English

• 1. Introduction

  As has been illuminated by groundbreaking research spanning the 1947–1961 period, messenger RNA (mRNA) functions as a transient intermediate that links genes to proteins [1]. The emergence of novel corona viruses in 2019 sparked investigations into the design of mRNA vaccines, which uncovered the immense therapeutic potential of these biomolecules. However, naked mRNA, a negatively charged hydrophilic macromolecule, confronts hurdles when it traverses cell membranes owing to repulsion, as the membrane itself is negatively charged and hydrophobic. Thus, a barrier exists for mRNA entry into cells. Furthermore, within the body's physiological milieu, mRNA is subject to rapid degradation by blood RNA enzymes [2-5]. Consequently, suitable nucleic acid carriers are needed to aid delivery of mRNAs to the target site. Ideal nucleic acid carriers must meet three properties including (1) robust stability to withstand degradation by nucleases and lysosomes within organism, (2) prolonged circulation time in vivo to evade capture by the reticuloendothelial system (RES) and minimize immune responses, (3) accurately target cells to promote efficient uptake and release from target cells [6,7].

  The carriers for nucleic acid delivery developed thus far are broadly categorized into viral and non-viral types. Viral vectors such as lentivirus (LV), adenovirus (Ad) and adeno-associated virus (AAV) have advantages over their non-viral counterparts in terms of high transfection efficiency for transient or stable expression [8]. However, viral vectors can elicit immunogenic reactions and the inserted gene fragments are susceptible to mutation, which pose potential carcinogenic risks that limit drug delivery applications. Consequently, current research on viral vectors focuses on reducing immunogenicity and toxicity without attenuating and even enhancing gene therapy capabilities. The non-viral vector group (Fig. 1) includes lipid-based (e.g., liposomes micelles, lipid nanoparticles (LNPs)) and polymer-based nanoparticles (e.g., chitosan, dendrimers, proteins), and inorganic material-based delivery systems (e.g., iron oxide, gold, quantum dots, carbon nanotubes). Non-viral vectors possess many advantages over the viral vectors. For example, compared to the viral counterparts, non-viral vectors have advantageously low immunogenicity, high biocompatibility, ease of synthetic modification and low production costs. Moreover, they not only are able to deliver genes into cells, through designed modification of their surfaces, they also play roles in enabling fluorescence imaging to monitor the progress of targeted delivery and biodegradation. Most importantly, non-viral vector systems are not limited by the size of the mRNA introduced [9].

  Figure 1

  

  Figure 1.  Non-viral vectors for mRNA delivery, including lipid-based nanoparticles, polymer-baser nanoparticles and inorganic material-based nanoparticles. Created with BioRender.com.

  DownLoad: Full-Size Img PowerPoint

  The U.S. Food and Drug Administration (FDA) approved the first siRNA based therapeutic drug, ONPATTRO^TM (Patisiran), a double-stranded interfering ribonucleic acid (siRNA) encapsulated in a LNP that delivers a siRNAs against mutant and wild-type transthyretin for the treatment of transthyretin-mediated amyloidosis [10]. The approval of ONPATTRO led to heightened awareness of the enormous potential of LNPs in delivering nucleic acids, and subsequently spurred a gradually increasing number of studies of these nanoparticles. LNPs have extensive applications in the field of drug delivery, especially in relationship to gene therapy and RNA interference (RNAi). In the context of the treatment of brain-related diseases, mRNAs are typically encapsulated by various modified LNPs so that they reach a target location via blood circulatory system and then cross the blood-brain barrier (BBB) for delivery to the brain. Additionally, LNPs are utilized for delivering other types of drugs, such as chemotherapeutic agents, for treatment of various diseases, including cancer and genetic disorders. Some FDA-approved drugs based on LNPs and those in clinical trials are listed in Table S1 (Supporting information). It is significant that the compositions of LNPs prevent degradation of delivered nucleic acids during their blood mediated transport to targets. However, challenges remain in developing new LNP platforms for broad use across disease applications given that LNPs primarily accumulate in the liver through the first-pass hepatic clearance effect and apolipoprotein E (ApoE)-mediated pathways [11,12].

  In this review, a brief overview of LNPs employed for nucleic acid delivery is given and then the development of LNP based approaches for delivery of RNAs to the liver as well as other organs is discussed. Several crucial challenges facing LNP formulation and optimization are presented at the end of the presentation. Finally, recent clinical advances made in studies of nucleic acid therapies utilizing LNP delivery vectors encouraged us to end the review by proposing some exciting prospects for future investigations of LNPs and mRNA therapy.

  2. Lipid nanoparticle composition

  LNPs typically contain four components (Fig. 2) including ionizable lipids, amphiphilic phospholipids (i.e., helper lipids), cholesterol and polyethylene glycol (PEG) lipids. LNPs utilized for therapeutic purposes have sizes typically in the 20–200 nanometer range, which enables their use for effective drug delivery in the body [13,14].

  Figure 2

  

  Figure 2.  Schematic representation of nucleic acid containing LNPs and their composition. The mRNA-loaded LNP is composed of an ionizable lipid, PEGylated lipid, phospholid and cholesterol. Created with BioRender.com.

  DownLoad: Full-Size Img PowerPoint

  2.1 Ionizable lipid

  Ionizable lipids (LNPs, Fig. 3) have been a primary focus of studies aimed at the development of new vectors because they have several advantageous features not possessed by their ionic counterparts. For example, traditional permanently charged cationic lipids are prone to interact with negatively charged serum proteins and aggregate in the bloodstream. This phenomenon leads to short half-lives and significant toxic side effects, which are major impediments to their use as vectors. In contrast, ionizable LNPs are uncharged at physiological pHs and, thus minimally interact with the anionic membranes of blood and making them biocompatible [15]. However, endocytosis promoted entry into the endosome, which has a lower local pH, results in protonation of free amine groups in LNPs (proton sponge effect). Creation of positively charged states plays a crucial role in promoting endosomal escape of the LNPs [16-18]. In the majority of current LNP systems, ionizable lipids have been a focal point of research. Screening ionizable lipids for optimal head groups, linkages, and tails to identify those with higher transfection efficiency is currently the most widely used research method.

  Figure 3

  

  Figure 3.  Types of ionizable lipids. The structure of an ionizable lipids is composed of three parts including heads, linkers and tails (First: first-generation-ionizable; Second: second-generation-ionizable; Third: third-generation-ionizable). Created with Marvinsketch.

  DownLoad: Full-Size Img PowerPoint

  Generally, the overall structure of ionizable lipides is comprised of three components, one being a positively charged head group. These groups are responsible for electrostatic interactions with other components and ion exchange, which contribute to maintenance of the structural integrity of the LNPs, thereby enhancing stability and biocompatibility in vivo. These positively charged head groups also interact with negatively charged regions on cell membranes, facilitating cellular uptake and release of LNPs. The ionization state of head groups also influences the rates of drug release in cells. Also, the ionization status of the LNP head group is affected by environmental factors, control over the location of drug release is possible. The second component is a linker that connects the hydrophilic head group and hydrophobic tail of ionizable lipids. The nature of the linker can be designed to adjust the internal structure and/or chemical environment of LNPs, both of which can affect the rate and mode of drug release. Moreover, linkers can have varying effects on maintaining the structural integrity of the nanoparticles, as well as controlling the blood circulation time and resistance to degradation in the body. Linkers can also possess targeting sites that enable LNPs to bind to specific receptors or biomolecules for drug delivery to target cells or tissues. Furthermore, the immunogenicity of LNPs can be reduced by utilizing certain linkers, to enhance their tolerance in vivo. Lastly, LNPs possess hydrophobic tails which primarily influence their pK_a values, lipophilicities and morphologies. By adjusting the length, structure and chemical properties of the hydrophobic tails, one can control the biological distribution, drug release rate and cellular uptake selectivity of LNPs.

  Ionizable lipids contain alkylated tertiary amine groups, and they are classified as either monoamine or diamine lipids based on the numbers of these basic groups. DLin-MC3-DMA (MC3), SM-102 and ALC-0315 are the most common monoamine lipids, which have gained FDA approval as ionizable cationic lipids for RNA delivery [19,20], with MC3 being used for siRNA delivery and SM-102 and ALC-0315 being used for mRNA delivery [21]. Through modification of their head groups, linkers and tails of these amine bearing lipids it is possible derive differently performing vectors.

  2.2 Phospholipid

  Phospholipids are ionizable amphiphilic molecules that are comprised of hydrophobic tails and phosphate ester containing hydrophilic head groups, which aid the formation of LNPs and escape from endosomes (Fig. 4). Phospholipids have a high phase transition temperature, which facilitates the nucleic acid adsorption ability and improves the stability of the corresponding LNPs. Properties of these LNPs, including as biological distribution, clearance rate, drug release propensity, permeability and surface charge, are significantly impacted by the properties of the component phospholipids [22,23]. Commonly used phospholipids for LNP formation includ distearoylphosphatidylcholine (DSPC), dioleoylphos-phatidylcholine (DOPC), dipalmitoylphosphatidylcholine (DPPC), dioleoylphosphatidylethanolamine (DOPE), and distearoylphos-phatidylethanolamine (DSPE) [24]. The nature of the component phospholipid can lead to disruption of the lipid bilayer formed by the LNPs, thereby promoting endosomal escape. Phospholipids such as dioleoylphosphatidylethanolamine (DOPE) and DSPC have been widely employed in clinical translation formulations like Onpattro, which utilizes MC3 as the ionizable lipid for siRNA delivery [25].

  Figure 4

  

  Figure 4.  Representative structures of phospholipid, cholesterol and PEGylated lipid in LNP formulations. Here listed phospholipids include DSPE, DSPC, distearoylphosphatidyl-glycerole (DSPG); cholesterol and derivatives include β-sitosterol, stigmasterol, fucosterol, daucosterol, lanosterin; and PEGylated lipids include 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-amino(polyethylene glycol)−2000 (DSPE-PEG2000), 1,2-distearoyl-rac-glycerin-3-methoxypolyethylene glycol 2000 (DSG-PEG2000). Created with Marvinsketch.

