Metabolizable binary gold supraclusters and uses thereof

Abstract

Disclosed herein are metabolizable binary gold supraclusters, which, in some implementations, comprise cationic gold nanoclusters laden with a therapeutic nucleic acid, such as a small interfering RNA (siRNA), and are intertwined through bioresponsive crosslinkers within a hydrophilic polymer matrix. Also provided herein are uses of the supraclusters, e.g., in treatment of diseases such as cancers.

Description

[0001] STDU2-43706.601

[0002] METABOLIZABLE BINARY GOLD SUPRACLUSTERS AND USES THEREOF

[0003] CROSS-REFERENCE TO RELATED APPLICATIONS

[0004] This application claims priority to and the benefit of U.S. Provisional Patent Application No. 63 / 707,912, filed on October 16, 2024, the disclosure of which is incorporated herein by reference in its entirety.

[0005] STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH This invention was made with Government support under contract DE029672 awarded by the National Institutes of Health. The Government has certain rights in the invention.

[0006] FIELD

[0007] Provided herein are metabolizable binary gold supraclusters, which, in some implementations, comprise cationic gold nanoclusters laden with a therapeutic nucleic acid (TNA), such as a small interfering RNA (siRNA), and are intertwined through bioresponsive crosslinkers within a hydrophilic polymer matrix. Also provided herein are uses of the supraclusters, e.g., in treatment of diseases such as cancers.

[0008] BACKGROUND

[0009] Imaging-guided stereotactic ablative radiotherapy (SABR) which precisely delivers high-dose radiation (RT) to tumor sites without disrupting normal tissues stands as a milestone of clinical cancer treatments. Because its effectiveness is primarily confined to localized RT field, growing clinical interests have been directed to synergize SABR with immune checkpoint inhibitors (ICIs) to potentiate systemic immunogenic responses for addressing distant metastasis and preventing local / regional recurrence. However, the ideal immunological planning remains controversial with existing clinical evidence. Radiosensitization through enhancing radiodynamic effects by high atomic number materials (e.g. gold, hafnium, and lanthanides) has manifested a powerful strategy to generate immunostimulatory adjuvants in response to RT, whereas their clinical utility is significantly constrained by chronic toxicities from inadequate clearance and prolonged retention in the body. Ultrasmall gold nanoclusters (< 3 nm) with excellent renal clearance offer unparalleled advantages in addressing this issue (Loynachan et al. Nat. Nanotechnol. 14, 883-890 (2019); Jiang et al. Nat. Nanotechnol. 14, 874-882 (2019); Huang et al. Nat. Nanotechnol. 18, 637-646 (2023); Du et al. Nat. Nanotechnol. 12, 1096-1102 (2017)). Explorative studies have showcased their excellent STDU2-43706.601 radiosensitization capabilities (Schwartz-Duval et al. ACS Nano 18, 1865-1881 (2024); Luo et al. Small 15, 1900968 (2019); Jia et al. ACS Nano 13, 8320-8328 (2019); Broekgaarden et al. Nanoscale 12, 6959-6963 (2020)). However, their potential to act as a simultaneous radiosensitizing adjuvant and immunomodulator to maximize the therapeutic efficacy of the SABR + immunotherapy clinical regimen has yet to be investigated. Additionally, achieving an optimal balance between their biosafety and efficacy profiles for clinical translation requires delicate molecular engineering as well as comprehensive evaluation, which are lacking in these explorative studies.

[0010] Galectin-1 (GaLl), a glycan-binding immunosuppressive mediator, is upregulated in various malignancies such as head and neck cancer (HNC), lung, pancreatic cancers and gliomas. Its overexpression correlates with tumor hypoxia, heightened tumor aggressiveness, and poor survival. In the tumor microenvironment (TME), Gal-l interacts with glycosylated receptors on immune cells to trigger effector T cell apoptosis and regulatory T (Treg) cell expansion, fostering tumor immune privilege. It has been shown that Gal-1 impedes T cell migration into tumor through upregulating programmed death ligand 1 (PD-L1) and galectin- 9 on tumor endothelium, culminating in poor response to anti-programmed death-1 (anti-PDl) therapy (Nambiar et al. J. Clin. Investig. 129, 5553-5567 (2019)). Additionally, tumor Gal-1 increases the release of certain chemokines that enhance recruitment of myeloid-derived suppressor cells (MDSCs) to tumor and premetastatic niches, promoting treatment resistance and metastasis (Nambiar et al. Cancer Res. 83, 3205-3219 (2023)). Thus, downregulating Gal- 1 in TME presents a promising immunotherapeutic strategy to improve response rates; whereas Gal-1 targeting therapeutics are currently unavailable in clinics.

[0011] SUMMARY

[0012] In one aspect, disclosed herein is a supracluster comprising a plurality of gold nanoclusters encapsulated within a reversibly crosslinked polymer matrix.

[0013] In some embodiments, the gold nanoclustcrs comprise surface amine groups. In some embodiments, the gold nanoclusters further comprise one or more capping moieties. In some embodiments, the capping moiety comprises an arginine-rich protein selected from protamine, polyarginine, penetratin, maurocalcine, Pep-1, Tat48-60, polyarginine, and polylysine. In some embodiments, the capping moiety comprises protamine.

[0014] In some embodiments, the gold nanoclusters are complexed with one or more therapeutic nucleic acids. In some embodiments, the one or more therapeutic nucleic acids are RNA or DNA. In some embodiments, the one or more therapeutic nucleic acids are STDU2-43706.601 selected from small interfering RNAs, antisense oligonucleotides, microRNAs, and PIWI- interacting RNAs. In some embodiments, the one or more therapeutic nucleic acids is a small interfering RNA (siRNA). In some embodiments, the siRNA downregulates a cancerpromoting or immunosuppressive target. In some embodiments, the siRNA is an anti-Gal-1 siRNA.

[0015] In some embodiments, the crosslinked hydrophilic polymer matrix comprises a hydrophilic polymer selected from polyethylene glycol, chitosan, hyaluronic acid, alginate, poly(acrylic acid), poly(vinyl alcohol), poly(acrylamide), poly(2-hydroxypropyl methacrylamide), poly (vinylpyrrolidone), poly(7V,7V-dimethylacrylamide), polyethylene imine), and poly(2-methyl-2-oxazoline). In some embodiments, the crosslinked hydrophilic polymer matrix comprises a polyethylene glycol. In some embodiments, the crosslinked hydrophilic polymer matrix comprises a branched polyethylene glycol. In some embodiments, the branched polyethylene glycol is an 8-armed branched polyethylene glycol having terminal amine groups. In some embodiments, the crosslinked hydrophilic polymer matrix comprises a crosslinker having a disulfide bond. In some embodiments, the crosslinker is 3,3’-dithiobis(sulfosuccinimidyl propionate).

[0016] In another aspect, disclosed herein is a pharmaceutical composition comprising a plurality of supraclusters disclosed herein, and a pharmaceutically acceptable carrier.

[0017] In another aspect, disclosed herein is a method of making a supracluster of claim 1 , the method comprising: reacting a gold(III) salt with a reducing agent and a polycationic protein in aqueous solution to generate cationic gold nanoclusters; and mixing the cationic gold nanoclusters with a hydrophilic polymer and a crosslinker comprising a disulfide bond.

[0018] In some embodiments, the gold(III) salt is gold(III) chloride trihydrate. In some embodiments, the reducing agent is glutathione. In some embodiments, the polycationic protein is protamine. In some embodiments, the hydrophilic polymer is a branched polyethylene glycol. In some embodiments, the branched polyethylene glycol is an 8-armed branched polyethylene glycol having terminal amine groups. In some embodiments, the crosslinker is 3,3’-dithiobis(sulfosuccinimidyl propionate).

[0019] In another aspect, disclosed herein is a method of treating cancer in a subject in need thereof, comprising: administering to the subject an effective amount of a pharmaceutical composition disclosed herein (i.e., a pharmaceutical composition comprising a plurality of supraclusters disclosed herein), and subjecting the subject to radiotherapy. STDU2-43706.601

[0020] In some embodiments, the cancer comprises a solid tumor. In some embodiments, the cancer is a head and neck cancer, lung cancer, pancreatic cancer, or a glioma. In some embodiments, the cancer is a head and neck cancer.

[0021] In some embodiments, the administering step comprises intravenous administration.

[0022] In some embodiments, the radiotherapy comprises stereotactic ablative radiotherapy.

[0023] BRIEF DESCRIPTION OF THE DRAWINGS

[0024] FIG. la shows a schematic illustration showing the construction of BSC and its spontaneous degradation in response to reducing pressure in vivo as well as the simultaneous release of therapeutic RNA cargo for immune activation. Meanwhile, BSC accumulates in the tumor and / or metastases via passive tumor targeting, where they radiosensitize SABR for radiodynamic therapy (RDT) and enhanced radio-vaccination. Eventually, the decomposed nano-segments are cleared from the living bodies through both hepatobiliary and renal clearance pathways.

[0025] FIG. lb shows zeta potential profiles of AuNC+, GSC and BSCgai.

[0026] FIG. 1c shows dynamic light scattering (DLS) profiles of AuNC+, GSC and BSCgai.

[0027] FIG. Id shows representative TEM images of BSCgai.

[0028] FIG. le shows TEM images of GSC.

[0029] FIG. If shows DLS profiles of GSC and BSCgaiafter GSH treatment.

[0030] FIG. 1g shows TEM images of GSC and BSCgaiafter GSH treatment.

[0031] FIG. 2a shows WB images of M0C2 cells at 48 h after treatment with PBS, free siRNA (4 pg), GSC (40 pg supracluster), BSCgai(0.5x) (20 pg supracluster containing 2 pg siRNA), BSCgai (40 pg supracluster containing 4 pg siRNA) and lipofectamine 3000 delivering Gal-1 siRNA (4 pg). The mass amount of supraclusters was indicated by the gold content.

[0032] FIG. 2b shows fluorescence microscopy of ROS production in M0C2 cells after incubation with PBS, GSC (50 pg / mL), or BSCgai(50 pg / mL) for 24 h and radiation at 0 or 6 Gy. ROS production was indicated by H2DCFDA (10 pM). Scale bar, 100 pm.

[0033] FIG. 2c shows mean fluorescence intensity (MFI) of ROS production in M0C2 cells after incubation with PBS, GSC, or BSCgaifor 24 h followed by radiation at 6 Gy. P value was determined by one-way analysis of variance (ANOVA); ns, not statistically significant;

[0034] 0.0001. STDU2-43706.601

[0035] FIG. 2d shows y-H2AX foci formation in M0C2 cells after incubation with PBS, GSC, or BSCgai for 24 h followed by radiation at 6 Gy. Red fluorescence indicated IF staining of y-H2AX and blue fluorescence indicated nucleus staining with DAPI. Scale bar, 50 itm.

[0036] FIG. 2e shows quantification of foci number per nucleus in FIG. 2d. P value was determined by ANOVA; ns, not statistically significant: ****P < 0.0001.

[0037] FIG. 2f shows a clonogenic survival assay of M0C2 cells after incubation with PBS, GSC, and BSCgaifollowed by radiation of 2 Gy. P value was determined by two-tailed unpaired Student’s / -test . *P < 0.05; **P < 0.01. Data were expressed as mean ± SD.

[0038] FIG. 2g shows a schematic illustration of in vivo clearance study of supraclusters and clearance pathways.

[0039] FIG. 2h shows accumulated clearance efficiencies (%ID) of GSC and BSCgaiin urine and feces at 1 week and 4 weeks after i.v. injection (n = 4). P value was determined by two- tailed unpaired Student’s t-test. ns, no significant difference.

[0040] FIG. 2i shows biodistribution of gold in heart, liver, spleen, lungs, and kidneys of mice at 1 d, 4 weeks, and 5 months after i.v. injection of GSC (n = 4). P value was determined by one-way ANOVA analysis of 5 groups. ***P = 0.0003;****P < 0.0001.

[0041] FIG. 2j shows absorption and fluorescence spectra of GSC-F7.

[0042] FIG. 2k shows In vivo NIRF imaging of GSC-F7 in subcutaneous M0C2 tumor bearing mice (n = 3).

[0043] FIG. 21 shows quantification of NIRF signals from tumor region in (k) (n = 3).

[0044] FIG. 3a shows representative light-sheet imaging (3D view) of iv injected GSC-F6 in part of a cleared tumor. Scale bar, 0.5 mm.

[0045] FIG. 3b shows 3D rendering of the tumor in FIG. 3a; scale bar, 0.5 mm. The right image shows a zoom-in view of 3D model; scale bar, 50 pin. White arrowhead indicates intravascular GSC-F6 signals and orange arrowhead indicates extravascular GSC-F6 signals.

[0046] FIGS. 3c-3d show representative light-sheet imaging (3D view) of part of a lung with metastases after iv injection of GSC-F6 and tissue clearing. Scale bar, 1.5 mm.

[0047] FIG. 3e shows 3D rendering of the lung tissue in FIGS. 3c-3d. Scale bar, 1.5 mm. The right image shows the GSC-F6 signal distribution in an individual metastatic nodule in the lung. Scale bar, 300 pm.

[0048] FIG. 3f shows a scheme showing the workflow of CyTOF analysis of injected GSC in tumor. GSC was i.v. injected to MOC2 tumor-bearing mice every two days starting at day 0 STDU2-43706.601 for a total of 6 doses (150 pg per dose). At day 1 and 11, tumors were collected from mice for single cell suspension, staining and analysis on mass cytometer.

[0049] FIG. 3g shows a summary of mean signal intensities of gold in tumor cells and CD45+leukocytes from GSC-0, GSC-1 and GSC-6 tumors (n = 3 mice / group). P value was determined by two-tailed unpaired Student’s t-test. ns, not statistically significant. *P (GSC- 1) = 0.0297; *P (GSC-6) = 0.0164. (h) Heatmap of median intensity of gold in different intratumoral immune cell populations (n = 3 mice / group).

[0050] FIG. 3i shows the SPADE tree plots of the gold distribution in different immune cell subpopulations (n = 3 mice / group). Color bar indicates the relative intensity of gold signal.

[0051] FIG. 4a shows a schematic illustration of the experiment design using BSCgaifor synergic in vivo gene silencing and enhanced SABR to treat HNC mouse model. BSCgaior GSC were i.v. injected to mice every 2 days starting at day 0 for a total of 6 doses (150 pg per dose). SABR were applied only to tumor site at day 3, 5, 7 at 6 Gy per fraction.

[0052] FIG. 4b shows tumor growth curves of M0C2 tumor-bearing mice after treatment with PBS, GSC or BSCgai(n = 8 / group). ns, not statistically significant; ****P < 0.0001.

[0053] FIG. 4c shows representative CT images showing X-ray beams were focused and targeted to the tumor site of the orthotopic HNC mouse model. The yellow crossing mark and yellow beams respectively indicated top and side views of X-ray beams.

[0054] FIG. 4d shows tumor growth curves of orthotopic mouse models after combinational supracluster + SABR or control treatments (n = 8 / group). ****P < 0.0001 .

[0055] FIG. 4e shows quantitation of lung metastatic nodules from orthotopic mouse models at 21 days after various treatments (n = 8 / group). **P - 0.0087. ****P < 0.0001.

[0056] FIG. 4f shows representative H & E staining images of the lungs in FIG. 4e. Scale bar: 200 pm.

[0057] FIG. 4g shows confocal fluorescence imaging of Gal- 1 in M0C2 tumor after various treatments. Red fluorescence indicates IF staining of Gal-1 and blue fluorescence indicated DAP1 staining of nuclei. Scale bar, 100 pm.

