Zwitterionic lipid nanoparticle compositions, and methods of use

Abstract

A lipid composition having the formula (1): wherein: R is a linear or branched hydrophobic group containing at least four carbon atoms and at least one linkage in which at least three carbon atoms are connected together, and wherein R optionally contains one or more heteroatoms selected from O, N, and S atoms; L is a bond or a linker containing 1-3 atoms, wherein at least one of the atoms is a heteroatom selected from O and N atoms; Ra, Rb, Rc, and Rd are independently selected from H, halogen, methyl group, halomethyl group, C≡N, hydroxy group, and methoxy group; and n is an integer of 1, 2, or 3. Also described herein are methods of producing LNPs in which the lipid composition of Formula (1) is incorporated and methods for delivering a therapeutic substance to a subject, such as an mRNA, by administering to the subject an LNP composition described above.

Description

ZWITTERIONIC LIPID NANOPARTICLE COMPOSITIONS, AND METHODS OF USECROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of priority from U.S. Provisional Application No. 63 / 723,698, filed on November 22. 2024, which is herein incorporated by reference in its entirety.GOVERNMENT SUPPORT

[0002] This invention was made with government support under Grant 1R01AI178125-01 and 1R01 AI193334- 01 awarded by the National Institutes of Health (NIH). The government has certain rights in the invention.FIELD OF THE INVENTION

[0003] The present invention generally relates to zwitterionic lipid nanoparticle (LNP) formulations for the encapsulation and delivery of therapeutic agents, such as nucleic acids. The present invention more specifically relates to the use of such LNP formulations, particularly non-PEGylated versions thereof, for delivery of therapeutic agents to treat a range of diseases or disorders, such as by gene therapy, or to vaccinate a subject.BACKGROUND

[0004] Most mRNAs inside lipid nanoparticles (LNPs) in vaccines become trapped in endosomes after their internalization, preventing them from entering the cytoplasm to exert their effects. Although COVID-19 mRNA vaccines have achieved great clinical success, cancer vaccines face greater challenges, as they require the processing of the host's selfantigens. Higher dosages are typically needed, and more efficient endosomal escape will improve the performance of cancer vaccines.

[0005] Another challenging issue is the reactogenicity of the LNP vectors. Reactogenicity represents the manifestation of inflammatory response to vaccination. This can include pain, redness, or swelling at the injection site, and even threaten the health of patients in severe cases. Vectors or adjuvants can be used to enhance the host's antigen-specific immune response, reducing the need for increased dosages, but often increase the reactogenicity. This indicates that the vectors or adjuvants themselves, independent ofmRNA cargoes, possess structural components that induce reactogenicity. Even FDA- approved LNP vectors, such as BNT162b2 from Pfizer-BioNTech COVID-19 vaccines, exhibits reactogenicity, as observed in its phase 2 / 3 trial safety results (F. P. Polack et al., New England journal of medicine 383, 2603-2615, 2020). Ionizable lipids, the key component of LNPs, have been identified as a significant contributor to the reactogenicity (T. Korzun et al., Pharmaceuticals 16, 1088, 2023). Additionally, the chemical structures of ionizable lipids also influence the mRNA endosomal escape, thereby affecting the expression efficiency of a vaccine and its subsequent immune activation. Thus, it is essential to simultaneously optimize both the endosomal escape (i.e., expression efficiency) and the reactogenicity of mRNA-LNP cancer vaccines, in order to balance the efficacy with clinically acceptable levels of pain and potential risks (S. Chatterjee et al.. Proceedings of the National Academy of Sciences 121, e2307800120, 2024). This undoubtedly puts higher demands on the chemical structure design of ionizable lipids for cancer vaccine applications.

[0006] Although efforts have been made to remedy these problems, these problems still persist and continue to pose a serious challenge to existing vaccines. Thus, a solution to either of these problems would represent a substantial advance in the art of vaccine technology.SUMMARY

[0007] The present invention solves the problems noted above by providing a novel zwitterionic lipid composition that, when incorporated into a lipid nanoparticle (LNP) composition, induces an increased and more effective endosomal escape compared to conventional LNP compositions. The result is a significantly increased mRNA expression compared to conventional LNP compositions. The novel lipid composition also advantageously reduces the reactogenicity of the final vector while retaining its desirable mRNA delivery ability.

[0008] The zwitterionic lipid composition that promotes endosomal escape has the following formula:wherein: R is a linear or branched hydrophobic group containing at least four carbon atoms and at least one linkage in which at least three carbon atoms are connected together, and wherein R optionally contains one or more heteroatoms selected from O, N, and S atoms; L is a bond or a linker containing 1 -3 atoms, wherein at least one of the atoms is a heteroatom selected from O and N atoms; Ra, Rb, Rc, and Rdare independently selected from hydrogen atom, halide atom, methyl group, halomethyl group, C=N, hydroxy group, and methoxy group; and n is an integer of 1, 2, or 3. In some embodiments, R is a linear or branched hydrophobic group containing at least four carbon atoms connected together, and wherein R contains at least one heteroatom selected from O and N. In separate or further embodiments, L = C(O)X, wherein X is O, NH, or S.

[0009] The zwitterionic lipid composition may more particularly have the following formula:wherein: X is O, NH, or S; and R, Ra, Rb, Rc, Rd, and n are as defined above.

[0010] In some embodiments, the zwitterionic lipid composition may more particularly have the following formula:wherein: L, Ra, Rb, Rc, Rd, and n are as defined above; L’ is a bond or a linker of the formula -(CH2)m- wherein m is an integer of 1-30; and R1and R2are independently selected from hydrogen atom and linear or branched hydrophobic groups containing at least four carbon atoms and at least one linkage in which at least three carbon atoms are connected together, and optionally containing one or more heteroatoms selected from O, N, and S atoms. In some embodiments, at least one (or both) of R1and R2has the following formula: — (CH2)P-L1-R3(3a), wherein L1is independently a linker containing 1-3 atoms, wherein at least one of the atoms is a heteroatom selected from O and N atoms; R3is independently a linear or branched hydrophobic group containing at least four carbon atoms and at least one linkage in which at least three carbon atoms are connected together, and optionally containing one or more heteroatoms selected from O, N, and S atoms; and p is an integer of 1 -30.

[0011] In other aspects, the present disclosure is directed to a lipid nanoparticle (LNP) composition comprising a lipid composition of the Formula (1) and a therapeutic molecule. In further embodiments, the LNP further includes at least one lipid not within the scope of Formula (1). The lipid not within the scope of Formula (1) may be selected from the following: (i) tertiary amine (ionizable) lipids, (ii) non-tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group (which may be uncharged or charged, wherein charged includes anionic, cationic, and zwitterionic forms); (iii) lipids containing an uncharged (typically hydrophilic) or zwitterionic polymeric group attached to a hydrophobic tail, and (iv) sterols. The LNP may be one-component, two-component, three- component, four-component, or five-component, wherein the number of components refers to the lipid components in the LNP and not the therapeutic component. In some embodiments, phospho-L-serine (PS) or sphingomyelin (SM) is added onto (attached to) one or more of the lipid components chemically. Alternatively, a natural PS-containing or SM-containing lipid can be included in the LNP. The therapeutic molecule may be one ormore selected from, for example, a nucleotide, gene editing system, protein, compound (e.g., pharmaceutical or medication), and / or an antibody.

[0012] In another aspect, the present disclosure is directed to a method of treating a subject by administering any of the LNPs described above to a subject. The LNP may be administered to the subject intravenously, intramuscularly, intradermally, subcutaneously, intranasally, or by inhalation. In particular embodiments, the LNP is administered by injection into the subject. In some embodiments, the LNP is delivered to cells of the subject. In some embodiments, the LNP is delivered by removing cells from the subject, administering the lipid nanoparticle to the removed cells, and then reintroducing the removed cells to the subject. In some embodiments, the LNP may be administered as part of protein replacement therapy, cancer immunotherapy, cancer vaccine therapy, infectious disease vaccines, gene editing, autoimmune disease treatment and cancer diagnosis. In particular embodiments, the LNP is administered for gene therapy comprising CRISPR-Cas gene editing, or for in vitro and in vivo production of extracellular vesicles, or for vaccination against coronavirus (e.g., SARS-CoV-2), or wherein the LNP is administered along with checkpoint inhibitor (e.g., anti- Programmed death-ligand 1 (anti-PD-Ll) antibody, anti- cytotoxic T-lymphocyte-associated protein 4 (anti-CTLA4) to treat cancer. In embodiments, the method results in a higher transfection efficiency compared to a lipid nanoparticle composition not containing the lipid composition of Formula (1).BRIEF DESCRIPTION OF THE FIGURES

[0013] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary' fee.

[0014] FIGS, la-lh. MeDZ lipid with reduced reactogenicity for efficient mRNA delivery. FIG. la is a scheme showing a synthesis route of MeDZ lipid. FIG. lb is a scheme showing response of MeDZ lipid to pH changes. FIG. 1c is a scheme showing endosomal membrane phospholipids (pink) adopt a lamellar phase. When the protonated MeDZ lipids (yellow) are inserted into the endosomal membrane, the formed cone-shaped ion pairs are incompatible with the bilayer due to the multi-tail structure of MeDZ lipid, enabling hexagonal II transformation and subsequent endosome disruption. MeDZ lipids are protonated in acidic endosomes, promoting endosomal escape and the release of mRNA. FIG. Id shows 'HNMR spectra of MeDZ lipid over the chemical shift range 5.0 ppm to 3.0 ppm after exposure to deionized water with different pH between 5.5 and 7.4. After exposure for 10 minutes, the MeDZ lipid solutions are lyophilized and the resulting lyophilized powders are dissolved in DMSO-t / 6. The peak shift from a to b indicates the protonation of PyCB headgroup. FIG. le shows patterns of hydrogen bonds in PyCBHBO and PyCB-H3O+(Gauss View 6.0.16; hydrogen bond, dashed line). Density functional theory (DFT) calculations based on B3LYP-D3(BJ) / 6-31g(d) were conducted to achieve the optimized geometry structure of these complexes, and solvation Model Based on Density (SMD) was used to simulate the water environment. Carbon, gray; Hydrogen, white; Oxygen, red; Nitrogen, blue. FIG. If shows binding energies of PyCB-H2O and PyCB-H3O+. Binding energies are calculated while basis set superposition error is considered based on M062X / def2TZVP level. FIG. 1g is an LNP formulation table for mRNA delivery validation. FIG. Ih is a graph showing zeta potentials of different LNPs in the buffers versus gradient pH values (pl = isoelectric point).

[0015] FIGS. 2a- 2i. MeDZ LNP facilitates mRNA expression in vitro and in vivo. FIG. 2a is a graph plotting EGFP mRNA expression efficacy of the LNPs containing different percentages of MeDZ lipids (mRNA, 1 pg mL1) tested on DC2.4 cells. FIG. 2b is a graph plotting internalization of DiR-labelled LNP formulations by DC2.4 cells after 3-hour incubation. FIG. 2c is a schematic illustration of the experimental design. Images were taken at 6 hours post-injection. FIG. 2d shows representative whole-body fluorescence (top) and bioluminescence (bottom) images of the mice after SC injection of different DiR- labelled Luc mRNA-loaded LNPs measured by IVIS imaging system (0.25 mg kg-1Luc mRNA, 5 pg per mouse). Representative ex vivo fluorescence (top) and bioluminescence (bottom) images of the main organs are also shown. LNs are marked with yellow arrows. FIG. 2e is a quantitative analysis plot of the bioluminescence intensity (luciferase expression) of LNs in each group. FIG. 2f is a quantitative analysis plot of the fluorescence intensity of LNs. FIG. 2g is a plot of relative ratio of bioluminescence and fluorescence of LNs (the relative ratio of bioluminescence to fluorescence for the BNT162b2 formulation was defined as 1). FIG. 2h schematically depicts the mechanism of Cre-mediated gene recombination in Ail4 reporter mice. The gene encoding tdTomato is blocked by a stop gene (red) between two loxP segments (green). When Cre mRNA is delivered to the cytoplasm and expressed, the loxP gene will be cut. This activates the tdTomato gene, leading to the expression of tdTomato in the cells. FIG. 2i shows a plot of percentage oftdTomato-positive (tdTomato+) cells in different types of immune cells after SC injection of Cre mRNA-loaded MeDZ LNP. Naked Cre mRNA is employed as control. Data are presented as mean ± standard deviation (s.d.) from n biologically independent samples (n = 3). Statistical significance was analyzed by one-way ANOVA with Tukey’s multiple comparisons test. (** *p < 0.001, *** < 0.0001).

[0016] FIGS. 3a-3h. Model membrane studies of MeDZ lipid-mediated endosomal disruption and antigen presentation. FIG. 3 a schematically shows the mechanism of MeDZ lipid-mediated endosomal disruption and production of FRET probes for a FRET assay. The endosomal mimic is prepared by mixing the DOPS, DOPC, and DOPE (molar ratio, 25 / 25 / 50), and being pre-labelled by NBD-PE and Rhod-PE. FIGS. 3b and 3c are graphs showing the capability of BNT162b2 LNPs and MeDZ LNPs in disrupting the endosomal mimicking liposomes. FIG. 3d is a31P NMR spectra of a mixture of endosomal mimicking liposomes and MeDZ lipids. The mixture of endosomal mimicking liposomes and ALC- 0315 lipids is used as control. FIG. 3e shows representative fluorescence images of the endosomal escape of MeDZ LNP and BNT162b2 LNP in DC2.4 cells. NBD-labelled LNP formulations are prepared (green) and incubated with cells. LysoTracker™ Red is added to stain the endosomes. Scale bar, 20 pm and 5 pm. FIG. 3f shows representative surface plots of quantification of LNPs (NBD, left) and endosomes (Lyso, right) within cells. FIG. 3g shows quantitative analysis of co-localization of NBD-labelled LNP formulations with endosomes labelled with LysoTracker™ Red. FIG. 3h shows the Pearson’s correlation coefficient values. Data are presented as mean ± s.d. from n biologically independent samples (n = 3). Statistical significance was analyzed by t test for b and c. Cp < 0.05, **p < 0.01).

[0017] FIGS. 4a-41. MeDZ LNP enhances antigen presentation and exhibits reduced reactogenicity. FIG. 4a shows Western blotting analysis of cellular OVA levels in BMDCs after incubating with different LNP formulations for 6 hours. FIG. 4b shows the quantitative analysis results obtained using Image J. Cellular OVA levels are evaluated after further 24-hour incubation. FIG. 4c shows a representative flow cytometry analysis. FIG. 4d shows a statistical analysis of the expression of H2kb-SIINFEKL in BMDCs after different treatments. FIG. 4e is a schematic illustration of the experimental design. C57BL / 6 mice were subcutaneously injected MeDZ LNPs (LNP, 30 pg per mouse). The skin samples from the injection sites were collected for analysis. FIGS. 4f, 4g, and 4h showLuminex analysis of the granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin- ip (IL-1 P ), and interleukin-6 (IL-6) in the skin samples. FIGS. 4i shows flow cytometry analysis of the proportion of neutrophils (gated on Ly-6G+CDl lb+CD45+) in CD45+cells. FIG 4j shows the gating strategy. FIG. 4k shows photographs of skin samples from the mice subcutaneously treated with different LNPs. FIG. 41 shows H&E staining images of the skin tissues collected from the injection sites. Scale bar, 300 pm. Data are presented as mean ± s.d. from n biologically independent samples (n = 3). Statistical significance was analyzed by one-way ANOVA with Tukey’s multiple comparisons test. (*P < 0.05, **p < 0.01, *** < 0.0001).

[0018] FIGS. 5a-5m. MeDZ LNP vaccine prevents tumorigenesis, lung metastasis, and recurrence. FIG. 5a is a schematic illustration of the timeline for vaccination (OVA mRNA, 5 pg per mouse), bleeding, and B16F10-OVA tumor inoculation (SC injection, 1 x 106per mouse) in a C57BL / 6 murine model. FIGS. 5b and 5c are graphs showing OVA-specific IgM (FIG. 5b) and IgG (FIG. 5c) levels in the blood on day 14 after vaccination. FIG. 5d shows photographs of B 16F10-OVA tumors after different treatments on day 30. FIG. 5e shows individual tumor growth curves (CR = complete response). FIG. 5f shows flow cytometry analysis of the expression of H2kb-SIINFEKL in DCs (gated on CD1 lc+cells) within the LNs. FIG. 5g shows flow cytometry analysis of the population of CD8+CD3+T cells that bear T cell receptors binding to H2Kb OVA tetramer-SIINFEKL in the blood. FIG. 5h shows flow cytometry analysis of the population CD62LlowCD44hlghT cells (gated on CD8+CD3+cells) within the spleens. FIG. 5i is a schematic illustration of the timeline for B16F10-OVA tumor inoculation (IV injection, 1 x 106per mouse) and vaccination. FIG. 5j shows H&E staining images of the lung tissues. Scale bar, 3000 pm. FIG. 5k is a schematic illustration of the timeline for B16F10-OVA tumor inoculation (IV injection, 1 x 106per mouse), vaccination (OVA mRNA, 5 pg per mouse), and tumor rechallenge. For rechallenge, B16F10-OVA tumor cells (1 x 106per mouse) were subcutaneously injected at the left flank of mice, and EO771 tumor cells (1 x 106per mouse) were inoculated in the opposite flank of the B 16F10-OVA tumors. FIG. 51 shows photographs of B 16F10-OVA and EO771 tumors. FIG. 5m shows individual tumor growth curves (PBS, gray; MeDZ LNP, red), wherein TFS = tumor-free survival. Data are presented as mean ± s.d. from n biologically independent samples (b, c, f, g, h, n = 3; e, n = 6; m, n = 5). Statisticalsignificance was analyzed by one-way ANOVA with Tukey’s multiple comparisons test. (*P < 0.05, **p < 0.01, ***p < 0.001, *** *p < 0.0001).

[0019] FIGS. 6a-6p. MeDZ LNP vaccine inhibits tumor growth. FIG. 6a is a schematic illustration of the timeline for B 16F10-OVA tumor inoculation (SC injection, 1 x 106per mouse) and treatments (OVA mRNA, 5 pg per mouse) in a C57BL / 6 murine model. FIG. 6b is a graph showing average tumor growth kinetics after different treatments. FIG. 6c is a graph showing survival of B16F10-OVA tumor-bearing mice. FIG. 6d shows H&E staining images of the tumor tissues. Scale bar, 100 pm. FIG. 6e is a graph showing flow cytometry analysis of the expression of H2kb-SIINFEKL in DCs (gated on CD1 lc+cells) within the LNs. FIG. 6f is a graph showing flow cytometry analysis of the population of CD8+CD3+T cells that bear T cell receptors binding to H2Kb OVA tetramer-SIINFEKL within the LNs. FIG. 6g is a graph showing flow analysis of the population of CD8+T cells (gated on CD3+CD45+cells) within the tumors. FIG. 6h is a graph showing flow cytometry analysis of the population of CD8+CD3+CD45+T cells that bear T cell receptors binding to H2Kb OVA tetramer-SIINFEKL within the tumors. FIG. 6i is a schematic illustration of the experimental design. FIG. 6j shows individual tumor growth curves (CR = complete response). FIG. 6k shows photographs of B16F10-OVA tumors on day 15. FIG. 61 is a graph showing average tumor growth kinetics (aPD-1 antibody, IP injection, 200 pg per dose). FIG. 6m is a graph showing survival of B16F10-OVA tumor-bearing mice. FIG. 6n shows representative ex vivo fluorescence (left) and bioluminescence (right) images of the main organs, including heart, liver, lung, kidney, and spleen. FIG. 6o is a graph showing quantitative analysis of the bioluminescence (luciferase expression) intensity of spleen in each group. FIG. 6p is an LNP formulation table for spleen-targeting mRNA delivery validation. Data are presented as mean ± s.d. from n biologically independent samples (b, c, n = 5; e, f, g, h, o, n = 3; j, 1, m, n = 6). Statistical significance was analyzed by one-way ANOVA with Tukey’s multiple comparisons test for e, f, g, and h, two-way ANOVA with Tukey’s test for b and 1, log-rank (Mantel-Cox) test for c and m, and t test for o. Cp < 0.05, * *p < 0.01, ** *p < 0.001, ****p < 0.0001).

[0020] FIGS. 7a-7f. Cholesterol- and PEGylated lipid-free zwitterionic ThrCo LNPs for spleen-specific mRNA translation. FIG. 7a shows the chemical structure of ALC-0315 IL (IL, ionizable lipid). FIG. 7b shows the chemical structure of zwitterionic PyCB IL. FIG. 7c schematically shows the pH-responsiveness of the PyCB group. The PyCB groupcomplexes with water (PyCB-I I2O. blue) via charge-assisted hydrogen bonding and exhibits a zwitterionic property at physiological pH. At a lower pH (~6.5), it rapidly transforms into a cationic state (PyCB-H3O+, magenta), thereby contributing towards endosomal escape. FIGS. 7d, 7e, and 7f are representative images of mRNA-LNP accumulation and translation in the main organs after administration of (FIG. 7d) clinically available BNT162b2 LNPs (high levels of accumulation and translation in the liver; low levels accumulation and translation in the spleen), (FIG. 7e) anionic spleen-targeting 18PA LNPs (low levels of accumulation and translation in the liver; low levels of accumulation and translation in the spleen), and (FIG. 7f) our zwitterionic ThrCo LNPs (a moderate level of accumulation and a low level of translation in the liver; a moderate level of accumulation and a high level of translation in the spleen), respectively. A, accumulation; T, translation; Mod, moderate.

[0021] FIGS. 8a-8q. PyCB ILs enable reformulation of ThrCo LNPs for spleen-specific mRNA transfection. FIG. 8a shows representative whole-body fluorescence and bioluminescence images of the mice after IV injection of different DiR-labeled Luc mRNA- loaded LNPs (mRNA, 5 pg per mouse). Representative ex vivo fluorescence and bioluminescence images of the main organs. Images were taken at 6 hours post-injection. FIG. 8b is a graph showing quantitative analysis of the spleen fluorescence intensity in each group. FIG. 8c is a graph showing quantitative analysis of the spleen bioluminescence intensity. FIG. 8d are graphs showing quantitative analysis of the spleen fluorescence and bioluminescence intensity in the BNT162b2 LNP and ThrCo(45%) LNP groups. FIG. 8e is a graph showing quantitative analysis of the liver fluorescence intensity. FIG. 8f provides a quantitative analysis of spleen bioluminescence intensity with increasing PyCB IL percentage in LNPs from 20% to 65%. FIG. 8g shows representative ex vivo bioluminescence images of the main organs. FIG. 8h is an LNP formulation table for the optimization of helper lipid molar ratios. FIG. 8i is a graph showing quantitative analysis of the spleen bioluminescence intensity. FIG. 8j shows representative ex vivo fluorescence (left, DiR; middle, Cy5) and bioluminescence images of the main organs. FIG. 8k is a graph showing quantitative analysis of the spleen DiR fluorescence intensity. FIG. 81 is a graph showing quantitative analysis of the spleen Cy5 fluorescence intensity. FIG. 8m is a graph showing quantitative analysis of the spleen bioluminescence intensity. FIGS. 8n-8q are graphs showing serum ALT and AST levels in the male and female mice after IV injection of BNT162b2 LNP and ThrCo LNP, respectively. Data are presented as mean ± standard deviation (s.d.) from n biologically independent samples (for FIGS. 8b, 8c, 8e, 8i,8k, 81, and 8m, n - 3; for FIGS. 8n, 8o, 8p, and 8q, n = 4). Statistical significance was analyzed by one-way ANOVA with Tukey’s multiple comparisons test for FIGS. 8b, 8c, 8i, 8n, 8o, 8p, and 8q, and t test for FIGS. 8d, 8e, 8k, 81, and 8m. (*P < 0.05, **p < 0.01, ***p < 0.001, *** < 0.0001).