  DownLoad: Full-Size Img PowerPoint

  2.3 Cholesterol

  LNP formulations often contain cholesterol (20%–50% of the total lipids) in their outer shells [26]. Because this steroid alters the fluidity of bilayers, it is frequently used as an additional component to control the stiffness and integrity of membranes in LNPs, and hence improve the stability of the LNPs. Cholesterol is also employed to stabilize formulations by preventing aggregation through steric repulsion and electrostatic effects. This steroid also decreases the permeability coefficients of negatively charged, neutral and positively charged membranes toward Cl⁻, K⁺, Na⁺, and glucose, and it stabilizes membranes by preventing decreases in permeability with increasing temperatures [27]. Cholesterol also altered order degree of phospholipid within bilayers, thus affecting bilayer fluidity and it regulates membrane protein interactions [28,29].

  2.4 PEGylated lipid

  PEG plays a crucial role in LNP formulations. When PEG is present in the LNP assembly process, highly hydrophilic PEG chains with molar masses in the 1–50 kDa accumulate on the surfaces of nanoparticles to create an external polymeric layer that hinders adsorption of serum proteins and the mononuclear phagocyte system (MPS). This process prolongs the circulation time of the nanoparticle by reducing metabolic inactivation and degradation by the renal and MPS [30-32]. PEGylation also prevents nanoparticle aggregation. As a result, the quantity of PEG included determines the sizes and zeta potentials of lipid particles, factors that impact drug delivery [33]. Another potentially important feature of PEGylation is that it can be used to functionalize of surfaces of LNPs, which enables bioconjugation of the lipids with ligands or biomacromolecules.

  Although PEG protects LNPs from uptake by the MPS, it reduces protein adsorption, cellular uptake and transfection LNPs, leading to the so-called "PEG dilemma" [34]. In addition, the immunogenicity of PEG and anti-PEG antibodies are important matters to consider. Specifically, these antibodies can be generated and participate in "antigen-antibody" complex formation and subsequent clearance by macrophages, leading to changes in the biodistribution/pharmacokinetics and reduced efficacy of the delivered drugs [35-38]. Furthermore, existing anti-PEG antibodies, especially IgG and IgM, in the blood of a patient could accelerate clearance, a phenomenon known as accelerated blood clearance [35]. Moreover, complexes formed between IgG and IgM anti-PEG antibodies with antigens can trigger severe adverse reactions such as hypersensitivity reactions, and even life-threatening events [39-41]. Therefore, addressing the important potentially detrimental issues related to the use of PEG lipids in LNPs is a major challenge.

  3. Organ targeting

  Currently, LNPs as target specific nucleic acid delivery vehicles is a topic of great interest. Strategies for intravenous administration of drug formulations typically rely on LNPs ApoE in circulation and internalization by hepatocytes via low-density lipoprotein receptor (LDLR). However, this approach restricts applications of LNPs as vectors because achieving extrahepatic targeting using this approach is challenging. The following discussion focusses on recent progress in developing liver and non-liver targeting LNPs.

  3.1 LNPs for liver targeting

  The liver possesses over 500 functions, including metabolism (such as lipids, carbohydrates, or amino acids), protein secretion (such as clotting factors, plasma proteins, or hormones), and immune response. It serves as one of the most vital organs in body for clearing endogenous and exogenous substances [42,43]. The liver is comprised of two major cell types including parenchymal cells (such as hepatocytes) and non-parenchymal cells. Additionally, approximately 60%–80% of hepatocytes in the liver, other non-parenchymal cells, include Kupffer cells and wandering macrophages, liver sinusoidal endothelial cells (LSECs), hepatic stellate cells (HSCs), bile duct epithelial cells (cholangiocytes), immune cells (dendritic cells (DCs), natural killer cells, and lymphocytes), and circulating blood cells [44-46]. LNPs have been proven to be highly effective agents for delivering nucleic acid drugs to the liver and silencing various gene targets in liver cells [47-49].

  The results of comprehensive research on cellular uptake mechanisms demonstrate that the ApoE-LDLR pathway enables LNP-mediated RNA delivery to the liver for clinical translation. A large amount of ApoE in the bloodstream is adsorbed on the surface of conventional LNPs to facilitate that drugs entry into liver cells through ApoE-LDLR interactions [19,49]. Many reports exist describing the use of ionizable lipids such as SM-102, ALC-0315, C12–200, CKK-E12, and DLin-MC3-DMA in LNP formulations that have demonstrated effectiveness liver gene silencing at lower RNA dose levels.

  Different liver cell types exhibit distinctly different LNP uptake behaviors. Therefore, researches on liver targeting with LNPs focuses on specific liver cell types associated with particular liver diseases. For example, LSECs are closely associated with chronic liver diseases such as liver fibrosis and cirrhosis. Han et al. prepared an anisamide ligand-tethered lipidoids (AA-lipids) library using a one-pot two-step modular approach and screened the members to assess their capability to efficiently deliver RNA payloads into activated fibroblasts [50]. The results showed that the lead AA-lipid, AA-T3A-C12, enhanced RNA delivery and transfection in activated fibroblasts more effectively than the related anisamide and FDA-approved MC3 ionizable lipid analogs. To achieve selective delivery of LNPs to LSECs, Kim et al. explored highly PEGylated LNPs containing LSEC-specific ligands [51]. For this purpose, mannose moieties were introduced onto the surface of LNPs using different concentrations of mannose-PEG lipids, and the generated LNPs were subjected to a test to determine whether the introduction of mannose-PEG lipids alters LNP cellular uptake relative to that of original LNPs in the absence of ApoE. Wang et al. reports a mannose-modified LNP (M-MC3 LNP@TNFα) loaded with tumor necrosis factor α (TNFα) siRNA for targeting liver macrophages, achieving effective inhibition of acute liver injury (Fig. 5a) [52]. The results show that to bring about targeted delivery to LSECs, the ApoE-dependent cellular uptake pathway can be replaced by a mannose ligand directed LNP binding. Several studies found that polylactic-co-glycolic acid (PLGA) nanoparticles that contain encapsulated ovalbumin (OVA) or the peanut allergen Ara h2 target stable scavenger receptors expressed on LSECs while alleviating allergic reactions [53-55]. Based on the results of both studies, Xu et al. found a liver-targeted LNP platform modified with mannose ligands was developed to deliver peanut allergen epitopes encoding non-allergenic MHC-Ⅱ binding T cell epitopes mRNA to LSECs [56]. LSECs, as potent tolerogenic antigen-presenting cells, induce FoxP3+ regulatory T cells (Tregs). Both ionizable lipids accumulated in liver cells and HSCs, which expressed abundant LDLR on their cell membranes. Ferraresso et al. directly compared LNPs containing ALC-0315 or MC3 for the knock-down of coagulation factor Ⅶ (FⅦ) in liver cells and ADAMTS13 in hematopoietic stem cells [57]. They found that ALC-0315 LNPs achieved more effective siRNA-mediated knock-down of target proteins in mouse liver cells and hematopoietic stem cells compared to MC3, but liver toxicity markers were observed at high doses, while no liver toxicity was found with MC3 (Fig. 5b). This comparison could provide more insights into the future development based on these two lipids and the accumulation of LNPs in different liver cells. Kupffer cells, the first liver cells observed to interact with LNPs, are a type of phagocytic cell that is part of the MPS, also known as the RES. These cells comprise 80% of total macrophages in body, indicating their importance in host defense and LNP clearance. Many studies have explored strategies to prevent Kupffer cell clearance for targeted delivery of LNPs to liver cells. For example, treatment with clodronate liposomes or transient depletion of Kupffer cells by knocking out the caveolin1 gene were found to be effective methods for this purpose [58,59].

  Figure 5

  

  Figure 5.  (a) Schematic representation the preparation process and anti-inflammatory effect of M-MC3 LNP@TNFα against mice with acute liver injury. Copied with permission [52]. Copyright 2024, Elsevier. (b) Targeting liver by mRNA/LNP platform. Copied with permission [57]. Copyright 2023, American Chemical Society. (c) Formulation scheme for siRNA SORT LNPs consisting of five lipids. Copied with permission [64]. Copyright 2024, Wiley-Blackwell.

  DownLoad: Full-Size Img PowerPoint

  3.2 LNPs for lung targeting

  Lungs, one of the most important organs in the human body, have been a focal point of nucleic acid drug therapy research, because of the severity of chronic respiratory diseases such as asthma, chronic obstructive pulmonary disease (COPD), cystic fibrosis (CF) and cancer. Many drug delivery carriers for formulations administered systemically encounter difficulties with delivering drugs to the lungs because special cell types, mucous barrier and mucociliary clearance favor accumulation LNPs in the liver rather than lung. Therefore, in recent efforts to enhance lung targeting attention has been given to manipulating formulations by altering lipid structure, composition and molar ratio. Other investigations have aimed at specific binding to relevant proteins or peptides in the target organs or changing the route of administration. In one study, Cheng et al. attempted to achieve extrahepatic targeting of LNPs by using a proposed selective organ targeting (SORT) strategy [60]. The authors added a supplementary component as SORT molecules to modulate the surface charge of LNPs. They found that when the proportion of a positively charged cationic lipid SORT molecules is increased, LNPs shifts from liver to spleen specific accumulation, and then almost entirely to lungs. This observation suggests that the organ selectivity of the new formulation originates from the charge state or function of serum proteins comprising the biomolecular corona. Based on this phenomenon, this team developed practical protocols for treatment of a CF model in the lungs involving in vitro (DOTAP10 LNPs) and in vivo (DOTAP40 LNPs) approaches to improve the LNPs formulation [60-64]. They observed that the devised SORT LNPs achieve precise homologous directed repair-mediated gene correction in the CF model. These efforts demonstrate the potential of the SORT strategy for therapeutic applications involving mRNA and clustered regularly interspaced short palindromic repeats/CRISPR-associated proteins system (CRISPR/Cas) (Fig. 5c).

