded nanoparticles were retained in the GI tract for at least 8 hours (Fig. 5(B)). Their study suggests that orally delivered LNPs should be protected in the stomach and upper intestine to promote siRNA delivery to intestinal epithelial cells.

In a study aimed at treating inflammatory bowel disease (IBD), the research team led by Sung J. developed a lipid nanoparticle (LNP) formulation loaded with mRNA encoding interleukin-22 (IL-22), a cytokine known for its protective and regenerative roles in intestinal tissue.⁸⁸ The LNP was uniquely composed of phosphatidic acid, monogalactosyldiacylglycerol (MGDG), and digalactosyldiacylglycerol (DGDG) at a molar ratio of 5 : 2 : 3. These specific lipid components were selected for their biocompatibility and their ability to enhance stability and delivery efficiency in the harsh gastrointestinal environment. The mRNA-loaded LNPs had a particle diameter of approximately 200 nm, which is within the optimal size range for mucosal uptake, and a surface charge of −18 mV, reducing the likelihood of aggregation and promoting cellular uptake in the colonic mucosa. Oral administration of the IL-22 mRNA-loaded LNPs resulted in significantly increased IL-22 expression in the colonic mucosa, demonstrating the ability of the LNPs to effectively protect the mRNA from enzymatic degradation during gastrointestinal transit and deliver it to target cells. This upregulation of IL-22 expression accelerated the healing of colitis in mouse models. The choice of lipid components in the LNP formulation was critical to its success. Phosphatidic acid, a natural phospholipid, provided a robust structural foundation for the particle and facilitated RNA encapsulation. MGDG and DGDG, both derived from plant glycolipids, contributed to the stability of the LNPs under acidic gastric conditions, ensuring efficient transit to the intestinal lumen. These glycolipids also enhanced uptake by epithelial cells in the colonic mucosa, enabling the targeted release of IL-22 mRNA. The inclusion of these lipids, coupled with the precise molar ratio used, highlights the importance of tailoring LNP composition to the unique challenges of oral RNA delivery.

To further enhance the clarity and practical utility of this review, we compiled a detailed summary of representative LNP formulations across various administration routes, along with their optimal physicochemical properties (Table 1). The table highlights specific examples of LNP compositions, particle size, surface charge, PEG content, and targeted tissues optimized for each administration route.

Table 1 Representative LNPs employed through different routes of administration and their optimized properties

Administration route	LNP composition	Size (nm)	Surface charge	PEG-lipid content (%)	Target tissue/cells	Ref.
i.v.	MC3/DSPC/Chol/PEG2k-DMG	80–100	Neutral to slightly negative	1–2	Liver (hepatocytes)	29,30
i.v.	Ionizable lipid/DSPC/Chol/PEG2k-DMG	50–80	Neutral	1–1.5	Hepatocytes	29
i.v.	DOTAP/Chol/PEG (SORT LNPs)	∼100	Cationic/anionic (depending on the 5th lipid)	Variable	Lungs/spleen/other	35,36
i.m.	DOTAP/DSPC/Chol/PEG	80–100	Slightly positive	2–5	Muscle tissue/APCs	45,46
i.d.	DOTAP/Chol/PEG	<150	Slightly positive	2–5	Dendritic cells, APCs	51
s.c.	Ionizable lipid/DSPC/Chol/PEG	40–60	Neutral	1–3	Liver via lymphatics	55
i.p.	Ionizable lipid/helper lipid/Chol/PEG	100–120	Neutral to slightly positive	1–2	Pancreas	60
Pulmonary (inhalation)	DOTAP/Chol/PEG or ZIP-LNPs (nebulized)	<100	Slightly anionic	5–10	Airway epithelial cells, endothelial cells	6,65,78
Intranasal	cKK-E12/DOPE/Chol/C14-PEG2k/sodium lauryl sulfate	80–100	Slightly negative	2.5	Epithelial cells, lung-resident APCs	82
Intranasal	SPC/HePc/little DOPE (0.1)	100–400	Negative	—	Brain (via olfactory pathways)	81
Subretinal/intravitreal	Typical LNPs/(functionalized PEG)	<100	Slightly positive (LNPa), neutral (LNPx) or slightly negative (LNPz)	2–5	Retinal pigment epithelium, photoreceptors	84,85
Intracerebral	MC3-based LNPs	<120	Neutral	1–2	Brain neurons, microglia	86
Oral	Phosphatidic acid/MGDG/DGDG	150–250	Negative (∼−18 mV)	—	Intestinal epithelial cells	88

---

3. Discussion

3.1 Clinical progress and translational challenges

The clinical translation of LNP-based RNA therapeutics has seen substantial progress in recent years, particularly in response to the global demand for rapid and scalable mRNA vaccine platforms.⁷ The success of intramuscularly delivered COVID-19 vaccines, such as BNT162b2 (Pfizer–BioNTech) and mRNA-1273 (Moderna), provided proof-of-concept that lipid-based RNA delivery systems can be safely and effectively administered in humans.⁴⁴ These formulations utilized ionizable lipids such as ALC-0315 or SM-102, designed to condense RNA into stable nanoparticles and facilitate endosomal escape following uptake by antigen-presenting cells. Their widespread use highlighted the importance of finely tuned lipid ratios, PEGylation levels, and particle sizes in optimizing immunogenicity and tolerability for systemic applications.

