WO2025151019A1 - Cationic liposome and method for preparing rna lipoplex using same - Google Patents

Abstract

The present invention relates to a method for preparing a cationic liposome specifically designed for the preparation of an RNA lipoplex for delivering RNA to a target tissue, and to a method for preparing an RNA lipoplex using the cationic liposome. According to the preparation method of the present invention, high RNA encapsulation efficiency can be achieved while ensuring structural stability and storage stability of the lipoplex, RNA can be effectively delivered to a desired target organ, and the lipoplex can be prepared without using shear stress and/or using an auxiliary surfactant having a high salt concentration.

Description

Cationic liposome and method for producing RNA lipoplex using the same

The present invention relates to a method for preparing cationic liposomes and a method for preparing RNA lipoplexes using the cationic liposomes. More specifically, the present invention relates to a method for preparing cationic liposomes specifically designed for preparing RNA lipoplexes for delivering RNA to target tissues and a method for preparing RNA lipoplexes using the cationic liposomes.

The development of gene delivery systems using nucleic acids and carrier molecules offers promising approaches to address a variety of diseases. This approach has the potential to precisely control the expression of disease-related genes, regulate the expression of functional proteins, and orchestrate regulated immune responses. In recent years, lipid-based nucleic acid delivery systems have emerged as a key technology to facilitate efficient transport and delivery of genetic material. This prominence is demonstrated by the successful application of lipid nanoparticles (LNPs) in the development of messenger RNA (mRNA) vaccines to address coronavirus disease-19 (COVID-19). This demonstrates the diverse potential of lipid-based gene delivery systems.

Although significant progress has been made in the area of lipid-based gene delivery systems, research efforts have primarily focused on exploring the use of different lipid formulations. These explorations have included immobilized cationic lipids (e.g., 1,2-dioleoyl-3-trimethylammonium-propane (DOTAP)) and polycationic lipids, each with unique functional effects. Additionally, neutral lipids, such as 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE), have been incorporated to induce hexagonal phases and promote membrane fusion. Furthermore, the inclusion of ionizable lipids has enabled endosomal escape, a critical step in gene delivery, and organ-specific targeting has been advanced by incorporating novel lipids into LNPs.

Despite significant progress in lipid-based nucleic acid delivery systems, there remains a significant knowledge gap regarding the influence of biophysical properties (e.g., size, lamellarity, shape, surface charge) on lipid assembly and mRNA complexes. Notably, the influence of these systems on the self-assembly process is further highlighted in ‘lipoplexes’, aqueous mixtures of cationic liposomes and nucleic acids, due to the inherent flexibility of preparation derived from various liposome preparation methods. The resulting assembly characteristics of lipoplexes, which are closely related to the preparation method used, have the potential to significantly influence key biological activities. However, previous studies have primarily focused on directly comparing preparation methods in terms of delivery efficiency, making it difficult to decipher whether lipoplex function is dependent on specific factors or whether specific variables are irrelevant due to a lack of a deep understanding of the underlying biophysical properties. Furthermore, the relationship between liposome size and biological activity has yielded conflicting results, creating ambiguity in this seemingly simple association. Therefore, understanding the complex biophysical relationships between the properties of cationic lipid assemblies and mRNA-lipoplexes is critical to fully realizing the therapeutic potential of lipid-based nucleic acid delivery systems.

Against this backdrop, the present inventors systematically investigated the biophysical properties of cationic liposomes and lipoplexes, paying particular attention to the effects of different preparation methods on the assembly structures of lipids. First, cationic liposomes prepared using various preparation methods, including short-chain lipids (1,2-dihexanoyl-sn-glycero-3-phosphocholine (DHPC)) in some formulations, were characterized, and biophysical factors of liposomes, such as size distribution, surface charge, lamellarity, and morphology, were investigated. Then, by varying the N/P ratio, which represents the ratio of positively charged cationic lipids and phosphate groups in mRNA, the effects of liposome properties on the biophysical properties of mRNA-lipoplexes were evaluated. In addition, the biological efficacy of these lipoplexes was evaluated to correlate the initial liposome structure with the biophysical properties and functionality of the lipoplexes. Finally, representative cationic and anionic lipoplexes were subjected to detailed morphological analyses. A comprehensive structural comparison revealed that the liposome structure and N/P ratio play an important role in forming the structure of the lipoplex, which confirmed that the self-assembly of liposomes resulted in distinct structural patterns, thereby completing the present invention.

To achieve the above purpose,

In one aspect, the present invention provides a method for preparing a cationic liposome, comprising the following steps:

(a) a step of preparing a lipid solution by dissolving long-chain cationic lipids and short-chain lipids in an organic solvent, and then drying under a nitrogen gas flow to evaporate the organic solvent and obtain a dried lipid film;

(b) a step of preparing a lipid suspension by hydrating the dried lipid film obtained in the above step with an aqueous solvent; and

(c) A step of immersing the hydrated lipid suspension prepared in the above step in liquid nitrogen, thawing it in a water bath at 10 to 80°C, and then vortexing it for 5 to 60 seconds, repeating the process 1 to 10 times.

In another aspect, the present invention provides a method for preparing a lipoplex comprising RNA, comprising the following steps:

(i) a step of preparing a lipid solution by dissolving a long-chain cationic lipid and a short-chain lipid in an organic solvent, and then drying the solution under a nitrogen gas flow to evaporate the organic solvent and obtain a dried lipid film;

(ii) a step of preparing a lipid suspension by hydrating the dried lipid film obtained in the above step with an aqueous solvent;

(iii) a step of preparing a cationic liposome by immersing the hydrated lipid suspension prepared in the above step in liquid nitrogen, thawing it in a water bath of 10 to 80°C, and then vortexing it for 5 to 60 seconds, repeating the process 1 to 10 times; and

(iv) A step of diluting RNA with an aqueous solvent and then adding the cationic liposome dispersion.

According to the manufacturing method of the present invention, cationic liposomes can be manufactured to have a concentric multilamellar structure without using shear stress and/or using an auxiliary surfactant having a high salt concentration, and the lipoplex can efficiently carry a large amount of drugs such as mRNA compared to conventional lipoplexes, while ensuring high structural stability of the lipoplex. In addition, by controlling the N/P ratio, the lipoplex has the effect of effectively delivering drugs such as mRNA to a desired organ by targeting the targeted location.

Figures 1a-c are schematics of the experimental framework to investigate the effect of initial liposome properties on lipoplex formation.

Specifically, FIG. 1a is a schematic diagram illustrating a method for preparing liposomes and a biophysical and biological evaluation method, along with a conventional extrusion method for comparison with the precisely controlled freeze-thaw based method, Liposome under Cryo-Assembly (LUCA) cycle used in the present invention.

Figure 1b is a drawing showing the chemical structures of lipids used in the present invention. DOTAP is a permanently charged single cationic lipid that promotes nucleic acid complex formation, DOPE is a neutral auxiliary lipid that promotes membrane fusion, and DHPC is a neutral short-chain lipid that modulates liposome properties.

Figure 1c is a table showing the types of cationic liposomes used for lipoplex synthesis using different charge ratios (N/P). N/P represents the molar ratio of positively charged amine (N = nitrogen) groups to negatively charged nucleic acid phosphate (P) groups. Q represents the molar ratio of long-chain lipids to short-chain lipids.

Figures 2a-d are diagrams showing the biophysical properties of cationic liposomes according to the method for preparing them and the inclusion of short-chain lipids.

Specifically, Fig. 2a shows a representative size distribution curve obtained by DLS in intensity-weighted mode.

Figure 2b is a plot of the mean diameter (n = 6) measured by DLS after fitting to a correlation function of single exponential decay.

Figure 2c is a cryo-EM micrograph showing the structural differences between cationic liposomes. Examples of liposome types are highlighted in boxes: blue, unilamellar; green, multilamellar; purple, bilamellar; orange, multivesicular.

Figure 2d is a diagram showing the results of qualitative analysis of particle morphology determined by cryo-EM (n > 80 for each sample).

Figures 3a-b are about biophysical property analysis and control for stable and efficient formation of lipoplexes through control of N/P charge ratio of each cationic liposome.

Specifically, Fig. 3a is a schematic diagram showing the effect of N/P charge ratio on the colloidal stability of lipoplexes, which can be inferred from the particle size (black line) and net surface charge (red line).

Figure 3b shows the results of DLS and zeta potential analyses to evaluate the size distribution, surface charge, and colloidal stability of lipoplexes. The diagram shows representative structural features present in each cationic liposome formulation. Note: The shaded region in the graph indicates the N/P charge ratio associated with unstable lipoplex formation. All data are expressed as mean ± standard deviation (n = 3).

Figures 4a-c show the results of confirming the in vitro transfection efficiency and in vivo biological efficacy of lipoplexes according to the N/P charge ratio of each cationic liposome.

Specifically, Fig. 4a is a graph showing the relative luciferase expression after 24 h in HEK293T cells treated with mRNA-lipoplexes (100 and 200 ng mRNA/well, n = 6). The orange region highlighted in the graph indicates the N/P charge ratio associated with unstable lipoplex formation determined through the biophysical analysis of Fig. 3 . The horizontal dashed line indicates the value where the expression level of 100 ng of E400-3 was set to 1.0 arbitrary units (a.u.). All data are presented as mean ± standard deviation (S.D.).