  Lungs as organs used for gas exchange, investigations have concentrated on the development of a non-invasive pulmonary therapeutic approach for treatment of lung diseases that involves inhalation as the protocol for delivery of mRNA-LNPs. For this purpose, drugs or nanoparticles need to be incorporated into nanoparticles with aerodynamic diameters between 1 µm and 5 µm, which enables to enter the pulmonary airway directly. This delivery mode permits transport of pulmonary genes to airway epithelial cells while avoiding transfecting to other organs. However, current clinical inhalers use compressed air or vibrating mesh spraying for administration, which creates strong shear forces that deform and then inactivate LNPs. Therefore, the development of stable inhalation nebulized LNP formulations is particularly important. In this regard, Jiang et al. screened a lipid library for LNPs that meet tare stable during nebulization and penetration through cellular and extracellular barriers [65]. In another investigation, Lokugamage et al. optimized the composition of LNPs [66]. Their observations show that LNPs with high PEG molar concentrations are stable under adverse nebulization conditions, and effectively deliver mRNA to the lung. Also, optimization of PEG molar concentration and sterol structure also enhanced the ability of the nebulized LNPs to deliver mRNA through the mucosal barrier to CF infected lungs.

  Some studies have concentrated on the use of dry powder inhalers as an alternative to traditional nebulizers to guide mRNA transfection to the lung [67]. In a study by Zimmermann et al., the feasibility of engineering spray-dried LNPs powders for this purpose was evaluated [68]. This group converted Onpattro^® formulations into spray-dried powders that maintain their physicochemical properties and siRNA integrity in pulmonary applications. Studies involving optimization of spray-drying parameters showed that the redispersed LNP size is ~150 nm, residual moisture content is below 5% and RNA loss is < 15%. Thus, this procedure is accompanied by significantly improved RNA activity and lower RNA loss compared to those of nebulizers.

  3.3 LNPs for eye targeting

  Ocular gene delivery therapies based on LNPs have attracted increasing attention, especially in light of the 2017 FDA approval of voretigene neparvovec (Luxturnatm), an AAV-based gene enhancement therapy for the treatment of hereditary retinal mutations in RPE65 double allele mutant dystrophy [69]. The cornea and retina are two of the main therapeutic targets in the eye for in vivo gene therapy, both of which can be accessed through multiple administration routes [70]. However, clinical development of viral vectors for the delivery of eye therapies has been hampered by the fact that most of these agents trigger an immune response. In contrast, LNPs are safer and less immunogenic than viral vectors for delivering gene drugs.

  The cornea is a transparent, avascular tissue located at the front of the eye, and its main role is to refract or bend light in the direction of the lens and retina, and to serve as an external barrier along with the sclera to protect intraocular structures [71]. For many corneal diseases, especially those that are hereditary, corneal transplantation is a superior treatment but the scarcity of high-quality corneal grafts makes it difficult to meet the high demand for treatment of large numbers of patients [72,73]. Consequently, gene delivery technologies utilizing LNPs as vectors have gained a great attention in therapies for corneal diseases. Disruption or silencing of gene targets associated with corneal diseases, such as corneal dystrophies (KRT3 and 12 genes), epidermal growth factor (EGF), transforming growth factor-β (TGF-β) and TNFα genes for epithelial healing, is currently the mainstay of gene therapy [74-77]. For example, Mohanna et al. used the gene editing technology to perform genome editing in mouse cornea [78]. The procedure, in which Cas9/gRNA ribonucleoprotein (RNP) complexes and DNA templates are delivered to mouse cornea through LNPs, demonstrated the effectiveness of this approach to corneal disease treatment. LNP is a mixture of 1,2-dioleyloxy-3-(dimethylamino)propane (DODMA)/DOPE/DSPC/cholesterol lipids in the molar ratio of 20/30/10/40.

  Retina is located in the posterior eye region and is comprised of multilayered tissue composed mainly of neurons and a layer of epithelial cells [79,80]. Vascular endothelial growth factor (VEGF) inhibitors are commonly employed in treatments for retinal diseases, but their use is limited to retinal neovascularisation [81]. Drugs are transmitted to the retina via three main routes including subretinal, intravitreal and suprachoroidal injection. However, the complex structure of the retina makes it difficult for mainstream therapeutic drugs such as VEGF inhibitors to move to the target location following intravitreal injection. Also, this method of administration carries risks such as infection and retinal detachment [82]. Therefore, approaches for gene therapy the utilize targeting LNPs have been widely investigated. In one effort, Patel et al. evaluated the ability of 11 different LNP vectors to deliver mRNA to the retina region of the eye [83]. They found that LNPs containing ionizable lipids with low pK_a values and unsaturated hydrocarbon chains promote the highest level of reporter gene transfection to the retina, suggesting that ionizable lipids have strongly influence targeting of LNPs, and that the surface charge is not negligible contribution to targeting. Gautam et al. also investigated the use of LNPs for targeting retina [84]. For this purpose, the group fabricated vectors containing a positively charged amine-modified (LNPa), negatively charged carboxylate (LNPz) and carboxyl ester (LNPx)-modified PEG-lipids, as well as neutral unmodified PEG-lipids (LNPs). The results show that LNPa, like conventional LNPs, locate in the retinal pigment epithelium (RPE), but LNPx and LNPz also localize in photoreceptor cells. Thus, these formulations are useful for gene therapy to treat inherited retinal diseases (IRDs) (loss or dysfunction of photoreceptor cells).

  Moreover, peptide coupling of LNPs can also facilitate LNP-mediated delivery of mRNAs to the RPE and glial cells, and even other cells. Herrera-Barrera et al. evaluating the potential use of peptide-ligand-modified LNPs in a heptameric peptide library from the M13 phage for mRNA delivery to cells [85]. The LNPs were observed to deliver mRNA to mouse photoreceptors, RPE and Müller glia, demonstrating that peptide-coupled LNPs can be utilized in the treatment of inherited blindness (Fig. 6a) [85-87].

  Figure 6

  

  Figure 6.  (a) Schematic of LNP formulation and conjugation with peptide via maleimide-thiol chemistry and Cre mouse model depicting both routes of administration trialed. Copied with permission [85]. Copyright 2023, American Association for the Advancement of Science. (b) Schematic of brain tumor immunotherapy mediated by BAMPA-O16B/siRNA lipoplex. Illustration of formulating bioreducible BAMPA-O16B/siRNA lipoplex. The BAMPA-O16B/siRNA lipoplexes successfully cross the BBB, deliver siRNA into tumor cells, and simultaneously downregulate the CD47 and PD-L1 expression in tumor cells, resulting in activation of a T cell-dependent antitumor immunity. Copied with permission [86]. Copyright 2022, Elsevier. (c) Schematic of targeted mIL-10@MLNPs. Schematic illustration of LNP-mediated delivery of mIL-10 to M2 microglia in cerebral ischemic regions for poststroke neurological repair. Copied with permission [87]. Copyright 2024, American Chemical Society.

  DownLoad: Full-Size Img PowerPoint

  3.4 LNPs for brain targeting

  The brain, perhaps the most important organ in the human body, has always been the focus of interest for the treatment of various critical diseases, such as ischaemic stroke, cerebral stroke and glioblastoma multiforme (GBM). However, LNPs encapsulating mRNAs cannot be used in brain disease therapies because they do not spontaneously cross the BBB. This issue prevents ready expression of exogenous proteins in the brain which is a major obstacle to the development of drug therapies for brain diseases [88]. The BBB is primarily composed of a continuous layer of nonfenestrated capillary endothelial cells covered by the glycocalyx and securely connected by a network of intercellular tight junctions (TJs) and adherens junctions, a basement membrane, pericytes and perivascular astrocyte end-foot processes [89]. Therefore, enabling passage through the BBB barrier is one of the most critical and difficult aspects of the delivery of nanomedicines to the brain.

  Various strategies have been devised to address this problem. The first one involves using a high-throughput screen of LNPs in which the four components are varied to obtain an optimal formulation. Employing this approach, Palanki et al. identified ionizable LNPs that display stronger brain-targeting ability than DLin-MC3-DMA upon delivery by intracerebroventricular (ICV) injection into the fetuses of mice and cynomolgus macaque [90]. By designing and screening cationic LNPs with different ionizable amine headgroups, Liu et al. also created a lipid (BAMPA-O16B) that has an optimal pK_a for significant enhancement of cellular uptake and endosomal escape of siRNA lipid complexes (Fig. 6b) [86]. Gao et al. synthesized DMG-PEG (DMG-PEG-mannose), containing a mannose tethered chain with high affinity for M2 microglia, and combined it with the amino-ionized lipid AA3-Dlin to construct MLNPs [87]. MLNPs, intravenously injected to the ischemic region of the brain, facilitate delivery of mIL-10 by recognizing the CD206 receptor, which is highly expressed on the surface of M2 microglia (Fig. 6c).

