WO2025212420A1 - Self-assembled multilayered supraparticles - Google Patents

Abstract

A supraparticle includes and can consist of, ultrasmall Au nanoparticles, a stabilizing matrix, and a lipidoid. The supraparticle can advantageously include a labile molecule payload. A method is disclosed to self-assemble the components into ordered multi-unit structures. The method provides a self-limiting assembly process that results in finite-size supraparticles that can store and transport sensitive labile molecule payloads (for example, labile RNAs) and release them selectively upon exposure to biochemical-stimuli within cells when introduced into cells.

Description

SELF-ASSEMBLED MULTILAYERED SUPRAP ARTICLES

PRIORITY CLAIM AND REFERENCE TO RELATED APPLICATION

[0007] The application claims priority under 35 U.S.C. §119 and all applicable statutes and treaties from prior United States provisional application serial number 63/573,648, which was filed April 3, 2025.

FEDERALLY SPONSORED RESEARCH AND DEVELOPMENT

[0008] This invention was made with government support under grant no. EB032869 awarded by tire National Institutes of Health. The government has certain rights in the invention.

FIELD

[0009] Fields of the invention include nanoparticles and complex structures including nanoparticles, as well as sensitive labile molecule payloads.

BACKGROUND

[0010] Supraparticles smaller than 300 nm in size and made up of individual subunits that are less than 3 nm, arranged in a layered or periodic pattern, and responsive to stimuli, can provide precisely targeted delivery' and favorable biodistribution in biological systems. Supraparticles are excellent vehicles for storing, protecting, and delivering delicate payloads such as RNA. Delicate RNAs can be stored and transported without degradation. Intricate structural design of the supraparticles ensures the safe transportation of RNA and other sensitive materials. This can provide great benefit, for example, to the gene therapy field. [0011] While significant progress has been made in managing the size, shape, and composition of supraparticles, achieving control over internal architecture remains a significant challenge. Currently available supraparticles use protein matrices to dictate the final assembly. The protein matrices make the synthesis complex, time-consuming, non- aqueous, and low yielding. Only a few supraparticle formation methods have demonstrated self-limiting formation.

[0012] Another well-known drawback with currently available supraparticles is they utilize large subunit-nanoparticles. This makes long-term biocompatibility questionable along with eventual clearance from the system.

[0013] At least one attempt has been made to create silicon microparticles with embedded 90nm traditional liposome nanoparticles containing siRNA. Silicon microparticles with embedded LNPs and surface conjugated with polymeric NPs are described in another publication. Hie supraparticles are described as having an approximate 2.6 pm size with 87 nm subparticles. This published process requires photolithography, etching, and sonication. See, e.g., Mi, Y. etal. “A Micro/Nano Composite for Combination Treatment of Melanoma Lung Metastasis,” Adv. Healthc. Mater. 5, 936-946 (2016.). The semiconductor-style fabrication is challenging to include the siRNA without damaging the delicate payload.

[0014] Very few publications describe supraparticles that carry labile molecule payloads. One publication describes AuNPs crosslinked by MMP-2/9-sensitive peptide. The size was about 75 nm with 5 nm subunits, and the payload was TRA-DM1. See, Liu, Y. et al. “A triple enhanced permeable gold nanoraspberry designed for positive feedback interventional therapy.” J. Controlled Release 345, 120-137 (2022). Another publication describes PFOA functionalized Nanodiamonds. Tire size was about 90 nm with 10 nm subunits. The payload was Cisplatin. See. Yu, Y. et al. Self-assembled nanodiamond supraparticles for anticancer chemotherapy. Nanoscale 10. 8969-8978 (2018). Another publication describes Dodecanethiol stabilized AuNPs within Hydrophobin film. The particle size was about 100 nm with 4 nm subunits. Tire payload was Cisplatin. Maiolo, D. et al. “Bioreducible Hydrophobin- Stabilized Supraparticles for Selective Intracellular Release.” ACS Nano 11, 9413-9423 (2017). Hexanethiol AuNPs are described with an approximate 200 nm size and 4 nm subunit in another publication. Wang, Y. et al. ‘Host-guest chemistry with water-soluble gold nanoparticle supraspheres.” Nat. Nanotechnol. 12, 170-176 (2017). Poly(bile acid) modified AuNPs with a 150 nm size and 5 nm subunit are described in another publication. The payload was Thiofavin. T. Sun, J. , Li, W., Xiao, L., Yu, G. & Shi, J. “Main chain poly(bile acid) directed plasmonic nanospheres with amphiphilic binding pockets and photo-triggcrcd destruction.” RSC Adv. 6, 62200-62207 (2016). BS2G crosslinked GSHDE-capped gold nanoclusters with a size of about 249 nm and 2 nm subunits are discussed in another publication. Tire payload was SQ740 Raman dye. Yu, J. H. et al. “Highly Excretable Gold Supraclusters for Translatable In Vivo Raman Imaging of Tumors,” ACS Nano 17, 2554-2567 (2023). Finally, a publication discloses 1-cysteine stabilized FeS2 and Fe2O3 NPs. The approximate size was 75 nm with a 4.5 nm subunit. The pay load was plasmid. Turali- Emre. E. S. et al. “Self-Organization of Iron Sulfide Nanoparticles into Complex ulticompartment Supraparticles.” Adv. Mater. 35, 2211244 (2023)

[0015] Tire United Sates Food and Drug Administration has approved lipidoid nanoparticles (LNPs) are the delivery platform in Onpattro, the first FDA-approved siRNA drug. See, Khare et al, “Development of Lipidoid Nanoparticles for siRNA Deliver}’ to Neural Cell,” AAPS J. 2021 Dec 6;24(1). The publication describes preparation of siRNA-LNPs using C12-200, an ionizable cationic lipidoid along with helper lipids. Inclusion of both siRNA was reported and PEG-lipid provided a stabilizing effect to the LNP particle diameters and polydispersity indices by minimizing aggregation.

[0016] Tire micro/nano composites discussed in the publications require multi-step active fabrication processes. The processes typically used are expensive, and time-consuming processes borrowed from semiconductor fabrication, specifically photolithography and electrochemical etching. Prior micro/nano composites commonly use silicon materials or PLGA, which present biocompatibility challenges.

SUMMARY

[0017] A preferred embodiment is a supraparticle including, and preferably consisting of, ultrasmall Au nanoparticles, a stabilizing matrix, and a lipidoid. The supraparticle can advantageously include a labile molecule payload. A preferred embodiment provides a synthetic method to self-assemble tire components into ordered multi-unit structures, which can be referred to as Self-Assembled Multilayered Supraparticles (SAMS) to reflect their structural nature. Tire preferred self-limiting assembly process results in a finite-size supraparticles that can store and transport sensitive labile molecule payloads (for example, labile RNAs) and release them selectively upon exposure to biochemical-stimuli within cells when introduced into cells.

BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Fig. 1 is a schematic cross-sectional diagram of a Self-Assembled Multilayered Sup raparticle (SAMS) according to a preferred embodiment. [0019] Fig. 2 shows a size distribution of Au nanoparticles fabricated for inclusion in SAMS synthesized in experiments.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] A preferred embodiment of the invention provides a supraparticle including, and preferably consisting of, three components: ultrasmall Au nanoparticles, a stabilizing matrix, and a lipidoid. The supraparticle can advantageously include a labile molecule payload. A preferred embodiment of the invention provides a synthetic method to self-assemble the components into ordered multi-unit structures, which can be referred to as Self-Assembled Multilayered Sup raparticles (SAMS) to reflect their structural nature. Tire preferred self-limiting assembly process results in a finite-size supraparticles that can store and transport sensitive labile molecule payloads (for example, labile RNAs) and release them selectively upon exposure to biochemicalstimuli within cells when introduced into cells. Preferred SAMS are multilayered supraparticles with capabilities that include protecting and releasing biologically labile payloads upon intracellular-stimuli driven disassembly with high efficiency in modulating the concentrations of oncoproteins both in vitro and in vivo.