Beyond intramuscular vaccines, numerous investigational LNP-based RNA therapeutics have undergone clinical evaluation across various administration routes (Table 2). For example, patisiran (Onpattro®), the first FDA-approved siRNA drug, utilizes an MC3-based LNP formulation for intravenous delivery to hepatocytes, where it silences transthyretin expression in patients with hereditary transthyretin-mediated amyloidosis.^93,94 The success of patisiran highlights how particle size (∼80 nm), near-neutral charge, and liver-targeted lipid chemistry can be leveraged to exploit the fenestrated hepatic endothelium. Similar i.v.-administered LNPs have also been used in clinical trials for protein replacement or gene modulation therapies in rare metabolic diseases such as methylmalonic acidemia (mRNA-3705) and propionic acidemia (mRNA-3927).^95–97 Clinical strategies have also been explored for inhaled LNPs.⁷ Translate Bio's MRT5005 was the first inhaled mRNA therapy to enter clinical trials for cystic fibrosis.⁹⁸ Delivered via nebulization, it aimed to restore CFTR function in lung epithelial cells. Phase 1/2 trials demonstrated that MRT5005 was generally safe and well-tolerated over a 28-day period, although some febrile and hypersensitivity reactions were noted, and no significant improvements in lung function (FEV1) were observed. The trial underscored the need to overcome formulation stability issues during nebulization, mucociliary clearance, and enzymatic degradation in the airways. Pulmonary LNPs often incorporate muco-inert PEG-lipids or surfactants to improve retention and epithelial uptake, and several preclinical studies have demonstrated the clinical translatability of this approach.^99–101 Although oral delivery remains a challenging route, early-phase research has focused on enteric-coated or mucosal-adhesive LNPs to enhance stability and gastrointestinal absorption.¹⁰² However, no clinically relevant data have been reported yet. In contrast, ocular administration via subretinal or intravitreal injection is gaining traction in gene therapy for inherited retinal disorders.^103,104 Companies such as Editas Medicine (www.editasmedicine.com) and EyeGene (https://www.genevant.com) have announced collaborations, in order to possibly leverage LNPs to deliver mRNA or CRISPR machinery directly to retinal pigment epithelium or photoreceptors. These routes demand formulations that are both non-inflammatory and stable under vitreal or subretinal conditions.^84,85

Table 2 Clinical progress and translational challenges of representative LNP-based RNA therapeutics across different administration routes

Administration route	Investigational product	Indication	Clinical trial number	Key challenges	Ref.
i.m.	BNT162b2 (Pfizer-BioNTech), mRNA-1273 (Moderna)	COVID-19	NCT04368728; NCT04283461	Repeat dosing, immunogenicity, anti-PEG antibodies	43,44,105,106
i.v.	Patisiran (Onpattro®)	hATTR amyloidosis	NCT01960348	Liver targeting, repeat administration	93,94
i.v.	mRNA-3705 (Moderna)	Methylmalonic acidemia	NCT04899310	Rare disease targeting, biodistribution	96,97
Inhalation	MRT5005 (Translate Bio)	Cystic fibrosis	NCT03375047	Nebulization stability, mucus penetration	6,98
Subretinal/intravitreal	LNP-delivered CRISPR/Cas9 (editas medicine)	Retinal diseases	N/A (preclinical)	Ocular inflammation, long-term expression	84,85
Intranasal	Tf-PTX-LNPs (preclinical)	Glioblastoma multiforme (GBM)	N/A (preclinical)	Nose-to-brain targeting, stability	81
Oral	Oravax (oral COVID-19 vaccine candidate)	COVID-19	N/A (early development)	GI degradation, poor absorption	102

---

Each of these clinical routes presents unique barriers to translation. Intravenous and intramuscular delivery often contend with systemic clearance, immunogenicity (especially anti-PEG antibodies), and the need for repeat dosing. However, while the intrinsic immunogenicity in in vitro transcribed (IVT) mRNA is beneficial to the LNP–mRNA vaccines for i.m. delivery, avoiding immune activation is critical for therapeutic applications of LNP–mRNA requiring protein replacement, where targeted mRNA expression and repetitive administration of high doses for a lifetime are required. Pulmonary and oral routes pose challenges with respect to mucosal transport, enzymatic degradation, and patient variability in deposition profiles. Ocular and intranasal formulations must overcome tight anatomical compartments and maintain bioactivity over prolonged exposure. Furthermore, large-scale, route-specific GMP manufacturing remains a major hurdle. Differences in shear sensitivity (e.g., for nebulized products), drying processes (e.g., lyophilization for pulmonary powders), and formulation buffers, as well as route-specific LNP properties all complicate the path to commercial viability.

3.2 Perspectives

The design and optimization of LNPs have made it a versatile platform for site-specific delivery across various routes of administration. The physicochemical properties of LNPs, such as size, lipid composition, and surface characteristics, are tunable, and thus can be modified to address the challenges posed by different delivery routes for enhanced therapeutic indices.¹⁰⁷ The absorption of LNPs and their distribution within the body are heavily influenced by these factors, which in turn affect the metabolism and excretion of the encapsulated drugs.