  GBM is a histological grade Ⅳ tumor that accounts for > 80% of malignant gliomas in primary brain cancer. Patients afflicted with GBM have an extremely low survival rate, with a 5% 5-year survival rate and a median survival of < 5 months [91,92]. Therefore, high interest exists in uncovering LNPs that cross the BBB to deliver clinical gene therapies for GBM. To break the BBB blockade, the surface modification of LNPs by linking ligands such as transferrin, insulin, hyaluronic acid (HA) and lipoproteins, is an effective strategy to prevent blockage by the BBB. HA, a natural target of the CD44 receptor that is overexpressed in many cancer cell lines, is often used to target cancer cells. Schäfer et al. devised HA-modified LNPs to deliver mRNA to the brain [93,94]. Considering the fact that programmed death ligand 1 (PD-L1) is highly expressed on glioma-associated myeloid cells, Zhang et al. designed LNP formulation containing a functionalized anti-PD-L1 therapeutic antibody (aPD-L1) modified on the surface, to enhance the efficiency for delivery to GBM [95]. The anti-tumor immunity in GBM is enhanced by activating the interferon gene-stimulating factor (STING) pathway, which induces the production of pro-inflammatory cytokines including type I interferon, that in turn stimulates infiltration and activation of effector T cells [96-98]. The bridging-lipid nanoparticle (B-LNP) with an anti-CD47/PD-L1 dual-targeting capability simultaneously blocks both innate (CD47) and effector (PD-L1) checkpoint molecules, and deliver the STING agonist diABZI to tumor-associated myeloid cells in GBM. This process enhances the targeting capability and promotes an anti-tumor immune response [99].

  When the above strategies are supplemented by using novel administration techniques, the LNPs better penetrate the BBB. For example, Ogawa et al. used a non-invasive administration method involving microbubble-assisted focused ultrasound (FUS) to enhance the permeability of the BBB [100]. This method to induce BBB opening is an emerging, less invasive brain drug delivery technique. In the procedure, ultrasound is focused on the target site of brain after intravenous injection, thereby generating oscillatory or cavitation energy in the vasculature of the irradiated site, which leads to instantaneous opening of the BBB. Ogawa et al. also utilized this method to transiently open the BBB to deliver mRNA-LNP to the mouse brain [101]. The results show that microbubble injection and FUS irradiation selectively increase the permeability of the BBB in the irradiated region, and enhance mRNA-LNP-induced exogenous protein expression.

  3.5 LNPs for spleen targeting

  The immune response is one of the tools currently used in the treatment of many diseases, and the spleen has a strong association with immunity due to its highly vascularized tissue and its very rich immune cell population. Therefore, many of the current vectors targeting the spleen may be designed with functionalized modifications to stimulate immunity. The size of splenic interendothelial clefts is in the 200–500 nm range, nanoparticles with diameters larger than 200 nm usually accumulate in the spleen [102]. Other physicochemical properties affect splenic targeting of LNPs including charge, shape, structure and surface composition. Ni et al. reported that piperazine-containing ionizable lipids (Pi-Lipids) preferentially deliver mRNAs to immune cells in vivo [103]. A high-throughput DNA barcoding protocol was employed to screen delivery efficiencies of 65 chemically distinct LNPs to 14 cell types in vivo. The results indicate that Pi-LNPs preferentially deliver mRNA to immune cells in the liver and spleen. Prior to this study, Fenton et al. screened a library of synthesized degradable diketopiperazine ionic lipid materials that have varying tail-chain lengths, tail geometries and linker spacings [104]. The diketopiperazine lipid were found to display high luciferase protein expression in the spleen. In addition, Zhao et al. described the properties of imidazole-containing lipids, which are particularly effective in T-lymphocyte transfection [105]. By designing, synthesizing and screening a library of imidazole-containing lipids, this group found that heteroatom substitutions (O or S) or disulfide bonding (S-S) is essential for efficient delivery of mRNAs to T-lymphocytes. Imidazole derivatives with longer carbon chains and an ester bond are more effective at delivery than those with shorter carbon chains and amide bonds in the tail. The screened lipids were observed to be 1.6-fold more effective for delivery to the spleen than to the liver, and they promoted delivery of mRNA well to primary T lymphocytes (Fig. 7a) [105-108].

  Figure 7

  

  Figure 7.  (a) Schematic illustration of rough-to-detailed screening. Copied with permission [105]. Copyright 2020, Wiley-VCH. (b) Rational design of one-component ionizable cationic lipids for mRNA delivery to the spleen. Copied with permission [106]. Copyright 2024, Wiley-VCH. (c) The targeted delivery of PS-LNP to the spleen and lymph node (or SLOs) was mediated by "monocytes/macrophages". Copied with permission [107]. Copyright 2022, American Chemical Society. (d) Designing LN-targeting LNPs prepared by microfluidic mixing. Copied with permission [108]. Copyright 2020, American Chemical Society.

  DownLoad: Full-Size Img PowerPoint

  Also, to avoid highly inflammatory and cytotoxic effects associated with substances having an excess of positive charge, Wang et al. designed non-cationic thiourea lipid nanoparticles (NC-TNP), which is responsible for complexation through the phosphate groups of mRNA [109]. NC-TNP had negligible inflammatory and cytotoxic side effects in vivo and exhibit spleen-targeted delivery capabilities that are comparable to SM-102. In addition, NC-TNP exhibit high mRNA-encoded antigen expression in the spleen and elicit strong immune responses. Interestingly, as opposed to substituting cationic lipids, Zhang et al. rationally designed a single component ionized cationic lipid that selectively delivers mRNA to spleen and T cells efficiently [106]. Screening results indicate that a single-component secondary amido lipids bind more tightly to mRNA and have higher cellular uptake compared to tertiary amido lipids. The single-component design also effectively avoids the effective avoidance the "PEG dilemma" caused by PEG-conjugated lipids (Fig. 7b).

  The previously mentioned SORT strategy can also be utilized for charge modulation. In studies based on this strategy, anionic lipids were incorporated as a fifth component into LNPs. The results reveal that original LNPs, which aggregate in the liver, are transformed by anion lipid incorporation to ones that aggregate in the spleen. Lipid particles generated in this manner were subsequently used to promote in situ production of chimeric antigen receptor T-cell immunotherapy (CAR-T) cells using mRNAs coding for CAR. In situ transfection of CAR-T cells was shown to be enhanced by increasing the number of tumor-infiltrating lymphocytes that reduce tumor metastasis to the liver and promote an immune response [61-63,110,111]. These studies have shown that the alteration of the surface charge of nanoparticles can be employed to modulate targeting specificity.

  3.6 LNPs for lymph node (LN) targeting

  LNs are important tissues in tumor immunotherapy. LNs have a broad bean-like structure composed of multiple lymph lobules, which are covered by the subcapsular sinus (SCS), and the SCS is further surrounded by the capsule. The lymphoid lobule is the smallest unit, and small lymph nodes may have only a few, or sometimes just one, lobule. The afferent lymphatic vessels connect to the LNs. Structurally, lymph nodes are divided into the cortex, paracortex, and medulla, and they are connected to the efferent lymphatic vessels. The lymphatic vessels are connected to blood vessels, allowing immune cells and molecules to flow into the lymphatic system. Therefore, the pathways into LNs can be mainly divided into two types: lymphatic transport and vascular transport [112,113].

  Since LNs are key organs for inducing immune responses against pathogens and cancers, transporting immune functional molecules (such as antigens and adjuvants) to lymph nodes through delivery systems is a useful strategy to enhance immune responses. The research on LNPs targeting lymph nodes has been under test for a long time. The relevant factors for LNPs to effectively target lymph nodes include particle size, charge, PEG modification and so on. Luozhong et al. added negatively charged phosphatidylserine (PS) to a standard four-component MC3 basic LNP formulation (PS-LNP) and found that PS-LNP exhibited efficient protein expression in lymph nodes after intravenous administration (Fig. 7c) [107]. Nakamura et al. study demonstrated that 30 nm LNP (30-LNP) was efficiently translocated to LNs and taken up by CD8⁺ DCs, whereas the efficiency of LNP translocation decreased dramatically at 100 and 200 nm [108]. Furthermore, a comparative study of neutral, positively and negatively charged 30-LNP showed that negatively charged 30-LNP moved more efficiently to LNs than the other LNPs, whereas positively charged nanoparticles were mainly restricted to direct lytic transport and were taken up by DCs at the injection site, which could possibly be attributed to the fact that the cationic liposomes mainly aggregated in tissue fluids leading to the formation of deposits at the injection site (Fig. 7d). All of the above studies confirm the conclusion that anionic charge enhances the aggregation of LNPs in LNs, representing a favorable charge profile for LN targeting. In addition to the effects of particle size and charge, the length and density of the PEG are also important factors for LNPs to achieve better translocation to and retention in LNs. As far as data are concerned, the use of longer PEG chains induces a rapid clearance of LNPs but a significantly lower retention in LNs, whereas higher PEG densities of LNPs accumulate faster in LNs than lower PEG densities of LNPs, but retention in LNs shows the opposite result [114,115]. All of these strategies provide insights into achieving LNPs-mediated immunomodulation of LNs, and show the great potential of an LNP system targeting LNs as a universal platform for next-generation mRNA vaccines.

  3.7 LNPs for pancrea targeting

  The pancreas is a very important organ in the human body with two major functions: exocrine and endocrine. The exocrine glands consist of acini, which secrete pancreatic juice, and ducts, through which pancreatic juice is discharged. The endocrine glands of the pancreas are known as pancreatic islets, which contain glucagon-secreting α-cells, insulin-secreting β-cells, and growth-inhibiting hormone-secreting δ-cells. Dysfunctions in the exocrine and endocrine secretion of the pancreas can lead to various diseases such as pancreatitis, diabetes, and pancreatic cancer. These pancreatic diseases pose a significant social and economic burden worldwide. In the case of diabetes, for example, the complications associated with it have led to limited therapeutic advances, especially in terms of medications [116-118].