Figure 4b is a bioluminescence image of BALB/c mice after intravenous injection of lipoplexes made of cationic liposomes with various charge ratios.

Figure 4c is a bioluminescence image of BALB/c mice after intravenous injection of lipoplexes made of anionic liposomes with various charge ratios. The pie charts in Figures 4b and c show the proportion of each organ in the total signal.

Figure 5 is a representative cryo-EM micrograph of (A) cationic and (B) anionic lipoplexes selected based on net charge and in vitro transfection efficiency. The schematic on the left shows representative structural features of the initial cationic liposomes for each lipoplex.

Figure 6 is a diagram showing a comparison of the self-assembly structures of E400 and LC-based lipoplexes. (A) Schematic of a monolayer E400 and multilamellar LC with an ordered structure of cationic and anionic lipoplexes highlighting the interlayer spacing. (B) Representative cryo-EM and FFT images of the LC showing a frequency of 8.89 nm/c (bilayer repetition). (C) Interlayer spacing of E400 and LC analyzed using pixel intensity profiles of cryo-EM images (n > 20 per sample, mean ± SD). (D) Representative cryo-EM images of cationic lipoplexes highlighting the ordered structures, corresponding FFT images, and pixel intensity profiles for repeat distance analysis. (E) Interlayer spacing of cationic lipoplexes analyzed from cryo-EM images (n > 20 per sample, I: LC-6 interior, O: LC-6 exterior). (F) SAXS profile of cationic lipoplexes with d-spacing calculated from peak positions. (G) Representative cryo-EM images of anionic lipoplexes highlighting the ordered structures, corresponding FFT images, and pixel intensity profiles for repeat distance analysis. (H) Interlayer spacing of anionic lipoplexes analyzed from cryo-EM images (n > 20 per sample). (I) SAXS profile of anionic lipoplexes with d-spacing calculated from peak positions. Data are reported as mean ± standard deviation, with a P value of <0.05 considered statistically significant (*p<0.05, **p<0.01, ***p<0.001).

Figure 7 is a cryo-electron microscopy (cryo-EM) image of DOTAP/DOPE liposomes prepared by the liposome preparation method (LUCA cycle) according to the present invention with DHPC at a ratio of 0.5 (L0.5). As can be seen in the image, a significant amount of DHPC micelles (indicated by black arrows) are present along with unilamellar and multivesicular liposomes.

Figure 8 is a diagram showing the in vitro transfection efficiency of VOR and E50-based lipoplexes at various N/P charge ratios in each cationic liposome (100 and 200 ng mRNA/well, n = 6). The orange highlighted region in the graph represents the N/P charge ratio associated with unstable lipoplex formation determined through the biophysical analysis of Figure 3 . The horizontal dashed line represents the expression level of E400-3 at 100 ng, which was set to 1.0 arbitrary units (a.u.). All data are presented as mean ± standard deviation (S.D.).

Figure 9 shows bioluminescence images of BALB/c mice after intravenous cationic E400-6 administration. Pie charts represent the relative contribution of each organ to the total signal.

Figures 10A and 10B show representative analytical results for comparison of self-assembled structures between E400 (Figure 10A) and LC-based lipoplexes (Figure 10B). In Figures 10A and 10B, (A) is a representative cryo-EM micrograph providing a macroscopic view of the overall particle structure. (B) is a magnified image of the red boxed area in panel (A) highlighting the ordered structure of the lipoplexes. (C) is the result of repeat distance analysis of the Fourier transform image of panel B. (D) is the pixel intensity profile of the blue square area in panel B, where lower intensities indicate darker regions with higher lipid density in the image.

Figure 11 is a diagram comparing the interlayer spacing of LC and LC-based lipoplexes analyzed using pixel intensity profiles in cryo-EM images (n > 20 for each sample, mean ± S.D.). I and O represent the inner and outer portions of LC-6 lipoplexes, respectively. A p value of <0.05 was considered statistically significant (*p<0.05, **p<0.01, ***p<0.001).

Hereinafter, the present invention will be described in more detail.

The present invention is characterized by producing cationic liposomes and lipoplexes using a precisely controlled freeze-thaw based method, the Liposome under Cryo-Assembly (LUCA) cycle.

In one aspect, the present invention relates to a method for preparing a lipoplex comprising RNA comprising the following steps.

(a) a step of preparing a lipid solution by dissolving a long-chain cationic lipid and a short-chain lipid in an organic solvent, and then drying the solution under a nitrogen gas flow to evaporate the organic solvent and obtain a dried lipid film;

(b) a step of preparing a lipid suspension by hydrating the dried lipid film obtained in the above step with an aqueous solvent; and

(c) A step of preparing a cationic liposome by immersing the hydrated lipid suspension prepared in the above step in liquid nitrogen, thawing it in a water bath at 10 to 80°C, and then vortexing it for 5 to 60 seconds, repeating this process 1 to 10 times.

In another aspect, the present invention relates to a method for preparing a lipoplex comprising RNA using the method for preparing the cationic liposome.

Specifically, the method for manufacturing the above lipoplex comprises the following steps:

In another aspect, the present invention provides a method for preparing a lipoplex comprising RNA, comprising the following steps:

(i) a step of preparing a lipid solution by dissolving a long-chain cationic lipid and a short-chain lipid in an organic solvent, and then drying the solution under a nitrogen gas flow to evaporate the organic solvent and obtain a dried lipid film;

(ii) a step of preparing a lipid suspension by hydrating the dried lipid film obtained in the above step with an aqueous solvent;

(iii) a step of preparing a cationic liposome by immersing the hydrated lipid suspension prepared in the above step in liquid nitrogen, thawing it in a water bath at 10 to 80°C, and then vortexing it for 5 to 60 seconds, repeating the process 1 to 8 times; and

(iv) a step of diluting RNA with an aqueous solvent to prepare an RNA solution, adding the cationic liposome prepared in step (iii) thereto and stirring to prepare a lipoplex containing RNA.

The method for producing a cationic liposome or a method for producing a lipoplex containing RNA according to the present invention is characterized by using a freeze-thaw-based method unique to the present invention followed by a thin-film hydration method.

The above step (a) or step (i) is a step for obtaining a dried lipid film according to a thin film hydration method, and is performed by dissolving a long-chain cationic lipid and a short-chain lipid in an organic solvent to prepare a lipid solution, and then drying under a nitrogen gas flow to evaporate the organic solvent to obtain a dried lipid film.

In the above steps, dissolving the long-chain cationic lipid and the short-chain lipid in the organic solvent can be performed by heating. For example, heating can be performed at 35°C to 55°C, preferably 40°C to 50°C.

In the present invention, "cationic lipid" refers to a lipid having a net positive charge. Cationic lipids bind to negatively charged RNA by electrostatic interaction with the lipid matrix. Generally, cationic lipids have lipophilic moieties such as sterol, acyl or diacyl chains, and the head group of the lipid is typically positively charged. In the present invention, the cationic lipid is intended to be used for capturing anionic hexane and forming a framework for forming liposomes.

In addition, the long-chain cationic lipid used in the present invention is a long-chain lipid that exhibits cationicity and may be a long-chain phospholipid. For example, 1,2-di-O-octadecenyl-3-trimethylammonium propane (DOTMA), dimethyldioctadecylammonium (DDAB); 1,2-dioleyl-3-trimethylammonium propane (DOTAP); 1,2-dioleyl-3-dimethylammonium-propane (DODAP); 1,2-diacyloxy-3-dimethylammonium propane; 1,2-dialkyloxy-3-dimethylammonium propane; The compound may be at least one selected from the group consisting of dioctadecyldimethyl ammonium chloride (DODAC), 2,3-di(tetradecoxy)propyl-(2-hydroxyethyl)-dimethylazanium (DMRIE), 1,2-dimyristoyl-sn-glycero-3-ethylphosphocholine (DMEPC), l,2-dimyristoyl-3-trimethylammonium propane (DMTAP), 1,2-dioleyloxypropyl-3-dimethyl-hydroxyethyl ammonium bromide (DORIE), and 2,3-dioleyloxy-N-[2(spermine carboxamide)ethyl]-N,N-dimethyl-l-propanamium trifluoroacetate (DOSPA), but is not limited thereto. DOTMA, DOTAP, DODAC or DOSPA is preferred. In certain embodiments, the one or more cationic lipids are DOTMA and/or DOTAP.

The content of long-chain cationic lipid used in the preparation of cationic liposomes or lipoplexes according to the present invention is 0.1 to 100 mM, but is not limited thereto.

The short-chain lipids used in the present invention can affect the shape and stability of liposomes by controlling the properties of the liposomes produced. Such short-chain lipids can be, for example, short-chain phospholipids, and specifically, 1,2-dipropionyl-sn-glycero-3-phosphocholine (3:0 PC), 1,2-dibutyryl-sn-glycero-3-phosphocholine (4:0 PC), 1,2-dipentanoyl-sn-glycero-3-phosphocholine (5:0 PC), 1,2-dihexanoyl-sn-glycero-3-phosphocholine (6:0 PC, DHPC), It may be at least one selected from the group consisting of 1,2-diheptanoyl-sn-glycero-3-phosphocholine (7:0 PC, DHPC) and 1,2-dioctanoyl-sn-glycero-3-phosphocholine (8:0 PC), but is not limited thereto. Preferably, the short-chain lipid may be DHPC.