Achieving optimal bioavailability and ideal pharmacokinetic profiles is a key challenge in LNP optimization, requiring a thorough understanding of the interactions between LNPs and biological systems across different administration routes and target sites. Particle size, for instance, is a critical factor in all routes but has different optimal ranges depending on the target tissue and route-specific barriers (Table 1). For intravenous (i.v.) administration, LNPs typically benefit from sizes between 80 and 150 nm to achieve efficient circulation and prevent rapid clearance by the reticuloendothelial system (RES). In contrast, for intranasal or pulmonary delivery, smaller particle sizes, typically under 100 nm, facilitate effective uptake through the mucosal barriers and improve distribution in the nasal cavity and lungs. For intradermal delivery, i.d. delivery benefits from smaller LNP sizes (<150 nm) to improve uptake by APCs and lymphatic drainage to regional lymph nodes. Surface modifications, such as mannose conjugation, can further enhance dendritic cell targeting. However, optimizing LNP stability in the highly immune-competent dermal environment remains a key challenge. Similarly, for subcutaneous injection, the local immune cell population, including macrophages and dendritic cells, facilitates efficient uptake of RNA-loaded LNPs. Particle size optimization is critical, with intermediate sizes (∼40–60 nm) balancing lymphatic drainage and systemic circulation. PEGylation density and lipid dissociation kinetics also play key roles in controlling release profiles.

Surface charge, dictated by the lipid composition, has profound implications for LNP interaction with biological barriers. Neutral or slightly negative charges reduce nonspecific interactions with serum proteins in systemic delivery, extending circulation time and enhancing biodistribution. On the other hand, positively charged LNPs may be beneficial for overcoming negatively charged cell membranes in mucosal delivery routes, such as intranasal and pulmonary administration. However, excessive positive charge can increase toxicity and trigger immune activation, highlighting the need for a delicate balance in design. In the ocular route, delivering RNA therapeutics to the retina for treating ocular diseases, such as macular degeneration or retinitis pigmentosa, presents distinct challenges due to the tightly organized retinal layers and the blood-retinal barrier. Subretinal injection allows for localized delivery but requires LNPs with high biocompatibility to minimize inflammation. Smaller particle sizes (<120 nm) improve diffusion across retinal layers, while neutral or slightly negative surface charges reduce toxicity and immunogenicity. Intravitreal delivery, while less invasive, demands stable LNP formulations capable of withstanding prolonged exposure to vitreous humor. Thus, LNPs must be optimized with ionizable lipids of reduced pK_a values to ensure longer circulation time in the retina.

The stability of LNPs in different biological environments is also a paramount consideration, requiring careful selection and combination of lipid materials. Modifications such as the incorporation of PEGylated lipids provide LNPs with stability and resistance to aggregation, but excessive PEGylation can reduce cellular uptake and endosomal escape, emphasizing the trade-offs inherent in optimizing LNP design. In i.v. delivery, strategies such as the incorporation of PEGylated lipids and SORT (selective organ targeting) LNPs have improved tissue specificity. In pulmonary delivery, LNPs must overcome mucociliary clearance and enzymatic degradation in the respiratory tract. Strategies such as incorporating bioadhesive lipids or muco-inert formulations have been employed to enhance particle retention in the lungs. For intranasal delivery targeting the brain, LNPs are designed for direct transport via olfactory and trigeminal pathways to reach target sites in the brain. LNP formulations incorporating brain-targeting ligands, such as transferrin or apolipoprotein E peptides, have shown promise in enhancing brain-specific delivery. Oral administration remains one of the most challenging routes for LNP-based RNA delivery due to harsh gastrointestinal (GI) conditions, including acidic pH, enzymatic degradation, and the mucus barrier. To address these barriers, LNPs must be optimized for stability during transit and efficient absorption in the intestinal epithelium.

The blood–brain barrier (BBB) is a functional barrier that allows only essential nutrients into the brain, excluding other foreign molecules. While its structure is essential for keeping harmful entities out, it is also a major roadblock for pharmacological treatment of brain diseases. Several alternative invasive drug delivery approaches, such as transcranial drug delivery and BBB disruption, have been explored with limited success and several challenges. The non-invasive intranasal delivery has shown potential as a solution to this challenge. Nasal administration of drug-loaded LNPs has shown effectiveness in treating central nervous system (CNS) disorders, particularly neurodegenerative diseases, because the nasal route allows direct nose-to-brain drug delivery and bypasses the BBB, while avoiding first-pass metabolism and gastrointestinal degradation. Nonetheless, the feasibility of this application remains an open field for researchers. Drawbacks such as subtherapeutic drug absorption levels and rapid mucociliary clearance must be overcome before clinical application. Intranasal administration of drugs for systemic absorption is also effective for treating other conditions, such as pulmonary diseases, cardiovascular diseases, infections, severe pain, and menopausal syndrome. The intracerebral route also offers a direct access to the brain, but in a localized administration manner. Through this route, modifiable LNPs provide a versatile platform for treating neurodegenerative diseases such as Alzheimer's and Parkinson's. However, despite these advantages, unlike the intranasal route, this direct route is both invasive and inconvenient.