[0058] FIG. 4h shows tumor growth curves of BSCgai+ SABR comparing to TDG + SABR or aGal-1 + SABR treatments in subcutaneous M0C2 tumor model (n = 8 / group). TDG was intratumorally (i.t.) injected to mice every two days starting from day 0 for 6 doses (150 pg per dose) in total, and aGal-1 was intraperitoneally (i.p.) injected at the same dosage. SABR was given on day 3, 5, 7 with 6 Gy per fraction. ****P < 0.0001.

[0059] FIG. 4i shows survival curves of mice in FIG. 4h (n = 8 / group). STDU2-43706.601

[0060] FIG. 4j shows a summary of maximum survival time (days) obtained after various treatments in different subcutaneous mouse tumor models (denoted as treatment / tumor models).

[0061] FIG. 4k shows tumor growth curves of P029 tumor bearing mice after combinational supracluster + SABR or control treatments (n = 5 / group). *P = 0.0124. ***P = 0.0005.

[0062] FIG. 41 shows quantitation of lung metastatic nodules from P029 tumor bearing mice at 33 days after various treatments (n = 5 / group). *P = 0.0133. ***P = 0.0002.

[0063] FIG. 4m shows representative H & E staining images of the lungs in FIG. 41. Scale bar, 2 mm.

[0064] FIG. 4n shows tumor growth curves of orthotopic mouse models treated with BSCgaiplus SABR comparing to chemoradiation (n = 8 / group). Cisplatin was i.p. injected to mice 1 hr prior to SABR at day 3 (one dose, 2.5 mg / kg). **P = 0.0081, ****P < 0.0001.

[0065] FIG. 4o shows tumor growth curves of orthotopic mouse models treated with PBS alone, or SABR plus one of the followings: PBS, aPDl, BSCgai, or BSCgai + aPDl (n = 8 / group). aPDl was i.p. injected to mice every two days starting from day 0 for 6 doses (150 pg per dose). P value was determined by two-tailed unpaired Student’s z-test. *P = 0.0352; **P - 0.0055, ****p < 0.0001. Data were expressed as mean ± SD.

[0066] FIG. 5a shows a schematic illustration of the workflow of immune analysis.

[0067] FIG. 5b shows fluorescence images of CRT exposure in tumors after various treatments. Scale bar, 50 pm. Green fluorescence indicates IF staining of CRT and blue fluorescence indicates nucleus staining with DAPI.

[0068] FIG. 5c shows quantification of CRT fluorescence intensity in FIG. 5b (n = 5 / group). Statistical significance among PBS, GSC and BSCgai was determined by one-way ANOVA. The other P values were determined by two-tailed unpaired Student’s Z-test. ns, not statistically significant. ***P = 0.0006, ****P < 0.0001.

[0069] FIG. 5d shows HMGB 1 concentration in the plasma of mice after various treatments (n = 5 / group). Statistical significance among PBS, GSC and BSCgai was determined by oneway ANOVA. I'he other P values were determined by two-tailed unpaired Student’s Z-test. ns, not statistically significant. **P (GSC + SABR vs PBS + SABR) = 0.0038, **P (BSCgai+ SABR vs BSCgai) = 0.003, *P (BSCgai + SABR vs PBS + SABR) = 0.0304.

[0070] FIG. 5e shows IFN-y concentration in the plasma of mice after various treatments (n =

[0071] 5 / group). P values were determined by two-tailed unpaired Student’s Z-test. ns, not STDU2-43706.601 statistically significant. *P (GSC vs BSCgai) = 0.0142; *P (GSC + SABR vs PBS + SABR) = 0.0164; *P (BSCgai + SABR vs GSC + SABR) = 0.0469; **P = 0.0095; ***P = 0.0005.

[0072] FIG. 5f shows the percentage of CD8+TCRb+T cells in tumor-infiltrating CD45+leukocytes from mice after various treatments (n = 5 / group). P values were determined by two-tailed unpaired Student’s / -test. ns, not statistically significant. *P = 0.0368; **P = 0.009; ***P (GSC vs BSCgai) = 0.0006; ***P (BSCgai+ SABR vs BSCgai) = 0.0002; ***P (BSCgai + SABR vs PBS + SABR) = 0.0004.

[0073] FIG. 5g shows the percentage of GzmB+cells in tumor-infiltrating CD8+T cells from mice after various treatments (n = 5 / group). P values were determined by two-tailed unpaired Student’s t-test. ns, not statistically significant. *P - 0.0458; **P - 0.0066; ***P = 0.0002.

[0074] FIG. 5h shows the percentage of Ki67+CD8+T cells in tumor- infiltrating CD45+leukocytes from mice after various treatments (n = 5 / group). P values were determined by two-tailed unpaired Student’s t-test. ns, not statistically significant. *P = 0.0319; **P (BSCgaivs GSC) = 0.0045; **P (BSCgai+ SABR vs BSCgai) = 0.0082.

[0075] FIG. 5i shows the percentage of CD4+TCRb+T cells in tumor-infiltrating CD45+leukocytes from mice after various treatments (n = 5 / group). P values were determined by two-tailed unpaired Student’s t-test. ns, not statistically significant. *P = 0.0207; **P (BSCgai vs GSC) = 0.0018; **P (BSCgai + SABR vs BSCgai) = 0.0055; *P (BSCgai+ SABR vs PBS + SABR) = 0.0164.

[0076] FIG. 5j shows the percentage of NK cells in tumor-infiltrating CD45+leukocytes from mice after various treatments (n = 5 / group). P values were determined by two-tailed unpaired Student’s t-test. ns, not statistically significant. *P (BSCgai+ SABR vs GSC + SABR) = 0.0158; *P (BSCgai + SABR vs PBS + SABR) = 0.0106; *P (BSCgai + SABR vs BSCgai) = 0.0185: ***P = 0.0008.

[0077] FIG. 5k shows tSNE-CUBA analysis of tumor infiltrating immune cells from mice after various treatments.

[0078] FIG. 51 shows schema summarizing the immune changes elicited by BSCgai+ SABR treatment. P value was determined by ANOVA or two-tailed unpaired Student's / -test; ns, not statistically significant. Data were expressed as mean ± SD.

[0079] FIG. 6a shows a schematic illustration of a mouse bearing bilateral MOC2 tumors on both flanks while only the right-side tumor was subject to SABR (deemed as irradiated primary tumor). Mice are also i.v. injected with BSCgaior PBS.

[0080] FIG. 6b shows a scheme showing the treatment design for study of abscopal effects. BSCgai (150 pg per dose) or PBS was i.v. injected to mice every two days starting at day 0 for STDU2-43706.601 a total of 6 doses. SABR was applied to right-side tumor at day 3. 5, 7 with 0 or 6 Gy per fraction.

[0081] FIG. 6c shows tumor growth curves of non-irradiated distant tumor (n = 5 / group). **P (BSCgai + SABR w PBS + SABR) = 0.0035.

[0082] FIG. 6d shows percentage of CD45+leukocytes in live cells from non-irradiated distant tumor after various treatments (n = 5 / group). **P = 0.0016.

[0083] FIG. 6e shows the percentage of CD8+TCRb+T cells in CD45+leukocytes from nonirradiated distant tumor after various treatments (n = 5 / group). **P (BSCgai+ SABR vs BSCgai) = 0.0031 ; **P (BSCgai+ SABR vs PBS + SABR) = 0.0017.

[0084] FIG. 6f shows the percentage of CD4+TCRb+T cells in CD45+leukocytes from nonirradiated distant tumor after various treatments (n = 5 / group). *P (BSCgai + SABR vs BSCgai) = 0.0125; **P (BSCgai + SABR vs PBS + SABR) = 0.0043. (g) Percentage of (NK1.1+) NK cells in CD45+leukocytes from non-irradiated distant tumor after various treatments (n = 5 / group). *P (BSCgai + SABR vs BSCgai) = 0.01; **P (BSCgai + SABR vs PBS + SABR) = 0.0433.

[0085] FIG. 6h shows tSNE-CLJBA analysis of tumor infiltrating immune cells from nonirradiated distant tumor after various treatments.

[0086] FIG. 6i shows flow cytometry plots showing the percentage of GzmB+cells in CD8+TCRb+T cells.

[0087] FIG. 6j shows the percentage of GzmB+cells in in CD8+TCRb+T cells from nonirradiated distant tumor after various treatments (n = 5 / group). **P = 0.0035; ****P < 0.0001. P value was determined by ANOVA or two-tailed unpaired Student’s t-test; ns, not statistically significant. Data were expressed as mean ± SD.

[0088] FIG. 7a shows a scheme showing supraclusters loaded with OVA mRNA (termed BSCova) and its stimulation on BMDC.

[0089] FIG. 7b shows flow cytometry plots showing the surface expression of antigen peptide on BMDCs after co-incubation with BSCova.

[0090] FIG. 7c shows flow cytometry plots showing the increased antigen presentation of BMDCs after co-incubation with BSCova.

[0091] FIG. 8a shows a schematic illustration showing the synthesis of AuNC+.

[0092] FIG. 8b shows absorption (yellow) and fluorescence (red) spectra of AuNC+in aqueous solution.

[0093] FIG. 8c shows representative TEM images of AuNC+. STDU2-43706.601

[0094] FIG. 9a shows a scheme showing the detailed crosslinking process of supraclusters.

[0095] FIG. 9b shows agarose gel electrophoresis of gold cluster-siRNA nanocomplexes with different gold-to-siRNA weight ratios (5:1, 10:1, 20:1, and 50:1).

[0096] FIG. 9c shows TEM images of GSC (upper panel) and BSCgai(lower panel).

[0097] FIG. 10 shows the hydrodynamic diameters of GSC and BSCgaiduring incubation in 1 x PBS or DMEM containing 10% FBS (physiologically mimicking condition) for 7 days. Measurements were performed in triplicate. Data were expressed as mean ± SD.

[0098] FIG. 1 la- 1 lb shows dissociation kinetics of supraclusters in the presence of 0 mM, 30 pM, 100 pM, 1 mM, and 10 mM GSH. Measurements were performed in triplicate. Data were expressed as mean ± SD.

[0099] FIGS. 1 lc-1 Id shows magnified TEM images of supraclusters after GSH treatment. Scale bar in magnified image: 10 nm.

[0100] FIG. 12 shows confocal fluorescence images of M0C2 cells treated with BSCcy3 (50 pg / mL) at different time points. Lysosomes were stained with LysoTracker™ Green and nuclei were stained with Hoechst 33342. Scale bar: 10 pm.

[0101] FIG. 13 shows qRT-PCR results showing Gal-1 mRNA level in cells after treatment with PBS, free siRNA (4 pg), GSC (40 pg), or BSCgai(40 pg) for 24 h. P value was determined by one-way analysis of variance (ANOVA) or two-tailed unpaired Student’s t- test. ns, no significant difference. **P = 0.0047. Data were expressed as mean ± SD.

[0102] FIG. 14 shows WB analysis of Gal-1 expression levels in M0C1, mEERL, and P029 cells after treatment with PBS (-) or BSCgai(+) (40 pg) for 48 h.

[0103] FIG. 15a shows a schematic illustration of co-culture experiment. T cells isolated from mouse spleen were activated with T-activator CD3 / CD28 Dynabeads before co-culture. M0C2 cells were pretreated with BSCgai(50 pg / mL) or GSC (50 pg / mL) for 48h before coculture with activated T cells.

[0104] FIG. 15b shows representative flow cytometry plots showing the apoptosis of CD8+T cells and CD4+T cells after co-culture with pretreated M0C2 cells. Cell death was assessed by Annexin V staining.

[0105] FIG. 15c shows percentage of apoptotic CD8+T cells (gated as Annexin- V positive cells among CD8+T cells). **P = 0.0025.

[0106] FIG. 15d shows the percentage of apoptotic CD4+T cells (gated as Annexin-V positive cells among CD4+T cells). **P - 0.0053. P value was determined by two-tailed unpaired Student’s West, ns, no significant difference. Data were expressed as mean ± SD. STDU2-43706.601

[0107] FIG. 16 shows fluorescence microscopy of ROS production in M0C2 cells after incubation with PBS, GSC (50 pg / mL), or BSCgai(50 pg / mL) for 24 h and radiation at 0 or 6 Gy. ROS production was indicated by H2DCFDA (10 pM). Fluorescence images were displayed individually or overlaid with bright-field images. Scale bar: upper panel, 100 pm; lower panel (magnified images), 20 pm.

[0108] FIG. 17a shows photos of colony formation of M0C2 cells at 10 days after treatment with PBS, GSC (50 pg / mL), BSCgai(50 pg / mL) for 48 h and irradiation at 0 or 2 Gy.

[0109] FIG. 17b shows the number of M0C2 colonies formed at 10 days after different treatments. P value was determined by two-tailed unpaired Student’s t-test. ns, no significant difference. ***P (GSC vs BSCgai) = 0.0003. **P (PBS+RT vs GSC+RT) = 0.0019. ***P (BSCgai +RT vs GSC+RT) = 0.0002. Data were expressed as mean ± SD.

[0110] FIG. 18a shows accumulated clearance efficiencies (%ID) of GSC (150 pg) in urine and feces at 1 d, 1 week, 2 weeks, 3 weeks, and 4 weeks after i.v. injection of GSC (n = 4).

[0111] FIG. 18b shows accumulated total clearance efficiencies of plasmonic gold nanoparticles (150 pg) in urine and feces at 1 week, 2 weeks, 3 weeks, and 4 weeks after i.v. injection (n = 4).

[0112] FIG. 19 shows biodistribution of gold in heart, liver, spleen, lungs, and kidneys of mice at 1 week, and 2 weeks after i.v. injection of GSC (n = 4). Statistical analyses among 5 groups were performed by one-way ANOVA. ****p < 0.0001.

[0113] FIG. 20a shows a schematic illustration of GSH depletion by DEM followed by analysis of GSC clearance. DEM was i.p. injected to mice 30 min prior to GSC injection. For control group, mice were pretreated with PBS.

[0114] FIG. 20b shows gold content in blood (%ID / g) at 24 h post-injection of GSC (n = 4). *P = 0.0196.

[0115] FIG. 20c shows gold content (%ID) in urine and feces as well as total clearance rates (from both urine and feces) of GSC (%ID) at 24 h post injection (n = 4). ***P (Urine) = 0.0007; ****P < 0.0001. P value was determined by two-tailed unpaired Student’s t-test. Data were expressed as mean ± SD.

[0116] FIG. 21 shows H&E staining images of major organs (heart, liver, spleen, lung, kidney) of mice at 4 weeks after iv injection of PBS, GSC (150 pg) and BSCgai(150 pg), respectively.

[0117] FIG. 22 shows results of blood chemistry parameters from mice at 4 weeks after iv injection with PBS, GSC (150 pg) and BSCgai(150 pg), respectively (n = 5). Blood chemistry STDU2-43706.601 parameters include aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), gamma-glutamyl transferase (GGT), cholesterol (Choi), blood urea nitrogen (BUN), creatinine (Crea), phosphorus (Phos), albumin (Alb). P value was determined by one-way ANOVA. ns, no significant difference. Data were expressed as mean + SD.

[0118] FIG. 23 shows plasma level of GRP, PCT, and IL-6 from mice at 4 weeks after iv injection with PBS, GSC (150 pg) and BSCgai(150 pg), respectively (n = 5). P value was determined by one-way ANOVA. ns, no significant difference. Data were expressed as mean ± SD.