[0021] Lipidoids are synthetic, lipid-like molecules typically designed as cationic amphiphiles. Their primary role is to complex with and facilitate the intracellular delivery of nucleic acids such as DNA or RNA. These molecules are distinct from traditional lipids in that they do not require defined head groups but rather achieve their functionality through their molecular structure and amphiphilic nature.

[0022] Preferred SAMS use biocompatabile materials. Gelatin is a material classified as GRAS (Generally Recognized as Safe) by the FDA — and gold nanoparticles, are among the FDA- approved metallic nanoparticles. The presence of metal nanoparticles SAMS contributes to functional advantages, including enhanced stability and tunable release properties. This provides functionality at the nanoscale, which changes the pharmacodynamics significantly compared to prior micro/nano composites that provide functionality only at the microscale.

[0023] Preferred SAMS are an ordered spherical lamellar arrangement of a lipidoid-gold nanoparticle complex stabilized in a gelatin matrix, capable of carrying negatively charged biorelevant molecules as payloads. The preferred SAMS resembled a 3-D onion-like layered structure. [0024] Preferred SAMS provide: (i) ultrasmall subunits assembled spontaneously into an ordered structure in aqueous media; (ii) capability of delivering biologically labile payloads followed by disassembly and efficient clearance.

[0025] A preferred embodiment provides a supraparticle formed from gold nanoparticles in a 3D layered structure, where the lamellar layers are arranged in concentric rings. The spacing between the layers provides adequate space to accommodate labile payloads. The supraprticle is formed from the interaction between hydroxymethyl phosphonium salt (THPC) and secondary amines. A theory of the reaction, which is not necessary to practice the present formation methods, is that a modified Mannich reaction starts w ith an immonium salt, resulting in a transient assembly of gold nanoparticles. If not stabilized, the assembly results in intractable gold aggregates. Gelatin is used in a preferred embodiment as a backbone to provide stabilization. Chemically, when gelatin is added to THPC, another Mannich reaction occurs, wherein three of the single-strand covalents bind with the single phosphorus unit. The combination of the tetrameric structure on the phosphorus and the transient gold assembly reacts to form a 3D layered supraparticles of defined size. The size of the final assembled supraparticles formed from homogeneous-size gold subunits is dictated by thermodynamic factors that governs the selfassembly process.

[0026] Methods of the invention provide several important features that contribute to the successful synthesis and utility of supraparticles. One feature is the use of aqueous-based synthesis methods, which ensures biocompatibility. Another important feature is a rapid selfassembly process, which facilitates efficient production. The self-limiting nature of preferred formation processes are important to control assembly to obtain definite and uniform-sized supraparticles.

[0027] A preferred fabrication method optimizes a ratio and combination sequence of the three components under specific conditions, whereby the unstable transient complexes are stabilized in a gelatin nanoparticle matrix, resulting in ordered spherical lamellar-like structures. The formation of SAMS requires a delicate balance betw een the attractive and repulsive forces during nanoassembly to control the aggregation process, while the Mannich reaction initiates the process. The resulting water- stable SAMS provides a supraparticle assembled using these interactions. The lipidoid that forms a part of the structure is cationic and lends itself to electrostatic binding of negatively charged payloads such as nucleic acids and phosphate containing drugs. [0028] Experiments demonstrated the ability of SAMS to bind, protect, and deliver labile nucleic acid payloads such as siRNA (97 ± 2.8% downregulation), mRNA (overexpression with -45% transfection efficiency), synthetic guide RNA(gene editing with -5% cleavage efficiency) and drugs such as Gemcitabine monophosphatc(cytoto.xicity) in cancer cell lines in vitro. Experiments also showed that these supraparticles are well-tolerated in mice and can deliver functional siRNA to subcutaneous tumors in mice, resulting in the downregulation of the target gene through intravenous administration.

[0029] Experiments have demonstrated superior efficacy of SAMS by delivering siRNA, mRNA, cre-mRNA, CRISPR-Cas9, and positively charged drugs to cells or animal models. The results demonstrate SAMS to be suitable as targeted and efficient delivery systems, and a platform that can advance many biomedical applications.

[0030] Preferred embodiments of the invention will now be discussed with respect to experiments and drawings. Broader aspects of the invention will be understood by artisans in view of the general knowledge in the art and the description of the experiments that follows.

[0031] Fig. 1 schematically illustrates a SAM 102 with Au nanoparticle subunits 104 arranged as lamellar layers 106 arranged in concentric rings. A spacing 108 between the rings provides for the carrying of labile molecules. A stabilizing matrix 110 maintains the stable arrangement of tire SAMS.

[0032] Method of Synthesis in Accordance with a Preferred Embodiment

[0033] Step A) treated a lipidoid, e.g., G0-C 14, with ultra-small 2 nm gold nanoparticles. Owing to their charges, artisans would expect these reactants to yield an electrostatic complex. Surprisingly, the phosphonium salt (tetrakis hydroxymethyl phosphonium chloride; THPC) coated on the surface of the AuNP as a stabilizer showed a quick complexation process with gold nanoparticles to yield the SAMS. A theory that can explain the formation is that a secondary amine on the lipidoid undergoes a Mannich-type reaction with available THPC and acts as a quasi-linker between gold and lipidoid. Covalent binding of secondary amine with THPC is known only in the chain reaction of trisubstitution with amino backbones. Structural variations of G0-C14 were tested and G0-C14 was the optimal lipidoid for formation of the present SAMS.

[0034] The m-complex is unstable and kinetically active. Typically, during the initial step, the gold-phosphonium salt units attach with lipidoid on two opposite faces of the sphere to minimize the steric interaction. If the process is not controlled, the subunits crosslink with others and yield an aggregate. [0035] To avoid this. Step B) immediately added the m-Complexes to a gelatin solution in the second step. The gelatin arrests the aggregation to stabilize the complex and consequently protect subunits from disintegration.

[0036] Payload addition is conducted in Step C). G0-C14 contains a cationic core and hydrophobic tails. Hie hydrophobic tails possess adequate space to accommodate siRNA without changing the functionality. The lipidoid was treated with siRNA and then that mixture with gold- THPC to form an m-complex. Encapsulating this payload within the space did not disturb the supraparticle layered structure, the structure's integrity was well maintained. A comparison of the empty particle and siRNA- loaded particle did not show any difference in the structure. By precisely controlling the solvent environment and concentration of payload and G0C14 an ideal ratio at which the payload successfully binds to G0C14 can be found to form stable complex with negligible loss. The cationic nature allows complexation of negatively charged molecules such as siRNA and mRNA. The minimum and optimum complexation ratio can be determined via a gel migration essay. In an experiment, complexation ratio for siRNA was determined by a gel migration assay. siRNA was successfully complexed at weight ratios of at least 20: 1 (G0- C14:siRNA) and ratios of 50: 1 and 100: 1 also were successful. Uris represents the complexation of G0-C14 and siRNA at different ratios. Gelatin is not incorporated in this test. In a separate experiment where gelatin was included, a ratio (5:4: 100) refers to G0-C14, AuNPs (gold nanoparticles), and gelatin. Inclusion of RNA makes the complete ratio RNA:G0- C14: AuNP: Gelatin (1:50:40: 1000). A final weight ratio ofG0-C14:AuNP:Gelatin (5:4: 100) at concentrations of AuNP (2 mg/mL), G0-C14 (2.5 mg/mL), Gelatin (1 mg/mL) was used for subsequent experiments.

[0037] Additional stabilization can be accomplished of the surface of the SAMS. Cross-linking of the gelatin matrix can be conducted, which provides additional stabilization of the SAMS surfaces.

[0038] Synthesis of subunit nanoparticles and lipidoids

[0039] Ultra-small gold nanoparticle cores (2 nm) were prepared using THPC as a reducing agent. Briefly. 90 mL of water was first added to a 200 mL round bottom flask (RBF) and deoxygenated for 5 minutes under vacuum. Separately, reducing agent solution was prepared by adding 24 pL of 80% THPC solution to 2 mL of water in a tube. Next, 1 mL of 1 M NaOH was added using a syringe into the RBF while stirring at 900 RPM (25 C). After one minute, the THPC solution was then quickly added into the RBF. After 5 minutes, 4 mL of 25 mM HAuC14 solution was quickly injected into tire RBF, and tire solution was allowed to change from yellow to dark brown color. After 15 minutes, the reaction was arrested in a cold bath (4C) and resultant product was concentrated and washed using an 3Kd amicon centrifuge tube (5500 g for 20 min; 5 w ashes with Dl-water) and stored at 4C.