  Diabetes is a disease of the pancreas that affects the endocrine system. This chronic disease depletes β-cells, leading to a decrease in their function and quality and to insufficient secretion of insulin, the only hormone responsible for lowering blood glucose levels [119]. It has been found that diacylglycerol kinase δ (DGKδ) is a factor associated with diabetes-affected β-cell regeneration and proliferation. DGKδ is localized in the nucleus of pancreatic β-cells, and inhibition of DGKδ promotes the proliferation of pancreatic β-cells and ameliorates the condition of diabetes [120,121]. Therefore, targeting DGKδ and delivering nucleic acid therapeutic drugs to pancreatic islets may be a new basic treatment for diabetes that does not involve transplantation. The clinical potential of pancreatic mRNA delivery has been demonstrated as a therapeutic approach for autoimmune diabetes through viral gene delivery to pancreatic regeneration of insulin-producing β-cells [122]. However, viral vectors limit their ability to be repeatedly administered due to their high immunogenicity and the risk of inducing complications. In contrast, LNPs, a drug delivery system technology for the stable delivery of nucleic acid drugs, demonstrates potential in the treatment of pancreatic diseases with nucleic acids. This is the fact that LNPs facilitate stable delivery of RNA, which is readily degraded in vivo, and that most LNPs accumulate in the liver when administered intravenously, limiting their practical use in a few organs. Oguma et al. went to improve the distribution of LNPs in the pancreas by changing the type and composition ratio of auxiliary lipids [123]. They prepared a three-component LNP consisting of phospholipids, cholesterol, and PEG lipids in order to elucidate the roles of the different phospholipids, and to compare their effects on the distribution of pancreas and pancreatic islets. The conclusion was obtained that the selection of DOPC as a phospholipid could promote the distribution of LNP in pancreatic islets and that changing the particle size could be used to change the distribution of nanoparticles in exocrine glands or endocrine tissues of the pancreas.

  In addition, intraperitoneal injection, which is an effective strategy to selectively deliver drugs to abdominal disease sites, such as ovarian and pancreatic tumors. Compared with intravenous administration, intraperitoneal administration reduces systemic toxicity, provides higher bioavailability and, due to the high retention of nanoparticles in the peritoneal cavity, prolongs the contact time with the peritoneal organ target and reduces to some extent the accumulation of LNP in the hepatic site [124,125]. Melamed et al. found that the provision of LNPs containing cationic auxiliary lipids by intraperitoneal administration could produce robust and specific protein expression in the pancreas [126]. This demonstrates that pancreatic mRNA delivery relies on a mechanism of horizontal gene transfer via exosomal secretion from peritoneal macrophages. These findings described above provide important information for gene therapy of pancreatic diseases such as diabetes and cancer, suggesting that LNPs are a viable, non-viral method of inducing protein expression in difficult-to-transfect pancreatic cells.

  3.8 LNPs for placenta targeting

  The placenta is an organ that develops throughout pregnancy to support the development of the fetus. Its main role is to act as a biological barrier between the maternal and fetal circulation, protecting the fetus from any harmful molecules, and providing a place for the exchange of nutrients and oxygen. Placental dysfunction can lead to serious obstetric complications, such as pre-eclampsia, hemolysis, elevated liver enzymes, low platelet count (HELLP) syndrome, and fetal growth restriction. Many of these complications are characterized by abnormal expression production of system-specific pathological proteins, which makes placental diseases an attractive application for LNP-mediated nucleic acid therapy. They address protein dysregulation at the genetic level and have higher specificity than small-molecule drugs that must be combined with pre-existing proteins [127].

  Several studies have demonstrated that pregnant women with these diseases have elevated blood levels of circulating soluble protein-like tyrosine kinase-1 (sFlt-1) and reduced levels of placental growth factor (PlGF). PlGF is involved in placental pro-angiogenic signaling by binding to vascular endothelial growth factor receptor 1 (VEGFR-1) on endothelial cells. sFlt-1 binds to and inactivates PlGF in circulation resulting in reduced VEGFR-1 signaling and endothelial dysfunction [128]. Young et al. based this conclusion on the delivery of PlGF mRNA in LNPs as a model for protein replacement therapy [129]. They also found a strong correlation between apparent pK_a and PEG content, with a decrease in the molar amount of PEG increasing apparent pK_a, and higher apparent pK_a, potentially driving trophoblast delivery in vitro (Fig. 8a). Swingle et al. have improved placental vasodilation and angiogenesis by up-regulating the incoming VEGF or PlGF or down-regulating sFlt-1 [130]. This would avoid the problem of abnormally low fetal growth rates caused by impaired placental vasodilatation and elevated maternal blood pressure. They designed ionizable LNP A4 for mRNA delivery to the placenta and applied it to mediate placental vasodilation using VEGF mRNA. LNP A4 mediated local VEGF expression and increased fetal vascular area in the placenta with a better safety profile than C12–200 LNPs, an industry standard ionizable formulation (Fig. 8b). These efforts also demonstrate the potential of the mRNA LNP platform for protein replacement therapy to treat placental insufficiency disease.

  Figure 8

  

  Figure 8.  (a) Ranges of parameters used in the DSD to make the library (A1-A18). Copied with permission [129]. Copyright 2024, KeAi Communications Co. (b) designed a library of ionizable lipids to formulate LNPs for mRNA delivery to placental cells and identified a lead LNP. Copied with permission [130]. Copyright 2023, American Chemical Society.

  DownLoad: Full-Size Img PowerPoint

  3.9 Summarization of organ targeting

  In addition to studies concentrating on the more common targets including various extrahepatic organs mentioned above, attention has also been given to specific targeting of other organs such as lymphatic system, bone, gastrointestinal tract, cardiovascular system, kidney and reproductive system. Delivery to each of the latter targets confronts unique biological barriers. To overcome these hurdles, it is important to explore the effects of charge, composition and particle size targeting propensities. For example, one research developed a non-invasive strategy for siRNA nasal delivery based on ionic liquids (ILs) and cationic lipid (2, 3-dioleoyloxy-propyl)-trimethylammonium-chloride (DOTAP) [131]. This overcomes the barrier of the nasal mucosa. It is also possible to integrate proteins with intrinsic targeting abilities into the protein crown of LNPs, to direct delivery to target sites through protein-receptor interactions [132,133]. In addition, it is also possible to enable LNPs to bypass biological barriers and improve accumulation in target organs through changing the administration protocol. The group of alternate administration methods includes but is not limited to ICV injection, intratympanic injection, intravitreal injection, subretinal injection, intramuscular injection, subcutaneous injection, intratumor injection, intranasal drip, and aerosol injection [134,135]. For example, one study validated the semi-elastic coreshell PLGA-LNPs as the carrier of the oral peptide drug [136].

  The mechanisms by which LNP target different organs are mainly involved in the following aspects: (1) Passive targeting, which is based on particle size and surface properties, the size of the LNP determines its distribution within the body. For example, LNP with small particle size can pass through the vascular endothelial cell gap into the tissue gap, and be enriched in organs such as the liver and spleen. Furthermore, physical properties, such as the surface charge of LNPs, also influence their distribution. Positively charged LNPs are more likely to bind to negatively charged cell surfaces, although positive charges may also trigger adverse effects such as immune responses, so the surface charge is usually finely regulated. (2) Active targeting, by modifying specific ligands on the surface of the LNPs, these ligands are able to recognize and bind to specific receptors on the cell surface of the target organ. For example, a ligand that binds specifically to the desialylate glycoprotein receptor (ASGPR) on the surface of liver cells can be attached to the LNP surface. Once the LNPs enter the circulation, it is able to precisely deliver the drug to the liver cells through ligand-receptor interactions. This targeting approach increases the concentration of the drug in the target organ and reduces side effects on non-target organs. (3) Localized administration targeting, for some organs that can be administered locally, such as the eye and lung, LNP can act directly on the target organ. For example, for ocular administration, LNP can be injected directly onto the ocular surface and then absorbed into the intraocular tissues through the cornea or conjunctiva. (4) Physical barrier breaking mechanism targeting, LNPs can be specially designed to break through the physical barriers of organs. For example, when crossing the BBB, the composition and structure of the LNP can be optimized so that it can encapsulate the drug to penetrate the BBB and enter the brain tissue. This may involve changing the lipid composition of the LNPs to make it more flexible, or adding molecules that interact with the BBB endothelium. (5) Endogenous targeting, where the molecular composition of the nanoparticles is adjusted to bind to specific proteins in the serum so that they can be delivered to the target site.

  These mechanisms work synergistically to enable LNP to target specific organs for precise drug delivery. By adjusting the composition and structure of LNP, its targeting can be optimized, resulting in improved therapeutic efficacy and fewer side effects. Some examples of improved organ-targeting formulations of LNPs by different organ targeting approaches are shown in Table S2 (Supporting information). As knowledge about the detailed mechanisms of passive targeting, active targeting and endogenous targeting intensifies, it should become possible to design nanoparticles that reach each organ through the vascular network.

  4. Conclusion and outlook

  The corona virus disease 2019 (COVID-19) pandemic has accelerated the development of mRNA vaccine technology and has significantly highlighted the potential of LNPs as a delivery mechanism for gene therapies. As one of the most promising nucleic acid delivery systems, LNPs have been the focus of extensive theoretical and clinical research aimed at exploring gene therapies for numerous diseases. However, despite being the subject of numerous studies, LNPs have not garnered substantial attention in the context of effective animal transformation. This indicates that there remain obstacles and constraints in the transition of research findings to clinical applications.

  This review summarizes studies aimed at optimizing nucleic acid delivery through modifications in the composition of LNPs, with a focus on both liver-targeting and non-liver-targeting approaches. Research in this field has demonstrated that adjustments to the structure, ratio, and number of LNP components can control their physical and chemical properties, including size, charge, surface characteristics, and pH responsiveness. For example, in the SORT strategy, negatively charged LNPs tend to be delivered to the lymphatic system, while positively charged LNPs tend to lung. This review also highlights efforts to overcome physiological barriers to LNP delivery to both liver and extrahepatic targets. Modifying the administration protocol can enhance LNP accumulation in target organs. For instance, for ocular administration, LNP can be injected directly onto the ocular surface. For brain administration, LNP can be delivered to the brain by ICV injection. In addition to commonly studied extrahepatic organs, attention has been given to specific targeting of organs such as the bone, gastrointestinal tract, cardiovascular system, kidney, and reproductive system. Each of these targets presents unique biological challenges. It is reasonable to believe that, as our understanding of passive targeting, active targeting, and endogenous targeting mechanisms deepens, it will become increasingly feasible to design LNP nanoparticles that can effectively deliver therapeutic agents to specific organs.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Jialin Guo: Writing – original draft, Investigation. Mingrui Gu: Writing – review & editing. Yahui Chen: Writing – review & editing. Tao Xiong: Writing – review & editing. Yiyang Zhang: Methodology. Simin Chen: Methodology. Mingle Li: Writing – review & editing. Xiaoqiang Chen: Writing – review & editing, Supervision, Investigation, Funding acquisition, Conceptualization. Xiaojun Peng: Writing – review & editing, Supervision, Investigation, Funding acquisition, Conceptualization.