[0119] FIG. 24a shows fluorescence intensity of major organs (heart, liver, spleen, lung, kidney) and tumor from MOC2 tumor-bearing mice at 1 week after iv injection of GSC-F7 ([Cy7] = 20 pM, 100 pL per mouse) or free Cy7 dye (([Cy7] = 20 pM, 100 pL per mouse), (n = 3). Excitation 745 nm; emission 790 nm. P value was determined by two-tailed unpaired Student’s Mest. *P = 0.0339; ****P < 0.0001.

[0120] FIG. 24b shows biodistribution of gold in heart, liver, spleen, lungs, kidneys and tumors of mice at 24 h, 48 h, and 1 week after i.v. injection of GSC (150 pg per mouse, n = 4). P value was determined by two-tailed unpaired Student's t-test. ***P (24 h) = 0.003; ***P(48 h) =0.0001; ***P (1 wk) = 0.0002.

[0121] FIG. 25a shows In vivo NIRF imaging of GSC-F7 in M0C2 orthotopic mouse model in the oral cavity region and lung with metastases (n = 3).

[0122] FIG. 25b shows quantification of NIRF signals from tumor region in orthotopic M0C2 tumor-bearing mice after iv injection with GSC-F7 ([Cy7] = 20 pM, 100 pL per mouse) or free Cy7 dye (([Cy7] = 20 pM, 100 pL per mouse) (n = 3).

[0123] FIG. 25c shows fluorescence intensity of major organs (heart, liver, spleen, lung, kidney) and tumor from orthotopic M0C2 tumor-bearing mice at 1 week after iv injection of GSC-F7 ([Cy7] = 20 pM, 100 pL per mouse) or free Cy7 dye (([Cy7] = 20 pM, 100 pL per mouse) (n = 3). Excitation 745 nm; emission 790 nm.

[0124] FIG. 26 shows absorption (blue) and emission (red) spectra of GSC-F6.

[0125] FIG. 27a shows a schematic illustration showing the workflow of nanoparticle injection, perfusion, tissue clearing, light-sheet imaging, and 3D reconstruction to visualize GSC-F6 in tumor and lung metastases.

[0126] FIG. 27b shows photos of lung (left) and tumor (right) from MOC2-tumor bearing mice after GSC-F6 injection, whole-body perfusion and tissue clearing. STDU2-43706.601

[0127] FIG. 28 shows images of two-dimensional slices of tumor captured by light-sheet imaging. MOC2 tumor bearing mice were treated with GSC-F6 injection, whole body perfusion and tissue clearing. Green fluorescence indicated vasculature staining by WGA- AF488 perfusion. Red fluorescence indicated signals from GSC-F6.

[0128] FIG. 29 shows a 3D rendering of the GSC-F6 signals (left) and vasculature structure in tumor.

[0129] FIG. 30 shows fluorescence microscopy images of a tumor section after GSC-F6 injection and IF staining with CD31. Scale bar, 50 pm.

[0130] FIG. 31 shows images of two-dimensional slices of lung with metastatic nodules captured by light-sheet imaging. MOC2 tumor bearing mice were treated with GSC-F6 injection, whole body perfusion and tissue clearing. Green fluorescence indicated vasculature staining by WGA-AF488 perfusion. Red fluorescence indicated signals from GSC-F6.

[0131] FIG. 32a fluorescence microscopy image of tissue section of lung with metastases. Green fluorescence indicates IF staining of Ki67 and red fluorescence indicates GSC-F6. Scale bar, 100 pm.

[0132] FIG. 32b shows the number of lung metastatic nodules in lung counted by light-sheet imaging or H&E staining. P value was determined by two-tailed unpaired Student’s t-test. ns, no significant difference.

[0133] FIG. 33 shows the gating strategy for CyTOF analysis.

[0134] FIG. 34a shows representative CT images showing X-ray beams (indicated in light yellow) were focused and targeted to only tumor site during SABR. In this mouse model, M0C2 tumor was inoculated on the right flank. The yellow crossing mark and yellow beams respectively indicated top and side views of X-ray beams.

[0135] FIG. 34b shows tumor growth curves of M0C2 tumor bearing mice after synergistic treatments of BSCgaiand SABR or control treatments (n = 8). P value was determined by two-tailed unpaired Student’s t-test. ***P = 0.0003, **P = 0.0011.

[0136] FIG. 35 shows tumor growth curves of M0C2 tumor bearing mice after synergistic treatments of BSCgai and SABR (8 Gy per fraction, 3 fractions in total) or control treatments (n = 8 / group). P value was determined by two-tailed unpaired Student’s t-test. ****P < 0.001.

[0137] FIG. 36 shows body weights of M0C2 tumor bearing mice after various treatments (n

[0138] = 8 / group). GSC or BSCgai were iv injected with 150 pg per dose at day 0, 2, 4, 6, 8, 10 for 6 doses in total. SABR was applied with 6 Gy per fraction at day 3, 5, 7 for a total of 18 Gy. STDU2-43706.601

[0139] FIG. 37 shows confocal fluorescence imaging of IF staining of Gal-1 in M0C2 tumor after various treatments. GSC or BSCgaiwere iv injected with 150 p.g per dose at day 0, 2, 4, 6, 8, 10 for 6 doses in total. SABR was administered at 6 Gy per fraction at day 3, 5, 7 for a total of 18 Gy. Red fluorescence indicates IF staining of Gal-1 and blue fluorescence indicated DAPI staining of nuclei. Scale bar, 500 pm.

[0140] FIG. 38 shows quantification of Gal-1 intensity in Supplementary Fig. 21 (n = 4 / group). P value was determined by two-tailed unpaired Student’s t-test. ns, no significant difference. ***P = 0.0004, *P = 0.0102.

[0141] FIG. 39 shows IHC staining of Gal-1 in M0C2 tumor after various treatments. Mice were administered with the same treatments as done in FIG. 28.

[0142] FIG. 40a shows tumor growth curves of MOC2 tumor bearing mice after treatments with PBS, TDG, antiGal-1 antibody (aGal-1), or BSCgai(n = 8 / group). TDG was intratumorally (i.t.) injected to mice every two days starting from day 0 for 6 doses (150 pg for each dose) in total, and aGal-1 was intraperitoneally (i.p.) injected at the same dosage. P value was determined by two-tailed unpaired Student’s t-test. ns, no significant difference. ****P < 0.0001.

[0143] FIG. 40b shows survival analyses of mice in (a) (n - 8 / group).

[0144] FIG. 41a shows tumor growth curves of P029 tumor bearing mice after treatments with PBS, GSC (150 pg per dose for 6 doses), or BSCgai(150 pg per dose for 6 doses) (n = 5 / group). ns, no significant difference. **P = 0.0062.

[0145] FIG. 41b shows quantification of lung metastasis nodules from mice mentioned in FIG. 41a (n = 5 / group). Lungs were harvested at day 21. ns, no significant difference. ***P = 0.0001.

[0146] FIG. 41c shows H&E staining images of lungs from in FIG. 41b. Scale bar, 200 pm.

[0147] FIG. 41d shows survival analyses of P029 tumor bearing mice after various treatments (n = 8 / group). P value was detemrined by two-tailed unpaired Student’s t-test. Data were expressed as mean ± SD.

[0148] FIG. 42a shows H&E staining images of lungs from orthotopic MOC2 tumor-bearing mice at 21 days after the first treatment. Mice were treated with PBS, cisplatin (2.5 mg / kg for only 1 dose), or BSCgai(150 pg per dose for 6 doses) and SABR treatment (6 Gy per fraction for 3 fractions). STDU2-43706.601

[0149] FIG. 42b shows quantification of lung metastasis nodules from mice mentioned in FIG. 42a (n = 8 / group). P value was determined by two-tailed unpaired Student’s / -test. ****? < 0.0001. *P = 0.0314.

[0150] FIG. 43a shows tumor growth curves of orthotopic M0C2 tumor bearing mice after treatments with cisplatin (1.25 mg / kg, one dose) + SABR (6 Gy per fraction, 3 fractions) or BSCgai (75 p per dose, 6 doses) + cisplatin (1.25 mg / kg, one dose) + SABR (6 Gy per fraction, 3 fractions) (n = 8 / group). P value was determined by two-tailed unpaired Student’s / -test. ***P = 0.0003.

[0151] FIG. 43b shows Survival analyses of mice in (a) (n = 8 / group).

[0152] FIG. 44a shows percentage of antigen presenting mature (GDI lc+MHC-II+) DCs among CD45+leukocytes in TDLNs from mice after various treatments (n = 5). MHC-I1, major histocompatibility complex class II. P value was determined by one-way analysis of variance or two-tailed unpaired Student’s t-test. ***P = 0.0007; ns, no significant difference. Data were expressed as mean ± SD.

[0153] FIG. 44b shows the gating strategy for the analysis of DCs in TDLNs.

[0154] FIG. 45 shows the gating strategy of flow cytometry analysis of CD8+T cells, CD4+T cells, NK cells and PMN-MDSC.

[0155] FIG. 46 shows the gating strategy of flow cytometry analysis of CD8+GzmB+T cells, CD8+Ki67+T cells and CD4+CD25+Foxp3+Tregs.

[0156] FIG. 47 shows the gating strategy of flow cytometry analysis of Ml (F4 / 80+MHC-II+) and M2 (F4 / 80+MHC-II CD206+) TAMs.

[0157] FIG. 48a-48b show percentages of Ml (FIG. 48a) and M2 (FIG. 48bb) macrophages in tumor associated macrophages (TAMs) from mice after various treatments (n = 5 / group). P value was determined by one way ANOVA or two-tailed unpaired Student’s / -test, ns, no significant difference. In FIG. 48a, *P (GSC + SABR vs PBS + SABR) = 0.0446. *P (BSCgai+ SABR vs GSC + SABR) = 0.0144. In FIG. 48b, *P (BSCgaivs GSC) = 0.0255.

[0158] FIG. 49a shows the percentage of PMN-MDSCs in tumor-infiltrating CD45+leukocytes from mice after various treatments (n = 5 / group). P values were determined by two-tailed unpaired Student’s t-test. ns, not statistically significant. *P = 0.0148; **P = 0.0088.

[0159] FIG. 49b shows the percentage of Tregs in tumor-infiltrating CD45+leukocytes from mice after various treatments (n = 5 / group). P values were determined by two-tailed unpaired STDU2-43706.601

[0160] Student’s t-test. ns, not statistically significant. *P (GSC vs BSCgai) = 0.0315; *P (GSC + SABR Vi BSCgai + SABR) = 0.0203.

[0161] FIG. 49c shows the ratio of CD8+T cells to Treg in tumor-infiltrating CD45+leukocytes from mice after various treatments (n = 5 / group). P values were determined by two-tailed unpaired Student’s t-test. ns, not statistically significant. *P (GSC vs BSCgai = 0.0117; *P (GSC + SABR w BSCgai+ SABR) = 0.0344; *P (PBS + SABR vs BSCgai+ SABR) = 0.0235.

[0162] FIG. 50 shows tumor growth curves of non-irradiated distant tumor (n = 5 / group). ***P ((BSCgai + SABR Vi PBS + SABR) = 0.0003; ***P ((BSCgai + SABR vs BSCgai) = 0.0002. P value was determined by two-tailed unpaired Student’s t-test. ns, no significant difference. Data were expressed as mean ± SD.

[0163] FIG. 51a shows flow cytometry plots showing the percentage of CD1 lc+MHC-II+mature dendritic cells in CD45+leukocytes from non-irradiated distant tumor after various treatments (n = 5 / group).

[0164] FIG. 51b shows the percentage of CD1 lc+MHC-II+mature dendritic cells in CD45+leukocytes from non-irradiated distant tumor after various treatments (n = 5 / group). **P = 0.0043; 0.0001.

[0165] FIG. 51c shows the percentage of MHC-IBCD206’ Ml macrophages in TAMs from non-irradiated distant tumor after various treatments (n = 5 / group). ***P = 0.0005. P value was determined by one-way analysis of variance or two-tailed unpaired Student’s / -lesl. ns, no significant difference. Data were expressed as mean ± SD.

[0166] FIG. 52a shows a schematic illustration of treatment plan for anti-CD8 (aCD8), anti- CD4 (aCD4), and anti-NKl.l (aNKl. l) depletion studies.

[0167] FIG. 52b shows the validation of immune cells depletion by flow cytometry analysis of splenic leukocytes (n = 5 / group). ****P < 0.0001.

[0168] FIG. 52c shows tumor growth curves of irradiated tumors after various treatments (n = 5 / group). ns, no significant difference.

[0169] FIG. 52d shows tumor growth curves of non-irradiated tumors after various treatments (n = 5 / group).

[0170] FIG. 52e shows the statistical analysis of tumor volumes on day 15 in FIG. 52d. ***P (BSCgai + SABR + aNKl.l) = 0.0002, ****P < 0.0001. P value was determined by one-way analysis of variance or two-tailed unpaired Student’s Mest. Data were expressed as mean ± SD. STDU2-43706.601

[0171] DETAILED DESCRIPTION

[0172] Disclosed herein are metabolizable binary gold supraclusters (referred to herein as BSCgai) which comprise cationic gold nanoclusters laden with a therapeutic nucleic acid (TNA), such as a small interfering RNA (siRNA), and intertwined through bioresponsive crosslinkers within a hydrophilic polymer matrix (FIG. la, FIG. 8a). The large supraclusters are strategically designed for passive targeting and retention in primary tumors and metastases. In response to reductive pressure in vivo, crosslinkers are cleaved to initiate the breakdown of supraclusters to miniscule fragments for gradual clearance and the liberation of the TNA (e.g., siRNA).

[0173] In an exemplary embodiment, the TNA is an siRNA that silences Gal-1 (FIG. la); these supraclusters are termed BSCgai. In multiple HNC models, BSCgaipotentiated SABR and resulted in remarkably improved tumor / metastasis inhibition and survival that outperformed the combination of SABR with Gal-1 inhibitors, blocking antibodies, chemotherapy, or anti-PDl treatment. BSCgaigreatly boosted immunogenicity and immune surveillance of SABR though a bifunctional mechanism involving immune checkpoint silencing and augmentation of radio- vaccination effects, culminating in the regression of not only primary tumors but also non-irradiated distant tumors through reinforced abscopal effects.

[0174] Definitions

[0175] Unless otherwise defined herein, scientific and technical terms used in connection with the present disclosure shall have the meanings that are commonly understood by those of ordinary skill in the art. For example, any nomenclatures used in connection with, and techniques of, cell and tissue culture, molecular biology, immunology, microbiology, genetics, and protein and nucleic acid chemistry and hybridization described herein are those that are well known and commonly used in the art. The meaning and scope of the terms should be clear; in the event, however of any latent ambiguity, definitions provided herein take precedent over any dictionary or extrinsic definition. Further, unless otherwise required by context, singular terms shall include pluralities and plural terms shall include the singular.

[0176] As used herein and in the appended claims, the singular forms “a”, “an”, and “the” include plural reference unless the context clearly dictates otherwise.

[0177] As used herein, the term “and / or” includes any and all combinations of listed items, including any of the listed items individually. For example, “A, B, and / or C” encompasses A, STDU2-43706.601

[0178] B, C, AB, AC, BC, and ABC, each of which is to be considered separately described by the statement “A, B, and / or C.”

[0179] For the recitation of numeric ranges herein, each intervening number there between with the same degree of precision is explicitly contemplated. For example, for the range of 6- 9, the numbers 7 and 8 are contemplated in addition to 6 and 9, and for the range 6.0-7.0, the number 6.0, 6.1, 6.2, 6.3, 6.4, 6.5, 6.6, 6.7, 6.8, 6.9, and 7.0 are explicitly contemplated.