[0040] Lipidoid G0-C14 was either synthesized according to previously described procedures or purchased from Ruixi Biotech. Briefly, G0-C14 was synthesized by reacting 1,2- epoxytetradecane with generation 0 of ethylenediamine core PAMAM dendrimer at a sub- stoichiometric molar ratio of 7: 1. The mixture was reacted at 90C under vigorous stirring for 2 days. The crude mixture was purified by column chromatography on silica with a gradient elution from CH2C12 to 75:22:3 CH2C12/MeOH/NH4OH.

[0041] Formulation of SAMS and siRNA loaded SAMS.

[0042] SAMS were synthesized using a self-assembly method. 200pg of AuNP-THPC in 100 pL of 50% Acetone (2mg/mL) was added dropwise to 625 pg of G0-C14 in 250pL of Acetone (2.5mg/mL). The solution was mixed gently(lOs) and then allowed to stand(30s) to fomr the m- complex. A gelatin solution w as prepared by heating gelatin in water to 70C and allowing it to cool to room temperature. The m-complex was added dropwise to 5mL of gelatin solution(lmg/mL) under stirring conditions(lOOOrpm) at room temperature. Hie supraparticles form instantly upon mixing. The mixture was allowed to stir(lOOOrpm) for 3 hours at room temperature. The particles w ere collected and washed thrice by centrifugation at 12000g for 8 minutes to remove organic solvent and free components. The final pellet was resuspended in Hy clone water by gentle sonication to the desired concentration. Preparation of siRNA-loaded particles required an additional step at the beginning wherein Inmole of siRNA in lOpL of water was gently mixed with 625 pg of G0-C14 in 250pL of Acetone and incubated at room temperature for 30s before proceeding with the procedure described above. The amount of loaded siRNA was determined by sequential digestion of the particles in trypsin and NaCN. 15pL of SAMS(lmg/mL) carrying Fluorescein tagged siRNA was digested in 150pL of 0.25% Trypsin for 3 hours at 30C followed by the addition of 135pL of 1.5M NaCN. The mixture was centrifuged, and the fluorescence intensity of the supernatant was analyzed at 490/525nm and compared to a standard curve of siRNA-Fluorescein in an equivalent mixture of Trypsin/NaCN.

[0043] For the experiments w ith alternate organic stabilizers. BSA/Casein/DSPE-PEG was dissolved in water by vigorous stirring at a concentration of 0. Img/mL at room temperature.

[0044] Experimental Findings

[0045] Experiments showed that SAMS exhibited outstanding stability and retained siRNA with no release or degradation. In contrast, within cells, SAMS disintegrates and releases tire siRNA within the cytoplasm. T disintegration of SAMS is likely mediated by pH and MMP proteins present in cancer cells, though tlris theory is not necessary to use the present SAMS as a payload delivery vehicle in cells.

[0046] A method of the invention synthesizes SAMS by mixing an appropriate ratio of G0-C14 and 2 nm gold nanoparticle components (and an optional payload) to form stable complexes that were further stabilized by a gelatin matrix producing self-assembled SAMS. More generally, the lipidoid should contain a specific length of alkyl tails in the final compound (cl 4). The G0-cl4 was synthesized using a ratio of 1 :7 (G0:epoxy). Degradation of payload is avoided (i) by ensuring conditions for complexation with the lipidoid, e.g., G0-C14, (ii) crosslinking the complex to 2 nm metal nanoparticles, e.g. Au-THPC nanoparticles, (iii) encapsulating the nanocomplexes within a gelatin matrix and (iv) subsequently crosslinking the gelatin on the surface of SAMS. The AuNPs allow high siRNA loading and packing within the gelatin matrix due to the ordered self-assembly process of SAMS synthesis. Uniform composition of the payload within the supraparticles is achieved. The gelatin provides robust protection of the nanocomplex, and functional groups such as amines and carboxyl groups allow attachment of targeting ligands on the surface of SAMS. Passive PEGylation optionally can allow additional high stability and solubility.

[0047] pH. ionic strength and temperature of the aqueous environment affect the interaction between reaction components given the nature of gelatin and the lipidoid. The concentration of the gelatin and lipidoid components can be adjusted within an optimum range to form well- defined nanoparticles. Too low concentration of lipidoid might not allow for sufficient interactions. While too high concentration could lead to aggregation rather than a well-defined nanoparticle formation. Moreover, gelatin comes in various molecular weights and structures which can influence the result of the interaction. However, the yield and stability of nanoparticles was suboptimal with noticeable loss of interacting gelatin was observed as filamentous residues in the reaction supernatant post-centrifugation. In an experimental formation process, tire parameters for gelatin solution without gold nanoparticles is pH (5.5), Ionic strength (Ultrapure water with very low ionic strength of 10 ⁷ M), Temperature (25 °C).

[0048] Given the importance of THPC in the formation of the ordered structure, an artisan could expect that the subunit nanoparticle core could be substituted with another metal provided the size is maintained. However, testing showed that switching gold to other transition metals does not result in a structure bearing any similarity to SAMS Tire size of the Au subunits required to form the ordered spherical lamellar suprastructurc was determined by experiments to demonstrate a clear transition point beyond which the structure is destabilized. Generally, for SAMS formation, the gold nanoparticle size should be below 5 nm (Ideally between 0.5 nm and 2 nm). Slightly larger particles around 3nm are expected to work well since the size distribution for 2nm subunits covers this size.

[0049] Experiments revealed some important parameters. a. Au nanoparticles led to formation of the SAMS, but Pd. Pt and Rh nanoparticles did not. b. Lipidoid chain lengths of between 12 and 14 led to formation of the SAMS, but shorter and longer chain lengths did not. c. Au nanoparticle sizes of between 1 nm and 2.5 nm to formation of the SAMS, but particles above 5 nm did not. d. Gelatin and Bovine serum albumin (BSA) provided a stabilizing matrix to fonn the SAMS, but Casein and DSPE-PEG did not. e. A nanoprecipitation temperature of 25C led to formation of the SAMS, but temperatures of 4 C did not. A temperature range of 25-50 C is preferred. f. A neutral nanoprecipitation pH of ~7 led to fonnation of the SAMS. g. Gelatin concentrations of 0.2 mg/ml to 4 mg/ml led to fonnation of the SAMS, while 1 mg/ml was found optimal. h. THPC was found to be a suitable capping agent. i. A lipidoid to siRNA complex ratio to form SAMS was 1:20 to 1:50.

[0050] Synthesis of G0C14

[0051] Tire synthesis of G0-C14 involves a ring opening of 1 .2-cpoxytctradccane by PAMAM generation 0 in tire presence of heat . Sub-stoichiometric molar ratio of the epoxide to the dendrimer ensures 1 less tail than the total reaction sites achieving a balance of hydrophobicity and net positive charge. The reaction as purified by silica column chromatography using a gradient from CH2CI2 to 75:22:3 CFHCE/McOH/NFEOH. The final product was characterized by TLC and NMR confirming addition of hydrophobic tails.

[0052] Synthesis of 2 nm gold nanoparticles with THPC.

[0053] AuNP-THPC was synthesized according to previously reported procedures. Transmission electron microscopy (TEM) was used to detennine the size distribution (2.12nm ± 0.52) of the final washed nanoparticles from an experiment. Fig. 2 shows the size distribution of the Au nanoparticles. Other experiments shows a range of sizes from 0.4 to 3.3 nm.

