CN118434400A - Compositions and methods for delivering an active agent comprising a nucleic acid - Google Patents

Abstract

Disclosed herein are pH-sensitive nanoemulsions and methods of use thereof. These pH sensitive nanoemulsions can comprise lipid particles encapsulating an active agent. The lipid particle can comprise one or more ionizable lipids; one or more neutral lipids; one or more pegylated lipids; and optionally one or more fusion oils. In some embodiments, these compositions are capable of buffering at an acidic pH (e.g., a pH of less than 6.5, such as a pH of 4 to 6.5, or a pH of 5.0 to 6.5). By buffering at an acidic pH, the delivery efficiency of the composition can be improved compared to an otherwise identical composition buffered at a pH of 7 or higher.

Description

Compositions and methods for delivering active agents comprising nucleic acids

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of U.S. Provisional Application No. 63/270,719 filed on October 22, 2021, U.S. Provisional Application No. 63/270,724 filed on October 22, 2021, and U.S. Provisional Application No. 63/288,152 filed on December 10, 2021, each of which is hereby incorporated by reference in its entirety.

Background technique

Therapeutic nucleic acids include, for example, small interfering RNA (siRNA), microRNA (miRNA), antisense oligonucleotides, ribozymes, plasmids, mRNA, sgRNA and immunostimulatory nucleic acids. These nucleic acids work through a variety of mechanisms. In the case of siRNA or miRNA, these nucleic acids can downregulate the intracellular level of specific proteins through a process called RNA interference (RNAi). After siRNA or miRNA is introduced into the cytoplasm, these double-stranded RNA constructs can enter a protein complex called RISC. The sense strand of siRNA or miRNA is displaced from the RISC complex, providing a template in RISC that can recognize and bind to mRNA with a complementary sequence to the bound siRNA or miRNA. After binding to complementary mRNA, the RISC complex cuts the mRNA and releases the cut strand. RNAi can downregulate specific proteins by targeting the corresponding mRNA that specifically destroys the synthesis of encoded proteins.

The therapeutic applications of RNAi are very broad, as siRNA and miRNA constructs can be synthesized with any nucleotide sequence against a target protein. To date, siRNA constructs have demonstrated the ability to specifically downregulate target proteins in both in vitro and in vivo models. In addition, siRNA constructs are currently being evaluated in clinical studies.

Although progress has been made recently, this area still needs the improved lipid therapeutic nucleic acid composition applicable to general therapeutic purposes.These compositions will for example efficiently encapsulate nucleic acid, have high medicine: lipid ratio, protect the nucleic acid of encapsulation from degradation and removal in serum, are suitable for systemic delivery, and provide the intracellular delivery of the nucleic acid of encapsulation.In addition, these lipid-nucleic acid particles should have good tolerability and provide enough therapeutic indexes, so that the nucleic acid therapy patient with effective dose will not produce significant toxicity and/or risk to the patient.

Summary of the invention

Provided herein are pharmaceutical compositions comprising lipid particles encapsulating active agents. The lipid particles may include one or more ionizable lipids; one or more neutral lipids; and one or more pegylated lipids. These compositions may be buffered at an acidic pH (e.g., a pH less than 6.5, such as a pH of 4 to 6.5, or a pH of 5.0 to 6.5). By buffering at an acidic pH, the delivery efficiency of the composition may be improved compared to other identical compositions buffered at a pH of 7 or higher.

In some embodiments, one or more ionizable lipids are present in lipid particles in an amount of 20 mol% to 65 mol% (e.g., 30 mol% to 50 mol%) of the total components forming lipid particles. In some embodiments, one or more ionizable lipids may include a lipid head group containing a tertiary amine. In certain embodiments, one or more ionizable lipids may include N, N-dimethyl-2,3-dioleyloxypropylamine (DODMA), [(4-hydroxybutyl) azanediyl] di(hexane-6,1-diyl) bis(2-hexyldecanoate) (ALC-0315); 8-{(2-hydroxyethyl)[6-oxo-6-(undecyloxy) hexyl] amino}octanoic acid 9-heptadecyl ester (SM-102), DLin-MC3-DMA; DLin-KC2-DMA; or any combination thereof.

In some embodiments, one or more neutral lipids are present in lipid granules with an amount of 35 mol % to 80 mol % (30 mol % to 50 mol %) of the total components forming lipid granules. One or more neutral lipids can comprise any suitable neutral lipid. For example, in some embodiments, one or more neutral lipids can comprise dipalmitoylphosphatidylcholine (DPPC), dioleoylphosphatidylethanolamine (DOPE), palmitoyloleoylphosphatidylcholine (POPC), egg phosphatidylcholine (EPC), distearoylphosphatidylcholine (DSPC), cholesterol or its any combination.

In some embodiments, one or more PEGylated lipids are present in lipid granules with an amount greater than 0 % by mole to 5 % by mole of the total components forming lipid granules. One or more PEGylated lipids can include, for example, PEG-di (tetradecyl) acetamide, PEG-myristoyl diglyceride, PEG-diacylglycerol, PEG dialkoxypropyl, PEG-phospholipids, PEG-ceramide or any combination thereof.

The average diameter of the lipid particles may be less than 1 micron, such as 50nm to 750nm, 50nm to 250nm, 50nm to 200nm, 50nm to 150nm or 50nm to 100nm. The polydispersity index (PDI) of the lipid particles may be less than 0.4.

The active agent encapsulated in the lipid particles can comprise any suitable active agent, such as a small molecule therapeutic agent, a diagnostic agent, a peptide, a protein, an antibody or a nucleic acid. In certain embodiments, the active agent can comprise a nucleic acid, such as siRNA, mRNA or any combination thereof.

Also provided is a pharmaceutical composition comprising a lipid granule of an encapsulated active agent, the lipid granule comprising: 20 mol % to 65 mol % of one or more ionizable lipids; 35 mol % to 80 mol % of one or more neutral lipids; greater than 0 mol % to 5 mol % of one or more PEGylated lipids; and 5 mol % to 50 mol % of one or more fusion oils. Compared to other identical compositions lacking one or more fusion oils, by incorporating one or more fusion oils, the delivery efficiency of the composition can be improved.

Optionally, these compositions can be buffered at an acidic pH (e.g., a pH less than 6.5, such as a pH of 4 to 6.5, or a pH of 5.0 to 6.5). By buffering at an acidic pH, the delivery efficiency of the composition can be improved compared to an otherwise identical composition buffered at a pH of 7 or higher.

One or more fusion oils can form 10 mol % to 40 mol % of the total components of lipid granule and be present in lipid granule.In some embodiments, fusion oil can comprise the C12-C40 hydrocarbon containing less than 3 rings.In some cases, C12-C40 hydrocarbon can comprise alkyl or alkylene chain, such as optionally comprising the alkylene chain of at least one cis double bond.For example, fusion oil can comprise squalene, squalane, pristane, pristene, farnesene, farnesane, retinol, phytol, carotene, tocopherol, tocotrienol, phytomenadione, menaquinone (wherein valence allows its ester) or its combination.In certain embodiments, fusion oil can comprise squalene.

In some embodiments, one or more ionizable lipids are present in lipid particles in an amount of 30 mol% to 50 mol% of the total components forming lipid particles. Fusion oil and one or more ionizable lipids can be present in lipid particles in a molar ratio of 0.25:1 to 1:1. In some embodiments, one or more ionizable lipids can include lipid head groups containing tertiary amines. In certain embodiments, one or more ionizable lipids can include N, N-dimethyl-2,3-dioleyloxypropylamine (DODMA), [(4-hydroxybutyl) azanediyl] di(hexane-6,1-diyl) bis(2-hexyldecanoate) (ALC-0315); 8-{(2-hydroxyethyl)[6-oxo-6-(undecyloxy) hexyl] amino}octanoic acid 9-heptadecyl ester (SM-102), DLin-MC3-DMA; DLin-KC2-DMA; or any combination thereof.

In some embodiments, one or more neutral lipids are present in lipid granules with an amount of 35 mol % to 80 mol % (30 mol % to 50 mol %) of the total components forming lipid granules. One or more neutral lipids can comprise any suitable neutral lipid. For example, in some embodiments, one or more neutral lipids can comprise dipalmitoylphosphatidylcholine (DPPC), dioleoylphosphatidylethanolamine (DOPE), palmitoyloleoylphosphatidylcholine (POPC), egg phosphatidylcholine (EPC), distearoylphosphatidylcholine (DSPC), cholesterol or its any combination.

In some embodiments, one or more PEGylated lipids are present in lipid granules with an amount greater than 0 % by mole to 5 % by mole of the total component forming lipid granules. Fusion oil and one or more PEGylated lipids can be present in lipid granules with a mol ratio of 5:1 to 20:1. One or more PEGylated lipids can comprise for example PEG-di (tetradecyl) acetamide, PEG-myristoyl diglyceride, PEG-diacylglycerol, PEG dialkoxypropyl, PEG-phospholipid, PEG-ceramide or its any combination.

The average diameter of the lipid particles may be less than 1 micron, such as 50nm to 750nm, 50nm to 250nm, 50nm to 200nm, 50nm to 150nm or 50nm to 100nm. The polydispersity index (PDI) of the lipid particles may be less than 0.4.

The active agent encapsulated in the lipid particles can comprise any suitable active agent, such as a small molecule therapeutic agent, a diagnostic agent, a peptide, a protein, an antibody or a nucleic acid. In certain embodiments, the active agent can comprise a nucleic acid, such as siRNA, mRNA or any combination thereof.

The compositions described herein can be used to deliver one or more active agents to cells (e.g., in vivo, ex vivo, or in vitro). Therefore, provided herein is a method for delivering an active agent to a cell (e.g., in vivo, ex vivo, or in vitro), the method comprising contacting the cell with a composition described herein. Also provided is a method for delivering an active agent to a cell in vivo, the method comprising administering a composition described herein to a mammalian subject (e.g., a human). In some embodiments, administration may include systemic administration (e.g., intravenous injection or infusion).

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 schematically illustrates the structure and chemical components of R848-loaded squalene emulsion.

Figures 2A-2C illustrate the particle characteristics of R848 nanoemulsions. Figure 2A shows the particle size of R848 NE with lipid to R848 weight ratios of 20:1, 15:1, 10:1, 5:1 and 2:1. Figure 2B shows the SEC chromatogram of R848 NE using a Sepharose CL-4B gel column. The absorbance at 320 nm was measured in the presence of R848. Figure 2C shows the particle stability of empty NE and R848 NE stored at 4°C for up to 3 weeks.

Figure 3 shows 200X bright field images of RAW 264.7 cells treated for 12 hours with (A) complete medium only, (B) empty NE, (C) free R848, (D) SD-101, (E) R848 NE, and (F) R848 NE with SD-101. R848 was treated alone or in combination at 50 μM in emulsion. SD-101 was treated alone or in combination at 300 nM.

Figures 4A-4C show the concentrations of TNF-α (Figure 4A), IL-6 (Figure 4B), and IL-12p70 (Figure 4C) secreted by RAW 264.7 cells after treatment with complete medium only, empty NE, free R848, free SD-101, R848 NE, or R848 NE/SD-101 combination for 12 hours. R848 was treated alone or in combination at 50uM in nanoemulsion. SD-101 was treated alone or in combination at 300nM. One-way ANOVA: *p<0.05, **: p<0.01, ***p<0.001.

Figure 5 illustrates the treatment of R848 NE and SD-101 in a syngeneic C57BL/6N mouse model of murine colon adenocarcinoma (MC38). Figure A shows a timeline for MC38 inoculation and treatment regimen. 500,000 MC38 were inoculated subcutaneously in the right flank of the mice. Treatment began on day 8 after inoculation when the tumor became palpable. Treatment was given every 3 days for up to 4 times. After the fourth dose on day 10, the mice were euthanized. Figures B-E show tumor growth over time in individual mice treated with (B) saline, (C) R848 NE, (D) SD-101, or (E) R848 NE/SD-101 combination (n=5). Figure F shows images of MC38 tumor tissues collected on day 10 with measured (G) tumor sizes. Figure H is a graph of the average tumor growth of each treatment group over 10 days. Data are expressed as mean ± SEM (n=5). One-way ANOVA: *p<0.05, **p<0.01, ***p<0.001.

Figure 6A shows spleen tissues collected from mice on day 10. Figure 6B is a graph of measured spleen weights, normalized to individual body weight (n=5). One-way ANOVA *p<0.05, **: p<0.01, ***p<0.001.

Figures 7A-7C show the concentrations of TNF-α (Figure 7A), IL-6 (Figure 7B) and IL-12p70 (Figure 7C) in mouse serum. Serum samples were isolated from whole blood on day 10 (n=3). One-way ANOVA: *p<0.05, **: p<0.01, ***p<0.001.

Figures 8A-8B show the gene regulation of Akt1, Bcl2, Pdl1, calreticulin, Hmgb1, Cd3e, Cd4 and Cd8a in tumor tissues collected from C57BL/6N mice (Figure 8A), and the gene regulation of Akt1, Bcl2, Pdl1, Foxp3, Ifng in spleen tissues collected from mice in the same study (Figure 8B). Data are expressed as mean ± SEM (n = 3). One-way ANOVA: *p < 0.05, **: p < 0.01, ***p < 0.001.

FIG. 9 includes graphs showing PSNE particle size (left) and polydispersity index (PDI, right).

Figures 10A-10B show the results of gel retardation assays for detecting oligonucleotide encapsulation (Figure 10A) and size exclusion chromatography for detecting resiquimod encapsulation (Figure 10B). In Figure 10A, Ln: 1: SD-101 ODN, 2: anti-PDL1 LNA ASO, 3: empty PSNE LNP, 4: empty PSNE LNP with resiquimod, 5: PSNE/SD-101 ODN, 6: PSNE/anti-PDL1 LNA ASO, 7: PSNE/(SD-101 ODN+anti-PDL1 LNA ASO), 8: PSNE/(SD-101 ODN+resiquimod.

FIG. 11 shows tumor growth curves of the MC-38 subcutaneous mouse syngeneic model treated with various PSNE lipid nanoparticle constructs throughout the treatment period.

Figure 12 shows tumor weight analysis and tumor size comparison.

FIG. 13 shows spleen index analysis and spleen size comparison.

Figure 14 shows the results of flow cytometric analysis of various immune cells of interest. Gating strategy: CD8 ⁺ T lymphocytes: CD3 ⁺ CD8 ⁺ ; CD4 ⁺ T lymphocytes: CD3 ⁺ CD4 ⁺ ; Regulatory T lymphocytes: CD3 ⁺ CD4 ⁺ Foxp3 ⁺ ; MDSCs: CD11b ^- CD11c ⁺ Gr-1 ⁺ .

FIG. 15 shows the results of flow cytometric analysis of surface PD-L1 expression on immune cells (left: CD8 ⁺ T lymphocytes, CD3 ⁺ CD8 ⁺ ; right: macrophages, CD11b ⁺ F4/80 ⁺ ).

FIG. 16 shows the results of RT-qPCR analysis of PD-L1 mRNA levels in the spleen of mice.

FIG. 17 shows the results of cytokine analysis (IL-10, IL-12p70, TNFα, IFNγ) by enzyme-linked immunosorbent assay (ELISA).

Figures 18A and 18B show the amino acid (SEQ ID NO: 1) and nucleotide (SEQ ID NO: 2) sequences of human PD-L1, respectively. The signal peptide is indicated in italics in the amino acid sequence, and the coding region is indicated in bold in the nucleotide sequence.

FIG. 19 is a graph showing the tumor progression profile of mice treated with various cationic nanoemulsions incorporating TLR agonists.

Figure 20 shows the tumor progression profile of mice treated with various cationic nanoemulsions incorporating TLR agonists in each group. Panel A. Normal saline. Panel B. Squalene vehicle control. Panel C. Poly(I:C) cationic nanoemulsion. Panel D. CpG 2216 cationic nanoemulsion. Panel E. Imiquimod squalene nanoemulsion.

FIG. 21 shows J558 tumor challenge of mice immunized with different cancer vaccine constructs.

FIG. 22 shows the tumor progression profile of mice treated with SD-101 CpG ODN and/or 2′-OMe anti-PDL1 ASO in standard LNP.

FIG. 23 shows body weight profiles of mice treated with SD-101 CpG ODN and/or 2′-OMe anti-PDL1 ASO in standard LNPs.

FIG. 24 shows gene downregulation efficacy analysis of different chemical modifications of anti-murine PD-L1 ASOs using RT-qPCR.

FIG. 25 shows the profile of MC-38 tumor progression in mice treated with PSNE-based lipid nanoparticles (Experiment 1).

FIG. 26 shows the particle size and polydispersity index analysis of different PSNE LNP constructs encapsulating SD-101, anti-PDL1 LNA ASO, and resiquimod.

FIG. 27 shows a gel retardation assay evaluating the oligonucleotide encapsulation efficiency of different PSNE LNPs.

FIG. 28 shows the size exclusion chromatography elution chromatogram of resiquimod encapsulated PSNE on a CL-4B SEC column.

FIG. 29 shows tumor progression curves of individual mice in different PSNE lipid nanoparticle treatment groups.

FIG. 30 shows the profile of MC-38 tumor progression in mice treated with PSNE-based lipid nanoparticles (Experiment 2).

FIG31 shows tumor weight analysis and tumor images for each PSNE lipid nanoparticle treatment group.

FIG. 32 shows spleen weight/index analysis and tumor images for each PSNE lipid nanoparticle treatment group.

FIG. 33 shows analysis of spleen cell populations in each PSNE lipid nanoparticle treatment group by flow cytometry.

FIG. 34 shows analysis of surface PD-L1 expression of splenic cytotoxic T lymphocytes and macrophages by flow cytometry.

FIG. 35 shows the results of RT-qPCR analysis of the splenocyte Pdl1 mRNA level for each PSNE lipid nanoparticle treatment group.

FIG. 36 shows the analysis of spleen cytokine (Il10 and Il12-p40) mRNA levels by RT-qPCR for each PSNE lipid nanoparticle treatment group.

FIG. 37 shows the analysis of serum cytokine levels for each PSNE lipid nanoparticle treatment group by cytokine ELISA.

FIG. 38 shows modulation of murine Pdl1 mRNA using anti-PDL1 LNA ASOs on different next generation PSNE LNP constructs.

FIG. 39 shows the regulation of murine Pdl1 mRNA in Hepa1-6 and MC-38 cells using next generation PSNE encapsulated anti-PDL1 LNA ASOs and IFN-γ induction.

40A-40B show murine PD-L1 surface protein expression of Hepa1-6 ( FIG. 40A ) and MC-38 ( FIG. 40B ) cells induced by IFN-γ using next generation PSNE encapsulated anti-PDL1 LNA ASOs and flow cytometry.

Figure 41 shows surface protein expression of LPS-induced RAW264.7 cells using next-generation PSNE encapsulated SD-101/anti-PDL1 LNA ASO and flow cytometry analysis. The left panel shows CD86 M1 macrophage activation marker BV605. The right panel shows PD-L1, APC.

FIG. 42 shows MC-38 cytotoxicity analysis using RAW264.7 conditioned medium.

FIG. 43 shows an overview of MC-38 tumor progression in mice treated with next generation PSNE based lipid nanoparticles throughout/selected treatment periods.

Figure 44 shows individual mouse tumor progression curves in different next generation PSNE lipid nanoparticle treatment groups. Panel A, saline; Panel B, PSNE-Chol/M5-SD-101; Panel C, PSNE-Chol/M5-anti-PDL1 LNA ASO; Panel D, PSNE-Chol/M5 mixture; and Panel E, PSNE-Chol/M5 co-loading.

Figure 45 shows body weight analysis of tumor-bearing mice treated with next-generation PSNE-based lipid nanoparticles. Panel A, saline; Panel B, PSNE-Chol/M5-SD-101; Panel C, PSNE-Chol/M5-anti-PDL1 LNA ASO; Panel D, PSNE-Chol/M5 mixture; and Panel E, PSNE-Chol/M5 co-loading.

FIG. 46 shows the tumor weight (Panel A) and spleen index (Panel B) analysis of each next-generation PSNE lipid nanoparticle treatment group.

Figure 47 shows the analysis of spleen cell populations by flow cytometry for each next generation PSNE lipid nanoparticle treatment group. Panel A, CD4+ T lymphocytes; and Panel B, CD8+ T lymphocytes (cytotoxic T lymphocytes).

FIG48 shows analysis of spleen regulatory T lymphocyte populations by flow cytometry for each next generation PSNE lipid nanoparticle treatment group.

Figure 49 shows the analysis of spleen mRNA levels by RT-qPCR for each next generation PSNE lipid nanoparticle treatment group. Panel A, Pdl1; Panel B, Siglech, a marker for plasmacytoid dendritic cells; Panel C, Foxp3, a marker for regulatory T lymphocytes.

FIG. 50 shows analysis of tumor Pdl1 mRNA levels by RT-qPCR for each next-generation PSNE lipid nanoparticle treatment group.

Figure 51 shows the results of RT-qPCR analysis of splenocyte cytokine mRNA levels for each next-generation PSNE lipid nanoparticle treatment group. Panel A, TNF-α; Panel B, IFN-γ; Panel C, IL-10; Panel D, IL-6; and Panel E, TGF-β.

FIG. 52 shows the results of RT-qPCR analysis of liver Pdl1 mRNA levels of anti-PDL1 LNA ASOs delivered by next generation PSNE lipid nanoparticles.

Figures 53A-53C show particle characteristics of R848-IVM NE. Figure 53A shows the particle size of empty NE, R848 NE, IVM NE and R848-IVM NE. Figure 53B shows the SEC chromatogram of R848-IVM NE using Sepharose CL-4B column. The absorbance at 320nm and 245nm is used to measure the presence of R848 and IVM, respectively. Figure 53C shows the solubility of R848 and IVM in squalene NE and PBS determined by HPLC.

Figures 54A-54B show cell viability and IC ₅₀ determinations for R848 and IVM. Figure 54A shows the results obtained from free R848 and R848 NE used to treat MC38 cells for 72 hours followed by MTS assays. Figure 54B shows the results obtained from free IVM and IVM NE used to treat MC38 cells for 72 hours followed by MTS assays. Data are expressed as mean ± SD (n = 3).

Figures 55A-55B show gene regulation of calreticulin, Hmgbl and Lc3b in MC38 cells. Cells were treated with (Figure 55A) free R848 and R848 NE or (Figure 55B) free IVM and IVM NE. R848 and IVM were treated at 8 μM as free drug or in squalene-based NE, respectively. Data are expressed as mean ± SD (n = 3). One-way ANOVA: *p < 0.05, **: p < 0.01, ***p < 0.001.

Figure 56A-Figure 56D shows MC38 migration in response to treatment with R848 and IVM. Figure 56A shows cell morphology before and after treatment with 8 μM R848 NE, IVM NE and R848-IVM NE. Figure 56B shows the percentage of MC38 cells that migrated to the wound area assessed 24 hours after a scratch wound was created on confluent cells. Figure 56C shows that the dose-dependent inhibition of MC38 migration by R848-IVM NE was determined. Figure 56D shows the gene regulation of calreticulin, Hmgb1 and Lc3b in MC38 migration studies treated with R848 NE, IVM NE and R848-IVM NE (n=3). One-way ANOVA: *p<0.05, **p<0.01, ***p<0.001.

FIG. 57 shows treatment of a murine colon adenocarcinoma (MC38) syngeneic C57BL/6N mouse model with R848 NE, IVM NE, and R848-IVM NE. (A) Schematic illustration of tumor inoculation and treatment regimen. One million MC38 cells were inoculated subcutaneously in the right flank of mice. Treatment was initiated 7 days after tumor inoculation when the tumor size reached 100 mm ^3. Treatment was given every 3 days for a total of 3 times. After the third dose on day 9, mice were euthanized. (B-E) Tumor growth over time for individual mice treated with (B) saline, (C) R848 NE, (D) IVM NE, and (E) R848-IVM NE (n=5). (F) Images of MC38 tumor tissue were collected on day 10, and tumor weights were measured (G). (H) Average tumor growth over 9 days for each treatment group. Data are expressed as mean ± SEM (n=5). One-way ANOVA: *p<0.05, **p<0.01, ***p<0.001.

Figure 58A-Figure 58D shows gene regulation and immune cell populations in tumor and spleen tissue from treated mice. (Figure 58A) Calreticulin, Hmgb1, Lc3b, Cd3e, Cd4 and Cd8a mRNA expression in tumor tissue collected from C57BL/6N mice. (Figure 58B-Figure 58C) The percentage (Figure 58B) of CTL in spleen tissue and the ratio (Figure 58C) of CTL to Tregs are determined by flow cytometry. (Figure 58D) The protein expression of HMGB1 and Ki67 is determined by Western blotting in the same tumor tissue of (Figure 58A). Data are expressed as mean ± SD (n = 3). Single factor ANOVA: *p < 0.05, **p < 0.01, ***p < 0.001.

59A-59B show the results of a study demonstrating the efficacy of squalene as a fusing agent in pH-sensitive nanoemulsions.

Figure 60 is a graph showing down-regulation of Akt-1 gene by PSNE-siRNA compositions formulated at pH 7.4, 6.5 and 5.5. The y-axis represents the relative mRNA level after administration of PSNE-siRNA compositions. Akt-1 mRNA was determined by qRT-PCR and a negative control was included for reference.

Figures 61A-61E show IVIS in vivo bioluminescence imaging of mice treated with intramuscular injections of FFLuc mRNA (naked mRNA), FFLuc mRNA encapsulated in LNPs, FFLuc mRNA encapsulated in pH sensitive micelles (PSMs), and FFLuc mRNA encapsulated in PSNE (buffered at pH 6). As shown in Figures 3A-3E, the PSM and PSNE formulations were much more active than the LNP formulation, with the PSNE formulation showing the highest activity in terms of mRNA delivery.

Figure 62 shows the delivery provided by PSNE based on DLin-MC3-DMA and DODMA carrying luciferase mRNA given to mice by intramuscular injection. High-level gene expression of PSNE was observed in 50mM pH 6 histidine buffer, but high-level gene expression of PSNE was not observed in 50mM phosphate buffer at pH 7.4. This shows that the preparation for intramuscular delivery of mRNA has a very strong pH dependence. The same principle also applies to LNPs containing appropriate ionizable lipids.