[0180] “Administration” encompasses any administration of the agents disclosed herein to a subject. Administration may include, for example, intratumoral, peritumoral, injection into blood vessels that feed or surround tumors, intravenous, intraperitoneal or other administration to the subject.

[0181] As used herein, the terms “condition,” “disease,” and “disorder” are used interchangeably.

[0182] An “effective amount” of a compound or composition refers to an amount sufficient to elicit a desired biological response (e.g., treating a condition). As will be appreciated by those skilled in the art, the effective amount of a compound may vary depending on such factors as the desired biological endpoint, the pharmacokinetics of the compound, the condition being treated, the mode of administration, and the age and health of the subject. An effective amount encompasses therapeutic and prophylactic treatment. For example, in treating cancer, an effective amount of a compound or composition may reduce tumor burden or stop the growth or spread of a tumor.

[0183] A “therapeutically effective amount” of a compound or composition is an amount sufficient to provide a therapeutic benefit in the treatment of a condition, or to delay or minimize one or more symptoms associated with the condition. In some embodiments, a therapeutically effective amount is an amount sufficient to provide a therapeutic benefit in the treatment of a condition or to minimize one or more symptoms associated with the condition. A therapeutically effective amount of a compound means an amount of therapeutic agent, alone or in combination with other therapies, that provides a therapeutic benefit in the treatment of the condition. The term “therapeutically effective amount” can encompass an amount that improves overall therapy, reduces or avoids symptoms or causes of the condition, or enhances the therapeutic efficacy of another therapeutic agent.

[0184] A “subject” to which administration is contemplated includes, but is not limited to, a human (i.e., a male or female of any age group, e.g., a pediatric subject (e.g., infant, child, adolescent) or adult subject (e.g., young adult, middle-aged adult, or senior adult)) and / or other non-human animals, for example, mammals (e.g., primates (e.g., cynomolgus monkeys, STDU2-43706.601 rhesus monkeys); commercially relevant mammals such as cattle, pigs, horses, sheep, goats, cats, and / or dogs) and birds (e.g., commercially relevant birds such as chickens, ducks, geese, and / or turkeys).

[0185] As used herein, the terms “treatment,” “treat,” and “treating” refer to reversing, alleviating, delaying the onset of, or inhibiting the progress of a disease or condition, or one or more signs or symptoms thereof. In some embodiments, “treatment,” “treat,” and “treating” require that signs or symptoms of the disease disorder or condition have developed or have been observed. In other embodiments, treatment may be administered in the absence of signs or symptoms of the disease or condition. For example, treatment may be administered to a susceptible individual prior to the onset of symptoms (e.g., in light of a history of symptoms and / or in light of genetic or other susceptibility factors). Treatment may also be continued after symptoms have resolved, for example, to delay or prevent recurrence.

[0186] Disclosed herein are supraclusters comprising a plurality of gold nanoclusters encapsulated within a reversibly crosslinked polymer matrix.

[0187] The gold nanoclusters (e.g., cationic gold nanoclusters (AuNC+) having surface amine groups) can be prepared by conventional methods, including those described in the Examples and the references cited therein. The compositions and methods of the present disclosure encompass any AuNC+s known in the art. The AuNC+s utilized herein include those synthesized by any method, including by template, ligand, reduction and other methodologies known in the art. In some embodiments, the AuNC+s are synthesized by a reduction reaction. Exemplary reducing agents include glutathione, sodium borohydride, ascorbic acid, citrate, TCEP, etc. Exemplary capping ligands include glutathione, BSA, Captopril, PANAM, etc. The gold nanoclusters may comprise any number of gold atoms per cluster, for example, from about 4 to 1000 atoms per cluster. Exemplary AuNC++ synthesis methods include those described in: Liu et al. J Mater Chem B 2019, 7(44):6924-6933; Jiang, et al. Bioconjugate Chem. 2020, 31(5): 1522-1528; Zhang et al. Adv. Healthc. Mater. 2014, 3(1);133-141 ; Lakshmi et al. Colloids Surf. B Biointerfaces, 2019, 178:230-237. One exemplary synthesis method involves mixing gold(III) chloride trihydrate with glutathione and protamine sulfate at a molar ratio of 13.3:19:1, followed by gentle shaking at 70 °C for 24 hours.

[0188] In some embodiments, the nanoclusters are cationic in order to facilitate complexing with oligonucleotides. Positive charges can be imparted to the nanoclusters by selection of suitable components. For example, polycationic proteins or peptides may be added to the STDU2-43706.601 reaction mixture during nanocluster formation as capping moieties. For example, arginine- rich or lysine-rich proteins or peptides such as protamine or cell penetrating peptides (CPPs) such as penetratin, maurocalcine, Pep-1, Tat48-60, polyarginine, or polylysine.

[0189] In some embodiments, for example, the nanoclusters are formed as follows: reduction of a gold(III) salt, such as gold(III) chloride trihydrate, can be conducted in aqueous solution using glutathione as a reducing agent. The glutathione also serves as a capping agent on the nanoparticles. Further inclusion of a polycationic protein, for example, an arginine- rich or lysine-rich protein in the reaction mixture, such as protamine and or penetrating peptides (CPPs) such as penetratin, maurocalcine, Pep-1, Tat48-60, polyarginine, or polylysine, produces cationic gold nanoclusters. The doping (molar) ratio of cationic proteins or peptides to the reducing agent (e.g. glutathione) can be maintained at approximately 1-10%, depending on the desired stability and positive charge. In some embodiments, the doping ration of cationic proteins or peptides to the reducing agent is maintained at about 1%, about 2%, about 3%, about 4%, about 5%, about 6%, about 7%, about 8%, about 9%, or about 10%.

[0190] The surface charge of the nanoclusters can be about 10 mV to about 40 mV, and the hydrodynamic diameter of the nanoclusters can be in a range of about 1 nm to about 5 nm. In some embodiments, the nanoclusters have a surface charge of about 10 mV, about 15 mV, about 20 mV, about 25 mV, about 30 mV, about 35 mV, or about 40 mV.

[0191] In some implementations, the supraclusters comprise complexes of positively charged gold nanoclusters. The positive surface charge of the gold nanoclusters allows them to complex to TNAs via electrostatic interactions. The nucleic acid-to-gold weight ratio can be approximately 1-10%. In some embodiments, the nucleic acid-to-gold weight ratio is about 1%, about 2%, about 3%, about 4%, about 5%, about 6%, about 7%, about 8%, about 9%, or about 10%. Gentle shaking at room temperature for 10 min to 1 h will facilitate the formation of gold nanocluster-TNA complexes. In some embodiments, gentle shaking at room temperature for about 10 min, about 15 min, about 20 min, about 25 min, about 30 min, about 35 min, about 40 min, about 45 min, about 50 min, about 55 min, or about 60 min facilitates the formation of gold nanocluster-TNA complexes.

[0192] Therapeutic nucleic acids (TNAs) include any oligonucleotides having a biological effect in vivo, for example, a therapeutic effect, for example, inducing or enhancing an immunotherapeutic effect. In some embodiments, the TNA is RNA. In some embodiments, the TNA is DNA. In some embodiments, the TNA comprises DNA and / or RNA nucleotides or non-natural nucleotides. STDU2-43706.601

[0193] In some embodiments, the TNA binds to a selected target polynucleotide sequence and effects a change in the target sequence function. For example, TNAs include gene silencing or downregulating oligonucleotides (e.g., small interfering RNAs (siRNAs), antisense oligonucleotides (ASOs), short-hairpin RNAs (shRNAs), microRNAs (miRNA), PlWI-interacting RNAs (piRNA), dicer-substrate RNAs, DNAzymes, small circular RNAs, aptamers targeting a gene or messenger RNAs), which can be used to modulate gene expression. TNAs also include oligonucleotides used for targeting effector molecules to a target sequence to facilitate structural or functional modifications (e.g., insertion, deletions, mutations, chemical modifications) on a target sequence which does not necessarily result in gene silencing or downregulation (e.g., guide RNAs, snoRNAs, snoRNA-like RNAs). In some embodiments, the TNA is a therapeutic mRNA (e.g., a neoantigen-encoding mRNA such as WGc-043 or mRESVIA, an antibody or antigen-encoding mRNA, or an mRNA encoding a CRISPR gene editing system). In some embodiments, the TNA is a siRNA molecule, a miRNA molecule, a small non-coding RNA (sncRNA) molecule, a piRNA, or an ASO. In particular embodiments, the TNA is an siRNA molecule. The TNA may be an agent that specifically disrupts a selected cancer-associated factor’s expression, for example, being selected from the following targets: CTLA, PD-1, PD-L1, XIAP, CD47, YAP1, BCL2, STAT3, STAT6, CLU, and HRAS.

[0194] In some embodiments, the TNA comprises an mRNA or other expression vector which induces the expression of a therapeutic protein, for example, a cytotoxic protein or an antigen which promotes an immunotherapeutic response, for example, resulting in antigen presentation and T-cell stimulation. Exemplary mRNA TNAs include, for example NY-ESO- 1, MAGE- A3, tyrosinase, and TPTE, or a patient- specific neoantigen. In some embodiments, the TNA comprises an mRNA or other expression vector which induces the expression of an immunologically active agent such proinflammatory cytokines, or immune-modulatory antibody or fragment thereof, such as an scFV. Exemplary expression products include IL12 CD137 ligand, INF, CD70, CD80, CD86, and CD40L

[0195] The therapeutic nucleic acids are not limited by size. In some embodiments, the TNAs are about 10 to about 10,000 nucleotides in length. In some embodiments, the TNAs are about 10 to about 50 nucleotides in length. For example, the TNAs may be about: 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 45, 46, 47, 48, 49, 50, at 100, 200, 500, one thousand, two thousand, three thousand, four thousand, five thousand, six thousand, seven thousand, eight thousand, nine thousand, or ten thousand or greater nucleotides in length. In some STDU2-43706.601 implementations the TNAs can be greater than 50 nucleotides in length, while having a target binding or antisense region of less than 50 nucleotides in length. For example, guide RNAs and snoRNAs have antisense elements of less than 50 nucleotides in length but the entire length of the molecule can be greater than 50 nucleotides, 100 nucleotides, 150 nucleotides, 200 nucleotides, or more in length. In other embodiments, the TNAs are more than 50 nucleotides in length (e.g., in the case of therapeutic mRNAs).

[0196] In some embodiments, the TNA downregulates or silences a cancer-promoting or immunosuppressive target. In an exemplary embodiment, the target is Gal-1. Accordingly, in some embodiments, the therapeutic nucleic acid is an anti-Gal-1 siRNA.

[0197] In some embodiments, the TNA is an siRNA selected from inclisiran, patisiran. givosiran (ALN-AS1), or lumasiran (ALN-G01). In some embodiments, the TNA is an ASO selected from fomivirsen, pegaptanib, mipomersen, nusinersen, inotersen, defibrotide, eteplirsen, golodirsen, viltolarsen, and casimersen.

[0198] The gold nanoclustcrs complexed to the therapeutic nucleic acid arc incorporated into supraclusters by their admixture with one or more polymer species and a subsequent crosslinking process resulting in the incorporation of polymer and numerous nanoclusters in a complex. In some embodiments, the nanoclusters are embedded within a matrix comprising crosslinked polymer. In some embodiments, the nanoclusters are linked to each other by the polymer. In some embodiments, a cross-linking agent is admixed with the polymer to promote crosslinking.

[0199] The supraclusters will be biodegradable, i.e., they will degrade over time in response to cellular conditions or processes. In one embodiment, the bonds forming the cross-links are biodegradable. In some embodiments, the polymers themselves, are biodegradable, for example, comprising copolymers comprising bioresponsive bonds. In one embodiment, the biodegradable crosslink comprises a disulfide bond.

[0200] In some embodiments, the polymer is hydrophilic. Exemplary hydrophilic polymers include, for example polyethylene glycol (PEG), chitosan, hyaluronic acid, alginate, poly(acrylic acid) (PAA), poly(vinyl alcohol) (PVA) or poly(acrylamide) (PAAm), poly(2- hydroxypropyl methacrylamide) (PHPMA), poly(vinylpyrrolidone) (PVP), poly(A,A- dimethylacrylamide) (PDMA), poly(ethylene imine) (PEI) and poly(2-methyl-2-oxazoline) (PMeOx), for example. In some embodiments, the polymers are branched, for example, in various embodiments comprising two, three, four, five, six, seven, eight, or more arms. In some embodiments, two or more different species of polymer are utilized, for example, two STDU2-43706.601 polymers having opposite charges or a pair of polymer species wherein each is functionalized with complementary reactive groups.

[0201] In one embodiment, the polymer comprises a polyamine species and is cross linked by reaction with an amine-responsive crosslinker, for example, a crosslinking agent comprising any of aldehydes, NHS esters, isothiocyanates, isocyanates, imidoesters, oxiranes, acyl azides, sulfonyl chlorides, glyoxals, epoxides, carbonates, aryl halides, carbodiimides. In one embodiment, the polymer comprises a hydrophilic polymer grafted or functionalized with amines. In one embodiment, the polymer comprises a polypeptide comprising an N-terminal amine and one or more lysine residues comprising amines.

[0202] Exemplary cross-linking agents include, for example, 3,3'- dithiobis(sulfosuccinimidylpropionate) (DTSSP), l-ethyl-3-(3-dimethylamino propyl) carbodiimide hydrocholoride, glutaraldehyde, MA'-dicyclohexylcarbodiimidc (DCC), polycarboxylic acids, 1 ,4-butanediol diglycidyl ether (BDDE), and trilysine. One exemplary crosslinking process includes mixing gold-TNA complexes with 8-arm PEG-amine and DTSSP at a weight ratio of 1:10:20, followed by gentle shaking at 4 °C overnight in the dark. After reaction, the crosslinked nanoparticles can be washed by ultracentrifugation at 2,500g for 30 min at 4 °C using a 10-kDa cutoff filter. The weight ratio of crosslinking reagents can be adjusted depending on the desired size of crosslinked nanoparticles. Crosslinking reactions can be performed at 4 °C to preserve RNA stability.

[0203] Alternative crosslinking chemistries include any known in the art wherein the crosslinked product is biodegradable. Exemplary processes facilitating the formation supraclusters include gelation reactions, charge interactions, hydrogen bonding, sulfone-thiol bonding, azide-alkyne cycloaddition, and photo-bonding facilitated by photosensitive functional groups.

[0204] The supraclusters may contain any number of nanoclusters, for example, in various embodiments, the supraclusters may comprise about 10-100, 100-500, 500-1000, 1000-5000, 5000-10000, 10000-20000, 20000-50000, 50000-100000, 100000-500,000, up to 106, up to 107, or up to or more than 108nanoclusters per supracluster. The average diameter of the supraclusters may be, in some embodiments, in the range of 10 to 500 nm, for example, in the range of 10-20, 20-50, 50-75, 75-150 nm, 150-300, or 300-500 nm, for example about 10, about 20, about 30, about 40, about 50, about 60, about 70, about 80, about 90, about 100, about 110, about 120, about 130 nm, about 140, about 150, about 160, about 180, about 200, about 250, about 300, about 400, or about 500 nm. In some cases, the foregoing values represent average numbers of nanoclusters within each supracluster and average supracluster STDU2-43706.601 sizes. It will be understood that a plurality of supraclusters formed by the methods of the invention will generally comprise a range of sizes. Where it is desirable to obtain supraclusters of a desired size range from heterogeneous and polydisperse batches of supraclusters, standard techniques known in the art may be applied. Exemplary techniques to isolate monodisperse supraclusters include, for example, size exclusion chromatography, size selective precipitation, magnetic field flow fractionation filtration, density gradient centrifugation, cross-flow filtration, gel electrophoresis or other mass- or density-based purification techniques.