[0022] FIGS. 9a-9s. Characterization of ThrCo LNPs. FIG. 9a is a graph plotting particle size and FIG. 9b is a TEM image of ThrCo LNPs. Scale bar, 200 nm. FIG. 9c is a graph showing stability analysis of ThrCo LNPs in the presence and absence of serum. FIG. 9d is a scheme showing DiR-labeled ThrCo LNPs loaded with EGFP mRNA co-incubated with DC2.4 cells in the presence of various internalization inhibitors. FIGS. 9e-9j are graphs showing the internalization of ThrCo LNPs and the translation efficiency of mRNA in DC2.4 cells in the presence of amiloride (FIGS. 9e and 9f), nystatin (FIGS. 9g and 9h), or chlorpromazine (FIGS. 9i and 9j). FIG. 9k shows representative fluorescence images of the endosomal escape of NBD-labeled ThrCo LNPs (green) in DC2.4 cells. Scale bar, 10 pm. FIG. 91 shows representative surface plot of quantification of ThrCo LNPs (NBD, left) and endosomes (Lyso, right) within cells. FIG. 9m is a schematic showing ThrCo LNP- mediated endosome disruption. FIG. 9n is a graph plotting endosome disruption determined by a FRET assay. The endosomal mimicking liposomes are prepared by mixing the DOPS, DOPC, and DOPE (25 / 25 / 50, mol / mol), and labeled by NBD-PE and Rhod-PE. FIG. 9o is a graph plotting ThrCo LNP dissociation determined by a FRET assay. The ThrCo LNPs are labeled by NBD-PE and Rhod-PE. FIG. 9p shows images of the hemolysis assay to evaluate the membrane disruption by ThrCo LNPs at pH 7.4 and pH 6.0. FIG. 9q shows images of the hemolysis assay to evaluate the biocompatibility of ThrCo LNPs at pH 7.4. FIG. 9r is a graph plotting zeta potentials of ThrCo LNPs in the buffers with a gradient pH value. FIG. 9s shows31P NMR spectra of a mixture of endosomal mimicking liposomes and PyCB ILs. The mixture of endosomal mimicking liposomes and ALC-0315 ILs is used as control. Data are presented as mean ± s.d. from n biologically independent samples (n = 3). Statistical significance was analyzed by one-way ANOVA with Tukey’s multiple comparisons test. (*P < 0.05, * < 0.01, ** < 0.001, *** < 0.0001).

[0023] FIGS. lOa-lOk. ThrCo LNPs as a robust mRNA vaccine carrier. FIG. 10a is a scheme showing Cre -mediated gene recombination in Ail 4 mice. The gene encoding tdTomato is initially blocked by a stop gene (yellow) located between two loxP segments. Upon expression of Cre mRNA in the cytoplasm, the loxP gene will be cut by the Cre. Thisremoval activates the tdTomato gene, resulting in tdTomato expression, which can be detected in the cells where the recombination event occurs. FIG. 10b shows the percentage of tdTomato-positive (tdTomato+) cells across different immune cell populations following IV injection of Cre mRNA-loaded LNPs. FIG. 10c is a schematic illustration of the timeline for multiple administrations of LNPs (OVA mRNA, 5 pg per mouse). On day 17, splenic T cells are isolated and re-stimulated with OVA to assess antigen-specific immune responses. FIG. lOd shows representative flow cytometry analysis and FIG. lOe shows statistical analysis of the population of CD8+CD3+T cells that bear T cell receptors binding to H2Kb OVA tetramer-SIINFEKL. FIG. 1 Of is a schematic illustration of the timeline for single injection and multiple injections of GFP mRNA-loaded LNPs. The images are taken at 24 hours post-injection of the LNPs encapsulating Luc mRNA. FIG. 10g shows representative ex vivo bioluminescence images of the livers and spleens. FIG. 1 Oh is a graph providing quantitative analysis of liver bioluminescence intensity in each group. FIG. lOi is a graph providing quantitative analysis of spleen bioluminescence intensity. FIG. lOj is a schematic illustration of the separation of free and bound proteins to investigate serum protein adsorption on the LNP surface. FIG. 10k shows quantitative measurements of serum protein amounts adsorbed by LNPs after incubation with mouse whole serum. Data are presented as mean ± s.d. from n biologically independent samples (n = 3). Statistical significance was analyzed by one-way ANOVA with Tukey’s multiple comparisons test for FIGS. lOe and 10k, and t test for FIGS. lOh and lOi. (p < 0.05, * *p < 0.01, ** *p < 0.001, *** *p < 0.0001).

[0024] FIGS. 1 la-1 In. ThrCo LNP-based vaccines inhibit tumor metastasis. FIG. I la is schematic illustration of the timeline for B16F10-OVA tumor inoculation and vaccination (mRNA, 15 pg per mouse). FIG. 1 lb shows photographs of the lungs. FIG. 11c shows H&E staining images of the lung tissues. Scale bar, 3000 pm. FIG. 1 Id is a graph showing flow cytometry analysis of the population of CD8+CD3+T cells that bear T cell receptors binding to H2Kb OVA tetramer-SIINFEKL in the spleens. FIG. 1 le is a graph showing flow cytometry analysis of the population of IFN-y+CD8+CD3+T cells in the spleens. FIG. 1 If is a graph showing flow cytometry analysis of the population of CD8+CD3+T cells that bear T cell receptors binding to H2Kb OVA tetramer-SIINFEKL in the blood. FIG. 11g is a graphs showing flow cytometry analysis of the population CD62LlowCD44hlg11T cells (gated on CD8+CD3+cells) in the spleens. FIG. 1 Ih is a schematic illustration of the timeline for B16F10-OVA tumor inoculation, vaccination (mRNA, 15 pg per mouse), and tumorrechallenge. FIG. Hi shows photographs of the tumors. FIG. 11 j shows individual tumor growth curves (PBS, gray; ThrCo LNP, steel blue). TFS: tumor-free survival. FIG. I lk is a graph showing flow cytometry analysis of the population of CD8+T cells (gated on CD3+CD45+cells) in B16F10-OVA tumors. FIG. 111 is a graph showing flow cytometry analysis of the population of IFN-y+CD8+T cells (gated on CD3+CD45+cells) in B 16F10- OVA tumors. FIG. 1 Im is a graph showing flow cytometry analysis of the population of CD8+T cells (gated on CD3+CD45+cells) in EO771 tumors. FIG. 1 In is a graph showing flow cytometry analysis of the population of IFN-y+CD8+T cells (gated on CD3+CD45+cells) in EO771 tumors. Data are presented as mean ± s.d. from n biologically independent samples (for FIGS, l id, He, I lf, 11g, Hi, Ilk, 111, 11m, and l ln, n = 3; J, n = 5). Statistical significance was analyzed by one-way ANOVA with Tukey’s multiple comparisons test for FIGS. 1 Id, 1 le, 1 If, and 11g, and t test for FIGS. I lk, 111, 1 Im, and l ln. Cp < 0.05, **p < 0.01, ***p < 0.001, *** *p < 0.0001).

[0025] FIGS. 12a-12j. ThrCo LNP-based vaccines inhibit tumor growth. FIG. 12a is a schematic illustration of the timeline for B16F10-OVA tumor inoculation and treatments (mRNA, 15 pg per mouse). FIG. 12b shows photographs of B16F10-OVA tumors in each group. FIG. 12c are graphs showing individual tumor growth curves. CR: complete response. FIG. 12d is a graph showing average tumor growth kinetics. FIG. 12e is a graph showing survival of B16F10-OVA tumor-bearing mice. FIG. 12f is a graph showing flow cytometry analysis of the population of CD8+CD3+T cells that bear T cell receptors binding to H2Kb OVA tetramer-SIINFEKL in the spleens. FIG. 12g is a graph showing flow cytometry analysis of the population of lFN-y+CD8+T cells (gated on CD3+CD45+cells) in the spleens. FIG. 12h is a graph showing flow cytometry analysis of the population of CD8+T cells (gated on CD3+CD45+cells) in B16F10-OVA tumors. FIG. 12i is a graph showing flow cytometry analysis of the population of CD8+CD3+T cells that bear T cell receptors binding to H2Kb OVA tetramer-SIINFEKL in B16F10-OVA tumors. FIG. 12j is a graph showing flow cytometry analysis of the population CD62LlowCD44hlghT cells (gated on CD8+CD3+cells) in the spleens. Data are presented as mean ± s.d. from n biologically independent samples (for FIGS. 12d and 12e, n - 5; for FIGS. 12b, 12f, 12g, 12h, 12i, and 12j, n = 3). Statistical significance was analyzed by two-way ANOVA with Tukey’s test for D, log -rank (Mantel-Cox) test for FIG. 12e, and one-way ANOVA with Tukey’s multiple comparisons test for FIGS. 12f, 12g, 12h, 12i, and 12j. (*P 0.05, ' < 0.01, ** < 0.001, *** *p < 0.0001).

[0026] FIGS. 13a-13b. Representative ex vivo fluorescence (FIG. 13a) and bioluminescence (FIG. 13b) images of the main organs at different time points after IV administration of new ThrCo (DOPS) LNP.

[0027] FIGS. 14a- 14b. FIG. 14a schematically shows the chemical structure of a zwitterionic ionizable (ZI) lipid and the structure of mRNA-loaded antibody-modified LNP. FIG 14b schematically shows a "two-step" targeting strategy: first, leveraging the intrinsic organ-targeting properties of LNPs to achieve efficient splenic accumulation; second, functionalizing the LNP surface with specific antibodies to enable precise recognition and internalization by predefined immune cell subsets.

[0028] FIGS. 15a-15b. FIG. 15a shows representative spleen sections from C57BL / 6 mice 2 and 8 hours after i.v. injection of Cy5-mRNA-loaded TwoCo LNPs, with multiplex fluorescence images (DAPI, f-actin, Cy5-mRNA-LNP). FIG. 15b shows zoomed-in spleen images highlighting Cy5 signal and the corresponding Cy5 intensity maps across the white pulp.

[0029] FIGS. 16a-16d. FIG. 16a shows Cy5-mRNA-LNP internalization in RAW264.7 cells treated in vitro with BioNTech (BNT) or TwoCo (STB) LNPs at high (H), normal (N), and low (L) doses, presented as a heatmap and corresponding median fluorescence intensity. FIG. 16b shows GFP expression in RAW264.7 cells under the same conditions. FIG. 16c shows Cy5-mRNA-LNP internalization in bone-marrow-derived macrophages (BMDMs) treated with BNT or STB LNPs at H, N, and L doses. FIG. 16d shows GFP expression in BMDMs under the same dosing conditions, with heatmaps and median fluorescence intensity plots.

[0030] FIGS. 17a-17b. FIG. 17a shows the preparation of the dorsal skinfold window chamber for real-time intravital imaging following LNP administration. FIG. 17b shows representative time-lapse fluorescence images before and after i.v. injection of Cy5-mRNA- loaded BioNTech-like (BNT) or TwoCo (STB) LNPs, with nuclei, blood vessels, macrophages, and Cy5-mRNA-LNP signal visualized.

[0031] FIGS. 18a-18c. FIG. 18a shows representative fluorescence (up) and bioluminescence (bottom) images of the mice and their main organs after IV administration of different LNPs. FIGS. 18b and 18c are graphs showing the percentage of GFP-positive(GFP+) cells across different immune cell populations following IV injection of GFP mRNA-loaded LNPs.

[0032] FIGS. 19a-19e. FIG. 19a shows representative fluorescence (up) and bioluminescence (bottom) images of the mice and their main organs after IV administration of different LNPs. FIGS. 19b and 19c are graphs showing the percentage of GFP-positive (GFP+) cells across different immune cell populations following IV injection of GFP mRNA-loaded LNPs. FIG. 19d is a graph showing the percentage of tdTomato+T-cell in Ail 4 mice after IV administration of Cre mRNA-loaded LNPs conjugated with anti -mouse CD5 (TwoCo+PCB vs. commercial four-component). FIG. 19e is a graph showing the percentage of GFP+B-cell after IV administration of GFP mRNA-loaded LNPs conjugated with different antibodies (TwoCo+PCB vs. commercial four-component).DETAILED DESCRIPTION

[0033] In one aspect, the present disclosure is directed to a lipid composition that facilitates endosomal escape while minimizing LNP reactogenicity. The present invention has achieved this by including a novel Membrane-Destabilizing Zwitterionic (MeDZ) ionizable lipid that is incorporated into the LNPs, thereby greatly facilitating mRNA expression by enhancing endosomal escape while minimizing LNP reactogenicity. The MeDZ lipid features a pyridine-based carboxybetaine (PyCB) zwitterionic headgroup, degradable multitailed alkyl chains, and a tertiary amine-based linker. As discussed in the Examples section, the molecular mechanisms behind the excellent mRNA expression ability of the MeDZ LNPs was explored. The PyCB headgroup forms a zwitterionic PyCB-FLO complex that protonates to a positively charged PyCB-HsO+state below a pH of 6.8. Thus, the PyCB headgroup shows excellent biocompatibility at physiological pH and a strong protonation ability in endosomes, thereby enabling earlier and efficient endosomal escape. The present disclosure also demonstrates that incorporating the MeDZ lipid into the BNT162b2 LNP vector from Pfizer-BioNTech boosts mRNA expression in antigen-presenting cells within lymph nodesfor cytotoxic T cell activation, thus resulting in superior anti-tumor efficacy compared to the Pfizer-BioNTech vaccine. LNPs containing MeDZ lipids exhibit significantly reduced inflammation and neutrophil infiltration in skin tissue at the injection site due to its zwitterionic property. It has herein been further demonstrated that the MeDZlipid can be integrated with the existing targeted LNP formulations to improve their mRNA expression.

[0034] The MeDZ ionizable lipid (i.e., lipid composition) has the following formula:

[0035] The variable R in Formula (1) is a linear or branched hydrophobic group containing at least four carbon atoms and at least one linkage in which at least three (or at least four, five, six, seven, or eight) carbon atoms are connected together. The variable R optionally contains one or more heteroatoms selected from O, N, and S atoms. The variable R may or may not include one or more carbon-carbon double or triple bonds. In different embodiments, the variable R is a linear branched hydrophobic group (e.g., alkyl, alkenyl, or alkynyl) containing precisely or at least 4, 5, 6, 7, 8, 10, 12, 15, 20, 25, 30, 35, 40, 45, or 50 carbon atoms or a number of carbon atoms within a range bounded by any two of the foregoing values (e.g., 4-50, 4-40, 4-30, 4-20, 5-50, 5-40, 5-30, 5-20, 6-50, 6-40, 6-30, 6-20, 7-50, 7-40, 7-30, 8-50, 8-40, 8-30, 10-50, 10-40, 10-30, 12-50, 12-40, or 12-30 carbon atoms). In some embodiments, R contains at least one, two, or more heteroatoms selected from O, N, and / or S atoms (or more particularly, at least one or two heteroatoms selected from O and / or N). In some embodiments, R contains one, two, or more heteroatomcontaining linkages, such as one or more ether (-O-), thioether (-S-), carbonyl (-C(O)-), thiocarbonyl (-C(S)-), ester (-C(O)O-), thioester (-C(S)O-), carbonate (-OC(O)O-),I secondary amine (-NH-), tertiary amine (— N— ), imine (-N=), diazo (-N=N-), disulfide (-S- S-), amide (-C(O)NH-), urea (-NHC(O)NH-), and / or thiourea (-NHC(S)NH-) linkages. In some embodiments, R contains precisely or at least one, two, or three tertiary amine linkages and may or may not also include precisely or at least one or two ester or amide linkages.

[0036] In some embodiments of Formula (1), R is a linear alkyl group, such as an n-butyl, n-pentyl, n-hexyl, n-heptyl, n-octyl, n-nonyl, n-decyl, n-undecyl, n-dodecyl, n-tridecyl, n-tetradecyl, / / -pentadecyl, zz-hexadecyl, zz-heptadecyl, n-octadecyl, zz-eicosyl, n-docosyl, n- tetracosyl, n-hexacosyl, zz-octacosyl, and zz-triacontyl groups, as well as those containing 40, 50, or more carbon atoms, wherein any of the foregoing groups may or may not contain one or more heteroatoms or heteroatom-containing linkages provided above. In other embodiments, R is a branched alkyl group, such as isobutyl (2-methylprop-l-yl), sec-butyl (2-butyl), / -butyl (1,1-dimethylethyl-l-yl), 2-pentyl, 3-pentyl, 2-methylbut-l-yl, isopentyl (3-methylbut-l-yl), 1,2-dimethylprop-l-yl, 1,1-dimethylprop-l-yl, neopentyl (2,2- dimethylprop-l-yl), 2-hexyl, 3-hexyl, 2-methylpent-l -yl, 3-methylpent-l -yl, isohexyl (4- methylpent-l-yl), 1,1-dimethylbut-l-yl, 1,2-dimethylbut-l-yl, 2,2-dimethylbut-l-yl, 2,3- dimethylbut- 1 -yl, 3,3-dimethylbut-l -yl, 1 , 1 ,2-trimethylprop- 1 -yl, 1 ,2,2-trimethylprop- 1 -yl, isoheptyl, isooctyl, and the numerous other branched alkyl groups having up to 20, 30, 40, 50, or more carbon atoms, wherein any of the foregoing groups may or may not contain one or more heteroatom-containing linkages provided above. In some embodiments, R is or includes a saturated or unsaturated cyclic hydrocarbon group, wherein some examples of saturated cyclic hydrocarbon (cycloalkyl) groups include cyclobutyl, cyclopentyl, cyclohexyl, cycloheptyl, and cyclooctyl groups, and some examples of unsaturated cyclic hydrocarbon groups include cyclopentenyl, cyclopentadienyl, cyclohexenyl, cyclohexadienyl, phenyl, benzyl, cycloheptenyl, cycloheptadienyl, cyclooctenyl, cyclooctadienyl, and cyclooctatetraenyl groups. Any of the cyclic hydrocarbon groups exemplified above may or may not include one or more nitrogen atoms substituting the equivalent number carbon atoms. Moreover, the cyclic hydrocarbon group(s), such as any of those exemplified above, may or may not be connected to a linear or branched alkenyl, alkynyl, or alkyl linker that links between the R cyclic group and L, or the R cyclic group may be bound directly to L while containing any one or more of the linear or alkenyl, alkynyl, or branched alkyl groups exemplified above as one or more substituents, wherein any of the substituents may or may not contain one or more heteroatoms or heteroatomcontaining linkages provided above.

[0037] The variable L in Formula (1) is a bond or a linker containing 1-3 atoms, wherein at least one of the atoms is a heteroatom selected from O, N, and S atoms. For purposes of the invention, an H atom is not counted as an atom in L. Any of the linkers exemplified earlier above which contain 1-3 atoms (excluding H) may be considered as L. In particular embodiments, L is C(O)X, wherein X is O, NH, or S (e.g., L is -C(O)O- or -C(O)NH-), wherein the C(O) portion is typically bonded to the pyridinium ring in Formula (1) and Xlinks between C(O) and R. Notably, any one of the foregoing L linkers may be combined with any one of the above-described R groups to result in any L-R combination.

[0038] The variables Ra, Rb, Rc, and Rdin Formula (1) are independently selected from hydrogen atom, halide (halogen) atom, methyl group, halomethyl group, C=N (cyano), hydroxy group, and methoxy group. Tn some embodiments, Ra, Rb, Rc, and Rdare all H. In some embodiments, at least one of Ra, Rb, Rc, and Rdis / are selected from non-H groups exemplified above. In some embodiments, precisely or at least one of Ra, Rb, Rc, and Rdis a halogen atom, such as F, Cl, or Br. In some embodiments, precisely or at least one of Ra, Rb, Rc, and Rdis a methyl group, halomethyl group, C=N (cyano), hydroxy group, or methoxy group. Notably, any of the above possibilities for Ra, Rb, Rc, and Rdmay be combined with any of the possible combinations of L and R disclosed above.

[0039] The variable n in Formula (1) is an integer of 1, 2, or 3. When n is 1, a methylene (-CH2-) linker results between the carboxylate group and pyridinium nitrogen in Formula (1). When n is 2, an ethylene (-CH2CH2-) linker results between the carboxylate group and pyridinium nitrogen in Formula (1). When n is 3, a propylene (-CH2CH2CH2-) linker results between the carboxylate group and pyridinium nitrogen in Formula (1). Notably, any particular n value or range thereof may be combined with any of the possible selections of Ra, Rb, Rc, and Rdand any of the possible combinations of L and R disclosed above.

[0040] In some embodiments, the lipid composition has a structure within the following sub-formula of Formula (1):

[0041] In Formula (2), X is O, NH, or S; and R, Ra, Rb, Rc, Rd, and n are as defined above under Formula (1). Notably, any of the above possibilities for R may be combined with any of the above possibilities for X and any of the above possibilities for Ra, Rb, Rc, and Rdand any of the above possibilities for n in Formula (2).

[0042] In some embodiments, the lipid composition has a structure within the following sub-formula of Formula (1):

[0043] In Formula (3), L, Ra, Rb, Rc, Rd, and n are as defined above under Formula (1). The linker L’ is a bond or a linker of the formula -(CFhlm- wherein m is an integer of 1-30. In different embodiments, m is an integer of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 15, 20, or 30 or m is within a range bounded by any two of these values, e.g., 1-20, 1-10, 1-6, 1-4, 1-3, 2-20, 2- 10, 2-6, 2-4, 3-20, 3-10, or 3-6. The variables R1and R2are independently selected from hydrogen atom and linear or branched hydrophobic (e.g., alkyl, alkenyl, or alkynyl) groups (R, as described above) containing at least four carbon atoms and at least one linkage in which at least three (or at least four, five, six, seven, or eight) carbon atoms are connected together. The linear or branched hydrophobic groups for R1and / or R2may or may not contain one or more heteroatoms selected from O, N, and S atoms. R1and R2may be independently selected from any of the R groups described and exemplified earlier above, i.e., linear or branched hydrophobic groups containing at least four carbon atoms and at least one linkage in which at least three (or at least four, five, six, seven, or eight) carbon atoms are connected together, and optionally containing one or more heteroatoms selected from O, N, and S atoms. In some embodiments, one or both of R1and R2is / are selected from linear or branched hydrophobic (e.g., linear or branched alkyl, alkenyl, or alkynyl) groups, wherein R1and R2may be the same or different. In some embodiments, at least one (or both) of R1and R2contain at least one heteroatom-containing linkage, such as one or more ether (-O-), thioether (-S-), carbonyl (-C(O)-), thiocarbonyl (-C(S)-), ester (-C(O)O-), thioester (-C(S)O-), carbonate (-OC(O)O-), secondary amine (-NH-), tertiary amine (— — ), imine (-N=), diazo (-N=N-), disulfide (-S-S-), amide (-C(O)NH-), urea (-NHC(O)NH-), and / or thiourea (-NHC(S)NH-) linkages. In particular embodiments, R1and / or R2may contain a tertiary amine linkage. In separate or further embodiments, R1and / or R2may contain an ester or amide linkage. In separate or further embodiments, at least one (or both)of R1and R2is / are branched alkyl groups or R1and R2is / are linear or branched alkenyl groups or linear or branched alkynyl groups. In any of the foregoing embodiments, R1and R2may be the same or different. Notably, as the L’-NR'R2moiety in Formula (3) further specifies the group R, the L'-NR ’ R2moiety should contain a total number of carbon atoms as defined above for the group R, i.e., at least four carbon atoms and at least one linkage in which at least three carbon atoms are connected together, and wherein R optionally contains one or more heteroatoms selected from O, N, and S atoms.

[0044] In some embodiments, at least one of R1and R2in Formula (3) has the following formula:— (CH2) -L'-R (3a).

[0045] The variable L1in Formula (3a) is a linker containing 1-3 atoms, wherein at least one of the atoms is a heteroatom selected from O and N atoms, such as described in detail above for the variable L under Formula (1). Any of the L groups described above under Formula (1) may serve as L1in Formula (3a). In particular embodiments, L1is C(O)X, wherein X is O, NH, or S (e.g., L1is -C(O)O- or -C(O)NH-). In some embodiments, R1and R2both independently have the Formula (3a), wherein R1and R2may be the same or different,

[0046] The variable R3in Formula (3a) is independently selected from any of the R groups described above under Formula (1), i.e., a linear or branched hydrophobic group containing at least four carbon atoms and at least one linkage in which at least three carbon atoms are connected together, and optionally containing one or more heteroatoms selected from O, N, and S atoms. Any of the R groups described above under Formula (1) may serve as R3in Formula (3a).

[0047] The variable p in Formula (3a) is an integer of 1-30. In different embodiments, p is an integer of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 15, 20, or 30 or p is within a range bounded by any two of these values, e.g., 1-20, 1-10, 1-6, 1-4, 1-3, 2-20, 2-10, 2-6, 2-4, 3-20, 3-10, or 3- 6. Notably, any particular p value or range thereof may be combined with any possible selection of Ra, Rb, Rc, and Rdand this may be combined with any selection of L, L’, L1, R3, R1and R2disclosed above.