Figure 63 is a graph quantifying the increase in bioluminescence observed after intramuscular injection of PSNE buffered in 50 mM pH 6 histidine buffer and 50 mM phosphate buffer at pH 7.4 based on DLin-MC3-DMA carrying luciferase mRNA. By lowering the pH from 7.4 to 6, a 30-50x increase in luminescence intensity was observed.

Figure 64 includes graphs showing the cellular uptake of fluorescently labeled PSNE in KB cells at different buffer pHs. The results showed that by flow cytometry analysis, the cellular uptake levels at pH 5 and pH 6 were much higher (increased mean fluorescence intensity) compared to pH 7. This illustrates the mechanism of pH-dependent cellular uptake/delivery of PSNE.

Figure 65 shows the design of a syringe assembly with an integrated size exclusion cartridge for online ethanol removal and buffer exchange.

Figures 66A-66C are photos showing a 4-syringe + 1 cartridge assembly for instant production of ethanol-free nanoparticles. The yield (nucleic acid recovery % compared to input) of ethanol-free nanoparticles produced using this device is greater than 60% (specifically 68.8%).

Detailed ways

definition

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs. Methods and materials for use in the present invention are described herein; other suitable methods and materials known in the art may also be used. The materials, methods, and examples are illustrative only and are not intended to be limiting. All publications, patent applications, patents, sequences, database entries, and other references mentioned herein are incorporated herein by reference in their entirety. In the event of a conflict, the present specification (including definitions) shall prevail.

"Aqueous solution" refers to a composition that comprises, in whole or in part, water.

"Organic lipid solution" refers to a composition that contains, in whole or in part, an organic solvent with lipids. In some embodiments, the organic lipid solution may contain an alkanol, most preferably ethanol. In certain embodiments, the compositions described herein may be free of organic solvents, such as ethanol.

"Lipids" refers to a group of organic compounds that are esters of fatty acids and are characterized by being insoluble in water but soluble in many organic solvents, such as fats, oils, waxes, phospholipids, glycolipids, and steroids.

"Amphipathic lipids" include lipids in which hydrophilicity originates from the presence of polar or charged groups, such as carbohydrates, phosphates, carboxyls, sulfates, aminos, sulfhydryls, nitros, hydroxyls and other similar groups, and hydrophobicity can be imparted by including polar groups, which include but are not limited to long-chain saturated and unsaturated aliphatic hydrocarbons and such groups substituted by one or more aromatics, alicyclics or heterocyclics. Examples include phospholipids, amino lipids and sphingolipids. Phospholipids include phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, phosphatidylinositol, phosphatidic acid, palmitoyloleoylphosphatidylcholine, lysophosphatidylcholine, lysophosphatidylethanolamine, dipalmitoylphosphatidylcholine, dioleoylphosphatidylcholine, distearoylphosphatidylcholine or dilinoleoylphosphatidylcholine. Amphipathic lipids may also lack phosphorus, such as sphingolipids, sphingolipid families, diacylglycerols and b-acyloxy acids.

An “anionic lipid” is any lipid that is negatively charged at physiological pH and includes phosphatidylglycerol, cardiolipin, diacylphosphatidylserine, diacylphosphatidic acid, N-dodecanoylphosphatidylethanolamine, N-succinylphosphatidylethanolamine, N-glutarylphosphatidylethanolamine, lysylphosphatidylglycerol, and other anionic modifying groups attached to neutral lipids.

"Cationic lipids" carry a net positive charge at a selective pH, such as physiological pH, and include N,N-dioleoyl-N,N-dimethylammonium chloride ("DODAC"); N-(2,3-dioleyl)propyl)-N,N,N-trimethylammonium chloride ("DOTMA"); N,N-distearyl-N,N-dimethylammonium bromide ("DDAB"); N-(2,3-dioleyl)propyl)-N,N,N-trimethylammonium chloride ("DOTAP"); and N-(l,2-dimyristyloxyprop-3-yl)-N,N-dimethyl-N-hydroxyethylammonium bromide ("DMRIE"). In addition, many commercial preparations of cationic lipids that can be used in the present invention are available. These include, for example and

Pharmaceutical composition

Provided herein are pharmaceutical compositions comprising lipid particles encapsulating active agents. The lipid particles may include one or more ionizable lipids; one or more neutral lipids; and one or more pegylated lipids. These compositions may be buffered at an acidic pH (e.g., a pH less than 6.5, such as a pH of 4 to 6.5, or a pH of 5.0 to 6.5). By buffering at an acidic pH, the delivery efficiency of the composition may be improved compared to other identical compositions buffered at a pH of 7 or higher.

Also provided is a pharmaceutical composition comprising a lipid particle encapsulating an active agent, the lipid particle comprising: 20 mol% to 65 mol% of one or more ionizable lipids; 35 mol% to 80 mol% of one or more neutral lipids; greater than 0 mol% to 5 mol% of one or more PEGylated lipids; and 5 mol% to 50 mol% of one or more fusion oils. Compared with other identical compositions lacking one or more fusion oils, by incorporating one or more fusion oils, the delivery efficiency of the composition can be improved. Optionally, these compositions can be buffered at an acidic pH (e.g., less than a pH of 6.5, such as a pH of 4 to 6.5, or a pH of 5.0 to 6.5). In some embodiments, these compositions can be buffered at a pH of 6.5 to less than 6.8, or 6.5 to less than 7. By buffering at an acidic pH, compared with other identical compositions buffered at a pH of 7 or higher, the delivery efficiency of the composition can be improved. In other embodiments, these compositions can be buffered at a pH of 7 to 7.4.

The average diameter of the lipid particles may be less than 1 micron, such as 50nm to 750nm, 50nm to 250nm, 50nm to 200nm, 50nm to 150nm or 50nm to 100nm. The polydispersity index (PDI) of the lipid particles may be less than 0.4.

The components of these compositions are described in more detail below.

Ionizable lipids

As described above, the compositions described herein may include one or more ionizable lipids."Ionizable lipids" are lipids that carry a pH-dependent charge. The one or more ionizable lipids in the compositions described herein may include ionizable cationic lipids that carry a positive charge or a neutral charge depending on pH.

Typically, in lipid-based formulations for nucleic acid delivery, cationic lipids or ionizable lipids are used to achieve electrostatic interactions with negatively charged cargoes. Cationic lipids are generally defined as lipids that carry a permanent positive charge, usually from a quaternary amine. Examples of cationic lipids include DOTAP, DOTMA, DDAB, and DODAC. In contrast, ionizable lipids include chemical moieties, such as tertiary amines, that are positively charged at acidic pH but uncharged at neutral to alkaline pH. Ionizable lipids can have pKa values within a biologically relevant range. However, the pKa value of this lipid is highly dependent on the method used to measure it, resulting in values for the same lipid differing by up to 3 units. This was documented in a recent article by Carrasco et al. Communications Biology, Vol. 4, Article No.: 956 (2021).

Examples of ionizable lipids are DODMA (N,N-dimethyl-2,3-dioleylpropylamine), DODAP, DLinDMA (1,2-dilinoleyl-3-dimethylaminopropane), DLinMC3DMA (dilinoleylmethyl-4-dimethylaminobutyrate), DLinKC2DMA (2-dilinoleyl-4-dimethylaminoethyl-[1,3]-dioxolane), ALC-0315 ([(4-hydroxybutyl)azanediyl]bis(hexane-6,1-diyl)bis(2-hexyldecanoate)), SM-102 (8-{(2-hydroxyethyl)[6-oxo-6-(undecanyloxy)hexyl]amino}octanoic acid 9-heptadecyl ester), Merck-32 (see, e.g., WO 2012/018754), Acuitas-5 (see, e.g., WO 2012/018754), 2015/199952), KL-10 (see, e.g., U.S. Patent Application Publication No. 2012/0295832), C12-200 (see, e.g., Love, K T et al., PNAS, 107:1864 (2009)), 3-(N-(N',N'-dimethylaminoethane)-carbamoyl cholesterol ("DC-Chol"), etc. Ionizable lipids also include U.S. Patent Nos. 8,158,601, 9,593,077, 9,365,610, 9,567,296, 9,580,711, and 9,670,152, International Publication Nos. WO 2012/018754, WO 2015/199952, WO No. 2019/191780 and those disclosed in U.S. Patent Application Publication Nos. 2012/0295832, 2017/0190661, and 2017/0114010, each of which is incorporated herein by reference in its entirety.

In some embodiments, the one or more ionizable lipids may comprise a lipid head group comprising a tertiary amine. In certain embodiments, the one or more ionizable lipids may comprise N,N-dimethyl-2,3-dioleyloxypropylamine (DODMA), [(4-hydroxybutyl) azanediyl] di(hexane-6,1-diyl) bis(2-hexyldecanoate) (ALC-0315); 8-{(2-hydroxyethyl)[6-oxo-6-(undecyloxy)hexyl] amino} octanoic acid 9-heptadecyl ester (SM-102), DLin-MC3-DMA; DLin-KC2-DMA; or any combination thereof.

In some embodiments, one or more ionizable lipids account for at least 20 mol% (e.g., at least 25 mol%, at least 30 mol%, at least 35 mol%, at least 40 mol%, at least 45 mol%, at least 50 mol%, at least 55 mol%, or at least 60 mol%) of the total components of the lipid particles. In some embodiments, one or more ionizable lipids account for 65 mol% or less (e.g., 60 mol% or less, 55 mol% or less, 50 mol% or less, 45 mol% or less, 40 mol% or less, 35 mol% or less, 30 mol% or less, or 25 mol% or less) of the total components of the lipid particles.

One or more ionizable lipids are present in lipid granules with the amount in the scope of above-mentioned any minimum value to above-mentioned any maximum value.For example, in some embodiments, one or more ionizable lipids are present in lipid granules with the amount of 20 % by mole to 65 % by mole (for example 30 % by mole to 50 % by mole) of the total component forming lipid granules.

Neutral lipids

As noted above, the compositions described herein may comprise one or more neutral lipids.

Examples of neutral lipids include phospholipids such as phosphatidylcholine, phosphatidylethanolamine, lysolecithin, lysolecithin, phosphatidylethanolamine, phosphatidylserine, phosphatidylinositol, sphingomyelin, egg sphingomyelin (ESM), cephalin, cardiolipin, phosphatidic acid, cerebrosides, dicetyl phosphate, distearoylphosphatidylcholine (DSPC), dioleoylphosphatidylcholine (DOPC), dipalmitoylphosphatidylcholine (DPPC), dioleoylphosphatidylglycerol (DOPG), dipalmitoylphosphatidylglycerol (DPPG), dioleoylphosphatidylethanolamine (DOPE), palmitoyloleoylphosphatidylcholine (POPC), palmitoyloleoylphosphatidylcholine (DPPC), and dioleoylphosphatidylglycerol (DOPG). Phosphatidylcholine, dilinoleylphosphatidylcholine and mixtures thereof can be used. Other diacylphosphatidylcholine and diacylphosphatidylethanolamine phospholipids can also be used. The acyl groups in these lipids are preferably acyl groups derived from fatty acids having C10-C24 carbon chains, such as lauroyl, myristoyl, palmitoyl, stearoyl or oleoyl.

Other examples of neutral lipids include sterols, such as cholesterol and its derivatives. Non-limiting examples of cholesterol derivatives include polar analogs, such as 5a-cholestanol, 5a-coprostanol, cholesteryl-(2'-hydroxy)-ethyl ether, cholesteryl-(4'-hydroxy)-butyl ether and 6-ketocholestanol; non-polar analogs, such as 5a-cholestanes, cholestenone, 5a-cholestanone, 5a-cholestanone and decanoic acid cholesterol ester; and mixtures thereof. In preferred embodiments, cholesterol derivatives are polar analogs, such as cholesteryl-(4'-hydroxy)-butyl ether. Other examples of neutral lipids include phosphorus-free lipids such as stearylamine, dodecylamine, hexadecylamine, acetyl palmitate, ricinoleylglycerol, cetyl stearate, isopropyl myristate, amphoteric acrylic acid polymers, triethanolamine-lauryl sulfate, alkyl-aryl sulfate polyethoxylated fatty acid amides, dioctadecyldimethylammonium bromide, ceramides, and sphingomyelin.

In some embodiments, the one or more neutral lipids can include dipalmitoylphosphatidylcholine (DPPC), dioleoylphosphatidylethanolamine (DOPE), palmitoyloleoylphosphatidylcholine (POPC), egg phosphatidylcholine (EPC), distearoylphosphatidylcholine (DSPC), cholesterol, or any combination thereof.

In some embodiments, one or more neutral lipids account for at least 35 mol % (e.g., at least 40 mol %, at least 45 mol %, at least 50 mol %, at least 55 mol %, at least 60 mol %, at least 65 mol %, at least 70 mol % or at least 75 mol %) of the total components forming lipid particles. In some embodiments, one or more neutral lipids account for 80 mol % or less (e.g., 75 mol % or less, 70 mol % or less, 65 mol % or less, 60 mol % or less, 55 mol % or less, 50 mol % or less, 45 mol % or less or 40 mol % or less) of the total components forming lipid particles.

One or more neutral lipids are present in lipid granules with the amount in the scope of above-mentioned any minimum to above-mentioned any maximum.For example, in some embodiments, one or more neutral lipids are present in lipid granules with the amount of 35 % by mole to 80 % by mole (30 % by mole to 50 % by mole) of the total component forming lipid granules.

PEGylated lipids

As described above, the compositions described herein may include one or more PEGylated lipids. One or more PEGylated lipids are useful because they can reduce or prevent aggregation of lipid particles.

PEG is a linear water-soluble polymer of ethylene PEG repeating units with two terminal hydroxyl groups. PEG is classified by its molecular weight; and includes the following: monomethoxypolyethylene glycol (MePEG-OH), monomethoxypolyethylene glycol-succinate (MePEG-S), monomethoxypolyethylene glycol-succinimidyl succinate (MePEG-S-NHS), monomethoxypolyethylene glycol-amine (MePEG-NEh), monomethoxypolyethylene glycol-trifluoroethanesulfonate (MePEG-TRES), monomethoxypolyethylene glycol-imidazolyl-carbonyl (MePEG-IM), and such compounds containing terminal hydroxyl groups instead of terminal methoxy groups (e.g., HO-PEG-S, HO-PEG-S-NHS, HO-PEG-NH2).

Examples of PEG-lipids include, but are not limited to, PEG coupled to a dialkoxypropyl group (PEG-DAA), PEG coupled to diacylglycerol (PEG-DAG), PEG coupled to a phospholipid such as phosphatidylethanolamine (PEG-PE), PEG conjugated to a glyceride forming a diol (e.g., 1,2-dimyristoyl-sn-glycerol, methoxy-PEG glycol (PEG-DMG)), PEG conjugated to ceramide, PEG conjugated to cholesterol, or derivatives thereof, and mixtures thereof. In some examples, one or more PEGylated lipids may include, for example, PEG-di(tetradecyl)acetamide, PEG-myristoyl diglycerol, PEG-diacylglycerol, PEG dialkoxypropyl, PEG-phospholipid, PEG-ceramide, or any combination thereof.

The average molecular weight of the PEG portion of the PEG-lipid conjugate described herein can be in the range of 550 Daltons to 10,000 Daltons. In some cases, the average molecular weight of the PEG portion is 750 Daltons to 5,000 Daltons (e.g., 1,000 Daltons to 5,000 Daltons, 1,500 Daltons to 3,000 Daltons, 750 Daltons to 3,000 Daltons, 750 Daltons to 2,000 Daltons). In some embodiments, the average molecular weight of the PEG portion is 2,000 Daltons or 750 Daltons.

In some cases, PEG can be optionally substituted by alkyl, alkoxy, acyl or aryl. PEG can be directly conjugated to lipid, or can be connected to lipid by linker moiety. Any linker moiety suitable for coupling PEG to lipid can be used, including, for example, a linker moiety without ester and a linker moiety containing ester. In one embodiment, the linker moiety is a linker moiety without ester. Suitable linker moieties without ester include, but are not limited to, amide (-C (O) NH-), amino (-NR-), carbonyl (-C (O)-), carbamate (-NHC (O) O-), urea (-NHC (O) NH-), disulfide (-S-S-), ether (-O-); succinyl (-(O) CCH2CH2C (O)-), succinyl (-NHC (O) CH2CH2C (O) NH-), ether, disulfide, and combinations thereof (such as joints containing carbamate linker moieties and amide linker moieties). In some embodiments, carbamate linkers are used to couple PEG to lipid. In other embodiments, ester-containing linker moieties can be used to couple PEG to lipids. Suitable ester-containing linker moieties include, for example, carbonate (—OC(O)O—), succinyl, phosphate (—O—(O)POH—O—), sulfonate, and combinations thereof.

The term "diacylglycerol" or "DAG" includes compounds having two fatty acyl chains ( ^R1 and ^R2 ), each independently having 2 to 30 carbons bonded to the 1-position and 2-position of glycerol via an ester bond. The acyl groups may be saturated or have varying degrees of unsaturation. Suitable acyl groups include, but are not limited to, lauroyl (C12), myristoyl ( _Ci4 ), palmitoyl (Ci6), stearoyl (Cis), and eicosanoyl (C20). In a preferred embodiment, ^R1 and ^R2 are the same, i.e., both ^R1 and ^R2 are myristoyl (i.e., dimyristoyl), and both ^R1 and ^R2 are stearoyl (i.e., distearoyl).

The term "dialkoxyalkyl" or "DAA" includes compounds having two alkyl chains (R and R') each independently having from 2 to 30 carbons. The alkyl groups may be saturated or have varying degrees of unsaturation.

Examples of PEG-DAA conjugates include PEG-didecyloxypropyl (C10), PEG-dilauryloxypropyl (C12), PEG-dimyristyloxypropyl (C14), PEG-dipalmityloxypropyl (C16), and PEG-distearyloxypropyl (C18). In some of these embodiments, the average molecular weight of PEG may be 750 or 2,000 Daltons. In certain embodiments, the terminal hydroxyl groups of PEG may be substituted with methyl groups.

In addition to the above, other hydrophilic polymers can be used to replace PEG. Examples of suitable polymers that can be used to replace PEG include but are not limited to polyvinylpyrrolidone, polymethyloxazoline, polyethyloxazoline, polyhydroxypropylmethacrylamide, polymethacrylamide and polydimethylacrylamide, polylactic acid, polyglycolic acid and derived cellulose such as hydroxymethylcellulose or hydroxyethylcellulose.

In some embodiments, one or more PEGylated lipids account for greater than 0 mol% (e.g., at least 0.5 mol%, at least 1 mol%, at least 1.5 mol%, at least 2 mol%, at least 2.5 mol%, at least 3 mol%, at least 3.5 mol%, at least 4 mol%, or at least 4.5 mol%) of the total components forming lipid particles. In some embodiments, one or more PEGylated lipids account for 5 mol% or less (e.g., 4.5 mol% or less, 4 mol% or less, 3.5 mol% or less, 3 mol% or less, 2.5 mol% or less, 2 mol% or less, 1.5 mol% or less, 1 mol% or less, or 0.5 mol% or less) of the total components forming lipid particles.

One or more PEGylated lipids are present in lipid granules with an amount in the scope of any minimum value to any maximum value mentioned above. For example, in some embodiments, one or more PEGylated lipids are present in lipid granules with an amount greater than 0 % by mole to 5 % by mole of the total component forming lipid granules.

Fusion Oil

In some embodiments, the compositions described herein may include one or more fusion oils.

In some embodiments, the fusion oil may include C12-C40 hydrocarbons (e.g., C12 hydrocarbons, C13 hydrocarbons, C14 hydrocarbons, C15 hydrocarbons, C16 hydrocarbons, C17 hydrocarbons, C18 hydrocarbons, C19 hydrocarbons, C20 hydrocarbons, C21 hydrocarbons, C22 hydrocarbons, C23 hydrocarbons, C24 hydrocarbons, C25 hydrocarbons, C26 hydrocarbons, C27 hydrocarbons, C28 hydrocarbons, C29 hydrocarbons, C30 hydrocarbons, C31 hydrocarbons, C32 hydrocarbons, C33 hydrocarbons, C34 hydrocarbons, C35 hydrocarbons, C36 hydrocarbons, C37 hydrocarbons, C38 hydrocarbons, C39 hydrocarbons, or C40 hydrocarbons). In some cases, the C12-C40 hydrocarbons may include an alkyl or alkylene chain. The alkylene chain may include one or more double bonds (e.g., one to five double bonds, or one to three double bonds). In some embodiments, the C12-C40 hydrocarbons may include an alkylene chain optionally including at least one cis double bond. In some embodiments, the fusion oil may contain fewer than three rings (eg, fewer than two rings, or no rings).

In some instances, the fusion oil may include squalene, squalane, pristane, pristene, farnesene, farnesane, retinol, phytol, carotene, tocopherol, tocotrienol, phytonadione, menaquinone (where valence allows esters thereof) or a combination thereof. In certain embodiments, the fusion oil may include squalene.

When present in compositions as described herein, in some embodiments, one or more fusion oils account for at least 10 mol% (e.g., at least 15 mol%, at least 20 mol%, at least 25 mol%, at least 30 mol%, or at least 35 mol%) of the total components forming lipid particles. When present in compositions as described herein, one or more fusion oils account for 40 mol% or less (e.g., 35 mol% or less, 30 mol% or less, 25 mol% or less, 20 mol% or less, or 15 mol% or less) of the total components forming lipid particles.

One or more fusion oils are present in lipid granules with the amount in the scope of above-mentioned any minimum value to above-mentioned any maximum value.For example, in some embodiments, one or more fusion oils can form the amount of 10 mol % to 40 mol % of the total component of lipid granules and are present in lipid granules.

When present in compositions described herein, in some embodiments, fusion oil and one or more PEGylated lipids can be present in lipid particles at a molar ratio of at least 5: 1 (e.g., at least 10: 1 or at least 15: 1). When present in compositions described herein, in some embodiments, fusion oil and one or more PEGylated lipids can be present in lipid particles at a molar ratio of 20: 1 or less (e.g., 15: 1 or less, or 10: 1 or less).

Fusion oil and one or more PEGylated lipids can be present in lipid particles in a molar ratio in the range of any minimum value to any maximum value. For example, in some embodiments, fusion oil and one or more PEGylated lipids can be present in lipid particles in a molar ratio of 5:1 to 20:1.

When present in compositions described herein, in some embodiments, fusion oil and one or more ionizable lipids can be present in lipid particles at a molar ratio of at least 0.25: 1 (e.g., at least 0.5: 1 or at least 0.75: 1). When present in compositions described herein, in some embodiments, fusion oil and one or more ionizable lipids can be present in lipid particles at a molar ratio of 1: 1 or lower (e.g., 0.75: 1 or lower, or 0.5: 1 or lower).

Fusion oil and one or more ionizable lipids can be present in lipid particles in a molar ratio in the range of any minimum value to any maximum value. For example, in some embodiments, fusion oil and one or more ionizable lipids can be present in lipid particles in a molar ratio of 0.25:1 to 1:1.

Active Agent

The active agent encapsulated in the lipid particle can comprise any suitable active agent, such as a small molecule therapeutic, a diagnostic agent, a peptide, a protein, an antibody, or a nucleic acid.

In certain embodiments, the active agent may comprise a nucleic acid.

Nucleic acids (NA, e.g., polynucleotides or oligonucleotides) encoding peptides can be used to produce antigenic peptides in vitro. NA can be, e.g., DNA, cDNA, PNA, CNA, RNA, single-stranded and/or double-stranded or natural or stable forms of polynucleotides, e.g., polynucleotides with a thiophosphate backbone, or a combination thereof, and it may or may not contain introns, as long as it encodes the peptide. In one embodiment, in vitro translation is used to produce the peptide. There are many exemplary systems that those skilled in the art can utilize.

In some embodiments, activating agent can include mRNA or expression vector capable of expressing polypeptide.Expression vectors for different cell types are well known in the art, and can be selected without excessive experiments.Usually, DNA is inserted into expression vector (such as plasmid) with appropriate direction and correct reading frame to express, if necessary, DNA can be connected with suitable transcription and translation regulation control nucleotide sequence identified by required host (such as bacterium), but such control can be obtained in expression vector usually.Then vector is introduced into host bacteria to clone using standard techniques (see, for example, Sambrook et al. (1989) MOLECULAR CLONING, A LABORATORY MANUAL, Cold Spring Harbor Laboratory, NY).

The term "nucleic acid encoding a polypeptide" encompasses NAs that contain only the coding sequence for the polypeptide as well as NAs that contain additional coding and/or non-coding sequences. NAs can be in the form of RNA or in the form of DNA. DNA includes cDNA, genomic DNA, and synthetic DNA; and can be double-stranded or single-stranded, and if single-stranded, can be a coding strand or a non-coding (antisense) strand.

NA may comprise a coding sequence for a peptide (antibody or antigen) fused in the same reading frame with, for example, a polynucleotide that facilitates expression and/or secretion of the polypeptide from a host cell (e.g., a leader sequence, which is used as a secretory sequence to control transport of the polypeptide from a cell). A polypeptide having a leader sequence is a pre-protein and may have a leader sequence that is cleaved by a host cell to form a mature form of the polypeptide.

An NA sequence encoding a polypeptide of interest is constructed by chemical synthesis using an oligonucleotide synthesizer. Such oligonucleotides can be designed based on the amino acid sequence of the desired polypeptide, and those codons that are favorable in the host cell producing the recombinant polypeptide of interest are selected. Standard methods can be applied to synthesize the isolated polynucleotide sequence encoding the isolated polypeptide of interest. Oligomers containing the NA sequence encoding a specific isolated polypeptide can be synthesized. For example, several small oligonucleotides encoding the parts of the desired polypeptide can be synthesized and then connected. A single oligonucleotide usually contains a 5' or 3' overhang for complementary assembly.