The content of short-chain lipids used in the preparation of cationic liposomes or lipoplexes according to the present invention is 0.1 to 100 mM, but is not limited thereto.

In another embodiment, the lipid solution can be prepared by including additional lipids to adjust the overall positive to negative charge ratio and physical stability of the cationic liposomes being prepared, to increase the strength and stability of the liposomes, and to promote membrane fusion. In a specific embodiment, the additional lipid is a neutral lipid. As used herein, "neutral lipid" refers to a lipid having a net charge of zero. Examples of neutral lipids include, but are not limited to, 1,2-di-(9Z-octadecenoyl)-sn-glycero-3-phosphoethanolamine (DOPE), 1,2-dioleyl-sn-glycero-3-phosphocholine (DOPC), diacylphosphatidylcholine, diacylphosphatidyl ethanolamine, ceramides, sphingomyelins, cephalins, cholesterol, and cerebrosides. In certain embodiments, the additional lipid is DOPE, cholesterol and/or DOPC.

In an exemplary embodiment, the long-chain cationic lipid is DOTAP, the short-chain lipid is DHPC, and the additional lipid is DOPE.

The content of additional lipid used in the preparation of cationic liposomes or lipoplexes according to the present invention is 0.1 to 100 mM, but is not limited thereto.

The above long-chain cationic lipid and short-chain lipid, or the long-chain cationic lipid and additional lipid and short-chain lipid can be mixed in an appropriate ratio as needed.

In some embodiments, the molar ratio of long-chain cationic lipid to short-chain lipid is from 4:1 to 1:2, or from 3:1 to about 1:1. In exemplary embodiments, the molar ratio of the cationic lipid to short-chain lipid is 2:1 or 1:1.

In some embodiments, when including an additional lipid, the molar ratio of long-chain cationic lipid to the additional lipid is from about 4:1 to about 1:2, or from about 3:1 to about 1:1. In an exemplary embodiment, the molar ratio of the cationic lipid to the additional lipid is about 1:1.

The organic solvent used in the above step (a) or (i) may be used without limitation as long as it can dissolve all of the long-chain cationic lipids, short-chain lipids, and additional lipids to prepare a lipid solution. As a non-limiting example, the organic solvent may be one or a combination of two or more selected from the group consisting of ethanol, methanol, dichloromethane, chloroform, tetrahydrofuran, ethyl acetate, and ether as volatile organic solvents.

In the above step (a), the lipid solution prepared by mixing the lipids and the organic solvent is gently dried under a nitrogen gas flow, and then the solvent is evaporated to obtain a dry film. At this time, the flow rate of nitrogen can be 2 to 20 L/min.

In a specific embodiment according to the present invention, the lipid solution was gently dried under a nitrogen gas flow and then incubated overnight in a vacuum desiccator to evaporate the solvent.

The above step (b) or step (ii) is a step of preparing a lipid suspension by hydrating the dried lipid film obtained in step (a) or (i) with an aqueous solvent. The aqueous solvent may be, but is not limited to, sterile purified water, distilled water, physiological saline solution, or water for injection. The aqueous solvent may be used in an appropriate amount adjusted to be sufficient to hydrate the dried lipid film.

The step (c) or step (iii) is a step of obtaining cationic liposomes from the hydrated lipid suspension prepared in the step (b) or (ii), characterized in that the lipid suspension is prepared using a precisely controlled freeze-thaw-based method, Liposome under Cryo-Assembly (LUCA) cycle.

Specifically, it can be performed by a process of immersing the lipid suspension in liquid nitrogen, thawing it in a water bath, and then vortexing it, repeating the process 1 to 8 times.

The temperature of the water bath during the above thawing can be 10 to 80°C or 60°C.

The vortexing time after thawing can be 5 to 60 seconds, or 30 seconds.

It is preferable to perform this process of liquid nitrogen immersion, thawing and vortexing 1 to 8 times, or 5 times.

By precisely controlling the freeze-thaw-based cycles in this way, the cationic liposomes produced according to the present invention have a unique concentric multilamellar structure that has not been reported in conventional liposome systems. In general, the freeze-thaw method reduces the lamellae to produce a single-lamellar structure, and in order to have a dense and highly ordered concentric multilamellar structure, the presence of an auxiliary surfactant having a shear stress and/or a high salt concentration is generally required. However, the production method of the present invention allows the cationic liposomes to have a concentric multilamellar structure without the use of such shear stress and/or the use of an auxiliary surfactant having a high salt concentration.

In addition, the step (iv) is a step for preparing a lipoplex containing RNA using the cationic liposome prepared in the step (iii), and is specifically performed by diluting RNA with an aqueous solvent to prepare an RNA solution, adding the cationic liposome prepared in the step (iii) thereto, and stirring to prepare a lipoplex containing RNA.

The “aqueous solvent” used in the above step may be, but is not limited to, sterile purified water, distilled water, saline solution or water for injection, similar to step (b) or step (ii). The aqueous solvent may be appropriately adjusted to be used in a sufficient amount to dilute RNA.

The term "RNA" as used herein relates to a nucleic acid molecule comprising ribonucleotide residues. In a preferred embodiment, the RNA contains all or most ribonucleotide residues. As used herein, "ribonucleotide" refers to a nucleotide having a hydroxyl group at the 2'-position of the β-D-ribofuranosyl group. RNA includes, without limitation, double-stranded RNA, single-stranded RNA, isolated RNA, such as partially purified RNA, essentially pure RNA, synthetic RNA, recombinantly produced RNA, and modified RNA that differs from naturally occurring RNA by addition, deletion, or substitution of one or more nucleotides. Such alterations may refer to the addition of non-nucleotide material within the RNA nucleotides or to the terminus(s) of the RNA. It is also contemplated that the nucleotides in RNA may be non-standard nucleotides, such as chemically synthesized nucleotides or deoxynucleotides. For the purposes of this specification, such modified RNA is considered an analogue of naturally occurring RNA. In certain embodiments of the present disclosure, the RNA is messenger RNA (mRNA), which is an RNA transcript encoding a peptide or protein. As is well known in the art, mRNA generally contains a 5' untranslated region (5'-UTR), a peptide coding region, and a 3' untranslated region (3'-UTR). In some embodiments, the RNA is produced by in vitro transcription or chemical synthesis. In one embodiment, the mRNA is produced by in vitro transcription using a DNA template, where DNA refers to a nucleic acid containing deoxyribonucleotides. In one embodiment, the RNA is in vitro transcribed RNA (IVT-RNA) and can be obtained by in vitro transcription of a suitable DNA template. The promoter for controlling transcription can be any promoter for any RNA polymerase. The DNA template for in vitro transcription can be obtained by cloning a nucleic acid, particularly a cDNA, and introducing it into a suitable vector for in vitro transcription. cDNA can be obtained by reverse transcription of RNA.

In certain embodiments of the present disclosure, the RNA used in the manufacturing method of the present invention is at a concentration of about 0.01 mg/mL to 1 mg/mL, or about 0.05 mg/mL to about 0.5 mg/mL, relative to the total lipoplex being manufactured. In certain embodiments, the RNA is present at about 0.05 mg/mL, about 0.06 mg/mL, about 0.07 mg/mL, about 0.08 mg/mL, about 0.09 mg/mL, about 0.10 mg/mL, about 0.11 mg/mL, about 0.12 mg/mL, about 0.13 mg/mL, about 0.14 mg/mL, about 0.15 mg/mL, about 0.16 mg/mL, about 0.17 mg/mL, about 0.18 mg/mL, about 0.19 mg/mL, about 0.20 mg/mL, about 0.21 mg/mL, about 0.22 mg/mL, about 0.23 mg/mL, about 0.24 mg/mL, about 0.25 mg/mL, about 0.26 mg/mL, about 0.27 mg/mL, about at a concentration of about 0.28 mg/mL, about 0.29 mg/mL, about 0.30 mg/mL, about 0.31 mg/mL, about 0.32 mg/mL, about 0.33 mg/mL, about 0.34 mg/mL, about 0.35 mg/mL, about 0.36 mg/mL, about 0.37 mg/mL, about 0.38 mg/mL, about 0.39 mg/mL, about 0.40 mg/mL, about 0.41 mg/mL, about 0.42 mg/mL, about 0.43 mg/mL, about 0.44 mg/mL, about 0.45 mg/mL, about 0.46 mg/mL, about 0.47 mg/mL, about 0.48 mg/mL, about 0.49 mg/mL, or about 0.50 mg/mL. In an exemplary embodiment, the RNA is at a concentration of 0.05 mg/mL.

As another specific example, the present invention is characterized in that the charge ratio is 1 to 7 for achieving stable and efficient lipoplex formation in the step (iv). In the present invention, the charge ratio (N/P) represents the molar ratio of the positively charged amine group (N = nitrogen) of the cationic lipid and the negatively charged phosphate group (P) in RNA. The N/P ratio determines the charge ratio and the overall charge of the lipoplex.

The charge ratio (N/P) of the lipoplex manufactured according to the present invention may be 1 to 7, preferably 1.5 to 6.