[0048] In some embodiments, the lipid composition has a structure within the following sub-formula of Formula (1):

[0049] In Formula (4) above, the variables L, L’, R1, R2, Ra, Rb, Rc, Rd, and n are as defined above under Formulas (1) and (3). The variable Z can be CH, N, or a branched alkyl linker containing 4-30 carbon atoms and optionally one, two, three, or more tertiary amine linkages. In some embodiments, compounds of Formula (4) in which Z is CH are excluded.

[0050] In some embodiments, for any of Formula (3) or (4), the R1and R2groups may be selected from any of the following groups, wherein R1and R2may be the same or different:R and / or R2=wherein n in any of the foregoing formulas is independently an integer of 1-30.

[0051] In some embodiments of Formula (4), R1and R2 each include a branched structure containing at least one tertiary amine linkage. Compounds having these features may be represented by the following sub-formula of Formula (4):

[0052] In Formula (4a) above, the variables L, L’, R1, R2, Ra, Rb, Rc, Rd, and n are as defined above under Formulas (1) and (3). The variables L3and L4, which may be the same or different, are independently selected from a linker of the formula -(CH2)r- , wherein r is an integer of 1-6. The variables R7, R8, R9, and R10, which may be the same or different, are independently selected from linear or branched alkyl, linear or branched alkenyl, or linear or branched alkynyl groups containing at least one, two, three, or four carbon atoms and up to 20, 30, 40, or 50 carbon atoms and optionally containing one or more heteroatoms selected from O, N, and S atoms. In some embodiments, any one or more (or each one) of R7, R8, R9, and R10may contain at least one heteroatom-containing linkage, such as one or more ether (-O-), thioether (-S-), carbonyl (-C(O)-), thiocarbonyl (-C(S)-), ester (-C(O)O-), thioester (-C(S)O-), carbonate (-OC(O)O-), secondary amine (-NH-), tertiary amine (— N— ), imine (-N=), diazo (-N=N-), disulfide (-S-S-), amide (-C(O)NH-), urea (-NHC(O)NH-), and / or thiourea (-NHC(S)NH-) linkages.

[0053] In another aspect, the present disclosure is directed to a lipid nanoparticle (LNP) composition that contains any of the lipid compositions according to Formula (1) or subformula thereof and a therapeutic molecule. Some LNPs of the art are described in, for example, X. Hou et al., Nature Reviews Materials, 6, 1078-1094, 2021, the contents of which are herein incorporated by reference. The term “lipid nanoparticle” refers to nanoparticles constructed, at least in part, of lipid molecules. As further described below,the lipid molecules include one or more lipid compositions described herein according to Formula (1) or sub-formula thereof and one or more lipid not within the scope of Formula (1). The lipid not within the scope of Formula (1) may be, for example, one or more noncationic lipids, cationic or ionizable lipids, and / or cholesterol. In different embodiments, the LNP has a size of precisely, about, at least, or up to, for example, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 125, 150, 175, 200, 225, or 250 microns, or a size within a range bounded by any two of the foregoing values.

[0054] In some embodiments, the LNP includes at least one zwitterionic polymer- containing lipid. In the zwitterionic polymer-containing lipid, a lipid moiety is covalently attached (i.e., linked) to a zwitterionic polymer. The zwitterionic polymers are generally derived from zwitterionic monomers, as well as monomers that can be converted to zwitterionic monomers, i.e., precursors of zwitterionic monomers. Zwitterionic monomers are electronically neutral monomers that include equal numbers of positive and negative charges (e.g., one of each). In some embodiments, the zwitterionic polymer contains a plurality of repeating units, each repeating unit comprising one positive and one negative charged moiety. The zwitterionic polymer typically contains at least or greater than 2, 5, or 10, and up to or less than 100, 200, 300, 400, 500, or 1000 units. For purposes of the present invention, the zwitterionic polymer has a molecular weight (Mwor Mn) of no more than or less than or about 4 kDa, 3.5 kDa, 3 kDa, 2.5 kDa, or 2 kDa.

[0055] ‘ ‘As used herein, the term “zwitterionic polymer” refers to a polymer prepared by polymerizing a polymerizable zwitterionic monomer, which provides a zwitterionic polymer having 100 mole percent zwitterionic moieties (i.e., each repeating unit of the zwitterionic polymer is a zwitterionic moiety; or refers to a polymer prepared by copolymerizing a polymerizable zwitterionic monomer and a polymerizable comonomer, which provides a zwitterionic polymer having less than 100 mole percent zwitterionic moieties (e.g., when the polymerizable zwitterionic monomer and the polymerizable comonomer are present in equal proportions in the polymerization mixture, the product is a zwitterionic polymer having 50 mole percent zwitterionic moieties).

[0056] The term “zwitterionic polymer” also refers to a polymer having a substantially equal number of negative (anionic) charges and positive (cationic) charges that is prepared by copolymerizing a polymerizable negatively charged monomer and a polymerizable positively charged monomer, each present in substantially equal proportions in thepolymerization mixture. The product of such a copolymerization is a zwitterionic polymer having 100 mole percent zwitterionic moieties, where each zwitterionic moiety is defined as a pair of repeating units: a repeating unit having a negative charge and a repeating unit having a positive charge. Such zwitterionic polymers are referred to as mixed charge copolymers. The term “zwitterionic polymer” also refers to a polymer prepared by copolymerizing a polymerizable negatively charged monomer, a polymerizable positively charged monomer, each present in substantially equal proportions in the polymerization mixture, and a polymerizable comonomer, which provides a zwitterionic polymer having less than 100 mole percent zwitterionic moieties (e.g., when the combination of polymerizable negatively charged monomer and polymerizable positively charged monomer and the polymerizable comonomer are present in equal proportions in the polymerization mixture (i.e., 50% combination of polymerizable negatively charged monomer and polymerizable positively charged monomer and 50% polymerizable comonomer), the product is a zwitterionic polymer having 50 mole percent zwitterionic moieties).”

[0057] The lipid moiety in the zwitterionic polymer-containing lipid is constructed of a polyol portion (e.g., a diol, glycerol, phosphatidylglycerol, phosphatidylethanolamine, or phosphatidylserine) that has been esterified with one or two fatty acid molecules to result in a monoacyl or diacyl lipid, wherein the term “acyl” refers to a RC(=O) group in which R is a linear or branched hydrocarbon (fatty) chain containing at least eight and typically up to 30 carbon atoms, wherein the hydrocarbon chain may be saturated or contain one or more carbon-carbon double bonds. The lipid moiety may be, for example, a diacyldiol (e.g., diacylethyleneglycol), diacylglycerol (diacylglyceride), diacylphosphatidylglycerol, diacylphosphatidylethanolamine, or diacylphosphatidylserine moiety. The lipid may be any of the lipids described in any one of Examples 1-14 provided in this application. The lipid may also be an of the lipids described in WO2011 / 057227, which is herein incorporated by reference. The fatty acyl portion may be derived from any of the known fatty acids. Some examples of fatty acyl portions include oleoyl, palmitoyl, lauryl, myristoyl, stearoyl, linoleoyl, and arachidonyl. The zwitterionic polymer is attached to the lipid moiety, such as any of the lipid moieties described above, typically via a carbon on the polyol. The zwitterionic polymer may contain the zwitterionic groups in side chains or the backbone (or combination thereof) of the polymer. In particular embodiments, the lipid in the zwitterionic polymer-containing lipid is 1,2-dimyristoyl-rac-glycero- (DMG) or 1,2- distearoyl-rac-glycero- (DSG).

[0058] In one embodiment, the zwitterionic polymer is a homopolymer prepared from zwitterionic monomers and has the formula:wherein B is a polymer backbone, such as a polyester, polyether, polyurethane, polyamide, or polyhydrocarbon (e.g., polyethylene or polypropylene) backbone, and P is a zwitterionic moiety. In particular embodiments, the backbone (B) may have any of the following structures:wherein R is selected from the group consisting of hydrogen and substituted or unsubstituted alkyl; and E is selected from the group consisting of substituted or unsubstituted alkylene, -(CH2)PC(O)O-, and -(CH2)PC(O)NR2-; p is typically an integer from 0 to 12; R2is selected from hydrogen and substituted or unsubstituted alkyl; and L is a straight or branched alkylene group optionally including one or more oxygen atoms. The subscript x is typically at least or greater than 2, 5, or 10, and up to or less than 100, 200, 300, 400, 500 or 1000 units.

[0059] In particular embodiments, P is selected from any of the following structures:wherein R3, R4, and R6are independently selected from the group consisting of hydrogen and substituted or unsubstituted alkyl group, R5or R5 is selected from the group consisting of substituted or unsubstituted alkylene, phenylene, and polyether groups, and m is an integer from 1 to 7; and x is an integer from 2 to 500.

[0060] In some embodiments, the zwitterionic polymer is a betaine polymer. In other embodiments, the zwitterionic polymer is a poly(phosphatidylcholine) polymer, poly(trimethylamine N-oxide) polymer, poly(zwitterionic phosphatidylserine) polymer, or glutamic acid-lysine (EK) -containing polypeptide. In some embodiments, zwitterionic phosphatidyl serine comprises one neighboring positive charged moiety to balance the negative charge of the phosphoserine. In some embodiments, zwitterionic phosphatidyl serine comprises a compound as described in “De novo design of functional zwitterionic biomimetic material for immunomodulation’’ Science Advances, 29 May 2020, Vol. 6, Issue 22, (DOI: 10. 1126 / sciadv.aba0754) which is hereby incorporated by reference in its entirety.

[0061] Some examples of betaine polymers include poly(carboxybetaine), poly(sulfobetaine), and poly(phosphobetaine) polymers. Suitable poly(carboxybetaine)s can be prepared from one or more monomers selected from, for example, carboxybetaine acrylates, carboxybetaine acrylamides, carboxybetaine vinyl compounds, carboxybetaine epoxides, and mixtures thereof. In one embodiment, the monomer is carboxybetaine methacrylate. Representative monomers for making carboxybetaine polymers useful in the invention include carboxybetaine methacrylates, such as 2-carboxy-N,N-dimethyl-N-(2’- methacryloyloxyethyl) ethanaminium inner salt; carboxybetaine acrylates; carboxybetaine acrylamides; carboxy betaine vinyl compounds; carboxybetaine epoxides; and other carboxybetaine compounds with hydroxyl, isocyanates, amino, or carboxylic acid groups. In a particular embodiment, the polymer is a poly(carboxybetaine methacrylate) (poly(CBMA)). In particular embodiments, the zwitterionic polymer is a PCB polymer having a molecular weight (Mwor Mn) of no more than or less than 4 kDa, 3.5 kDa, 3 kDa, 2.5 kDa, or 2 kDa.

[0062] The zwitterionic polymer can be prepared by any suitable polymerization method, such as atom transfer radical polymerization (ATRP), reversible addition fragmentation chain transfer (RAFT) polymerization, and free radical polymerization. Any suitable radical initiators for polymerizing such monomers including those well known in the art,may be used. In some embodiments, to prepare the zwitterionic polymer-containing lipid or other polymer-containing lipid, a zwitterionic or other monomer or precursor thereof is attached to a lipid, and the monomer is polymerized while attached to the lipid. Alternatively, an already produced polymer may be attached to a lipid by means well known in the art.

[0063] In another embodiment, the zwitterionic polymer is a homopolymer that has a positive charge in the polymer backbone and a pendant carboxylic acid group and has the formula:wherein R is selected from the group consisting of hydrogen and substituted or unsubstituted alkyl; Li and L2 are independently a straight or branched alkylene group optionally including one or more oxygen atoms; and x is an integer from 2 to 500.

[0064] In another embodiment, the zwitterionic polymer is a mixed charge copolymer and has the general formula:wherein Bi and B2 are independently selected from Xi, X2, and X3 as described earlier above; R is selected from hydrogen and substituted or unsubstituted alkyl; E is selected from substituted or unsubstituted alkylene, -(CH2)PC(O)O-, and -(CH2)PC(O)NR2-, wherein p is an integer from 0 to 12; R2is selected from hydrogen and substituted or unsubstituted alkyl; L is a straight or branched alkylene group optionally including one or more oxygen atoms; Pi is a positively charged group; P2 is a negatively charged group, such as a carboxylic acid group; m is an integer from 1 to 500; and n is an integer from 1 to 500. In some embodiments, Pi is nitrogen in an aromatic ring or NR5R6, wherein R5 and Re are independently substituted or unsubstituted alkyl group.

[0065] The positively charged unit (Pi containing unit) of the zwitterionic polymer can be derived from a monomer having a positively charged pendant group. Representative monomers that can be used to derive the positively charged unit in the polymers of the present invention include 2-(dimethylamino)ethyl methacrylate, 2-(diethylamino)ethyl methacrylate, [2-(methacryloyloxy)ethyl] trimethylammonium chloride, and N- acetylglucosamine.

[0066] In one embodiment, the negatively charged unit of the zwitterionic polymer is derived from 2-carboxyethyl acrylate (CA), and the positively charged unit is derived from 2-(dimethylamino)ethyl methacrylate (DM). In another embodiment, the negatively charged unit is derived from 2-carboxyethyl acrylate (CA), and the positively charged unit is derived from 2-(diethylamino)ethyl methacrylate (DE). In another embodiment, the negatively charged unit is derived from 2-carboxyethyl acrylate (CA), and the positively charged unit is derived from [2-(methacryloyloxy)ethyl]trimethylammonium chloride (TM). In another embodiment, the negatively charged unit is derived from 2-carboxyethyl acrylate (CA), and the positively charged unit is derived from 2-aminoethyl methacrylate hydrochloride (NH2).

[0067] In some embodiments, the zwitterionic polymer excludes a poly alkylene oxide (polyalkylene glycol) segment, or the zwitterionic polymer more specifically excludes a polyethylene oxide or polypropylene oxide segment. In some embodiments, all of component (i) excludes a polyalkylene oxide segment, or component (i) more specifically excludes a polyethylene oxide or polypropylene oxide segment. In some embodiments, the lipid nanoparticle (LNP) as a whole excludes a polyalkylene oxide segment or molecule.

[0068] The term “lipid” (i.e., in the zwitterionic polymer-containing lipid) refers to organic compounds that include, but are not limited to, esters of fatty acids and are characterized by being insoluble in water, but soluble in many organic solvents. They are usually divided into at least three classes: (1) “simple lipids,” which include fats and oils as well as waxes; (2) “compound lipids,” which include phospholipids and glycolipids; and (3) “derived lipids” such as steroids. In one embodiment, lipid includes diacylglyceride.

[0069] In one embodiment, the zwitterionic polymer is linked with “compound lipids”. Some examples of such lipids include dioleoylphosphatidylglycerol (DOPG), dipalmitoylphosphatidylglycerol (DPPG), dioleoylphosphatidylethanolamine (DOPE), palmitoyloleoylphosphatidylethanolamine (POPE), dipalmitoylphosphatidylethanolamine(DPPE), dimyristoylphosphoethanolamine (DMPE), distearoylphosphatidylethanolamine (DSPE), 16-O-monomethyl-phosphoethanolamine, 16-O-dimethyl-phosphoethanolamine, 18-1-trans-phosphoethanolamine, l-stearoyl-2-oleoyl-phosphatidyethanolamine (SOPE), and l,2-dioleoyl-sn-glycero-3-phophoethanolamine (transDOPE). In another embodiment, the zwitterionic polymer is linked with “simple lipids”. In another embodiment, the zwitterionic polymer is linked with “derived lipids”. In some embodiments, the zwitterionic polymer comprises zwitterionic compounds as disclosed in WO2011057225A2, which is incorporated herein by reference.

[0070] In separate or further embodiments, the LNP includes at least one non-cationic lipid selected from charged and uncharged lipids not attached to a polymer. The term “noncationic lipid,” as used herein, refers to a lipid that is not positively charged and not capable of being ionized to a positively charged state. However, the non-cationic lipid may be neutral charged by containing a zwitterion (positive and negative charge within the polymer), such as any of the zwitterionic groups and moieties described earlier above.

[0071] In some embodiments, the non-cationic lipid contains a zwitterionic moiety. The zwitterionic moiety can be any such moieties described in detail earlier above. The zwitterionic moiety may be, for example, a phosphobetaine, phosphatidylcholine, carboxybetaine, sulfobetaine, trimethylamine N-oxide, glutamic acid-lysine (EK)- containing, or zwitterionic phosphatidyl serine (phosphoserine) moiety, or a combination thereof. In some embodiments, the zwitterionic lipid is a phospholipid, such as a phosphatidylcholine or phosphatidylserine lipid.

[0072] In other embodiments, the non-cationic lipid is not zwitterionic but negatively charged by containing a negatively charged group (e.g., phosphoserine). A metal (e.g., alkali) or ammonium counteranion may be ionically and fluxionally associated with the negatively charged group.

[0073] In another set of embodiments, the non-cationic lipid is uncharged by not containing any charged groups. The uncharged non-cationic lipid may be, for example, a phosphatidylglycerol lipid, phosphatidylethanolamine lipid, or sphingolipid. The noncationic lipid may, in some embodiments, be a simple lipid, such as a fat, oil, and / or wax. The non-cationic lipid may, in some embodiments, be a compound lipid, such as a phospholipid or glycolipid. The non-cationic lipid may, in some embodiments, be a derived lipid, such as a steroid, a phospholipid, a sphingolipid, and / or a sterol. In someembodiments, the non-cationic lipid is selected from a diacylphosphatidylethanolamine, a ceramide, a sphingomyelin, a dihydrosphingomyelin, a cephalin, or a cerebroside. In particular embodiments, the non-cationic lipid is selected from one or more of a phosphatidylethanolamine (PE), a phosphatidylglycerol (PG), a phosphatidic acid (PA), or a phosphatidylinositol (PI). In other particular embodiments, the non-cationic lipid is a dioleoylphosphatidylglycerol (DOPG), dipalmitoylphosphatidylglycerol (DPPG), dioleoylphosphatidylethanolamine (DOPE), palmitoyloleoyl-phosphatidylethanolamine (POPE), dipalmitoylphosphatidylethanolamine (DPPE), dimyristoylphosphoethanolamine (DMPE), distearoylphosphatidylethanolamine (DSPE), 16-O-monomethyl-phospho ethanolamine, 16-O-dimethyl-phosphoethanolamine, 18-1-trans-phosphoethanolamine, 1- stearoyl-2-oleoyl phosphatidyethanolamine (SOPE), and 1,2-dioleoyl-sn glycero-3- phophoethanolamine (transDOPE). The non-cationic lipid may or may not be attached to a polymer. In the event the non-cationic lipid is attached to a polymer, the non-cationic lipid preferably excludes a polyalkylene oxide (polyalkylene glycol) segment.

[0074] In particular embodiments, non-cationic lipids containing an ionic moiety are phospholipids. In one embodiment, the non-cationic lipid containing an ionic moiety is a lipid conjugated with one or more carboxy betaine groups. In one embodiment, the noncationic lipid containing an ionic moiety is a lipid conjugated with one or more sulfobetaine groups. In one embodiment, the non-cationic lipid containing an ionic moiety is a lipid conjugated with one or more trimethylamine N-oxide groups.

[0075] In separate or further embodiments, the LNP may optionally include one or more cationic or ionizable lipids. The cationic or ionizable lipid may or may not contain a lipid attached to a polymer that has a cationic group. In some embodiments, the cationic or ionizable lipid is not attached to a polymer. The term “cationic lipid,” as used herein, refers to a positively charged lipid (typically, by possessing a quaternary ammonium group). In the cationic lipid, the positively charged group is not associated with a negative charge within the cationic lipid. Thus, the cationic lipid is not a zwitterionic lipid. The term “ionizable lipid,” as used herein, refers to lipids that contain one or more groups capable of being ionized to result in a positive charge in the polymer. The ionizable lipid generally possesses a secondary or tertiary amino group, or particularly an alkylated amine, or more particularly, a monoalkylamine or dialkylamine group, any of which can be protonated or alkylated to result in an alkylated ammonium group. The cationic lipid may contain atrialkylamine group, which is necessarily positively charged when bound to the lipid. In particular embodiments, the cationic or ionizable lipid possesses a dimethylamino or trimethylamino (or dimethylammonium or trimethylammonium) group. In particular embodiments, the cationic or ionizable lipids are selected from l,2-dioleoyl-3- dimethylammonium-propane (DODAP), 1 ,2-dilinoleyloxy-N,N-dimethyl-3-aminopropane (DLinDMA), 2,2-dihnoleyl-4-(2-dimethylaminoethyl)-[l ,3]-dioxolane (DLinKC2DMA), and [(6Z,9Z,28Z,3 lZ)-heptatriaconta-6,9,28,31 -tetraen- 19-yl] 4-(dimethylamino)butanoate (DLinMC3DMA). In some embodiments, the LNP excludes a cationic or ionizable lipid.

[0076] In some embodiments, the ionizable lipid contains a lipid moiety attached to a secondary or tertiary amine group along with a functional group, which is negatively charged under physiological conditions. This lipid moiety can beRN-L-A-(X)nR2n = 0 or 1 whereR1 / R2 = H or an alkyl group. The alkyl can be saturated or unsaturated, branched or unbranched, all-carbon and hydrogen or containing heteroatoms such as but not limited to N, O, F, Si, P, S, Cl, Br, and F.L is a covalent linker group between N and A. The linker may be all carbon and hydrogen, or containing heteroatoms such as but not limited to N, O, F, Si, P, S, Cl, Br and F. The structure of L is exemplified by, but not limited to: -CH2-, - CH2CH(OH)-, -CH2CHCICH2-, -CH2OCH2-, -CH2SCH2-, -CH2SSCH2-, - CH2COOCH2-.A-(X)n is a functional group that is negatively charged under certain pH conditions. Structures are exemplified by, but not limited to the following:(i) A(X)n is a carboxylic acid group (A = -COOH and n=0) or(ii) A(X)n is a phosphate, where A =and n = 1. X = H or an alkyl group that is saturated or unsaturated, branched or not branched, all-carbon or containing heteroatoms such as but not limited to N, O, F, Si, P, S, Cl, Br, and F or(iii) A(X)n is a sulfonic acid group, where A =0-S ii-OH0 and n = 0(iv) A(X)n is a sulfonamide group, where A = T oNoA o and n = 1, X = H or an alkyl group that is saturated or unsaturated, branched or not branched, all carbon and hydrogen or containing heteroatoms such as but not limited to N, O, F, Si, P, S, Cl, Br, and F.Some examples of ionizable (tertiary amine) lipids include:Some examples of cationic (quaternary amine) lipids include:Where n=l to 10

[0077] In some embodiments, the cationic or ionizable lipid excludes a polyalkylene oxide (polyalkylene glycol) segment, or the cationic or ionizable lipid more specifically excludes a polyethylene oxide or polypropylene oxide segment. In some embodiments, the cationic or ionizable lipid excludes a polyalkylene oxide segment, or the cationic or ionizable lipid more specifically excludes a polyethylene oxide or polypropylene oxide segment. In some embodiments, the lipid nanoparticle excludes a polyalkylene oxide segment.

[0078] In separate or further embodiments, the LNP includes cholesterol or a derivative thereof, which is considered herein to be an optional further component of the LNP. In certain embodiments, the cholesterol derivative is a phytosterol, e.g., P-sitosterol, campesterol, stigmasterol, fucosterol, or stigmastanol. In certain embodiments, the cholesterol derivative is dihydrocholesterol, ent-cholesterol, epi-cholesterol, desmosterol, cholestanol, cholestanone, cholestenone, cholesteryl-2'-hydroxyethyl ether, cholesteryl-4'- hydroxybutyl ether, 3P[N — (N'N'-dimethylaminoethyl)carbamoyl cholesterol (DC-Chol), 24(S)-hydroxycholesterol, 25-hydroxycholesterol, 25(R)-27-hydroxycholesterol, 22- oxacholesterol, 23 -oxacholesterol, 24-oxacholesterol, cycloart enol, 22-ketosterol, 20- hydroxysterol, 7-hydroxycholesterol, 19-hydroxycholesterol, 22-hydroxycholesterol, 25- hydroxycholesterol, 7-dehydrocholesterol, 5a-cholest-7-en-3P-ol, 3,6,9-trioxaoctan-l-ol- cholesteryl-3e-ol, dehydroergosterol, dehydroepiandrosterone, lanosterol, dihydrolanosterol, lanostenol, lumisterol, sitocalciferol, calcipotriol, coprostanol, cholecalciferol, lupeol, ergocalciferol, 22-dihydroegocalciferol, ergosterol, brassicasterol, tomatidine, tomatine, ursolic acid, cholic acid, chenodeoxycholic acid, zymosterol, diosgenin, fucosterol, fecosterol, or fecosterol, or a salt or ester thereof, e.g., sodium cholate. In some embodiments, the LNP contains an ionizable or cationic lipid, DSPC, cholesterol, and zwitterionic polymer-containing lipid in a molar ratio of 50:40:38.5-x:x, wherein x is a value within a range of 1 .5-5, or more particularly, x may be precisely or about 1 .5, 2, 2.5, 3, 3.5, 4, 4.5, 5, or 5.5, or within a range bounded by any two of the foregoing values.