Other delivery routes such as the intraperitoneal offer their own unique advantages in targeted LNP–RNA delivery. The peritoneal cavity, for instance, is enriched in immune cells such as macrophages that can adhere to or internalize LNPs, and subsequently transfer the LNP cargo to target sites outside of the systemic liver, spleen, and lungs such as the pancreas. In fact, owing to this physiology, optimized LNPs can easily be adapted for targeted delivery against diseases such as diabetes and other autoimmune diseases. In addition, compared to the less efficient systemic route for pancreatic targeting, the intraperitoneal route offers a less invasive alternative route through which LNPs can easily be adapted to meet therapeutic needs.

Future advancements in LNP optimization will likely arise from the integration of cutting-edge technologies in materials science and RNA biology. For example, machine learning and computational modelling are being increasingly employed to predict how specific lipid compositions and particle characteristics influence pharmacokinetics and biodistribution. These tools can accelerate the discovery of novel lipid libraries and optimize LNP formulations for specific routes and diseases. Moreover, the use of modular LNP platforms, which allow for interchangeable components tailored to distinct administration routes, could provide scalable and adaptable solutions for a wide range of therapeutic applications. Advances in stimuli-responsive materials, such as lipids that change properties in response to local pH or redox conditions, could further enhance delivery precision and efficacy. As the field evolves, the focus will also shift towards improving the safety and tolerability of LNPs, particularly for chronic or repeated dosing. Current clinical efforts have not only validated the potential of LNP–RNA therapeutics across multiple delivery routes, but also emphasized that successful translation hinges on precision-engineered, route-specific designs. Innovations in lipid chemistry, targeting ligands, excipient selection, and scalable formulation technologies will be critical for broadening the therapeutic reach of LNPs in future clinical landscapes. Efforts to develop biodegradable lipids that degrade into nontoxic metabolites post-delivery are also gaining traction, addressing concerns related to long-term toxicity and systemic accumulation.

Despite the challenges in LNP optimization for site-specific administration routes, there is great optimism that advancements in chemical synthesis and nanotechnology will lead to increasingly effective treatments. Our review findings contribute to understanding the versatility of LNPs and how they can be tailored to enhance delivery efficiency across different routes, thereby improving the therapeutic potential of the RNA cargo. Different administration routes hold various advantages and challenges; thus, more research into each route will provide further advancements and a reliable database for optimal LNP optimization.

Author contributions

Muhammed Boye Jallow: writing – original draft, Conceptualization, software, writing – review and editing. Kun Huang: software. All authors discussed and commented on the manuscript; all authors have read and agreed to the submitted version of the manuscript. Min Qiu: fund acquisition, supervision, writing – review and editing, software.

Data availability

No primary research results, software or code have been included and no new data were generated or analysed as part of this review.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

This research was supported in part by the National Key Research and Development Program of China (2022YFC3400100), the National Natural Science Foundation of China (52303369), and Shanghai Science and Technology Project (23ZR1405400). Apart from Fig. 1, 4(A)–(C) and 5(B), all other figures, including the graphical abstract, were created with https://Biorender.com.