The present inventors have confirmed that when the charge ratio of the lipoplex is adjusted in the manufacturing method of the lipoplex, a drug can be selectively delivered to a desired organ. In a specific example, the present inventors confirmed that when the N/P of the lipoplex is adjusted, a cationic lipoplex having an N/P ratio of 3 showed luciferase expression mainly in the spleen, a cationic lipoplex having an N/P ratio of 6 showed luciferase expression in both the lungs and the spleen, and an anionic lipoplex showed spleen-specific targeting ability. When the N/P of the lipoplex is adjusted according to the manufacturing method of the present invention, a desired drug can be delivered to a desired organ, preferably the lung or spleen, and more preferably the lung, in an organ-specific manner. This is significantly different from the conventional neutral lipid nanoparticles (LNPs) that mainly target and deliver the drug to the liver. That is, while neutral lipid nanoparticles primarily target the liver, lipoplexes manufactured by the manufacturing method according to the present invention can target other specific organs by adjusting the N/P charge ratio. Cationic lipoplexes have shown the ability to target the lung or both the lung and spleen, whereas anionic lipoplexes specifically target the spleen. This phenomenon may lead to the development of vaccines targeting splenic dendritic cells, for example. In addition, similar surface charge-dependent selective organ targeting (SORT) LNPs have been designed by incorporating charged lipids into existing LNPs.

The lipoplex manufactured according to the method for manufacturing a lipoplex of the present invention has a particle size of 200 nm to 2000 nm, 200 nm to 1000 nm, 200 nm to 700 nm, 300 nm to 2000 nm, 300 nm to 1000 nm, 300 nm to 700 nm, preferably about 300 nm to 500 nm.

In addition, the polydispersity index (PDI) of the lipoplex manufactured according to the method for manufacturing the lipoplex of the present invention is 0 to 0.4, preferably 0.1 to 0.4. In addition, the zeta potential of the lipoplex of the present invention is 10 to 60 mV, preferably about 30 to 50 mV.

Additionally, the lipoplex manufactured according to the lipoplex manufacturing method of the present invention exhibits a Q ratio of 0.2 to 8.0, 0.3 to 6.0, 0.4 to 5.0, 0.5 to 4.0, 0.5 to 3.0, or 0.5 to 2.0.

The term “Q ratio” refers to the molar ratio of long-chain lipids to short-chain lipids in isotropic bicelles, and in the present invention, represents the molar ratio of the sum of long-chain cationic lipids (the sum of long-chain cationic lipids and additional lipids when additional lipids are further included) to the sum of all short-chain lipids in a lipoplex. In one embodiment of the present invention, DOTAP was used as the long-chain cationic lipid, DOPE was used as the additional neutral lipid, and DHPC was used as the short-chain lipid for the production of lipoplexes, and in this case, the Q ratio is the molar ratio of the sum of DOTAP and DOPE/DHPC. In a specific embodiment of the present invention, the structural characteristics of the lipoplex according to the Q ratio were confirmed. As a result, it was confirmed that when the Q ratio is infinite (i.e., only long-chain cationic lipids are present, LC), it contains the most multilamellar structures, but as the Q ratio approaches 2 (L-2.0) (i.e., as short-chain lipids are included relatively more), the proportion of multilamellar structures decreases. When more multilamellar structures are included, more RNA can be loaded, but if too much drug is loaded, it may have the disadvantage of making the lipoplex structure unstable.

Therefore, in the present invention, when the lipoplex maintains this Q ratio, the drug delivery efficiency and safety of the lipoplex can be improved, and the multilamellar structure is affected.

According to the manufacturing method of the present invention, cationic liposomes can be manufactured to have a concentric multilamellar structure without using shear stress and/or using an auxiliary surfactant having a high salt concentration. In addition, by having a specific Q ratio during the manufacture of the lipoplex, the lipoplex can efficiently carry a large amount of drugs such as mRNA, etc., compared to conventional lipoplexes, while maintaining structural stability. In addition, by controlling the N/P ratio, the lipoplex has the effect of effectively delivering drugs such as mRNA to a desired organ by targeting the targeted location.

In one embodiment of the present invention, it was confirmed that lipoplexes prepared according to the present invention exhibited distinct structural differences compared to conventional lipoplexes when stored at 4°C, exhibited consistent colloidal stability for one week, and exhibited low cytotoxicity and effective in vitro transfection.

The present invention will be described below through examples. However, the examples are merely introduced to illustrate preferred embodiments of the present invention, and the scope of the present invention should not be limited thereto.

Example

The development of mRNA delivery systems using lipid-based assemblies holds enormous potential for precise control of gene expression and targeted therapeutic intervention. Despite remarkable advances in lipid-based gene delivery systems, there remains a significant knowledge gap in understanding the biophysical properties of lipid assemblies and the impact of mRNA complexes on these systems. In this study, we investigated the biophysical properties of cationic liposomes and their impact on mRNA-lipoplex formation by comparing different manufacturing methods. Notably, a novel manufacturing technique called Liposome under Cryo-Assembly (LUCA) cycle involving a precisely controlled freeze-thaw-vortexing process produces a unique onion-like concentric multilamellar structure in cationic DOTAP/DOPE liposomes, in contrast to the unilamellar liposomes produced by conventional extrusion methods. Inclusion of short-chain DHPC lipids further modifies the structure of cationic liposomes, converting them from multilamellar to unilamellar structures during the LUCA cycle. Furthermore, biophysical and biological evaluations of mRNA-lipoplexes revealed that the optimal N/P charge ratio in lipoplexes can vary depending on the structure of the initial cationic liposomes. Cryo-EM structural analysis demonstrates that multilamellar cationic liposomes induce two distinct lamellar spacings in the cationic lipoplexes, highlighting the significant influence of liposome structure on the final structure of mRNA-lipoplexes. Taken together, the present invention provides an interesting relationship between the structure of lipid assemblies and the biophysical properties of the resulting lipoplexes. This relationship may open the way to developing lipid-based mRNA delivery systems through more streamlined manufacturing processes.

Materials and Methods

Reagents

1,2-dioleoyl-3-trimethylammonium-propane (chloride) (DOTAP), 1,2-di-(9Z-octadecenoyl)-sn-glycero-3-phosphoethanolamine (DOPE), and 1,2-dihexanoyl-sn-glycero-3-phosphocholine (DHPC) lipids were purchased in chloroform from Avanti Polar Lipids (Alabaster, AL). CleanCap Firefly Luciferase mRNA (5 moU) was purchased from TriLink BioTechnologies (San Diego, CA).

Mice

BALB/c mice were purchased from InVivos. Mice were housed and maintained in a pathogen-free environment, strictly following institutional protocols established by the Biological Resource Centre (BRC), A*STAR, Singapore. All procedures involving animals were approved by the Institutional Animal Care and Use Committee (IACUC# 191452).

Liposome Preparation

Positively charged liposomes were prepared by thin-film hydration followed by extrusion or LUCA cycle. First, appropriate amounts of long-chain cationic DOTAP, neutral DOPE, and short-chain DHPC lipids dissolved in chloroform were added to a glass vial, gently dried under a stream of nitrogen gas, and incubated in a vacuum desiccator overnight to evaporate the solvent. Next, the dried lipid film was rehydrated with Milli-Q treated water (MilliporeSigma, Burlington, MA) or UltraPure DNase/RNase-free distilled water (Thermo Fisher Scientific, Waltham, MA) and vortexed. To prepare E50 and E400 liposomes, the resulting suspensions were extruded using track-etched polycarbonate filter membranes with 50 or 400 nm diameters, respectively. To prepare LC, L4.0, and L2.0 liposomes, the hydrated suspensions were subjected to five LUCA cycles including the following steps: (1) immersion in liquid nitrogen for 1 min, (2) thawing in a 60 °C water bath for 5 min, and (3) vortexing for 30 s. The concentrations of DOTAP and DOPE were fixed at 10 mM unless otherwise specified.

Lipoplex Preparation

Lipoplexes were prepared by slightly modifying the protocol described in a previous study (Kranz, LM et al., Systemic RNA delivery to dendritic cells exploits antiviral defense for cancer immunotherapy. Nature 2016, 534 (7607), 396-401 and Barichello, JM et al., Complexation of siRNA and pDNA with cationic liposomes: the important aspects in lipoplex preparation. Liposomes: Methods and Protocols, Volume 1: Pharmaceutical Nanocarriers 2010 , 461-472). A diversity of formulations complexed with the reporter firefly luciferase (Luc)-encoding mRNA was assembled with cationic liposomes to create various N/P ratios, which defined the charge ratio and overall lipoplexes net charge. Lipoplexes with various N/P ratios were prepared by combining mRNA expressing firefly luciferase (Luc) with various combinations of cationic liposomes. This N/P ratio determines the charge ratio and overall charge of the lipoplex. The charge ratio was calculated from the molar ratio of the positively charged amine (N = nitrogen) groups to the negatively charged phosphate (P) groups presented by the mRNA nucleotides. An average molar mass of 330 Da per nucleotide was assumed to calculate the molar ratio between the cationic lipid and mRNA. The mRNA was provided dissolved in sodium citrate buffer at a mRNA concentration of 1 mg/mL. Lipoplexes were prepared by diluting the mRNA with ultrapure DNase/RNase-free distilled water and adding an appropriate amount of cationic liposome dispersion to achieve the selected charge ratio.