Once assembled (e.g., by synthesis, site-directed mutagenesis or another method), the polynucleotide sequence encoding the specific isolated polypeptide of interest is inserted into an expression vector and optionally operably linked to expression control sequences suitable for expressing the protein in the desired host. Correct assembly can be confirmed by nucleotide sequencing, restriction mapping and expression of the biologically active polypeptide in a suitable host. As is well known in the art, in order to obtain high expression levels of the transfected gene in the host, the gene can be operably linked to transcriptional and translational expression control sequences that are functional in the selected expression host.

Recombinant expression vectors can be used to amplify and express DNA encoding antibodies or antigenic peptides. Recombinant expression vectors are reproducible DNA constructs having synthetic or cDNA-derived DNA fragments that are operably linked to suitable transcription or translation regulatory elements derived from mammalian, microbial, viral or insect genes. Transcription units are generally comprised of a combination of one or more genetic elements that have a regulatory effect in gene expression, such as a transcription promoter or enhancer, a structure or coding sequence that is transcribed into mRNA and translated into protein, and appropriate transcription and translation initiation and termination sequences, as described in detail herein. Such regulatory elements may include an operator sequence that controls transcription. Typically, operably linked means continuous, and in the case of a secretory leader sequence, means continuous and in a reading frame. Structural elements intended for use in yeast expression systems include leader sequences that enable host cells to secrete translated proteins outside the cell. Alternatively, when a recombinant protein is expressed without a leader sequence or a transport sequence, it may include an N-terminal methionine residue. This residue may optionally be subsequently cut from the expressed recombinant protein to provide a final product.

"Ribonucleic acid" or "RNA" refers to a polymer containing at least two ribonucleotides. "Ribonucleotides" comprise the sugar ribose, a base, and a phosphate group. The nucleotides are linked together by the phosphate group. "Bases" include purines and pyrimidines, and also include the natural compounds adenine, thymine, guanine, cytosine, uracil, inosine, and natural analogs; as well as synthetic derivatives of purines and pyrimidines, including but not limited to modifications that place new reactive groups, such as, but not limited to, amines, alcohols, thiols, carboxylates, and alkyl halides.

The RNA may be in the form of oligonucleotide RNA, tRNA (transfer RNA), snRNA (small nuclear RNA), rRNA (ribosomal RNA), mRNA (messenger RNA), antisense RNA, siRNA (small interfering RNA), self-replicating RNA, ribozyme, chimeric sequence or derivatives of these groups.

The RNA may include (in addition to any 5' cap structure) one or more nucleotides having modified nucleobases, including m5C (5-methylcytidine), m5U (5-methyluridine), m6A (N6-methyladenosine), s2U (2-thiouridine), Um (2'-0-methyluridine), mlA (l-methyladenosine); m2A (2-methyladenosine); Am (2'-0-methyladenosine); ms2m6A (2-methylthio-N6-methyladenosine); i6A (N6-isopentenyladenosine); ms2i6A (2-methylthio-N6 isopentenyladenosine); io6A (N6-(cis-hydroxy)- isopentenyl)adenosine); ms2io6A (2-methylthio-N6-(cis-hydroxyisopentenyl)adenosine); g6A (N6-glycylaminoformyladenosine); t6A (N6-threonylaminoformyladenosine); ms2t6A (2-methylthio-N6-threonylaminoformyladenosine); m6t6A (N6-methyl-N6-threonylaminoformyladenosine); hn6A (N6-hydroxynorvalylaminoformyladenosine); ms2hn6A (2-methylthio-N6-hydroxynorvalylaminoformyladenosine); Ar(p) (2'-0-riboadenosine (phosphate)); I (inosine); ml l(1-methylinosine); m'lm(l,2'-0-dimethylinosine); m3C(3-methylcytidine); Cm(2T-0-methylcytidine); s2C(2-thiocytidine); ac4C(N4-acetylcytidine); f5C(5-formylcytidine); m5Cm(5,2-O-dimethylcytidine); ac4Cm(N4 acetyl 2TO methylcytidine); k2C(methylisopyrazole); mlG(l-methylguanosine; m2G(N2-methylguanosine); m7G(7-methylguanosine); Gm(2'-0-methylguanosine); m22G(N2,N2-dimethylguanosine); m2Gm (N2, 2'-0-dimethylguanosine); m22Gm (N2, N2, 2'-0-trimethylguanosine); Gr(p) (2'-0-ribosylguanosine (phosphate)); yW (wybutin); o2yW (peroxywybutin); OHyW (hydroxywybutin); OHyW* (unmodified hydroxywybutin); imG (wyother); mimG (methylguanosine); Q (quercetin); oQ (epoxyquercetin); galQ (galactosyl-quercetin); manQ (mannose-quercetin); preQo (7-cyano-7-deazaguanosine); preQi (7-aminomethyl-7 -deazaguanosine); G* (archauridine); D (dihydrouridine); m5Um (5,2'-0-dimethyluridine); s4U (4-thiouridine); m5s2U (5-methyl-2-thiouridine); s2Um (2-thio-2'-0-methyluridine); acp3U (3-(3-amino-3-carboxypropyl)uridine); ho5U (5-hydroxyuridine); mo5U (5-methoxyuridine); cmo5U (uridine 5-oxyacetic acid); mcmo5U (uridine 5-oxyacetic acid methyl ester); chm5U (5-(carboxyhydroxymethyl)uridine); mchm5U (5-(carboxyhydroxymethyl)uridine) methyl) uridine methyl ester); mcm5U (5-methoxycarbonylmethyl uridine); mcm5Um (S-methoxycarbonylmethyl-2-O-methyl uridine); mcm5s2U (5-methoxycarbonylmethyl-2-thiouridine); nm5s2U (5-aminomethyl-2-thiouridine); mnm5U (5-methylaminomethyl uridine); mnm5s2U (5-methylaminomethyl-2-thiouridine); mnm5se2U (5-methylaminomethyl-2-seleno-uridine); ncm5U (5-carbamoylmethyl uridine); ncm5Um (5-carbamoylmethyl-2'-0-methyl uridine); cmn m5U (5-carboxymethylaminomethyluridine); cnmm5Um (5-carboxymethylaminomethyl-2-L-0-methyluridine); cmnm5s2U (5-carboxymethylaminomethyl-2-thiouridine); m62A (N6,N6-dimethyladenosine); Tm (2'-0-methylinosine); m4C (N4-methylcytidine); m4Cm (N4,2-0-dimethylcytidine); hm5C (5-hydroxymethylcytidine); m3U (3-methyluridine); cm5U (5-carboxymethyluridine); m6Am (N6,T-0-dimethyladenosine); rn62Am (N6,N6,0 -2-trimethyladenosine); m2'7G (N2,7-dimethylguanosine); m2'2'7G (N2,N2,7-trimethylguanosine); m3Um (3,2T-0-dimethyluridine); m5D (5-methyldihydrouridine); f5Cm (5-formyl-2'-0-methylcytidine); mlGm (l,2'-0-dimethylguanosine); m'Am (1,2-0-dimethyladenosine) isomethyluridine); tm5s2U (S-taurinemethyl-2-thiouridine); imG-l4 (4-demethylguanosine); imG2 (isoguanosine); or ac6A (N6-acetyladenosine) uracil, 5-(C2-C6)-alkenyl uracil, 5-(C2-C6)-alkynyl uracil, 5-(hydroxymethyl) uracil, 5-chlorouracil, 5-fluorouracil, 5-bromouracil, 5-hydroxycytosine, 5-(C2-C6)-alkyl cytosine, 5-methyl cytosine, 5-(C2-C6)-alkenyl cytosine, 5-(C2-C6)-alkynyl uracil, guanine, 7-deaza-7-substituted guanine, 7-deaza-7-(C2-C6)alkynylguanine, 7-deaza-8-substituted guanine, 8-hydroxyguanine, 6-thioguanine, 8-oxoguanine, 2-aminopurine, 2-amino-6-chloropurine, 2,4-diaminopurine, 2,6-diaminopurine, 8-azapurine, substituted 7-deazapurine, 7-deaza-7-substituted purine, 7-deaza-8-substituted purine or abasic nucleotide.

The RNA can optionally comprise one or more UNA molecules, e.g., as disclosed in U.S. Pat. Nos. 8,314,227, 9,051,570, 9,303,260, 9,297,009, and 9,340,789, and U.S. Patent Publication No. 2016/0168567, which are incorporated herein by reference in their entireties.

The RNA or self-replicating RNA may include one or more modified pyrimidine nucleobases, such as pseudouridine and/or 5-methylcytosine residues. [0038] The RNA may include a 5' cap containing 7'-methylguanosine, and the first 1, 2 or 3 5' ribonucleotides may be methylated at the 2' position of the ribose. The RNA may include a 5' trinucleotide cap structure as described in U.S. Application No. 15/788,742 filed by Tanis et al. on October 19, 2017, which is incorporated herein by reference in its entirety.

While naturally occurring RNA has a phosphate backbone, the RNA described herein may contain other types of backbones and bases, including peptide nucleic acids, phosphothionates, phosphoramidates, phosphorothioate, and/or methylphosphonate linkages.

"Antisense" is a polynucleotide that interferes with the function of DNA and/or RNA. This may result in inhibition of expression.

"Gene" refers to a nucleic acid (e.g., DNA) sequence that contains the coding sequence necessary to produce a polypeptide or precursor. A polypeptide can be encoded by the full-length coding sequence or any portion of the coding sequence, as long as the desired activity or functional properties (e.g., enzymatic activity, ligand binding, signal transduction, etc.) of the full-length polypeptide or its fragment are retained.

Instructions

These compositions can be prepared as described herein or elsewhere, and can be administered by a variety of routes, depending on whether local or systemic treatment is needed and the area to be treated. Administration can be topical (including transdermal, epidermal, ocular and mucosal, including intranasal, vaginal and rectal delivery), pulmonary (e.g., by inhalation or insufflation of powder or aerosol, including by nebulizer; intratracheal or intranasal), oral or parenteral. Parenteral administration includes intravenous, intraarterial, subcutaneous, intraperitoneal intramuscular or injection or infusion; or intracranial (e.g., intrathecal or intraventricular administration). Parenteral administration can be in the form of a single bolus administration, or can be performed, for example, by a continuous infusion pump. In some embodiments, provided herein is a compound or a pharmaceutically acceptable salt thereof suitable for parenteral administration. In some embodiments, provided herein is a compound suitable for intravenous administration. In some embodiments, provided herein is a compound suitable for oral administration. In some embodiments, provided herein is a compound suitable for topical administration.

Pharmaceutical compositions and preparations for topical administration may include, but are not limited to, transdermal patches, ointments, lotions, creams, gels, drops, suppositories, sprays, liquids and powders. Conventional pharmaceutical carriers, aqueous, powdered or oily bases, thickeners, etc. may be necessary or desirable. In some embodiments, the pharmaceutical compositions provided herein are suitable for parenteral administration. In some embodiments, the pharmaceutical compositions provided herein are suitable for intravenous administration. In some embodiments, the pharmaceutical compositions provided herein are suitable for oral administration. In some embodiments, the pharmaceutical compositions provided herein are suitable for topical administration.

The compositions described herein can be used to deliver one or more active agents to cells (e.g., in vivo, ex vivo, or in vitro). Therefore, provided herein is a method for delivering an active agent to a cell (e.g., in vivo, ex vivo, or in vitro), the method comprising contacting the cell with a composition described herein. Also provided is a method for delivering an active agent to a cell in vivo, the method comprising administering a composition described herein to a mammalian subject (e.g., a human). In some embodiments, administration may include systemic administration (e.g., intravenous injection or infusion).

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. In view of the specification and practice disclosed herein, other embodiments will be apparent to those skilled in the art. The embodiments are further explained in the following examples. These examples do not limit the scope of the claims, but are merely used to illustrate certain embodiments. This specification and examples are considered to be exemplary only, and their true scope and spirit are indicated by the appended claims.

Example

The present invention will be described in more detail by specific examples. The following examples are provided for illustrative purposes and are not intended to limit the present invention in any way. Those skilled in the art will readily recognize various non-critical parameters that can be changed or modified to produce substantially the same result.

Example 1: Squalene-based nanoemulsions for therapeutic delivery of TLR agonists.

Overview

Toll-like receptor (TLR) agonists have shown promising anticancer activity. In this embodiment, squalene-based nanoemulsions are loaded with resimod, a TLR7/8 agonist for therapeutic delivery. When combined with the CpG-containing TLR9 agonist SD-101, strong antitumor activity was observed in the MC38 mouse colon cancer model. The treatment induced upregulation of PD-L1 in tumors, indicating potential therapeutic synergy with immune checkpoint inhibitors.

background

Toll-like receptors (TLRs) play a key role in immune responses by recognizing pathogen-associated molecular patterns (PAMPs), subsequently inducing cytokine production and activating adaptive immunity. TLRs are expressed on the plasma membrane (TLR1/2/4/5/6/10) or endosomes (TLR3/7/8/9) in antigen presenting cells (APCs) such as dendritic cells and macrophages. TLR activation leads to the induction of the MyD88/NF-κB pathway and activation of the naive T cell pool in the adaptive immune response [1]. Studies have shown that endosomal TLR agonists are effectively used as adjuvants in cancer vaccines due to their potent immunostimulatory activity [2]. It has been shown that endosomal TLR agonists activate plasmacytoid dendritic cells (pDCs) and cytotoxic T lymphocytes (CTLs), thereby enhancing T cell-mediated immunity. The FDA has approved three drugs with TLR agonist activity for cancer treatment, including BCG (TLR2 & 4 agonist mixture), monophosphoryl lipid A (TLR2/4 agonist mixture), and imiquimod (TLR7 agonist) [2]. Overall, the clinical efficacy of TLR agonists is mixed [3-4]. Intratumoral injection of CpG TLR9 agonists has recently been studied. However, this mode of administration is difficult in clinical practice for most solid tumors.

The use of TLR7/8 and TLR9 combinations can achieve powerful innate and adaptive immune system activation and promising anti-tumor efficacy [5-7]. Synergistic cytokine release and antibody production were observed using a Schistosoma japonicum DNA vaccine containing a combination of TLR 7/8 and TLR9 agonists [6]. Another study showed that TLR7/8/9 combination therapy had significant tumor suppression and synergistic IFN-γ secretion effects [5-8]. However, these results on dual TLR activation lack an effective platform for delivering TLR agonists.

Resiquimod (R848) is a TLR7/8 agonist that has shown antitumor activity in murine tumor models [9-12]. However, due to its limited solubility, the clinical application of R848 requires an injectable formulation. Oil-in-water nanoemulsions (NE) are effective delivery systems for hydrophobic drugs [13-16]. NEs consist of a surfactant-stabilized oil core, which can serve as an effective reservoir for poorly water-soluble drugs [17]. In addition, squalene-based NEs have been shown to be an effective vaccine adjuvant through adaptive immune activation [18]. Squalene-based NE vaccine adjuvants MF59 and AddaVax have vaccinated more than 100 million people in more than 30 countries against seasonal and pandemic influenza. It is also noted that the squalene core in NE (which was originally derived from shark liver oil) is reported to enhance immune responses and antitumor efficacy [19].

In this study, we developed a squalene-based NE using 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) and polysorbate 80 (Tween 80) as surfactants to encapsulate R848 (Figure 1). Here, the squalene core was used not only to enhance the immune response but also to solubilize R848, a poorly water-soluble molecule. We evaluated the encapsulation efficiency of R848 in squalene-based NE and optimized the drug loading by adjusting the lipid to drug ratio. The R848-loaded squalene-based NE also showed long-term stability for up to 1 month at 4°C.

In addition to R848, we also incorporated SD-101 as another component that promotes immune activation. SD-101 is a second-generation TLR9 agonist containing cytidine phosphate guanosine (CpG) dinucleotides. This C-class oligonucleotide stimulates TLR9, which is mainly expressed on pDCs, and enhances innate and adaptive immune responses [20]. In addition, in clinical trials, SD-101 has been shown to have good anti-tumor efficacy in combination with immune checkpoint inhibitors or radiotherapy [20-22]. Our study demonstrated the synergistic anti-tumor activity of NE loaded with R848 and SD-101 through adaptive immune stimulation. In mouse models, the combination treatment of NE loaded with R848 and SD-101 also showed upregulation of Pdl1 mRNA levels, suggesting a therapeutic strategy based on combined TLR activation and PDL1 targeting.

Materials and methods

Materials. Squalene was purchased from Sigma-Aldrich (St. Louis, MO, USA). 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC) was purchased from Avanti Polar Lipids (Birmingham, AL). Resiquimod (R848) was purchased from MedChemExpress (Monmouth Junction, NJ), and SD-101 was synthesized by Alpha DNA (Montreal, Quebec, Canada). TNF-α, IL-6, and IL-12p70 mouse uncoated ELISA kits and high capacity cDNA reverse transcription kits were purchased from Invitrogen (Waltham, MA). SsoAdvanced ^™ Universal Green Supermix was purchased from Bio-Rad Laboratories (Hercules, CA). Pre-designed primers for real-time PCR of mouse Akt1, Bcl2, Hif1a, Pdl1 (Cd274), calreticulin, Hmgb1, and Actb were purchased from Sigma-Aldrich (St. Louis, MO). Primers for mouse Cd3e, CD4, CD8a, Foxp3, and Ifng were designed and synthesized by ThermoFisher Scientific (Waltham, MA). Polysorbate 80 (Tween 80) and all other chemicals and buffers described above were purchased from FisherScientific (Hampton, NH).

R848-NE formulation and characterization. Squalene-based NEs were prepared by manual rapid injection of the oil-lipid mixture into phosphate buffered saline (PBS). Squalene, DOPC, and Tween 80 were prepared in ethanol at a molar ratio of 1/1/1. R848 was then added to the lipid-ethanol solution, maintaining a lipid to R848 ratio of 10:1 (w/w). The final total lipid concentration of the nanoemulsion was 8 mg/mL, and the final R848 concentration was 0.8 mg/mL. Based on particle size and R848 solubility, R848-loaded NEs (R848 NEs) with lipid to R848 weight ratios of 20:1, 15:1, 10:1, 5:1, and 2:1 were developed to optimize the R848 loading in squalene-based nanoemulsions. Particle size was measured by dynamic light scattering (DLS) using a NICOMP NANO ZLS Z3000 (Entegris, Billerica, MA). A similar procedure was used to generate empty nanoemulsion (empty NE) without the addition of R848. Empty NE and R848 NE were stored at 4°C prior to characterization and long-term stability.

Sepharose CL-4B size exclusion chromatography was performed to examine the encapsulation efficiency of R848 in squalene nanoemulsion. The R848 concentration was determined by UV-visible spectroscopy at 320 nm using a NanoDrop 2000 spectrophotometer [23]. The loading efficiency of resiquimod was determined by the following formula:

Cell culture. RAW 264.7 murine macrophage cell line and MC38 murine colon cancer cell line were gifts from Dr. Peixuan Guo and Dr. Christopher Coss, Ohio State University College of Pharmacy, respectively. RAW 264.7 and MC38 were grown in DMEM supplemented with 10% FBS and 1x antibiotic-antimycotic and incubated at 37°C in a humidified atmosphere containing 5% _CO2 .

In vitro macrophage stimulation imaging. RAW 264.7 cells were seeded at a density of 1.5 x 10 ⁵ cells/well in 24-well plates 24 hours before treatment. Cells were treated with empty NE, R848, SD-101, R848 NE, or R848 NE/SD-101 combination for 24 hours. R848 was treated alone or in combination with SD-101 at 50 μM in nanoemulsion. SD-101 was treated alone or in combination at 300 nM. After incubation for 24 hours, morphological changes of RAW 264.7 cells were observed under 200X bright field by Nikon Eclipse Ti-S microscope (Nikon, Tokyo, Japan).

In vitro cytokine induction assessment by enzyme-linked immunosorbent assay (ELISA). RAW 264.7 cells were seeded at a density of 3 x 10 ⁵ cells/well in 6-well plates 24 hours before treatment. Cells were treated with empty NE, R848, SD-101, R848 NE, or R848 NE/SD-101 combination for 24 hours. R848 was treated alone or in combination at 50 μM in nanoemulsion. SD-101 was treated alone or in combination at 300 nM. Supernatants were collected and stored at -80 °C and then subjected to cytokine quantification by ELISA. Supernatants were pre-diluted 6-fold with PBS and TNF-α, IL-6, and IL-12p70 concentrations were measured by mouse uncoated ELISA kits according to the manufacturer's protocol.

In vivo antitumor efficacy. The MC38 murine colorectal syngeneic model was generated by subcutaneously inoculating C57BL/6N mice (available from Charles River Labortories) in the right flank with 0.5 x 10 ⁶ cells per mouse. Treatment was initiated once tumors reached approximately 100 mm ³ . Mice (n=5) were treated intraperitoneally with saline, 4 mg/kg R848 NE, 2 mg/kg SD-101, or a combination of R848 NE/SD-101 (4 mg/kg R848 NE and 2 mg/kg SD-101). All treatment solutions were prepared in PBS. Mice were dosed every 3 days for a total of 4 times. Tumor growth and body weight were monitored, and tumor volume was calculated according to the following formula:

All animal studies were reviewed and approved by the Ohio State University Institutional Laboratory Animal Care and Use Committee (IACUC). On day 10, 6 hours after the fourth dose, all mice were euthanized to allow serum cytokine concentrations to peak. Whole blood was collected by cardiac puncture. Tumor and spleen tissues were collected and weighed for comparison. Spleen weights were normalized to individual body weights for comparison between treatment groups. Tissues and serum were stored at -80°C prior to in vivo cytokine and gene regulation studies. Tumor growth inhibition (%TGI) on day 10 was determined by the following formula:

Wherein _T10 represents the mean tumor volume of the treatment group on day 10, _T0 represents the mean tumor volume of the treatment group on day 0, _C10 represents the mean tumor volume of the control group on day 10, and _C0 represents the mean tumor volume of the control group on day 0. %TGI>50% was considered significant.

In vivo cytokine measurements. Mouse serum was collected by placing whole blood at room temperature for 30 minutes and randomly centrifuged at 2000 x g for 20 minutes. Samples were collected and stored at -80°C before cytokine quantification. Murine TNF-α, IL-6, and IL12p70 cytokine concentrations were determined by ELISA according to the manufacturer's protocol.

In vivo gene regulation by real-time qPCR. Tumor and spleen tissues were homogenized in TRI reagent using probe sonication, and total RNA was extracted according to the manufacturer's protocol. cDNA was prepared using a high-capacity cDNA reverse transcription kit according to the manufacturer's protocol. Real-time PCR was performed on the QuantStudio 7Flex Real-Time PCR System for the target genes Akt1, Bcl2, and Pdl1 in spleen tissue samples and Akt1, Bcl2, Hif1a, Pdl1, calreticulin, and Hmgb1 in tumor tissue samples. All genes were normalized to Actb as a housekeeping gene. The relative amounts of RNA levels were calculated and compared according to the 2 ^-ΔΔCt method [24,25].

Statistical Analysis. All studies were performed in triplicate. Unless otherwise stated, data are presented as mean ± standard deviation. Statistical analysis will be performed using Microsoft Excel. One-way ANOVA will be used to determine the differences in means between two or more treatment groups. A p value of 0.05 was selected as the critical value for statistical significance.

Results and discussion

Particle Characterization. As a TLR7/8 agonist, resiquimod has been shown to have stronger cancer immunotherapy efficacy compared with other imidazoquinoline analogs [26]. However, tolerance induction and adverse reactions limit its development as a strong candidate for marketed drugs. Some studies have used polymer-based nanoparticles (such as polylactic acid (PLA) or β-cyclodextrin) to carry R848 to overcome these limitations [27-29]. However, the disadvantages of polymer nanoparticles include toxic degradation and self-aggregation of polymer materials [30]. In our study, we used squalene-based oil-in-water nanoemulsion as a carrier of R848 (Figure 1). Neutral charged components (squalene, DOPC, and Tween 80) can eliminate the cytotoxicity caused by cationic lipid nanoparticles and polymer nanoparticles [31]. The size of R848 NE is about 50-100 nm (Figure 2A). There was no significant change in particle size between the lipid to R848 weight ratio of 20:1 and 2:1, indicating that the addition of R848 did not affect the structural stability of the nanoemulsion. However, R848 precipitation due to insufficient oil phase was observed in R848 NE samples with a lipid to R848 weight ratio of 2:1 after storage at 4°C overnight and a lipid to R848 weight ratio of 5:1 after storage at 4°C for 1 week (data not shown). Therefore, R848 NE with a lipid to R848 weight ratio of 10:1 was selected for further study to maximize the loading amount of R848. Size exclusion chromatography using a Sepharose CL-4B gel column (Figure 2B) showed that there was 35.9% ± 0.53% R848 in the NE encapsulated fraction, which is much higher than another liposomal formulation of R848, whose encapsulation efficiency was only 7% [32]. The results show that water-in-oil nanoemulsions are superior to liposomes in encapsulating poorly water-soluble drugs. We further demonstrated that both empty NE and R848 NE exhibited high colloidal stability after storage at 4 °C for 3 weeks (Figure 2C). However, there was an unexpected particle size shrinkage between empty NE and R848 NE, where the median particle size of R848 NE was approximately 50 nm smaller than that of empty NE (R848 NE approximately 100 nm and empty NE approximately 150 nm). The reduction in particle size after R848 loading may be due to the hydrophilic interaction between R848 (Figure 1) and the aqueous phase, which reduces the hydrophobic interaction between the emulsion particles and the aqueous phase.

In vitro macrophage stimulation. The morphological changes of macrophages upon activation can be observed under a microscope, which can be further used to study the factors that regulate pro-inflammatory (M1) and anti-inflammatory (M2) activation [33,34]. TLR7, 8, and 9 are mainly expressed on pDCs and macrophages [26,35-39], which further polarizes naive macrophages to M1 activation [40,41]. We used RAW264.7 mouse macrophages to detect immune stimulation produced by R848 NE treatment and the addition of SD-101 treatment. Our results showed that untreated RAW 264.7 generally appeared round, while empty NE-treated RAW 264.7 appeared round but formed filopodia (Figure 3, panels A and B). Partially activated macrophages were shown in R848-treated and SD-101-treated RAW 264.7, which partially expanded and formed lamellipodia (Figure 3, panels C and D). Finally, RAW 264.7 treated with R848 NE and the combination of R848 NE/SD-101 showed fully activated macrophages with accelerated spreading and formation of lamellipodia (Fig. 3, Panels E and F).