References

1. M. Cornebise, E. Narayanan, Y. Xia, E. Acosta, L. Ci, H. Koch, J. Milton, S. Sabnis, T. Salerno and K. E. Benenato, Adv. Funct. Mater., 2021, 32, 2106727.
2. L. Cui, M. R. Hunter, S. Sonzini, S. Pereira, S. M. Romanelli, K. Liu, W. Li, L. Liang, B. Yang, N. Mahmoudi and A. S. Desai, Small, 2022, 18, e2105832.
3. C. Hald Albertsen, J. A. Kulkarni, D. Witzigmann, M. Lind, K. Petersson and J. B. Simonsen, Adv. Drug Delivery Rev., 2022, 188, 114416.
4. Y. Eygeris, M. Gupta, J. Kim and G. Sahay, Acc. Chem. Res., 2022, 55, 2–12.
5. E. Haghighi, S. S. Abolmaali, A. Dehshahri, S. A. Mousavi Shaegh, N. Azarpira and A. M. Tamaddon, J. Nanobiotechnol., 2024, 22, 710.
6. M. P. Lokugamage, D. Vanover, J. Beyersdorf, M. Z. C. Hatit, L. Rotolo, E. S. Echeverri, H. E. Peck, H. Ni, J. K. Yoon, Y. Kim, P. J. Santangelo and J. E. Dahlman, Nat. Biomed. Eng., 2021, 5, 1059–1068.
7. X. Hou, T. Zaks, R. Langer and Y. Dong, Nat. Rev. Mater., 2021, 6, 1078–1094.
8. M. S. Alqahtani, M. Kazi, M. A. Alsenaidy and M. Z. Ahmad, Front. Pharmacol., 2021, 12, 618411.
9. D. Wu, Q. Chen, X. Chen, F. Han, Z. Chen and Y. Wang, Signal Transduction Targeted Ther., 2023, 8, 217.
10. T. Zhang, H. Yin, Y. Li, H. Yang, K. Ge, J. Zhang, Q. Yuan, X. Dai, A. Naeem, Y. Weng, Y. Huang and X. J. Liang, iScience, 2024, 27, 109804.
11. M. M. Billingsley, N. Singh, P. Ravikumar, R. Zhang, C. H. June and M. J. Mitchell, Nano Lett., 2020, 20, 1578–1589.
12. K. Su, L. Shi, T. Sheng, X. Yan, L. Lin, C. Meng, S. Wu, Y. Chen, Y. Zhang, C. Wang, Z. Wang, J. Qiu, J. Zhao, T. Xu, Y. Ping, Z. Gu and S. Liu, Nat. Commun., 2024, 15, 5659.
13. J. R. Melamed, K. A. Hajj, N. Chaudhary, D. Strelkova, M. L. Arral, N. Pardi, M. G. Alameh, J. B. Miller, L. Farbiak, D. J. Siegwart, D. Weissman and K. A. Whitehead, J. Controlled Release, 2022, 341, 206–214.
14. M. K. Grun, A. Suberi, K. Shin, T. Lee, V. Gomerdinger, Z. M. Moscato, A. S. Piotrowski-Daspit and W. M. Saltzman, Biomaterials, 2021, 272, 120780.
15. K. Lam, P. Schreiner, A. Leung, P. Stainton, S. Reid, E. Yaworski, P. Lutwyche and J. Heyes, Adv. Mater., 2023, 35, e2211420.
16. X. Xu and T. Xia, ACS Nanosci. Au., 2023, 3, 192–203.
17. K. Mrksich, M. S. Padilla, R. A. Joseph, E. L. Han, D. Kim, R. Palanki, J. Xu and M. J. Mitchell, J. Biomed. Mater. Res., Part A, 2024, 112, 1494–1505.
18. S. M. Lee, Q. Cheng, X. L. Yu, S. Liu, L. T. Johnson and D. J. Siegwart, Angew. Chem., Int. Ed., 2021, 60, 5848–5853.
19. J. Di, Z. Du, K. Wu, S. Jin, X. Wang, T. Li and Y. Xu, Pharm. Res., 2022, 39, 105–114.
20. H. Wen, H. Jung and X. Li, AAPS J., 2015, 17, 1327–1340.
21. N. Veiga, M. Goldsmith, Y. Granot, D. Rosenblum, N. Dammes, R. Kedmi, S. Ramishetti and D. Peer, Nat. Commun., 2018, 9, 4493.
22. C. Lopes, J. Cristóvão, V. Silvério, P. R. Lino and P. Fonte, Expert Opin. Drug Delivery, 2022, 19, 1381–1395.
23. L. Hourdel, N. Lebaz, F. Peral, M. Ripoll, S. Briançon, F. Bensaid, S. Luthra and C. Cogné, Int. J. Pharm., 2025, 672, 125297.
24. D. Jung, S. Jang, D. Park, N. H. Bae, C. S. Han, S. Ryu, E.-K. Lim and K. G. Lee, BioChip J., 2025, 19, 79–90.
25. C. Atallah, B. Piegay, V. Chiavazza and C. Charcosset, Chem. Eng. Sci., 2024, 290, 119877.
26. H. Seo, L. Jeon, J. Kwon and H. Lee, Adv. Healthcare Mater., 2023, 12, e2203033.
27. M. Wacker, Int. J. Pharm., 2013, 457, 50–62.
28. R. Chu, Y. Wang, J. Kong, T. Pan, Y. Yang and J. He, J. Mater. Chem. B, 2024, 12, 4759–4784.
29. J. Kim, A. Jozic, A. Mukherjee, D. Nelson, K. Chiem, M. S. R. Khan, J. B. Torrelles, L. Martinez-Sobrido and G. Sahay, Adv. Sci., 2022, 9, e2202556.