[0079] In some embodiments, the LNP includes at least the lipid within the scope of Formula (1) or sub-formula therein and one or more lipids not within the scope of Formula (1), along with the therapeutic substance. The lipid not within the scope of Formula (1) may be selected from any of those described above, such as, for example, ionizable (tertiary amine) lipids, cationic (quaternary amine) lipids, and / or non-cationic lipids. In someembodiments, the one or more lipids not within the scope of Formula (1) include one or more lipids selected from: (i) tertiary amine (ionizable) lipids, (ii) non-tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group; (iii) lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, and / or (iv) sterols.

[0080] In some embodiments, the LNP includes a tertiary amine (ionizable) lipid, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance. Some examples of tertiary amine lipids include those already described above. Some particular examples of tertiary amine lipids include ALC-0315 lipid, SM-102 lipid, and DLin-MC3-DMA lipid, which have the following structures:

[0081] In some embodiments, the LNP includes a non-tertiary amine lipid containing a hydrophobic tail attached to a hydrophilic head group, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance. In some embodiments, the LNP includes a tertiary amine lipid and non-tertiary amine lipid, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance. The non-tertiary amine lipid may be, for example, a phospholipid or nonphospholipid (e.g., carboxybetaine lipid). The non-tertiary amine is often zwitterionic, such as any of the zwitterionic lipids described above. Some examples of non-tertiary amine lipids include distearoylphosphatidylcholine (DSPC), dimyristoylphosphatidylcholine (DMPC), l,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE), 1,2-dioleoyLsn- glycero-3 -phosphocholine (DOPC), dioleoylphosphatidyl glycerol (DOPG), dipalmitoylphosphatidyl glycerol (DPPG), palmitoyloleoyl-phosphatidyl ethanolamine (POPE), dipalmitoylphosphatidyl ethanolamine (DPPE), dimyristoylphosphoethanolamine (DMPE), distearoylphosphatidyl ethanolamine (DSPE), 16-O-monomethyl- phosphoethanolamine, 16-O-dimethyl-phosphoethanolamine, 18-1- transphosphoethanolamine, l-stearoyl-2-oleoyl phosphatidy ethanolamine (SOPE), and 1,2- dioleoyl-sn glycero-3-phophoethanolamine (transDOPE).

[0082] In some embodiments, the LNP includes one or more lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, in addition to the lipid within the scope of Formula (1 ) or sub-formula therein and therapeutic substance. In some embodiments, the LNP includes a lipid containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail and a tertiary amine lipid, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance. In other embodiments, the LNP includes a lipid containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail and a non-tertiary amine lipid, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance. In other embodiments, the LNP includes a lipid containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, a tertiary amine lipid, and non-tertiary amine lipid, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance. In lipids containing an uncharged polymeric group attached to a hydrophobic tail, the uncharged polymeric group may be, for example, a polyalkylene oxide (e.g., a polyethylene glycol) group. In lipids containing a zwitterionic polymeric group attached to a hydrophobic tail can be any of the zwitterionic polymer-containing lipids described above. In particular embodiments, the lipids containing a zwitterionic polymeric group attached to a hydrophobic tail may be one or more of poly(carboxybetaine) (PCB), poly(sulfobetaine), poly (phosphobetaine), poly(phosphatidylcholine), poly(trimethylamine N-oxide), and glutamic acid-lysine (EK) containing polypeptide.

[0083] In some embodiments, the LNP includes one or more sterols, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance. In some embodiments, the LNP includes one or more sterols and one or more lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance. In some embodiments, the LNP includes one or more sterols, one or more lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, and one or more tertiary amine lipids, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance. In other embodiments, the LNP includes one or more sterols, one or more lipids containing an uncharged or zwitterionic polymeric groupattached to a hydrophobic tail, and one or more non-tertiary amine lipids, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance. In other embodiments, the LNP includes one or more sterols, one or more lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, one or more tertiary amine lipids, and one or more non-tertiary amine lipids, in addition to the lipid within the scope of Formula (1) or sub-formula therein and therapeutic substance.

[0084] As well known, the sterols are a family of lipophilic molecules typically bearing four or more fused rings and which typically function as precursors for steroid hormones. The sterols typically contain a cholestan-3-ol moiety and have a hydroxyl group at carbon 3 in the fused ring system. In typical embodiments, the sterol is cholesterol or a cholesterol derivative. The derivative of cholesterol typically preserves the four-ring fused system of cholesterol but may have the OH group or alkene bond removed, changed in position, altered (e.g., OH may be esterified), or one or more additional groups (e.g., OH or double bond) may be included. Some examples of cholesterol derivatives include 20- hydroxycholesterol, 22-hydroxycholesterol, 24-hydroxycholesterol, 25-hydroxycholesterol, 27 -hydroxy cholesterol, 7a-hydroxycholesterol, 7 P-hydroxy cholesterol, 7-ketocholesterol, 5,6a-epoxycholesterol, 5,6P-epoxycholesterol, lanosterol, dihydrolanosterol, lathosterol, zymosterol, A7-cholesterol, A8-cholesterol, A8,14-dimethylsterol, A5,7-cholestadienol, and A8,24-dimethylsterol. Any of the exemplary LNP compositions described above that include a sterol may select cholesterol or one or more of the above cholesterol derivatives as the sterol.

[0085] In a first set of embodiments, the LNP includes a lipid within the scope of Formula (1) along with at least one, two, or three (or all) of: i) tertiary amine (ionizable) lipids, (ii) non-tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group; (iii) lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, and (iv) sterols, along with therapeutic substance. In a second set of embodiments, the LNP includes a lipid within the scope of Formula (1) and a tertiary amine (ionizable) lipid along with at least one, two, or three (or all) of: (ii) non-tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group; (iii) lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, and (iv) sterols, along with therapeutic substance. In a third set of embodiments, the LNP includes a lipid within the scope of Formula (1) and a non-tertiary amine (ionizable) lipidalong with at least one, two, or three (or all) of: (i) tertiary amine (ionizable) lipids, (iii) lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, and (iv) sterols, along with therapeutic substance. In a fourth set of embodiments, the LNP includes a lipid within the scope of Formula (1) and one or more lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail along with at least one, two, or three (or all) of: (i) tertiary amine (ionizable) lipids, (ii) non-tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group: and (iv) sterols, along with therapeutic substance. In a fifth set of embodiments, the LNP includes a lipid within the scope of Formula (1) and one or more sterols, along with at least one, two, or three (or all) of: (i) tertiary amine (ionizable) lipids, (ii) non-tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group: and (iii) lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail.

[0086] In some embodiments, the membrane-destabilizing zwitterionic (MeDZ) lipids possess a multi-tailed structure and can switch from zwitterionic to positive charge at a pH between 5.5-7.0. As an example, the tails can be designed based on ionizable lipid ALC- 0315, SM102 and DLin-MC3-DMA. Substituents can be introduced into the ortho- and para- positions of the pyridine ring in the headgroup, including H, F, Cl, Br or OCH3, to adjust the pKa values of the MeDZ lipids. The following schemes show some exemplary chemical structures of the MeDZ lipids along with the mechanism involved in switching from zwitterionic to positive charge at a pH within a range of 5.5-7.

[0087] In some embodiments, any of the above MeDZ lipids is mixed with one or more cargoes (therapeutic substances) to form one-component cargo-lipid nanoparticles (LNPs), wherein the cargo may be selected from, for example, RNAs, DNAs, CRISPR / Cas RNP, and numerous others as described earlier above. In some embodiments, any of the MeDZ lipids is mixed with one or more helper lipids and one or more cargoes to form two- component cargo-LNPs. Helper lipids are a well known class of lipid components in LNPs that include various biocompatible phospholipids. The helper lipid may be selected from, for example, distearoylphosphatidylcholine (DSPC), l,2-dioleoyl-sn-glycero-3- phosphoethanolamine (DOPE), l,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), or carboxybetaine (CB) lipids. In some embodiments, one or more lipids in the one- component or two-component lipids or any other type of LNP described above may be attached with functional ligands, such as targeting ligands to the cells, including peptides (e.g., RGD, TAT, LyP-1, Ang), proteins (e.g., anti-CD3. anti-CD8, anti-CD19, anti-CD47), carbohydrates (e.g., dextran, mannose, chitosan, galactose, hyaluronic acid), and small molecules (e.g., phospho-L-serine (PS)). In some embodiments, any of the LNPs described herein may further include functional lipids, such as l,2-dioleoyl-sn-glycero-3-phospho-L- serine (DOPS), sphingomyelin (SM), and peptide or antibody-functionalized lipids. As mentioned above, functional lipids include various types of lipids with targeting capability.In some embodiments, any of the LNPs described herein may further include one or more of the following components: (1) protecting lipids (e.g., l,2-dimyristoyl-rac-glycero-3- methoxypolyethylene glycol (DMG-PEG) and poly(carboxybetaine) (PCB)-based lipid); (2) cholesterol; and / or (3) ionizable lipids (e.g., ALC-0315 lipid, SM-102 lipid, and DLin- MC3-DMA lipid), to form 3-5 component cargo-LNPs. A five-component LNP may include, for example, MeDZ lipid + helper lipid + ionizable lipid + DMG-PEG + cholesterol, along with the therapeutic substance, such as an RNA. A three-component LNP may include, for example, MeDZ lipid + helper lipid + ionizable lipid, along with the therapeutic substance, such as an RNA. As other examples, a three-component LNP may include, for example, MeDZ lipid + helper lipid + DOPS, or MeDZ lipid + helper lipid + antibody-functionalized lipid, along with the therapeutic substance, such as an RNA.

[0088] The LNP further includes a therapeutic substance (which may be a therapeutic molecule) incorporated into or encapsulated by the LNP. The therapeutic substance may be incorporated into any of the lipid components described above. The therapeutic substance can be any substance having therapeutic value for a living organism, particularly a mammal, such as a human or animal subject. Exemplary therapeutic agents that may be used in accordance with the present invention include small molecules, organometallic compounds, nucleic acids (e.g., nucleotides or oligonucleotides), proteins (including multimeric proteins, protein complexes, peptides), lipids, carbohydrates, hormones, metals, compounds (e.g., drugs), radioactive elements and radioactive-containing compounds, vaccines, immunological agents, antibodies, and / or combinations thereof. The therapeutic substance may alternatively be any one of a nucleotide, gene editing system, protein, compound, and / or an antibody.

[0089] In particular embodiments, the therapeutic substance is a nucleic molecule, which is typically negatively charged. The nucleic molecule may be, for example, a nucleotide, nucleoside, nucleobase, or a nucleic acid (e.g., DNA or RNA). The therapeutic molecule is or includes one or more of DNA, RNA, ssDNA, dsDNA, ssRNA, dsRNA, and hybrids thereof. In some embodiments, the therapeutic molecule is or includes one or more of plasmid DNA or linearized DNA. In other embodiments, the therapeutic molecule is or includes one or more of messenger RNA (mRNA), small interfering RNA (siRNA), microRNA (miRNA), circular RNA (circRNA), and / or long-noncoding RNA (IncRNA). In other embodiments, the therapeutic molecule is or includes antisense oligonucleotide(ASO). In other embodiments, the therapeutic molecule is or includes Cas nuclease mRNA and guide RNA nucleic acid, wherein the guide RNA may be single-guide RNA (sgRNA). In particular embodiments, the therapeutic substance contains one or more such nucleic molecules. In some embodiments, the therapeutic substance contains RNA, or more particularly, mRNA, or more particularly viral mRNA. In other particular embodiments, the therapeutic substance is a spike protein of a virus, such as a coronavirus, SARS-COV2 (COVID-19), or HIV virus.

[0090] The lipid nanoparticles and compositions of the present invention may be used for a variety of purposes, including the delivery of nucleic acid molecules, ribonucleoprotein (RNP) and numerous other therapeutic substances. Examples of nucleic acids include messenger RNA (mRNA), ribosomal RNA (rRNA), transfer RNA (tRNA), antisense oligonucleotide (ASO), short interfering RNAs (siRNA), microRNA(miRNA), miRNA inhibitors (antagomirs / antimirs), messenger-RNA-interfering complementary RNA (micRNA), multivalent RNA, circular RNA (circRNA), crispr RNA (crRNA), long noncoding RNA (IncRNA), plasmid DNA, oligo DNA, and complementary DNA (cDNA), any one of which may function as a therapeutic molecule in the LNP. In particular embodiments, the therapeutic molecule comprises one or more of DNA, RNA, ssDNA, dsDNA, ssRNA, dsRNA, and hybrids thereof. In other embodiments, the therapeutic molecule comprises one or more of plasmid DNA or linearized DNA. In other embodiments, the therapeutic molecule comprises one or more of messenger RNA (mRNA), small interfering RNA (siRNA), microRNA (miRNA), circular RNA (circRNA), and / or long-noncoding RNA (IncRNA). In other embodiments, the therapeutic molecule comprises antisense oligonucleotide (ASO). In other embodiments, the therapeutic molecule comprises Cas nuclease mRNA and / or guide RNA nucleic acid. The guide RNA nucleic acid may be, for example, a single-guide RNA (sgRNA). Any one or more of the foregoing types of nucleic acids may be incorporated into LNPs of the present invention as a therapeutic substance. In some embodiments, the therapeutic substance may function as a vaccine against SARS-Cov-2, particularly wherein the vaccine is an mRNA vaccine or wherein the mRNA vaccine corresponds to a spike protein or portion thereof.

[0091] In any of the foregoing embodiments, the nucleotide may encode fusion biological moieties comprising protective domains and functional domains. In some embodiments, the functional domains are fused to the protective domains directly or via a linker consisting ofamino acids. In further particular embodiments, the protective domain may comprise: a plurality of negatively charged amino acids (e.g., aspartic acid, glutamic acid, and derivatives thereof); a plurality of positively charged amino acids (e.g., lysine, histidine, arginine, and derivatives thereof); and a plurality of additional amino acids independently selected from the group consisting of proline, serine, threonine, asparagine, glutamine, glycine, and derivatives thereof, wherein the ratio of the number of positively charged amino acids to the number of positively charged amino acids is from about 1:0.5 to about 1 :2. In some embodiments, the protective domain can be selected from other amino acid polymers (e.g., extended recombinant polypeptide (XTEN), proline-alanine-serine and elastin-like polypeptides). In any of the foregoing embodiments, the protective domain can be selected from natural half-life extension domains (e.g., Fc fragment).

[0092] In some embodiments, the therapeutic agent is a small molecule (e.g., organic compound) with pharmaceutical activity. In some embodiments, the small molecule is a clinically-used drug. In some embodiments, the drug is an anti-cancer agent, antibiotic, anti-viral agent, anti-HIV agent, anti-parasite agent, anti-protozoal agent, anesthetic, anticoagulant, inhibitor of an enzyme, steroidal agent, steroidal or non-steroidal antiinflammatory agent, antihistamine, immunosuppressant agent, anti-neoplastic agent, antigen, vaccine, antibody, decongestant, sedative, opioid, analgesic, anti-pyretic, birth control agent, hormone, prostaglandin, progestational agent, anti-glaucoma agent, ophthalmic agent, anti-cholinergic, analgesic, anti-depressant, anti-psychotic, neurotoxin, hypnotic, tranquilizer, anti-convulsant, muscle relaxant, anti-Parkinson agent, antispasmodic, muscle contractant, channel blocker, miotic agent, anti-secretory agent, antithrombotic agent, anticoagulant, anti-cholinergic, 0-adrenergic blocking agent, diuretic, cardiovascular active agent, vasoactive agent, vasodilating agent, anti-hypertensive agent, angiogenic agent, modulators of cell-extracellular matrix interactions (e.g. cell growth inhibitors and anti-adhesion molecules), or inhibitor of DNA, RNA, or protein synthesis.

[0093] In certain embodiments, a small molecule can be any drug. In some embodiments, the drug is one that has already been deemed safe and effective for use in humans or animals by the appropriate governmental agency or regulatory body. For example, diugs approved for human use are listed by the FDA under 21 C.F.R. §§330.5, 331 through 361, and 440 through 460, incorporated herein by reference; drugs for veterinary use are listedby the FDA under 21 C.F.R. §§500 through 589. incorporated herein by reference. All listed drugs are considered acceptable for use in accordance with the present invention.

[0094] A more complete listing of classes and specific drugs suitable for use in the present invention may be found in Pharmaceutical Drugs: Syntheses, Patents, Applications by Axel Kleemann and Jurgen Engel, Thieme Medical Publishing, 1 99 and the Merck Index: An Encyclopedia of Chemicals, Dings and Biologicals, Ed. by Budavari et al, CRC Press, 1996, both of which are incorporated herein by reference.

[0095] The LNP may also include lipid components with combined functionality. In embodiments, any of the lipid components, including but not limited to the zwitterionic polymer-containing lipid, the non-cationic lipid, the cationic lipid, and the cholesterol and / or cholesterol derivative, can include one or more functionalities of a different lipid component. In embodiments, the zwitterionic polymer-containing lipid includes the functionality of a zwitterionic polymer-containing lipid and a cationic or ionizable lipid. In embodiments, any of the zwitterionic polymer-containing lipid, the non-cationic or ionizable lipid, the cationic lipid, and the cholesterol and / or cholesterol derivative can also include one or more functionalities of one or more of the zwitterionic polymer-containing lipid, the non-cationic lipid, the cationic lipid, and the cholesterol and / or cholesterol derivative. In embodiments, a lipid component can include the functionality of another lipid component, thereby eliminating the need for an additional lipid component with that functionality. In embodiments, the zwitterionic polymer-containing lipid can include the functionality of the non-cationic lipid, thereby eliminating or reducing the need for the noncationic lipid. In embodiments, the zwitterionic polymer-containing lipid can include the functionality of the cationic lipid, thereby eliminating or reducing the need for the cationic lipid. In embodiments, the zwitterionic polymer-containing lipid can include the functionality of the cholesterol and / or cholesterol derivative, thereby eliminating or reducing the need for the cholesterol and / or cholesterol derivative.

[0096] In embodiments, any lipid component (zwitterionic polymer modified lipids, cationic lipids, non-cationic lipids, and cholesterol or its derivative) may be chemically combined with any other lipid components. In embodiments, the zwitterionic polymer modified lipid and cationic lipid are chemically combined into one lipid. In embodiments, the zwitterionic polymer modified lipid and non-cationic lipid are chemically combined into one lipid. In embodiments, the cationic lipid and non-cationic lipid are chemicallycombined into one lipid. In embodiments, the zwitterionic polymer modified lipid, cationic lipid and non-cationic lipid are chemically combined into one lipid. In embodiments, the zwitterionic polymer modified lipid and cholesterol or a derivative are chemically combined into one lipid. In embodiments, the zwitterionic polymer modified lipid, cationic lipid and cholesterol or a derivative are chemically combined into one lipid. In embodiments, the zwitterionic polymer modified lipid, non-cationic lipid and cholesterol or a derivative are chemically combined into one lipid. In embodiments, the zwitterionic polymer modified lipid, non-cationic lipid, non-cationic lipids and cholesterol or a derivative are chemically combined into one lipid. In other embodiments, each lipid component is separate, distinct, and not combined with another lipid component. In embodiments, the tertiary amine (ionizable) lipid or non-tertiary amine lipid is chemically combined with a lipid containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail.

[0097] Moreover, any of the components of the lipid nanoparticle described above may be chemically modified to contain a targeting ligand, such as an antibody or antibody fragment (e.g., Fab or Fc). In particular embodiments, the zwitterionic polymer-containing lipid possesses a targeting ligand to deliver LNPs loaded with a therapeutic or diagnostic agent or both of them to a targeted area within the special organ in the body. The targeting agent may be, for example, a peptide (e.g., RGD), a lipid (e.g., phosphoserine -containing lipid), a protein (e.g., apolipoprotein E), an aptamer (e.g., anti-VEGF aptamer), a sugar (e.g., Sialic acid), a compound, or an antibody (e.g., anti-PDl) or antibody fragment (e.g., Fab or Fc). In some embodiments, at least one component of the LNP, such as a lipid not within the scope of Formula (1), is / are selected from peptides (e.g., RGD, TAT, LyP-1, Ang), proteins (e.g., anti-CD3, anti-CD4, anti-CD8, anti-CD19, anti-CD47), carbohydrates (e.g., dextran, mannose, chitosan, galactose, hyaluronic acid), small molecules (e.g., phospho-L-serine (PS)), l,2-dioleoyl-sn-glycero-3-phospho-L-serine (DOPS), and sphingomyelin (SM)).

[0098] In another aspect, the present disclosure is directed to a method of delivering a therapeutic substance to a subject by administering to the subject any one or more of the LNP compositions described above. The subject is typically a mammal, more typically a human subject, but may also be another type of mammal, such as a pet or farm animal, such as a dog, cat, cow, or sheep. In some embodiments, the method of delivering the therapeutic substance results in a method of treating the subject. As a method of treatment, the LNP composition can be administered for the purpose of, for example, proteinreplacement therapy, cancer immunotherapy, cancer vaccine therapy, infectious disease vaccines, gene editing, autoimmune disease treatment, and / or cancer diagnosis. In particular embodiments, the LNP composition is administered for gene therapy comprising CRISPR-Cas gene editing, or for in vitro and in vivo production of extracellular vesicles, or for vaccination against coronavirus (e.g., SARS-CoV-2). In more particular embodiments, the LNP is used for clustered regularly interspaced short palindromic repeats-Cas endonuclease (CRISPR-Cas) gene editing in vitro and in vivo, for example including but not limited to delivering one or more nucleic acids that encode for one or more CRISPR associated proteins such as Cas protein. In some embodiments, the LNP is administered along with a checkpoint inhibitor (e.g., anti- Programmed death-ligand 1 (anti-PD-Ll) antibody or anti-cytotoxic T-lymphocyte-associated protein 4 (anti-CTLA4) to treat cancer.

[0099] In embodiments, the method results in a higher transfection efficiency compared to a lipid nanoparticle composition not containing the lipid composition of Formula (1) or subformula therein.

[0100] In some embodiments, the method of treatment includes targeted delivery of a therapeutic agent to a secondary lymphoid organ (SLO) in a subject, wherein the subject is administered lipid nanoparticles comprising a phosphoserine-containing lipid and the therapeutic agent. The SLO may be, for example, spleen and / or lymph nodes. The phosphoserine-containing lipid may be, for example, l,2-dioleoyl-sn-glycero-3-phospho-L- serine (DOPS), or a naturally-occurring PS-lipid, such as L-a-phosphatidylserine (brain). In some embodiments, the targeted delivery results in cancer immunotherapy, autoimmune disease immunotherapy, or gene editing.

[0101] The LNP composition is typically administered in the form of a pharmaceutical composition containing the LNP. In the pharmaceutical composition, the LNP may be dissolved or suspended in, or admixed with, a pharmaceutically acceptable carrier, which may be a liquid, semi-solid (e.g., gel or wax), or solid, as well known in the art. The phrase “pharmaceutically acceptable” refers herein to those compounds, materials, compositions, and / or dosage forms which are, within the scope of sound medical judgment, suitable for administration to a subject. Each carrier should be “acceptable” in the sense of being compatible with the other ingredients of the formulation and physiologically safe to the subject. Any of the carriers known in the art can be suitable herein depending on the mode of administration.