TNF-α is a proinflammatory cytokine that is indispensable for the generation of early immune responses. We found that all R848 and SD-101 treatments produced significant levels of TNF-α induction compared to the untreated group. Empty NE also induced moderate levels of TNF-α due to the immunostimulatory properties of squalene production (Figure 4A). This result is consistent with previous studies on the induction of TNFa by TLR7, 8, and 9 activation [42,43]. However, squalene-triggered TNF-α production was not significant compared to R848 and R848 NE treatments. In addition, treatment with SD-101 or the R848 NE/SD-101 combination showed slightly higher levels of TNF-α secretion compared to R848 or R848 NE treatment (Figure 4A), indicating that activation of TLR9 by SD-101 favors TNF-α production compared to activation of TLR7/8 by R848.

Both TLR7/8 and TLR9 stimulate IL-6 production, which functions as both a pro-inflammatory and anti-inflammatory cytokine [44-47]. Here, we report that R848 NE treatment promoted IL-6 production compared to R848 treatment, whereas empty NE did not show significant IL-6 production (Figure 4B). We also report that our R848 NE and SD-101 (acting as TLR7/8 and TLR9 agonists, respectively) synergistically increased IL-6 production compared to either treatment alone (Figure 4B).

Finally, IL-12p70 production is biased towards Th1 activation, leading to a cellular immune response [48,49]. There was moderate IL-12p70 production in both R848 NE and SD-101 treatments alone or in combination (Figure 4C). We observed that squalene-based NE could enhance R848-stimulated IL-12p70 production. In addition, the total IL-12p70 levels triggered by SD-101 and the R848 NE/SD-101 combination were slightly higher than those triggered by R848 or R848 NE alone. In contrast to previous reports that dual TLR coactivation synergizes IL-12p70 production [8,50-52], we found that TLR7/8 and TLR9 coactivation did not synergistically increase IL-12p70 production (Figure 4C).

In vivo antitumor efficacy. Studies have shown that TLR7 and TLR9 agonists have the highest immunogenicity when administered intranasally or orally compared with other endosomal TLR agonists [53]. In addition, intratumoral combination therapy with TLR7/8 and TLR9 agonists has also been shown to induce the highest tumor-specific immunity compared with each agent alone [5]. However, intraperitoneal administration is simple, rapid, and minimally stressful for both animal studies and metastatic cancer patients in practice. In this study, we used 4 mg/kg R848 and 2 mg/kg SD-101 alone or in combination, and showed synergistic antitumor efficacy by intraperitoneal administration. Compared with saline control (Figure 5, Panel B), moderate antitumor efficacy of R848 NE and SD-101 monotherapy was observed (Figure 5, Panels C, D, and F), which is consistent with previous studies of R848 and SD-101 [54, 55].

Although single treatment with R848 NE or SD-101 showed significant tumor growth inhibition (50.72% ± 16.83% and 65.65% ± 12.88%), R848 NE/SD-101 combination treatment achieved approximately 84.62% ± 28.05% total tumor growth inhibition at the end of the study (Table 1), with a median tumor growth inhibition of 98.05%, indicating a synergistic antitumor efficacy produced by the R848 NE/SD-101 combination treatment. The lack of significant differences in body weights indicated mild systemic toxicity in mice treated with R848 NE and SD-101 alone or in combination. Among individual mice treated with the R848 NE/SD-101 combination, one mouse was observed to have an initial tumor size of more than 200 mm ³ and a slightly higher tumor growth rate compared to other individuals in the same group (Figure 5, Panel E). This suggests that additional treatments should be combined with the R848 NE/SD-101 combination strategy to eliminate large solid tumors via intraperitoneal administration.

Table 1. Tumor growth inhibition (TGI %) at day 10 by R848 NE or SD-101 alone or in combination.

Splenomegaly indicates T cell activation and natural killer cell (NK cell) expansion [56]. In addition to the significant tumor suppression in the R848 NE/SD-101 combination treatment group, we also observed significant splenomegaly in the R848 NE/SD-101 combination treatment group compared with the R848 NE or SD-101 alone treatment groups and saline controls (Figures 6A and 6B). The significant splenomegaly demonstrates the synergistic antitumor immune activation achieved by the R848 NE/SD-101 combination treatment.

Cytokine production in vivo. The presence of TNF-α indicates strong proinflammatory cytokine release, which further enhances tumor cell apoptosis [57]. In mouse sera, we observed a significant increase in TNF-α levels in mice treated with the R848 NE/SD-101 combination (Figure 7A). This again demonstrates the synergistic antitumor immune activation and significant splenomegaly and tumor inhibition produced by the R848 NE/SD-101 combination treatment. Compared with saline controls, no significant changes in IL-6 and IL-12p70 levels were observed in mice treated with R848 NE and SD-101 alone or in combination (Figure 7C), but TLR7 activation-biased IL-12p70 production was observed in mice treated with R848 NE compared to mice treated with SD-101 or R848 NE/SD-101.

In vivo gene regulation. The antitumor immunity generated by the R848 NE/SD-101 combination can be attributed to cell-mediated cytotoxicity, such as CTL or NK cell activation consistent with the immunogenic cell death (ICD) process [58,59]. In tumor tissues of mice treated with the R848 NE/SD-101 combination, Cd8a mRNA was significantly upregulated, while calreticulin and Cd3e were elevated (Figure 8A), indicating that systemic treatment with R848 NE and SD-101 synergized antitumor immunity through CTL and NK cell activation. Different immune regulation was observed in mice treated with R848 NE or SD-101. In spleens treated with SD-101 or the R848 NE/SD-101 combination, the transcription factor Foxp3, known as a regulatory T cell marker, was downregulated, suggesting that fewer regulatory T cells may be generated after TLR9 activation, thereby inhibiting the antitumor immunity of CTL and NK cells. Compared with mice treated with SD-101, the expression of Cd3e and Cd4 mRNA in tumors of mice treated with R848 NE was significantly reduced (Figure 8A), indicating that the amount of CD4 T cells infiltrating into the tumor microenvironment (TME) was lower in mice treated with R848 NE and the R848 NE/SD-101 combination. However, the low expression of Cd3e and Cd4 mRNA did not interfere with the high expression of Cd8a mRNA in the TME (Figure 8A), which may indicate that the cytotoxicity of CTL and NK cells directed by R848 NE or SD-101 is independent of T helper cells in the TME. Treatment with R848 NE also successfully induced the expression of Ifng in the spleen of mice, while treatment with SD-101 suppressed the expression of Ifng in the spleen of mice (Figure 8B). SD-101-directed TLR9 activation may increase TNF-α production (Figure 7A), which enhances the cytotoxicity of CTL and NK cells in the TME. Tumor cell lysis guided by elevated TNF-α will further trigger tumor antigen presentation and more rapidly promote immune evasion by CTLs and NK cells in the TME [60,61].

Upregulation of calreticulin mRNA was observed in tumors of mice treated with R848 NE and SD-101 alone or in combination (Figure 8A), where increased exposure of calreticulin would lead to ICD of tumor antigens engulfed by APCs [62]. HMGB1 is known to negatively regulate cell proliferation, where the release of HMGB1 induces ICD [63,64]. In mice treated with SD-101 or R848 NE/SD-101, Hmgb1 mRNA was significantly downregulated (Figure 8A), suggesting that R848 NE/SD-101 treatment produces a TME that is unfavorable for cell proliferation. AKT-1, a member of the AKT family, has been shown to play an important role in cell survival and is associated with tumorigenesis [65,66]. In addition, the presence of AKT-1 has also been shown to be consistent with BCL-2 expression, the function of which is to inhibit apoptosis and promote tumor proliferation [67-69]. ICD induced by R848 NE and SD-101 alone or in combination also showed mild downregulation of Akt1 and Bcl2 mRNA ( Figure 8A ), also indicating a negative effect on the proliferation of the TME.

TLR7/8 and TLR9 activation in cancer cells, lymphocytes, and pDCs is also associated with upregulation of PD-L1, an immune checkpoint for tumor immune escape [70-73]. Mice treated with the R848 NE/SD-101 combination had the highest levels of Pdl1 mRNA in both spleen and tumor tissues compared with mice treated with either single therapy or saline control (Figure 8A and Figure 8B). Although the concept of immune escape established by the binding between PD-L1 on cancer cells and PD-1 on T cells is generally accepted [74], more studies now show that cancer cells can express both PD-1 and PD-L1, and PD-L1 is also found on certain immune cells such as macrophages and dendritic cells [75-77]. Considering the role of PD-L1 on antigen-presenting cells that may inhibit T cell function [77], we believe that R848 NE/SD-101 combination treatment by intraperitoneal injection may further demonstrate higher antitumor efficacy by eliminating large tumors or metastatic tumors in combination with anti-PD-L1 therapeutic agents.

in conclusion

Our study demonstrated a squalene-based nanoemulsion formulation loaded with R848 that exhibited improved encapsulation efficiency compared to reported liposomal formulations. We also showed that R848 NE was highly stable during long-term storage at 4°C. R848 NE and SD-101 had strong immune activation effects, respectively, as evidenced by increased TNF-α levels in vitro. Bright-field images and cytokine levels in vitro and in vivo further illustrated the synergistic immune activation upon combined treatment of R848 NE and SD-101. Finally, we demonstrated that R848 NE/SD-101 combination treatment greatly enhanced the in vivo antitumor efficacy by greatly inhibiting tumor growth by more than 80%. Our work suggests that combined intraperitoneal administration of R848 NE and SD-101 may be a powerful approach for CTL- and NK cell-dependent systemic antitumor therapy. Furthermore, when R848 NE/SD-101 treatment was combined with anti-PD-L1 therapy, the synergistic PD-L1 upregulation by the R848 NE/SD-101 combination in vivo suggested a greater anti-tumor potential.

Example 2: pH sensitive lipid nanoemulsion (PSNE) formulation as an effective delivery system for the delivery of CpG oligonucleotides and therapeutic oligonucleotides for anti-cancer treatment.

Overview

Cancer vaccines aim to trigger specific and durable immune responses against tumor antigens. Well-designed vaccines should trigger activation of dendritic cells (DCs) by defined tumor antigens and include adjuvants that stimulate expansion of naive T cells into effector T cells. It has been shown that the combination of toll-like receptor (TLR) agonists with tumor-associated antigens (TAAs) triggers DC maturation, cytotoxic T cell (CTL) activation, and tumor regression. However, TLR activation induces PD-L1 expression by monocytes and dendritic cells, even cancer cells, thereby protecting them from CTL attack. Elevated PD-L1 on cancer cells or monocytes may lead to CTL exhaustion and dysfunction. Here, we hypothesized that co-delivery of TLR agonists and anti-PD-L1 (LNA ASO/siRNA) results in higher levels of dendritic cell activation and maturation, thereby inducing potent T cell activation and anti-tumor activity. In addition, by co-delivering TLR agonists and anti-PDL1 ASOs, the overall immunosuppressive environment may be tilted toward an immune-permissive condition, which is associated with a reduced population of regulatory T cells and repolarization of macrophages. A novel pH-sensitive lipid nanoemulsion formulation or PSNE was developed to co-deliver CpG oligonucleotide-based TLR agonists and anti-PDL1 LNA ASOs in a single particle. The particle size of PSNE lipid nanoparticles is approximately 85-95nm, which is very suitable for delivery into the tumor microenvironment (TME) through the enhanced permeation and retention (EPR) effect. In the MC-38 syngeneic mouse tumor model, single-agent oligonucleotide-based TLR agonists (C-class CpG) or anti-PDL1 LNA ASOs in PSNE can lead to tumor growth inhibition (TGI) of about 45%, while co-encapsulation of CpG and ASO into a single particle leads to a higher TGI of about 73%. Flow cytometric analysis showed that the frequency of suppressive regulatory T lymphocyte (Treg) population in the spleen was reduced in the single-agent group and further reduced in the combination group, indicating a reversal of the immunosuppressive environment. In addition, PD-L1 expression of splenic macrophages, T lymphocytes, and whole splenocytes was evaluated by flow cytometry and RT-qPCR. The results showed that PD-L1 expression was downregulated in macrophages in all treatment groups, but not in splenic T lymphocytes. In addition, the PD-L1 mRNA level in the group treated with anti-PDL1 ASO was reduced by about 60% in both the single-agent and combination groups, indicating that PSNE lipid nanoparticles were taken up by phagocytes and overcame PD-L1 upregulation through TLR activation.

In conclusion, we demonstrated that co-encapsulation of CpG-oligonucleotide-based TLR agonists and anti-PDL1 ASOs in PSNE formulations enhanced the delivery of both agents to tumors and spleen and was more effective in activating the anti-tumor immune system.

Materials and methods

Squalene was purchased from Sigma-Aldrich (St. Louis, MO). DOPC and DOPE were purchased from Avanti Polar Lipids (Alabaster, AL). DODMA and DMG-PEG ₂₀₀₀ were purchased from NOF America (White Plains, NY). Unless otherwise stated, any chemicals or buffers were purchased from Fisher Scientific (Hampton, NH).

DOPC, DOPE, squalene, DODMA, and DMG-PEG2000 were dissolved in ethanol at a molar ratio of 15:28:10:45:2 as a mixture. Next, the lipid ethanol solution was quickly injected into an acidic buffer to form an empty pH-sensitive nanoemulsion (PSNE) with a lipid concentration of 8 mg/mL. At the same time, nucleic acid cargoes including SD-101 and anti-PDL1 locked nucleic acid (LNA) antisense oligonucleotides (ASOs) were dissolved in nuclease-free water at 0.4 mg/mL. Before mixing, the empty PSNE and nucleic acid cargo were heated to 60°C. Then, the cargo was added dropwise to the slowly vortexed empty PSNE at a weight ratio of 1 to 20 until nucleic acid-encapsulated PSNE was formed. The product was incubated at 37°C for 10 minutes and stored at 4°C before use. The particle size and zeta potential (ζ) of nucleic acid-loaded PSNE were analyzed by dynamic light scattering on a NICOMP Z3000 Nano DLS/ZLS system (Entegris, Billerica, MA).

Animal Model. C57BL/6 mice were purchased from Charles River Laboratory. Animals were housed in a temperature-controlled room with a 12-h light/12-h dark cycle and fed a normal chow diet. All animal studies were reviewed and approved by the Institutional Animal Care and Use Committee of The Ohio State University. Both male and female mice were used in the experiments.

The MC-38 murine colorectal cancer cell line was a gift from Dr. Christopher Coss, College of Pharmacy, Ohio State University. The MC-38 syngeneic subcutaneous murine colorectal cancer model was established by injecting 1 million MC-38 cells suspended in PBS subcutaneously into the right flank of each mouse. Tumor size and body weight were monitored daily using digital calipers and laboratory scales, respectively.

Treatment was started when the mean tumor size reached 80 mm ^3. Mice were randomly divided into 6 groups of 5 mice each. Treatments were given every three days for a total of 5 doses, including saline control, SD-101 oligonucleotide, PSNE with SD-101, PSNE with anti-PDL1 LNA, PSNE with SD-101 and anti-PDL1 LNA, and PSNE with resiquimod and SD-101. Mice were sacrificed 6 hours after the last dose, and organ and whole blood samples were collected for further analysis.

Serum samples were obtained from whole blood samples by incubating the blood at room temperature for 30 minutes followed by centrifugation at 2,000 x g for 20 minutes. Samples were stored at -80°C prior to analysis. Cytokine concentrations were measured by uncoated ELISA kits purchased from Invitrogen according to the manufacturer's protocol.

Single cell splenocyte suspension was harvested from spleen by the following procedure. The spleen was gently screened through a 70um nylon mesh (Thermo Scientific) using a sterile 5-mL syringe piston and washed twice in a six-well plate with cold RPMI medium. The crude splenocyte suspension was then centrifuged at 500x g for 5 minutes at 4°C to obtain a cell pellet. The cell pellet was resuspended in 1xRBC lysis buffer to lyse red blood cells at room temperature for 5 minutes, and 10 times the volume of PBS was added to terminate the reaction. After centrifugation at 500xg for 5 minutes, the cell pellet was then resuspended in FACS staining buffer to form a single cell splenocyte suspension.

Antibodies were purchased from Biolegend (San Diego, CA) and surface cell staining was performed according to the manufacturer's protocol. Intracellular staining was performed using the Biolegend True-NuclearTM transcription factor buffer set according to the manufacturer's protocol. Stained cells were analyzed on a BD LSR Foretessa flow cytometer at the Ohio State University Comprehensive Cancer Center's Flow Cytometry Shared Resource.

Splenocyte messenger RNA for gene expression analysis was extracted from spleen cell suspension using TRI reagent according to the manufacturer's protocol. cDNA was prepared using a high-capacity cDNA reverse transcription kit (Invitrogen). Real-time PCR was performed on a QuantStudio7Flex Real-Time PCR system. The relative amounts of RNA levels were calculated and compared according to the 2 ^-ΔΔCt method.

Results and discussion

PSNE encapsulating oligonucleotides and small molecules with high colloidal stability were successfully formulated. Particle sizes of about 80-100 nm were obtained for empty PSNE and PSNE loaded with SD-101 CpG ODN or anti-PDL1 LNA ASO. The polydispersity index (PdI) was about 0.15-0.25, indicating that a narrow particle size distribution was obtained in all samples (Figure 9). Empty PSNE had an average zeta potential of +8.06 mV in 20 mM Pb at pH 4, while encapsulated PSNE had an average zeta potential close to neutral in 10 mM phosphate buffer at pH 7.

Next, the encapsulation of ODN and hydrophobic small molecule cargo in PSNE was analyzed by gel retardation assay and size exclusion chromatography, respectively. The gel images showed that when resiquimod was added to the formulation, ODN was highly encapsulated in PSNE, and a small amount of SD-101 was adsorbed onto the PSNE surface. By calculating the ratio of the two peaks on the chromatogram obtained by OD327 using a NanoDrop spectrophotometer, the encapsulation efficiency of resiquimod in PSNE was about 53.8% (Figure 10A-Figure 10B).

In the MC-38 mouse syngeneic colon cancer model, the combination of SD-101 CpG ODN and anti-PDL1 LNA ASO in PSNE showed good anti-tumor effects, with an average tumor growth inhibition rate (TGI) of about 73%. Both groups of mice treated with a single agent (SD-101 ODN or anti-PDL1 LNA ASO) in PSNE showed a TGI of about 46%, while the average TGI of mice treated with PSNE encapsulated with SD-101 and resimod was 57% (Figure 11). Tumor weights were recorded and represent the actual tumor size after treatment. The results showed a similar trend to tumor volume, as the PSNE combination group showed significant tumor growth inhibition compared to saline controls using a one-way ANOVA (Figure 12).

The spleen index (spleen weight expressed as a percentage of body weight, Figure 13) results showed that PSNE had similar immune system activation effects as the SD-101 oligonucleotide. In addition, the combination of dual TLR agonists further activated the immune system (by expanding the immune cell population). In addition, the results also showed that PSNE encapsulated with anti-PDL1 LNA ASO did not have an immune system activation effect as a monotherapy. Tumor growth inhibition by PSNE/anti-PDL1 LNA ASO may be the result of PD-L1 downregulation at the tumor site, whether from LNA delivery to cancer cells or TME immune cells.

In flow cytometry analysis (Figure 14), the frequency of T lymphocyte populations (whether CD4+ T helper cells, CD8+ cytotoxic T cells or regulatory T cells) in all PNSE-based treatment groups was significantly reduced compared with the saline control group. The results can be interpreted as infiltration of T cells into the TME after activation. IHC staining of T cell populations can be performed to confirm this hypothesis. There was no significant change in MDSC frequency after all treatments in the MC-38 tumor model.

The surface PD-L1 marker of lymphocytes and bone marrow cells was detected by flow cytometry to study the PD-L1 protein expression of immune cells after treatment (Figure 15). The results showed that the surface PD-L1 protein level of macrophages was downregulated in the group treated with anti-PDL1 LNA ASO, because macrophages have the ability to phagocytose PSNE, and anti-PDL1 LNA can induce mRNA downregulation. Nevertheless, the PD-L1 protein expression of cytotoxic T lymphocytes was not significantly downregulated after treatment with anti-PDL1 LNA, indicating that the transfection rate of T cells was relatively low. As we expected, the PD-L1 protein expression level of the TLR (SD-101ODN and Resimod) treatment group was upregulated due to the NFκB-MyD88 activation pathway. This also shows that anti-PDL1 LNA can overcome the PD-L1 upregulation caused by TLR agonists.

By performing splenocyte mRNA analysis, we evaluated the cytokine secretion profile and overall PD-L1 levels after treatment with TLR agonists and/or anti-PDL1 LNA. The results demonstrated the excellent activity of LNA in downregulating PD-L1 mRNA in splenocytes (Figure 16).

Based on the cytokine ELISA results (Figure 17), it is clear that mice treated with PSNE/anti-PDL1 LNA were not in an inflammatory state (basal levels of proinflammatory cytokines TNFα and IFNγ); therefore, tumor growth inhibition was associated with the delivery of anti-PDL1 LNA ASO to the tumor. In addition, the results also showed that TLR agonists are able to drive the immune system into a proinflammatory phase by increasing IL-12 levels.

in conclusion

In conclusion, this study shows that PSNE is an effective delivery vehicle for anti-PD-L1 LNA ASOs for anti-tumor therapy and is able to co-deliver ASOs with resiquimod. This shows that anti-PD-L1 synergizes with TLR activation and, therefore, the combination of anti-PD-L1 oligonucleotide (also siRNA) therapy with TLR agonists is highly synergistic. It was further shown that PSNE-delivered resiquimod was very effective and synergistic with the CpG ODN SD-101, providing a new approach to delivering TLR combinations for therapeutic purposes. The triple combination of dual TLR agonists and anti-PD-L1 produced the best overall therapeutic response.

Example 3: Nanoemulsion-based lipid nanoparticles as a delivery platform for Toll-like receptor agonist-based anticancer immunotherapy

Traditional cancer therapies, including surgery, radiotherapy, and chemotherapy, have been practiced for decades to eliminate tumor tissue, but these treatments have significant side effects. To reduce side effects, next-generation drugs including targeted therapies and immunotherapies entered the clinic after the first approval of the monoclonal antibody rituximab in 1997. Compared with targeted therapies that target cancer cells themselves, immunotherapies have opened up new battlefields by targeting the immune system. Well-known immunotherapies, including immune checkpoint blockade (ICB), chimeric antigen receptor (CAR) T cell therapy, oncolytic viruses, and cancer vaccines, have attracted increasing attention in treating cancer patients.

Cancer vaccines have established a promising strategy to trigger specific and durable immune responses against tumor antigens. Well-designed cancer vaccines should trigger the activation of dendritic cells (DCs) through defined tumor antigens and appropriate adjuvants, thereby stimulating the expansion of the naive T cell pool into effector T cells. Many research groups have shown that combining toll-like receptor (TLR) agonists with tumor-associated antigens (TAAs) can initiate DC maturation, cytotoxic T cell (CTL) activation, and tumor shrinkage.

Tumor progression can be described as three stages by immune surveillance: the immune system clears cancer cells, cancer balance, and tumor escape. Tumor escape evolves by genetic changes in tumor antigens that avoid immune recognition and tumor-extrinsic mechanisms with active immunosuppression. Low expression levels of major histocompatibility complex (MHC) class I on the surface of cancer cells reduce the recognition and removal of CD8 ⁺ T cells. Tumors also produce immunosuppressive cytokines, such as transforming growth factor β (TGF-β) or soluble Fas ligand, to mediate the response of regulatory T cells (T _reg ) to suppress anti-tumor effector T cells. In order to overcome the immunosuppressive barrier established by TME or reduced functional CTL, immunomodulators are introduced into the clinic. Immune checkpoint blockade (ICB) agents can help restore CTL function by inhibiting negative regulatory signals on the surface of T cells. Typical targets are cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) or programmed cell death 1 (PD-1). Blocking CTLA-4 or PD-1 with monoclonal antibodies shows enhancement of effector T cell activation and proliferation.

An alternative pathway to upregulate the immune system is to activate T cell function by stimulating dendritic cells. Dendritic cells are the most potent antigen presenting cells (APCs) in the innate immune system and can present exogenous antigens to CD4 ⁺ T cells and CD8 ⁺ T cells via MHC class II and class I, respectively. Maturation and activation of DC cells are initiated by the uptake and recognition of antigens by pattern recognition receptors (PRRs) such as toll-like receptors (TLRs). Mature DCs migrate to the draining lymph nodes with high expression of MHC and antigen presentation capacity. In the draining lymph nodes, DCs encounter naive T cells and stimulate T cell activation. T cell activation requires three signals: interaction of antigen-associated MHC with the T cell receptor (TCR), recognition of co-stimulatory ligands (CD80, CD86, CD40), and proinflammatory cytokines. All of these signals promote the differentiation of CD4 ⁺ naive T cells into T _h1 cells and enhance the cytotoxic response of CTLs. Therefore, the basic principle behind DC vaccines requires incubation of DCs with a tumor antigen/stimulatory agent mixture to generate mature DCs and promote T cell activation after injection into the patient. However, the objective response rate (ORR) of cancer patients to DC vaccines rarely exceeds 15%.

To enhance DC maturation, antigen uptake, and T cell activation, TLR agonists have been introduced as stand-alone antitumor agents or as adjuvants in combination with TAAs. The TLR signaling cascade induces the production of type 1 interferons and inflammatory mediators, which promote further activity of T lymphocytes. Previously published studies have shown that injection of free CpG ODN or CpG-ODN nanoparticles loaded with tumor antigens showed potent antitumor responses in vivo. In addition to CpG ODN, the synthetic TLR3 agonist poly(I:C) has also been used in multiple studies to show the promotion of DC maturation and the induction of T cell-related antitumor activity.