[0205] In some embodiments, the supraclusters are formed by admixture of the nanocluster- TNA complexes with a hydrophilic polymer and a crosslinker containing a disulfide bond. This produces supraclusters in which the AuNC+-TNA complexes are directly or indirectly linked to each other, forming supraclusters of AuNC+-TNAs encapsulated within an interconnected polymer matrix. The use of a disulfide bond-containing crosslinker allows the crosslinked polymer to degrade in response to reducing pressure in vivo, with concomitant release of the therapeutic nucleic acid.

[0206] In some embodiments, the hydrophilic polymer is a polyalkylene glycol, such as polyethylene glycol. In some embodiments, the hydrophilic polymer is a branched polyethylene glycol, such as a 4-armed polyethylene glycol or an 8-armed polyethylene glycol. In other embodiments, the hydrophilic polymer is a straight-chain polyethylene glycol. The polyalkylene glycol, such as polyethylene glycol, can be terminated by at least two reactive groups, such as amino groups, to provide a reactive handle for crosslinking. For example, in some embodiments, the AuNC+-TNA complexes are reacted with 8-arm PEG- amine, along with an amine-reactive crosslinker such as a bis(succinimidyl) or bis (sulfosuccinimidyl) compound, such as 3,3’-dithiobis(sulfosuccinimidyl propionate).

[0207] Prior to crosslinking, the hydrophilic polymer can have an average molecular weight of about 1 kDa to about 50 kDa, or about 10 kDa to about 40 kDa.

[0208] Pharmaceutical

[0209] The disclosed supraclusters may be incorporated into pharmaceutical compositions suitable for administration to a subject (such as a patient, which may be a human or nonhuman).

[0210] The pharmaceutical compositions may include pharmaceutically acceptable carriers. The term “pharmaceutically acceptable carrier,” as used herein, means a non-toxic, inert solid, semi-solid or liquid filler, diluent, encapsulating material or formulation auxiliary of STDU2-43706.601 any type. Some examples of materials which can serve as pharmaceutically acceptable carriers are pyrogen- free water; isotonic saline; Ringer’s solution; ethyl alcohol, and phosphate buffer solutions.

[0211] The supraclusters may be formulated for administration by, for example, injection. Techniques and formulations may generally be found in “Remington' s Pharmaceutical Sciences,” (Meade Publishing Co., Easton, Pa.).

[0212] Methods of Use

[0213] The disclosed supraclusters can be administered to a subject and used to treat a variety of disorders. In some embodiments, disclosed herein is a method of treating cancer in a subject in need thereof, comprising administering to the subject an effective amount of a pharmaceutical composition comprising a plurality of supraclusters disclosed herein, and subjecting the subject to radiotherapy. As disclosed herein, supraclusters can synergize localized radiosensitization with systemic specific target silencing for personalized radi oimmunotherapy .

[0214] In some embodiments, the cancer to be treated comprises a solid tumor. For example, in some embodiments, the cancer is a head and neck cancer, lung cancer, pancreatic cancer, or a glioma. In some embodiments, the cancer is a head and neck cancer.

[0215] The radiotherapy can be any suitable radiation-based therapy, including therapy using external or internal radiation sources. In these methods, the supraclusters act as radiosensitizing agents that enhance the therapeutic effect of the administered radiation. In some embodiments, the radiotherapy comprises fractionated radiation. In some embodiments, the radiotherapy comprises brachytherapy. In some embodiments, the radiotherapy comprises stereotactic ablative radiotherapy. In some embodiments, the radiotherapy comprises photothermal therapy, wherein the supraclusters generate heat and resultant tumor killing upon irradiation by a suitable source, for example NIR light. In some embodiments, the radiotherapy is photodynamic therapy (PDT) wherein supraclusters act as photosensitizing agents, wherein, upon irradiation by suitable sources, such as visible light, the supraclusters promote generation of reactive oxygen species (ROS), killing the targeted cells.

[0216] The supraclusters described herein may be used in combination with other known therapies. Administered “in combination,” as used herein, means that two (or more) different treatments are delivered to the subject during the course of the subject's affliction with the disorder, e.g., the two or more treatments are delivered after the subject has been diagnosed with the disorder and before the disorder has been cured or eliminated or treatment has STDU2-43706.601 ceased for other reasons. In some embodiments, the delivery of one treatment is still occurring when the delivery of the second begins, so that there is overlap in terms of administration. This is sometimes referred to herein as “simultaneous” or “concurrent delivery.” In other embodiments, the delivery of one treatment ends before the delivery of the other treatment begins. In some embodiments of either case, the treatment is more effective because of combined administration. For example, the second treatment is more effective, e.g., an equivalent effect is seen with less of the second treatment, or the second treatment reduces symptoms to a greater extent, than would be seen if the second treatment were administered in the absence of the first treatment, or the analogous situation is seen with the first treatment. In some embodiments, delivery is such that the reduction in a symptom, or other parameter related to the disorder is greater than what would be observed with one treatment delivered in the absence of the other. The effect of the two treatments can be partially additive, wholly additive, or greater than additive. The delivery can be such that an effect of the first treatment delivered is still detectable when the second is delivered.

[0217] In some embodiments, the supraclusters and radiotherapy described herein are used in combination with other therapeutic treatment modalities, including surgery, chemotherapy, hormone therapy, immunotherapy, cryotherapy, targeted therapy, cell therapy, theragnostic treatment, and thermotherapy. Combination therapies may advantageously utilize lower dosages of the administered agent and / or other chemotherapeutic agent, thus avoiding possible toxicities or complications associated with the various therapies.

[0218] In some embodiments, the supraclusters described herein are administered with at least one additional therapeutic agent, such as a chemotherapeutic agent. In certain embodiments, the compound described herein is administered in combination with one or more additional chemotherapeutic agents. The chemotherapeutic agent may be a chemotherapeutic agent identified on the “A to Z List of Cancer Drugs” published by the National Cancer Institute.

[0219] In some embodiments, the supraclusters are administered for use as imaging agents. For example, supraclusters may be used to image tumors or other target structures. Imaging may be by any suitable modality, including for example, X-ray computed tomography, NIR fluorescent imaging, positron emission tomography (PET), and magnetic resonance imaging (MRI). STDU2-43706.601

[0220] Materials. All chemicals were purchased from Sigma- Aldrich unless otherwise stated. Gold(III) chloride trihydrate (520918), protamine (P4005), N-Ethyl-N' -(3- dimethylaminopropyl)carbodiimide hydrochloride (EDC, E6383), indium chloride (203440), DAPI (MBD0015), L-GSH (G4251) were ordered from Sigma-Aldrich. 8-Arm PEG-NH2 HCI MW 40k was ordered from Advanced Biochemicals (8AP0804-1 G). 3,3'- dithiobis(sulfosuccinimidyl propionate) (DTSSP, 21578), Silencer® Select Pre-designed Gal- 1 siRNA (s201592), RNA gel loading dye (R0641), SYBR™ Gold Nucleic Acid Gel Stain (SI 1494), TRIzol™ reagent (15596026), Alexa Fluor™ 647-conjugated donkey anti-rabbit secondary antibody (A-31573), Alexa Fluor 488 conjugated wheat germ agglutinin (WGA- AF488, W11261), FITC conjugated goat anti-rabbit secondary antibody (65-6111) were ordered from Thermo Fisher Scientific. Immunotag™ mouse hmgbl ELISA kit was ordered from G-Biosciences (IT5108). LEGEND MAX™ Mouse IFN-y ELISA kit was ordered from Biolegend (430807).

[0221] Example 1 Synthesis and Characterization of Supraclusters

[0222] Synthesis of AuNC+. 100 pl. gold (III) chloride trihydrate aqueous solution (20 mM) is diluted with 750 pL deionized water followed by addition of pre-mixed solution containing 142.5 pL reduced L-GSH (20 mM) and 7.5 pL protamine sulfate (20 mM). The mixture solution is vortexed for 5 s and subject to gentle shaking at 70°C for 24 h. To remove impurities, the raw nanoclusters obtained after reaction are precipitated in cold ethanol and centrifuged at 17,000 rpm for 40 min at 0°C. The pellets are air-dried and reconstituted under sonication in DNase / RNase-free distilled water containing 1 mM HC1. The purified nanocluster solution is then stored at 4°C at dark.

[0223] Nanoparticle characterization. Zeta potential and DLS of nanoparticles were measured on Malvern Zetasizer Nano ZSP. Optical spectra were recorded on Bioteck Synergy Hl microplate reader (Agilent Technologies, Inc.). Fluorescence spectrum was recorded on Horiba Jobin-Yvon Spex Fluorolog-3 fluorimeter. TEM images were captured on El Tecnai Transmission Electron Microscope with 200 kV operating voltage.

[0224] Preparation of Supraclusters. To prepare BSCgai, 100 pL purified AuNC+nanocluster (0.5 pg / pL) were complexed with anti-Gal-1 siRNA (5 pg) under gentle shaking at room temperature for 1 hr (gold-to-siRNA weight ratio, 10: 1), followed by addition of 50 pL 8-Arm PEG-amine (10 mg / mL), 300 pL DNase / RNase-free 1 X PBS and 100 pL freshly made DTSSP STDU2-43706.601 solution (10 pg / pL). The mixture was gently shaken at 4°C overnight at dark. For the construction of BSCcy3, 5 pg Silencer™ Cy™34abeled Negative Control No. 1 siRNA (AM4621, Thermo Fisher Scientific) instead of Gal-1 siRNA was used. After reaction, the mixture was washed 3 times with DNase / RNase-free distilled water by ultracentrifugation at 4500 rpm for 30 min at 4°C using 10 kDa cut-off filter. The purified nanocomplexes were then diluted with DNase / RNase-free 1 x PBS buffer and stored at 4°C at dark shortly before use. GSC was prepared following the same steps without complexation with siRNA.

[0225] Agarose gel electrophoresis. Gold cluster-siRNA nanocomplexes with different gold- to-siRNA weight ratios (5: 1, 10: 1, 20:1, and 50: 1) were prepared and purified following the steps mentioned above. These nanocomplexes together with free anti-Gal-1 siRNA were then respectively mixed with RNA gel loading dye, and subject to gel electrophoresis with 5% agarose gel in DNase / RNase-free 1 x TAE (tris-acetate-EDTA) buffer at 150 V for 30 min. After gel running, the gel was stained with SYBR™ Gold Nucleic Acid Gel Stain and imaged with gel imager (Bio-Rad).

[0226] Supracluster dissociation assay. GSC and BSCgai(0.5 pig / ptl.) were respectively incubated with 30 pM. 100 pM, 1 mM, and 10 mM GSH at 37°C and the HDs were analyzed by DLS at different time points. The treated nanoparticles were washed three to five times to remove excess GSH and diluted before characterization with DLS or TEM.

[0227] Cell lines. M0C2 and M0C1 murine HNC cell lines were obtained from Dr. Uppaluri's lab. They were cultured in Iscove's modified Dulbecco's medium (IMDM) and Dulbecco's modified eagle medium / nutrient mixture F-12 (DMEM / F12) (2: 1) supplemented with 5% fetal bovine serum (FBS), 1% penicillin / streptomycin, 5 pg / mL insulin, 5 ng / mL epidermal growth factor (EGF) and 40 ng / mL hydrocortisone. mEERL murine oral cancer cells were obtained from Dr. William Spanos. The cells were cultured in IMDM and Ham’s F-12 nutrient mixture (F-12) (2: 1) supplemented with 10% fetal bovine serum (FBS), 1% penicillin / streptomycin, 25 pg / mL hydrocortisone, 5 pg / mL insulin, 5 ng / mL EGF, 5 pg / mL transferrin, and 1.36 ng / mL triiodothyronine. P029 cell line was obtained from Prof Xiao-Jing Wang (U.C. Davis) and cultured with DMEM / F12 supplemented with 10% FBS and 1% penicillin / streptomycin. Cells were grown at 37°C in a humidified atmosphere containing 95% air and 5% CO2.

[0228] Lysosomal escape of BSCcy.v MOC2 cells were seeded to confocal dishes (75856-742, VWR International LLC) and were grown to approximately 70% confluency. Cells were then respectively incubated with BSCcy3 (50 pg / mL). At different time points, cells were gently washed and co-stained with LysoTracker™ Green DND-26 (L7526, Invitrogen) as well as STDU2-43706.601

[0229] Hoechst 33342 (H3570, Invitrogen), followed by imaging with Zeiss LSM980 Plus confocal microscope.

[0230] In vitro Gal-1 knockdown by BSCgai. MOC2 cells (1 x 1025cells per well) were seeded to 12- well plate and cultured for 24 h. Cells were then respectively treated with PBS, free siRNA (4 pg), GSC (40 pg), BSCgai(0.5x) (20 pg supracluster containing 2 pg siRNA), BSCgai(40 pg supracluster containing 4 pg siRNA) and lipofect amine 3000 delivering Gal-1 siRNA (4 pg) according to manufacturer’s protocol. The gold content in supraclusters was quantified by ICP-MS and was used to indicate the mass amount of supraclusters. After 48 h, cells were trypsinized and prepared to cell lysates. Protein concentration was measured by Pierce™ BCA Protein Assay kit (23225, Thermo Fisher Scientific) and lysates were subject to WB analysis following standard procedure. Immunoblots were probed by rabbit monoclonal Gal-1 antibody (1 :1000, abl38513, Abeam) and horseradish peroxidase (HRP)-linked goat anti-rabbit secondary antibody ( 1:3000, G-21234, Invitrogen). 0-actin was used as internal loading control and was probed with HRP-actin antibody (1 : 10000, sc-47778 HRP, Santa Cruz Biotechnology). Immunoblots were developed with SuperSignal™ West Pico PLUS Chemiluminescent Substrate (34578, Thermo Fisher Scientific) and imaged with ChemiDoc imaging system (Bio-Rad). M0C1, mEERL and P019 cells were assayed with BSCgai(40 pg) or PBS following the same steps. Each treatment was assessed in triplicate.

[0231] Gal-1 mRNA levels in M0C2 cells were evaluated at 24 h after various treatments. Total RNA from cells were extracted by TRIzol™ reagent and converted to cDNA using High- Capacity cDNA Reverse Transcription Kit (4368813, Thermo Fisher Scientific). RT-PCR was performed on 7900HT Fast Real-Time PCR system (Thermo Fisher Scientific) using PowerUp™ SYBR™ Green Master Mix for qPCR (A25777, Thermo Fisher Scientific) and Igalsl mouse qPCR primers (MP207325, Origene).

[0232] Tumor cells and T cells co-culture assay. T cells were isolated from spleen of C57BL / 6 mouse using EasySep™ Mouse T Cell Isolation Kit (STEMCELL Technologies) according to manufacturer’s protocol. Isolated T cells were activated with mouse T-activator CD3 / CD28 Dynabeads (Gibco) followed by culture in RPMI 1640 medium supplemented with 2 mM L-glutamine, 10% FCS, 100 U / mL % penicillin / streptomycin, and 50 U / mL interleukin- 2. MOC2 cells were treated with PBS, GSC (50 pg / mL), or BSCgai (50 pg / mL) for 48 h. Thereafter, pretreated M0C2 cells were cultured together with activated T cells for 24 h. Cells were then harvested and blocked with blocked with TrueStain FcX™ (anti-mouse CD16 / 32) antibody (1 : 100, 101319, Biolegend) for 10 min on ice. Then cells were stained with anti- STDU2-43706.601 mouse BV421-CD3e, PE-CD4, APC-CD8, and FITC-Annexin V followed by flow cytometry analysis.