[0054] Nanocomplex formation with gold nanoparticles and G0C14-siRNA [0055] The ability of G0-C14 to interact with AuNP-THPC was explored by mixing G0-C14 solubilized in ethanol with a colloidal suspension of AuNP-THPC in a volume of water that does not affect the solubility of G0-C14. Tire nature of mixing is not critical but dropwise addition followed by gentle mixing is preferred. When allowed to incubate at 25C we noticed gradual precipitation until all visible gold nanoparticles appeared irreversibly bound to the tube. Sonication or vigorous vortexing could not reverse the precipitation observed. We termed the product of the AuNP- G0C14 interaction: m-Complex

[0056] Optimization : Importance of THPC on AuNP for m-Complex formation

[0057] Considering tire cationic nature of G0-C 14 and the overall negative charge of AuNP- THPC we sought to understand the driving force behind the observed interaction. We synthesized similar sized negatively charged gold nanoparticles through Sodium Borohydride. The NaBH4 AuNPs showed no interaction with the lipidoid and remained stable in solution unlike the AuNP- THPC nanoparticles. Electrostatic charge appeared to be insufficient for the formation of m- Complexes pointing towards an interaction between THPC and G0-C14.

[0058] Optimization: Importance of gelatin stabilization matrix

[0059] Gelatin was tried as an aqueous organic stabilizer. Upon nanoprecipitation of the freshly prepared m-Complexes (before instability) into gelatin solution (1 mg/mL), we observed a homogenous solution with colloidal properties. When left overnight we did not observe any increased instability in the homogenous reaction. In the absence of gelatin, water alone could not stabilize tire precipitation quickly resulting in a black mass stuck to the stirring magnetic bead.

[0060] Reproducible synthesis of self-assembled lamellar onion-like nanoparticles (SAMS) or SAMS.

[0061] To understand tire role and importance of each of the 3 components we performed tire same reaction with different combinations of each component (G0-C14. Gelatin, AuNP-THPC). As observed previously, G0-C14 and AuNP-THPC formed an irreversible precipitate seen as heterogenous masses in TEM. AuNP-THPC and gelatin did not seem to interact as evidenced by free floating AuNP-THPC observed in TEM. G0-C14 and gelatin interaction resulted in tire formation of gelatin nanoparticles which was unexpected. When all three components were incorporated, ordered multilayer supraparticles form. The stabilized m-Complexes appeared to self-assemble themselves into an ordered supraparticle in a self-limiting manner. The presence of both lipidoid and gelatin was required for the formation of SAMS.

[0062] Optimization: Importance of THPC on AuNP for SAMS formation [0063] We further characterized the importance of THPC in the formation by attempting to synthesize SAMS with NaBfty AuNPs instead of AuNP-THPC. The lack of m-Complex formation resulted in the formation of empty Gelatin nanoparticles with free floating NaBH AuNPs. The presence of NaBH4 AuNPs in the reaction did not affect the interaction of G0-C14 and gelatin to fomr Gelatin nanoparticles.

[0064] Optimization: Gelatin concentration optimization.

[0065] A first SAMS synthesis used a gelatin concentration of 1 mg/m. T o understand the impact of organic stabilizer concentration we attempted synthesis over a range of concentrations (0.2mg/mL to 4mg/mL). The hydrodynamic size and zeta potential were relatively unaffected by the change in organic stabilizer concentration.

[0066] Optimization: Importance of size of AuNP THPC for SAMS formation

[0067] Subunit nanoparticle size dependence on the ordered arrangement was tested. AuNP- THPC of sizes 0.5-1.5nm, 2.5nm, 5nm and lOnm were synthesized and characterized. Our results showed that 5nm and lOnm subunits disrupted the arrangement and resulted in heterogeneous aggregations within the supraparticle and displayed aggregations of gelatin and gold nanoparticles outside the boundary of the supraparticle.

[0068] Optimization: Alternative Lipidoid carbon chain length

[0069] Changing the length of the alkyl chains would affect the solubility, hydrophobicity, and size of the lipidoid molecule. We attempted the synthesis of SAMS with GO-Cn (n=8, 12, 14, 18). G0-C12 and GO- C14 yielded the ordered SAMS structure. G0-C8 and G0-C18 formed incomplete arrangements with free floating subunit nanoparticles outside the structures.

[0070] Optimization: pH of aqueous phase during synthesis.

[0071] The nature of gelatin in solution is dependent of the charges on the amino acid sidechains, which are sensitive to pH changes. Amino-rich PAMAM containing G0-C14 is also influenced by the pH of the environment. We attempted to synthesize SAMS with the pH of the aqueous gelatin solution adjusted to 2.5 and 9.0. Both conditions adversely affected the formation of tire supraparticle. At pH 2.5 the edge of the supraparticle appeared diffused and merged supraparticle were observed. At pH 9.0, the boundary of the supraparticles were well defined; however, the internal order was disrupted.

[0072] Optimization: pH impact on SAMS

[0073] The pH sensitivity of the components prompted us to explore the stability of the formed SAMS at different pH. At pH 9.0 similar smaller clusters were observed with some order retained within tire clusters in addition to free nanoparticlcs. There was a pronounced impact on the gelatin mesh at pH 9.0 with large disc like formation on the periphen’ of the supraparticles. At pH 3, the supraparticle structure is disrupted with smaller collapsed, disordered clusters visible.

[0074] Characterization: Self-assembly process over time

[0075] Tire nanoprecipitation process allows the self-limiting formation of SAMS. To capture the formation process we sampled the reaction in under 5 seconds and at 500 seconds and prepared TEM grids immediately. We observed partially formed SAMS at the early timepoint with large deposits undistributed m -Complex visible on SAMS. At 500 seconds only ordered SAMS were visible. The data suggests that the deposited m-Complexes distributes overtime forming the final ordered SAMS structure.

[0076] Optimization: Alternative stabilizing matrix.

[0077] Bovine Serum Albumin (BSA), Casein, 1, 2-Distearoyl-sn-glycero-3- phosphoethanolamine-Polyethylene glycol (DSPE-PEG) and Polyvinylpyrrolidone (PVP) were tested as stabilizers. BSA was able to recapitulate the formation of SAMS. The other stabilizers resulted in disrupted supraparticles (Casein) or heterogeneous aggregates (PVP, DSPE-PEG).

[0078] Optimized Formation

[0079] A final weight ratio of G0-C14:AuNP:Gelatin (5:4: 100) at concentrations of AuNP (2 mg/mL), G0-C14 (2.5 mg/mL), Gelatin (1 mg/mL) was used for many experiments. A library’ of TEM images were generated for SAMS of different sizes. The lamellar arrangement was easier to visualize at larger sizes where the supraparticle have more subunits, appearing dense. High-angle annular dark-field imaging (HAADF) showed that the AuNP subunits retained their crystalline nature. Three-dimensional nature of SAMS was confirmed through Tilt TEM tomography. Analysis of the ‘roundness’ over 80 degrees revealed a fairly spherical supraparticle.

[0080] Inter-Layer distance between layers.

[0081] The multilayered structure was further characterized by measuring the interlayer distance for differently sized SAMS. The interlayer distance was found to be consistent (3.5±0.004) regardless of supraparticle size.

[0082] Application : Radiotherapy enhancement.

[0083] Gold nanoparticles can significantly enhance tire efficacy of radiotherapy by increasing the absorption of radiation, through secondary electron generation and subsequent DNA damage and generation of reactive oxygen species. Since each sup raparticle is made up of multiple gold nanoparticle subunits, SAMS may have the potential to enhance radiotherapy. We explored this potential through radiotherapy of A549 lung cancer cell lines and qualified the y-H2AX foci indicative of DNA damage. SAMS treatment significantly enhances radiotherapy compared to nanoparticle free control.

[0084] Optimization: Controlling siRNA complexation with GOC 14

[0085] G0-C14 is a lipid like compound that consists of a cationic core with alkyl hydrocarbon tails. Its cationic nature allows complexation of negatively charged molecules such as siRNA and mRNA. Tire optimum complexation ratio for siRNA was determined by a gel migration assay. siRNA was successfully complexed at weight ratios of 20: 1 (G0-C14:siRNA) and above 50: 1 was chosen for further experiments.

[0086] Characterization: siRNA load estimation in SAMS-siRNA.