[0102] Some examples of pharmaceutically acceptable liquid carriers include alcohols (e.g., ethanol), glycols (e.g., propylene glycol and polyethylene glycols), polyols (e.g., glycerol), oils (e.g., mineral oil or a plant oil), paraffins, and aprotic polar solvents acceptable for introduction into a mammal (e.g., dimethyl sulfoxide or N-methyl-2-pyrrolidone) any of which may or may not include an aqueous component (e.g., at least, above, up to, or less than 10, 20, 30, 40, or 50 vol% water). Some examples of pharmaceutically acceptable gels include long-chain polyalkylene glycols and copolymers thereof (e.g., poloxamers), cellulosic and alkyl cellulosic substances (as described in, for example, U.S. Patent 6,432,415), and carbomers. The pharmaceutically acceptable wax may be or contain, for example, carnauba wax, white wax, bees wax, glycerol monostearate, glycerol oleate, and / or paraffins.

[0103] In some embodiments, the pharmaceutical composition contains solely the LNP and one or more solvents or the carrier. In other embodiments, the pharmaceutical composition includes one or more additional components. The additional component may be, for example, a pH buffering agent, mono- or poly-saccharide (e.g., lactose, glucose, sucrose, trehalose, lactose, or dextran), preservative, electrolyte, surfactant, or antimicrobial. If desired, a sweetening, flavoring, or coloring agent may be included. Other suitable excipients can be found in standard pharmaceutical texts, e.g. in “Remington’s Pharmaceutical Sciences”, The Science and Practice of Pharmacy, 19th Ed. Mack Publishing Company, Easton, Pa., 1995.

[0104] The LNP composition, typically in the form of a pharmaceutical composition in which the LNP is admixed with or suspended in a liquid or solid pharmaceutically acceptable carrier, can be administered to the subject by any suitable route. The LNP may be administered intravenously, orally, intramuscularly, intradermally, subcutaneously, intranasally, or by inhalation. In particular embodiments, the LNP is administered by injection into the subject. In some embodiments, the LNP is delivered to cells of the subject. In some embodiments, the LNP is delivered by removing cells from the subject, administering the lipid nanoparticle to the removed cells, and then reintroducing the removed cells to the subject. In some embodiments, the LNP is injected directly in vivo and delivered into the host cells in vivo. In other embodiments, the LNP is transfected into the host cells ex vivo and the resulting cells are then infused in vivo.

[0105] Examples have been set forth below for the purpose of illustration and to describe the best mode of the invention at the present time. However, the scope of this invention is not to be in any way limited by the examples set forth herein.Example 1Five-component LNPs formulated by MeDZ lipid, ionizable lipid, cholesterol, DMG-PEG, and helper lipid for enhanced mRNA expression

[0106] Design and Synthesis of an MeDZ Ionizable Lipid

[0107] The present work describes the design and synthesis of a Membrane-Destabilizing Zwitterionic (denoted MeDZ) ionizable lipid. The structure of the MeDZ lipid is shown in FIG. la. The MeDZ lipid minimized the reactogenicity of the final LNP vector while retaining its desirable mRNA delivery property. The MeDZ lipid features a hydrophilic headgroup composed of a pyridine-based carboxybetaine (PyCB), degradable hydrophobic multi-tailed alkyl chains, and a tertiary amine-based linker. As depicted in FIG. lb, the PyCB headgroup can complex with water (PyCB-H2O, gray) via hydrogen bonding and exhibit a zwitterionic property at physiological pH. This zwitterionic molecular structure functions to reduce the reactogenicity of the obtained LNP. As also shown in FIG. lb, in a lower pH (~6.8) environment, such as early endosomes, the PyCB-H2O rapidly transforms into a positively charged state (PyCB-HaCF, yellow circle).

[0108] Computational simulation results show that PyCB-fLO' is thermodynamically more stable than PyCB-H2O, which explains this pH-responsiveness of the PyCB headgroup. In other words, the MeDZ lipids participate in rapid protonation, thereby contributing towards a robust endosomal escape. Previous results show that the mRNA transfection and expression of conventional LNPs relies on a narrow endosomal escape window (typically within 2 hours of internalization) prior to lysosomal sequestration and / or exocytosis (E. J. Sayers et al., Mol. Ther. 27, 1950-1962 (2019); Y. Zhao et al., Adv. NanoBiomed Res. 1, 2000078 (2021). By virtue of the excellent protonation ability of the MeDZ lipids, the MeDZ lipids can exhibit stronger interactions with the negatively charged endosomal membranes in acidic endosomes (pH ~6.3) than the conventional ionizable lipids. As depicted in FIG. 1c, at the molecular level, the MeDZ lipids interact with anionic endosomal phospholipids to form cone-shaped ion pairs that are incompatible with the bilayer due to its multi-tail structure. These lipid pairs promote the transition from thebilayer structure to the inverted hexagonal II phase, thereby facilitating membrane disruption and endosomal escape in the narrow time window for robust mRNA expression.

[0109] As a proof of concept, MeDZ lipids were incorporated into the commercially available BNT162b2 LNP from Pfizer-BioNTech (denoted MeDZ LNP) and the resulting mRNA expression efficiency and reactogenicity of the resulting LNP were investigated. The data revealed that the new mRNA vaccine resulting from incorporation of MeDZ LNP into the vaccine exhibited significantly increased mRNA expression (approximately 1.7- fold over BNT162b2 LNP) after entering lymph nodes (LNs) through subcutaneous (SC) injection. Cell type analysis revealed that the mRNA is predominantly expressed in antigen presenting cells APCs, including dendritic cells (DCs) and macrophages. Consequently, the increased levels of tumor antigen expression and robust antigen presentation in LNs after administration of MeDZ LNP vaccine were achieved. The MeDZ LNP vaccine demonstrated excellent efficacy in preventing tumorigenesis and lung metastasis in murine melanoma models. Furthermore, a significant reduction in vector-caused reactogenicity in the skin tissue at the injection site (approximately 41.8-65.7% of BNT162b2 LNP vector) was observed, which can be attributed the zwitterionic surface of the MeDZ LNPs. In addition, it has herein been shown that the MeDZ lipid can be used with known LNP formulations to improve their efficacy. For example, by integrating the MeDZ lipid into a spleen-specific selective organ targeting (SORT) LNP, this approach increased the mRNA expression in the spleens. Collectively, the present work demonstrates a new ionizable lipid that simultaneously increases mRNA expression efficacy and reduces vector reactogenicity.

[0110] Synthesis and characterization of MeDZ-containing LNP

[0111] The MeDZ lipid was first synthesized. The pH -responsiveness of the PyCB headgroup was assessed by comparing the1H NMR spectra of MeDZ lipids exposed to different pH levels, ranging from 7.4 and 5.5 (FIG. Id). After exposing the MeDZ lipids to deionized water with varying pH for 10 minutes, the samples underwent lyophilization to remove free water, and the bound water molecules in these samples were retained. When exposing MeDZ lipids to varying pH values from 7.4 to 5.5, ’ l l NMR spectra of hydrogen in bound water demonstrated gradual broadening and disappearance of the peak at 3.3 ppm (arrow a, high field), accompanied by the appearance of a broad new peak at 3.6-4.0 ppm (arrow b, low field). This result confirmed the formation of positively charged PyCB-HsCF, providing a structural foundation for its pH-responsiveness. Computational simulationswere conducted to explain the pH-responsiveness. The highest occupied molecular orbital (HOMO) analysis showed that the electron-donating site in PyCB is located at the negatively charged oxygen in carboxylate group. The lowest unoccupied molecular orbital (LUMO) analysis showed that the electron-accepting site is located at the positively charged nitrogen in pyridine ring. The molecular electrostatic surface potential was then mapped on the van der Waals surface of PyCB, which determined that the carboxylate group is electron-rich and the pyridine ring is electron-deficient. Due to the large charges of the oxygen and nitrogen atoms in PyCB, it was herein speculated that the interactions between PyCB and H2O or H3O+are typical charge-assisted hydrogen bonds (also known as salt bridge interactions). The hydrogen bonds were determined based on the geometrical characteristic, with a distance of approximately 0.3 nm. As demonstrated in FIG. le, both H2O and H fO+can create a multi-salt bridge structure with oxygen and nitrogen atoms in PyCB, leading to formation of PyCB-H2O and PyCB-H3O+. A shown in FIG. If, PyCB- H3O+exhibits a higher binding energy than PyCB-H2O, which shows that the PyCB headgroup has a greater tendency to complex with H3O+when both H2O and H3O+co-exist. This means that when the pH value of the aqueous solution decreases, the zwitterionic PyCB-H2O transforms to a positively charged PyCB-H3O+.

[0112] To formulate MeDZ LNPs, the present work modified an existing BNT162b2 LNP formulation by replacing the ionizable lipid ALC-0315 with the MeDZ lipids. As shown in FIG. 1g, the molar percentages of the MeDZ lipid were then systematically increased while simultaneously decreasing the molar percentages of ALC-0315 lipid from 0% to 50%. Dynamic light scattering measurements showed that the LNPs containing different percentages of MeDZ lipids and BNT162b2 LNP had similar hydrodynamic diameters ranging from 100 to 120 nm. Transmission electron microscopy (TEM) imaging showed a spherical morphology with a solid core. The mRNA encapsulation efficiencies of these formulations exceeded 80%. By increasing the ratio of MeDZ lipids within the LNP, it was herein found that the isoelectric point (pl) increases from 6.06 at 0% MeDZ lipids to 6.81 at 50% MeDZ lipids. This can be attributed to the excellent pH-responsiveness of the PyCB headgroup in MeDZ lipid (see results in FIG. Ih). Cytotoxicity experiments demonstrated that these LNPs exhibited negligible toxicity.

[0113] MeDZ LNP facilitates mRNA expression in lymph nodes

[0114] The present work first evaluated the mRNA delivery efficacies of LNPs containing different percentages of the MeDZ lipid. Enhanced green fluorescent protein (EGFP) mRNA was employed, and the levels of EGFP expression in DC2.4 cell line were investigated by flow cytometry. BNT162b2 LNP was used for comparison. As shown in FIG. 2a, MeDZ(25%) LNP exhibited the highest EGFP expression efficiency. When the percentages of MeDZ lipids in the LNPs exceeded 37.5%, the mRNA expression levels reduced. This may be because the excess zwitterionic groups on the LNP surface may lead to reduced cellular uptake. This hypothesiswas tested by evaluating the cellular uptake using l,l’-dioctadecyl-3,3,3',3'-tetramethylindotricarbocyanine iodide (Di R) -labelled MeDZ LNPs with different percentages of MeDZ lipids (FIG. 2b). Significantly, MeDZ(25%) LNPs elicited higher mRNA expression (1.7-fold) compared to BNT162b2 LNPs, despite showing similar cellular internalization abilities. This could be attributed to the efficient endosomal escape caused by the excellent pH -responsiveness of the MeDZ lipids.

[0115] The present work further investigated the efficacy of MeDZ LNPs on in vivo mRNA delivery. Firefly luciferase (Luc) mRNA was encapsulated in different LNPs, which allows the expression of luciferase to be visualized in vivo by the IVIS imaging system (PerkinElmer). DiR-labelled Luc mRNA-loaded LNPs were subcutaneously injected into C57BL / 6 mice at the tail base, as schematically shown in FIG. 2c. As shown in Fig. 2d-f, MeDZ(25%) LNP accumulated in LNs and exhibited the highest Luc expression efficiency, which was consistent with the results of cell experiments. Considering that both cellular internalization and endosomal escape play important roles in determining mRNA expression, the endosomal escape effects of MeDZ(25%) LNP and BNT162b2 LNP were assessed by normalizing the cellular uptake of these two formulations in the LNs. The ratio of bioluminescence and fluorescence in LNs is positively correlated with the endosomal escape efficiency. As shown by the results in FIG. 2g, the MeDZ(25%) LNP showed higher endosomal escape efficiency (1.7-fold) than BNT162b2 LNP. This optimized MeDZ(25%) LNP was herein renamed as MeDZ LNP in the following experiments. Next experiments explored the cell types in the LN’ s that were targeted by the MeDZ LNP. Genetically-engineered Ail 4 mice were used because the red florescence protein gene, tdTomato, can be activated and stably expressed after intracellular delivery of Cre recombinase (Cre) mRNA, with the mechanism of Cre-mediated gene recombination inAil4 reporter mice shown in FIG. 2h. The present work used flow cytometry to detect the proportion of the cells expressing tdTomato. Cre mRNA, acting as an activator, was encapsulated into MeDZ LNP. As shown in FIG. 2i, SC injection of Cre mRNA-loaded MeDZ LNPs resulted in 5.06% of DCs and 4.57% of macrophages expressing tdTomato. This result indicates that MeDZ LNPs can effectively deliver the Cre mRNA to APCs. A potential mechanism is that MeDZ LNPs are uptaken up by APCs at the injection site, including DCs and macrophages, and are then trafficked to LNs via afferent lymphatic vessels.

[0116] Mechanism of MeDZ lipid-mediated endosomal membrane destabilization

[0117] The present work then explored the potential mechanism giving rise to the excellent mRNA expression ability of the MeDZ LNPs. The pl value of MeDZ LNP (pl = 6.61) was much higher than that of BNT162b2 LNP (pl =6.06), which means that the former underwent a higher degree of protonation in the acidic endosomes (pH ~6.3) and thus exhibited a stronger interaction with the endosomal membranes. To verify this hypothesis, a Forster resonance energy transfer (FRET) assay was performed to evaluate MeDZ LNP- induced membrane fusion and / or disruption. Two FRET probes, 7V-(7-nitrobenzo-2-oxa- 1,3-diazole-phosphatidylethanolamine (NBD-PE) and rhodamine-phosphatidylethanolamine (Rhod-PE), were formulated into a single endosomal mimicking liposome, which results in attenuated NBD fluorescence because of FRET to Rhod. As schematically shown in FIG. 3a, when the disruption occurs, the resulting enhanced distance between the two probes increases the NBD signal. The FRET efficiency of the mixture of MeDZ LNP and endosomal mimicking liposome under different pH conditions was measured. BNT162b2 LNP was used as the control. As shown in FIG. 3b and FIG. 3c, MeDZ LNP exhibited higher membrane disruption than BNT162b2 LNP at pH 6.0, demonstrating the crucial role of MeDZ lipid with excellent pH-responsiveness. As shown in FIG. 3d, further analysis of the phase transition using established31P nuclear magnetic resonance (NMR) spectroscopy techniques revealed that MeDZ lipid (right) and ALC-0315 lipid (left) exhibited similar peak shapes and shifts. This result indicated that the mechanism of membrane disruption caused by MeDZ lipids also works by promoting the transition from the bilayer structure to the inverted hexagonal II phase.

[0118] To confirm the endosomal escape ability of the MeDZ LNP, the present work observed the escape on DC2.4 cell line using confocal laser scanning microscopy (CLSM).The LNPs were pre -labelled by NBD-PE and incubated with DC2.4 cells for 1.5 hours. As demonstrated in FIG. 3e and FIG. 3f, compared to BNT162b2 LNP, a greater amount of green fluorescence signal was observed in the cytoplasm when the cells were incubated with MeDZ LNP, thus indicating efficient endosomal escape. To further confirm the above results, a co-localization analysis was conducted using the CLSM images in FIG. 3e. Pearson’s co-localization coefficient (PCC) between the fluorescence signals of different NBD-labelled LNPs and LysoTracker™ Red were calculated. A PCC value close to 1 indicates a high degree of co-localization between these two signals. As shown in FIG. 3g and FIG. 3h, the PCC value in the MeDZ LNP group was lower (PCC = 0.78) than BNT162b2 LNP group (PCC = 0.90), thus confirming the enhanced endosomal escape efficacy.

[0119] MeDZ LNP enhances antigen presentation and exhibits reduced reactogenicity

[0120] Efficient endosomal escape of mRNA leads to enhanced expression of target protein. The antigen expression in murine bone marrow-derived dendritic cells (BMDCs) was then tested by Western blotting. mRNA expressing the model antigen ovalbumin (OVA) was used, and the Lipofectamine 2k-based vector was used as the positive control. As shown by the data in FIG. 4a and FIG. 4b, semi-quantitative analysis showed that intracellular delivery of OVA mRNA by MeDZ LNPs exhibited significantly higher OVA expression than BNT162b2 LNP. The present work then assessed the presentation of OVA fragments in BMDCs by flow cytometry. As shown in FIG. 4c and FIG. 4d, BMDCs treated with OVA mRNA-loaded MeDZ LNPs for 24 hours showed the highest OVA fragment presentation on the major histocompatibility complex class I (MHC-I), which was demonstrated by staining H2kb-SIINFEKL complex on the BMDC surface. These results showed that MeDZ LNPs could serve as a vaccine vector to improve tumor antigen presentation in DCs, which is essential for cytotoxic T cell priming.

[0121] Before evaluating the potential of MeDZ LNPs as mRNA vaccines, the reactogenicity of this LNP vector was analyzed. MeDZ LNPs were subcutaneously injected to C57BL / 6 mice at the tail base. As schematically shown in FIG. 4e, after 3 days, the mice were sacrificed and the skin samples (about 1 cm2) at the injection site were harvested. As shown in FIG. 4f, FIG. 4g, and FIG. 4h, SC administration of BNT162b2 LNP resulted in high levels of pro-inflammatory cytokines within the skin tissues, including GM-CSF, IL- ip, and IL-6, whereas the samples in the MeDZ LNP group exhibited significantly reducedpro-inflammatory signs (approximately 41.8-65.7% of BNT162b2 LNP). Moreover, as indicated in FIG. 4k, MeDZ LNP alleviated the inflammation-induced redness and swelling at the injection site, compared to BNT162b2 LNP. It is known that neutrophils are the first cells to infiltrate inflamed tissues. In this respect, as shown by the flow cytometry results in FIG. 4i and FIG. 4j, the proportion of neutrophils in CD45+cells in the MeDZ LNP group was lower than that of the BNT162b2 LNP. As shown in FIG. 41, the hematoxylin and eosin (H&E) images of skin samples from the treated mice showed similar trends, with fewer pro-inflammatory signs observed in the MeDZ LNP group. These results showed that MeDZ LNP has reduced vector reactogenicity.

[0122] MeDZ LNP vaccine prevents tumorigenesis, lung metastasis, and recurrence

[0123] The present work then evaluated the tumor prophylactic effect of MeDZ LNP serving as a mRNA vaccine vector. OVA mRNA was encapsulated into the MeDZ LNP, then vaccinations were performed according to the timeline in FIG. 5a, with a prime dose on day 0 and booster doses on days 5 and 10. The levels of OVA-specific immunoglobulin M (IgM) and immunoglobulin G (IgG) in the blood on day 14 were measured by enzyme- linked immunosorbent assay (ELISA). As shown by the data presented in FIG. 5b and FIG. 5c, comparable levels of IgM and significantly higher levels of IgG were detected in the blood samples from the MeDZ LNP group compared to these in the BNT162b2 LNP group. These results confirmed the activation of tumor antigen-specific immune responses induced by MeDZ LNP vaccines. To evaluate the prophylactic effect, these mice were inoculated with B16F10 tumor cells that express OVA antigens (B16F10-OVA) on day 15. As demonstrated by the data shown in FIG. 5d and FIG. 5e, the tumors in the unvaccinated mice grew rapidly until reaching the humane endpoint. The tumorigenesis was completely inhibited in four of the six mice in the BNT162b2 LNP group. Significantly, neither tumor development nor growth were found in mice after vaccination with MeDZ LNP vaccines, suggesting an excellent prophylactic effect.

[0124] Next, the efficacy of MeDZ LNP vaccines on amplifying antigen presentation and triggering antigen-specific cytotoxic T cell activation were evaluated by flow cytometry. As shown by the data in FIG. 5f, administration of MeDZ LNP vaccine exhibited significantly higher presentation of OVA fragments (H2kb-SIINFEKLhlgh, 4.97%) on the DC surface within LNs, compared to other treatment groups (phosphate-buffered saline (PBS), 0. 12%; BNT162b2 LNP, 2.46%). The improved OVA presentation should becorrelated with the excellent endosomal escape efficacy of MeDZ LNP in DCs. Next, the activation of cytotoxic CD8+T cells in the blood was assessed because they are the major elfectors for the suppression of tumorigenesis. It was herein observed that MeDZ LNP vaccine -treatment induced increased IFN-y expression (a cytotoxic cytokine that initiates tumor cell apoptosis, IFN-y+, 2.00%) on the surface of CD8+T cells, which was 1.4-fold higher than that of BNT162b2 LNP vaccine (1.39%). To further confirm the activation of tumor antigen-specific CD8+T cells, the population of OVA-specific CD8+T cells in the blood was investigated using a tetramer assay. As shown by the data in FIG. 5g, negligible OVA-specific CD8+T cells were observed in PBS control (OVA Tetramer -, 0.30%), while treatment with MeDZ LNP vaccines showed a significantly increased level of OVA-specific T cells (2.80%). Efficient cytotoxic T cell activation also facilitated the establishment of strong immune memory, as confirmed by evaluating the effector memory T (TF.M) cell and central memory T (TCM) cell levels within spleens. As shown by the data in FIG. 5h, MeDZ LNP vaccine-treatment resulted in significant increase in the proportions of TEM cells (CD62LlowCD44hlgh, 12.5%) and TCM cells (CD62LhlghCD44lllgh, 23.2%), compared to other treatment groups.

[0125] The present work then explored the efficacy of MeDZ LNP vaccines on preventing the tumor metastasis to lungs. As schematically shown in FIG. 5i, the C57BL / 6 mice were intravenously inoculated with B16F10-OVA tumor cells, followed by vaccination with different formulations on days 1, 6, and 11. At day 20, these mice were euthanized and the lung tissues were collected. As demonstrated by the H&E staining images of the lung tissues in FIG. 5j, numerous black tumor nodules in the lung tissue were observed in the PBS control group. The BNT162b2 LNP vaccine treatment significantly reduced the number of tumor nodules in lung tissue. In the MeDZ LNP group and no nodules were observed in the lung. Histological analysis of H&E-stained lung sections confirmed this result (tumor nodules were marked with yellow arrows (Fig. 5j).

[0126] T cell-dependent immune responses shaped by DCs are antigen-specific. The present work then explored the efficacy of MeDZ LNP vaccines in intervention of cognate tumor recurrence. As schematically shown in FIG. 5k, the mice were first intravenously injected with B16F10-OVA tumor cells and vaccinated. To mimic tumor recurrence, these mice were inoculated with B16F10-OVA (left flank) and non-cognate EO771 murine mammary (right flank) tumor cells via SC injection on day 30. Normal mice wereinoculated with these two tumor cells, serving as a control group. FIG. 51 shows photographs of B16F10-OVA and EO771 tumors. FIG. 5m shows individual tumor growth curves (PBS-treated mice, gray; MeDZ LNP-treated mice, red). TFS = tumor-free survival. The recurrence of B16F10-OVA tumor was suppressed completely in three out of five mice after treatment with the MeDZ LNP vaccines. However, only slight suppression of EO771 tumor growth was found. The above-mentioned selective inhibition of B16F10-OVA tumor growth can be attributed to the infiltration and activation of OVA-specific T cells within tumors. This was confirmed by detecting the intra-tumoral CD8+T cell infiltration (MeDZ LNP, 40.2%; PBS, 19.7%) and the population of CD8+T cells that bear T cell receptors binding to H2Kb OVA tetramer-SIINFEKL (MeDZ LNP, 5.52%; PBS, 0.28%). These results indicate that MeDZ LNPs can serve as a promising vector for mRNA vaccines to induce robust T cell-dependent adaptive immunity and immune memory.

[0127] MeDZ LNP vaccine inhibits tumor growth

[0128] The therapeutic effect of MeDZ LNP vaccine was then assessed. As schematically shown in FIG. 6a, B16F10-OVA tumor-bearing mice were injected subcutaneously with different formulations. As shown by the data in FIG. 6b and FIG. 6c, treatments with MeDZ LNP vaccines effectively suppressed the tumor growth and improved the survival rate of mice to 40% from 0% over a 40-day observation period. Tumors in different treatment groups were then collected for histological analysis, with the images shown in FIG. 6d. Karyopyknosis and a drastic decrease in the number of nuclei were found in tumor tissues collected from the MeDZ LNP group. The anti-tumor immune response caused by the MeDZ LNP vaccine was then evaluated. As shown by the data in FIG. 6e, significantly enhanced OVA fragment presentation (H2Kb-SIINFEKLhlsh, 5.44%) on the DC surface within LNs was observed in the MeDZ LNP group. In contrast, a relatively moderate OVA fragment presentation was observed in BNT162b2 LNP group (2.32%). A tetramer assay was then performed to evaluate the APC-mediated CD8+T cell activation within LNs. As shown by the data in FIG. 6f, the MeDZ LNP vaccine induced a higher level of OVA-specific CD8+T cells (OVA Tetramer+, 1.09%) within LNs than that of BNT162b2 LNP (0.45%). Next, by analysis of the activation of intra-tumoral CD8+T cells, it was herein observed that the MeDZ LNP treatment induced the highest CD8+T cell infiltration (CD8+, 31.0%) (data shown in FIG. 6g). The level of OVA-specific CD8+T cells within tumor tissues was also assessed. As shown by the data in FIG. 6h, MeDZ LNPvaccine-treatment triggered significantly higher levels of OVA-specific CD8+T cells infiltration (4.76%) within tumor tissues compared to BNT162b2 LNP (OVA Tetramer-, 1.22%), which confirms the activation of OVA-specific T cell-mediated anti-tumor immune response.