It is well known that CpG oligonucleotides have the ability to activate TLR9, produce type 1 interferon, and trigger downstream processes of innate and adaptive immune activation. Many studies have used CpG ODN of different subclasses as vaccine adjuvants to stimulate dendritic cell maturation and initiate T cell ( _Th1 or _Th2 )-mediated immunity. In one study, researchers showed that the application of CpG ODN2006 (class B CpG ODN) can increase the infiltration of T cells and activated DCs in the draining lymph nodes, indicating that CpG ODN 2006 has the ability to stimulate DC activation as a vaccine adjuvant. In addition, the administration of cell lysate as an antigen source can not only promote the DC maturation process, but also promote CTL production. In addition, they demonstrated that the application of CpG ODN 2006/lysate combination can significantly prolong the survival of tumor-bearing mice compared with CpG ODN 2006 monotherapy, indicating the supportive role of CpG ODN in vaccine construction and play the role of adjuvant.

CpG ODN is a ligand for TLR9 that is mainly expressed in human plasmacytoid dendritic cells that specialize in secreting type 1 interferon. However, due to the oligonucleotide structure, different subclasses of CpG ODN have different abilities to stimulate type 1 interferon production. Class A CpG ODN includes poly-G motifs at the 5' and 3' ends and a self-complementary palindromic sequence containing one or more CpG motifs. Class A CpG ODN is a strong IFN-α inducer. On the other hand, class B CpG ODN has a complete thiophosphate backbone but no higher-order structure, and has a weaker ability to induce IFN-α production, but a stronger ability to stimulate B cell TLR9. It is reported that class A CpG is more active in supporting natural killer cell (NK cell) and CTL function and granzyme-B content in CTL, which is mainly mediated by type 1 interferon. In addition, IFN-α is a cytokine required for the activation of NK cells and partial activation and proliferation of memory CD8 ⁺ T cells. Interferon-γ (IFN-γ) secreted by activated NK cells also binds to IL-12 secreted by activated DCs to trigger _Th1 cell-mediated immunity. Therefore, the use of class A CpG ODNs rather than class B CpG ODNs should elicit a large number of innate and adaptive immune as well as proinflammatory _Th1 and CTL responses.

In this embodiment, we developed a series of highly fused nanoemulsion-based lipid nanoparticle constructs, co-encapsulated negatively charged ODN with different immune system enhancing factors (such as anti-PDL1 antisense oligonucleotides or tumor antigen peptides) at high loading, and used TLR agonist compounds (CpG ODN, poly (I: C) or resiquimod) as vaccine adjuvants to react with different DC subsets to enhance antigen presentation. We also evaluated TLR agonist combinations to demonstrate their synergistic effects relative to single agonists. Co-encapsulated TLR agonists produce higher proinflammatory cytokine secretion and stronger T cell activation. These strategies are likely to overcome the current obstacles to the low stimulation of the immune system in vivo due to insufficient delivery to APCs.

Materials and methods

Materials. Squalene, polyinosinic acid:polycytidylic acid (poly I:C), monophosphoryl lipid A (MPLA), and D-α-tocopheryl polyethylene glycol 1000 succinate (TPGS) were purchased from Sigma-Aldrich (St. Louis, MO, USA). DSPC, DOPC, DOPE, and cholesterol were purchased from Avanti Polar Lipids (Alabaster, AL, USA). DODMA and DMG-mPEG ₂₀₀₀ were purchased from NOFAmerica (White Plains, NY, USA). DLin-MC3-DMA was purchased from DC Chemical (Shanghai, China). Resiquimod (R848), imiquimod (R837), and amphotericin B were purchased from MedChemExpress (Monmouth Junction, NJ, USA). Tocopheryl succinate (TS) and didodecyldimethylammonium bromide (DDAB) were purchased from TCI America (Tokyo, JP). SD-101 CpG oligodeoxynucleotide (SD-101), CpG 2216 oligodeoxynucleotide, and 2'-O methyl modified mouse anti-PDL1 antisense oligodeoxynucleotide gapmer (2'-OMe anti-PDL1 ASO) were synthesized by Alpha DNA (Montreal, CA). Locked nucleic acid modified mouse anti-PDL1 antisense oligodeoxynucleotide gapmer (anti-PDL1 LNA ASO) was synthesized by Integrated DNA Technologies (IDT, Coralville, IA, USA). Any chemicals or buffers stated otherwise were purchased from Fisher Scientific (Hampton, NH, USA).

Preparation of cationic nanoemulsions incorporating TLR agonists. For CpG 2216 and poly (I: C) cationic nanoemulsions, squalene, DDAB, and TPGS were mixed into the lipid-ethanol mixture at a molar ratio of 50:35:15. Next, the lipid-ethanol solution was rapidly injected into a 20 mM HEPES buffer at pH 7.4 to form an empty cationic nanoemulsion with a lipid concentration of 25.0 mg/mL. To produce cationic nanoemulsions encapsulating oligonucleotides, 2.0 mg/mL of CpG 2216 or poly (I: C) oligonucleotide solution (in DEPC water) was added dropwise to the slowly vortexed empty lipid nanoemulsion at a weight ratio of 1 to 12.5 until a cationic nanoemulsion encapsulating nucleic acids was formed. For imiquimod nanoemulsions, squalene, TS, and TPGS were mixed into the lipid-ethanol mixture at a molar ratio of 75:15:10. Prior to ethanol diffusion, imiquimod (in DMSO) was added to the lipid-ethanol mixture to form a homogeneous solution. Next, the lipid-ethanol solution was rapidly injected into a 20 mM acetate buffer at pH 4.0 to form an imiquimod-incorporated nanoemulsion with a lipid concentration of 30.0 mg/mL and an imiquimod concentration of 1.5 mg/mL. The particle size and zeta potential of the pH-sensitive nanoemulsion loaded with nucleic acids were analyzed by dynamic light scattering using a NICOMP Z3000 Nano DLS/ZLS system (Entegris, Billerica, MA).

Preparation of cationic nanoemulsions incorporating tumor antigen peptides and TLR agonists.

Squalene, DOTAP and TPGS were mixed into the lipid ethanol mixture at a molar ratio of 68:30:2. Next, the synthetic peptide P1A (LPYLGWLVF, SEQ ID NO:3) dissolved in DMSO, resimod or MPLA was added to the lipid mixture to form a homogenous solution, and the lipid/peptide/TLRa mixture was then rapidly injected into 20 mM phosphate buffer (pH 7.4, containing 10% sucrose) to form a cationic nanoemulsion loaded with peptide/TLRa at a lipid concentration of 10.0 mg/mL. To form a cationic nanoemulsion encapsulating oligonucleotides, SD-101 oligonucleotide solution (in DEPC water) was added dropwise to slowly vortexed empty lipid nanoparticles at a weight ratio of 1 to 10 until a cationic nanoemulsion encapsulating nucleic acids was formed. The final concentrations of P1A peptide, MPLA, resimod and SD-101 were 1.0, 0.2, 0.2 and 0.2 mg/mL, respectively (or according to the formulation table). The control of each group was performed under the same formulation but without P1A peptide. The product was stored at -20°C before use. The particle size and zeta potential of the pH-sensitive nanoemulsion loaded with nucleic acid were analyzed by dynamic light scattering using a NICOMP Z3000 NanoDLS/ZLS system.

Preparation of standard lipid nanoparticles incorporating CpG oligonucleotides and anti-PDL1 LNA Gapmer. DSPC, cholesterol, DLin-MC3-DMA and DMG-mPEG ₂₀₀₀ were mixed into the lipid ethanol mixture at a molar ratio of 10:38.5:50:1.5. Next, the lipid ethanol mixture was rapidly injected into an acidic phosphate buffer to form empty lipid nanoparticles with a lipid concentration of 10.0 mg/mL. To form lipid nanoparticles encapsulating oligonucleotides, the oligonucleotide solution (in DEPC water) was added dropwise to the slowly vortexed empty lipid nanoparticles at a weight ratio of 1 to 10 until lipid nanoparticles encapsulating nucleic acids were formed. The product was incubated at 37°C for 10 minutes and stored at 4°C before use. The particle size and zeta potential of the pH-sensitive nanoemulsion loaded with nucleic acids were analyzed by dynamic light scattering using a NICOMP Z3000 Nano DLS/ZLS system.

Preparation of pH-sensitive nanoemulsion (PSNE) lipid nanoparticles. DOPC, DOPE, squalene, DODMA, and DMG-mPEG ₂₀₀₀ were mixed into a lipid-ethanol mixture at a molar ratio of 15:28:10:45:2. Next, the lipid-ethanol mixture was rapidly injected into an acidic phosphate buffer to form an empty pH-sensitive nanoemulsion with a lipid concentration of 8.0 mg/mL. At the same time, nucleic acid cargoes including SD-101 and anti-PDL1 locked nucleic acid (LNA) antisense oligonucleotides (ASOs) were dissolved in DEPC water at 0.4 mg/mL. Before mixing, the empty pH-sensitive nanoemulsion and nucleic acid cargo were heated to 60°C. Then, the cargo was added dropwise to the slowly vortexed empty pH-sensitive neutral nanoemulsion at a weight ratio of 1 to 20 until lipid nanoparticles encapsulating nucleic acids were formed. The product was incubated at 37°C for 10 minutes and stored at 4°C before use. The particle size and zeta potential of the pH-sensitive nanoemulsions loaded with nucleic acids were analyzed by dynamic light scattering using a NICOMP Z3000 NanoDLS/ZLS system.

Preparation of Next Generation pH-Sensitive Nanoemulsion (PSNE-Chol) Lipid Nanoparticles. DOPE, cholesterol, squalene, DLin-MC3-DMA, and DMG-mPEG ₂₀₀₀ were mixed into a lipid-ethanol mixture at a molar ratio of 7:38:5:48:2. Next, the lipid-ethanol mixture was rapidly injected into an acidic phosphate buffer/tris buffer to form an empty pH-sensitive nanoemulsion with a lipid concentration of 8.0 mg/mL. Simultaneously, nucleic acid cargos including SD-101, anti-PDL1 locked nucleic acid (LNA) antisense oligonucleotides (ASOs), or a mixture of both, were dissolved in DEPC water at 0.4 mg/mL. Prior to mixing, the empty pH-sensitive nanoemulsion and nucleic acid cargo were heated to 60°C. Then, the cargo was added dropwise to the rapidly vortexed empty pH-sensitive nanoemulsion at a weight ratio of 1 to 20 until nucleic acid-encapsulated lipid nanoparticles were formed. The product was titrated to pH 7 with 0.1 N NaOH, incubated at 37° C. for 10 min, and stored at 4° C. before use. The particle size and zeta potential of the nucleic acid loaded pH sensitive nanoemulsions were analyzed by dynamic light scattering using a NICOMP Z3000 Nano DLS/ZLS.

Cell culture. Hepa1-6, MC-38, and RAW264.7 were gifts from Dr. Kalpana Ghoshal, Ohio State University College of Medicine, Christopher Coss, and Peixuan Guo, Ohio State University College of Pharmacy, respectively, and were cultured in DMEM (Millipore Sigma) supplemented with 10% FBS (Millipore Sigma) and antibiotic-antimycotic (Invitrogen) at 37°C in a humidified atmosphere containing 5% _CO2 .

In vitro gene regulation assessment was performed by RT-qPCR. Cells were seeded in 6-well plates at a density of 0.25-0.5 million/well 24 h before LNP treatment. Next, cells were treated with LNPs in complete medium and incubated for 24 h before harvesting. Total RNA from cells was extracted using TRI reagent (Zymo Research) following the manufacturer's protocol. cDNA was prepared by a high capacity cDNA reverse transcription kit (Invitrogen). Real-time PCR was performed using Universal SYBR Green Supermix (Bio-Rad) on a QuantStudio™ 7Flex Real-Time PCR System. The relative amounts of RNA levels were calculated and compared according to the 2 ^-ΔΔCt method.

In vitro surface protein expression was evaluated by flow cytometry. 24 hours before LNP treatment, cells were seeded in 60 mm cell culture dishes at a density of 500,000 to 700,000 cells/well. Next, cells were treated with different concentrations of LNP, mouse interferon-γ or LPS in complete medium. Cells were washed twice with PBS and then harvested using Hank's enzyme-free cell dissociation solution (Millipore Sigma). Cell suspensions were centrifuged at 500 x g at 4 ° C, and the precipitate was washed once with PBS and then resuspended in FACS staining buffer. Single cell suspensions were fixed at room temperature in PBS containing 1% PFA for 45 minutes, and then stained with two biomarkers according to the manufacturer's protocol: mCd86-BV650 and mCD274-PE (Biolegend, San Diego, CA, USA). Stained cells were analyzed on a BD LSR Foretessa flow cytometer in the Flow Cytometry Shared Resource Core of the Ohio State University Comprehensive Cancer Center. Data were analyzed in FlowJo.

Cytotoxicity in macrophage conditioned medium. RAW264.7 cells were seeded in 6-well plates at a density of 500,000/well 24 hours before LNP treatment. Next, RAW cells were treated with next-generation PSNE (PSNE-Chol) LNPs incorporating SD-101CpG ODN or anti-PDL1 LNA, LPS, or a combination thereof. The cultured conditioned medium was harvested, centrifuged to remove cell debris, and stored at -80°C before use. MC-38 and Hepa1-6 were seeded in 96-well plates at a density of 3000-5000 cells/well 24 hours before conditioned medium treatment. The cells were treated in quadruplicate with 100 μL conditioned medium and incubated for 72 hours. Cell viability was detected by CellTiter Glo (Promega) using a Biotek Synergy H1 plate reader.

Mouse studies and MC-38/Hepa1-6 syngeneic tumor model. CD-1 Swiss mice and C57BL/6N mice were purchased from Charles River Laboratory. Animals were housed in a temperature-controlled room with a 12-hour light/12-hour dark cycle and fed a normal diet. All animal studies were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of The Ohio State University. Both male and female mice were used in the experiments.

The MC-38 murine colorectal cancer cell line MC-38 syngeneic subcutaneous murine colorectal cancer model was established by injecting 1 million MC-38 cells in PBS subcutaneously into the right flank of each C57 mouse. The Hepa1-6 syngeneic subcutaneous murine hepatocellular carcinoma model was established by injecting 1 million Hepa1-6 cells in FBS subcutaneously into the right flank of each C57 mouse. Tumor size and mouse weight were monitored daily using digital calipers and analytical scales, respectively.

In vivo antitumor activity evaluation of cationic nanoemulsions incorporating TLR agonists on the Hepa1-6 murine HCC syngeneic model. Treatment was initiated when the mean tumor size reached 80-100 mm ^3. Mice were randomly divided into 4 groups of 5 mice each. Treatments were given twice a week for a total of 8 doses, and the treatments included saline, TLR9a CNE (CpG 2216), TLR3a CNE (poly (I:C)) and TLR7a NE at doses of 50, 50 and 100 μg per injection. Mice were sacrificed when tumor size reached early resection criteria, and tumors were collected for further analysis.

In vivo vaccine efficacy evaluation and tumor challenge in the J558 mouse bone marrow syngeneic model. Wild-type (WT) Balb/c mice (2 mice/control group and 3 mice/peptide group) were treated subcutaneously with 100 μL of peptide/control vaccine for 7 days. Mice were bled for P1-CTL (cytotoxic T lymphocytes, CD8 ⁺ T lymphocytes) detection 7 days after immunization, and mice were challenged with J558 mouse myeloma cells on day 10. Mice were sacrificed 40 days after immunization, and activated P1-CTLs in spleen and tumors were detected by flow cytometry.

Evaluation of the in vivo antitumor activity of standard lipid nanoparticles incorporating CpG oligonucleotides and anti-PDL1 LNA gapmers in the MC-38 murine colon cancer syngeneic model. Treatment was initiated when the mean tumor size reached 80-100 mm ^3. Mice were randomly divided into 6 groups of 5 mice each. Treatments were given every three days for a total of 5 doses, including saline control, PBS containing anti-PDL1 2'-OMe ASO gapmer, PBS solution containing doxorubicin, LNP containing SD-101, LNP containing anti-PDL1 ASO, LNP containing SD-101 and LNP containing anti-PDL1 ASO in combination, and LNP containing anti-PDL1 ASO in combination with doxorubicin. Mice were sacrificed when tumor size reached early resection criteria, and tumors were collected for further analysis.

In vivo antitumor activity evaluation of pH-sensitive nanoemulsion (PSNE) lipid nanoparticles on the MC-38 murine colon cancer syngeneic model. Treatment was initiated when the mean tumor size reached 80-100 mm ^3. Mice were randomly divided into 6 groups of 5 mice each. Treatments were given every three days for a total of 5 doses, and the treatments included saline control, SD-101 oligonucleotide, PSNE with SD-101, PSNE with anti-PDL1 LNA, PSNE with SD-101 and anti-PDL1 LNA, and PSNE with resiquimod and SD-101. Mice were sacrificed 6 hours after the last dose, and various organs and whole blood samples were collected for further analysis.

Evaluation of the in vivo antitumor activity of next generation pH sensitive nanoemulsion (PSNE-Chol) lipid nanoparticles on the MC-38 murine colon cancer syngeneic model. Treatment was initiated when the mean tumor size reached 80-100 mm ^3. Mice were randomly divided into 5 groups of 6 mice each. Treatments were given every three days for a total of 5 doses, including saline control, PSNE-Chol with SD-101, PSNE-Chol with anti-PDL1 LNA, PSNE-Chol with SD-101 and anti-PDL1 LNA, and a mixture of PSNE-Chol with SD-101 and PSNE-Chol with anti-PDL1 LNA. Mice were sacrificed when tumor size reached early resection criteria, and various organs and whole blood samples were collected for further analysis.

Quantification of splenocyte and tumor tissue messenger RNA by RT-qPCR. Splenocyte messenger RNA for gene expression analysis was extracted from splenocyte suspension using TRI reagent according to the manufacturer's protocol. cDNA was prepared using a high-capacity cDNA reverse transcription kit (Invitrogen). Real-time PCR was performed on a QuantStudio 7Flex Real-Time PCR system. The relative amounts of RNA levels were calculated and compared according to the 2-ΔΔCt method.

Splenocyte and tumor-infiltrating immune cell population detection by flow cytometry. Single-cell splenocyte suspensions were harvested from the spleen by the following procedure. The spleen was gently sieved through a 70um nylon mesh (ThermoScientific) using a sterile 5-mL syringe piston and washed twice in a six-well plate with cold RPMI medium. The crude splenocyte suspension was then centrifuged at 500x g for 5 minutes at 4°C to obtain a cell pellet. The cell pellet was resuspended in 1x RBC lysis buffer to lyse red blood cells at room temperature for 5 minutes, and 10 volumes of PBS were added to terminate the reaction. After centrifugation at 500x g for 5 minutes, the cell pellet was then resuspended in FACS staining buffer to form a single-cell splenocyte suspension.

Antibodies were purchased from Biolegend (San Diego, CA) and surface cell staining was performed according to the manufacturer's protocol. Intracellular staining was performed using the Biolegend True-NuclearTM Transcription Factor Buffer Set according to the manufacturer's protocol. Stained cells were analyzed on a BD LSR Fretessa flow cytometer in the Flow Cytometry Shared Resource Core e at the Ohio State University Comprehensive Cancer Center.

Cytokine ELISA. Serum samples were obtained from whole blood samples by incubating the blood at room temperature for 30 minutes followed by centrifugation at 2,000 x g for 20 minutes. Serum was collected and stored at -80°C prior to cytokine analysis. Cytokine concentrations were measured by uncoated ELISA kits purchased from Invitrogen according to the manufacturer's protocol.

Statistical analysis. Statistical analysis was performed using GraphPad PRISM software version 9 (GraphPad Software, San Diego, CA, USA). p values ≤ 0.05 were considered significant.

result

In vivo evaluation of the antitumor activity of cationic nanoemulsions incorporated with TLR agonists in the Hepa1-6 murine HCC syngeneic model.

In the first preliminary study, several different toll-like receptor agonists were examined using squalene-based nanoemulsions and intratumoral injection. Toll-like receptors are known to trigger immune responses because different receptors are responsible for sensing different classes of PAMPs. Here, TLR3 agonist poly(I:C), TLR7 agonist imiquimod, and TLR9 agonist CpG 2216 were selected and encapsulated into nanoemulsions via hydrophobic interactions (imiquimod) or electrostatic interactions (poly(I:C) or CpG 2216). The drug nanoparticles were stable when stored at 4°C and did not form aggregates. Due to the limited cavities in tumor tissue, the injection volume was limited to 50-100 μL per injection. Administration was performed by injecting the nanoemulsion directly into the central cavity of the tumor under careful monitoring. In Figure 19, the tumor progression curve shows a clear trend in the effect of TLR agonists in tumor growth inhibition. Although the inhibition of tumor growth did not reach statistical significance, it was clear that CpG 2216 (a TLR9 agonist and A-type CpG oligodeoxynucleotide) was the most effective of all in inhibiting tumor growth. Surprisingly, poly (I:C) (a TLR3 agonist shown to trigger dendritic cell maturation and tumor regression) had no anti-tumor effect compared to the squalene vehicle control. The potency of the TLR agonist squalene nanoemulsion was compared as follows: CpG 2216 (TLR9) was the most potent of all drugs, followed by imiquimod (TLR7), and poly (I:C) (TLR3) was the least potent.

FIG. 19 is based on the treatment period before the early removal criteria. Since the Hepa1-6 syngeneic tumor is an aggressive tumor during progression and may form bleeding ulcers in the late stage, mice must be killed early to comply with the regulations. Therefore, if the survival rate of those mice in good condition is considered ( FIG. 20 ), the results show that CpG 2216 squalene nanoemulsion and imiquimod squalene nanoemulsion can cause complete tumor regression in two or three mice, respectively ( FIG. 20 , D and FIG. 20 , E). The anti-tumor results are promising, and additional tumor challenges were performed on those mice that completely regressed. For tumor challenge of immunized mice, 1-2 million Hepa1-6 cells were injected subcutaneously at a site similar to the previous tumor, and the mice were monitored for tumor uptake and progression for another two weeks. In untreated C57BL/6N mice, cancer cells should be absorbed and progress rapidly one week after inoculation. During the tumor attack period, the monitoring time was extended to two weeks to ensure that the inoculated cancer cells were completely eliminated. The results showed that none of the five mice with complete tumor regression had tumor uptake after tumor challenge, suggesting that the mice developed anticancer immunity specific to Hepa1-6 HCC.

In vivo cancer vaccine efficacy evaluation and tumor challenge in the J558 murine bone marrow syngeneic model.

In the first pilot study, only TLR agonists were delivered using nanoparticle carriers. Tumor-associated antigens are endogenously provided in the tumor microenvironment. During intratumoral injection, dendritic cells and other phagocytes can take up TLR agonist nanoparticles and/or tumor-associated antigens to trigger an inflammatory response and dendritic cell activation or maturation. Antigen presentation in dendritic cells is a highly specific and regulated process that requires degraded pathogen proteins (antigens) and activation signals (TLR agonists). Plasmacytoid dendritic cells (pDCs) activated by CpG TLR9 agonists are professional antigen-presenting cells (APCs) that present antigens to CD8 ⁺ T lymphocytes through cross-antigen presentation. However, a successful antigen presentation process requires the simultaneous presence of PAMPs (here TLR agonists) and antigenic proteins in APCs. Therefore, in order to develop a clear cancer vaccine for a specific cancer subtype, in a second pilot study, a combination of well-defined antigenic peptides and TLR agonists were added to squalene nanoemulsions.

Since single TLR activation did not produce a substantial response, dual or triple TLR agonist combinations were considered. Poly(I:C) was abandoned due to its low efficacy. MPLA, a TLR4 agonist that acts similarly to lipopolysaccharide (LPS) by simulating bacterial infection, was selected as one of the components. In addition, imiquimod was replaced by resiquimod, a modified version of the TLR7/8 agonist, to improve drug solubility and process operations. In addition, SD-101 CpG oligonucleotide (C-type CpG oligonucleotide) was added to the product combination together with CpG 2216 (A-type CpG oligonucleotide). The efficacy of each combination (Table 2) was evaluated by subcutaneously immunizing mice with peptide-incorporated TLR agonist nanoemulsions followed by tumor attack.

Table 2. TLR agonist compositions of cationic nanoemulsions incorporating tumor antigen peptides and TLR agonists.

P1A is a well-defined antigenic peptide of murine myeloma cell J558. The peptide can be synthesized by solid-state peptide synthesis with high purity and precision. The sequence of the P1A peptide is LPYLGWLVF (SEQ ID: 3), and more than half of the residues are hydrophobic or aromatic amino acids (highlighted in bold). The high lipophilicity of the P1A antigenic peptide is dissolved in the squalene core and a cationic nanoemulsion is formed by adding permanently charged DOTAP to the formulation. DOTAP is used as a static charge source to pair with oligonucleotides to form stable cationic nanoparticles or nanolipid complexes.

As a cancer vaccine candidate, mice were immunized once to illustrate the efficacy of immune system stimulation, and tumor attacks were performed to demonstrate the protective ability after immunization. In Figure 21, the results clearly show that the PC with a weight ratio of 1:1:1 combined with CpG 2216, MPLA and Resimod can produce J558-specific anti-peptide cytotoxic T lymphocytes, and cause the growth delay of J558 tumors during the tumor attack process. Other combinations (such as F1 to F3 and the combination of double TLR agonists of single lipid nanoparticles and P1A antigen peptides) show significant activation of J558-specific immunity, and can delay or eliminate the occurrence of J558 attacks. Preliminary studies using TLR agonist combinations and adding tumor-specific antigen peptides show that the enhancement of cancer vaccines depends on the co-delivery of tumor-associated antigens and TLR agonists as key stimulating factors.

Evaluation of the in vivo antitumor activity of standard lipid nanoparticles incorporating CpG oligonucleotides and 2'-OMe anti-PDL1 gapmer in the MC-38 murine colon cancer syngeneic model.