[0087] Tire siRNA load in SAMS-siRNA was estimated directly by digesting the supraparticle and releasing the siRNA with a fluorescent tag (Fluorescein). The gelatin matrix was digested by Trypsin and the gold nanoparticles were digested by Sodium Cyanide (NaCN). Any debris were centrifuged, and the fluorescence of the supernatant was compared with a standard curve of known concentrations of siRNA-Fluorescein in an identical mixture of Trypsin and NaCN (Fig. 29a). Three separate batches were analyzed with average siRNA loading of 4.85 ± 0.22 pg/mg. In the absence of AuNP-THPC, the Gelatin -GOC 14-siRNA particles showed remarkably lower siRNA loading of only 0.68 ± 0.23 pg/mg.

[0088] Characterization: siRNA distribution in SAMS.

[0089] TEM showed the ordered lamellar nature of SAMS. We sought to understand the distribution of the payload siRNA within the structure since the complexation with G0-C 14 is performed at the first step. Since G0-C 14 is integral to the formation and arrangement of SAMS any differential distribution of complexed/uncomplexed G0-C14 would be interesting to note. We perfonned STED (stimulated emission depletion) super resolution microscopy of single SAMS carrying siRNA attached to Cy5 fluorescence reporter, followed by software deconvolution. The confocal sections were reconstructed in 3D to give us a complete view of the supraparticle. The siRNA- Cy5 appeared evenly distributed throughout the SAMS structure when viewed through all axes.

[0090] Characterization: siRNA release study

[0091] The siRNA retention ability of SAMS was demonstrated by gel migration assay. Intact SAMS showed good siRNA retention with no free siRNA migrating through the gel. This was confirmed by release of the siRNA through trypsin digestion of gelatin and competitive release of siRNA from tire cationic complexes by SDS (Sodium Dodecyl Sulfate), where all the siRNA migrated through the gel. [0092] Cellular uptake study

[0093] Tire cellular uptake percentage of a particle reflects how efficiently nanoparticles can be internalized by target cells, directly influencing their therapeutic potential. We used flow cytometry to quantify tire uptake percentage of SAMS-siRNA-Cy5 on A549 lung cancer cell line after 6 hours of uptake. Tire results showed remarkable 99.87 ± 0.04% efficiency.

[0094] Cellular uptake study: TEM

[0095] High cellular uptake is insufficient for efficient therapy. The fate of the supraparticle within a cell would be integral to its function. Payload release would require disassembly of the SAMS structure and the small size of the subunits would facilitate subsequent clearance from the body.

[0096] TEM images of cells at various stages of uptake (0, 0.5. 2, 6 , 24 hours) were taken to characterize the intracellular fate of SAM). Within 30 minutes SAMS were seen inside the cells and at various stages of vesicle escape and release over the next few hours (2/6 hours). By the 24- hour timepoint, disassembly of the supraparticle into its individual subunits was observed.

[0097] SAMS-siRNA cellular localization.

[0098] The deliver}’ of labile siRNA in a functional form is highly dependent on successful early endosomal escape. The engineering of a carrier vehicle must be able to bind siRNA, protect the siRNA from degradation, be taken up into the cell efficiently, escape the endosome and release the payload within the cytoplasm . We performed real time confocal microscopy of SAMS- siRNA-Cy5 in A549 cells over time to colocalize the supraparticle with endosomes stained by a LysoView dye. We observed rapid uptake of SAMS- siRNA-C5 into the cell and over time few spots are seen colocalized with the LysoView dye and many spots were seen separated from the Cy5 fluorescence indicating escape. Tire gradual increase in acidic pH facilitates disassembly of SAMS exposing G0-C14. which when present in an acidic environment can promote disruption of the endosome layer through proton sponge effect and through its pronounced cone shaped structure.

[0099] Characterization: SAMS-siRNA serum-stability

[0100] The protection of siRNA from enzymatic degradation during circulation and biodistribution is a crucial aspect of RNA-based therapies. A gel migration assay was used to assess the protection afforded by SAMS against RNase degradation. SAMS-siRNA-Cy5 only showed partial degradation within the first 20 minutes and remained stable for the rest of the assay duration (4 hours), whereas free siRNA was completely degraded within minutes.

[0101] SAMS-siRNA penetrability in spheroids. [0102] Tumors have complex and dense 3D structures that can impede diffusion and deep penetration. Effective tumor penetration ensures the carrier can distribute to the difficult to access regions of the tumor. To assess tire tumor penetrability of SAMS an experiment analyzed confocal sections of A549 spheroids after 18 hours of treatment with SAMS-siRNA-Cy5. Cy5 signal was seen deep into the spheroid evidenced by a minimal drop-off in tire ratio of mean fluorescence between the inner and outer region of the treated spheroids.

[0103] SAMS-siRNA in vitro functionality: Gene Downregulation

[0104] Tire high uptake, disassembly, endosomal escape, and siRNA protection shown by SAMS could lead to efficient gene downregulation. AXL was selected as a model gene for downregulation in NSCLC cell line A459 due to its short half-life and negligible effect on cell viability. Three separate batches were tested for gene downregulation via western blot. SAMS- siAXL treated in A549 cell line(48hr) at 33 to 37 nM showed 97 ± 2.8% AXL downregulation in A549 and was comparable to 50nM siAXL delivered with gold standard transfection agent. The siRNA remained stable in the construct and did not show a dip in downregulation efficiency (97%) after 2 weeks of storage at 4C. Long term single dose in vitro gene silencing (AXL) was also evaluated (in A549). After a single treatment of SAMS-siAXL for 8 hours at 20nM siRNA eq. concentration, significant downregulation was observed for 9 days with a small drop in downregulation by day 16. SAMS-siAXL was further compared with standard transfection agent at various siRNA concentrations, showing comparable if not better downregulation at siAXL equivalent concentrations of 20nM and lOnM. At very low siAXL concentration of InM, the transfection agent showed slightly better downregulation (63% vs 42%). This can be attributed to absolute particle numbers dropping below a critical point at InM equivalent siAXL dose. Furthermore, in the absence of gold nanoparticle subunits, downregulation was poor even at higher equivalent siRNA concentrations.

[0105] SAMS-siRNA in vivo toxicity.

[0106] Tire in vivo toxicity profile of SAMS was investigated in immunocompetent CF1 mice at 3 different intravenous (IV) doses (16, 32 and 64 mg/kg dry weight; 0.2mL per mouse). The serum biochemistry profile showed that SAMS were well tolerated at 16 and 32 mg/kg but showed an increase in liver related enzymes at very high dosage (64 mg/kg). Similarly, the complete blood cell counts showed no remarkable changes at 16 and 32 mg/kg except for a drop in platelet count at all doses. At high doses (64 mg/kg), elevated WBC counts, and suppressed lymphocyte counts were also observed. [0107] Histological analysis showed no significant microscopic lesion in all examined organs (Lungs, heart, kidneys, liver, pancreas, spleen, and brain) at 16 and 32 mg/kg. In the high dose group (64 mg/kg), one animal showed multifocal myocardial necrosis with mineralization, and all the mice in the group had mild hepatocellular hypertrophy with mild cytomegaly and glycogen depletion. Uris partially explains the discrepancy observed in the blood reports.

[0108] SAMS-siRNA in vivo functionality: Gene downregulation

[0109] The downregulation efficacy of SAMS-siAXL was evaluated in 3 separate studies with variation in the target gene or tumor type. The first study evaluated the downregulation of AXL upon intratumoral administration of SAMS-siAXL in a A549 subcutaneous cell derived xenograft (CDX) model (single dose, 1.6ug siAXL per tumor). All mice except one showed significant downregulation. High interstitial pressure in A549 tumors was observed, which could lead to ejection of injected material resulting in variations in downregulation.

[0110] The second study evaluated the downregulation of YAP upon intravenous administration of SAMS-siYAP in a A549-AXL knockout subcutaneous CDX model (0. 16mg/kg siAXL equivalent dosed twice at Ohrs and 48hrs). 48 hours after the final dose, tumors were collected and evaluated for YAP expression by western blot. An average of 36.2% gene downregulation was observed.

[0111] The third study aimed to evaluate SAMS in a more relevant Patient derived xenograft model. In this study SAMS-siAXL was intravenously administered intravenously on an NSCLC subcutaneous PDX model that exhibited high AXL basal expression (0.08 and 0. 16mg/kg siAXL eq. for 3 consecutive days). Dose-dependent down regulation of the target gene was observed (27.9% at 0.16mg/kg and 15.6% at 0.08mg/kg).