[0129] Next experiments investigated the ability of MeDZ LNP vaccines to improve the therapeutic outcomes of clinically used immune checkpoint blockade (ICB) therapy. In this combination therapy, B16F10-OVA tumor-bearing mice were treated with MeDZ LNP vaccines on days 0, 5, and 10. As schematically shown in FIG. 6i, the anti-PD-1 (aPD-1) antibodies were administered by intraperitoneal (IP) injection on days 1, 6, and 11. The aPD-1 monotherapy was performed for comparison.

[0130] As shown by the data in FIG. 6j, FIG. 6k, and FIG. 61, aPD-1 monotherapy slowed the tumor growth, and one of the six tumors was completely suppressed over a 60-day observation period. However, it was herein observed that the combination of aPD- 1 and MeDZ LNP completely inhibited the tumor growth in three of the six mice. As a result, as shown by the data in FIG. 6m, this combination therapy effectively prolonged the survival of tumor-bearing mice (survival rate, 50.0%). These results showed that MeDZ LNP vaccines can synergize with ICB therapy to improve the therapeutic outcomes.

[0131] Recently, intravenously injected spleen-targeting LNPs have shown great potential for enhancing cancer immunotherapy. However, their ability to promote mRNA expression remains limited. Thus, further experiments explored the use of MeDZ lipids to improve the mRNA expression of these LNP formulations. As a proof of concept, a spleen-specific SORT LNP was chosen, with the results shown in FIG. 6p. As shown by the data in FIG. 6n and FIG. 6o, it was herein found that the integration of MeDZ lipids into spleen-specific SORT LNP (denoted SORT MeDZ LNP) did not affect its accumulation in the spleen, but improved the mRNA expression efficiency. This improvement can be attributed to the excellent ability of MeDZ lipids to facilitate endosomal escape.

[0132] Methods

[0133] The ability of MeDZ LNP to facilitate EGFP mRNA expression in DC2.4 cells. Flow cytometry was used to study the cellular uptake of DiR-labelled different LNPs and the ability of these LNPs to facilitate EGFP mRNA expression in DC2.4 cell line. For flow cytometry assays. DC 2.4 cells (5 x 104cells per well) were seeded in a 12-well plate andincubated overnight for cell attachment. DiR-labelled EGFP mRNA-loaded different LNPs (EGFP mRNA, 1 pg mL'1; LNP, 15 gig mL ) were incubated with DC 2.4 cells for 3 hours. To study the cellular uptake, the cells were then washed with PBS three times, followed by digesting using 0.05% trypsin for flow cytometry analysis. To study the EGFP mRNA expression, the cells were washed with PBS three times after another 24-hour incubation, followed by digesting using 0.05% trypsin for flow cytometry analysis.

[0134] Delivery of Luc mRNA to LNs using MeDZ LNP in C57BL / 6 mice. All animal procedures were performed with ethical compliance and approval by the Institutional Animal Care and Use Committee (IACUC) at the Cornell University (Animal Protocol# 2023-0101). C57BL / 6 mice (Jackson Lab) and DiR-labelled Luc-loaded formulations were employed. 75 pg of DiR-labelled LNPs (Luc mRNA, 5 pg per mouse) were subcutaneously injected into the mice at the tail base. After 6 hours, 100 pL of D-Luciferin potassium salt solution, at a concentration of 15 mg mL1, was intraperitoneally injected into the mice. After 10 minutes, the biodistribution of DiR-labelled LNPs and Luc mRNA expression were visualized using an IVIS imaging system.

[0135] Delivery of Cre mRNA to LNs using MeDZ LNP in AH4 reporter mice. Ail4 mice were subcutaneously injected with Cre mRNA-MeDZ LNPs containing 5 pg mRNA at the tail base. After 48-hour injection, the mice were euthanized and inguinal LNs were collected. The single-cell suspensions were prepared by grinding and filtrating through a 70 pm strainer. Then, 1 x 106cells were stained with 100 pL of flow cytometry staining buffer (eBioscience) containing different fluorophore -conjugated antibody of interest at the recommended concentration at 4 °C for 1 hour. The cells were kept at 4 °C after washing twice with staining buffer for flow cytometry analysis.

[0136] Evaluation of lipid fusion and / or disruption by FRET assay. Endosomal mimicking liposomes were prepared by mixing DOPS, DOPC, DOPE, NBD-PE, and Rhod-PE at a molar ratio of 25: 25: 48: 1 : 1 in chloroform, followed by rotary evaporation and vacuum dry to obtain a thin lipid film. The dried film was hydrated in PBS and sonicated for 10 min, and the concentration of total lipid was fixed at 1 mM. As a result, two FRET probes was formulated into the same endosomal mimicking liposomes, which resulted in the attenuated NBD flurescence due to the FRET to Rhod. When the fusion occurred, NBD flurescence increased due to the enhanced distance between these to probes. The FRET efficiency of the mixture of MeDZ LNP and endosomal mimicking liposome underdifferent pH conditions was measured. Fluorescence intensities were recorded at Ex / Em = 465 / 520 nm on a microplate reader.

[0137] Phase transition analysis using31P NMR spectroscopy techniques. Endosomal mimicking liposomes were prepared by mixing DOPS, DOPC, and DOPE at a molar ratio of 25: 25: 50 in chloroform, followed by rotary evaporation and vacuum dry to obtain a thin lipid film. The films were hydrated in D2O (10 mM HEPES, pH 5.5), and the solution was vortexed vigorously to obtain a white dispersion. Then, the endosomal mimicking liposomes and MeDZ lipid suspension were mixed thoroughly by sonication (MeDZ lipid: DOPS: DOPC: DOPE molar ratio, 20: 20: 20: 40), and were characterized by31P NMR. The mixture of endosomal mimic and ALC-0315 lipid was used as control.

[0138] The ability of MeD LNP to facilitate endosomal escape. Confocal laser scanning microscope (CLSM) was used to study the ability of MeDZ LNP in facilitating endosomal escape in DC2.4 cells. NBD-labelled different LNPs were employed. For CLSM observation, DC 2.4 cells were seeded in confocal dishes and allowed to grow until approximately 70-80% confluence. NBD-labelled LNPs (15 pg mL1) were incubated with DC 2.4 cells for 1.5 hours. The cells were then washed with PBS three times, and the endosomes were stained with LysoTracker™ Red. Then, the cells were fixed using 4% paraformaldehyde and the nucleus was stained with 4',6-diamidino-2-phenylin-dole (DAPI, blue).

[0139] The ability of MeDZ LNP to enhance OVA mRNA presentation. Bone marrow- derived dendritic cells (BMDCs) were generated from the bone marrow of the C57BL / 6 mice. Western blotting was performed to evaluate the cellular OVA levels, and flow cytometry analysis was performed to study the antigen presentation in BMDCs.

[0140] For western blotting, BMDCs (1 x 105cells per well) were seeded in a 6-well plate and incubated overnight for cell attachment. Different OVA mRNA-loaded LNPs (1 pg / mL-1) were incubated with BMDCs for 6 hours, and then replaced by fresh culture medium for additional 24-hour incubation. The cells were then washed with PBS three times and lysed in 0.5 mL of lysis buffer (PBS containing 1% Triton X-100 and 1 mM phenylmethylsulfonyl fluoride (PMSF)) at 4 °C for 30 min. Equal amounts protein loadings were separated by SDS-PAGE and electrophoretically transferred to a PVDF membrane. Non-specific binding sites were blocked with 5% nonfat milk in PBS and 0.05% Tween 20 (PBST) for 2 hours at room temperature. Membranes were incubated in the primaryantibodies at appropriate concentrations for 24 hours at 4 °C overnight, followed by three times rinse with PBST buffer for 15 min. The membranes were then incubated in the secondary antibodies (1 : 10000) for 1 hour at room temperature. The membranes were detected and analyzed using a computerized chemiluminescent imaging system. Antibody list: anti-chicken ovalbumin antibody (polyclonal) and anti-mouse GAPDH antibody (polyclonal).

[0141] For flow cytometry assays, different OVA mRNA-loaded LNPs (1 pg / mL1) were incubated with BMDCs for 24 hours. BMDCs were then collected and stained with the flow cytometry staining buffer containing different fluorophore-conjugated antibody of interest at the recommended concentration at 4 °C for 1 hour. Antibody list: PE-Cy7-anti- mouse CD11c antibody and APC-anti-mouse H2kb-SIINFEKL antibody.

[0142] Reactogenicity ofMeDZ LNP tested in C57BL / 6 mice. Different LNPs were subcutaneously injected to C57BL / 6 mice at the tail base. After 3 days, the mice were sacrificed and the skin samples (about 1 cm2) at the injection site were harvested to prepare single cell suspensions for flow cytometry analysis. Antibody list: FITC-anti-mouse CD45 antibody, PE-Cy7-anti -mouse Ly-6G antibody, and APC-anti-mouse CDl lb antibody.

[0143] Prophylactic effect of Mel) / . LNP vaccine in the murine melanoma models. The female C57BL / 6 mice at 6~8-week old and B16F10 tumor cells expressing ovalbumin (B16F10-OVA) were used in this study. The mice were randomly divided into three groups (n = 6). OVA mRNA was encapsulated into MeDZ LNP (OVA mRNA, 5 pg per mouse), and the vaccinations (at the tail base) were performed with a prime dose on day 0 and booster doses on days 5 and 10. The mice treated with PBS and BNT162b2 vaccine were used for comparison. The levels of OVA-specific IgM and IgG in the blood on day 14 were measured by ELISA. Then, the mice were challenged by SC injection of B 16F10-OVA tumor cells (l x 106per mouse) into the left flank. The growth of tumors was then monitored. The tumor volume (V) was calculated according to the formula: V = L x W2 / 2, where L and W were the longest and shortest diameter (mm) of the tumor, respectively. The mice were euthanized when exhibiting signs of impaired health (e.g., exhibited >15% loss of body weight) or when the volume of the tumor exceeded 1.5 cm3.

[0144] For flow cytometry assays, freshly harvested LNs were made into cell suspensions. The single-cell suspensions were prepared by grinding and filtrating through a 70 pm strainer. Then, 1 x 106cells were stained with 100 pL of flow cytometry staining buffercontaining different fluorophore-conjugated antibody of interest at the recommended concentration at 4 °C for 1 hour. The cells were kept at 4 °C after washing twice with staining buffer for flow cytometry analysis. Antibody list: eFluor 450-anti-mouse CD19 antibody, FITC-anti-mouse CD3 antibody, PE-Cy7 -anti -mouse CDl lc antibody, and APC- anti-mouse H2kb-SIINFEKL antibody.

[0145] Similar procedures were performed to analyze the activation of immune cells in the blood and spleens using fluorophore-conjugated antibodies. Red blood cells in these samples were removed by using red blood cell lysis buffers. Antibody list: FITC-anti- mouse CD3 antibody, PE-Cy7-anti-mouse CD8a antibody, APC-H2Kb OVA tetramer- SIINFEKL, PE-anti-mouse IFN-y antibody, Super Bright 600-anti-mouse CD62L antibody, and AF700-anti-mouse CD44 antibody.

[0146] Prevention of tumor metastasis to lungs using MeDZ LNP vaccine. The C57BL / 6 mice were randomly divided into three groups (n = 3). The mice were then intravenously challenged with B16F10-OVA tumor cells (1 x 106per mouse) on day 0, followed by vaccination with different formulations (OVA mRNA, 5 pg per mouse) on days 1, 6, and 11. At day 20, these mice were sacrificed and the lung tissues were collected for histological analysis.

[0147] Effect of MeDZ LNP vaccine on intervention of cognate tumor recurrence. The C57BL / 6 mice were randomly divided into two groups (n - 5). The mice were then intravenously injected with B16F10-OVA tumor cells (1 x 106per mouse) and vaccinated by MeDZ LNP vaccine (OVA mRNA, 5 pg per mouse). To mimic tumor recurrence, these mice were rechallenged with B16F10-OVA and non-cognate E0771 murine mammary tumor cells via SC injection on day 30. B16F10-OVA tumor cells (1 x 106per mouse) were subcutaneously injected at the left flank of mice, and E0771 tumor cells (1 x 106per mouse) were inoculated in the opposite flank of the B16F10-OVA tumors. Normal mice were inoculated with these two tumor cells, serving as a control group. The growth of tumors was then monitored.

[0148] For flow cytometry assays, freshly harvested tumor tissues were digested with collagenase IV and made into cell suspensions according to the manufacturer’s instructions. The single-cell suspensions were prepared by grinding and filtrating through a 70 pm strainer. Then, I x I06cells were stained with 100 pL of flow cytometry staining buffer containing different fluorophore-conjugated antibody of interest at the recommendedconcentration at 4 °C for 1 hour. The cells were kept at 4 °C after washing twice with staining buffer for flow cytometry analysis. Antibody list: FITC-anti-mouse CD45 antibody, PerCP-Cy5.5-anti-mouse CD3 antibody, PE-Cy7 -anti -mouse CD8a antibody, and APC-H2Kb OVA tetramer-SIINFEKL.

[0149] Therapeutic effect of MeD LNP vaccine in the established murine melanoma model. Tumor-bearing mice were established by SC injection of B 16F10-OVA tumor cells (1 x 106per mouse) into the left flank of C57BL / 6 mice. The mice were then randomly divided into three groups (n = 5) after six days of tumor inoculation. The tumor growth was monitored after SC administration of the mice by different formulations (OVA mRNA, 5 pg per mouse) at the tail base on days 0, 5, and 10. The growth of tumors was then monitored.

[0150] For flow cytometry assays, freshly harvested tumor tissues were digested with collagenase IV and made into single -cell suspensions. Then, 1 x 106cells were stained with 100 pL of flow cytometry staining buffer containing different fluorophore-conjugated antibody of interest at the recommended concentration at 4 °C for 1 hour. The cells were kept at 4 °C after washing twice with staining buffer for flow cytometry analysis. Antibody list: FITC-anti-mouse CD45 antibody, PerCP-Cy5.5-anti-mouse CD3 antibody, PE-Cy7- anti-mouse CD8a antibody, and APC-H2Kb OVA tetramer-SIINFEKL.

[0151] Similar procedures were performed to analyze the activation of immune cells in the LNs using fluorophore-conjugated antibodies. Antibody list: eFluor 450-anti-mouse CD 19 antibody, FITC-anti-mouse CD3 antibody, PE-Cy7-anti-mouse CD 11c antibody, APC-anti- mouse H2kb-SIINFEKL antibody, PE-Cy7-anti-mouse CD8a antibody, and APC-H2Kb OVA tetramer-SIINFEKL.

[0152] The ability of MeDZ LNP vaccine to improve 1CB therapy. The B 16F10-OVA tumor-bearing mice were randomly divided into three groups (n = 6). The mice were vaccinated by different formulations (OVA mRNA, 5 pg per mouse) on days 0, 5, and 10. The aPD-1 antibodies (200 pg per dose) were administrated by IP injection on days 1, 6, and 11.

[0153] The ability of spleen-specific SORT MeDZ LNP to enhance mRNA expression. Spleen-specific mRNA expression induced by spleen-specific SORT MeDZ LNPs was evaluated in C57BL / 6 mice using DiR-labelled Luc -loaded formulation. 75 pg of DiR- labelled LNPs (Luc mRNA, 5 pg per mouse) were intravenously injected into the mice.After 6 hours, 100 pL of D-Luciferin was intraperitoneally injected into the mice. After 10 min, the biodistribution of DiR-labelled LNPs and the mRNA expression were visualized using an IVIS imaging system.

[0154] Synthesis of MeDZ lipid

[0155] Synthesis of Compound 3

[0156] Compound 1 (4.1 mmol, 1.0 eq), 10 mL of dichloromethane, Compound 2 (4 mmol, 1.0 eq), EDC (6 mmol, 1.5 eq), DMAP (0.4 mmol, 0.1 eq), EhN (8 mmol, 2.0 eq) were added to the reaction flask in sequence and reacted at room temperature. After 4-hour reaction, the mixture was diluted with dichloromethane (30 mL) and washed with saturated sodium bicarbonate ( NaHCOq. The combined organic layer was separated, washed with saturated sodium chloride (NaCl), and dried with anhydrous magnesium sulfate (MgSCU). The organic layer was then filtered and the solvent was removed under reduced pressure. The residue was purified by silica gel column chromatography to yield Compound 3 (1.38 g, 45%).

[0157] Synthesis of Compound 4

[0158] Compound 3 (1.8 mmol, 1.2 eq), 3 mL of anhydrous ethanol, 4-amino-l -butanol (1.5 mmol, 1.0 eq), and Et3N (3 mmol, 2.0 eq) were added to the reaction flask and heated to 80°C for overnight reaction. After the reaction was completed, the solvent was removed under reduced pressure and the residue was dissolved in ethyl acetate and water. The organic layer was separated and washed with water and saturated NaCl, and dried with anhydrous MgSCh. The organic layer was then filtered and the solvent was removed under reduced pressure. The residue was purified by silica gel column chromatography to yield Compound 4 (1.27 mmol, 85%). ESI-MS: m / z [M + H+] C48H95NO5 calculated: 765.72; found: 766.00.

[0159] Synthesis of Compound 5

[0160] Compound 4 (1.27 mmol, 1.0 eq), 10 mL of di chloromethane, and Et.rN (2.54 mml, 2.0 eq) were added to the reaction flask. Then, isonicotinoyl chloride (1.91 mmol, 1.5 eq) was added dropwise at 0°C. The reaction was stirred overnight at room temperature. After the reaction was completed, the solvent was removed under reduced pressure and the residue was dissolved in ethyl acetate and water. The organic layer was separated and washed with water and saturated NaCl, and dried with anhydrous MgSCk The organic layer was then filtered and the solvent was removed under reduced pressure. The residue was purified by silica gel column chromatography to yield Compound 5 (0.82 mmol, 65%). ESI-MS: m z [M + H+] C54H98N2O6 calculated: 870.74; found: 870.90. m / z [M / 2 + H+]: found: 436.10. [M / 2 + K+]: found: 476.90.

[0161] Synthesis of MeDZ lipid

[0162] Compound 5 (0.5 mmol, 1.0 eq), 2 mL of anhydrous ethanol, and bromoacetic acid (0.75 mmol, 1.5 eq) were added to the reaction flask. The reaction was stirred overnight at 80°C. After the reaction was completed, the solvent was removed under reduced pressure and the residue was dissolved in di chloromethane. The organic layer was separated and washed with water and saturated NaCl, and dried with anhydrous MgSC The organic layer was then filtered and the solvent was removed under reduced pressure. The residue was purified by silica gel column chromatography to yield the MeDZ lipid. (350 mg, 66%).

[0163] Preparation of different LNPs

[0164] Preparation of the LNPs containing different percentages of MeDZ lipids

[0165] These LNPs were formulated with ALC-0315 lipid, MeDZ lipid, Choi, DSPC, and DMG-PEG2k using an ethanol dilution method. The formulations with different percentages of MeDZ lipids (from 0% to 50% molar ratio) were prepared below.BNT162b2 LNP: the molar ratio of ALC-0315 lipid: Choi: DSPC: DMG-PEG2k was 50:38.5: 10:1.5.MeDZ(12.5%) LNP: the molar ratio of ALC-0315 lipid: MeDZ lipid: Choi: DSPC: DMG- PEG2kwas 37.5: 12.5: 38.5: 10: 1.5.MeDZ(25%) LNP (the optimized MeDZ LNP): the molar ratio of ALC-0315 lipid: MeDZ lipid: Choi: DSPC: DMG-PEG2kwas 25: 25: 38.5: 10: 1.5.MeDZ(37.5%) LNP: the molar ratio of ALC-0315 lipid: MeDZ lipid: Choi: DSPC: DMG- PEG2kwas 12.5: 37.5: 38.5: 10: 1.5.MeDZ(50%) LNP: the molar ratio of ALC-0315 lipid: MeDZ lipid: Choi: DSPC: DMG- PEG2kwas 0: 50: 38.5: 10: 1.5.

[0166] Preparation of spleen-specific SORT LNP and spleen-specific SORT MeDZ LNP

[0167] These two LNPs were formulated with Dlin-MC3-DMA lipid, 18PA, MeDZ lipid, Choi, DSPC, and DMG-PEG2k using an ethanol dilution method.SORT LNP: the molar ratio of Dlin-MC3-DMA lipid: 18PA: Choi: DSPC: DMG-PEG2kwas 35: 30: 27: 7: 1.SORT MeDZ LNP: the molar ratio of Dlin-MC3-DMA lipid: 18PA: MeDZ lipid: Choi: DSPC: DMG-PEG2kwas 20: 25 20: 27: 7: 1.

[0168] Preparation of DiR-labelled LNP formulations

[0169] For the preparation of DiR-labelled LNPs, the mixture of total lipids and DiR were dissolved at a weight ratio of 100 / 1 in ethanol.

[0170] Characterization of MeDZ LNP

[0171] To prepare the TEM sample, 10 pL of sample solution was dropped onto a carbon- coated copper grid for 10 min and blotted with filter paper to remove excess liquid. Then, the sample was negatively stained with phosphotungstic acid (EPTA) (10 pL) for 2 min,blotted again and air-dried before analysis on TEM. The hydrodynamic size of different formulations was evaluated in phosphate-buffered saline (PBS) at 37 °C.Example 2Three-component LNPs formulated by MeDZ lipid, helper lipid, and ionizable lipid for spleen-specific mRNA expression

[0172] The spleen is emerging as a key vaccination target. However, existing lipid nanoparticle (LNPs) primarily accumulate in the liver, limiting their efficacy in cancer vaccine therapy. The cholesterol in current LNP formulations promote their uptake by hepatocytes while the PEGylated lipids induce PEG immunogenicity, further reducing the efficacy in the setting of repeated administrations. The present work has developed a three- component (ThrCo) LNP by replacing cholesterol and PEGylated lipids in Pfizer-BioNTech LNPs with zwitterionic pyridine carboxybetaine (PyCB) ionizable lipids (ILs), thereby achieving -70% lower liver accumulation and a 4.5-fold increase of spleen-specific mRNA translation. PyCB ILs enhance LNP hydrophilicity, thereby stabilizing the outer membrane to compensate for cholesterol removal. PyCB groups also exhibit strong protonation at endosomal pHs, facilitating mRNA translation. Additionally, the zwitterionic surface of ThrCo LNP reduces protein adsorption, thereby preventing the accelerated blood clearance effect caused by PEGylated lipids following repeated administrations. Thus, ThrCo LNP- based vaccines efficiently deliver mRNA to splenic antigen presenting cells, thereby boosting immune responses and improving therapeutic outcomes.