30. J. P. Han, M. Kim, B. S. Choi, J. H. Lee, G. S. Lee, M. Jeong, Y. Lee, E. A. Kim, H. K. Oh, N. Go, H. Lee, K. J. Lee, U. G. Kim, J. Y. Lee, S. Kim, J. Chang, H. Lee, D. W. Song and S. C. Yeom, Sci. Adv., 2022, 8, eabj6901.
31. B. B. Mendes, J. Conniot, A. Avital, D. Yao, X. Jiang, X. Zhou, N. Sharf-Pauker, Y. Xiao, O. Adir, H. Liang, J. Shi, A. Schroeder and J. Conde, Nat. Rev. Methods Primers, 2022, 2, 24.
32. C. Huang, X. Duan, J. Wang, Q. Tian, Y. Ren, K. Chen, Z. Zhang, Y. Li, Y. Feng, K. Zhong, Y. Wang, L. Zhou, G. Guo, X. Song and A. Tong, Adv. Sci., 2023, 10, e2205532.
33. L. Catenacci, R. Rossi, F. Sechi, D. Buonocore, M. Sorrenti, S. Perteghella, M. Peviani and M. C. Bonferoni, Pharmaceutics, 2024, 16, 1521.
34. X. Xu, X. Wang, Y. P. Liao, L. Luo, T. Xia and A. E. Nel, ACS Nano, 2023, 17, 4942–4957.
35. Q. Cheng, T. Wei, L. Farbiak, L. T. Johnson, S. A. Dilliard and D. J. Siegwart, Nat. Nanotechnol., 2020, 15, 313–320.
36. S. A. Dilliard, Q. Cheng and D. J. Siegwart, Proc. Natl. Acad. Sci. U. S. A., 2021, 118, e2109256118.
37. M. Qiu, Y. Tang, J. Chen, R. Muriph, Z. Ye, C. Huang, J. Evans, E. P. Henske and Q. Xu, Proc. Natl. Acad. Sci. U. S. A., 2022, 119, e2116271119.
38. C. I. Colino, J. M. Lanao and C. Gutierrez-Millan, Front. Immunol., 2020, 11, 218.
39. K. Paunovska, A. J. Da Silva Sanchez, C. D. Sago, Z. Gan, M. P. Lokugamage, F. Z. Islam, S. Kalathoor, B. R. Krupczak and J. E. Dahlman, Adv. Mater., 2019, 31, e1807748.
40. C. D. Sago, M. P. Lokugamage, K. Paunovska, D. A. Vanover, C. M. Monaco, N. N. Shah, M. Gamboa Castro, S. E. Anderson, T. G. Rudoltz, G. N. Lando, P. Munnilal Tiwari, J. L. Kirschman, N. Willett, Y. C. Jang, P. J. Santangelo, A. V. Bryksin and J. E. Dahlman, Proc. Natl. Acad. Sci. U. S. A., 2018, 115, e9944–e9952.
41. M. Kim, M. Jeong, S. Hur, Y. Cho, J. Park, H. Jung, Y. Seo, H. A. Woo, K. T. Nam, K. Lee and H. Lee, Sci. Adv., 2021, 7, eabf4398.
42. J. M. Richner, S. Himansu, K. A. Dowd, S. L. Butler, V. Salazar, J. M. Fox, J. G. Julander, W. W. Tang, S. Shresta, T. C. Pierson, G. Ciaramella and M. S. Diamond, Cell, 2017, 168, 1114–1125.
43. F. Saadati, S. Cammarone and M. A. Ciufolini, Chemistry, 2022, 28, e202200906.
44. L. Zhang, K. R. More, A. Ojha, C. B. Jackson, B. D. Quinlan, H. Li, W. He, M. Farzan, N. Pardi and H. Choe, npj Vaccines, 2023, 8, 156.
45. A. K. Blakney, S. Ip and A. J. Geall, Vaccines, 2021, 9(2), 97.
46. P. F. McKay, K. Hu, A. K. Blakney, K. Samnuan, J. C. Brown, R. Penn, J. Zhou, C. R. Bouton, P. Rogers, K. Polra, P. J. C. Lin, C. Barbosa, Y. K. Tam, W. S. Barclay and R. J. Shattock, Nat. Commun., 2020, 11, 3523.
47. A. K. Blakney, P. F. McKay, B. I. Yus, Y. Aldon and R. J. Shattock, Gene Ther., 2019, 26, 363–372.
48. N. Pardi, M. J. Hogan, F. W. Porter and D. Weissman, Nat. Rev. Drug Discovery, 2018, 17, 261–279.
49. A. Selmi, F. Vascotto, K. Kautz-Neu, Ö. Türeci, U. Sahin, E. von Stebut, M. Diken and S. Kreiter, Cancer Immunol. Immunother., 2016, 65, 1075–1083.
50. G. Anderluzzi, G. Lou, S. Woods, S. T. Schmidt, S. Gallorini, M. Brazzoli, R. Johnson, C. W. Roberts, D. T. O'Hagan, B. C. Baudner and Y. Perrie, J. Controlled Release, 2022, 342, 388–399.
51. W. Zhou, L. Jiang, S. Liao, F. Wu, G. Yang, L. Hou, L. Liu, X. Pan, W. Jia and Y. Zhang, Viruses, 2023, 15, 1760.
52. H. N. Jung, S. Y. Lee, S. Lee, H. Youn and H. J. Im, Theranostics, 2022, 12, 7509–7531.
53. I. Noh, K. Lee and Y. S. Rhee, J. Pharm. Invest., 2022, 52, 273–292.
54. E. Kenjo, H. Hozumi, Y. Makita, K. A. Iwabuchi, N. Fujimoto, S. Matsumoto, M. Kimura, Y. Amano, M. Ifuku, Y. Naoe, N. Inukai and A. Hotta, Nat. Commun., 2021, 12, 7101.
55. S. Chen, Y. Y. Tam, P. J. Lin, A. K. Leung, Y. K. Tam and P. R. Cullis, J. Controlled Release, 2014, 196, 106–112.