In the third study, SD-101 was combined with anti-PDL1 antisense oligonucleotides (anti-PDL1 ASOs) to detect anti-tumor efficacy not only by stimulating immune responses, but also by downregulating the checkpoint molecule PD-L1 or programmed cell death ligand 1. PD-L1, also known as cluster of differentiation 274 (CD274), is a protein that acts as a brake to stop immune responses by binding to its receptor PD-1 and inducing inhibitory signals. By downregulating PD-L1 expression in immune cells or cancer cells, T lymphocyte responses are not attenuated by signaling pathways triggered by PD-1/PD-L1 interactions. To demonstrate the activity of both SD-101 CpG ODN and anti-PDL1 ASOs, the two oligonucleotides were completely encapsulated in standard lipid nanoparticles with gold standard ionizable lipids DLin-MC3-DMA, with a gold composition of DSPC/cholesterol/DLin-MC3-DMA/DMG-mPEG ₂₀₀₀ = 10/38.5/50/1.5 (m/m). Mice bearing MC-38 tumors were treated once every three days for a total of five doses. The results (Figure 22) demonstrated that mice treated with SD-101LNP or a combination of SD-101LNP and anti-PDL1ASO LNP had significant tumor regression 21 days after the start of treatment compared to saline controls, indicating that the overall immunosuppressive environment of tumor-bearing animals can be reversed by systemic application of TLR agonists. On the other hand, anti-PDL1 ASO or anti-PDL1ASO LNP groups alone did not show significant tumor growth inhibition compared to saline controls. In addition, when comparing the tumor burden of mice treated with doxorubicin and mice treated with a combination of doxorubicin and anti-PDL1 ASO LNP, the results showed that anti-PDL1 ASO LNP had no therapeutic activity compared to doxorubicin chemotherapy drugs. When the monitoring period was extended to three weeks after the final dose (Figure 22), the results showed that mice treated with the LNP combination had a slower tumor growth rate, but the mice ultimately suffered from a massive tumor burden and the combination produced no preventive or therapeutic effects.

In addition to tumor burden analysis, body weight analysis during the treatment period is an important parameter for evaluating the overall systemic toxicity caused by the therapeutic agent. In Figure 23, the results clearly show that the mice in the LNP combined treatment group lost weight significantly during the treatment period (the first 15 days) and had to suspend treatment due to significant toxicity. In addition, the LNP monotherapy group experienced at least 10% weight loss during the first administration, and this shows that the LNP platform has significant intrinsic toxicity. It has been reported that LNP is known to trigger cytokine release syndrome, which may worsen the health of mice. However, as the treatment continued, those mice treated with LNP monotherapy adapted to the LNP carrier and had a weight gain curve similar to that of the saline control. The overall results show that SD-101LNP treatment is an effective anti-cancer treatment option, while 2'-OMe anti-PDL1gapmer ASO has less effect on tumor regression.

In vitro gene regulatory activity assessment of anti-PDL1 Gapmer ASOs by RT-qPCR.

The anti-PDL1 gapmer ASO was originally designed by Roche with locked nucleic acid (LNA) modification at both ends and labeled with GalNac moieties for liver targeting. In our study, the GalNac moiety was removed to obtain a single-stranded 16-mer anti-mouse PD-L1 gapmer antisense oligonucleotide. In the initial test, 2'-OMe-modified nucleotides were used at both ends to protect the antisense oligonucleotide from nuclease degradation. However, studies have shown that 2'-OMe modification provides less improvement over the original deoxynucleotide and less enhancement of messenger RNA binding. A newer version of nucleotide modification is locked nucleic acid (LNA), in which the 2'-O on the ribose is linked to the 4'-C via a methylene bridge. The binding affinity of LNA-modified ASOs is significantly stronger than that of 2'-OMe-modified ASOs, and the former can increase the binding of RNase H to the complementary sequence to degrade mRNA after binding to the target messenger RNA. Therefore, the original design of LNA-modified anti-mouse PD-L1 gapmer ASO was tested for target gene downregulation relative to 2'-OMe-modified ASO in vitro on different mouse cell lines by different transfection methods. In Figure 24, Hepa1-6 mouse HCC cell line and MC-38 mouse colon cancer cell line were evaluated using DOTAP/DOPE as transfection agent in serum-free medium or oligofectamine in complete medium. The results show that LNA-modified ASO has higher gene downregulation activity than the 2'-OMe version and that the downregulation efficiency is a cell line-dependent process. It is clearly shown that the Pdl1 gene in MC-38 is not as easily downregulated as the gene in Hepa1-6, and this trend is shown in both DOTAP/DOPE- and oligofectamine-treated cells. In short, in order to obtain a better therapeutic response by downregulating PD-L1 in murine tissues, LNA-modified ASOs must be introduced into LNP constructs to ensure significant downregulation of the target gene.

In vivo evaluation of the antitumor activity of pH-sensitive nanoemulsion (PSNE) lipid nanoparticles in the MC-38 murine colon cancer syngeneic model.

To evaluate the in vivo anticancer activity of PSNE lipid nanoparticles encapsulating SD-101CpG ODN, anti-PDL1 LNA ASO and resiquimod, mice were treated with single or combined treatment at a dose of 2.0 mg/kg, once every three days, for a total of five doses. In Figure 25, the tumor growth curve shows that resiquimod encapsulating PSNE has a small effect on tumor growth inhibition, while the combination of resiquimod with anti-PDL1 LNA ASO and resiquimod with SD-101CpG oligonucleotide has a significant effect on tumor growth inhibition, with TGIs of about 57.1% and 70.4%, respectively.

In the second experiment, different parameters and properties of PSNE lipid nanoparticles were examined and evaluated. PSNE lipid nanoparticles encapsulating oligonucleotides and small molecules with high colloidal stability were successfully formulated. Small particle sizes of about 80-100 nm were obtained for both empty PSNELNPs and LNPs encapsulated with SD-101 CpG ODN or anti-PDL1 LNA ASO. The polydispersity index (PdI) was about 0.15-0.25, indicating that uniform particle size distribution was obtained in all samples (Figure 26). Empty PSNE lipid nanoparticles had an average zeta potential of +8.06 mV in 20 mM PB at pH 4, while encapsulated PSNE lipid nanoparticles had an average zeta potential close to neutral in 10 mM PB at pH 7.

Next, the encapsulation of ODN and small hydrophobic cargo in PSNE LNPs was analyzed by gel retardation assay (Figure 27) and size exclusion chromatography (Figure 28), respectively. The gel image showed that when resiquimod was added to the formulation, ODN was highly encapsulated in PSNE LNPs, and a small amount of SD-101 was adsorbed onto the PSNE surface. By calculating the ratio of the two peaks on the chromatogram obtained by OD327 using a NanoDrop spectrophotometer, the encapsulation efficiency of resiquimod in PSNE lipid nanoparticles was about 53.8% (Figure 28).

Table 3. Lane descriptions of samples loaded in agarose gel retardation assay (Figure 27).

Lanes Lanes 1 SD-101 5 PSNE/SD-101 2 Anti-PDL1 LNA 6 PSNE/LNA 3 PSNE-2% PEG 7 PSNE/SD÷LNA 4 PSNE-2％/R848 8 PSNE/R848+SD

The combination of SD-101 CpG ODN and anti-PDL1 LNA ASO in PSNE lipid nanoparticles showed good tumor growth inhibition results in the MC-38 mouse isogenic colon cancer model (Figure 29, Figure 30), with an average tumor growth inhibition (TGI) of about 73%. Both groups of mice treated with a single drug (SD-101 ODN or anti-PDL1 LNA ASO) in PSNE showed a TGI of about 46%, while the average TGI of mice treated with PSNE encapsulating SD-101 and resimod was 57% (Figure 30). Tumor weights were recorded and represent the actual tumor size after treatment. The results showed a similar trend to tumor volume, as the PSNE combination group showed significant tumor growth inhibition compared to saline controls using a one-way ANOVA. (Figure 31).

The spleen index (spleen weight expressed as a percentage of body weight, Figure 32) results showed that PSNE has similar immune system activation ability as SD-101 oligonucleotide. In addition, co-treatment with dual TLR agonists did further activate the immune system (by expanding the immune cell population). In addition, the results also showed that PSNE lipid nanoparticles encapsulating anti-PDL1 LNA ASOs do not have immune system activation ability. Tumor growth inhibition by PSNE/anti-PDL1 LNA ASOs may be caused by PD-L1 downregulation at the tumor site, whether the LNA is delivered to cancer cells or TME immune cells.

In flow cytometry analysis (Figure 33), the frequencies of T lymphocyte populations (whether CD4 ⁺ T helper cells, CD8 ⁺ cytotoxic T cells or regulatory T cells) were significantly reduced in all PNSE-based treatment groups compared to the saline control group. The results can be described as T cell activation and, therefore, infiltration into the TME for immune response. IHC staining of T cell populations can be performed to verify this hypothesis. There was no significant downregulation of MDSC frequency after all treatments in the MC-38 tumor model.

The surface PD-L1 markers of lymphocytes and bone marrow cells were detected by flow cytometry to study the PD-L1 protein expression of immune cells after treatment (Figure 34). The results showed that the surface PD-L1 protein level of macrophages was downregulated in the group treated with anti-PDL1LNA ASO, because macrophages have the ability to engulf LNP and take up LNA for mRNA downregulation. Nevertheless, the PD-L1 protein expression of cytotoxic T lymphocytes was not significantly downregulated after treatment with anti-PDL1LNA, indicating the difficulty of transfecting T cells. As we expected, the PD-L1 protein expression level of the TLR (SD-101ODN and Resimod) treatment group was upregulated due to the NFκB-MyD88 activation pathway. This also shows that anti-PDL1 LNA can overcome the PD-L1 upregulation caused by TLR agonists. By performing splenocyte mRNA analysis, we can understand the cytokine secretion profile and overall PD-L1 levels after treatment with TLR agonists and/or anti-PDL1 LNA. The results demonstrated the excellent activity of LNA in downregulating PD-L1 mRNA in splenocytes (Figure 35). Based on the cytokine ELISA results (Figure 36, Figure 37), we can easily notice that mice treated with PSNE/anti-PDL1 LNA are not in an inflammatory state (basal levels of proinflammatory cytokines TNF-α and IFN-γ); therefore, tumor growth inhibition is associated with the delivery of anti-PDL1 LNA ASO to tumors. In addition, the results also showed that TLR agonists are able to put the immune system into a proinflammatory phase by increasing IL-12 levels.

In vitro gene regulation assessment by RT-qPCR.

PSNE was further developed by applying knowledge from COVID-19 vaccines to lipid nanoparticle delivery. The next generation PSNE consists of DOPE, squalene, DLin-MC3-DMA, cholesterol, and DMG-mPEG _2000. Gene delivery (antisense oligonucleotides, siRNA, mRNA) was evaluated by in vitro RT-qPCR and in vivo luciferase bioluminescence (data not shown). The composition was fine-tuned and 11 different formulations were examined and presented as a series of next-generation PSNE for different applications. Based on the evaluation results (data not shown), several candidates have high in vivo expression for systemic administration, such as PSNE-Chol/M5. In addition, PSNE-Chol/M9-M11 was tested in vitro together with M5, as M9, M10, and M11 showed high mRNA delivery efficiency in vitro to various cell lines (HEK293T, A549, Hepa1-6, and HepG2).

In the gene delivery evaluation study, three mouse cell lines with different characteristics were used, including the Hepa1-6 HCC cell line, the MC-38 colon cancer cell line, and the RAW264.7 macrophage cell line. The first two represent tumor burden, while the latter represents the immune cell population. All gene expression levels were normalized to the PBS control group, allowing us to make clear comparisons in each treatment group. As shown in Figure 38, the results again showed that Hepa1-6 was more sensitive to treatment with various forms of anti-PDL1 LNA ASO. Surprisingly, all next-generation PSNE platforms induced the expression of mouse PD-L1 in Hepa1-6, but this upregulation was significantly inhibited by anti-PDL1 LNA ASO. That is, by using the next-generation PSNE as a delivery platform, high gene regulation efficacy of mouse PD-L1 was shown. As shown in the previous section, MC-38 was not as sensitive to anti-PDL1 LNA ASO treatment as Hepa1-6. However, in both cell lines, it is clear that PSNE-Chol/M5 has the best overall mouse PD-L1 downregulation efficiency among all next-generation PSNEs tested.

On the other hand, the results from RAW cells showed that immune cells were not sensitive to gene modulation by anti-PDL1 LNA ASO using next-generation PSNE. The expression levels of murine PD-L1 did not change significantly when treated with control oligonucleotides or anti-PDL1 LNA ASO. Murine PD-L1 expression was slightly upregulated in each treatment group, suggesting that the next-generation PSNE platform may have the ability to activate RAW cells, such as cytokine release, which has been reported for most lipid nanoparticle formulations.

Next, the next generation of PSNE, especially PSNE-Chol/M5, was tested in vitro in Hepa1-6 and MC-38 to evaluate the efficiency of mouse PD-L1 downregulation under cytokine stimulation. Cancer cells are highly sensitive to the surrounding cytokines and chemokines and will respond accordingly to environmental stimuli. It is well known that the proinflammatory cytokine interferon gamma (IFN-γ) induces surface PD-L1 expression in cancer cells in response to the proinflammatory state and activation of the adaptive immune system. In Figure 39, RT-qPCR results confirmed that both Hepa1-6 and MC-38 responded to IFN-γ induction in PD-L1 expression, and Hepa1-6 was more dominant in upregulation. Next, PSNE-Chol/M5 LNPs encapsulated with anti-PDL1 LNA ASO were treated with IFN-γ to verify the downregulation efficacy of LNPs. The results showed that PSNE-Chol/M5 could successfully deliver anti-PDL1 LNA ASO to cells and down-regulate PD-L1 mRNA expression back to basal levels after induction of IFN-γ. In addition, RT-qPCR results showed that even free anti-PDL1 LNA antisense oligonucleotides could be taken up by cells at a certain level and led to significant downregulation of PD-L1 mRNA, but not as effectively as the PSNE-Chol/M5 delivery platform. Overall, RT-qPCR demonstrated the reliable efficacy of PSNE-Chol/M5 LNP with anti-PDL1 LNA ASO for gene-level regulation, even for pro-inflammatory cytokine-induced PD-L1 cells.

In vitro surface protein expression assessment was performed by flow cytometry.

In the previous section, gene regulation of murine PD-L1 mRNA by IFN-γ and/or PSNE-Chol/M5-anti-PDL1 LNA ASO was confirmed by RT-qPCR, and the results demonstrated promising results using PSNE-Chol/M5-anti-PDL1 LNA ASO as an agent to reduce mRNA expression levels at the mRNA stage. However, the protein expression level of surface PD-L1 must be confirmed to further demonstrate the relevance of using PSNE-Chol/M5-anti-PDL1 LNA ASO as an anticancer agent in regulating immune checkpoint expression. To investigate surface PD-L1 expression in cancer cells and immune cells, Hepa1-6 and MC-38 cells treated with IFN-γ and PSNE-Chol/M5-anti-PDL1 LNA ASO were scanned using flow cytometry to quantify surface immune checkpoint expression. In addition, surface activation marker CD86 as well as surface PD-L1 expression were detected in RAW264.7 cells treated with PSNE-Chol/M5-SD-101. In Figures 40A-40B, the results show that the surface expression of PD-L1 is highly correlated with the gene expression results obtained from RT-qPCR. When incubated with IFN-γ, the protein expression of Hepa1-6 was significantly upregulated, and the protein expression was inhibited by PSNE-Chol/M5-anti-PDL1 LNA ASO. Compared with the positive control group treated with IFN-γ, cells treated with IFN-γ and free anti-PDL1LNA ASO showed no significant changes in surface marker expression, indicating that free antisense oligonucleotides have less effect on downregulating target genes in the absence of any delivery or targeting ligand or platform. On the other hand, the increase in surface PD-L1 expression was less significant in MC-38 cells compared with Hepa1-6 cells, which is also related to the RT-qPCR data. However, even at a higher treatment level of 200nM, PSNE-Chol/M5-anti-PDL1 LNA ASO could not completely reverse the protein expression of MC-38 after IFN-γ induction. The overall data suggest that MC-38 may not be a good model for PSNE-Chol/M5-anti-PDL1 LNA ASO in patients with elevated PD-L1 expression.

RAW264.7 cell surface markers were examined after treatment to study the immune activation effects of each agent and combination. As a positive control for macrophage activation, LPS (a TLR4 agonist) showed a significant increase in CD86 expression and PD-L1 expression after overnight treatment. However, cells treated with free SD-101 oligonucleotides (a TLR9 agonist) did not show the same strong activation as LPS at 5 μg/mL levels (increased surface CD86 expression). On the other hand, RAW cells treated with LPS and PSNE-Chol/M5-anti-PDL1 LNA ASO showed reduced PD-L1 expression compared to the LPS-only group, but PD-L1 expression was still significantly induced by LPS (Figure 41).

Cytotoxicity in macrophage conditioned medium.

After validating the direct effects of PSNE-Chol/M5-SD-101LNP and PSNE-Chol/M5-anti-PDL1 LNA ASO LNP formulations on cancer cells and macrophages by RT-qPCR and flow cytometry, the indirect effects of PSNE-Chol/M5-SD-101 and PSNE-Chol/M5-anti-PDL1 LNA ASO LNP formulations on cancer cells were determined by examining the viability of cancer cells after incubation with macrophage conditioned medium. Conditioned medium was prepared by harvesting cytokine-containing medium from macrophages treated overnight with next-generation PSNE formulations. Short-lived cytokines secreted by macrophages were retained and interacted directly with cancer cells. Cancer cells Hepa1-6 and MC-38 were treated with conditioned medium containing RAW264.7 macrophage cytokines and viability was determined after 72 hours. Interestingly, Hepa1-6 (data not shown) had no significant cytotoxic effect on drug-treated conditioned medium compared to normal conditioned medium (PBS control). The morphology of Hepa1-6 cells was found to be unchanged or unaffected under bright light microscopy. On the other hand, MC-38 cells (Figure 42) were more sensitive to conditioned medium treatment, and under bright light microscopy, the cells underwent apoptosis or necrosis in morphology. The positive control group was cells treated with LPS conditioned medium known to contain a large amount of cytotoxic cytokine TNFα, which induced significant cancer cell death after 72 hours of incubation. The addition of free anti-PDL1 LNA ASO to macrophage treatment did not change the cytotoxic effect of MC-38, but the addition of PSNE-Chol/M5-anti-PDL1 LNA ASO LNPs and LPS can further secrete more cytotoxic cytokines, which may or may not be limited to TNF-α, and lead to a higher cell death rate. Next, the free SD-101CpG oligonucleotide treatment was compared with LPS, and the results showed that 5μg/mL SD-101CpG oligonucleotide had a similar effect to 10μg/mL LPS in conditioned medium treatment. Furthermore, when SD-101 oligonucleotides were encapsulated into the next generation PSNE (PSNE-Chol/M5), the nanoparticle carrier effect further increased the response and further enhanced the cell killing effect of the conditioned medium, and the carrier effect was also shown on cells treated with conditioned medium from macrophages treated with PSNE-Chol/M5-anti-PDL1 LNA ASO or PSNE-Chol/M5-control ODN, which had a significant cytotoxic effect on MC-38 compared with PBS conditioned medium or fresh complete medium.

In vivo evaluation of the antitumor activity of next-generation pH-sensitive nanoemulsion (PSNE-Chol) lipid nanoparticles in the MC-38 murine colon cancer syngeneic model.

The in vitro validation of next-generation PSNE LNPs in cancer cell lines and macrophage cell lines gave us some insight into their in vivo therapeutic effects. Therefore, next-generation PSNE LNPs incorporating SD-101 and anti-PDL1 LNA ASO were tested as single agents or in combination on the MC-38 syngeneic model. Based on this hypothesis, the anticancer efficacy of next-generation PSNE should be higher than that of the previous generation PSNE without cholesterol. The delivery efficiency of anti-PDL1 LNA ASO to cancer cells and spleen cells should also be higher than that using the previous generation PSNE. Mice were treated with saline control, PSNE-Chol/M5-SD-101 LNPs, PSNE-Chol/M5-anti-PDL1 LNA ASO LNPs, a 1:1 mixture of the two LNPs, or LNPs co-loaded with oligonucleotides at a 1:1 weight ratio. In Figures 43 and 44, the tumor growth curves again show that the PSNE-Chol/M5-anti-PDL1 LNA ASO LNP monotherapy group is not as effective as the PSNE-Chol/M5-SD-101LNP monotherapy group in inhibiting MC-38 tumor growth. In addition, no combination group is worse than the PSNE-Chol/M5-SD-101LNP single agent. The tumor growth inhibition (TGI) rates of PSNE-Chol/M5-SD-101LNP, PSNE-Chol/M5-anti-PDL1 LNA ASO LNP and the mixture combination were 76.5%, 60.5% and 72.9%, respectively. Compared with the saline control group, the PSNE-Chol/M5-SD-101LNP and mixture combination groups both showed significant in vivo tumor growth inhibition. In addition, according to Figure 43, the tumor growth curves show that these two groups can inhibit MC-38 tumor growth during the treatment period, but the tumor will begin to progress after the treatment is stopped.

However, the toxicity of LNP drugs is monitored by weight loss and early vitality during the treatment period.Weight curve (Figure 45) shows that the initial combination group (including mixture combination and co-load combination) significantly loses about 20% of body weight after the first administration, and causes several early deaths on the third day for the treatment, which shows that the severe toxicity caused by the LNP delivery platform (especially DLin-MC3-DMA ionizable lipids), as shown in the upper part, the high-dose DLin-MC3-DMA containing LNP may cause severe toxicity.Toxicity may come from cytokine release syndrome or the off-target effect to important tissues.From the second administration, the dosage treatment combination group is reduced (half) to control the delivery platform toxicity, the combination group has the lipid dosage identical with the single drug group, and all surviving mice can tolerate the drug effect until the treatment is terminated.

After the treatment was completed and the mice were sacrificed, several important parameters were measured and examined (Figure 46) to investigate possible mechanisms of tumor growth inhibition. Since the mice were sacrificed at a time point that met the removal criteria (i.e., 1.6 cm in length) to obtain survival, there was no significant difference in the average tumor size when measured after sacrifice, but the PSNE-Chol/M5-SD-101LNP monotherapy group showed an average lower tumor burden. Nevertheless, when examining splenomegaly, the results clearly showed that TLR agonist treatment can induce immune cell proliferation, leading to splenomegaly. As shown in Figure 46, Figure B, the PSNE-Chol/M5-SD-101LNP monotherapy group showed a significant increase in spleen size after five doses of treatment, which is consistent with the previous trend, but the effect is even better than the dual TLR agonist group using the first-generation PSNE LNP. This finding suggests that the next generation of PSNE can induce immune system activation at a higher level.

Next, the spleen cell population was examined to illustrate the effect of the next generation PSNE delivery platform on immune cells. As shown in Figure 47, in all treatment groups, the population frequencies of CD4 ⁺ and CD8 ⁺ T lymphocytes in the spleen decreased after treatment, consistent with the previous trend, but with lower frequencies. Interestingly, the PSNE-Chol/M5-anti-PDL1 LNA ASO LNP monotherapy group had a wider distribution in both T lymphocyte populations. The reduction of both T lymphocyte populations was related to the dose of SD-101CpG ODN, because all treatments had the same lipid dose throughout the treatment period. At the same time, the regulatory T lymphocyte (Figure 48, CD4 ⁺ Foxp3 ⁺ ) population was not significantly affected by treatment, because the PSNE-Chol/M5-SD-101LNP monotherapy group had a trend of reducing T _reg populations after treatment. Surprisingly, the co-loaded combination group had a significantly increased population after treatment, while the mixture combination did not. The same weight of oligonucleotides and lipids were administered to these mice, but in different loading modes, suggesting that the encapsulation method may lead to different treatment outcomes. In addition, the genes of interest in splenocytes were examined by RT-qPCR. The results are shown in Figure 49, showing some interesting findings. First, in the group treated with SD-101CpG ODN, splenocyte Pdl1 expression was significantly downregulated while anti-PDL1 LNA ASO was not downregulated, but PSNE-Chol/M5-anti-PDL1 LNA ASO LNP still caused a downward trend in Pdl1 expression. Siglech and Foxp3 mRNA levels were also examined because these genes represent characteristic cell populations plasmacytoid dendritic cells (pDC) and regulatory T lymphocytes, respectively. RT-qPCR results demonstrated that all treatment groups significantly reduced the expression level of Siglech in splenocytes, indicating that next-generation PSNE treatment can lead to a reduction in the pDC population, and this reduction may be due to DC activation or redistribution of pDC in the spleen after immune system activation. A similar trend was also shown in Foxp3 expression, with the PSNE-Chol/M5-SD-101LNP monotherapy group showing the least expression in Foxp3 mRNA, consistent with the results obtained from flow cytometry.

Not only spleen cell mRNA expression was examined, but also tumor single cell mRNA expression was examined. Tumor single cell suspensions were obtained by using protease to dissociate tumor tissue to remove the extracellular matrix. The RT-qPCR results of tumor single cell suspensions showed that there was no significant difference in Pdl1 expression, but the PSNE-Chol/M5-anti-PDL1 LNA ASO LNP monotherapy group showed lower Pdl1 expression (Figure 50). In addition, different cytokine mRNA expression levels were evaluated to indirectly clarify the cytokine levels in the tumor microenvironment. Although the results (Figure 51) did not show any significant differences in the expression of Tnfa, Ifng, Il10, Il6 or Tgfb, several key findings can be drawn from the results. Compared with the saline control group, the expression of Tnfa in the PSNE-Chol/M5-SD-101LNP monotherapy group was significantly increased, indicating that the secretion of the proinflammatory cytokine TNF-α in the tumor microenvironment was higher, resulting in a more cytotoxic environment.