[0173] The next experiments demonstrate that cholesterol and PEGylated lipids are not essential for LNP functionality, thereby permitting mitigation of the adverse events associated with these two components through simplifying LNP composition. Based on the BNT162b2 from Pfizer-BioNTech, a three-component (denoted as ThrCo) LNP formulation was produced. The three-component LNP contains a zwitterionic pyridine carboxybetaine (PyCB) IL, clinically available ALC-0315 IL, and l,2-distearoyl-sn-glycero-3- phosphocholine (DSPC). Replacement of cholesterol and PEGylated lipids with PyCB ILs redirects a significant portion of LNPs, which typically accumulate in the liver, to the spleen, thereby specifically enhancing mRNA translation in the spleen. Crucially, PyCB ILs allow ThrCo LNPs to maintain relatively high mRNA translation efficiency following repeated administrations, thereby achieving robust antigen expression and eliciting tumorspecific adaptive immune activation. To better align with clinical standards, PyCB IL isderived from ALC-0315 IL (FIG. 7A). Through covalent chemistry, a zwitterionic PyCB structure is attached to the hydroxyl group in ALC-0315 IL to increase the headgroup’s hydrophilicity (FIG. 7B). The increased hydrophilicity enhances the stability of the LNP outer membrane in aqueous environments, thereby compensating for the instability caused by the absence of cholesterol in LNPs and preventing LNP disintegration. As a result, such a reformulated ThrCo LNP attenuates the cholesterol-dependent hepatic tropism and increases the spleen accumulation. Furthermore, the zwitterionic surface offers another advantage to ThrCo LNPs by reducing non-specific protein adsorption in the bloodstream, mitigating the ABC effect and improving the vaccine bioavailability in the setting of multiple injections. Additionally, the PyCB headgroup exhibits zwitterionic properties at physiological pH and undergoes strong protonation in acidic endosomes, facilitating efficient endosomal escape for mRNA cytoplasmic delivery (FIG. 7C).

[0174] The present work revealed that, compared to BNT162b2 LNPs, ThrCo LNPs exhibited a 4.5-fold increase in mRNA translation in the spleen (FIGS. 7D and 7F). Compared to previously reported anionic spleen-targeting C-18 phosphatidic acid (18PA) LNPs, ThrCo LNPs boosted mRNA translation in the spleen by more than 40-fold (FIG. 7E). Cell type analysis showed that mRNA was predominantly translated in splenic APCs, including dendritic cells (DCs) and macrophages. Administration of ThrCo LNP-based vaccines led to a significant increase in antigen-specific cytotoxic T cell activation and demonstrated excellent efficacy in suppressing tumor progression, inhibiting lung metastasis, and reducing recurrence in mouse melanoma models. Collectively, the present work demonstrates a new chemical design paradigm for LNPs by replacing both cholesterol and PEGylated lipids with zwitterionic PyCB ILs. The PyCB IL provides multiple critical advantages. First, PyCB IL enhances the hydrophilicity of the LNP surface, thereby stabilizing the nano-structure and maintaining high mRNA encapsulation efficiency in the absence of cholesterol. Second, the removal of cholesterol in LNPs minimizes lipoprotein binding, effectively redirecting their biodistribution from the liver to the spleen for spleenspecific mRNA translation. Third, PyCB IL serves as a PEG-mimetic component that minimizes LNP aggregation and prevents anti-PEG antibody-mediated ABC, thereby enabling repeated administrations without loss of efficacy. Furthermore, the present work elucidates a structure-function relationship that provides a molecular principle for the rational design of new ILs for mRNA-LNPs. The present work also outlines chemicalstrategies and methodologies to facilitate mRNA delivery to extrahepatic organs, offering broad translational potential in vaccine and immunotherapy applications.

[0175] PyCB ILs enable development of ThrCo LNPs for spleen-specific mRNA transfection

[0176] The PyCB IL was synthesized and evaluated its spleen-targeting capability by substituting cholesterol and PEGylated lipids in BNT162B2 LNPs with PyCB ILs. The BNT162B2 LNP consists of ALC-0315 IL, DSPC, cholesterol, and 1 ,2-dimyristoyl-rac- glycerol-methoxy(poly(ethylene glycol)) (DMG-PEG; 50 / 10 / 38.5 / 1.5, mol / mol) and showed significant hepatic tropism following IV injection. The structural gaps in the LNPs, caused by the removal of cholesterol and PEGylated lipids from BNT162B2 LNPs, were compensated by the inclusion of PyCB ILs. Based on the above criteria, a series of LNPs were formulated by fixing the DSPC level and gradually increasing the percentage of PyCB ILs from 0 to 67.5% of the total lipids. Luciferase (Luc) mRNA was used, and these LNPs were labeled with l,T-dioctadecyl-3,3,3',3'-tetramethylindotricarbocyanine iodide (DiR), a lipophilic near-infrared fluorescent dye, to enable the observation of LNP distribution across different organs in C57BL / 6 mice using the IVIS imaging system.

[0177] As shown in FIG. 8a, the total removal of cholesterol and PEGylated lipids, leaving only ALC-0315 ILs and DSPC in the LNP formulation (denoted as ThrCo(0%) LNP), resulted in a significant reduction in both LNP accumulation and Luc mRNA translation in the liver. Unfortunately, although the accumulation of ThrCo(0%) LNPs in the spleen increased, the luciferase expression remained low. One possible explanation is the structural instability of ThrCo(0%) LNPs, as evidenced by our measurements of the polydispersity index (PDI) and particle size. Such instability caused by the absence of key stabilizing component leads to inconsistent mRNA encapsulation and reduced delivery efficiency. The mRNA encapsulation efficiency of ThrCo(0%) LNPs was only 25.23%. As the molar percentage of PyCB ILs increased while the ratio of ALC-0315 IL to DSPC is fixed, the Luc mRNA translation level in the spleen initially increased and then decreased, peaking at a PyCB IL percentage of 45%. When the PyCB IL percentage was increased to 67.5% (denoted as ThrCo(67.5%) LNP), a reduction in mRNA translation efficiency was observed (FIGS. 8b-8c). This may be attributed to the excessive zwitterionic groups on the LNP surface, which could hinder cellular uptake or lead to LNPs adhering to the cell membrane without being effectively internalized. These results underscore the critical roleof both the presence and optimal proportion of PyCB ILs in achieving efficient spleenspecific mRNA translation.

[0178] Then, the accumulation of ThrCo(45%) LNPs was compared with that of BNT162B2 LNPs in the liver and spleen. As shown in FIG. 8e, these two formulations exhibited distinct in vivo distribution profiles. Compared to BNT162b2 LNPs, ThrCo(45%) LNPs showed the reduced liver accumulation while enhancing spleen accumulation following IV injection. Consequently, injection of ThrCo(45%) LNPs resulted in higher spleen-specific mRNA translation than BNT162B2 LNPs (FIG. 8d). To optimize mRNA translation efficiency, a detailed investigation of the PyCB IL percentages in LNPs was conducted. Consistent with the results from the initial screening, the optimal percentage of PyCB ILs was found to be 50%, corresponding to an ideal molar ratio of PyCB IL to ALC- 0315 IL of 5:4 (FIG. 8f). Finally, further experiments fixed the molar ratio of PyCB IL and ALC-0315 IL and explored the effect of the DSPC helper lipid percentages on the final mRNA translation efficiency. According to the result, a 20% DSPC level was optimal for delivering and translating mRNA in the spleen (FIGS. 8g-8i). The following studies used this optimal LNP formulation, referred to as ThrCo LNP (ALC-0315 IL / PyCB IL / DSPC = 35.6 / 44.4 / 20, mol / mol). The above series of optimizations did not significantly alter the in vivo distribution of LNPs, further confirming the beneficial effect of cholesterol removal in LNPs for reducing liver tropism and enhancing spleen accumulation. Hepatic mRNA delivery typically relies on cholesterol-mediated adsorption of lipoproteins, which facilitates uptake via receptor-mediated endocytosis in hepatocytes. ThrCo LNPs lack cholesterol and thus have limited lipoprotein binding capacity. As a result, even though a portion of LNPs accumulate in the liver, they fail to enter hepatocytes efficiently, thereby leading to minimal mRNA expression.

[0179] Further experiments compared the optimized ThrCo LNPs with the BNT162b2 LNPs. Cy5-labeled Luc mRNA was encapsulated within DiR-labeled LNPs, enabling simultaneous tracking of both LNP and mRNA distribution. As shown in FIGS. 8j- 81, compared to BNT162b2 LNPs, the liver accumulation of ThrCo LNPs was reduced to 30.9%, while spleen accumulation increased by 2.6-fold, resulting in a substantial accumulation of mRNA levels in the spleen. Consequently, mRNA expression in the spleen was 4.5 times higher than that achieved with BNT162b2 LNPs (FIG. 8m). Further experiments evaluated mRNA expression at 24 hours and 48 hours post-injection. At allassessed time points, ThrCo LNPs consistently demonstrated stronger spleen-specific mRNA expression compared to BNT162b2 LNPs. Furthermore, when benchmarked against the previously reported spleen-targeting 18PA LNPs and 93-0170 LNPs, ThrCo LNPs exhibited a remarkable 42.1-fold and 9.3-fold increase in splenic mRNA expression, respectively.

[0180] Then, serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were measured to assess potential liver toxicity following IV administration of ThrCo LNPs. Phosphate buffered saline (PBS) was used as a negative control, while BNT162B2 LNPs served as the control group. Two different mRNA-LNP concentrations were tested: 1 mg / kg and 5 mg / kg. At the lower dose, no significant elevation in ALT and AST levels was observed in either the BNT162B2 LNP or ThrCo LNP groups. However, at the higher dose, the BNT162B2 LNP group showed a trend of increased ALT and AST levels compared to the ThrCo LNP group (FIGS. 8n-8q). This suggests that ThrCo LNPs induce lower liver toxicity, which could be attributed to their weaker hepatic tropism and the negligible expression of mRNA in the liver. These findings highlight the potential advantage of ThrCo LNPs in reducing off-target hepatic effects while maintaining their therapeutic efficacy. Further experiments evaluated the long-term stability of ThrCo LNPs under 4°C (standard storage condition) and -80°C storage, as well as short-term stability at 37°C by measuring the mRNA expression efficiency in vivo. BNT162b2 LNPs were employed as a benchmark. From the results, it was observed that ThrCo LNPs exhibited comparable stability to BNT162b2 LNPs under standard storage conditions (4 °C, 48 hours) and following one-week storage at -80 °C. Notably, ThrCo LNPs demonstrated superior stability after 24-hour incubation at 37 °C, compared to BNT162b2 LNPs.

[0181] Characterization of ThrCo LNPs

[0182] Further experiments conducted a comprehensive characterization of the formulated ThrCo LNPs, which exhibited a particle size of around 200 nm with a narrow PDI (FIG. 9a). Transmission electron microscopy (TEM) image showed a spherical morphology with a solid core (FIG. 9b). The stability of ThrCo LNPs was assessed by incubating them in PBS at 37°C for 48 hours, revealing no significant changes in particle size. A similar trend was observed in the presence of the serum, thus highlighting the robustness of the formulation under physiologically relevant conditions (FIG. 9c). To elucidate the cellular uptake mechanism of ThrCo LNPs, DiR-labeled LNPs encapsulating Enhanced GreenFluorescent Protein (EGFP) mRNA were co-incubated with DC2.4 cells in the presence of different pathway-specific inhibitors (FIG. 9d). Treatment with amiloride (macropinocytosis inhibitor), nystatin (caveolae-mediated endocytosis inhibitor), and chlorpromazine (clathrin-mediated endocytosis inhibitor) permitted an elucidation of the internalization pathways involved. As shown in FIGS. 9e-9j, both nystatin and chlorpromazine significantly reduced cellular uptake and mRNA expression, indicating that ThrCo LNPs are primarily internalized via endocytosis. These results suggest that the uptake mechanism of ThrCo LNPs closely resembles that of clinically approved LNP platforms. I ntracellular trafficking of ThrCo LNPs was then visualized using confocal laser scanning microscopy (CLSM). ThrCo LNPs were fluorescently labeled with N-(7- nitrobenzo-2-oxa-l,3-diazole)-phosphatidylethanolamine (NBD-PE), thus enabling tracking within DC2.4 cells. LysoTracker™ Red was added to stain the endosomes. As shown in FIGS. 9k and 91, a strong green fluorescence signal was observed in the cytoplasm after 3 hours of incubation, indicating efficient endosomal escape.

[0183] To verify the endosomal membrane disruption and LNP dissociation, Forster resonance energy transfer (FRET) assays were performed (FIG. 9m). Two FRET probes, NBD-PE and rhodamine-phosphatidylethanolamine (Rhod-PE), were formulated into a single endosomal mimicking liposome, which leads to attenuated NBD fluorescence because of FRET from NBD to Rhod. When membrane disruption occurs, the increased distance between the two probes enhances the NBD signal. The FRET efficiency of the mixture of ThrCo LNPs and endosomal mimicking liposomes was measured under different pH conditions. As shown in FIG. 9n, ThrCo LNPs exhibited effective membrane disruption capability as the pH decreased. Subsequently, ThrCo LNPs incorporating NBD- PE and Rhod-PE were prepared, and a similar assay was conducted to assess the dissociation of the LNPs. ThrCo LNPs were easy to disassemble once mixing with the endosomal mimicking liposomes, resulting in mRNA release (FIG. 9o). The endosomolytic activity of ThrCo LNPs was then visualized using a hemolysis assay, a surrogate model that assesses the hemoglobin release following the incubation of LNPs with red blood cells (RBC).

[0184] The endosomolytic activity at physiological pH (7.4) reflects the cytotoxicity of the nanoparticles, while lysis at pH values below 6.0 serves as a model for the ability of LNPs to escape from endosomes. ThrCo LNPs demonstrated significant hemolytic activity andshowed no detectable RBC lysis at pH 7.4 (FIG. 9p). In addition, no RBC lysis was observed following prolonged co-incubation of ThrCo LNPs with RBCs under physiological conditions, confirming the excellent biocompatibility of ThrCo LNPs in the blood (FIG. 9q).

[0185] The effective disruption of the endosomal membrane may be attributed to the fact that PyCB IL retains key characteristics of the clinically used ALC-0315 IL, including its protonation capacity and cone-shaped structure, both of which facilitate endosomal escape. Beyond these similarities, the PyCB group itself exhibits a strong protonation ability.Under neutral pH conditions, the zwitterionic PyCB groups interact with water molecules through charge-assisted hydrogen bonding, forming PyCB-H2O complexes. As the pH decreases, these bound water molecules undergo protonation, converting the PyCB groups into a cationic state through forming PyCB-HaCL complexes (FIG. 7c). Thus, the pH- responsiveness of the PyCB headgroups can be assessed by monitoring the protonation of their associated water molecules, as indicated by comparing the ’ H NMR spectra of PyCB ILs exposed to pH 7.4 and 6.5. After incubating the PyCB ILs to deionized water at different pH for 10 minutes, the samples were lyophilized to remove free water while preserving the bound water molecules. When PyCB ILs were exposed to a pH shift from 7.4 to 6.5, 'H NMR spectra of hydrogen in bound water showed disappearance of the peak at 3.3-3.4 ppm (upfield), along with the appearance of a broad new peak at 3.6-3.7 ppm (downfield). This spectral shift confirms the protonation of the PyCB headgroup, highlighting its pH-responsiveness. Zeta potential analysis further demonstrated that ThrCo LNPs undergo a sensitive transition to a positively charged state (FIG. 9r). The phase transition analysis using established31P nuclear magnetic resonance (NMR) spectroscopy techniques revealed that PyCB ILs and ALC-0315 ILs exhibited similar peak shapes and chemical shifts (FIG. 9s). This finding suggests that PyCB ILs mediate membrane disruption by promoting the transition from a bilayer structure to the inverted hexagonal II phase(46), which is the same as ALC-0315 ILs.

[0186] ThrCo LNPs as a robust mRNA vaccine carrier

[0187] A robust cancer vaccine carrier should efficiently deliver mRNA to APCs and enhance mRNA expression within APCs to promote antigen presentation and subsequent immune responses. Next experiments investigated the specific immune cells types within the spleen targeted by the ThrCo LNPs. To enable precise tracking, the followingexperiments utilized genetically engineered Ail 4 mice, in which the red fluorescent protein tdTomato is stably expressed following intracellular delivery of Cre recombinase (Cre) mRNA (FIG. 10a). Flow cytometry was employed to quantify the proportion of tdTomato- expressing cells. Cre mRNA, serving as an activator, was encapsulated within LNPs and administered via IV injection. Both ThrCo LNPs and BNT162b2 LNPs successfully induced tdTomato expression in DCs and macrophages, demonstrating effective delivery of Cre mRNA to APCs. Notably, ThrCo LNPs achieved significantly higher tdTomato expression levels compared to BNT162b2 LNPs, highlighting the great potential of ThrCo LNPs as robust vaccine carriers (FIG. 10b). In this study, Ail4 mice were used to evaluate the distribution of Cre mRNA delivery. The binary nature of Cre-loxP recombination enables detection of even low levels of Cre mRNA, thus permitting assessment of how broadly the mRNA was delivered across different cell populations. In contrast, FIG. 8j evaluates the expression of luciferase, where the signal is more directly proportional to the amount of mRNA delivered and translated in each cell. Together, these data suggest that ThrCo LNPs not only reach a broader range of cells (as shown by Cre expression in Ail4 mice) but also support higher expression levels within those cells (as shown by luciferase activity). These results are not contradictory but rather complementary in supporting the enhanced translation efficiency of ThrCo LNPs.

[0188] Next experiments selected ovalbumin (OVA) as a model antigen and vaccinated mice intravenously with 5 pg of OVA mRNA per mouse to compare the ability of ThrCo LNPs and BNT162b2 LNPs to induce antigen-specific immune responses. Briefly, mice received a prime dose on day 0, followed by booster doses on days 7 and 14. Splenocytes were then collected on day 17 to assess OVA-specific CD8+T cell activation in the spleen after vaccination using a tetramer assay (FIG. 10c). As shown in FIGS. lOd and lOe, negligible levels of OVA-specific CD8+T cells were detected in PBS control group (OVA Tetramer'. 0.57%). In contrast, vaccination with LNP-based formulations led to a significant increase in OVA-specific T cell responses. Notably, the ThrCo LNP-based vaccine induced a markedly higher frequency of CD8+T cell activation (9.06%) compared to the BNT162b2 LNP-based vaccine (3.96%). This finding underscores the exceptional potential of spleen-targeting ThrCo LNPs as an advanced carrier for cancer vaccination, facilitating robust antigen-specific cellular immunity.

[0189] Effective therapeutic outcomes often require repeated administrations. However, for commercially available LNP formulations containing PEG components, PEG immunogenicity has been widely reported, leading to reduced efficacy upon repeated dosing. Thus, further experiments evaluated the Luc mRNA delivery efficiencies of ThrCo LNPs and BNT162b2 LNPs under repeated (multiple) injection conditions following the schedule outlined in FIG. lOf. Luciferase expression in the livers and spleens was quantified using IVIS imaging system. Mice that received a single injection of Luc mRNA- loaded LNPs served as the control groups, providing a baseline for assessing any efficacy loss resulting from repeated administrations. By using EGFP mRNA instead of Luc mRNA for the first two injections, the present work aimed to prevent the induction of proteinspecific immunogenicity that could potentially hinder protein expression during the final injection. As a result, the decreased efficacy observed in the last injection can be attributed to ABC effects, caused by the immunogenicity of the components in LNPs. As shown in FIGS. 4g-4i, when BNT162b2 LNPs were employed as the carrier, a significant reduction in luciferase expression was observed in both the livers and spleens in multiple injection groups. In contrast, no significant decline in luciferase expression was observed after multiple injections upon using ThrCo LNPs as the carrier. This highlights the superior efficacy of ThrCo LNPs in maintaining robust mRNA delivery upon repeated administrations. This advantage may be attributed to the zwitterionic surface of ThrCo LNPs, which exhibits low protein adsorption, thereby reducing the antibody recognition and binding. Consequently, this leads to a diminished antibody-mediated ABC effect, preserving the delivery efficiency upon repeated administrations. To verify this hypothesis, the protein adsorption on the ThrCo LNP surface was evaluated through incubating LNPs with whole mouse serum at 37 °C. Compared to BNT162b2 LNPs, ThrCo LNPs, which contain abundant zwitterionic groups, exhibit significantly lower protein adsorption (FIG. lOj and FIG. 10k).

[0190] ThrCo LNP-based vaccines inhibit tumor growth and metastasis

[0191] The spleen is a critical organ for orchestrating systemic immune responses. Targeted cancer vaccine delivery to the spleen facilitates the direct activation of APCs, thereby enhancing cytotoxic T cell priming and promoting their entry into circulation. (52) This process leads to a significant increase in the level of activated lymphocytes in the bloodstream, highlighting their potential for enhancing systemic anti -tumor immunity.Initial experiments investigated the efficacy of ThrCo LNP-based vaccines in eradicating circulating tumor cells and preventing their metastasis to lungs. C57BL / 6 mice were intravenously inoculated with B16F10 tumor cells that express OVA (B 16F10-OVA), followed by vaccination with different LNP formulations encapsulating OVA mRNA on days 1, 6, and 11 (FIG. Ila). On day 21, the mice were euthanized, and lung tissues were collected. As shown in FIG. 11b, numerous black tumor nodules were observed in the lung tissues of the PBS control group. Treatment with the BNT162b2 LNP-based formulations reduced the number of tumor nodules. Notably, in the ThrCo LNP -treated group, no detectable tumor nodules were observed, suggesting a complete blockade of metastatic seeding. Histological analysis of H&E-stained lung sections further corroborated these findings, confirming the absence of tumor lesions in the ThrCo LNP-treated mice (FIG. 11c).

[0192] Next experiments evaluated the activation of cytotoxic CD8+ T cells in the spleen and blood, as they are the primary effector cells in combating tumor cells. The efficacy of ThrCo LNP-based vaccines in triggering OVA-specific cytotoxic T cell activation was assessed using flow cytometry. As shown in FIG. lid, administration of ThrCo LNP-based vaccines resulted in significantly enhanced population of OVA-specific CD8+ T cells (OVA Tetramer+, 9.22%) in the spleens, compared to BNT162b2 LNP-based formulations (3.17%). A similar trend was observed when analyzing OVA-specific CD8+ T cells in the blood (FIG. Ilf)- Moreover, ThrCo LNP-based vaccine treatment triggered increased IFN- y expression (a cytotoxic cytokine that initiates tumor cell apoptosis, IFN-y+, 7.43%) on the CD8+ T cell surface, which was 2.3-fold higher than that of BNT162b2 LNP-based formulations (3.24%) (FIG. lie). The improved T cell activation can be correlated with the excellent spleen-targeting delivery capability of the ThrCo LNPs. Efficient cytotoxic T cell activation also contributed to the establishment of robust immune memory, as evidenced by the evaluation of effector memory T (TEM) cell and central memory T (TCM) cell levels in the spleens. As shown in FIG. 11g, ThrCo LNP-based vaccine treatment resulted in a significant increase in the proportions of TEM cells (CD62LlowCD44hlgh, 17.5%) and TCM cells (CD62LhlghCD44hlgh, 31.6%), compared to other treatment groups.

[0193] T cell -dependent immune responses, shaped by APCs, are inherently antigenspecific. Next experiments investigated the efficacy of ThrCo LNP-based vaccines in preventing cognate tumor recurrence. As shown in FIG. llh, the mice were initiallyintravenously injected with B16F10-OVA tumor cells followed by vaccination. To model tumor recurrence, these mice were inoculated with B16F10-OVA (left flank) and noncognate EO771 murine mammary (right flank) tumor cells via subcutaneous (SC) injection on day 21. Normal mice, which received these two tumor cell types, served as a control group (gray lines; FIG. Hi and FIG. llj). The recurrence of B16F10-OVA tumor was completely suppressed in four out of five mice treated with the ThrCo LNP-based vaccines, whereas only slight suppression of EO771 tumor growth was observed. This selective prevention of B16F10-OVA tumor recurrence can be attributed to the infiltration and activation of CD8+ T cells within the B16F10-OVA tumors. This was confirmed by detecting the intra-tumoral CD8+ T cell infiltration and activation. Only a significant increase in the CD8+ T cell population (CD8+, 21.1%) and enhanced IFN-y expression (IFN-y+, 12.3%) was observed in the B16F10-OVA tumors (FIGS, llk-lln).

[0194] The therapeutic efficacy of ThrCo LNP-based vaccines was then evaulated in B 16F10-OVA tumor-bearing mice, which received IV injections of different formulations (FIG. 12a). As shown in FIGS. 12b-12e, treatment with ThrCo LNP-based vaccines effectively suppressed the tumor growth and improved the survival rate of mice from 0% to 60% over a 50-day observation period. To investigate the underlying anti-tumor immune response, a tetramer assay was performed to assess APC-mediated antigen-specific CD8+ T cell activation in the spleens. ThrCo LNP-based vaccines induced a significantly higher level of OVA-specific CD8+ T cells (OVA Tetramer+, 10.8%) compared to BNT162b2 LNP-based formulations (4.43%) in the spleens (FIG. 12f), accompanied by a notable increase in IFN-y secretion (lFN-y+. ThrCo LNP, 6.56%; BNT162b2 LNP, 2.70%; see FIG. 12g).