[0112] SAMS-mRNA

[0113] Unlike siRNAs compact double stranded nature, mRNA is significantly larger, more fragile and a more complex secondary structure. The parameters required for efficient packaging and deli very could be different such as the time required to condense such a large structure, the pH during tire condensation and the effect on the self-limiting ability of SAMS to accommodate this larger mRNA. While siRNA efficacy can be assessed through gene silencing, mRNA could require other techniques to quantitate gene expression and transfection efficacy. A good starting point could be to apply the sequence and conditions used for SAMS- siRNA synthesis and subsequently optimize the parameters required to improve reproducibility and efficacy.

[0114] SAMS-mRNA was first synthesized using the parameters optimized for SAMS-siRNA. Briefly, G0-C14 in acetone was added to an aqueous mRNA solution (mRNA:G0-C14: 1:50 w/w and 1:20 v/v). The mRNA was allowed to condense with the lipidoid for 1 minute. AuNP-THPC (50% acetone) was then added dropwise to the mixture (GO-C 14: AuNP-THPC; 5:4 w/w) and mixed gently forming m-complexes. The m-complexes w ere added dropwise to a gelatin solution (GO- C14:AuNP:Gelatin (5:4: 100 w/w) at room temperature. Tire mixture was stirred for 2 hours and collected and washed by centrifugation. Tire final supraparticles were resuspended in water for analysis.

[0115] EGFP mRNA is an excellent choice for supraparticle iteration due to its nominal size (~1000bp) and easy to detect fluorescent expression. There was no visual difference during the synthesis of SAMS using this mRNA, TEM analysis showed the formation of dense supraparticles with characteristic SAMS layers. However, most of the supraparticles showed the presence of something embedded within the structure, disturbing the density of AuNPs in that region. This data suggests that the large mRNA complexes are occluded within the structure unlike siRNA which do not affect the structure . Both DLS and TEM showed an increase in the overall size of the supraparticle (~460nm) compared to SAMS-siRNA and empty SAMS (~250nm). It is possible that the size of the complexed payload could influence the limit of the self-assembly. Tire overall charge of the nanoparticle remained moderately positive (+12mV).

[0116] SAMS -mRNA Optimization

[0117] Initial experiments showed that SAMS were able to successfully transfect functional mRNA to cancer cells (A549 lung cancer; 48 hours). Some settling was observed in the wells which could be due to the larger size of the supraparticles.

[0118] Previous research has shown that the lipidoids are more positively charged at acidic pH which can aid mRNA condensation and endosomal escape. Research has also shown that complexation time is important depending on the properties of tire cationic and anionic moieties. We varied the complexation pH (pH 6.4 and pH 4) and complexation time (1 minute or ~15 minutes) and assessed the impact on transfection efficience.. Flow' cytometry was used to quantitate the transfection efficiency of EGFP mRNA in a medium-high throughput manner.

[0119] Our results showed 10 minutes of G0-C14-mRNA complexation time at pH4(using ImM Citrate buffer) yielded the most consistent transfection efficiency (41.3% ± 5% GFP positive cells). Complexation with other shorter alkyl chain lipidoid (G0-C8/C12) were briefly explored. However, these SAMS-mRNA showed poor transfection efficiency.

[0120] SAMS- Cre-mRNA - HDS and Zeta

[0121] Tire mT/mG Cre system offers a powerful tool for spatiotemporally controlled mRNA delivery. This system utilizes a cell line or mouse model constitutively expressing a membrane- targeted tdTomato fluorophore. Upon Cre recombinase introduction, the mT cassette is excised resulting in green fluorophore expression. Here, we compared the efficacy of SAMS-mRNA-Cre with standard Lipofectamine transfection agent through the detection of mG activation with flow cytometry.

[0122] SAMS-mRNA

[0123] Unlike siRNAs compact double stranded nature, mRNA is significantly larger, more fragile and a more complex secondary' structure. The mT/mG Cre system offers a powerful tool for spatiotemporally controlled mRNA delivery'. This system utilizes a cell line or mouse model constitutively expressing a membrane-targeted tdTomato fluorophore. Upon Cre recombinase introduction, the mT cassette is excised resulting in green fluorophore expression. Here, we compared the efficacy of SAMS-mRNA-Cre with standard Lipofectamine transfection agent through the detection of mG activation with flow cytometry.

[0124] The Cre mRNA (1.3kb) is of comparable size to the GFP mRNA(lkb) used in the optimization stage. The hydrodynamic size was smaller than the SAMS-mRNA-EGFP, but significantly larger than SAMS-siRNA, with similar positive zeta potential (+7mV).

[0125] SAMS-mRNA-Cre - in vitro delivery

[0126] SAMS-mRNA-Cre showed successful expression of Cre recombinase when treated for ~36 hours in mT/mG fibroblast cell line. The successful gene editing was analyzed by the percentage of GFP positive cells indicating mG activation. Comparable gene editing efficiency between was observed for Lipofectamine 2000 mediated Cre mRNA transfection and SAMS- mRNA-Cre.

[0127] SAMS -Gemcitabine Monophosphate (GMP).

[0128] Gemcitabine monophosphate was chosen for the proof-of-concept study in complexing an anionic drug within tire supraparticle due to its structural similarity to nucleoside analogs. This similarity is relevant because SAMS has demonstrated the capability to deliver nucleic acids, suggesting potential compatibility with GMP’s structure.

[0129] A typical siRNA contains 21 base pairs which would have approximately 40 phosphate groups. Since each GMP contains one phosphate group, an equimolar amount of GMP relative to the number of siRNA phosphate groups (used in the preparation of SAMS-siRNA) was taken for complexation with G0-C14. The rest of the procedure remained the same as SAMS-siRNA (including volume ratios).

[0130] SAMS-GMP characterization - TEM, HDS and Zeta [0131] To confirm the interaction between G0-C14 and GMP, the shift zeta potential was analyzed for different concentrations of GMP complexed with a fixed amount of G0-C14. We observed a reduction in the overall zeta potential of the G0-C14-GMP mixture indicating interaction between the components.

[0132] SAMS-GMP in vitro efficacy

[0133] The efficacy of SAMS-GMP was assessed by treating A549 NSCLC cell lines for 48 hours. GMP was replaced with an equimolar amount of gemcitabine during synthesis to make SAMS-Gem which was used as a control. GMP lacks the anionic phosphate group and therefore should not be expected to complex with cationic GO-C 14. There was a significant reproducible increase in efficacy of SAMS-GMP (two batches) when compared to SAMS-Gem. Tire efficacy of SAMS-Gem could be the result of passive occlusion of gemcitabine within the SAMS structure.

[0134] SAMS- mRNA-Cas9 and SAMS- sgRNA-AXL in vitro gene editing

[0135] We were able to successfully encapsulate Cas9mRNA and sgRNA separately within SAMS. For the Cas9 mRNA we used the parameters optimized for SAMS-mRNA-EGFP. For sgRNA we used the parameters optimized for SAMS-siRNA.

[0136] The genomic cleavage assay leverages tire mismatch endonuclease activity of T7 Endonuclease 1 (T7E1) to cleave the amplified target site, which can then be visualized on a gel. If no CRISPR mediated editing was achieved, no T7E1 cleavage would occur resulting in a single band on the gel. Co-treatment of A549 NSCLC cells with SAMS-mRNA-Cas9 and SAMS- sgRNA-AXL showed significant gene editing activity demonstrated by the genome cleavage assay. We were able to achieve up to 6.4% gene editing efficiency in our proof-of-concept study. Further optimization would be needed to improve the efficiency, such as timing of the cotreatment. , parameters required for maximum sgRNA packing, duration of treatment, among others.

[0137] Dual -Targeted RNA Delivery System for AXL and TIGIT siRNAs to Tumor

Microenvironment.