[0233] ROS production assay. MOC2 cells were seeded to confocal dishes (75856-742, VWR International LLC) and were grown to approximately 70% confluency. Cells were then respectively incubated with PBS, 50 pg / mL GSC or BSCgaiovernight. The treated cells were then cultured with H2DCFDA ( 10 pM, ab 113851 , Abeam) for 45 min at dark. After incubation, cells were treated with or without RT at 6 Gy, followed by washing, fixation with 4% paraformaldehyde (PF A) solution and staining with DAPI. Cells were then imaged by Leica DMi8 fluorescence microscope. Fluorescence signals were analyzed by ImageJ.

[0234] DNA damage assay. MOC2 cells were seeded to confocal dishes. Cells were then respectively cultured with PBS, 50 pg / mL GSC or BSCgaifor 24 h. After incubation, cells were then treated with or without RT at 6 Gy. Cells were incubated for another 24 h after RT treatment, and subject to fixation with 4% PFA and IF staining with y-H2AX antibody ( 1 : 1000, 9718S, Cell Signaling Technology) and Alexa Fluor™ 647-conjugated donkey anti-rabbit secondary antibody (1 : 400). Cells were stained with DAPI before imaging under Zeiss LSM980 Plus confocal microscope. Fluorescence signals were analyzed by ZEN microscopy software and ImageJ.

[0235] Clonogenic assay. MOC2 cells were seeded in 12-well plate and then treated with PBS, 50 pg / mL GSC or BSCgaifor 48 h. After incubation, cells were exposed to 0 Gy or 2 Gy RT. Irradiated cells were trypsinized, counted and seeded in 60 mm culture dish (300 cells per dish). Cells were grown in culture media respectively containing PBS, 50 pg / mL GSC or BSCgai for 10 days. Then the culture media was removed, and dishes were gently rinsed with 1 X PBS. Cells were then stained with 0.5% crystal violet (50 / 50 methanol / water) for 15 min at room temperature, followed by gentle washing with water three to five times and air-drying. The colonies were then counted manually.

[0236] In vivo clearance, biodistribution and biocompatibility study. All the animal experiments have been approved by the Institutional Animal Care and Use Committee of Stanford University and conducted under Stanford University's Administrative Panel for Laboratory Animal Care (APLAC) protocol 15106. Four to six-week-old female C57BL / 6 mice were ordered from The Jackson Laboratory. Healthy mice were subject to iv injection of PBS, GSC (150 pg), BSCgai (150 pg), orplasmonic gold nanoparticles (150 pg, 100 nm, CP11- 100-3KM, Nanopartz) respectively and accommodated in metabolic cages (n = 4). Urine and feces were collected at designated time points post injection. The collected excreta were STDU2-43706.601 weighed and then digested in aqua regia under heating at 100°C for 2 h. The gold amount in digested excreta was then analyzed by ICP-MS (Thermo XSeries II).

[0237] For biodistribution study, healthy mice were respectively iv injected with PBS or GSC (150 pg) (n = 4). At 24 h, 1 wk, 2 wk, 4 wk and 5-month post-injection, mice were sacrificed. Major organs including hearts, livers, spleens, lungs and kidneys were harvested, weighed and digested in aqua regia under heating at 100°C for 2 h. Gold content in organs were then analyzed by ICP-MS.

[0238] For biocompatibility study, healthy mice were respectively iv injected with PBS, GSC (150 pg) and BSCgai(150 pg) (n = 5). At 4 wk post-injection, blood samples were collected, and blood chemistry analysis was conducted by The Diagnostic Laboratory at Stanford University. At this timepoint, major organs including hearts, livers, spleens, lungs and kidneys were also collected and subject to fixation, paraffin embedding, sectioning and hematoxylin and eosin (II&E) staining according to standard procedure. After staining, organ slides were imaged with Leica microscope.

[0239] In vivo GSH depletion study. The hepatic GSH was depleted by i.p. injection of DEM (1 mL / kg) to six-week-old female C57BL / 6 mice. At 30 min post DEM treatment, GSC (150 pg) was iv injected to the mice followed by collection of excreta in metabolic cages for 24h (n = 4). Healthy C57BL / 6 mice injected with GSC (150 pg) were used as control group. At 24 h, blood from mice were collected retro-orbitally and digested in aqua regia for ICP-MS analysis. Urine and feces were processed as described above for ICP-MS analysis.

[0240] Preparation of GSC-F7 and GSC-F6. To prepare GSC-F7, 1 pL sulfo-Cy7 NHS ester (25 mM in DMSO, 15320, Lumiprobe) and 5 pL EDC solution (10 pg / pL) was added to 80 pL GSC (0.8 pg / pL) in PBS solution followed by addition of 5 pL EDC solution (10 pg / pL) and reaction under gentle shaking for at least 3 h to overnight at dark. After reaction, the mixture is washed 3-10 times with PBS solution by ultracentrifugation at 17000 rpm for 10 min using 10 kDa cut-off filter. Purified GSC-F7 was diluted in PBS and stored at 4°C at dark. GSC-F6 was prepared following the same steps except using AF647 NHS ester (16820, Lumiprobe) instead of sulfo-Cy7 NHS ester.

[0241] In vivo fluorescence imaging. Four to six-week-old female NU / J nude mice were ordered from The Jackson Laboratory. MOC2 cells were subcutaneously (sc) injected to the right flank (2.5 X 10’ cells in 10% Matrigel (354230, Coming)) or orthotopically injected to the buccal cavity (1 x 105cells in 10% Matrigel) of mice. At 10 days after tumor inoculation, mice were respectively iv injected with GSC-F7 ([Cy7] = 20 pM, 100 pL per mouse) or free STDU2-43706.601

[0242] Cy7 dye (([Cy7] = 20 pM, 100 pL per mouse). Fluorescence signals from mice were closely monitored by SII Lago X with excitation at 745 nm, emission at 790 nm and X-ray overlay. At day 7 post-injection, mice were euthanized, and major organs including hearts, livers, spleens, lungs, and tumors were harvested and imaged with SII Largo-X. Fluorescence signals were analyzed by Aura Imaging Software.

[0243] Biodistribution studies in MOC2 tumor models using ICP-MS. MOC2 tumor models were established as described above in in vivo fluorescence imaging. GSC (150 pg per mouse) was iv injected to mice (n = 4). Tumors and major organs were harvested at 24 h, 48 h, and 1-week post-injection and digested in aqua regia for quantitative analysis using ICP-MS. The tissue digestion process was described above in In vivo clearance, biodistribution and biocompatibility study.

[0244] Light-sheet imaging. Four to six- week-old female NU / J nude mice were ordered from The Jackson Laboratory. Subcutaneous and orthotopic M0C2 tumor models were established as mentioned above. At 10 days after tumor inoculation, mice bearing subcutaneous tumors were respectively iv injected with GSC-F6 ([AF647] = 20 pM, 100 pL per mouse). At 48 h post-injection, mice were deeply anesthetized and transcardially perfused with 15 ml PBS (with 20 pg / mL heparin), 5 mL WGA-AF488 solution (0.1 nig / mL), 10 mL PBS (with 20 pg / mL heparin) and 20 mL 4% PFA. Orthotopic tumor models were treated following the same steps at 14-21 days after tumor inoculation. All perfusion reagents were kept ice cold before use. After perfusion, tumors from subcutaneous tumor models and lungs from orthotopic tumor models were harvested, post-fixed in 4% PFA overnight and washed twice with PBS for 1 h. Tumors and lungs were then immersed in 50 mL 50% v / v THF / H2O at dark overnight, and subsequently incubated in 50 mL 80% v / v THF / H2O for 1 h and washed with 50 mL 100% THF for 1 h twice. The obtained tumors and lungs were subject to incubation with dichloromethane until the tissues sank to the bottom of the container. Thereafter, tumors and lungs were immersed in dibenzyl ether (DBE) until they were clear. The cleared tissues were preserved in DBE at room temperature at dark and imaged by LaVision Biotec Ultramicroscope II light sheet microscope. Light-sheet images were analyzed with Imaris 9.9 software (Oxford Instruments).

[0245] Tumors harvested were also subject to IF staining with rabbit monoclonal CD31 antibody (1 : 100, ab222783, Abeam) and FITC conjugated goat anti -rabbit secondary antibody (1 :1000, 65-61 11 , Thermo Fisher Scientific). Lung collected were subject to IF staining with rabbit monoclonal Ki67 antibody (1 :250, abl6667, Abeam) and FITC conjugated goat anti- STDU2-43706.601 rabbit secondary antibody (1:1000). Tissue slides were co-stained with DAPI before imaging with Zeiss Axiolmager fluorescence microscope.

[0246] CyTOF analysis. Subcutaneous M0C2 tumor models on female C57BL / 6 mice were established and as mentioned above. When tumor size reaches ~75 mm3, mice were randomized to different groups and iv injected with PBS or GSC (150 pg per dose) at day 0, 2, 4, 6, 8, 10 for a total of 6 doses. At day 1 or day 11 post-injection, injected mice were euthanized, and tumors were harvested for digested using mouse tumor dissociation kit (30- 096-730, Miltenyi) according to manufacturer’s protocol. Dissociated tissues were then smashed and passed through 70 pm cell strainer for single-cell suspension and subject to RBC lysis using ACK lysing buffer. Live dead stain was obtained by in-house preparation based on maleimido-mono-amide-DOTA (B-272, Macrocyclics) and indium chloride, and 1 :3000 dilution of 5 mg / mL stock was used for staining. Cells were then stained with 16-marker Maxpar® Mouse Sp / LN Phenotyping Panel Kit (201306, Fluidigm), and anti-mouse I Io 165- EpCAM (3165014B, Standard BioTools Inc.) according to manufacturer’s protocol. CyTOF of stained cells were recorded on Helios mass cytometer. CyTOF data and SPADE clustering analysis were performed on Cytobank (Beckman Coulter).

[0247] Combined BSCgai and SABR on subcutaneous MOC2 tumor models. Four to six- week-old female C57BL / 6 mice were ordered from The Jackson Laboratory. Subcutaneous M0C2 tumor models were established as mentioned above. When tumor size reaches ~ 75 mm3, mice were randomized to different groups for various treatments. PBS, GSC (150 pg per dose), and BSCgai(150 pg per dose) were respectively iv injected to mice at day 0, 2, 4, 6, 8, 10 for a total of 6 doses. CT imaging of mice was conducted on PXi X-Rad SmART cabinet irradiator (Precision X-ray Inc) with 40 kVp beam energy. Treatment planning was accomplished through RT image software package. SABR was performed on this PXi irradiator with 225 kVp X-ray energy. SABR was applied to injected mice on day 3, 5, 7 with 0, 6, or 8 Gy per fraction for a total dose of 0, 18, and 24 Gy, respectively. Tumor volumes were measured with a digital caliper every 3 days. Tumor volume was calculated as v = (a x b2) / 2, where a and b are respectively the length and width of tumor. Body weights of mice were monitored every week.

[0248] In vivo comparison of BSCgai with TDG and aGal-1. Subcutaneous M0C2 tumor models on female C57BL / 6 mice were established as mentioned above. When tumor volume reaches ~75 mm3, mice were randomized to different groups and subject to intratumoral injection of TDG (150 pg per dose), intraperitoneal injection of aGal-1 (150 pg per dose), or STDU2-43706.601 iv injection of PBS or BSCgai(150 pg per dose) at day 0, 2, 4, 6, 8, 10 for a total of 6 doses. The injected mice were treated with SABR on day 3, 5, 7 with 0 or 6 Gy per fraction for a total dose of 0 or 18 Gy, respectively. Tumor sizes were measured with a digital caliper every 3 days. Survival of mice were closely recorded.

[0249] To evaluate in vivo Gal-1 knockdown efficiency, at day 12 after the first injection, tumors from mice after various treatments were harvested and subject to IF and IHC staining of Gal-1 with rabbit monoclonal Gal-1 antibody (1:50 for IF and 1 :400 for IHC, abl38513, Abeam). IF tumor slides were further co-stained with DAPI before imaging under Zeiss LSM980 Plus confocal microscope. Fluorescence signals were analyzed with ImageJ.

[0250] In vivo anti-tumor therapies on P029 tumor models. P029 tumor model was established on four to six-week-old female C57BL / 6 mice by subcutaneous injection of P029 cells (6 X 105cells in 10% Matrigel per mouse) to the right flank of mice. When tumor size reaches ~75 mm3, mice were randomized to different groups and iv injected with PBS, GSC or BSCgai (150 pg per dose) at day 0, 2, 4, 6, 8, 10 for a total of 6 doses. The injected mice were treated with SABR on day 3, 5, 7 with 0 or 6 Gy per fraction for a total dose of 0 or 18 Gy. Tumor volumes were recorded every 3 days. For lung metastases study, at designated time points, mice with various treatments were euthanized and lungs were harvested. Lung metastases nodules were counted. Lungs harvested were subject to H&E staining and lung slides were imaged by a Leica microscope.

[0251] In vivo anti-tumor therapies on orthotopic MOC2 tumor models. Orthotopic M0C2 tumor models on female C57BL / 6 mice were established as mentioned above. When tumor sizes reach ~75 mm3, mice were randomized to different groups and iv injected with PBS, GSC (150 pg per dose) or BSCgai(150 pg per dose) at day 0, 2, 4, 6, 8, 10 for a total of 6 doses, followed by SABR treatment at day 3, 5, and 7 with 6 Gy per fraction. For chemoradiation group, mice were i.p. injected with cisplatin (2.5 mg / kg) 1 h prior to SABR at day 3 (6 Gy), followed by irradiation with another 2 fractions at day 5 and day 7. Tumor volumes were measured with caliper.

[0252] To evaluate the therapeutic efficacy of combined BSCgaiand chemoradiation, mice bearing orthotopic M0C2 tumors (-75 mm3) were randomized to different groups and iv injected with PBS or BSCgai(75 p per dose) at day 0, 2, 4, 6, 8, 10 for a total of 6 doses. PBS treated mice were subject to SABR at day 3, 5, 7 with 6 Gy per fraction. BSCgaitreated mice were i.p. injected with cisplatin (1.25 mg / kg) 1 h prior to SABR at day 3 (6 Gy), followed by another 2 fractions of SABR at day 5 and day 7. Tumor volumes were closely STDU2-43706.601 recorded. At day 15 after the first injection, mice with various treatments were euthanized and lungs were harvested. Lung metastases nodules were counted. Collected lungs were subject to H&E staining and lung sections were imaged with Leica microscope.

[0253] To assess the therapeutic outcome of combined BSCgai, aPDl and SABR therapy, mice bearing orthotopic M0C2 tumors (~75 mm3) were randomized to different groups and subject to iv injection of PBS, i.p. injection of aPDl (150 pg for each dose), iv injection of BSCgai(150 pg for each dose), or combined i.p. injection of aPDl and iv injection of BSCgaiat day 0, 2, 4, 6, 8, 10 for a total of 6 doses. The injected mice were treated with SABR at day 3, 5, and 7 with 0 or 6 Gy per fraction for a total dose of 0 or 18 Gy. Tumor growth of mice was recorded every 3 days.