56. C. Zeng, C. Zhang, P. G. Walker and Y. Dong, Curr. Top. Microbiol. Immunol., 2022, 440, 71–110.
57. B. Bittner, W. Richter and J. Schmidt, BioDrugs, 2018, 32, 425–440.
58. N. Mirahmadi, M. H. Babaei, A. M. Vali and S. Dadashzadeh, Int. J. Pharm., 2010, 383, 7–13.
59. S. Dadashzadeh, N. Mirahmadi, M. H. Babaei and A. M. Vali, J. Controlled Release, 2010, 148, 177–186.
60. J. R. Melamed, S. S. Yerneni, M. L. Arral, S. T. LoPresti, N. Chaudhary, A. Sehrawat, H. Muramatsu, M. G. Alameh, N. Pardi, D. Weissman, G. K. Gittes and K. A. Whitehead, Sci. Adv., 2023, 9, eade1444.
61. Z. Chen, Y. Tian, J. Yang, F. Wu, S. Liu, W. Cao, W. Xu, T. Hu, D. J. Siegwart and H. Xiong, J. Am. Chem. Soc., 2023, 145, 24302–24314.
62. Y. Y. Zuo, W. E. Uspal and T. Wei, ACS Nano, 2020, 14, 16502–16524.
63. B. Wang, L. Wang, Q. Yang, Y. Zhang, T. Qinglai, X. Yang, Z. Xiao, L. Lei and S. Li, Mater. Today Bio, 2024, 25, 100966.
64. S. P. Newman, Ther. Delivery, 2017, 8, 647–661.
65. A. Tam, J. Kulkarni, K. An, L. Li, D. R. Dorscheid, G. K. Singhera, P. Bernatchez, G. Reid, K. Chan, D. Witzigmann, P. R. Cullis, D. D. Sin and C. J. Lim, Eur. J. Pharm. Sci., 2022, 176, 106234.
66. Z. Deng, G. T. Kalin, D. Shi and V. V. Kalinichenko, Am. J. Respir. Cell Mol. Biol., 2021, 64, 292–307.
67. X. Bai, Q. Chen, F. Li, Y. Teng, M. Tang, J. Huang, X. Xu and X.-Q. Zhang, Nat. Commun., 2024, 15, 6844.
68. S. Jabbal, G. Poli and B. Lipworth, J. Allergy Clin. Immunol., 2017, 139, e2011.
69. T. Y. Poh, N. A. t B. M. Ali, M. Mac Aogáin, M. H. Kathawala, M. I. Setyawati, K. W. Ng and S. H. Chotirmall, Part. Fibre Toxicol., 2018, 15, 46.
70. T. Praphawatvet, J. I. Peters and R. O. Williams, Int. J. Pharm., 2020, 587, 119671.
71. Z. Jin, Q. Gao, K. Wu, J. Ouyang, W. Guo and X.-J. Liang, Adv. Drug Delivery Rev., 2023, 202, 115111.
72. W. Kong, Y. Wei, Z. Dong, W. Liu, J. Zhao, Y. Huang, J. Yang, W. Wu, H. He and J. Qi, J. Nanobiotechnol., 2024, 22, 553.
73. S. Behzadi, V. Serpooshan, W. Tao, M. A. Hamaly, M. Y. Alkawareek, E. C. Dreaden, D. Brown, A. M. Alkilany, O. C. Farokhzad and M. Mahmoudi, Chem. Soc. Rev., 2017, 46, 4218–4244.
74. V. Forest and J. Pourchez, Mater. Sci. Eng., C, 2017, 70, 889–896.
75. V. Gyanani and R. Goswami, Pharmaceutics, 2023, 15(4), 1184.
76. S. D. McCarthy, H. E. González and B. D. Higgins, J. Pers. Med., 2020, 10(2), 37.
77. E. W. X. Leong and R. Ge, Biomedicines, 2022, 10, 2179.
78. A. Y. Jiang, S. Lathwal, S. Meng, J. Witten, E. Beyer, P. McMullen, Y. Hu, R. S. Manan, I. Raji, R. Langer and D. G. Anderson, J. Am. Chem. Soc., 2024, 146, 32567–32574.
79. L. Battaglia, P. P. Panciani, E. Muntoni, M. T. Capucchio, E. Biasibetti, P. De Bonis, S. Mioletti, M. Fontanella and S. Swaminathan, Expert Opin. Drug Delivery, 2018, 15, 369–378.
80. S. V. Dhuria, L. R. Hanson and W. H. Frey, 2nd, J. Pharm. Sci., 2010, 99, 1654–1673.
81. P. Sandbhor, J. Goda, B. Mohanty, P. Gera, S. Yadav, G. Chekuri, P. Chaudhari, S. Dutt and R. Banerjee, Drug Delivery Transl. Res., 2023, 13, 608–626.
82. B. Li, A. Y. Jiang, I. Raji, C. Atyeo, T. M. Raimondo, A. G. R. Gordon, L. H. Rhym, T. Samad, C. MacIsaac, J. Witten, H. Mughal, T. M. Chicz, Y. Xu, R. P. McNamara, S. Bhatia, G. Alter, R. Langer and D. G. Anderson, Nat. Biomed. Eng., 2025, 9, 167–184.
83. B. L. Hartwell, M. B. Melo, P. Xiao, A. A. Lemnios, N. Li, J. Y. H. Chang, J. Yu, M. S. Gebre, A. Chang, L. Maiorino, C. Carter, T. J. Moyer, N. C. Dalvie, S. A. Rodriguez-Aponte, K. A. Rodrigues, M. Silva, H. Suh, J. Adams, J. Fontenot, J. C. Love, D. H. Barouch, F. Villinger, R. M. Ruprecht and D. J. Irvine, Sci. Transl. Med., 2022, 14, eabn1413.