Dynamic Light Scattering (DLS) and Zeta Potential Measurements

Size distribution and zeta potential were investigated using a 90Plus particle size/zeta PALS analyzer (Brookhaven Instruments Corporation, NY). For size distribution analysis, measurements based on dynamic light scattering were performed at a scattering angle of 90° to minimize reflection effects. All obtained autocorrelation functions were analyzed by the cumulative method and fitted with a multimodal distribution to obtain size distributions.

Cryogenic Electron Microscopy (Cryo-EM)

Samples for cryo-EM imaging were prepared by glow-discharge treatment of lacey carbon-coated 300-mesh copper grids (Electron Microscopy Sciences, Hatfield, PA). 4 μL of sample solution was applied onto the grids at 100% humidity, blotted with filter paper (2 s blotting time, 0 blotting force), and immersed in liquid ethane (Vitrobot, FEI Company). Cryogrids were imaged using a FEG 200 keV transmission electron microscope (Arctica, FEI Company) equipped with a direct electron detector (Falcon II, FEI Company). Images were recorded at a nominal magnification of ×53,000 and an integration time (exposure time) of 1 s. Fast Fourier transforms (FFTs) of cryo-EM images were obtained using ImageJ software.

In vitro Luciferase Transfection Assay

HEK293T cells (ATCC, Manassas, VA) were cultured in Dulbecco's modified Eagle's medium (Gibco) supplemented with 10% heat-inactivated fetal bovine serum (Gibco) and 1% penicillin/streptomycin (Hyclone). All cultures were grown at 37°C in a 5% CO ₂ incubator and maintained according to the supplier's instructions. For assays, cells were seeded at a density of 5,000 per well in 96-well plates and allowed to attach overnight. mRNA-lipoplexes were then added to the cell culture medium in which the cells were cultured at 100 ng and 200 ng of mRNA per well. Luciferase expression data were collected 24 hours after treatment using a PierceTM Firefly Luciferase Glow Assay Kit (Thermo Fisher Scientific, Waltham, MA) using a microplate reader and normalized to the expression level of 100 ng of E400-3 according to the supplier's instructions.

In vivo Bioluminescence Imaging

Five to eight weeks old male BALB/c mice were injected intravenously (i.v.) with 10 μg of FLuc mRNA complexed with lipoplexes via the lateral tail vein. Six hours after lipoplex administration, mice were injected intraperitoneally with 150 mg/kg D-luciferin (Thermo Fisher Scientific, Waltham, MA) dissolved in phosphate-buffered saline (PBS), and were allowed to sit for 10 minutes to allow complete systemic circulation of luciferin before performing in vivo luminescence imaging. Twenty-four hours later, mice were reinjected with 150 mg/kg D-luciferin in PBS, allowed to sit for 10 minutes, and then liver, lungs, and spleen were removed for luminescence imaging. Bioluminescence image acquisition was performed using an IVIS® Spectrum In Vivo Imaging System (PerkinElmer, Shelton, CT). All image postprocessing and analysis were performed using Living Image® 4.8.0 software (PerkinElmer, Shelton, CT). Bioluminescence intensity was quantified by drawing regions of interest (ROIs) around the liver, lungs, and spleen and measuring the luminescence within each ROI.

Results and Discussion

Experimental design

As depicted in Figure 1a, we started by testing cationic liposomes and lipoplexes using two different fabrication methods, with a focus on elucidating their biophysical properties. Initially, we chose extrusion, one of the most commonly used methods for producing unilamellar liposomes with a monodisperse size distribution. Despite the time-consuming multi-step process, this technique is still widely used due to its excellent reproducibility, homogeneous size distribution, and ability to bypass the removal step of organic solvents or surfactants. Furthermore, the fabrication method according to the present invention introduces a novel fabrication technique called Liposome under Cryo-Assembly (LUCA) cycle, which is based on a freeze-thaw process. Through repetition of this cycle, the formation of ice crystals expands the inner liquid phase and disrupts the lamellar structure of the liposomes, resulting in smaller and more homogeneous lipid self-assembly populations.

For biophysical characterization, the size distribution and surface charge of the lipid assemblies were analyzed using dynamic light scattering (DLS) and zeta potential measurements, respectively. These parameters are important for understanding the self-assembly dynamics of cationic liposomes and mRNA, and the size and surface charge of the formed lipoplexes are affected by the intrinsic properties of the initial cationic liposomes and the charge ratio between the cationic lipids and mRNA. In addition, the structural characteristics, including lamellae and morphology, were investigated using cryo-electron microscopy (cryo-EM), and the self-assembled structures can show considerable variations depending on the changes in parameters. For biological evaluation, the in vitro transfection efficiency and in vivo biodistribution of luciferase-encoding mRNA within the lipoplexes were measured to elucidate how the biophysical properties of liposomes and lipoplexes affect the function of mRNA-lipoplexes.

To prepare cationic liposomes, DOTAP and DOPE lipids were chosen as the basic lipid composition based on their proven success in nucleic acid delivery ( Figure 1b ). The DOTAP/DOPE system has been studied for decades and presents a potential mechanism for nucleic acid delivery, where membrane fusion is facilitated by the reversed hexagonal phase and the reduced bending modulus of the unsaturated portion within this complex. Nevertheless, establishing a direct correlation between these properties and biological efficacy remains difficult, highlighting the need for further exploration in lipoplex systems. In addition, short-chain DHPC lipids were added to control the size and structure of the cationic liposomes. DHPC, which is commonly used in disc-shaped bilayer mixtures containing long-chain saturated lipids, disrupts the spherical assembly of liposomes and stabilizes the bilayer discs through edge interactions. However, when combined with long-chain unsaturated lipids such as DOPC, DHPC destabilizes the lipid bilayer due to its high curvature preference, leading to structural deformations including vesiculation and formation of smaller aggregates. Therefore, DHPC can be used as a tool to control the size distribution and structure of liposomes.

Thus, six different types of cationic liposomes were used for lipoplex synthesis with different N/P charge ratios ( Figure 1c ). The vortexed liposomes, denoted as VOR, represent large liposomes with broad size distribution and random structure. E50 and E400, prepared by conventional extrusion method, represent liposomes extruded to 50 nm and 400 nm, respectively. E50 was designed to exhibit a unilamellar morphology with small and uniform size distribution, whereas E400 was expected to have a larger size distribution that is controlled by extrusion. Similarly, cationic liposomes were prepared by applying the preparation method according to the present invention (LUCA cycle) and were designated as LC. To further control the size and structure of LC, short-chain DHPC lipids were subsequently added to the LC formulation. Considering the important role of q-ratio in forming bilayer phase, liposomes containing DHPC were prepared with q-ratios of 4.0 and 2.0, respectively, and were denoted as L4.0 and L2.0.

Biophysical properties of cationic liposomes.

Size distribution . The size distribution of the liposomes was first analyzed by DLS technique. After film formation and hydration, the VORs exhibited a broad size distribution with two distinct multimodal peaks at around 180 nm and 760 nm ( Figure 2a ). The average diameter of the VORs measured from multiple samples was 393.9 ± 28.9 nm, consistent with the polydispersity index (PDI) of 0.34, reflecting their less reproducible and heterogeneous nature ( Figure 2b , Table 1 ). These results are consistent with the expectation that hydration of dried phospholipids in aqueous media leads to uncontrolled self-assembly of heterogeneous bi- and/or multilamellar liposomes with a broad size distribution.

[Table 1]

Average diameter, PDI, and zeta potential values of cationic liposomes used in this experiment (n = 6, mean ± S.D.).

Subsequent extrusion through a 50 nm filter resulted in the formation of E50, which was characterized by a very narrow size distribution with a mean diameter of 73.7 ± 8.3 nm and a PDI of 0.13. Notably, E50 exhibited a single narrow peak in multimodal analysis, indicating its homogeneity. On the other hand, E400 exhibited a relatively broader size distribution and dual multimodal peaks compared to E50, which was presumably due to some degree of heterogeneity caused by its larger pore size. Nevertheless, E400 exhibited a smaller and more homogeneous size distribution compared to VOR, suggesting the potential advantage of the extrusion technique in producing liposomes with consistent properties.

The average diameter of LCs manufactured by introducing the manufacturing method (LUCA cycle) according to the present invention was 321.6 ± 20.6 nm, which corresponds to a size between VOR and E400, and surprisingly exhibited a low PDI of 0.20. This means that when liposomes are manufactured by the manufacturing method according to the present invention, the freeze-thaw mechanism effectively reduces the size and size distribution of liposomes, which may occur due to lipid fragmentation and subsequent lipid membrane rupture by ice crystals during the freezing process. Interestingly, the addition of short-chain DHPC lipids resulted in a broadening of the size distribution along with a slight decrease in liposome size as the PDI increased from 0.20 to 0.22 in L4.0. This effect was more pronounced as the DHPC concentration was doubled in L2.0. This observed size decrease and distribution increase are expected results due to the destabilization and disordering of the lipid membrane induced by the difference in chain length between long-chain and short-chain lipids.