  Acknowledgments

  This research was supported by GuangDong Basic and Applied Basic Research Foundation (No. 2023B1515120001) and Shenzhen University 2035 Program for Excellent Research (Nos. 00000208 and 00000225).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.110849.

  
  
• 1. [1]

     M. Cobb, Curr. Biol. 25 (2015) R526–R532. doi: 10.1016/j.cub.2015.05.032

  2. [2]

     D. Castanotto, J.J. Rossi, Nature 457 (2009) 426–433. doi: 10.1038/nature07758

  3. [3]

     A.D. Judge, V. Sood, J.R. Shaw, et al., Nat. Biotechnol. 23 (2005) 457–462. doi: 10.1038/nbt1081

  4. [4]

     S.R. Suter, A. Ball-Jones, M.M. Mumbleau, et al., Org. Biomol. Chem. 15 (2017) 10029–10036. doi: 10.1039/C7OB02654D

  5. [5]

     A. Khvorova, M.F. Osborn, M.R. Hassler, Nat. Biotechnol. 32 (2014) 1197–1198. doi: 10.1038/nbt.3091

  6. [6]

     S.M. Hammond, E. Bernstein, D. Beach, G.J. Hannon, Nature 404 (2000) 293–296. doi: 10.1038/35005107

  7. [7]

     C. Xu, D. Li, Z. Cao, et al., Nano Lett. 19 (2019) 2688–2693. doi: 10.1021/acs.nanolett.9b00657

  8. [8]

     R.M. David, A.T. Doherty, Toxicol. Sci. 155 (2017) 315–325. doi: 10.1093/toxsci/kfw220

  9. [9]

     Y.G. Gao, Y.D. Shi, Y. Zhang, et al., Chem. Commun. 51 (2015) 16695–16698. doi: 10.1039/C5CC06753G

  10. [10]

      K. Garber, Nat. Biotechnol. 36 (2018) 777–778. doi: 10.1038/nbt0918-777

  11. [11]

      D. Loughrey, J.E. Dahlman, Acc. Chem. Res. 55 (2021) 13–23.

  12. [12]

      C. Lorenzer, M. Dirin, A.M. Winkler, V. Baumann, J. Winkler, J. Control. Release 203 (2015) 1–15. doi: 10.1016/j.jconrel.2015.02.003

  13. [13]

      T.M. Allen, P.R. Cullis, Adv. Drug Deliv. Rev. 65 (2013) 36–48. doi: 10.1016/j.addr.2012.09.037

  14. [14]

      C. Wang, Y. Zhang, Y. Dong, Acc. Chem. Res. 54 (2021) 4283–4293. doi: 10.1021/acs.accounts.1c00550

  15. [15]

      C. Meng, Z. Chen, G. Li, T. Welte, H. Shen, Adv. Therapeutics 4 (2021) 2000099. doi: 10.1002/adtp.202000099

  16. [16]

      K.A. Hajj, K.A. Whitehead, Nat. Rev. Mater. 2 (2017) 17056. doi: 10.1038/natrevmats.2017.56

  17. [17]

      P.S. Kowalski, A. Rudra, L. Miao, D.G. Anderson, Mol. Ther. 27 (2019) 710–728. doi: 10.1016/j.ymthe.2019.02.012

  18. [18]

      K. Ewert, H.M. Evans, A. Ahmad, et al., Adv. Genet. 53 (2005) 119–155.

  19. [19]

      A. Akinc, M.A. Maier, M. Manoharan, et al., Nat. Nanotechnol. 14 (2019) 1084–1087. doi: 10.1038/s41565-019-0591-y

  20. [20]

      M. Jayaraman, S.M. Ansell, B.L. Mui, et al., Angew. Chem. Int. Ed. 51 (2012) 8529–8533. doi: 10.1002/anie.201203263

  21. [21]

      H.Y. Cho, T.H. Chuang, S.N. Wu, Biomedicines 9 (2021) 1367. doi: 10.3390/biomedicines9101367

  22. [22]

      J.C. Colas, W. Shi, V.S.N.M. Rao, et al., Micron 38 (2007) 841–847. doi: 10.1016/j.micron.2007.06.013

  23. [23]

      I.F. Uchegbu, A.T. Florence, Adv. Colloid Interfac. 58 (1995) 1–55. doi: 10.1016/0001-8686(95)00242-I

  24. [24]

      A. Samad, Y. Sultana, M. Aqil, Curr. Drug Deliv. 4 (2007) 297–305. doi: 10.2174/156720107782151269

  25. [25]

      S.M. Hoy, Drugs 78 (2018) 1625–1631. doi: 10.1007/s40265-018-0983-6

  26. [26]

      F. Sebastiani, M. Yanez Arteta, M. Lerche, et al., ACS Nano 15 (2021) 6709–6722. doi: 10.1021/acsnano.0c10064

  27. [27]

      D. Papahadjopoulos, S. Nir, S. Ohki, Biochim. Biophys. Acta 266 (1972) 561–583. doi: 10.1016/0005-2736(72)90354-9

  28. [28]

      A. Nagumo, Y. Sato, Y. Suzuki, Chem. Pharm. Bull. 39 (1991) 3071–3074. doi: 10.1248/cpb.39.3071

  29. [29]

      L. Coderch, J. Fonollosa, M. De Pera, et al., J. Control. Release 68 (2000) 85–95. doi: 10.1016/S0168-3659(00)00240-6

  30. [30]

      J. Kim, Y. Eygeris, M. Gupta, G. Sahay, Adv. Drug Deliv. Rev.. 170 (2021) 83–112. doi: 10.1016/j.addr.2020.12.014

  31. [31]

      K. Knop, R. Hoogenboom, D. Fischer, U.S. Schubert, Angew. Chem. Int. Ed. 49 (2010) 6288–6308. doi: 10.1002/anie.200902672

  32. [32]

      A.S. Lewin, R.C. Ryals, S. Patel, et al., Plos One 15 (2020) e0241006. doi: 10.1371/journal.pone.0241006

  33. [33]

      Y. Bao, Y. Jin, P. Chivukula, et al., Pharmcol. Res. 30 (2012) 342–351.

  34. [34]

      S. Schöttler, G. Becker, S. Winzen, et al., Nat. Nanotechnol. 11 (2016) 372–377. doi: 10.1038/nnano.2015.330

  35. [35]

      B.M. Chen, T.L. Cheng, S.R. Roffler, ACS Nano 15 (2021) 14022–14048. doi: 10.1021/acsnano.1c05922

  36. [36]

      Q. Yang, S.K. Lai, WIREs Nanomed. Nanobi. 7 (2015) 655–677. doi: 10.1002/wnan.1339

  37. [37]

      D. Shi, D. Beasock, A. Fessler, et al., Adv. Drug Deliv. Rev. 180 (2022) 114079. doi: 10.1016/j.addr.2021.114079

  38. [38]

      A. Judge, K. McClintock, J.R. Phelps, I. MacLachlan, Mol. Ther. 13 (2006) 328–337. doi: 10.1016/j.ymthe.2005.09.014

  39. [39]

      M. Ibrahim, E. Ramadan, N.E. Elsadek, et al., J. Control. Release 351 (2022) 215–230. doi: 10.1016/j.jconrel.2022.09.031

  40. [40]

      E. Chen, B.M. Chen, Y.C. Su, et al., ACS Nano 14 (2020) 7808–7822. doi: 10.1021/acsnano.9b07218

  41. [41]

      J. Szebeni, D. Simberg, Á. GonzálezFernández, Y. Barenholz, M.A. Dobrovolskaia, Nat. Nanotechnol. 13 (2018) 1100–1108. doi: 10.1038/s41565-018-0273-1

  42. [42]

      E. Trefts, M. Gannon, D.H. Wasserman, Curr. Biol. 27 (2017) R1147–R1151. doi: 10.1016/j.cub.2017.09.019

  43. [43]

      M. Krawczyk, R. Müllenbach, S.N. Weber, V. Zimmer, F. Lammert, Nat. Rev. Gastro. Hepat. 7 (2010) 669–681. doi: 10.1038/nrgastro.2010.170

  44. [44]

      Z. Zhang, C. Wang, Y. Zha, et al., ACS Nano 9 (2015) 2405–2419. doi: 10.1021/nn505166x

  45. [45]

      R. Li, A. Oteiza, K.K. Sørensen, et al., Am. J. Physiol-Gastrl. 300 (2011) G71–G81. doi: 10.1152/ajpgi.00215.2010

  46. [46]

      J. Poisson, S. Lemoinne, C. Boulanger, et al., J. Hepatol. 66 (2017) 212–227. doi: 10.1016/j.jhep.2016.07.009

  47. [47]

      T.S. Zimmermann, A.C.H. Lee, A. Akinc, et al., Nature 441 (2006) 111–114. doi: 10.1038/nature04688

  48. [48]

      C. Wolfrum, S. Shi, K.N. Jayaprakash, et al., Nat. Biotechnol. 25 (2007) 1149–1157. doi: 10.1038/nbt1339

  49. [49]

      A. Akinc, W. Querbes, S. De, et al., Mol. Ther. 18 (2010) 1357–1364. doi: 10.1038/mt.2010.85

  50. [50]