[0254] Immunogenic cell death and cytokine analysis. Subcutaneous M0C2 tumor models on female C57BL / 6 mice were established as mentioned above. When tumor reaches ~75 mm3, mice were randomized to different groups and iv injected with PBS, GSC (150 pg per dose), or BSCgai(150 pg per dose) at day 0, 2, 4, 6, 8, 10 for a total of 6 doses. SABR was applied to mice at day 3, 5, and 7 with 0 or 6 Gy per fraction for a total of 0 or 18 Gy. To assess CRT exposure, mice were euthanized at day 8 after the first injection, and tumors were harvested for IF staining with CRT antibody (MAB38981, R&D Systems) and FITC conjugated goat antirabbit secondary antibody. Tumor slides were co-stained with DAPI before imaging with Leica DMi8 fluorescence microscope. Fluorescence intensity was analyzed with Image.!. Blood was sampled from living mice at day 8 and day 12 after the first injection in heparinized capillary tubes and plasma was separated by centrifugation at 3000 rpm for 10 min at 4°C. HMGB 1 level in plasma collected at day 8 and IFN-y level in plasma collected at day 12 were measured by ELISA kits according to manufacturers’ protocols.

[0255] Flow cytometry analysis. Subcutaneous M0C2 tumor models on female C57BL / 6 mice were established and treated as mentioned above. At day 12 after various treatments, tumors were harvested from mice and subject to dissociation using mouse tumor dissociation kit (30-096-730, Miltenyi) following manufacturer’s protocol. Processed tumor tissue was gently smashed and passed through 70 pm cell strainer for single-cell suspension. Red blood cell lysis was then performed using ACK lysis buffer, followed by gently washing of single cell suspension before staining. TDLNs were harvested, gently smashed, and passed through 70 pm cell strainer for single-cell suspension, followed by staining.

[0256] Cells were first stained with Live / Dead Zombie NIR™ Fixable Viability Kit (1:1000, 423105, Biolegend) for 10 min on ice at dark. Thereafter, cells were washed three times with STDU2-43706.601 eBioscience™ flow cytometry staining buffer (00-4222-26, Thermo Fisher Scientific) and blocked with TrueStain FcX™ (anti-mouse CD16 / 32) antibody ( 1 : 100, 101319, Biolegend) for 10 min on ice. After washing with staining buffer, cells were then stained with a panel of flow cytometry antibodies for 30 mins at dark on ice. Fluorochrome-labeled anti-mouse cell surface staining antibodies include BV421-CD45.1 (103133, Biolegend), BV711-CD45.1 (103147, Biolegend), BV421-TCRb (109229, Biolegend), BV510-CD4 (100489, Biolegend), PE / Cy7- CD8a (100489, Biolegend), APC-CDl lb (101211, Biolegend), PE-Ly6G (127607, Biolegend), BV785-Ly6C (128041 , Biolegend), FITC-NK1.1 (108706, Biolegend), PerCP / Cy5.5-F4 / 80 (123128, Biolegend), BV605-CDllc (117333, Biolegend), FITC-CDl lc (117305, Biolegend), AF700-MHC-II (107622, Biolegend), PE-CD25 (12-0251-81A, eBiosence). Cells were fixed and permeabilized with eBioscience™ Intracellular Fixation & Permeabilization Buffer Set (88-8824-00, Invitrogen) according to manufacturer’s protocol before intracellular staining. Fluorochrome-labeled anti-mouse intracellular staining antibodies include BV711-CD206 (141727, Biolcgcnd), PE / Cy7-CD206 (141720, Biolcgcnd), PE / Cy5- GzmB (372225, Biolegend), APC-FoxP3 (17-5773-82, Invitrogen), and AF700-Ki67 (652420, Biolegend). Compensations were performed using BD™ CompBeads Compensation Particles Anti-Rat / Hamster Ig, K Set (552845, BD Biosciences), BD™ CompBeads Anti-Mouse Ig, K / Negative Control Compensation Particles Set (552843, BD Biosciences), and ArC™ Amine Reactive Compensation Bead Kit (A10628, Thermo Fisher Scientific). Stained cells were recorded on LSR Fortessa flow cytometer (BD Biosciences). Flow cytometry data were analyzed by FlowJo software. tSNE-CUBA analysis was performed using Cytobank (Beckman Coulter).

[0257] Abscopal effect of combinational BSCgai and SABR. M0C2 tumors were inoculated onto both left and right flanks of female C57BL / 6 mice to establish bilateral tumor models. When tumor reaches -75 mm3, mice were randomized to different groups and iv injected with PBS or BSCgai (150 pg per dose) at day 0, 2, 4, 6, 8, 10 for a total of 6 doses. SABR was applied only to the right-side tumor of mice at day 3, 5, and 7 with 0 or 6 Gy per fraction for a total of 0 or 18 Gy. Growth of tumors on both flanks were closely recorded. At day 15, mice were euthanized, and the non-irradiated distant tumors were harvested and subject to single cell suspension and staining for flow cytometry analysis.

[0258] For depletion studies, tumor inoculation and BSCgai + SABR treatments were performed as described above. Anti-mouse CD8 (300 pg per mouse, BE0117, BioXCell), antimouse CD4 (300 pg per mouse, BE0003, BioXCell), and anti-mouse NK1.1 (BE0036, STDU2-43706.601

[0259] BioXCell) antibodies were respectively i.p. administered every 3 days, starting the day before the first BSCgaiinjection and continuing until the end of the experiment. Splenocytes from control and depleted mice were harvested on day 15, prepared as single cell suspensions, stained, and analyzed on flow cytometer to verify the depletion efficiency.

[0260] Statistical analysis. Quantitative experiments were conducted with a minimum of three replicates. Data were presented as mean ± SD unless otherwise stated. Statistical analysis was performed with GraphPad Prism 9.0 Software. P values were determined by two-tailed unpaired Student’s / -test or one-way analysis of variance (ANOVA). P value less than 0.05 was considered as statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

[0261] Results and Discussion

[0262] To synthesize cationic gold clusters, protamine (a clinically available arginine-rich protein) was doped with the reducing thiol ligand tripeptide glutathione (GSH) to template and stabilize the formation of nanoclusters during successive reduction of Au3+(FIG. la, FIG. 8a) (Loynachan 2019; Mo et al. Angew. Chem. Int. Ed. 53, 5815-5820 (2014).; Luo et al. J. Am. Chem. Soc. 134, 16662-16670 (2012).; Brewer et al. Science 286, 120-123 (1999)). The purified cluster (AuNC+) had a positive surface charge of 20.9 mV (FIG. lb) and red fluorescence centered at 620 nm upon UV excitation (FIG. 8b). Dynamic light scattering (DLS) indicated that AuNC+had a hydrodynamic diameter of ~2.7 nm (FIG. 1c), which was confirmed by transmission electron microscopy (TEM) (FIG. 8c). Gal-1 siRNA was then complexed with AuNC+via electrostatic interaction at different mass ratios (FIG. la). To protect siRNA cargo and improve bioavailability, AuNC+-siRNA complex with abundant amine groups on the surface was crosslinked with 8-Arm PEG-amine by the amine-reactive and biothiol-cleavable reversible crosslinker 3,3’-dithiobis(sulfosuccinimidyl propionate) (DTSSP) (FIG. la, FIG. 9a). This crosslinking process allowed AuNC+to be directly or indirectly linked to each other, forming supraclusters consisting of AuNC+-siRNA encapsulated within interconnected polymer matrix (FIG. la, FIG. 9a). Agarose gel electrophoresis indicated that most siRNA was adsorbed onto nanoclusters starting from a mass ratio of 10: 1 (gold / siRNA) (FIG. 9b), and the obtained supracluster was named BSCgai. A counterpart supracluster without siRNA was prepared (termed GSC). Zeta potential indicated that both supraclusters had negative surface charge (GSC: -6.7mV, BSCgai: -7.8 mV) (FIG. lb). DLS indicated that GSC and BSCgaihad hydrodynamic diameters (HDs) of 93 nm and 128 nm, STDU2-43706.601 respectively (FIG. 1c). Transmission electron microscopy (TEM) confirmed the spherical morphology and the microstructure of GSC and BSCgai, which consisted of AuNC+clusters (FIG. Id-e, FIG. 9c). Furthermore, no obvious changes in the diameters of GSC and BSCgaiwere observed during incubation in either PBS or physiologically mimicking conditions, indicating their excellent colloidal stability (FIG. 10). Next, GSH was used as a reducing agent mimicking reductive environment in vivo to treat GSC and BSCgai(FIG. 11). Both supraclusters remained stable at 30 pM GSH (plasma level). In tumor cell and hepatocyte-mimicking GSH conditions (1-10 mM), both supraclusters demonstrated sharp HD reduction to small segments (~8 nm) within 2-12 h (FIG. If, FIGS. 1 la-1 lb). This decomposition was further confirmed by TEM (FIG. 1g, FIGS, l lc-l ld).

[0263] To study the behaviors of supraclusters in HNC cells, fluorescent Cy3 dye labeled siRNA instead of Gal-1 siRNA was used to construct a supracluster named BSCcy3. Confocal fluorescence microscopy indicated that Cy3 signals largely co-localized with lysosomes when M0C2 HNC cells were treated with BSCcy3 for 6 h (FIG. 12). Subsequent observations at 12 h and 24 h revealed Cy3 signals in the cytoplasm outside the lysosomes, suggesting the successful lysosomal escape of siRNA delivered by the supraclusters (FIG. 12). Western blot (WB) indicated a significant reduction of Gal-1 expression in BSCgaitreated M0C2 cells compared to cells treated with PBS, free siRNA, or GSC (FIG. 2a). Consistently, quantitative reverse transcription polymerase chain reaction (qRT-PCR) showed a 53% downregulation of Gal-1 mRNA in BSCgaitreated cells, whereas GSC did not induce Gal-1 mRNA downregulation (FIG. 13). Downregulation of Gal-1 by BSCgaiwas validated in three other HNC cell lines including MOC1, mEERL, and P029 cells (FIG. 14). We further pretreated M0C2 cells with either GSC or BSCgaior neither, then co-culture these cells with isolated activated T cells (FIG. 15a). Cell death analysis by Annexin V staining indicated that BSCgaipretreatment significantly mitigated Gal-1 induced apoptosis in both CD8+and CD4+T cells (FIGS. 15b- 15d), whereas GSC failed to provide such protection for T cells.

[0264] We then assessed radiosensitization effects of supraclusters. M0C2 cells were pretreated with GSC or BSCgaiand subject to RT. Both supraclusters induced ~2-fold higher ROS production than the RT only group as shown by H2DCFDA assay (FIGs. 2b-2c; FIG. 16). Similarly, GSC and BSCgaiinduced ~1.5 times more yH2AX foci formation in cells after RT than the RT only group, reflecting more DNA double-strand break with either treatment (FIGS. 2d, 2e). Clonogenic survival assay suggested that BSCgaialone resulted in -75% fewer colonies than GSC or PBS (FIG. 17), indicating more cell death and / or decreased cell proliferation with STDU2-43706.601

[0265] Gal- 1 downregulation. When combined with RT, BSCgaipre- treated MOC2 cells hardly formed any colony and had the lowest survival fraction compared to PBS or GSC pre-treated cells (FIG. 2f).

[0266] In vivo clearance and traceable delivery of supraclusters

[0267] We evaluated in vivo clearance of superclusters. GSC and BSCgaiwere respectively intravenously (i.v.) injected to mice followed by excreta collection in metabolic cages (FIG. 2g). Inductive coupled plasma-mass spectrometry (ICP-MS) analysis of collected excreta indicated that at 24 h post-injection, 15.6% of injected dose (ID) of GSC was cleared through the kidney whereas 9.8% ID through the hepatobiliary pathway (FIG. 18a). At 4 weeks after injection, renal and hepatobiliary clearance of GSC increased to 37.1% and 48.4%, respectively, yielding a total clearance rate of 85.5% ± 2.9% (FIG. 2h). Similarly, BSCgaiachieved a total clearance rate of 90.3% ± 3.6% at 4 weeks (FIG. 2h). The clearance efficiencies of supraclusters were consistent with the reported clearance rates of gold clusters (Jiang 2019; Zhang et al. Biomaterials 33, 4628-4638 (2012); Yu et al. ACS Nano 17, 2554-2567 (2023)), and were ~4 times higher than that of commercialized PEG-coated plasmonic gold nanoparticles (22.4%) (FIG. 18b). Biodistribution studies confirmed that the injected gold was predominantly found in kidneys and liver at 24 h after injection (FIG. 2i), followed by fast clearance from the kidneys and relatively slow clearance from the liver (FIG. 19). This was evidenced by a sharp decrease of gold content in kidneys at 1 week post-injection (0.85% ID) compared to 24 h (3.4% ID), while liver retention slightly increased over this time period (FIG. 19). This observation might be explained by a temporarily high liver accumulation rate that initially exceeded clearance rate, which decreased over time and eventually fell below the clearance rate, resulting in reduction of gold retention. At 5 months post-injection, gold signal was almost undetectable in all major organs (FIG. 2i). This excellent clearance of supraclusters should be ascribed to their dissociation and biotransformation by the local high GSII content (~ 10 mM) in the liver sinusoids, where the supraclusters are transformed into smaller segments that are suitable for renal clearance or extravasation to the space of Disse for endocytosis by hepatocytes followed by excretion to bile duct (Zhang et al. J Control Release 240, 332-348 (2016)). To confirm the clearance mechanism, we depleted hepatic GSH with diethyl maleate (DEM) and compared the clearance rates of GSC in both depleted and non-depleted mice (FIG. 20a). Quantitative analysis indicated a notable increase of gold content in the blood and a significant drop of total clearance rates in GSH-depleted mice compared to the control group STDU2-43706.601 at 24 h post-injection (FIGS. 20b-20c). These results validated the pivotal role of hepatic GSH in the efficient clearance of supraclusters through both renal and hepatobiliary pathways.

[0268] Histological analyses indicated no pathological damage in major organs after injection of GSC or BSCgai (FIG. 21), suggesting their good biocompatibility. Blood chemistry assays revealed no differences in the liver and kidney function between the mice treated with GSC, BSCgai, or PBS control (FIG. 22). Moreover, enzyme linked immunosorbent assay (ELISA) indicated that GSC and BSCgaitreatments did not trigger an elevation of inflammatory markers such as c-reactive protein (CRP), procalcitonin (PCT) and interleukin 6 (IL-6) in mouse serum, further verifying the excellent biocompatibility of supraclusters (FIG. 23).

[0269] Gold nanoclusters by themselves require ultraviolet excitation for red fluorescence (FIG. 8b), which is not ideal for in vivo bioimaging due to the high tissue autofluorescence and shallow imaging depth. Since GSC and BSCgai shared similar chemical construction, clearance rates, and biosafety profiles, GSC was selected as the representative supracluster to investigate their in vivo trajectory. A near infrared (NIR) fluorophorc Cy7 was conjugated to GSC, yielding GSC-F7 that had NIR emission at 777 nm (FIG. 2j). Following i.v. injection of GSC-F7 to MOC2-tumor bearing mice, the NIR fluorescence (NIRF) signals in the tumor gradually increased and reached the maximum at 2 days post-injection (FIGS. 2k-21), possibly due to the enhanced permeability and retention (EPR) effect of nanosized particles. In contrast, mice injected with free Cy7 did not have any tumor NIRF signals owing to their fast clearance (FIGS. 2k- 21). At this time point, tumor signal from GSC-F7 injected mice was 60.4 times higher than mice injected with free Cy7. Biodistribution study at 7 days post-injection indicated that GSC- F7 predominantly accumulated in the tumor over major organs (FIG. 24a), showing a tumor- to-liver ratio (TLR) of 3.1.