Morphology. Next, the morphology of the cationic liposomes was characterized by cryo-EM ( Figure 2c ). VORs showed heterogeneous sizes and shapes, with some liposomes exhibiting elongated vesicle morphologies encapsulated within larger liposomes. Quantitative analysis revealed that VORs were mainly composed of unilamellar and multivesicular structures, with both types accounting for 44% of the population ( Figure 2d ) . This heterogeneity was closely consistent with the results of DLS analysis. In contrast, extruded liposomes showed a distinct lamellar tendency compared to VORs. E50 was composed entirely of unilamellar structures, indicating that extended extrusion led to the rupture of the multilamellar structures and the generation of small, uniform unilamellar liposomes. In contrast, E400 was mainly unilamellar, with 19% bilamellar structures.

Notably, the LCs according to the present invention exhibited a distinct “onion-like” concentric multilamellar morphology, contrary to expectations, as freeze-thaw methods typically reduce lamellarity to produce single-lamellar structures. Approximately 65% of the LCs exhibited multilamellar structures, suggesting that although the LUCA cycle tends to induce relative homogeneity in the size distribution, it favors the formation of multilamellar structures over single-lamellar structures in DOTAP/DOPE cationic liposomes. Such dense and highly ordered concentric multilamellar structures have been observed previously in liposomes, but generally required the presence of co-surfactants with shear stress and/or high salt concentrations. Considering the system consisting solely of cationic and neutral lipids with similar chain lengths in pure water without salt, it is likely that the interplay between strong electrostatic charge effects and external forces induced by repetitive phase transitions played a role during this unusual assembly process, but further studies are needed for a comprehensive understanding of this phenomenon.

When the short-chain phospholipid DHPC was included, a decrease in the onion-like multilamellar structure was observed. This decrease was more pronounced with increasing DHPC concentration, especially when switching from L4.0 to L2.0. For example, when DHPC was added at the same molar ratio as DOPE (L2.0), the percentage of multilamellar liposomes decreased approximately 10-fold from 65% to 7%, whereas unilamellar liposomes increased 3-fold from 30% to 82% compared to the case without DHPC (LC). Consistent with the DLS results, L4.0 and L2.0 showed more heterogeneous sizes and shapes, such as elongated shapes and multivesicular structures. This change in liposome morphology is likely due to membrane destabilization, which may be induced by the difference in chain length between long- and short-chain lipids. These observations may provide an explanation for the decrease in lamellarity during the LUCA cycle. When DHPC was added at a 4-fold ratio of DOPE (L0.5), numerous mixed micelles appeared along with unilamellar and multivesicular liposomes, which was due to the high concentration of short-chain lipids ( Figure 7 ). Based on these results, L4.0 and L2.0 were selected for comparative analysis with LC to further understand the lipoplex formation and investigate the influence of the initial structure of liposomes on this process.

Biophysical properties of mRNA-lipoplexes.

Next, lipoplexes were systematically prepared by varying the molar ratio between cationic lipids and mRNA to achieve stable and efficient lipoplex formation. Here, the N/P ratio represents the molar ratio of DOTAP (positively charged amine group, N) to nucleic acid (negatively charged phosphate group, P) of mRNA. During the complexation process of cationic liposomes and nucleic acids, liposomes initially bind to nucleic acid molecules, and aggregation begins when the zeta potential of the lipoplex approaches zero ( Figure 3A ). The colloidally unstable region is generally observed near the neutral surface charge, which can lead to potential aggregation of lipoplexes. When the negative or positive charge is excessive, colloidally stable lipoplexes with distinct particle sizes can be prepared due to repulsive electrostatic forces. Since the N/P ratio acts as an important parameter for colloidal stability, particle properties, and target selectivity, it is essential to investigate the effect of the N/P ratio to optimize lipoplex formulations.

In general, at high charge ratios (N/P > 6), predominantly positively charged particles were generated, whereas at low charge ratios (N/P < 1.0), negatively charged lipoplexes were generated. Notably, most of the lipoplexes exhibited near-neutral surface charge and reached their largest size at an N/P ratio of 1.5, where nearly equal molar ratios of cationic lipids and nucleic acids coexist. However, an exceptional case was observed for LC-based lipoplexes, where the largest size was reached at an N/P ratio of 3.0 ( Figure 3B ). This unusual observation in LC-based lipoplexes can be explained by the relatively high fraction (~65%) of multilamellar structures generated by freeze–thaw cycling. A higher N/P ratio was required to neutralize the overall charge when the amount of cationic lipids exposed on the surface was small. Similarly, for VOR and L4.0, relatively large lipoplexes (~1,000 nm) were formed at an N/P ratio of 3.0. This phenomenon can be explained by the relatively low ratio of single- and double-lamellar structures. Interestingly, E50-derived lipoplexes with an N/P ratio of 1.5 (E50-1.5) formed significantly larger aggregates than E400-derived lipoplexes (E400-1.5). This result can be explained by the considerably large surface area of E50 due to the 100% single-lamellar structure with a small diameter of about 70 nm. As the surface charge of the lipoplexes shifted from neutral to negative (at lower N/P ratios), the complexes exhibited enhanced stability and reduced size due to the excess of mRNA that counteracted the positive charge of DOTAP in the cationic liposomes. This trend in size and zeta potential changes was consistent for all lipoplexes. In particular, liposomes with multilamellar or multivesicular structures tended to aggregate at relatively high N/P ratios compared to unilamellar or bilamellar liposomes. This behavior is likely due to the limited exposure of cationic lipids on the surface of multilamellar or multivesicular liposomes. Collectively, these results demonstrate that the region of colloidal instability of lipoplexes is significantly affected by both the initial liposome morphology and the N/P ratio.

Biological efficacy of mRNA-lipoplexes

To evaluate the biological efficacy of the lipoplexes, in vitro cell transfection experiments were performed using HEK293T cells. After 24 h of treatment with lipoplexes, cells were lysed and the relative intensity of luciferase expression was measured. The predominant trend was that lipoplexes with aggregated particles, such as LC-derived lipoplexes with an N/P ratio of 3, such as E400-1.5 and LC-3, consistently exhibited lower transfection efficiencies ( Figure 4A ). Conversely, positively charged adjacent lipoplexes near the aggregation threshold (e.g., E400-3 and LC-6) consistently exhibited higher efficiencies. However, at excessively high N/P ratios (~9.0), transfection efficiencies were reduced in all lipoplexes due to excess cationic lipids and/or inefficient release of mRNA5, 56. Interestingly, different patterns were observed depending on the initial liposome structure within the N/P ratio range of approximately 3.0 to 6.0. For example, L4.0-based lipoplexes with low ratio of unilamellar/bilamellar liposomes exhibited the highest luciferase expression at an N/P ratio of 6.0 and the lowest at an N/P ratio of 3.0. In contrast, L2.0-based lipoplexes exhibited a similar trend in transfection efficiency to E400-based lipoplexes, which was consistent with a high ratio of unilamellar structures similar to E400. Notably, negatively charged LC-based lipoplexes (e.g., LC-1.5 and LC-0.75) exhibited similar luciferase expression levels to cationic lipoplexes such as E400-3 and LC-6, which was different from the results observed with other types of lipoplexes.

VOR-based lipoplexes showed significant variation in luciferase expression levels without a clear trend ( Figure 7 ). This variation is likely due to the variability in complex formation resulting from the heterogeneity of vortexed liposomes. On the other hand, E50-derived lipoplexes showed low luciferase expression at all N/P ratios. Although the size effect of lipoplexes needs to be approached cautiously, the lower transfection efficiency in E50-derived lipoplexes may be related to the smaller size of the lipid assemblies. This hypothesis is supported by the notable performance differences observed between E400-derived and E50-derived lipoplexes. Despite similar liposome morphologies achieved by extrusion, E400-based lipoplexes derived from larger liposomes showed higher transfection efficiency than E50-derived lipoplexes at the same N/P ratio. This highlights the importance of lipid assembly size on the transfection outcome of lipoplexes, especially in lipoplexes derived from extruded liposomes.

Neutral LNPs primarily target the liver, but it has been reported that lipoplexes can be targeted to specific organs by adjusting the N/P charge ratio. Cationic lipoplexes have shown the ability to target the lung or both the lung and spleen, whereas anionic lipoplexes specifically target the spleen. This phenomenon has led to the development of vaccines targeting splenic dendritic cells. In addition, similar surface charge-dependent selective organ targeting (SORT) LNPs have been designed by incorporating charged lipids into conventional LNPs. Based on this background, the inventors evaluated the in vivo biological efficacy and organ-selective biodistribution of cationic and anionic lipoplexes, focusing on lipoplexes prepared by extrusion and LUCA cycles, which showed effective in vitro transfection and structurally distinct differences.

As shown in Fig. 4b , cationic lipoplexes derived from E400 with an N/P ratio of 3, E400-3, mainly expressed luciferase in the spleen, consistent with previous results observed at the same charge ratio. E400-6 with a higher positive charge exhibited enhanced targeting to the lung and spleen, suggesting that increasing the N/P ratio may enhance lung targeting ( Fig. 9 ). Notably, this targeting pattern was sustained in LC-derived lipoplexes, as confirmed by targeting to the lung and spleen by LC-6. In contrast, anionic E400-0.75 and LC-0.75 exhibited clear spleen-specific targeting abilities ( Fig. 4c ). This consistent trend of organ-targeting abilities was observed even after the inclusion of short-chain DHPC, demonstrating the organ-targeting biological efficacy of lipoplexes with diverse structural properties.