[0138] This system relates generally to nucleic acid delivery systems, and more specifically to a system for the targeted delivery of small interfering RNA (siRNA) to tumor microenvironments, focusing on the simultaneous suppression of AXL and TIGIT gene expressions. Cancer therapies often face challenges due to the complex interactions within the tumor microenvironment (TME). AXL and TIGIT are two key immune checkpoints that, when overexpressed in tumors, can lead to immune evasion and resistance to therapy. AXL, a receptor tyrosine kinase, is known for its role in promoting tumor survival, invasion, and metastasis. On the other hand, TIGIT, an inhibitory receptor on T cells, helps in dampening the immune response against cancer cells. Current therapeutic approaches often target these molecules separately, but there is a growing need for an integrated strategy to simultaneously inhibit both pathways for enhanced therapeutic efficacy.

[0139] A preferred immunotherapeutic paradigm simultaneously targeting three critical pathways is demonstrated: AXL-mediated drug resistance, PD-1 checkpoint inhibition, and TIGIT-associated T-cell exhaustion. We found that AXL inhibition triggers a compensatory upregulation of PD-1 and TIGIT. This discovery provides critical insights into the failure of current clinical trials focusing on dual AXL and PD-1 inhibition. Only one-thirds of patient respond to this dual therapy, and we attribute to TIGIT upregulation in tumor microenvironment. Cancer therapies often face challenges due to the complex interactions within the tumor microenvironment (TME). In the current situation, AXL and TIGIT are two key immune checkpoints that, when overexpressed in tumors or TME, can lead to immune evasion and resistance to therapy. AXL, a receptor tyrosine kinase, is known for its role in promoting tumor survival, invasion, and metastasis. On the other hand, TIGIT, an inhibitory receptor on T cells, helps in dampening the immune response against cancer cells.

[0140] There are no FDA-approved AXL inhibitors or TIGIT-targeting antibodies, although several candidates are in development. This embodiment provides a nanoparticle-based system delivering dual siRNAs to inhibit both AXL and TIGIT simultaneously. This novel approach aims to overcome resistance by simultaneously silencing upstream immune suppression and compensatory immune checkpoint pathways, providing a new and effective strategy for resistant NSCLC. There are several advantages of using nanoparticle-siRNA over antibody-based therapies. First, while antibodies block immune checkpoint proteins on the cell surface, siRNA prevents their formation at the mRNA level, ensuring sustained T-cell activity and prolonged antitumor responses. This enables continuous cancer cell elimination, reducing tumor escape and relapse, with potentially improved clinical outcomes. Second, nanoparticles address tumor penetration issues that often limit efficacy of antibody-based therapies. Their small size allows infiltration of dense tumor, delivering siRNA to cells antibodies may fail to reach.

[0141] SAMS-AXL-T1GIT: We synthesized a dual siRNA construct (using SAMS platform technology) by incorporating equal amounts of AXL and TIGIT siRNAs into the self-assembly process. Ensuring consistent composition across nanoparticles remains a significant challenge in developing effective delivery’ systems. To address this, we developed a self-assembly process that generates SAMS with uniform composition, akin to a well-controlled assembly line where each block is constructed from the same amount of materials. Our data demonstrate that SAMS exhibit uniform composition, excellent stability, enhanced cellular deli ven . and robust knockdown efficacy in vitro and in vivo. Notably, we observed that SAMS-mediated AXL downregulation in immunogenic murine models led to a significant increase in PD-1 and TIGIT levels within CD4+ and CD8+ T cells, replicating the compensatory immune escape mechanism triggered by AXL inhibition in our co-culture system.

[0142] The Self-Assembled Multilayered Supraparticle (SAMS) platform for the co-delivery of AXL and TIGIT siRNAs, overcomes resistance mechanisms and maximizing therapeutic efficacy. Unlike traditional antibody-based therapies, this siRNA-loaded nanoparticle system: Prevents compensatory upregulation of immune checkpoints by simultaneously silencing AXL and TIGIT; Enhances tumor penetration due to its sub-200 nm size; Ensures uniform siRNA composition per particle, mitigating inconsistencies in RNA delivery; Leverages tumor microenvironmental triggers (pH and MMP-2 activity ) for controlled siRNA release, reducing systemic exposure and off-target effects. By addressing the limitations of current therapies, this invention provides a scalable, efficient, and clinically translatable solution for improving NSCLC treatment outcomes.

[0143] The SAMS provides an ultra-small gold nanoparticle provides structural stability while layers enable strategic positioning of different siRNAs with distinct release kinetics. The nanoparticle facilitates strong charge separation, which drives siRNA release within the cell via the proton sponge effect. The self-assembly process is highly scalable and ensures the exact amount of siRNA is packed into each nanoparticle, addressing a critical issue in therapeutic nanoparticle production. The gelatin-based structure responds specifically to tumor microenvironmental triggers such as MMP-2 activity and acidic pH, facilitating precise spatial and temporal siRNA release. Importantly, SAMS can simultaneously deliver AXL and TIGIT siRNAs, synergistically offering a combinatorial therapeutic solution that targets drug resistance and immune checkpoint pathways.

[0144] SAMS simultaneously downregulate both AXL and TIGIT: We demonstrated the efficacy of the SAMS-AXL-TIGIT construct in down regulating tire target proteins in both in vitro using 4T1 cells and in vivo using tire B16F10 mouse model. In 4T1 cells, the construct achieved efficient co-delivery of siRNAs, resulting in simultaneous knockdown of AXL and TIGIT. In the B16F10 mice model, SAMS nanoparticles effectively downregulated the target proteins in both tumor and T-cells. PCR analysis revealed that while SAMS-AXL treatment increased TIGIT expression, the SAMS-AXL-TIGIT construct successfully reduced TIGIT levels, demonstrating its effectiveness. This dual-targeted approach demonstrates the versatility of SAMS for combinatorial gene silencing and highlights its potential to overcome compensatory immune checkpoint upregulation in cancer. This spatially coordinated deliver}’ maximizes the therapeutic synergy by addressing both tumor-intrinsic resistance (via AXL inhibition) and tumor-induced immune suppression (via TIGIT blockade), overcoming compensatory escape mechanisms.

[0145] SAMS are uniform composition nanoparticles with oncentric architecture: The SAMS integrate gold nanoparticles (AuNPs; 2 nm) stabilized by tris(hydroxymethyl)-phosphonium chloride (THPC), a lipidoid (G0-C14), siRNA, and gelatin, resulting in a stable particle with a size of -155 nm and an interlayer spacing of -3.5 nm. Hie siRNA (loading 5-10 pg/mg) is strategically positioned within the interlayer space, ensuring efficient protection against serum degradation and enzymatic activity. Each component plays a role in forming the SAMS structure. The gold nanoparticles (AuNPs) serve as functional "helpers" that enhance nanoparticle performance. Their negative charge facilitates the stability of the layered structure and strengthens electrostatic interactions, improving siRNA encapsulation and enabling controlled release. Like negatively charged polymers (e.g.. polystyrenesulfonate) that aid RNA delivery in lipid nanoparticle systems, AuNPs within SAMS contribute to efficient endosomal escape and precise siRNA release. SAMS without AuNPs fail to form a layered structure and exhibit poor gene downregulation efficacy. Gelatin serves as a biocompatible coating, integrating all components while providing stability and functional groups for antibody conjugation. Using STED microscopy, we confirmed the uniform distribution of siRNA across SAMS’ concentric layers, enabling consistent and controlled release. To achieve targeted delivery, SAMS were functionalized with HS-PEG(2000) and a target-specific antibody. The thiol-PEG permeates within the SAMS structure to covalently bind to the AuNP core, while the antibody is conjugated to the surface via EDC/NHS chemistry.

[0146] SAMS respond to tumor-specific triggers for siRNA deliver}’: SAMS demonstrate exceptional stability under physiological conditions and disassemble in response to TME cues, low pH (5-6) and MMP-2/MMP-9 enzymatic activity. For example. MMP-2 facilitates degradation of the gelatin matrix, enabling targeted siRNA release within cancer cells. To demonstrate this, we treated the SAMS with MMP-2 (80 nM; matching the concentration present in TME or tumor cells which is 80-160 nM; in blood it ranges between 10 to 20 nM) and monitored structural changes at 30 minutes and 10 hours using TEM. Within 30 minutes, the TEM images revealed disassembly in the outermost layers of the SAMS, resulting in the release of free AuNPs. After 10 h, the entire SAMS had disassembled, with only scattered AuNP clusters observed. No disassembly if the concentration of MMP-2 is less than 80nM. Similarly, SAMS respond to low pH, enhancing tire delivery of siRNA within the TME.