In addition, based on the biodistribution properties of the nanoparticles, the anatomical structure of the liver, and the biodistribution assay of PSNE-Chol/M5 modified with DiR lipid near-infrared fluorescent dye, RT-qPCR was used to analyze the Pdl1 mRNA levels in the liver of the PSNE-Chol/M5-anti-PDL1 LNA ASO LNP treatment group, where most of the systemically administered lipid nanoparticles accumulated in the liver, with less accumulation in the spleen (3% of the FL intensity of the liver, data not shown). The RT-qPCR results (Figure 52) showed that liver Pdl1 mRNA was significantly downregulated, indicating that most PSNE-Chol/M5-anti-PDL1 LNA ASO LNPs entered the liver and were effectively absorbed by hepatocytes, resulting in degradation of mRNA by the LNA ASO and RNase H pathways.

discuss

In this embodiment, several strategies for stimulating the immune system to fight cancer were tested. We chose to activate dendritic cells by toll-like receptor activation as the central concept, and tested multiple strategies and combinations using different lipid nanoparticle preparations. In the early stages of this project, cationic nanoemulsions were used to encapsulate hydrophobic agents (such as lipophilic TLR agonists or antigenic peptides) and oligonucleotides into single particles for delivery. In the later stages of this project, a newly developed pH-sensitive nanoemulsion lipid nanoparticle delivery platform was used to enhance the delivery of antisense oligonucleotides and CpG oligonucleotides.

To enhance immune system activation, tumor-associated antigens and toll-like receptor agonists are required. In our first series of constructs, different TLR agonists were encapsulated into cationic nanoemulsions and injected intratumorally. The source of antigens for dendritic cell activation was defined as tumor cell fragments of necrotic tumor cells, proteins/fragments released in exosomes, or phagocytosed dead cancer cells. When the tumor microenvironment is replete with useful antigens, the internal environment is anti-inflammatory, and cytotoxic immune cell function is inhibited by released anti-inflammatory cytokines such as TGF-β. Therefore, TLR agonists are needed to reactivate immune cells, and agonists are supplemented by nanoemulsions carrying imiquimod, CpG 2216, or poly(I:C). Preliminary studies did not show significant inhibition of tumor growth, but some mice did completely regress after treatment in the imiquimod (TLR7 agonist) and CpG2216 (TLR9 agonist) treatment groups. Additionally, tumor model selection is an important factor to evaluate, as the Hepa1-6 syngeneic tumor model tends to produce larger ulcers and necrosis in the center of tumor tissue, which is a good key factor for tumor-associated antigen production, but also leads to premature resections, as disease and unhealthy ulcers can affect overall animal health. The bleeding ulcers produced by the Hepa1-6 model led to premature resections of mice in the saline control group, thus biasing the treatment groups. Additionally, mice in the treatment groups were removed due to the development of bleeding ulcers at an early stage of tumor progression, which led to biases in judging the therapeutic efficacy of cancer vaccines. Overall, the preliminary experiments still provide us with some useful information, including the efficacy of different TLR agonists delivered intratumorally in vivo.

Next, we evaluated an alternative strategy in which defined tumor antigens were encapsulated with TLR agonists together or in combination into a comprehensive series of cancer vaccine constructs. The P1A peptide is a well-defined peptide antigen for murine J558 myeloma and has been reported to successfully trigger a protective response against J558 myeloma. Since P1A is a hydrophobic peptide, it can be successfully dissolved in the squalene oil core as a solvate. In addition to adding P1A to the nanoemulsion constructs, several TLR agonist combinations were tested, including MPLA (TLR4), resiquimod (TLR7/8), and CpG 2216 or SD-101 (TLR9). MPLA and resiquimod are lipophilic substances and can also be dissolved in the squalene core. On the other hand, CpG 2216 and SD-101 are oligonucleotide-based cargoes and must be encapsulated in lipid nanoparticles through electrostatic interactions. Combination protective therapy or cancer vaccines did show some promising results when three different TLR agonists were co-loaded into a single particle (rather than two or one of them). The results are interesting because cancer vaccines are injected subcutaneously and may have to boost the response, as the general concept suggests to reactivate memory immune cells. The data suggest that a single injection of cancer vaccines may not generate anti-cancer immunity or that other routes of administration may be needed.

In addition to the application of tumor-associated antigens as well as TLR agonists, we further considered delivering anti-PDL1 antisense oligonucleotides to downregulate PD-L1 surface expression of tumor cells to enhance immune responses, as achieved by anti-PDL1 or anti-PD1 antibodies as immune checkpoint blockade therapy. Immune checkpoint signaling requires the interaction between PD-L1 and PD-1 or PD-1 and CTLA-4 to inhibit the immune activation response of CD4 ⁺ and CD8 ⁺ T lymphocytes, the two major populations of anticancer immunity. Therefore, the TLR9 agonist SD-101 and anti-PDL1 antisense oligonucleotides were encapsulated in standard lipid nanoparticles, respectively, to enhance the delivery of oligonucleotides to cells and increase circulation time. However, the overall results of using anti-PDL1 ASO LNPs as monotherapy were not favorable, while SD-101 LNPs worked effectively as a single agent. In fact, lipid nanoparticles are designed to deliver cargo to the liver, as the liver is the main organ for the clearance of foreign substances. Due to the similar size, sinusoids and capillaries in the liver can capture nanoparticles. Most of the nanoparticles enter the liver, which is the essence of lipid nanoparticle physiology. In addition, we found that modifications to the ASO significantly affected efficacy, as previously reported and as assessed in vitro by comparing 2'-OMe-modified ASOs and LNA-modified ASOs. The results also suggest that the failure of anti-PDL1 ASO LNP monotherapy may be due to ineffective cargo carried in the lipid nanoparticles, as standard lipid nanoparticles with DLin-MC3-DMA are the gold standard for high delivery efficiency.

To find a solution and develop a delivery platform for anti-cancer immune system activation therapy, a pH-sensitive nanoemulsion-based lipid nanoparticle delivery platform was developed to accommodate hydrophobic and nucleic acid materials in a single construct. Several combinations were evaluated, including TLR agonist combinations or TLR agonists in combination with anti-PDL1 LNA ASOs. The preliminary results presented in this example are promising, with a high TGI of approximately 70%.

Example 4: Squalene emulsion co-loaded with ivermectin and resiquimod promotes systemic anti-tumor immunity.

Overview

Endosomal TLR agonists effectively function as anticancer agents by promoting the antigen presentation process in dendritic cells (DCs) and enhancing the maturation of CD8+T lymphocytes (or cytotoxic T lymphocytes, CTLs), which can rapidly recognize and kill cancer cells through T cell-mediated immunity. In this study, a squalene-based nanoemulsion (NE) formulation was developed to co-deliver the TLR7/8 agonist resimod (R848) and the antiparasitic drug ivermectin (IVM), which has been used worldwide since 1975. NE co-loaded with R848-IVM was developed and its stability was characterized. The antitumor activity of R848-IVM NE was also evaluated in vitro and in vivo. In vivo studies have shown that IVM can increase R848-induced immunogenic cell death and show strong antitumor activity, with tumor growth inhibition exceeding 80%. Significant HMGB1 release into the tumor microenvironment was observed in mice treated with NE co-loaded with R848-IVM. A more than 3-fold increase in Cd8a expression was also observed in tumor tissues. The results suggest a potential combination therapy against solid cancers by systemic co-delivery of IVM with other immunostimulants.

introduce

Toll-like receptors (TLRs) play a key role in immune responses by recognizing pathogen-associated molecular patterns (PAMPs), subsequently inducing cytokine production and activating adaptive immunity. TLRs are expressed on the cell surface (TLR1/2/4/5/6/10) or endosomal surface (TLR3/7/8/9) of antigen presenting cells (APCs), such as dendritic cells (DCs) or macrophages. TLRs are ready to recognize foreign molecular patterns, initiate the MyD88/NF-κB transduction pathway, and activate the naive T cell pool in the adaptive immune system. Studies have shown that endosomal TLR agonists are effectively used as adjuvants in cancer vaccines due to their powerful immunostimulatory capacity and anti-tumor efficacy. Endosomal TLR agonists have been shown to promote the antigen presentation process in DCs and enhance the maturation of CD8+T lymphocytes (or cytotoxic T lymphocytes, CTLs), which can ultimately inhibit cancer cell growth through T cell-mediated immunity. The anti-tumor activity achieved by endosomal TLR agonists suggests that cell-mediated immunity derived from TLR activation may be beneficial for anti-cancer therapy when combined with personalized antigen-based therapy, immune checkpoint blockade, chemotherapy, or radiotherapy. The FDA has approved three TLR agonists for cancer treatment, including Bacillus Calmette-Guérin (TLR2&4 agonist mixture), monophosphoryl lipid A (TLR2/4 agonist), and Imiquimod (TLR7 agonist). Overall, the clinical results of TLR agonists are mixed. Intratumoral or intradermal injection of TLR agonists has recently been studied. However, these routes of administration are difficult to obtain during clinical practice for most solid tumors.

During TLR agonist anticancer therapy, immunogenic cell death (ICD) plays an important role in antitumor immunity by rapidly inducing surface exposure of high mobility group protein B1 (HMGB1), calreticulin, and release of tumor antigens to immune cells. Chronic exposure of these damage-associated molecular patterns (DAMPs) will recruit DCs and promote CTL activation through antigen presentation. Activated DCs and CTLs will accelerate the phagocytosis of antigen components in the tumor microenvironment (TME) and produce long-term antitumor immunity. Patients treated with traditional chemotherapy show ICD-mediated antitumor immunity, with an increased ratio of CTLs to regulatory T cells (Tregs) in their TME. ICD-mediated antitumor responses can also be enhanced by immune checkpoint blockade, which prevents immune escape between tumor cells and immune cells. However, chemotherapy is often associated with high cytotoxicity, and even when applied locally to reduce systemic side effects, it can cause dose-dependent damage to normal cells. On the other hand, immune checkpoint blockade has become a revolutionary method for activating patients' immune systems to treat cancer. However, the anticancer efficacy of immune checkpoint inhibitors is limited to "hot tumors" with a large number of tumor-infiltrating immune cells, while "cold tumors" with limited immune cell infiltration are less responsive to immune checkpoint inhibitors. Therefore, immunomodulatory therapies with enhanced systemic immune activation, rapid induction of ICDs, and low toxicity would be an ideal approach for systemic anticancer treatment.

Resiquimod (R848) is a TLR7/8 agonist that has been shown to have desirable antitumor activity in murine tumor models. However, monotherapy with R848 is insufficient to induce a systemic immune response against tumors. Ivermectin (IVM) is an antiparasitic drug used worldwide since 1975. Studies have shown that IVM has the potential to alter the release of HMGB1 and calreticulin in the TME, making it an ideal candidate for inducing ICD in addition to R848 treatment. Nevertheless, due to the limited solubility of R848 and IVM, injectable formulations are required for clinical translation. Oil-in-water nanoemulsions (NEs) are considered to be promising non-viral delivery systems for hydrophobic drugs. NEs consist of a surfactant-encapsulated oil core, which can be used as an effective reservoir to dissolve poorly water-soluble drugs. Our above work demonstrates a squalene-based NE encapsulating R848, a TLR7/8 agonist, which shows moderate antitumor activity by systemic administration.

In this example, squalene-based NE was developed to co-encapsulate R848 and IVM. Squalene NE greatly increased the solubility of R848 and IVM in aqueous solution, which makes hydrophobic drugs useful for systemic administration. Squalene-based NE eliminated autophagy-associated cytotoxicity in IVM, but maintained its ability to promote ICD. NE co-loaded with R848-IVM showed high stability when stored at 4°C and -20°C. IVM NE treatment successfully induced the release of HMGB1 from TME and increased Cd8a mRNA expression in tumor tissues. The anti-tumor efficacy of NE co-loaded with R848-IVM (R848-IVM NE) was superior to that of R848NE or IVM NE, indicating the potential for combined therapy with TLR agonists and IVM.

Materials and methods

Materials. Squalene was purchased from Sigma-Aldrich (St. Louis, MO, USA). 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC) was purchased from Avanti Polar Lipids (Birmingham, AL, USA). Polysorbate 80 (Tween 80) was purchased from Fisher Scientific (Hampton, NH, USA). R848 and IVM were purchased from MedChemExpress (Monmouth Junction, NJ, USA) and Sigma-Aldrich, respectively. Any other reagents, including but not limited to buffers, can be purchased from Fisher Scientific.

R848-IVM NE formulation and characterization. Squalene-based NE was prepared by manual rapid injection of the oil-lipid mixture into phosphate-buffered saline (PBS). Squalene, DOPC, and Tween80 were prepared in ethanol at a molar ratio of 1/1/1. R848 and IVM were then added to the lipid-ethanol solution alone or in combination, keeping the lipid to R848/IVM ratio at 10:1 (w/w). The final lipid concentration of the NE was 8 mg/mL, and the final drug concentration of both was 0.8 mg/mL. Particle size was measured by dynamic light scattering (DLS) using a NICOMPNANO ZLS Z3000 (Entegris, Billerica, MA, USA). Empty NE was generated using the same procedure in the absence of R848 or IVM. Empty NE and R848-IVM NE were stored at 4°C prior to characterization and at -20°C for long-term stability testing. Sepharose CL-4B size exclusion chromatography was performed to detect the encapsulation efficiency of R848 or IVM in squalene nanoemulsions. Drug concentrations were determined by UV-visible spectroscopy at 320 nm (R848) or 245 nm (IVM) using a NanoDrop 2000 spectrophotometer. The loading efficiency was determined by the following equation:

In the saturation solubility solution study, R848 or IVM was dissolved in PBS or prepared in NE at 2mg/ml and incubated at room temperature for 24 hours. Insoluble solid drug was precipitated by centrifugation. Supernatant was collected for concentration analysis. R848 concentration was quantified by UV-visible spectroscopy at 320nm using NanoDrop 2000 spectrophotometer. Using an isocratic mobile phase composed of water/methanol/acetonitrile (8:36:56, v/v/v), a C18 column (Kromasil150-C-18, 4.6x150mm) and PDA detection at 230nm, IVM concentration was quantified by high performance liquid chromatography (HPLC).

Cell culture.RAW 264.7 murine macrophage cell line and MC38 murine colorectal cancer cell line were gifts from Dr. Peixuan Guo and Dr. Christopher Coss, College of Pharmacy, Ohio State University, respectively.RAW 264.7 and MC38 were grown in DMEM supplemented with 10% FBS and 1x antibiotic-antimycotic. Cells were maintained at 37°C and grown in a humidified atmosphere containing 5% _CO2 .

Cell viability assay. MC38 cells were seeded at 3000 cells/well in 96-well plates 24 hours before treatment. Cells were treated with increasing concentrations of R848, IVM, R848 NE, or IVM NE from 1 μM to 400 μM. A separate experiment was set up to evaluate the potential cytotoxicity of empty NE. Empty NE was treated with concentrations ranging from 10 μg/mL to 4000 μg/mL. After 72 hours of treatment, cell viability was checked by CellTiter 96R AQueous One Solution (Promega, Madison, WI) according to the manufacturer's protocol. The IC50 of R848 or IVM was determined by R programming.

In vitro gene regulation of R848 NE, IVM NE and R848-IVM NE. MC38 cells were seeded in 6-well plates at 3x 105 cells/well 24 hours before treatment. Cells were treated with R848, IVM, R848 NE, IVM NE or R848-IVM NE in complete medium and incubated for 24 hours. Both R848 and IVM were treated with 8 μM as free drug solution or in squalene NE. Total RNA was extracted using TRI reagent (Zymo Research) according to the manufacturer's protocol. cDNA was prepared by high-capacity cDNA reverse transcription kit (Invitrogen, Waltham, MA, USA), and real-time qPCR (RT-qPCR) was performed on QuantStudio 7FlexReal-time PCR system using SsoAdvancedTM UniversalSYBRR Green Supermix (Bio-Rad Laboratories, Hercules, CA). RT-qPCR primers for mouse calreticulin, Hmgb1, Lc3b, and Actb were purchased from Sigma-Aldrich. Actb was selected as a housekeeping gene control. The relative amounts of RNA levels were calculated and compared according to the 2-ΔΔCt method.

In vitro tumor cell migration assay. Scratch healing model was performed to examine the migration ability of MC38 cells after treatment. 24 hours before treatment, MC38 cells were seeded in 6-well plates at a density of 5x 105 cells/well. A scratch was made on the well using a 10ul pipette tip immediately before treatment. The cells were washed with PBS and incubated with complete medium containing 8 μM R848, IVM, R848NE, IVM NE or R848-IVM NE. Cells were propagated at 37°C for 24 hours. The distance between the wound edges was measured by a Nikon Eclipse Ti-S microscope (Nikon, Tokyo, Japan).

In vivo antitumor activity. The MC38 mouse syngeneic colorectal cancer model was generated by inoculating 1 x 10 ⁶ cells/mouse in PBS on the right flank of C57BL/6N mice (Charles River Laboratories). Treatment was initiated once the tumor size reached approximately 100 mm3. Mice (n=5) were treated intraperitoneally with saline, 4 mg/kg R848 NE, 4 mg/kg IVM NE, or R848-IVM NE (4 mg/kg R848 and 4 mg/kg IVM) every 3 days for a total of 3 times. Tumor growth and body weight were monitored, and tumor volume was calculated according to the following equation:

All mice were maintained and treated according to the guidelines of the Institutional Animal Care and Use Committee (IACUC) of The Ohio State University. All groups were euthanized on the 9th day, and whole blood samples were collected by cardiac puncture. Mouse serum was collected by placing the whole blood samples at room temperature for 30 minutes, followed by centrifugation at 2000 x g for 20 minutes at room temperature. Before cytokine quantification, samples were stored at -80 ° C. Mouse TNF-a and IL-6 concentrations were determined by TNF-a and IL-6 mouse uncoated ELISA kits (Invitrogen, Waltham, MA, USA) according to the manufacturer's protocol. Tumor and spleen tissues were collected and weighed for comparison. Spleen weight was normalized to individual body weight for comparison between treatment groups. Tumor growth inhibition (% TGI) on day 10 was determined by the following equation:

Wherein _T10 represents the mean tumor volume of the treatment group on day 10, _T0 represents the mean tumor volume of the treatment group on day 0, _C10 represents the mean tumor volume of the control group on day 10, and _C0 represents the mean tumor volume of the control group on day 0. %TGI>50% was considered significant.

In vivo gene regulation of R848 NE, IVM NE and R848-IVM NE. Tumor and spleen tissues were harvested immediately at the end of treatment. The spleen was homogenized into a single cell suspension by squeezing a 70 μm cell strainer with a syringe plunger. The spleen cells were washed and resuspended in FACS staining buffer at 2 x 10 ⁷ cells/mL. 1 x 10 ⁶ spleen cells were saved for flow cytometric analysis. The remaining spleen cells were lysed in TRI reagent to extract total RNA. Tumor tissue was directly homogenized in TRI reagent by probe sonication. Total RNA was isolated according to the manufacturer's protocol. RT-qPCR was completed according to the procedures described in Section 2.5. The primer sequences for real-time qPCR assays of mouse Cd3e, Cd4 and Cd8a are listed below.

In vivo protein expression analysis by Western blotting. Tumors were harvested and homogenized in Pierce RIPA buffer (Thermo Fisher Scientific) using a handheld homogenizer. Total protein was extracted after incubation on ice for 30 minutes, and protein samples were denatured and quantified by centrifugation at 14,000 x g for 30 minutes at 4°C, and equal amounts of protein were loaded and electrophoresed on 10% SDS-polyacrylamide gels and transferred to nitrocellulose membranes. The transfer membranes were blocked with Tris-buffered saline (TBS) containing 5% skim milk. HMGB1 rabbit monoclonal antibody and anti-rabbit HRP-conjugated secondary antibody were purchased from Cell Signaling (Danvers, MA, USA). Ki-67 rabbit monoclonal antibody was purchased from Thermo Fisher Scientific. Specific target protein bands were generated using an enhanced chemiluminescence (ECL) detection system and total protein was detected using Ponceau S as a control.

Flow cytometry. APC/Cyanine7 anti-mouse CD3e (145-2C11), FITC anti-mouse CD4 (RM4-5), PE/Cyanine7 anti-mouse CD8a (53-6.7), PE anti-mouse FOXP3 (MF-14), and a eukaryotic transcription factor buffer set for FOXP3 staining were purchased from Bio-Legend (San Diego, CA, USA). Single cell suspensions of splenocytes in FACS staining buffer were stained according to the manufacturer's protocol. Stained cells were analyzed using an LSR II flow cytometer in the Flow Cytometry Shared Resources (FCSR) at the Ohio State University Comprehensive Cancer Center.

Statistical Analysis. All studies were performed in triplicate. Unless otherwise stated, data are presented as mean ± standard deviation. Statistical analysis will be performed using Microsoft Excel. One-way ANOVA was used to determine the mean differences between two or more treatment groups. Student's t-test was used as a post hoc analysis to determine the statistically significant differences between any two groups. A p-value of 0.05 was selected as the critical value for statistical significance.

Results and discussion

Particle properties and solubility.IVM is a broad-spectrum antiparasitic drug targeting many internal and external paratopic sites. Recent studies have shown that IVM exhibits certain antitumor activity in various types of tumors through multiple pathways. However, it has been reported that when high doses of IVM are administered, there are serious side effects. Previous studies have introduced different lipid-based nanoparticles to deliver IVM against parasites. However, the systemic delivery platform for anti-tumor IVM has not been well established. In our previous examples, squalene-based NE was able to encapsulate R848 and achieved great potential for eliminating tumor growth by systemic administration in tumor-bearing mice when used in combination with other TLR agonists. Here, squalene-based NE is able to co-encapsulate IVM and R848. The size of R848-IVM NE is about 140-160nm, which is larger than IVM NE or R848 NE (Figure 53A). Nevertheless, according to published studies, the particle size of R848-IVM NE is considered suitable for cellular uptake. R848-IVM NE also exhibited high colloidal stability when stored at 4°C and -20°C for more than 6 months, with an encapsulation efficiency (%) of 21.77±2.10 (STD) for R848 232 and 22.80±0.13 (STD) for IVM. Compared with the solubility in PBS solvent, squalene-based NE successfully increased the solubility of R848 by 2-fold and the solubility of IVM by 100-fold ( FIG. 53C ), which provides an effective systemic delivery platform and can expand the indications of R848 and IVM in the clinic.

In vitro cell viability. Although most TLR7/8 agonists (such as R848) have no direct cytotoxic effects in vitro or in vivo, many reports show that IVM produces significant toxicity through autophagy and DNA damage. At the highest dose of 4000 μg/ml, empty NE treatment in MC38 cells did not show significant cytotoxicity. Similarly, analysis by MTS assay, treatment with free R848 and R848 NE did not result in significant cytotoxicity (IC50>100 μM) (Figure 54A). However, compared with free IVM (Figure 54B) with an IC50 of 10.94±4.15 μM (SEM), the IC50 of IVM NE was higher, at 43.52±23.53 μM (SEM), indicating that squalene-based NE successfully reduced the cytotoxicity produced by IVM.

In vitro gene regulation of free drugs and NE preparations. The process of immunogenic cell death in tumors is mainly mediated by DAMPs released by cancer cells, which include surface exposure of calreticulin, HMGB1, and release of type I interferon (IFN). DAMPs are further recognized by APCs and CTLs to induce antitumor immunity. Many studies have shown that calreticulin translocation and exposure are two important components of the ICD checkpoint, and knockdown of calreticulin completely abolishes immunogenicity in tumors during the ICD process. In addition, HMGB1 has been shown to promote ICD of the extracellular compartment, including triggering tumor necrosis factor-α (TNFa) release, APC maturation, and CTL recruitment. However, endogenous calreticulin and HMGB1 exhibit controversial roles in cancer progression. Endogenous HMGB1 has been shown to promote cancer cell proliferation, and CRT has a pro-angiogenic function due to its ability to promote the expression of vascular endothelial growth factor (VEGF), leading to cancer cell proliferation and migration.

In MC38 cells, calreticulin mRNA levels decreased slightly after R848 NE treatment, indicating a potential for cell growth inhibition (Figure 55A). No significant Hmgb1 mRNA regulation was observed in R848 NE treatment, but free R848 showed downregulation of Hmgb1 (Figure 55A). IVM and IVM NE did not show significant calreticulin regulation at the mRNA level, while IVM NE significantly downregulated Hmgb1 (Figure 55B), suggesting a novel anti-tumor pathway for IVM NE through inhibition of cell proliferation.

IVM has been shown to trigger autophagy-mediated cell death by blocking the PAK1/Akt axis and generating LC3-II in autophagosomes, where LC3 is a key protein involved in initiating autophagy. In this study, squalene-based NE alleviated IVM-induced autophagy-mediated cell death, as determined by the reduction of Lc3b mRNA expression to normal levels in IVM NE treatment compared to free IVM treatment (Figure 55B). The results suggest that IVM NE would be suitable for systemic administration with low cytotoxicity, which is consistent with the cell viability results of free IVM and IVM NE (Figure 54B).

Wound healing. Wound healing studies were performed to monitor the relative mobility of MC38 after treatment with R848 NE, IVM NE or R848-IVM NE (Figure 56A). R848 NE does not significantly reduce cell migration in the wound area because R848 does not have any inhibitory effect on cell growth, but only has immunostimulatory properties. The motility of cells treated with IVM NE was reduced to 37.57% ± 8.33%, and the mobility of cells treated with R848-IVM NE was reduced to 19.37% ± 4.89% (Figure 56B). The reduction in cell motility may be due to the significant downregulation of Hmgb1 in IVM NE and R848-IVM NE treatment (Figure 56D), which is related to the established role of HMGB1 in promoting cell proliferation. The lower motility in R848-IVM NE may be attributed to the inhibition of addictive cell growth by calreticulin (no statistical significance) and the downregulation of Hmgb1 mRNA in NE by R848 and IVM. R848-IVM NE treated at 2 μM, 4 μM and 8 μM also exhibited concentration-dependent locomotion inhibition of 43.85%, 40.32% and 19.37%, respectively ( FIG. 56C ).