It should be emphasized that the in vivo organ targeting ability of lipoplexes is significantly influenced by the N/P ratio. In contrast, the N/P ratio for optimal in vitro transfection efficiency is intricately regulated by the initial liposome morphology. Taken together, these results suggest that the variability of the optimal N/P ratio in lipoplexes depends on the structural characteristics of cationic liposomes, which profoundly influences their biological efficacy. Beyond the influence of lipid composition, this highlights the importance of simultaneously considering both cationic liposome structure and N/P ratio when designing mRNA delivery systems.

Structural characteristics of cationic and anionic lipoplexes

To better understand the influence of the structure of the lipid assemblies on the lipoplex architecture, cryo-EM imaging experiments were performed next. Cationic and anionic lipoplexes were selected based on their net charge and biological efficacy, focusing on the N/P ratio near the aggregation point. For lipoplexes prepared from E50, E400, and L2.0, the neutral charge point was identified at N/P = 1.5. Therefore, cationic and anionic lipoplexes were studied at N/P ratios of 3 and 0.75, respectively. On the other hand, lipoplexes derived from VOR, LC, and L4.0 showed relative instability in the N/P ratio range from 3 to 1.5. Therefore, for these lipoplexes, cationic and anionic cases were selected at N/P ratios of 6 and 0.75, respectively. Notably, the initial liposome structures in the latter cases had relatively lower contents of unilamellar/bilayer structures compared to the former cases. Representative cryo-EM images are presented in Fig. 5 .

In cationic lipoplexes ( Figure 5a ), VOR-6 exhibited high-contrast thick layers and voids, indicating a condensed complex structure formed by the interaction between the nucleic acid and the cationic lipid within the lipoplex. This structural feature was likely influenced by the high proportion of multivesicular structures of VOR (44%), because smaller mRNA molecules may not have a significant effect on the initial liposomal structure of the much larger and more flexible VOR. When anionic VOR-0.75 was prepared with excess mRNA ( Figure 5b ), paired lamellar structures were formed within and between lipoplex particles due to the strong electrostatic interactions between the cationic lipid membrane and the mRNA molecules. This sandwiched lamellar phase produced the characteristic 'fingerprint' pattern commonly observed in DOTAP/DOPE lipoplex systems, but voids were observed within the lipoplexes due to the large, multivesicular liposome structure of VOR.

In contrast, E50-derived lipoplexes exhibited dense lamellar structures when complexed with small (E50-3) or large (E50-0.75) amounts of mRNA. This dense internal structure may be induced by the high surface area of E50. In addition, E50-3 exhibited connected and elongated structures compared to other cationic lipoplexes, which may partly explain the lower transfection efficiency observed with E50-3. Similarly, E400-3 exhibited a series of closely packed lamellar structures, representing the ‘fingerprint’ pattern of lipoplexes. These lipoplexes exhibited larger size and relatively spherical shape compared to E50-3, which indicated higher transfection efficiency. In contrast, E400-0.75 exhibited elongated structures connected by short-length multilamellar complexes. These elongated structures are consistent with those observed in previous studies, showing that excess mRNA under low N/P ratio conditions induces spaghetti-like tubular projections or elongated periodic multilamellar structures— ^{61, 62.} Thus, the elongated structures of E400-0.75 appear to be induced by excess mRNA, contributing in part to the reduced transfection efficiency in vitro.

Surprisingly, the concentric multilamellar structure of LC was remarkably preserved after mRNA incorporation in cationic LC-6. The outer layer of LC-6 was thicker than that of LC, and the inner layer was deformed into a dense state due to binding to mRNA. In lipoplexes, this preservation of liposome structure was observed only when LC was characterized by a densely packed concentric multilamellar structure. This behavior is in sharp contrast to conventional unilamellar liposomes, such as E50 and E400, which tend to deform into distorted complex assemblies upon complexation with nucleic acids. In contrast, anionic LC-0.75 showed a typical lamellar structure upon addition of mRNA, suggesting that strong electrostatic interactions induced by excess mRNA reorganized the concentric multilamellar structure into a fingerprint pattern. Upon addition of short-chain DHPC, L4.0-6 exhibited hybrid structures with both concentric multilamellar and thick layers, which appear to be induced by the mixed multilamellar, unilamellar/bilamellar, and multivesicular morphologies of L4.0. With increasing DHPC content, L2.0-3 exhibited multilamellar structures with fingerprints similar to those of E400-3, which are mainly due to the high proportion of unilamellar structures. In anionic lipoplexes, both L4.0-0.75 and L2.0-0.75 exhibited typical dense fingerprint patterns when exposed to excess mRNA.

In conclusion, our results highlight that both the structure of cationic liposomes and the N/P ratio significantly influence the structure of mRNA-lipoplexes. This highlights the importance of the manufacturing process in controlling the structure of lipid assemblies and the potential for developing manufacturing processes to control the structure of lipid-based assemblies and improve mRNA delivery systems.

Structural comparison of liposomes and lipoplexes produced by extrusion and LUCA circulation methods

To gain a deeper understanding of the morphological architecture of the lipid assemblies, we performed a comparative analysis of the detailed repeat distances within the lamellar layers of representative cationic/anionic lipoplexes from 400 nm-extruded E400 and LUCA cycle LCs ( Figure 6A ). Prior to mRNA complexation, cryo-EM and fast Fourier transform (FFT) images of the LCs showed the lamellar pattern arising from unique concentric multilamellar structures with an interlayer spacing of 9.68 ± 1.61 nm ( Figure 6B, C ). To further analyze the repeat distances of the lamellar layers in the lipoplexes, characteristic lamellar structures were selected from the original cryo-EM images, and then FFT and pixel intensity profile analyses were performed ( Figure 10 ).

When complexed with mRNA in cationic E400-3, a single-sided repeating lamellar structure was observed, with periodic layers of high electron density clearly observed by FFT image analysis and intensity line profiling. In contrast, cationic LC-6 showed distinctly two separate interlayer spacings (the sum of the thickness of the lipid bilayer and the thickness of the mRNA-containing aqueous layer) with dominant frequencies of 7.8 nm/c and 14.4 nm/c, which were supported by line plots in the same region ( Figure 6D ). The average interlayer spacing of E400-3 was 6.60 ± 1.16 nm, which was smaller than that observed in LC-6, indicating that the lamellar structure of E400-3 was more densely packed. The outer lamellar layers showed a spacing of 12.74 ± 1.98 nm, whereas the inner lamellar layers were more densely packed with a spacing of 7.83 ± 0.93 nm ( Figure 6E ). The distinct interlayer spacing in LC-6 suggests that mRNA may be encapsulated or packaged differently in the inner or outer domains. The lamellar structure within the cationic lipoplexes was also confirmed by small-angle X-ray scattering (SAXS) analysis ( Figure 6F ), although E400-3 and LC-6 showed similar d-spacings. This difference in observational results may be due to the difference in sensitivity of the techniques: SAXS characterizes a large-scale bulk analysis, whereas cryo-EM provides local morphology focused on individual particles. Therefore, these measurements can synergistically contribute to a comprehensive analysis of the lipoplex structure. When excess mRNA was incorporated, both E400-0.75 and LC-0.75 showed characteristic lamellar patterns with single-frequency diffraction spots in FFT analysis, with similar interlayer spacings of 6.99 ± 0.31 nm and 6.51 ± 0.45 nm, respectively ( Figures 6G–H ). These results were also supported by SAXS analysis ( Figure 6I ). Overall, the observed interlayer spacing was consistent with the values reported for multilamellar structures in lipoplexes, suggesting that the excess mRNA modifies the lipoplex structure to exhibit similar interlayer spacing regardless of the structure of cationic liposomes.

It is interesting that the concentric multilamellar structure of LC liposomes transforms into two distinct parts in cationic LC-6. In LC-6, the inner region is compressed, while the outer interlamellar gap expands upon mRNA complexation ( Figure 11 ). In the presence of excess mRNA, this dynamic change disappeared, and anionic LC-0.75 exhibited a fingerprint pattern with a typical interlamellar gap. Although direct observation of nucleic acid molecules such as mRNA using cryo-EM is still difficult, nucleic acid molecules can be inferred from their structural features and their influence on electron density within the lipid assembly. In particular, the contrast between the lipid bilayer planes inside the lipoplex was more pronounced than that at the outer edge (see Figure 5 ). This observation suggests that electron-dense mRNA molecules are sandwiched between lipid bilayers, which provides a possible explanation for the formation of a dense multilamellar structure in lipoplexes, as mRNA molecules can facilitate the attachment between adjacent cationic lipid bilayers. Surprisingly, the mRNA molecules inside LC-6 showed a 'bulged' distribution in the outer lamella, with a broad, high-contrast appearance. This suggests that a significant amount of mRNA is trapped in this region, possibly due to significant electrostatic interactions with the cationic DOTAP lipids present in large quantities within the multilamellar bilayer. In contrast, the inner region of LC-6 has a more densely packed lamellar structure and appears to accommodate fewer mRNA molecules. This may be due to the lower amount of cationic lipids in the inner layer and the less accessible mRNA in the inner layer of the liposome. Nevertheless, it is noteworthy that the interlamellar spacing in the inner region of LC-6 lipoplexes is larger than that observed in E400-3 lipoplexes. This suggests that the monolayer lipid membrane of E400 liposomes tends to pack the mRNA closely, and this may be because the monolayer lipid membrane of E400 liposomes is relatively flexible compared to the multilamellar membrane, making it more amenable to deformation. In contrast, the similar interlayer spacing observed for anionic E400-0.75 and LC-0.75 suggests that at very low N/P ratios, strong electrostatic interactions by excess mRNA molecules overcome the inherent rigidity of the multilamellar lipid bilayer, leading to the induction of a dense fingerprint pattern within the lipoplexes.