[0147] SAMS protect siRNA, enable intracellular delivery, and achieve efficient gene knockdown: Encapsulated siRNA remains intact for up to 24 hours in serum, in contrast to free siRNA, which degrades within 20 minutes. Intracellular studies using TEM confirm successful endosomal escape and cytoplasmic release of siRNA demonstrating that SAMS achieve functional deli very of encapsulated siRNAs. SAMS demonstrate superior penetration into dense tumor spheroids (-300 pm), with fluorescent siRNA detected as deep as 100 pm, effectively overcoming physical barriers within tumor structures. In A549 lung cancer cells overexpressing AXL, SAMS achieve 97% gene knockdown at 33-37 nM. Notably, the downregulation efficiency of SAMS surpasses that of transfection agents (TA), which achieve 90% knockdown at 50 nM or higher concentrations. Importantly, at 20 nM concentrations of siAXL, SAMS, with no AuNP, showed <50% downregulation at the same concentration (named as Gel(GOSi). Importantly. SAMS exhibit long-term silencing of AXL, even after 16 days.

[0148] SAMS demonstrate effective and dose-responsive gene knockdown in NSCLC cell- derived xenograft (CDX) and patient-derived xenograft (PDX) model.: Our in vivo studies aimed to achieve the following objectives: evaluate the extent of AXL downregulation following a single intratumoral (IT) injection of SAMS and assess the impact of three systemic intravenous (IV) injections on gene knockdown to determine the dose-dependent efficacy of SAMS in a clinically relevant patient-derived xenograft (PDX) model. In the A549 CDX model, a single intratumoral administration of SAMS-siRNA-AXL achieved a 63.5% reduction in AXL expression within 48 hours. Next, we used the NSCLC PDX model TM00784, with high basal AXL expression, which provided a more realistic tumor microenvironment compared with CDX models. In this model, the primary objectives were to assess both gene knockdown efficacy and dose-dependent responses. Two dose concentrations — 0.08 mg/kg and 0.16 mg/kg siRNA- AXL equivalents — were administered IV for three consecutive days. We observed a dose-dependent downregulation of AXL, with 15.6% reduction at 0.08 mg/kg and 27.9% reduction at 0.16 mg/kg. highlighting SAMS’ capability for adjustable gene-silencing through dose modulation. The observed dose-response demonstrates SAMS’ potential for therapeutic window optimization, ensuring effective gene silencing while minimizing off-target effects. [0149] AXL inhibition using SAMS modulates the TME in syngeneic models: To establish the role of AXL inhibition in modulating the TME, we utilized the B16F10 (melanoma) and LLC (Lewis lung carcinoma) syngeneic tumor models, both of which naturally express AXL and serve as ideal systems for this study. Using SAMS-siAXL, we assessed tire downstream effects of AXL inhibition over 96 hours. AXL inhibition significantly enhanced the recruitment and activation of multiple immune populations, including pan T cells, NK cells, and B cells, demonstrating a robust activation of the tumor immune response. Notably, this immune activation was accompanied by a significant increase in PD-1 and TIGIT expression across multiple T cell subsets, indicating activation of the compensatory immune checkpoint mechanism. These results validate the utility of SAMS in syngeneic models and emphasize the necessity of combinatorial strategies targeting AXL, PD-1, and TIGIT to overcome immune resistance in NSCLC.

[0150] SAMS-AXL-TIGIT demonstrates strong therapeutic efficacy in syngeneic tumor models: We examined preclinical efficacy of SAMS co-loaded with AXL and TIGIT siRNAs in Bl 6F 10 and LLC models and demonstrated significant tumor control, indicating the ability of SAMS to effectively target tumors in an immunocompetent setting. Our data also indicated that AXL inhibition, along with PD-1 or TIGIT, leads to some reduction in tumor size, but the triple combination is more effective at controlling tumor growth. These results confirm the therapeutic potential of SAMS-mediated co-inhibition of AXL and TIGIT.

[0151] SAMS is non -toxic in murine models: The in vivo toxicity profile of SAMS carrying non-targeting siRNA was evaluated in immuno-competent 6-week-old CF1 mice at two IV doses: 16 and 32 mg/kg dry weight. Serum biochemistry indicated that SAMS were well tolerated at both dose levels. A slight reduction in serum albumin levels observed at 32 mg/kg dry weight may be attributed to protein corona formation on the SAMS, potentially sequestering free albumin in the blood. Importantly, no significant alterations were noted in liver function parameters. Complete blood cell counts revealed no remarkable changes. Histological analysis of major organs (lungs, heart, kidneys, liver, pancreas, spleen, and brain) showed no significant microscopic lesions. These results establish the safety profile of SAMS at 16 and 32 mg/kg dry weight, supporting their selection as dose levels for further in vivo evaluation (10 mg/Kg BW). Following disassembly, the individual AuNPs, below the renal clearance size threshold, cleared from the body. To assess systemic clearance, we evaluated mouse kidneys 72 h post-lV SAMS administration, well beyond the 24-h window of rapid renal clearance of sub-6 nm AuNPs. Qualitative TEM revealed sparse disassembled-SAMS on the luminal side of renal tubules. [0152] While specific embodiments of the present invention have been shown and described, it should be understood that other modifications, substitutions and alternatives are apparent to one of ordinary skill in the art. Such modifications, substitutions and alternatives can be made without departing from the spirit and scope of the invention, which should be determined from the appended claims.

[0153] Various features of the invention are set forth in the appended claims.

Claims

Claims

1. A supraparticle comprising Au nanoparticle subunits, a stabilizing matrix and a lipidoid.

2. A supraparticle consisting Au nanoparticle subunits, a stabilizing matrix and a lipidoid, and optionally a payload,

3. The supraparticle of claim 1 or 2, carrying a labile molecule payload.

4. The supraparticle of claim 3, wherein the labile molecule payload comprises an RNA molecule.

5. The supraparticle of claim 3, wherein the RNA molecule comprises siRNA.

6. The supraparticle of claim 3, wherein the RNA molecule comprises mRNA solution.

7. Tire supraparticle of claim 3, wherein the labile molecule payload comprises an DNA molecule.

8. The supraparticle of claim 3, wherein the labile molecule payload comprises Crispr moleclue.

9. The supraparticle of claim 3, wherein the labile molecule payload comprises AXL inhibitors or TIGIT-targeting antibodies.

10. The supraparticle of any previous claim, wherein the Au nanoparticle subunits are arranged as lamellar layers arranged in concentric rings.

11. Tire supraparticle of claim 7, comprising a labile molecule in a spacing between the concentric rings.

12. The supraparticle of claim 1 or 2, formed as THPC stabilized AuNP subunits, Gelatin, and Lipidoid complex.

13. The supraparticle of any previous claim, wherein the Au nanoparticle subunits are below 5 nm

14. The supraparticle of any previous claim, wherein the Au nanoparticle submits are in the size range of 0.5 - 2.5 nm.

15. The supraparticle of claim 1 or 2, wherein the lipidoid is GO-C12 orG0-C14.

16. The supraparticle of any previous claim, wherein the stabilizing matrix comprises gelatin.

17. The supraparticle of any previous claim, wherein the stabilizing matrix comprises Bovine Serum Ablumin.

18. A method for forming a supraparticle, comprising adding AuNP -THPC in Acetone to a a lipidoid in Acetone to form a solution, gently mixing the solution, allowing an m-complex to form, adding the m-complex to gelatin solution while stirring.

19. The method according to claim 17, wherein the lipidoid is GO-C12 or G0-C14.

20. The method of claim 17 or 18 wherein tire lipidoid comprises a hydrophobic tail.

21. The method of claim 17 or 18 comprising mixing labile molecules with the lipidoid in Acetone prior to adding the AuNP-THPC.

22. The method according to claim 17, wherein the formation of supraparticles is conducted for at least 500 seconds.