84. M. Gautam, A. Jozic, G. L. Su, M. Herrera-Barrera, A. Curtis, S. Arrizabalaga, W. Tschetter, R. C. Ryals and G. Sahay, Nat. Commun., 2023, 14, 6468.
85. M. Herrera-Barrera, R. C. Ryals, M. Gautam, A. Jozic, M. Landry, T. Korzun, M. Gupta, C. Acosta, J. Stoddard, R. Reynaga, W. Tschetter, N. Jacomino, O. Taratula, C. Sun, A. K. Lauer, M. Neuringer and G. Sahay, Sci. Adv., 2023, 9, eadd4623.
86. J. Tuma, Y. J. Chen, M. G. Collins, A. Paul, J. Li, H. Han, R. Sharma, N. Murthy and H. Y. Lee, Biochemistry, 2023, 62, 3533–3547.
87. Z. R. Cohen, S. Ramishetti, N. Peshes-Yaloz, M. Goldsmith, A. Wohl, Z. Zibly and D. Peer, ACS Nano, 2015, 9, 1581–1591.
88. J. Sung, Z. Alghoul, D. Long, C. Yang and D. Merlin, Biomaterials, 2022, 288, 121707.
89. A. A. Date, J. Hanes and L. M. Ensign, J. Controlled Release, 2016, 240, 504–526.
90. X. Jiang, N. Wang, C. Liu, Y. Zhuo, L. Liang, Y. Gan and M. Yu, Drug Discovery Today, 2023, 28, 103507.
91. Z. Xi, E. Ahmad, W. Zhang, J. Li, A. Wang, Faridoon, N. Wang, C. Zhu, W. Huang, L. Xu, M. Yu and Y. Gan, J. Controlled Release, 2022, 342, 1–13.
92. R. L. Ball, P. Bajaj and K. A. Whitehead, Sci. Rep., 2018, 8, 2178.
93. A. Akinc, M. A. Maier, M. Manoharan, K. Fitzgerald, M. Jayaraman, S. Barros, S. Ansell, X. Du, M. J. Hope, T. D. Madden, B. L. Mui, S. C. Semple, Y. K. Tam, M. Ciufolini, D. Witzigmann, J. A. Kulkarni, R. van der Meel and P. R. Cullis, Nat. Nanotechnol., 2019, 14, 1084–1087.
94. X. Zhang, V. Goel, H. Attarwala, M. T. Sweetser, V. A. Clausen and G. J. Robbie, J. Clin. Pharmacol., 2020, 60, 37–49.
95. H. Attarwala, M. Lumley, M. Liang, V. Ivaturi and J. Senn, Nucleic Acid Ther., 2023, 33, 141–147.
96. R. Baek, K. Coughlan, L. Jiang, M. Liang, L. Ci, H. Singh, H. Zhang, N. Kaushal, I. L. Rajlic, L. Van, R. Dimen, A. Cavedon, L. Yin, L. Rice, A. Frassetto, L. Guey, P. Finn and P. G. V. Martini, Nat. Commun., 2024, 15, 3804.
97. Y. Suzuki, Y. Katsurada and K. Hyodo, Mol. Pharmaceutics, 2023, 20, 4883–4892.
98. S. M. Rowe, J. B. Zuckerman, D. Dorgan, J. Lascano, K. McCoy, M. Jain, M. S. Schechter, S. Lommatzsch, V. Indihar, N. Lechtzin, K. McBennett, C. Callison, C. Brown, T. G. Liou, K. D. MacDonald, S. Z. Nasr, S. Bodie, M. Vaughn, E. B. Meltzer and A. J. Barbier, J. Cystic Fibrosis, 2023, 22, 656–664.
99. J. Kim, A. Jozic, Y. Lin, Y. Eygeris, E. Bloom, X. Tan, C. Acosta, K. D. MacDonald, K. D. Welsher and G. Sahay, ACS Nano, 2022, 16, 14792–14806.
100. S. Liu, Y. Wen, X. Shan, X. Ma, C. Yang, X. Cheng, Y. Zhao, J. Li, S. Mi, H. Huo, W. Li, Z. Jiang, Y. Li, J. Lin, L. Miao and X. Lu, Nat. Commun., 2024, 15, 9471.
101. A. Sarode, P. Patel, N. Vargas-Montoya, A. Allawzi, A. Zhilin-Roth, S. Karmakar, L. Boeglin, H. Deng, S. Karve and F. DeRosa, Drug Delivery Transl. Res., 2024, 14, 360–372.
102. K. Suri, L. Pfeifer, D. Cvet, A. Li, M. McCoy, A. Singh and M. M. Amiji, Drug Delivery Transl. Res., 2025, 15, 1755–1769.
103. P. Antas, C. Carvalho, J. Cabral-Teixeira, L. de Lemos and M. C. Seabra, Trends Mol. Med., 2024, 30, 136–146.
104. A. L. Yan, S. W. Du and K. Palczewski, Vision Res., 2023, 206, 108192.
105. Y. N. Lamb, Drugs, 2021, 81, 495–501.
106. F. P. Polack, S. J. Thomas, N. Kitchin, J. Absalon, A. Gurtman, S. Lockhart, J. L. Perez, G. P. Marc, E. D. Moreira, C. Zerbini, R. Bailey, K. A. Swanson, S. Roychoudhury, K. Koury, P. Li, W. V. Kalina, D. Cooper, R. W. Frenck, L. L. Hammitt, Ö. Türeci, H. Nell, A. Schaefer, S. Ünal, D. B. Tresnan, S. Mather, P. R. Dormitzer, U. Şahin, K. U. Jansen and W. C. Gruber, New Engl. J. Med., 2020, 383, 2603–2615.
107. N. Pardi, S. Tuyishime, H. Muramatsu, K. Kariko, B. L. Mui, Y. K. Tam, T. D. Madden, M. J. Hope and D. Weissman, J. Controlled Release, 2015, 217, 345–351.

---