In summary, these structural comparisons highlight an interesting relationship between the structure of the lipid assembly and the resulting structure of the lipoplex. These results emphasize that the interaction between the cationic liposome and the mRNA complex is a complex process, and that the structure of the cationic liposome significantly influences the final structure of the mRNA-lipoplex. This knowledge further emphasizes the importance of understanding and controlling the structural aspects of the liposome for the optimized performance of lipid-based mRNA delivery systems.

conclusion

The present invention systematically explores the biophysical complexity of lipid assembly in mRNA delivery systems, revealing that liposome structure plays a pivotal role in forming mRNA-lipoplex structures. Prior to mRNA complexation, cationic DOTAP/DOPE liposomes showed significant differences in size distribution and morphological characteristics, which mainly depended on the preparation method. In particular, the LUCA cycle included in the preparation method according to the present invention generated a unique concentric multilamellar structure in liposomes that has not been reported in existing liposome systems. These structural features of cationic liposomes significantly affected the optimal N/P charge ratio in mRNA-lipoplexes, since multilamellar liposomes aggregate at a higher N/P ratio than unilamellar liposomes due to less exposure of cationic lipids. In addition, biological analysis results showed that the variability of the optimal N/P ratio within lipoplexes, which is determined by the structural characteristics of cationic liposomes, significantly affected the biological efficacy. Cryo-EM analysis revealed structural differences between lipoplexes, which are mainly influenced by the cationic liposome architecture and the N/P ratio. Comparative structural analysis of extruded and LUCA cycle liposomes and lipoplexes showed that the concentric multilamellar structure is preserved in cationic lipoplexes derived from LUCA cycle liposomes, exhibiting two distinct interlamellar spacings between the inner and outer lamellae. Overall, our results provide valuable insights into the complex relationship between the architecture of lipid assemblies and the properties of the resulting lipoplexes, laying the foundation for the development of lipid-based mRNA delivery systems.

Claims (18)

A method for preparing cationic liposomes comprising the following steps: (a) a step of preparing a lipid solution by dissolving long-chain cationic lipids and short-chain lipids in an organic solvent, and then drying under a nitrogen gas flow to evaporate the organic solvent and obtain a dried lipid film; (b) a step of preparing a lipid suspension by hydrating the dried lipid film obtained in the above step with an aqueous solvent; and (c) A step of immersing the hydrated lipid suspension prepared in the above step in liquid nitrogen, thawing it in a water bath at 10 to 80°C, and then vortexing it for 5 to 60 seconds, repeating this process 1 to 8 times. A method in claim 1, wherein additional lipids are further added in step (a). In the first aspect, the long-chain cationic lipid is selected from the group consisting of 1,2-di-O-octadecenyl-3-trimethylammonium propane (DOTMA), dimethyldioctadecylammonium (DDAB); 1,2-dioleyl-3-trimethylammonium propane (DOTAP); 1,2-dioleyl-3-dimethylammonium-propane (DODAP); 1,2-diacyloxy-3-dimethylammonium propane; 1,2-dialkyloxy-3-dimethylammonium propane; A method, wherein at least one is selected from the group consisting of dioctadecyldimethyl ammonium chloride (DODAC), 2,3-di(tetradecoxy)propyl-(2-hydroxyethyl)-dimethylazanium (DMRIE), 1,2-dimyristoyl-sn-glycero-3-ethylphosphocholine (DMEPC), l,2-dimyristoyl-3-trimethylammonium propane (DMTAP), 1,2-dioleyloxypropyl-3-dimethyl-hydroxyethyl ammonium bromide (DORIE), and 2,3-dioleyloxy- N-[2(spermine carboxamide)ethyl]-N,N-dimethyl-l-propanamium trifluoroacetate (DOSPA). A method in claim 2, wherein the additional long-chain neutral lipid is at least one selected from the group consisting of 1,2-di-(9Z-octadecenoyl)-sn-glycero-3-phosphoethanolamine (DOPE), 1,2-dioleyl-sn-glycero-3-phosphocholine (DOPC), diacylphosphatidylcholine, diacylphosphatidylethanolamine, ceramide, sphingomyelin, cephalin, cholesterol, and cerebroside. In the first aspect, the short-chain lipid is 1,2-dipropionyl-sn-glycero-3-phosphocholine (3:0 PC), 1,2-dibutyryl-sn-glycero-3-phosphocholine (4:0 PC), 1,2-dipentanoyl-sn-glycero-3-phosphocholine (5:0 PC), 1,2-dihexanoyl-sn-glycero-3-phosphocholine (6:0 PC, DHPC), A method, wherein at least one is selected from the group consisting of 1,2-diheptanoyl-sn-glycero-3-phosphocholine (7:0 PC, DHPC) and 1,2-dioctanoyl-sn-glycero-3-phosphocholine (8:0 PC). A method in claim 1, wherein the molar ratio of the long-chain cationic lipid to the short-chain lipid is 10:0 to 1:9. A method for preparing a lipoplex comprising RNA comprising the following steps: (i) a step of preparing a lipid solution by dissolving a long-chain cationic lipid and a short-chain lipid in an organic solvent, and then drying the solution under a nitrogen gas flow to evaporate the organic solvent and obtain a dried lipid film; (ii) a step of preparing a lipid suspension by hydrating the dried lipid film obtained in the above step with an aqueous solvent; (iii) a step of preparing a cationic liposome by immersing the hydrated lipid suspension prepared in the above step in liquid nitrogen, thawing it in a water bath at 10 to 80°C, and then vortexing it for 5 to 60 seconds, repeating the process 1 to 8 times; and (iv) a step of diluting RNA with an aqueous solvent to prepare an RNA solution, adding the cationic liposome prepared in step (iii) thereto and stirring to prepare a lipoplex containing RNA. A method in claim 7, wherein additional lipids are further added in step (i). In claim 7, the long-chain cationic lipid is selected from the group consisting of 1,2-di-O-octadecenyl-3-trimethylammonium propane (DOTMA), dimethyldioctadecylammonium (DDAB); 1,2-dioleyl-3-trimethylammonium propane (DOTAP); 1,2-dioleyl-3-dimethylammonium-propane (DODAP); 1,2-diacyloxy-3-dimethylammonium propane; 1,2-dialkyloxy-3-dimethylammonium propane; A method, wherein at least one is selected from the group consisting of dioctadecyldimethyl ammonium chloride (DODAC), 2,3-di(tetradecoxy)propyl-(2-hydroxyethyl)-dimethylazanium (DMRIE), 1,2-dimyristoyl-sn-glycero-3-ethylphosphocholine (DMEPC), l,2-dimyristoyl-3-trimethylammonium propane (DMTAP), 1,2-dioleyloxypropyl-3-dimethyl-hydroxyethyl ammonium bromide (DORIE), and 2,3-dioleyloxy- N-[2(spermine carboxamide)ethyl]-N,N-dimethyl-l-propanamium trifluoroacetate (DOSPA). A method in claim 8, wherein the additional long-chain neutral lipid is at least one selected from the group consisting of 1,2-di-(9Z-octadecenoyl)-sn-glycero-3-phosphoethanolamine (DOPE), 1,2-dioleyl-sn-glycero-3-phosphocholine (DOPC), diacylphosphatidylcholine, diacylphosphatidylethanolamine, ceramide, sphingomyelin, cephalin, cholesterol, and cerebroside. In claim 7, the short-chain lipid is 1,2-dipropionyl-sn-glycero-3-phosphocholine (3:0 PC), 1,2-dibutyryl-sn-glycero-3-phosphocholine (4:0 PC), 1,2-dipentanoyl-sn-glycero-3-phosphocholine (5:0 PC), 1,2-dihexanoyl-sn-glycero-3-phosphocholine (6:0 PC, DHPC), A method, wherein at least one is selected from the group consisting of 1,2-diheptanoyl-sn-glycero-3-phosphocholine (7:0 PC, DHPC) and 1,2-dioctanoyl-sn-glycero-3-phosphocholine (8:0 PC). A method in claim 7, wherein the molar ratio of the long-chain cationic lipid to the short-chain lipid is 10:0 to 1:9. In claim 7, the organic solvent of step (a) is a volatile organic solvent, and is a mixed solvent of ethanol, methanol, dichloromethane, chloroform, tetrahydrofuran, ethyl acetate, and ether. A method in claim 7, wherein the molar ratio (N/P ratio) of the positively charged amine group (N) of the cationic lipid and the negatively charged phosphate group (P) in RNA is 1 to 7. A method according to claim 7, wherein the particle size of the manufactured lipoplex is 200 nm to 2000 nm. A method according to claim 7, wherein the Q ratio of the particles of the manufactured lipoplex is 0.5 to 2.0. A method for increasing the selectivity of a target organ that can be delivered by adjusting the N/P ratio to 1 to 7 in the production of a lipoplex. A method in claim 17, wherein the organ is a lung or a spleen.