In vivo anti-tumor activity. In animal studies evaluating the anti-tumor efficacy of R848-IVM NE, 4 mg/kg R848NE and IVM NE were administered alone or in the form of R848-IVM NE to show combined anti-tumor efficacy by intraperitoneal administration. Compared with saline controls, moderate anti-tumor efficacy of R848 NE treatment was observed and consistent with our previous results. Studies have shown that IVM enhances the ICD process and promotes anti-tumor immunity of immune checkpoint inhibitors. However, no significant anti-tumor effect was observed in mice treated with IVM NE compared with saline controls, which is consistent with previous studies on IVM (Figure 57, C Figure, F Figure-H Figure). Nevertheless, compared with R848 NE and saline controls, strong anti-tumor effects were observed in mice treated with R848-IVM NE (Figure 57, F Figure-H Figure).

At the end of the study, the tumor growth inhibition (TGI%) of R848-IVM NE reached 88.66% ± 14.91%, which was superior to R848 NE and IVM NE treatment (Table 4). No significant weight loss was observed during the treatment regimen, indicating that R848-IVMNE treatment had less systemic toxicity. Mice treated with R848 NE or R848-IVM NE showed mild splenomegaly. The increase in spleen weight was caused by immune activation achieved by R848. No significant changes in spleen weight were observed in mice treated with IVM NE. No significant changes in serum cytokine levels were observed.

Table 4. Tumor growth inhibition (TGI %) at day 10 by R848 NE, IVM NE and R848-IVM NE.

therapy group Average TGI% standard deviation R848NE 64.33 10.75 IVM 43.27 22.12 R848-IVM NE 88.66 14.91

Effects of R848 and IVM on gene regulation and immune cell populations in vivo. Although no significant increase in cytokines in serum was observed in mice treated with R848 NE, IVM NE, or R848-IVM NE in 359, RT-qPCR results of tumor tissues showed that IVM NE and R848-IVM NE induced significant upregulation of Hmgb1 mRNA (Figure 58A). At the same time, Western blot results showed that lower HMGB1 was detected in tumor tissues collected from mice treated with IVM NE and R848-IVM NE, indicating that Hmgb1 mRNA is overexpressed in the TME but HMGB1 protein is released into the extracellular compartment during the ICD process, which is consistent with previous studies on the regulation of HMGB1 by IVM treatment. Significant downregulation of calreticulin mRNA and low Ki67 protein levels were observed in tumor tissues of mice treated with R848-IVM NE ( FIGS. 58A and 58D ), indicating that R848-IVM NE treatment generated a TME that was unfavorable for proliferation.

Flow cytometry data showed that after treatment with R848 NE and R848-IVM NE, the CTL population in the spleen of mice was reduced (Figure 58B and Figure 58C), which was not related to the immune activation effect of R848 in splenomegaly. However, a significant increase in Cd8a mRNA levels was observed in tumor tissues collected from mice treated with R848-IVM NE (Figure 58A), which may indicate that R848-IVM NE treatment can successfully induce CTL and/or NK cells to infiltrate from the spleen into the TME. The upregulation of Cd8a mRNA was also attributed to IVM, which was shown in tumor tissues of the IVM NE monotherapy group (Figure 58A).

in conclusion

Although ICD-mediated antitumor immunity has been shown to promote antitumor responses in many cancer patients by conventional chemotherapy and immune checkpoint blockade, chemotherapy is often associated with undesirable side effects and toxicity, which can cause dose-dependent damage to normal cells even when administered locally to reduce drug concentrations within the systemic circulation. On the other hand, the anticancer efficacy of immune checkpoint inhibitors is limited to tumors that have been infiltrated by T cells, while tumors with less T cell infiltration are less responsive to immune checkpoint inhibitors. This study shows a promising strategy to enhance the process of ICD and thus achieve systemic antitumor activity by using a combination therapy of squalene-based NE co-encapsulating R848 and IVM. R848-IVM NE was highly stable during long-term storage at 4°C and -20°C. Squalene-based NE greatly reduced the cytotoxicity produced by IVM. Ultimately, R848-IVMNE strongly enhanced the in vivo antitumor activity by greatly inhibiting tumor growth by more than 80%. Our results suggest that intraperitoneal administration of R848-IVM NE may be a promising strategy to induce ICD and recruit CTL and NK cells to the TME, suggesting a broader application of IVM in inducing ICD in combination with TLR agonists or other anticancer immunotherapeutics.

Example 5: pH-sensitive nanoemulsion (PSNE) for delivery of nucleic acid cargo

PNSE contains ingredients similar to conventional LNPs (e.g., ionizable lipids, neutral lipids, cholesterol, and PEG-lipids). In some embodiments, PNSE may also contain a fusion oil, such as squalene or squalene, to promote membrane fusion and endosomal release of nucleic acid (NA) drugs. The addition of fusion oil can increase the intracellular delivery efficiency of PSNE cargo.

NA delivery based on PSNE or LNP is largely affected by the surface properties of the particles. Because lipid particles include ionizable lipids (e.g., DLinMC3DMA, SM-102 or ALC-0315), the pH of the composition (and the pKa of the ionizable lipids) may significantly affect the delivery efficiency. However, previously reported formulations all have a neutral or slightly alkaline final pH. For example, the pH of Patisiran (LNP and siRNA) is about 7; the pH of Moderna and Pfizer coronavirus vaccines are both (7-8 or 7.5). Within these pH ranges, most of the ionizable lipids are uncharged, resulting in very low zeta potentials of the particles. Since the electrostatic interactions between the particles and the negatively charged cell surface are negligible, this in turn leads to minimal cellular uptake, as well as low delivery efficiencies in vitro and in vivo at physiological extracellular pH (7.4).

In some embodiments, PSNE can be buffered at an acidic pH (eg, a pH less than 6.5, such as a pH of 4 to 6.5, or a pH of 5 to 6.5), which results in a significant increase in cellular delivery efficiency in vitro and in vivo.

Materials and methods

Squalene was purchased from Sigma-Aldrich (St. Louis, MO). DOPC and DOPE were purchased from Avanti Polar Lipids (Alabaster, AL). DODMA and DMG-PEG2000 were purchased from NOF America (White Plains, NY). DLinDMA and DLin-MC3-DMA were purchased from MedChemExpress (Monmouth Junction, NJ). Any chemicals or buffers not otherwise specified were purchased from Fisher Scientific (Hampton, NH).

DOPC, DOPE, squalene, ionizable lipids (including but not limited to DODMA and DLin-MC3-DMA) and DMG-PEG2000 are mixed into the lipid ethanol mixture at a molar ratio of 15:28:10:45:2. Next, the lipid ethanol mixture is quickly injected into an acidic phosphate buffer to form an empty pH-sensitive nanoemulsion. At the same time, nucleic acid cargo, including but not limited to messenger RNA (mRNA), small interfering RNA (siRNA) or antisense oligonucleotide (ASO), is dissolved in DEPC water at a desired concentration. Before mixing, the empty pH-sensitive nanoemulsion and nucleic acid cargo are heated to 60°C. Then, the cargo is added dropwise to the slowly vortexed empty pH-sensitive neutral nanoemulsion at a weight ratio of 1 to 15-25 until lipid nanoparticles encapsulating nucleic acids are formed. The product is incubated at 37°C for 10 minutes and stored at 4°C before use. The particle size and zeta potential (ζ) of the nucleic acid-loaded pH-sensitive nanoemulsions were analyzed by dynamic light scattering using a NICOMP Z3000 Nano DLS/ZLS system (Entegris, Billerica, MA).

Cells were maintained in Dulbecco's Modified Eagle Medium (DMEM) or Minimum Essential Medium (MEM ₎ supplemented with L-glutamine (2 mM), 10% FBS, sodium pyruvate (1 mM) and antibiotic-antimycotic (Thermo Fisher Scientific, Pittsburg, PA) at 37°C and 5% CO2. Cells were homogenized in TRI reagent and total RNA was extracted according to the manufacturer's protocol. cDNA was prepared using a high-capacity cDNA reverse transcription kit according to the manufacturer's protocol. Real-time PCR was performed on a QuantStudio 7Flex Real-Time PCR system. The relative amounts of RNA levels were calculated and compared according to the 2 ^-ΔΔCt method.

C57BL/6 and ICR mice were purchased from Charles River Laboratory. Animals were housed in a temperature-controlled room and fed a normal diet under a 12-hour light/12-hour dark cycle and Helicobacter-free conditions. All animal studies were reviewed and approved by the Institutional Animal Care and Use Committee of The Ohio State University. Both male and female mice were used in the experiments.

In vivo mRNA expression was quantified by measuring firefly luciferase bioluminescence. 5 minutes before bioluminescence measurement, treated mice injected with IM FFLuc mRNA lipid nanoparticles were injected intraperitoneally with luciferin. Bioluminescence images were taken at optimal exposure settings by an in vivo imaging system (IVIS).

result

Evaluation of squalene as a fusion oil. In order to compare the fusibility of different oils, a simple anionic nanoemulsion was prepared. The composition evaluated was: oil/DOPC/CHEMS (30:40:40 mol/mol). The oil was selected from squalene, medium chain triglycerides (MCT) from Abitec (Captex 300), long chain triglycerides (LCT, soybean oil from Sigma-Aldrich). DOPC: dioleoylphosphatidylcholine, CHEMS: cholesterol hemisuccinate. Various formulations were prepared in PBS at pH 8.0, then diluted into various buffers with different pH values and incubated at room temperature for 5 minutes, followed by OD measurements. OD500 and OD600 were measured to indicate particle size, with larger particle sizes indicating increased fusibility.

The results are shown in Figures 59A and 59B. As shown in Figures 59A and 59B, squalene exhibited improved fusibility compared to other representative oils (medium chain triglycerides and/or long chain triglycerides).

In vitro activity of PSNE loaded with siRNA against Akt-1 in KB cells. In this proof-of-principle study, PSNE preparations loaded with siRNA showed greater downregulation of the target gene Akt-1 when cells were cultured at acidic pH (5.5 and 6.5) compared to culture at a typical culture medium pH of 7.4 (see Figure 60). These results have been confirmed using a variety of siRNA and mRNA cargoes and are highly reproducible.

Evaluation of PSNE nucleic acid delivery in vivo. PSNE formulations loaded with luciferase mRNA reporter were used to evaluate the efficacy of PSNE formulation delivery in vivo. PSNE loaded with FFLuc mRNA buffered at pH 6 was administered to mice by intramuscular injection. For comparison, mice were similarly treated with FFLuc mRNA alone (naked mRNA), FFLuc mRNA encapsulated in LNPs, and FFLuc mRNA encapsulated in pH-sensitive micelles (PSMs).

Figures 61A-61E show in vivo bioluminescence imaging of mice treated with intramuscular injections of FFLuc mRNA (naked mRNA), FFLuc mRNA encapsulated in LNPs, FFLuc mRNA encapsulated in pH sensitive micelles (PSMs), and FFLuc mRNA encapsulated in PSNE (buffered at pH 6). As shown in Figures 61A-61E, the PSM and PSNE formulations were much more active than the LNP formulation, with the PSNE formulation showing the highest activity in terms of mRNA delivery.

Figure 62 shows the delivery provided by PSNE based on Dlin-MC3-DMA and DODMA carrying luciferase mRNA given to mice by intramuscular injection. High-level gene expression of PSNE was observed in 50mM pH 6 histidine buffer. However, when the preparation was adjusted to pH 7.4 using phosphate buffer, gene expression in vivo was no longer detectable. The result is quantitative in Figure 63, which includes a figure of bioluminescence increase observed after PSNE buffered in 50mM pH 6 histidine buffer and pH 7.4 of Dlin-MC3-DMA carrying luciferase mRNA. By reducing pH from 7.4 to 6, it was observed that the luminous intensity increased by 30-50x. These results show that the preparation for intramuscular delivery of mRNA has a strong pH dependence.

Similar results were observed in in vitro studies. As shown in Figure 64, the cellular uptake of fluorescently labeled PSNE in KB cells at different buffer pHs was shown to show that, by flow cytometry analysis, the cellular uptake levels at pH 5 and pH 6 were much higher (increased mean fluorescence intensity) compared to pH 7. This illustrates the mechanism of pH-dependent cellular uptake/delivery of PSNE.

There are no previous examples of LNP formulations with an acidic final pH for in vitro or in vivo applications. All previously reported formulations had a neutral or slightly alkaline pH for the final formulation, which roughly matches the physiological value of pH (7.4).

discuss

Without wishing to be bound by theory, it is believed that inducing a positive zeta potential on the lipid nanoparticles will increase cellular uptake due to electrostatic interactions with negatively charged cell surfaces. In the case of lipid-based delivery vehicles comprising ionizable lipids, this positive charge is only present at acidic pH. By buffering such compositions at acidic pH values, these compositions can exhibit improved cellular uptake.

In vivo, low pH formulations of PSNE or LNP (thus having a positive surface charge) can similarly interact with target cells (e.g., during intramuscular injection). Compared with formulations in buffers at pH 7-8, this can result in greatly improved intramuscular delivery efficiency. This is confirmed by luciferase mRNA reporter genes and intramuscular injections of mice in pH 6 histidine buffers, showing highly superior gene expression compared to the same formulations in pH 7 phosphate buffers. Alternatively, compared with pH 7-8 (when the particles are mostly uncharged), particles can be coated with plasma proteins to a greater extent at lower pH (e.g., during intravenous injection). Protein coatings that are rapidly achieved based on cationic surface charges will result in different biodistribution patterns, and adsorbed proteins can promote target cell uptake (e.g., binding of acidic albumin (pI 4.7) can promote albumin receptor-mediated endocytosis).

There are several important consequences of LNP or PSNE formulation in an acidic final buffer. First, the optimal pKa of ionizable lipids may now deviate significantly from previous reports. Due to the change in buffer pH, a pKa of 6.2-6.5 may no longer be the optimal pKa, and thus commonly available ionizable lipids (such as DODMA) may now show much higher activity than their use in previous studies (when the pH of the formulation was 7-8).

Practical applications include the following:

(1) Previously it was difficult to demonstrate NA delivery in vitro using formulations containing ionizable lipids due to low activity, making comparisons between ionizable lipids difficult. Using an acidic pH strategy, it is now possible to produce highly efficient delivery in vitro, even in cells that are difficult to transfect. This has potential applications in ex vivo transfection protocols, using ionizable lipids that are less cytotoxic than permanently charged lipids. Ex vivo NA delivery is indeed used for AAV production plasmid transfection, CAR-T, and gene editing. This concept of acidic pH transfection can be extended to ionizable polymers, such as polyhistidine and polyethyleneimine, by enhancing their in vitro gene transfer activity for ex vivo applications.

(2) For intramuscular and other local injection routes (intrathecal, intracerebral, intravitreal, etc.), acidic buffers (e.g., 50 mM histidine pH 6) can maintain local pH and facilitate the delivery of NA (e.g., mRNA) to PSNE or LNP into local tissues (e.g., muscle fibers). Clinical applications include mRNA vaccines for viruses and cancers (e.g., coronavirus vaccines).

(3) When an acidic buffer containing LNPs or PSNE is administered intravenously, the pH equilibrates with the plasma pH, however, before this, the particles have been coated with negatively charged plasma proteins (such as albumin), and this promotes cellular uptake in the liver or tumors and leads to increased tissue uptake and cellular delivery efficiency.

In summary, lipid nanoparticles comprising ionizable lipid components can be formulated in acidic buffers to improve delivery efficiency. In addition, we have also demonstrated that fusion oils (e.g., squalene) can be incorporated into lipid nanoparticles comprising ionizable lipids to improve delivery efficiency. These concepts can be used independently or in combination. Applications include preparations for in vitro, ex vivo, local injection in vivo, and systemic administration of active agents such as, but not limited to, nucleic acids.

Example 6: Device for rapid synthesis of ethanol-independent lipid nanoparticles

Lipid nanoparticles are typically synthesized by mixing lipids and nucleic acids dissolved in ethanol in an aqueous buffer at a volume ratio of 1:3. However, this process produces nanoparticles containing 25% ethanol. Removal of ethanol is typically accomplished by small-scale dialysis or large-scale diafiltration, a time-consuming process that is clearly beyond the capabilities of hospital pharmacies, thus precluding bedside or hospital pharmacy-based manufacturing.

In this embodiment, we integrated a cartridge of size exclusion resin into a syringe pump system for ethanol removal. Therefore, nanoparticles free of ethanol and buffer exchange are generated in real time in one step. This has never been achieved before and greatly simplifies the nanoparticle synthesis process, making it possible to generate nanoparticles at the bedside using a kit formulation or a prefilled syringe assembly. This will be a decisive advantage for personalized cancer vaccine applications that require small-scale and individualized synthesis of nanoparticles. Nanoparticle manufacturing can be achieved by pressing a button in a hospital pharmacy by a technician with limited training. This is crucial in the implementation of personalized cancer vaccine production. Tumor antigens encoded by different mRNAs can be produced in large quantities, and personalized nanoparticles are produced using the selected mRNA.

Figure 65 shows the design of a syringe assembly with an integrated size exclusion cartridge for online ethanol removal and buffer exchange. Nanoparticle formation and ethanol removal are achieved by pressing a single button using a preassembled disposable syringe-cartridge assembly. The syringe can be pre-filled or loaded on-site from a kit-based vial labeled for each syringe (each kit has a total of 4 vials). The system is capable of generating nanoparticles with high reproducibility. The mixer can simply narrow the fluid path via a double hub needle or an online micro mixer. The pump pushes the solution through the mixer element and then through the size exclusion cartridge column to produce ethanol-free nanoparticles by limiting the total volume to less than 75% of the total volume of the cartridge column. Ethanol is trapped in the dead volume of the assembly and the resin in the cartridge column. For example, an 8-mL cartridge column is capable of producing 6mL of ethanol-free nanoparticle product.

Figure 66A-Figure 66C is a photo showing 4 syringes + 1 cartridge assembly for instant production of ethanol-free nanoparticles. The cartridge shown is a Sorbent Technologies Flash cartridge with an 8-mL volume. In a typical run, the cartridge is filled with a Sephadex G-50 resin pre-balanced in PBS. The syringe is equipped with each 1.5-ml of mRNA, lipid or buffer solution. The pump is started by the start button, and 6mL of ethanol-free and buffer-exchanged nanoparticles (in PBS) are immediately produced in the receiving bottle. The particle size obtained by dynamic light scattering is 100-200nm. The yield (nucleic acid recovery rate compared to input) of the nanoparticles produced without ethanol with this device is 68.8%.

Thus, the “injection device” is essentially a pump-driven device that mixes an ethanol lipid solution with an aqueous nucleic acid solution for instant self-assembly of nanoparticles (via nanoprecipitation). The specific configuration that enables a 1:3 mixing ratio and pump setup is unique, convenient to operate, and compatible with standard syringe pump drivers and standard syringe kits. Integration with an online cartridge for instant ethanol removal is also a unique feature. Although nucleic acid recovery is limited to approximately 50%, the product is ethanol-free. Instant ethanol removal has never been achieved before and is critical for personalized nanoparticle manufacturing. The device is equally effective for the instant production of microparticles or microspheres. By making the mRNA reagents interchangeable while maintaining the remainder of the disposable components, the device can be used to manufacture personalized cancer vaccines. In addition, ethanol can be removed immediately without the need for dialysis or diafiltration, making on-site nanoparticle preparation feasible.

The scope of the compositions and methods of the appended claims is not limited by the specific compositions and methods described herein, which are intended to illustrate several aspects of the claims. Any functionally equivalent compositions and methods are intended to fall within the scope of the claims. In addition to the compositions and methods shown and described herein, various modifications of the compositions and methods are intended to fall within the scope of the appended claims. In addition, although only certain representative compounds, components, compositions, and method steps disclosed herein are specifically described, other combinations of compounds, components, compositions, and method steps are intended to fall within the scope of the appended claims even if they are not specifically listed. Therefore, the combination of steps, elements, components, or ingredients may be explicitly mentioned or less mentioned herein, however, even if not explicitly stated, other combinations of steps, elements, components, and ingredients are also included.

As used herein, the term "comprising" and its variations are used synonymously with the term "including" and its variations, and are open, non-limiting terms. Although the terms "comprising" and "including" are used herein to describe various embodiments, the terms "consisting essentially of" and "consisting of" can be used in place of "comprising" and "including" to provide more specific embodiments of the present invention and are also disclosed. Except as otherwise noted, all numbers used in the specification and claims to indicate geometric shapes, dimensions, etc. should at least be understood to be interpreted according to the number of significant digits and ordinary rounding methods, rather than attempting to limit the application of the doctrine of equivalents to the scope of the claims.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the disclosed invention belongs.Publications cited herein and the materials for which they are cited are specifically incorporated by reference.

SEQ ID NO:1

SEQ ID NO:2

Claims (30)

1. A pharmaceutical composition comprising lipid particles encapsulating an active agent, wherein the lipid particles comprise: one or more ionizable lipids; one or more neutral lipids; and one or more PEGylated lipids; wherein the composition is buffered at a pH of 5.0 to 6.5. 2. compositions as claimed in claim 1, one or more ionizable lipids are present in the lipid granule in an amount of 20 % by mole to 65 % by mole of the total components forming the lipid granule. 3. compositions as described in any one in claim 1 to 2, one or more neutral lipids are present in described lipid granule with the amount of 35 % by mole to 80 % by mole of the described total component forming described lipid granule. 4. compositions as described in any one in claim 1 to 3, one or more PEGylated lipids are present in the lipid granule with the amount greater than 0 % by mole to 5 % by mole of the described total component forming the described lipid granule. 5. A pharmaceutical composition comprising lipid particles encapsulating an active agent, wherein the lipid particles comprise: 20 mol % to 65 mol % of one or more ionizable lipids; 35 mol % to 80 mol % of one or more neutral lipids; greater than 0 mol % to 5 mol % of one or more PEGylated lipids; and 5 to 50 mole % of one or more fusion oils. 6. The composition of claim 5, wherein the composition is buffered at an acidic pH. 7. The composition of claim 6, wherein the acidic pH is from 5.0 to 6.5. 8. compositions as described in any one in claim 5 to 7, wherein said one or more fusion oils are present in said lipid granule with the amount of 10 % by mole to 40 % by mole of the described total component forming said lipid granule. 9. The composition of any one of claims 5 to 8, wherein the fusion oil comprises C12-C40 hydrocarbons containing less than 3 rings. 10. The composition of claim 9, wherein the C12-C40 hydrocarbon comprises an alkyl or alkylene chain. 11. The composition of claim 10, wherein the C12-C40 hydrocarbon comprises an alkylene chain optionally comprising at least one cis double bond. 12. The composition of any one of claims 5 to 11, wherein the fusion oil comprises squalene, squalane, pristane, pristene, farnesene, farnesane, retinol, phytol, carotene, tocopherol, tocotrienol, phytomenadione, menaquinone, wherein valency allows esters thereof, and combinations thereof. 13. The composition of any one of claims 5 to 12, wherein the fusion oil comprises squalene. 14. compositions as described in any one in claim 1 to 13, wherein said one or more ionizable lipids are present in described lipid granule with the amount of 30 % by mole to 50 % by mole of the described total component that forms described lipid granule. 15. A composition as described in any one of claims 1 to 14, wherein the one or more ionizable lipids comprise a lipid head group containing a tertiary amine. 16. A composition as described in any one of claims 1 to 15, wherein the one or more ionizable lipids comprise N,N-dimethyl-2,3-dioleyloxypropylamine (DODMA), [(4-hydroxybutyl)azanediyl]bis(hexane-6,1-diyl)bis(2-hexyldecanoate) (ALC-0315); 8-{(2-hydroxyethyl)[6-oxo-6-(undecanyloxy)hexyl]amino}octanoic acid 9-heptadecyl ester (SM-102), DLin-MC3-DMA; DLin-KC2-DMA; or any combination thereof. 17. compositions as described in any one in claim 5 to 16, wherein said fusion oil and said one or more ionizable lipids are present in said lipid particle with the molar ratio of 0.25:1 to 1:1. 18. compositions as described in any one in claim 1 to 17, wherein said one or more neutral lipids are present in said lipid granule with the amount of 30 % by mole to 50 % by mole of the described total component forming said lipid granule. 19. The composition of any one of claims 1 to 18, wherein the one or more neutral lipids comprise dipalmitoylphosphatidylcholine (DPPC), dioleoylphosphatidylethanolamine (DOPE), palmitoyloleoylphosphatidylcholine (POPC), egg phosphatidylcholine (EPC), distearoylphosphatidylcholine (DSPC), cholesterol, or any combination thereof. 20. compositions as described in any one in claim 1 to 19, wherein said one or more PEGylated lipids are present in said lipid granule with the amount that is greater than 0 % by mole to 10 % by mole of the described total component that forms said lipid granule. 21. The composition of any one of claims 1 to 20, wherein the one or more PEGylated lipids comprise PEG-di(tetradecyl)acetamide, PEG-myristoyl diglyceride, PEG-diacylglycerol, PEG dialkoxypropyl, PEG-phospholipid, PEG-ceramide, or any combination thereof. 22. compositions as described in any one in claim 5 to 21, wherein said fusion oil and said one or more PEGylated lipids are present in said lipid particle with the molar ratio of 5:1 to 20:1. 23. compositions as described in any one in claim 1 to 22, the mean diameter of wherein said lipid granule is less than 1 micron, such as 50nm to 750nm, 50nm to 250nm, 50nm to 200nm, 50nm to 150nm or 50nm to 100nm. 24. compositions as described in any one in claim 1 to 23, the polydispersity index (PDI) of wherein said lipid granules is less than 0.4. 25. The composition of any one of claims 1 to 24, wherein the active agent comprises a nucleic acid. 26. The composition of claim 25, wherein the nucleic acid comprises siRNA, mRNA, or any combination thereof. 27. A method of delivering an active agent to a cell, the method comprising contacting the cell with the composition of any one of claims 1 to 26. 28. A method for delivering an active agent to a cell in vivo, the method comprising administering to a mammalian subject a composition as claimed in any one of claims 1 to 26. 29. The method of claim 28, wherein the mammal is a human. 30. The method of any one of claims 28 to 29, wherein the administration is performed intravenously.