      X. Han, N. Gong, L. Xue, et al., Nat. Commun. 14 (2023) 75. doi: 10.1038/s41467-022-35637-z

  51. [51]

      M. Kim, M. Jeong, S. Hur, et al., Sci. Adv. 7 (2021) eabf4398. doi: 10.1126/sciadv.abf4398

  52. [52]

      Q. Wang, Q. Jiang, D. Li, et al., Chin. Chem. Lett. 35 (2024) 108683. doi: 10.1016/j.cclet.2023.108683

  53. [53]

      Q. Liu, X. Wang, X. Liu, et al., ACS Nano 13 (2019) 4778–4794. doi: 10.1021/acsnano.9b01444

  54. [54]

      Q. Liu, X. Wang, X. Liu, et al., ACS Nano 15 (2020) 1608–1626.

  55. [55]

      Q. Liu, X. Wang, Y.P. Liao, et al., Nano Today 42 (2022) 101370. doi: 10.1016/j.nantod.2021.101370

  56. [56]

      X. Xu, X. Wang, Y.P. Liao, et al., ACS Nano 17 (2023) 4942–4957. doi: 10.1021/acsnano.2c12420

  57. [57]

      F. Ferraresso, A.W. Strilchuk, L.J. Juang, et al., Mol. Pharm. 19 (2022) 2175–2182. doi: 10.1021/acs.molpharmaceut.2c00033

  58. [58]

      Y.N. Zhang, W. Poon, A.J. Tavares, I.D. McGilvray, W.C.W. Chan, J. Control. Release 240 (2016) 332–348. doi: 10.1016/j.jconrel.2016.01.020

  59. [59]

      C.D. Sago, M.P. Lokugamage, G.N. Lando, et al., Nano Lett. 18 (2018) 7590–7600. doi: 10.1021/acs.nanolett.8b03149

  60. [60]

      Q. Cheng, T. Wei, L. Farbiak, et al., Nat. Nanotechnol. 15 (2020) 313–320. doi: 10.1038/s41565-020-0669-6

  61. [61]

      S.A. Dilliard, Q. Cheng, D.J. Siegwart, Proc. Natl. Acad. Sci. U. S. A. 118 (2021) e2109256118. doi: 10.1073/pnas.2109256118

  62. [62]

      T. Wei, Y. Sun, Q. Cheng, et al., Nat. Commun. 14 (2023) 7322. doi: 10.1038/s41467-023-42948-2

  63. [63]

      X. Wang, S. Liu, Y. Sun, et al., Nat. Protoc. 18 (2022) 265–291. doi: 10.1504/ijmpt.2022.122896

  64. [64]

      A. Vaidya, S. Moore, S. Chatterjee, et al., Adv. Mater. 36 (2024) 2313791. doi: 10.1002/adma.202313791

  65. [65]

      A.Y. Jiang, J. Witten, I.O. Raji, et al., Nat. Nanotechnol. 19 (2023) 364–375.

  66. [66]

      M.P. Lokugamage, D. Vanover, J. Beyersdorf, et al., Nat. Biomed. Eng. 5 (2021) 1059–1068. doi: 10.1038/s41551-021-00786-x

  67. [67]

      Y. Qiu, R.C.H. Man, Q. Liao, et al., J. Control. Release 314 (2019) 102–115. doi: 10.1016/j.jconrel.2019.10.026

  68. [68]

      C.M. Zimmermann, D. Baldassi, K. Chan, et al., J. Control. Release 351 (2022) 137–150. doi: 10.1016/j.jconrel.2022.09.021

  69. [69]

      H. Ameri, J. Curr. Ophthalmol. 30 (2018) 1–2. doi: 10.1016/j.joco.2018.01.006

  70. [70]

      R. Zhou, R. Rigden, R. Horai, R. Villasmil, R. Caspi, Invest. Ophth. Vis. Sci. 51 (2010) 1170.

  71. [71]

      M. Wels, D. Roels, K. Raemdonck, S.C. De Smedt, F. Sauvage, J. Control. Release 333 (2021) 560–578. doi: 10.1016/j.jconrel.2021.04.008

  72. [72]

      P. Gain, R. Jullienne, Z. He, et al., JAMA Ophthalmol. 134 (2016) 167–173. doi: 10.1001/jamaophthalmol.2015.4776

  73. [73]

      W. Kah Hie, K. Ka Wai, C. Li Jia, L.Y. Alvin, Int. J. Ophthalmol. 10 (2017) 1154–1162.

  74. [74]

      J.S. Weiss, H.U. Moller, W. Lisch, et al., Cornae 27 (2008) S1–S42.

  75. [75]

      V.F. Oliver, K.A. van Bysterveldt, M. Cadzow, et al., Ophthalmology 123 (2016) 709–722. doi: 10.1016/j.ophtha.2015.12.008

  76. [76]

      M.E. Fini, B.M. Stramer, Cornea 24 (2005) S2–S11. doi: 10.1097/01.ico.0000178743.06340.2c

  77. [77]

      R.R. Mohan, Acta Ophthalmol. 97 (2019) 343. doi: 10.1109/iciip47207.2019.8985692

  78. [78]

      S.Z. Mirjalili Mohanna, D. Djaksigulova, A.M. Hill, et al., J. Control. Release 350 (2022) 401–413. doi: 10.1016/j.jconrel.2022.08.042

  79. [79]

      S. Veleri, C.H. Lazar, B. Chang, et al., Dis. Model. Mech. 8 (2015) 109–129. doi: 10.1242/dmm.017913

  80. [80]

      A. Auricchio, A.J. Smith, R.R. Ali, Hum. Gene Ther. 28 (2017) 982–987. doi: 10.1089/hum.2017.164

  81. [81]

      M. Tolentino, Surv. Ophthalmol. 56 (2011) 95–113. doi: 10.1016/j.survophthal.2010.08.006

  82. [82]

      K. Ghasemi Falavarjani, Q.D. Nguyen, Eye 27 (2013) 787–794. doi: 10.1038/eye.2013.107

  83. [83]

      S. Patel, R.C. Ryals, K.K. Weller, M.E. Pennesi, G. Sahay, et al., J. Control. Release 303 (2019) 91–100. doi: 10.1016/j.jconrel.2019.04.015

  84. [84]

      M. Gautam, A. Jozic, G.L.N. Su, et al., Nat. Commun. 14 (2023) 6468. doi: 10.1038/s41467-023-42189-3

  85. [85]

      M. HerreraBarrera, R.C. Ryals, M. Gautam, et al., Sci. Adv. 9 (2023) eadd4623. doi: 10.1126/sciadv.add4623

  86. [86]

      M.J.W. Evers, W. Du, Q. Yang, et al., J. Control. Release 343 (2022) 207–216. doi: 10.1016/j.jconrel.2022.01.027

  87. [87]

      E. Belykh, K.V. Shaffer, C. Lin, et al., Front. Oncol. 10 (2020) 739. doi: 10.3389/fonc.2020.00739

  88. [88]

      R. Palanki, S.K. Bose, A. Dave, et al., ACS Nano 17 (2023) 13594–13610. doi: 10.1021/acsnano.3c02268

  89. [89]

      S. Liu, J. Liu, H. Li, et al., Biomaterials 287 (2022) 121645. doi: 10.1016/j.biomaterials.2022.121645

  90. [90]

      M. Gao, Y. Li, W. Ho, et al., ACS Nano 18 (2024) 3260–3275. doi: 10.1021/acsnano.3c09817

  91. [91]

      A. Omuro, L.M. DeAngelis, JAMA-J. Am. Med. Assoc. 310 (2013) 1842–1850. doi: 10.1001/jama.2013.280319

  92. [92]

      R. Stupp, W.P. Mason, M.J. van den Bent, et al., N. Engl. J. Med. 352 (2005) 987–996. doi: 10.1056/NEJMoa043330

  93. [93]

      A. Schäfer, L. Evers, L. Meier, et al., Front. Oncol. 12 (2022) 826273. doi: 10.3389/fonc.2022.826273

  94. [94]

      T. Rezaei, M. Hejazi, B. Mansoori, et al., Eur. J. Pharmacol. 888 (2020) 173483. doi: 10.1016/j.ejphar.2020.173483

  95. [95]

      P. Zhang, J. Miska, C. Lee-Chang, et al., Proc. Natl. Acad. Sci. U. S. A. 116 (2019) 23714–23723. doi: 10.1073/pnas.1906346116

  96. [96]

      G.N. Barber, Nat. Rev. Immunol. 15 (2015) 760–770. doi: 10.1038/nri3921

  97. [97]

      J. Ahn, T. Xia, A. Rabasa Capote, D. Betancourt, G.N. Barber, Cancer Cell 33 (2018) 862–873. doi: 10.1016/j.ccell.2018.03.027

  98. [98]

      Y. Zhu, X. An, X. Zhang, et al., Mol. Cancer 18 (2019) 1–15. doi: 10.1186/s12943-018-0930-x

  99. [99]

      P. Zhang, A. Rashidi, J. Zhao, et al., Nat. Commun. 14 (2023) 1610. doi: 10.1038/s41467-023-37328-9

  100. [100]

       K. Ogawa, N. Kato, S. Kawakami, Chem. Pharm. Bull. 68 (2020) 567–582. doi: 10.1248/cpb.c20-00041

  101. [101]

       K. Ogawa, N. Kato, M. Yoshida, et al., J. Control. Release 348 (2022) 34–41. doi: 10.1016/j.jconrel.2022.05.042

  102. [102]

       E. Blanco, H. Shen, M. Ferrari, Nat. Biotechnol. 33 (2015) 941–951. doi: 10.1038/nbt.3330

  103. [103]

       H. Ni, M.Z.C. Hatit, K. Zhao, et al., Nat. Commun. 13 (2022) 4766. doi: 10.1038/s41467-022-32281-5

  104. [104]

       O.S. Fenton, K.J. Kauffman, R.L. McClellan, et al., Angew. Chem. Int. Ed. 57 (2018) 13582–13586. doi: 10.1002/anie.201809056