[0195] From next experiments analyzing the activation of intra-tumoral CD8+ T cells, it was observed that ThrCo LNP-based vaccine treatment resulted in the highest CD8+ T cell infiltration (CD8+, 15.9%) (FIG. 12h). Additionally, the levels of OVA-specific CD8+ T cells within tumor tissues were evaluated. As shown in FIG. 12i, ThrCo LNP-based vaccines significantly enhanced OVA-specific CD8+ T cells infiltration (OVA Tetramer+, 16.1%) within tumor tissues compared to BNT162b2 LNP-based formulations (7.72%), confinning the activation of a potent OVA-specific T cell-mediated anti-tumor immune response. Moreover, the ThrCo LNP-based vaccines elicited a robust immune memory response (FIG. 12j). These results suggest that ThrCo LNPs represent a promising carrierfor mRNA vaccines, capable of eliciting robust T cell-dependent adaptive immunity and long-lasting immune memory.

[0196] Discussion

[0197] The clinical translation of LNP platforms faces significant challenges related to their composition. Cholesterol, a common LNP component, facilitates lipoprotein binding, which enhances hepatic uptake but concurrently restricts biodistribution to extrahepatic organs, thereby limiting therapeutic flexibility. In parallel, PEGylated lipids — widely used to prolong circulation time — can induce the formation of anti-PEG antibodies, leading to ABC effects and reduced efficacy upon repeated dosing. Notably, the prevalence of preexisting or treatment-induced anti-PEG immunity appears to be increasing in the general population, raising the efficacy concerns for repeated administration in clinical settings. Through a combination of rational lipid design and in vivo performance optimization, the present work developed a cholesterol- and PEGylated lipid-free ThrCo LNP to achieve spleen-specific mRNA translation by employing zwitterionic PyCB ILs as a crucial component. Mechanistically, the superior hydrophilicity of PyCB ILs enhance the stability of the LNP outer membrane in aqueous environments, thereby preventing LNP disintegration and avoiding the low encapsulation efficiency caused by the absence of cholesterol. The zwitterionic surface formed by PyCB ILs also eliminates the need for PEG. Consequently, unlike traditional LNPs that incorporate cholesterol and PEGylated lipids, the present design circumvents their associated limitations, including hepatic tropism and ABC effects, thereby enhancing spleen-targeted delivery and improving vaccine bioavailability in the setting of multiple (repeated) administrations. The rationally designed ThrCo LNP allows for efficient mRNA vaccine delivery to the spleen, significantly boosting antigen presentation by splenic APCs and promoting robust antigen-specific adaptive immune activation that surpass those elicited by the BNT162b2 LNP formulation from Pfizer-BioNTech.

[0198] Compared to BNT162b2 LNPs, the present formulation exhibited markedly increased spleen accumulation and increased mRNA translation within the spleen by more than 4-fold, while maintaining high efficacy across repeated administrations. This selective targeting resulted in potent antigen-specific cytotoxic CD8+ T cell activation, substantial suppression of tumor growth, inhibition of lung metastasis, and prevention of tumor recurrence in mouse melanoma models. These findings highlight the clinical potential ofThrCo LNPs as a promising platform for mRNA vaccine delivery. This work not only demonstrates a robust ionizable lipid but also provides the molecular design principles for the rational design of better ionizable lipids and LNPs to facilitate extrahepatic delivery, offering a blueprint for designing new LNPs with improved targeting efficiency and therapeutic efficacy.

[0199] Despite these promising results, several limitations warrant consideration. While ThrCo LNPs demonstrated reduced hepatic accumulation relative to BNT162b2 LNPs, complete avoidance of liver uptake was not achieved. Given the liver’s central role in nanoparticle clearance and lipid metabolism, even low levels of hepatic accumulation may pose risks of hepatotoxicity in long-term or high-dose vaccine regimens.

[0200] Moreover, the current spleen-targeting selectivity, although improved, may be sub- optimal for achieving cell-type-specific delivery within the complex microarchitecture of lymphoid tissues. Further refinement of lipid composition, incorporation of active targeting ligands (e.g., DC-specific aptamers or antibodies), or modular design of responsive elements that selectively activate in the APC-rich zones of the spleen may further improve delivery precision. In addition, integration with personalized neoantigen identification pipelines could position ThrCo LNPs as a platform technology for individualized cancer vaccination.

[0201] In summary, ThrCo LNPs represent a rationally engineered, biocompatible, and spleen-directed mRNA delivery system. By overcoming key limitations of existing LNP platforms, including liver tropism, PEG-related immunogenicity, and poor efficiency for repeated dosing, ThrCo LNPs pave the way for the development of next-generation mRNA vaccines targeting cancer and other chronic diseases.Example 3Three-component LNPs formulated by MeDZ lipid, helper lipid, and DOPS for robust spleen-specific mRNA expression

[0202] It was herein found that the LNPs composed only of MeDZ lipid and DSPC helper lipid (TwoCo LNP) showed excellent spleen accumulate ability. However, when loading mRNA, its effectiveness in promoting mRNA expression was not ideal. This may be because the zwitterionic surface of the LNPs reduced the interactions with cells, and thus failed to efficiently deliver the mRNA to cells. Introducing different lipids targetingdifferent immune cells could improve the efficacy. As a proof of concept, 1,2-dioleoyl-sn- glycero-3-phospho-L-serine (DOPS) lipid that can target macrophages was integrated to prepare a new three-component LNP (ThrCo (DOPS) LNP).

[0203] As shown in FIGS. 13a-13b, compared to TwoCo LNP, ThrCo (DOPS) LNPs did not compromise the ability to efficiently accumulte within the spleen. Furthermore, the latter significantly improved the expression efficiency of mRNA. This shows that the introduction of DOPS lipids makes the original TwoCo LNP more easily taken up by target cells, thereby promoting mRNA expression.Example 4Three-component LNPs formulated by zwitterionic ionizable lipid, helper lipid, and antibody -functionalized lipid for cell-specific targeting in the spleen

[0204] The spleen, the largest secondary lymphoid organ in humans, plays a central role in the initiation and regulation of adaptive immune responses. Its structure is divided into two main functional areas: the red pulp and the white pulp. The red pulp is rich in innate immune cells such as macrophages, dendritic cells, and natural killer cells, which are responsible for clearing blood-borne pathogens and presenting antigens. The white pulp is densely populated with T cells and B cells and serves as a critical site for antigen-specific immune responses. In recent years, nanocarrier-based targeting strategies toward the spleen have garnered significant attention in vaccine development and immunotherapy. However, most existing delivery systems rely on the efficient phagocytic activity of macrophages for splenic accumulation. This single-cell targeting approach limits the interaction of carriers with other key immune cells in the spleen, such as T cells, B cells, and NK cells, thereby hindering the precise modulation of diverse immune responses.

[0205] To address this limitation, the present work developed a two-component lipid nanoparticle (TwoCo LNP) system based on a novel zwitterionic ionizable (ZI) lipid (FIG. 14a). Following intravenous (IV) administration, these LNPs exhibit significant accumulation in the spleen. Notably, due to their zwitterionic surface characteristics, they avoid nonspecific cellular uptake, particularly by macrophages, which prevents premature clearance and allows prolonged presence in the splenic microenvironment. Building on this, the present work proposes a "two-step" targeting strategy: first, leveraging the intrinsic organ-targeting properties of LNPs to achieve efficient splenic accumulation; second,functionalizing the LNP surface with specific antibodies to enable precise recognition and internalization by predefined immune cell subsets (FIG. 14b).

[0206] This approach not only extends the potential for non-macrophage targeting within the spleen but also offers a versatile platform for precision immunotherapy. Antibody- conjugated LNPs can efficiently deliver genetic cargo into target cells, enabling in vivo generation or expansion of engineered immune cells, such as CAR-T cells, CAR-NK cells, and engineered B cells. This strategy provides a safe and efficient delivery platform that could synergize with next-generation cellular immunotherapies and spleen microenvironment modulation, opening new avenues for advanced immune engineering.

[0207] Part 1. Two-component LNPs as a platform to minimize nonspecific macrophage uptake

[0208] LNPs composed only of a zwitterionic (ZI) ionizable lipid and DSPC (“TwoCo LNPs”) exhibited reduced macrophage uptake, suggesting utility as a platform for precise recognition and internalization by predefined immune cell subsets after antibody -binder modification.

[0209] In vivo, the present work characterized TwoCo LNP biodistribution between the white and red pulp of the spleen following i.v. administration. C57BL / 6 mice received Cy5- mRNA-loaded TwoCo LNPs; spleens were harvested and stained. From 2 to 8 hours postinjection, signal shifted from the macrophage-rich white pulp to the red pulp, where other immune cell types predominate (FIGS. 15a and 15b).

[0210] In vitro, TwoCo LNPs showed markedly lower uptake than a commercial BioNTech-like LNP in both RAW264.7 cells and bone-marrow-derived macrophages across multiple doses (FIGS. 16a-d).

[0211] Real-time imaging assessed co-localization of LNPs with macrophages. At multiple time points post-injection, TwoCo LNPs showed significantly lower co-localization with macrophages than the commercial BioNTech LNP (FIG. 17a, 17b, and 17c).

[0212] Part 2. Three-component LNPs containing an antibody-functionalized PEG lipid

[0213] The LNPs composed only of ZI lipid and DSPC helper lipid (TwoCo LNP) showed excellent spleen accumulate ability. By introducing different lipids that target different immune cells, it is expected to achieve effective cell-specific targeting delivery of mRNA in the spleen. Here, antibody-functionalized lipids that can target certain cell type (e.g., anti-CD5 antibody, anti-CD8 antibody, anti-CD20 antibody, etc.) was integrated to prepare a new three-component LNP (including ThrCo(aCD5) LNP, ThrCo(aCD20) LNP, ThrCo(aCD2) LNP, and ThrCo(aCD205) LNP).

[0214] The present work constructed a cell-specific targeting formulation by incorporating an antibody-functionalized PEG-lipid into the TwoCo formulation. As shown in FIG. 18a, compared to TwoCo LNP, antibody -modified ThrCo LNP did not compromise the ability to efficiently accumulte within the spleen. Furthermore, the latter significantly improved the expression efficiency of mRNA, confirming antibody-mediated cell specific uptake. To further determine cell targeting, next experiments performed flow cytometry analysis. The results showed that ThrCo (aCD5) LNP, which can target T cells, can lead to specific mRNA expression of T cells in the spleen (FIG. 18b). Similarly, ThrCo (aCD20) LNP targeting B cells can lead to specific mRNA expression of B cells in the spleen (FIG. 18c). These results indicate that the presently descried strategy achieves specific targeting of cells in the spleen.

[0215] Part 3. Three-component LNPs containing an antibody -functionalized PCB lipid

[0216] We constructed a cell-specific targeting formulation by incorporating an antibody- functionalized poly(carboxybetaine)(PCB)-lipid into the TwoCo formulation. As shown in FIG. 19a, compared to TwoCo LNP, antibody-modified ThrCo LNP did not compromise the ability to efficiently accumulte within the spleen. Furthermore, the latter significantly improved the expression efficiency of mRNA, confirming antibody-mediated cell specific uptake. To further determine cell targeting, next experiments performed flow cytometry analysis. The results showed that ThrCo (aCD5) LNP, which can target T cells, can lead to specific mRNA expression of T cells in the spleen (FIG. 19c and FIG. 19d). Similarly, ThrCo (aCD20) LNP targeting B cells can lead to specific mRNA expression of B cells in the spleen (FIG. 19b and FIG. 19e). These results indicate that the presently described strategy achieves specific targeting of cells in the spleen.

[0217] While there have been shown and described what are at present considered the preferred embodiments of the invention, those skilled in the art may make various changes and modifications which remain within the scope of the invention defined by the appended claims.

Claims

WHAT IS CLAIMED IS:

1. A lipid composition having the following formula:wherein:R is a linear or branched hydrophobic group containing at least four carbon atoms and at least one linkage in which at least three carbon atoms are connected together, and wherein R optionally contains one or more heteroatoms selected from O, N, and S atoms;L is a bond or a linker containing 1-3 atoms, wherein at least one of the atoms is a heteroatom selected from O, N, and S atoms;Ra, Rb, Rc, and Rdare independently selected from hydrogen atom, halide atom, methyl group, halomethyl group, C=N, hydroxy group, and methoxy group; and n is an integer of 1, 2, or 3.

2. The lipid composition of claim 1, wherein R is a linear or branched hydrophobic group containing at least four carbon atoms connected together, and wherein R contains at least one heteroatom selected from O and N.

3. The lipid composition of any one of claims 1-2, wherein L = C(O)X and the lipid composition has the following formula:wherein X is O, NH, or S; and R, Ra, Rb, Rc, Rd, and n are as defined in claim 1.

4. The lipid composition of claim 1, wherein the lipid composition has the following formula:wherein:L, Ra, Rb, Rc, Rd, and n are as defined in claim 1 ;L’ is a bond or a linker of the formula -(Cthlm- wherein m is an integer of 1-30: andR1and R2are independently selected from hydrogen atom and linear or branched hydrophobic groups containing at least four carbon atoms and at least one linkage in which at least three carbon atoms are connected together, and optionally containing one or more heteroatoms selected from O, N, and S atoms.

5. The lipid composition of claim 4, wherein at least one of R1and R2has the following formula:— (CH2)P-L1-R3(3 a) wherein:L1is independently a linker containing 1-3 atoms, wherein at least one of the atoms is a heteroatom selected from O and N atoms;R3is independently a linear or branched hydrophobic group containing at least four carbon atoms and at least one linkage in which at least three carbon atoms are connected together, and optionally containing one or more heteroatoms selected from O, N, and S atoms; and p is an integer of 1-30.

6. The lipid composition of claim 5, wherein L1has the formula C(O)X, wherein X is O, NH, or S.

7. The lipid composition of any one of claims 5-6, wherein R1and R2both independently have the formula:— (CH2)p-L1-R3(3a)8. The lipid composition of any one of claims 1-7, wherein Ra, Rh, Rc, and Rdare all hydrogen atom.

9. A lipid nanoparticle composition comprising a lipid composition of any one of claims 1-8 and a therapeutic molecule.

10. The lipid nanoparticle composition of claim 9, further comprising at least one lipid not within the scope of Formula (1) in claim 1.

11. The lipid nanoparticle composition of claim 10, wherein said lipid not within the scope of Formula (1) is selected from the group consisting of: (i) tertiary amine (ionizable) lipids, (ii) non-tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group; (iii) lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, and (iv) sterols.

12. The lipid nanoparticle composition of claim 10, wherein said lipid not within the scope of Formula (1) comprises: (i) a tertiary amine (ionizable) lipid.

13. The lipid nanoparticle composition of claim 12, wherein said tertiary amine (ionizable) lipid is selected from the group consisting of ALC-0315 lipid, SM-102 lipid, and DLin-MC3-DMA lipid.

14. The lipid nanoparticle composition of any one of claims 12-13, further comprising at least one of: (ii) non-tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group; (iii) lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, and (iv) sterols.

15. The lipid nanoparticle composition of claim 10, wherein said lipid not within the scope of Formula (1) comprises: (ii) a non-tertiary amine lipid containing a hydrophobic tail attached to a hydrophilic head group.

16. The lipid nanoparticle composition of claim 15, wherein said non-tertiary amine lipid is a phospholipid or non-phospholipid (e.g., carboxybetaine lipid).

17. The lipid nanoparticle composition of claim 16, wherein said phospholipid is selected from the group consisting of distearoylphosphatidylcholine (DSPC), dimyristoylphosphatidylcholine (DMPC), l,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE), l,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), dioleoylphosphatidyl glycerol (DOPG), dipalmitoylphosphatidyl glycerol (DPPG), palmitoyloleoyl-phosphatidyl ethanolamine (POPE), dipalmitoylphosphatidyl ethanolamine (DPPE), dimyristoylphosphoethanolamine (DMPE), distearoylphosphatidyl ethanolamine (DSPE),16-O-monomethyl-phosphoethanolamine, 16-O-dimethyl-phosphoethanolamine, 18-1- trans-phosphoethanolamine, l-stearoyl-2-oleoyl phosphatidy ethanolamine (SOPE), and 1,2- dioleoyl-sn glycero-3-phophoethanolamine (transDOPE).

18. The lipid nanoparticle composition of any one of claims 15-17, further comprising at least one of: (i) tertiary amine (ionizable) lipids, (iii) lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail, and (iv) sterols.

19. The lipid nanoparticle composition of claim 10, wherein said lipid not within the scope of Formula (1) comprises: (iii) lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail.

20. The lipid nanoparticle composition of claim 19, wherein said lipid not within the scope of Formula (1) comprises: (iii) lipids containing an uncharged polymeric group attached to a hydrophobic tail.

21. The lipid nanoparticle composition of claim 20, wherein said uncharged polymeric group is a polyalkylene oxide (e.g., polyethylene glycol) group.

22. The lipid nanoparticle composition of claim 19, wherein said lipid not within the scope of Formula (1) comprises: (iii) lipids containing a zwitterionic polymeric group attached to a hydrophobic tail.

23. The lipid nanoparticle composition of claim 22, wherein the zwitterionic polymeric group is selected from the group consisting of poly(carboxybetaine) (PCB), poly(sulfobetaine), poly(phosphobetaine), poly(phosphatidylcholine), poly(trimethylamine N-oxide), and glutamic acid-lysine (EK) containing polypeptide.

24. The lipid nanoparticle composition of any one of claims 19-23, further comprising at least one of: (i) tertiary amine (ionizable) lipids, (ii) non-tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group: and (iv) sterols.

25. The lipid nanoparticle composition of any one of claims 22, further comprising non- tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group.

26. The lipid nanoparticle composition of claim 25, wherein the non-tertiary amine lipids are selected from distearoylphosphatidylcholine (DSPC), dimyristoylphosphatidylcholine (DMPC), 1 ,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE), l,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), dioleoylphosphatidyl glycerol (DOPG), dipalmitoylphosphatidyl glycerol (DPPG), palmitoyloleoyl-phosphatidyl ethanolamine (POPE), dipalmitoylphosphatidyl ethanolamine (DPPE), dimyristoylphosphoethanolamine (DMPE), distearoylphosphatidyl ethanolamine (DSPE),16-O-monomethyl -phosphoethanolamine, 16-O-dimethyl-phosphoethanolamine, 18-1- trans-phosphoethanolamine, l-stearoyl-2-oleoyl phosphatidyethanolamine (SOPE), and 1,2- dioleoyl-sn glycero-3-phophoethanolamine (transDOPE).

27. The lipid nanoparticle composition of claim 10, wherein said lipid not within the scope of Formula (1) comprises: (iv) sterols.

28. The lipid nanoparticle composition of claim 27, wherein said sterols is cholesterol or a derivative thereof.

29. The lipid nanoparticle composition of claim 28, further comprising at least one of: (i) tertiary amine (ionizable) lipids, (ii) non-tertiary amine lipids containing a hydrophobic tail attached to a hydrophilic head group; and (iii) lipids containing an uncharged or zwitterionic polymeric group attached to a hydrophobic tail.

30. The lipid nanoparticle composition of any one of claims 10-29, wherein at least one of said lipid not within the scope of Formula (1) is attached to a functional ligand selected from the group consisting of peptides (e.g., RGD, TAT, LyP-1, Ang), proteins (e.g., anti- CD3, anti-CD4, anti-CD8, anti-CD19, anti-CD47), carbohydrates (e.g., dextran, mannose, chitosan, galactose, hyaluronic acid), small molecules (e.g., phospho-L-serine (PS)), 1,2- dioleoyl-sn-glycero-3-phospho-L-serine (DOPS), and sphingomyelin (SM)).

31. The lipid nanoparticle composition of any one of claims 1-30, wherein a component of the lipid nanoparticle composition is chemically modified with an antibody (or Fab or Fc).

32. The lipid nanoparticle composition of any one of claims 1-31, wherein the therapeutic molecule comprises one or more of a nucleotide, gene editing system, protein, compound, and / or an antibody.

33. The lipid nanoparticle composition of any one of claims 1-31, wherein the therapeutic molecule comprises a nucleotide.

34. The lipid nanoparticle composition of claim 33, wherein the therapeutic molecule comprises one or more of DNA, RNA, ssDNA, dsDNA, ssRNA, dsRNA, and hybrids thereof.

35. The lipid nanoparticle composition of claim 33, wherein the therapeutic molecule comprises one or more of plasmid DNA or linearized DNA.

36. The lipid nanoparticle composition of claim 33, wherein the therapeutic molecule comprises one or more of messenger RNA (mRNA), small interfering RNA (siRNA), microRNA (miRNA), circular RNA (circRNA), and / or long-noncoding RNA (IncRNA).

37. The lipid nanoparticle composition of claim 33, wherein the therapeutic molecule comprises antisense oligonucleotide (ASO).

38. The lipid nanoparticle composition of claim 33, wherein the therapeutic molecule comprises Cas nuclease mRNA and guide RNA nucleic acid.

39. The lipid nanoparticle composition of claim 38, wherein the guide RNA is singleguide RNA (sgRNA).

40. The lipid nanoparticle composition of claim 33, wherein the nucleotide encodes fusion biological moieties comprising protective domains and functional domains.

41. The lipid nanoparticle composition of claim 40, wherein the functional domains are fused to the protective domains directly or via a linker consisting of amino acids.

42. The lipid nanoparticle composition of claim 40, wherein the protective domain comprises: a plurality of negatively charged amino acids (e.g., aspartic acid, glutamic acid, and derivatives thereof); a plurality of positively charged amino acids (e.g., lysine, histidine, arginine, and derivatives thereof); and a plurality of additional amino acids independently selected from the group consisting of proline, serine, threonine, asparagine, glutamine, glycine, and derivatives thereof, wherein the ratio of the number of positively charged amino acids to the number of negatively charged amino acids is from about 1:0.5 to about 1:2.

43. The lipid nanoparticle composition of claim 40, wherein the protective domains are selected from extended recombinant polypeptide (XTEN), proline-alanine-serine, and elastin-like polypeptides.

44. The lipid nanoparticle composition of claim 40, wherein the protective domain is selected from natural half-life extension domains (e.g., Fc fragment).

45. The lipid nanoparticle composition of any one of claims 1-44, wherein the therapeutic molecule comprises a vaccine against SARS-Cov-2.

46. The lipid nanoparticle composition of claim 45, wherein the vaccine is an mRNA vaccine.

47. The lipid nanoparticle composition of claim 46, wherein the mRNA vaccine corresponds to a spike protein, or portion thereof.

48. A method of treating a subject, the method comprising administering the lipid nanoparticle of any one of claims 1-47 to a subject.

49. The method of claim 48, wherein the lipid nanoparticle is administered by injection into the subject.

50. The method of any one of claims 48-49, wherein the lipid nanoparticle is delivered to cells of the subject.

51. The method of claim 48, wherein lipid nanoparticle is delivered by removing cells from the subject, administering the lipid nanoparticle to the removed cells, and then reintroducing the removed cells to the subject.

52. The method of claim 48, wherein the lipid nanoparticle is administered intravenously, intramuscularly, intradermally, subcutaneously, intranasally, or by inhalation.

53. The method of any one of claims 48-52, wherein the lipid nanoparticle is administered as part of protein replacement therapy, cancer immunotherapy, cancer vaccine therapy, infectious disease vaccines, gene editing, autoimmune disease treatment and cancer diagnosis.

54. The method of any one of claims 48-52, wherein the lipid nanoparticle is administered for gene therapy comprising CRISPR-Cas gene editing.

55. The method of any one of claims 48-52, wherein the lipid nanoparticle is administered for in vitro and in vivo production of extracellular vesicles.

56. The method of any one of claims 48-52, wherein the lipid nanoparticle is administered for vaccination against coronavirus (e.g., SARS-CoV-2).

57. The method of any one of claims 48-52, wherein the lipid nanoparticle is administered along with checkpoint inhibitor (e.g., anti- Programmed death-ligand 1 (anti- PD-L1) antibody, anti- cytotoxic T-lymphocyte-associated protein 4 (anti-CTLA4) to treat cancer.

58. The method of any one of claims 48-57, wherein the method results in a higher transfection efficiency compared to a lipid nanoparticle composition not containing the lipid composition of Formula (1) in claim 1.

Images