  105. [105]

       X. Zhao, J. Chen, M. Qiu, et al., Angew. Chem. Int. Ed. 59 (2020) 20083–20089. doi: 10.1002/anie.202008082

  106. [106]

       C. Wang, C. Zhao, W. Wang, X. Liu, H. Deng, Proc. Natl. Acad. Sci. U. S. A. 120 (2023) e2311276120. doi: 10.1073/pnas.2311276120

  107. [107]

       X. Zhang, K. Su, S. Wu, et al., Angew. Chem. Int. Ed. 63 (2024) e202405444. doi: 10.1002/anie.202405444

  108. [108]

       J.E. Dahlman, K.J. Kauffman, Y. Xing, et al., Proc. Natl. Acad. Sci. U. S. A. 114 (2017) 2060–2065. doi: 10.1073/pnas.1620874114

  109. [109]

       E. Álvarez-Benedicto, Z. Tian, S. Chatterjee, et al., Angew. Chem. Int. Ed. 62 (2023) e202310395. doi: 10.1002/anie.202310395

  110. [110]

       C.L. WillardMack, Toxicol. Pathol. 34 (2006) 409–424. doi: 10.1080/01926230600867727

  111. [111]

       O. Ohtani, Y. Ohtani, Arch. Histol. Cytol. 71 (2008) 69–76. doi: 10.1679/aohc.71.69

  112. [112]

       S. Luozhong, Z. Yuan, T. Sarmiento, et al., Nano Lett. 22 (2022) 8304–8311. doi: 10.1021/acs.nanolett.2c03234

  113. [113]

       T. Nakamura, M. Kawai, Y. Sato, et al., Mol. Pharm. 17 (2020) 944–953. doi: 10.1021/acs.molpharmaceut.9b01182

  114. [114]

       S.M. Moghimi, Biomaterials 27 (2006) 136–144. doi: 10.1016/j.biomaterials.2005.05.082

  115. [115]

       Y. Zhuang, Y. Ma, C. Wang, et al., J. Control. Release 159 (2012) 135–142. doi: 10.1016/j.jconrel.2011.12.017

  116. [116]

       M. Arnold, C.C. Abnet, R.E. Neale, et al., Gastroenterology 159 (2020) 335–349. doi: 10.1053/j.gastro.2020.02.068

  117. [117]

       A. McGuigan, P. Kelly, R.C. Turkington, et al., World J. Gastroenterol. 24 (2018) 4846–4861. doi: 10.3748/wjg.v24.i43.4846

  118. [118]

       W.R. Rowley, C. Bezold, Y. Arikan, E. Byrne, S. Krohe, Popul. Health Manag. 20 (2016) 6–12.

  119. [119]

       K.J. Tien, Y.J. Hung, J.F. Chen, et al., J. Diabetes Invest. 10 (2019) 560–570. doi: 10.1111/jdi.12984

  120. [120]

       T. Sawatani, Y.K. Kaneko, T. Ishikawa, J. Pharmacol. Sci. 140 (2019) 178–186. doi: 10.1016/j.jphs.2019.06.001

  121. [121]

       T. Sato, C. Ishiwatari, Y.K. Kaneko, et al., FASEB J. 35 (2021) e21420. doi: 10.1096/fj.202001279RR

  122. [122]

       X. Xiao, P. Guo, C. Shiota, et al., Cell Stem Cell 22 (2018) 78–90. doi: 10.1016/j.stem.2017.11.020

  123. [123]

       T. Oguma, T. Kanazawa, Y.K. Kaneko, et al., J. Control. Release 373 (2024) 917–928. doi: 10.1016/j.jconrel.2024.07.059

  124. [124]

       S. Dadashzadeh, N. Mirahmadi, M.H. Babaei, A.M. Vali, J. Control. Release 148 (2010) 177–186. doi: 10.1016/j.jconrel.2010.08.026

  125. [125]

       N. Mirahmadi, M.H. Babaei, A.M. Vali, S. Dadashzadeh, Int. J. Pharm. 383 (2010) 7–13. doi: 10.1016/j.ijpharm.2009.08.034

  126. [126]

       J.R. Melamed, S.S. Yerneni, M.L. Arral, et al., Sci. Adv. 9 (2023) eade1444. doi: 10.1126/sciadv.ade1444

  127. [127]

       K.E. Shields, A.D. Lyerly, Obstet. Gynecol. 122 (2013) 1077–1081. doi: 10.1097/AOG.0b013e3182a9ca67

  128. [128]

       H. Zeisler, E. Llurba, F. Chantraine, et al., N. Engl. J. Med. 374 (2016) 13–22. doi: 10.1056/NEJMoa1414838

  129. [129]

       R.E. Young, K.M. Nelson, S.I. Hofbauer, et al., Bioact. Mater. 34 (2024) 125–137.

  130. [130]

       K.L. Swingle, H.C. Safford, H.C. Geisler, et al., J. Am. Chem. Soc. 145 (2023) 4691–4706. doi: 10.1021/jacs.2c12893

  131. [131]

       Z. Luyu, D. Zirong, Y. Shuai, et al., Chin. Chem. Lett. 35 (2024) 109101. doi: 10.1016/j.cclet.2023.109101

  132. [132]

       S.A. Dilliard, D.J. Siegwart, Nat. Rev. Mater. 8 (2023) 282–300. doi: 10.1038/s41578-022-00529-7

  133. [133]

       X. Li, X. Guo, M. Hu, R. Cai, C. Chen, J. Mater. Chem. B 11 (2023) 2063–2077. doi: 10.1039/D2TB02455A

  134. [134]

       P.S. Apaolaza, A. del PozoRodríguez, M.A. Solinís, et al., Biomaterials 90 (2016) 40–49. doi: 10.1016/j.biomaterials.2016.03.004

  135. [135]

       H. Tanaka, T. Nakatani, T. Furihata, et al., Mol. Pharm. 15 (2018) 2060–2067. doi: 10.1021/acs.molpharmaceut.7b01084

  136. [136]

       Z. Yang, W. Fazhan, Y. Ting, et al., Chin. Chem. Lett. 31 (2020) 1147–1152. doi: 10.1016/j.cclet.2019.07.009

• 

  Figure 1  Non-viral vectors for mRNA delivery, including lipid-based nanoparticles, polymer-baser nanoparticles and inorganic material-based nanoparticles. Created with BioRender.com.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Schematic representation of nucleic acid containing LNPs and their composition. The mRNA-loaded LNP is composed of an ionizable lipid, PEGylated lipid, phospholid and cholesterol. Created with BioRender.com.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Types of ionizable lipids. The structure of an ionizable lipids is composed of three parts including heads, linkers and tails (First: first-generation-ionizable; Second: second-generation-ionizable; Third: third-generation-ionizable). Created with Marvinsketch.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Representative structures of phospholipid, cholesterol and PEGylated lipid in LNP formulations. Here listed phospholipids include DSPE, DSPC, distearoylphosphatidyl-glycerole (DSPG); cholesterol and derivatives include β-sitosterol, stigmasterol, fucosterol, daucosterol, lanosterin; and PEGylated lipids include 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-amino(polyethylene glycol)−2000 (DSPE-PEG2000), 1,2-distearoyl-rac-glycerin-3-methoxypolyethylene glycol 2000 (DSG-PEG2000). Created with Marvinsketch.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) Schematic representation the preparation process and anti-inflammatory effect of M-MC3 LNP@TNFα against mice with acute liver injury. Copied with permission [52]. Copyright 2024, Elsevier. (b) Targeting liver by mRNA/LNP platform. Copied with permission [57]. Copyright 2023, American Chemical Society. (c) Formulation scheme for siRNA SORT LNPs consisting of five lipids. Copied with permission [64]. Copyright 2024, Wiley-Blackwell.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) Schematic of LNP formulation and conjugation with peptide via maleimide-thiol chemistry and Cre mouse model depicting both routes of administration trialed. Copied with permission [85]. Copyright 2023, American Association for the Advancement of Science. (b) Schematic of brain tumor immunotherapy mediated by BAMPA-O16B/siRNA lipoplex. Illustration of formulating bioreducible BAMPA-O16B/siRNA lipoplex. The BAMPA-O16B/siRNA lipoplexes successfully cross the BBB, deliver siRNA into tumor cells, and simultaneously downregulate the CD47 and PD-L1 expression in tumor cells, resulting in activation of a T cell-dependent antitumor immunity. Copied with permission [86]. Copyright 2022, Elsevier. (c) Schematic of targeted mIL-10@MLNPs. Schematic illustration of LNP-mediated delivery of mIL-10 to M2 microglia in cerebral ischemic regions for poststroke neurological repair. Copied with permission [87]. Copyright 2024, American Chemical Society.

  下载: 全尺寸图片 幻灯片
  

  Figure 7  (a) Schematic illustration of rough-to-detailed screening. Copied with permission [105]. Copyright 2020, Wiley-VCH. (b) Rational design of one-component ionizable cationic lipids for mRNA delivery to the spleen. Copied with permission [106]. Copyright 2024, Wiley-VCH. (c) The targeted delivery of PS-LNP to the spleen and lymph node (or SLOs) was mediated by "monocytes/macrophages". Copied with permission [107]. Copyright 2022, American Chemical Society. (d) Designing LN-targeting LNPs prepared by microfluidic mixing. Copied with permission [108]. Copyright 2020, American Chemical Society.

  下载: 全尺寸图片 幻灯片
  

  Figure 8  (a) Ranges of parameters used in the DSD to make the library (A1-A18). Copied with permission [129]. Copyright 2024, KeAi Communications Co. (b) designed a library of ionizable lipids to formulate LNPs for mRNA delivery to placental cells and identified a lead LNP. Copied with permission [130]. Copyright 2023, American Chemical Society.

  下载: 全尺寸图片 幻灯片
•