[0270] This biodistribution was further confirmed by quantitative analysis of gold content in tumors and major organs using ICP-MS (FIG. 24b). At 48 h post-injection, gold content in tumor reached 9.7% ID with a TLR of 3.2, followed by a decline to 7.5% ID and a TLR of 2.6 at 1 week post-injection. The small difference in the TLRs and biodistribution measured by ICP-MS and NIRF imaging at 1 week post-injection should be ascribed to the different detection sensitivities of these two approaches. GSC-F7 also accumulated in cancer tissues of an orthotopic HNC mouse model with oral cavity tumor and spontaneous lung metastases (FIG. 25a). NIRF signals in the orthotopic tumor reached a plateau at 2 days post-injection of GSC- F7 (FIG. 25b) and was 44.1 times higher than tumor signals in Cy7-injected mice. Biodistribution study revealed that the orthotopic tumor had the highest GSC-F7 signals with an TLR of 2.0 (FIG. 25c). STDU2-43706.601

[0271] Characterization of supraclusters in the TME

[0272] Extravasation of nanoscale agents is a fundamental prerequisite for the proficient delivery of therapeutic cargo (e.g. siRNA) to target cells in TME. We delineated the three- dimensional (3D) distribution of supraclusters in primary tumor and lung metastases using light-sheet imaging. GSC was covalently labeled with a fluorophore AF647 to generate GSC- F6, showing peak emission at 675 nm (FIG. 26). At 2 days after i.v. injection of GSC-F6, M0C2 tumor-bearing mice were perfused with triticum vulgaris lectin-fluorescein (WGA- AF488) for vasculature staining followed by whole animal fixation (FIG. 27a). Fixed tumor were harvested for tissue clearing following a simplified iDISCO procedure and subjected to fast light-sheet scanning immersed in refractive index matching solvent (FIG. 27b) (Renier et al. Cell 159, 896-910 (2014).). The stacked dual-channel 3D image of cleared tumor tissue showed the distribution of GSC-F6 signals against tumor vessels (FIG. 3a, FIG. 28). 3D reconstruction was rendered based on the stacked image (FIG. 3b, FIG. 29). GSC-F6 channel was mapped with pseudo-color to visualize signal intensity distribution and blood vasculature was reconstructed in semitranslucent white (FIG. 3b). The reconstructed model indicated notably stronger GSC-F6 intensity in the extravascular regions compared to the intravascular counterpart, showing that the preponderance of supraclusters extravasated though vasculature to tumor interstitium. This observation was further validated by confocal fluorescence imaging of GSC-F6 in tumor sections against CD31 staining (FIG. 30).

[0273] We also investigated the distribution of supraclusters in spontaneous lung metastases derived from M0C2 orthotopic tumors. The cleared lungs with metastases were subject to light-sheet scanning, resulting in a stacked 3D image that revealed GSC-F6 signals distributed in the form of large clusters resembling metastatic nodules (FIGS. 3c-3d, FIG. 31). Confocal fluorescence imaging validated that GSC-F6 signals colocalized with metastatic biomarker Ki67 in the lung sections (FIG. 32a). 3D reconstruction of lung metastases was generated from the stacked image, with the GSC-F6 channel pseudocolored to illustrate signal intensity distribution, and the autofluorescence channel displayed in semitranslucent green to map the tissue (FIG. 3e). The brightest GSC-F6 signals were found concentrated in the center of the clusters, gradually diminishing in intensity towards the outer rim (FIG. 3e). Moreover, the number of GSC-F6 clusters detected in the 3D model was consistent with metastatic nodules counted by histopathological analysis (FIG. 32b).

[0274] We next sought to elucidate the intricate interactions between extravasated supraclusters and cells in TME through single-cell mass cytometry (CyTOF). GSC was i.v. injected to M0C2 tumor-bearing mice for 6 doses (FIG. 3f). At 24 h after the 1stand 6th STDU2-43706.601 injections, tumors (denoted as GSC-1 and GSC-6, respectively) were harvested, followed by CyTOF analysis together with a control cohort receiving PBS injection (GSC-0) (FIG. 33). FIG. 3g shows a gradual increase of gold signal intensity from GSC-0 to GSC-6 in both tumor cells and CD45+tumor-infiltrating leukocytes. Interestingly, the leukocyte uptake was ~3 folds more than that of cancer cells, indicating the therapeutic capacity of supraclusters targeting both tumor cells and immune leukocytes in TME.

[0275] SPADE (spanning-tree progression analysis for density-normalized events) clustering algorithm was applied to leukocyte populations. Principal immune cell populations were identified via conventional cell surface markers and were graphically designated as distinct nodes. SPADE tree plots indicated a discernible elevation of gold intensity across all immune cell subpopulations in GSC-1 and GSC-6 compared to PBS group (FIGS. 3h-3i). In GSC-1 samples, tumor associated macrophages (TAMs) had the highest median gold signals among all identified leukocyte subpopulations. In GSC-6 samples, most immune cells had higher gold signal intensity compared to GSC-1, with TAMs and dendritic cells (DCs) showing the highest median gold intensities, followed by natural killer (NK) cells and cytotoxic T lymphocytes (CTLs). In contrast, Tregs exhibited the lowest and the least change from GSC-1 to GSC-6. These findings suggest that TAMs were likely the first immune cells to engage with the supraclusters in the TME followed by DCs. Engagements of supraclusters with NK cells and CTLs might be secondary to these interactions, followed by other immune cells excluding Tregs.

[0276] BSCgai potentiates SABR in preclinical models

[0277] To investigate whether in vivo Gal-1 silencing with SABR radiosensitization could enhance antitumor response, we treated subcutaneous MOC2 tumors with BSCgai, GSC or PBS + / - SABR (FIG. 4a). Based on reported treatment regimens for siRNA therapeutics (Wang et al. Mol. Ther. 21, 1919-1929 (2013); Lei et al. Nat. Commun. 8, 15130 (2017)), supraclusters were administered every other day for a total of six doses. BSCgai injection alone modestly suppressed tumor growth compared to PBS control mice, while GSC induced negligible tumor inhibition (FIG. 4b). For SABR, we used the computed tomography (CT)-guided irradiator PXi X-Rad SmART to target only the primary tumor with 6 Gy X 3 fractions applied every two days (FIG. 34a). This regimen was adopted from a clinical study of SABR. For combination treatment, SABR fractions were delivered between supracluster injections to ensure better enrichment of supraclusters in the tumor during SABR than applying both treatments on the same day. GSC and SABR significantly induced tumor regression compared to SABR alone STDU2-43706.601

[0278] (FIG. 34b), ascribed to the radiosensitization and radiodynamic effect. Notably, BSCgai+ SABR exhibited superior tumor control compared to GSC + SABR (FIG. 34b). When the SABR dose was increased to 8 Gy X 3 fractions, BSCgai+ SABR almost completely inhibited tumor growth, whereas GSC + SABR failed to do so (FIG. 35).

[0279] We further assessed therapeutic efficacies of supraclusters + SABR treatments in orthotopic mouse models. X-ray beams were carefully guided by CT to focus on tumor region and avoid damage to the mouse brain (FIG. 4c). Both GSC + SABR and BSCgai + SABR significantly retarded the growth and metastases of oral cavity tumor compared to SABR alone (FIGS. 4d-4f). However, BSCgai + SABR induced significantly more inhibitory effects on both tumor growth and lung metastasis than GSC + SABR (FIGS. 4d-4f). There was no significant change in body weights of mice treated with either supraclusters + / - SABR (FIG. 36), suggesting the biosafety of these treatments. Confocal fluorescence microscopy showed that BSCgai alone significantly reduced Gal-1 expression in M0C2 tumors (FIG. 4g, FIG. 37, FIG. 38). Gal-1 downregulation was even stronger when BSCgaiwas combined with SABR (FIG. 4h, FIG. 37, FIG. 38), which was also validated by Gal-1 immunohistochemical (IHC) staining (FIG. 39). The enhanced Gal-1 reduction in BSCgai + SABR might be attributed to combinational Gal- 1 silencing and enhanced cell death.

[0280] We next compared the treatment efficacy of BSCgaito other Gal-1 antagonists, including small molecule inhibitor thiodigalactoside (TDG) and a blocking antibody (aGal-1), each administered at the same dosage and frequency as BSCgaibut through their own preferred injection routes. MOC2-tumor bearing mice received PBS, TDG, aGal-1 or BSCgai alone. The tumor control efficacy of BSCgaiwas similar to that of aGal-1, whereas it was significantly better than TDG (FIG. 40a). BSCgaitreatment yielded the longest survival (45d) compared to aGal-1 (33d) or TDG (30d) treatment (FIG. 40b). When combined with SABR, BSCgaiachieved considerably better tumor control compared to both aGal-1 and TDG (FIG. 4h), which translated to a significant extension in survival (FIGS. 4i-4j).

[0281] BSCgaiwas also assessed in highly metastatic P029 HNC tumor model (FIG. 41). BSCgaialone resulted in significantly slower tumor growth and fewer lung metastases compared to PBS or GSC alone (FIGS. 41a-41c). When combined with SABR, GSC exhibited tumor growth inhibition which surpassed SABR alone (FIG. 4k); and BSCgai+ SABR demonstrated the best tumor growth inhibition (FIG. 4k) with the fewest number of lung metastases compared to GSC + SABR or PBS + SABR (FIGS. 4i-4m). This combined STDU2-43706.601 treatment also led to a considerable extension in survival (57d), compared to BSCgal (33d) or SABR (39d) alone (FIG. 4j, FIG. 41d).

[0282] Since HNC patients are primarily treated with chemoradiation, we also compared the treatment efficacy of BSCgai+ SABR to chemoradiation in orthotopic M0C2 model. BSCgaiwas administered to mice for 6 doses and the oral cavity tumor received conformal SABR of 6 Gy x 3 fractions (FIG. 4c). For chemoradiation, cisplatin was not administered as frequently as BSCgaidue to its known dose-related systemic toxicity. Instead, a therapeutic dose of cisplatin comparable to that used in reported clinical studies was i.p. injected prior to SABR (Perse et al. Biomedicines 9, 1406 (2021).; McKibbin et al. Support. Care Cancer 24, 1789- 1793 (2016).; De Felice et al Crit. Rev. Oncol. / Hematol. 162, 103345 (2021)). BSCgai+ SABR treatment induced superior tumor inhibition efficiency and fewer lung metastases compared to cisplatin + SABR (FIG. 4n, FIG. 42). We also evaluated the treatment efficacy of combining BSCgai with chemoradiation, with BSCgaiand cisplatin each administered at half dose. BSCgai+ chemoradiation almost completely abrogated orthotopic tumor growth and prolonged survival of mice (FIG. 43), much superior to chemoradiation alone. These findings suggested that BSCgaican either replace or allow for dose reduction of cisplatin for toxicity mitigation.

[0283] Clinically available ICIs blocking PD-1 (e.g. pembrolizumab) have been extensively investigated to combine with RT in locally advanced clinical settings, whereas most trials yielded negative results to date. We thus investigate whether BSCgaiwould improve therapeutic effectiveness of combinational aPDl + SABR. aPDl was i.p. injected to an orthotopic HNC model, either with SABR alone or with BSCgai+ SABR. (FIG. 4o). aPDl + SABR resulted in improved tumor growth control compared to SABR alone but was inferior to BSCgai+ SABR (FIG. 4o). Notably, the combinational BSCgai+ aPDl + SABR regimen yielded the best tumor inhibition efficacy among all treatments.

[0284] BSCgai reinforces systemic antitumor immunity of SABR

[0285] We investigated the underlying immunomodulatory mechanism to elaborate the therapeutic enhancement of BSCgaito SABR. M0C2 tumor models were treated with supraclusters + / - SABR (FIG. 5a). At 2 (day 8) and 6 (day 12) days after the last SABR treatment, blood and tumor specimens were collected from mice for immune analysis. We first evaluated immunogenic cell death (ICD) by assessing danger-associated molecular patterns, which include surface expression of calreticulin (CRT) and extracellular translocation of high mobility group box-1 (HMGB1). IF staining of CRT on day 8 tumor sections showed that both GSC + SABR and BSCgai+ SABR significantly increased CRT expression compared to SABR STDU2-43706.601 alone (FIGS. 5b-5c). Enzyme-linked immunosorbent assay (ELISA) indicated that both supraclusters with SABR induced substantial elevation of plasma HMGB1 over SABR alone (FIG. 5d). Consistently, more mature antigen-presenting DCs were found in tumor draining lymph nodes (TDLNs) from mice after GSC + SABR and BSCgai+ SABR treatments than those treated with SABR alone (FIG. 44). Such enhanced ICD and DC activation triggered by combined treatment should be ascribed to the radiosensitizing and radiodynamic effects of supraclusters.

[0286] BSCgai alone induced a discernible elevation in proinflammatory cytokine interferon-y (IFN-y) production in plasma compared to GSC or PBS alone (FIG. 5e), which should be related to Gal-1 silencing. SABR combined with either supracluster elicited a marked increase in IFN-y production over SABR alone, with BSCgai+ SABR yielded the highest level. A similar trend was noted in the flow cytometry analyses of tumor-infiltrating CTLs (FIG. 5f, FIG. 45), CD4+T helper cells (FIG. 5i) and NK cells (FIG. 5j). Within CTLs, the same trend was noted for activated subtypes including GzmB (granzyme B)+CD8+and Ki67+CD8+T cells (FIGS. 5g-5h; FIG. 46). Similarly, BSCgai+ SABR evoked the most prominent macrophage polarization toward the proinflammatory Ml subtype in the TME (FIG. 47, FIG. 48a). The reverse was seen for immunosuppressive populations. BSCgaialone led to a decrease in the number of polymorphonuclear MDSCs (PMN-MDSCs), Tregs and M2 macrophages in the TME compared to GSC or PBS treatment (FIG. 48b, FIG. 49a). Combined with SABR, both supraclusters resulted in further reduction of PMN-MDSCs in the TME than SABR alone. Notably, BSCgai+ SABR induced the lowest percentage of tumor-infiltrating Tregs and the highest CTL to Treg ratio among various treatments (FIGS. 49b-49c). The tSNE-CUBA analysis shows the changes in CTLs and MDSCs across different treatments (FIG. 5k). The orchestrated bifunctional immunomodulatory mechanism of BSCgai+ SABR is summarized in FIG. 51.

[0287] Considerable clinical interest has been centered on ‘abscopal effect’ which harnesses single-site RT to inhibit non-irradiated malignancies via systemic antitumor immunity. However, clinical results of this strategy have been disappointing, albeit in combination with ICIs, which were mainly ascribed to the inadequate activation of anti-tumor immune cells (Brooks et al. Nat Rev Clin Oncol 16, 123-135 (2019)). We thus sought to investigate whether the bifunctional immunostimulatory BSCgaicould potentiate the tumor- suppressive abscopal responses of SABR. We established bilateral murine tumor models and mice were subject to SABR to the right tumor + / - concurrent BSCgaitreatment (FIGS. 6a, 6b). Besides inhibiting STDU2-43706.601 irradiated tumors (FIG. 50), BSCgai+ SABR induced notably superior growth regression of non-irradiated distant tumors compared to monotherapy (FIG. 6c). Flow cytometry analysis indicated significantly more infiltrated CD45+leukocytes in non-irradiated tumors from mice treated with SABR compared to those without (FIG. 6d). The addition of BSCgaito SABR resulted in the highest population of mature (CDl lc+MHC-II+) DCs and Ml macrophages polarization (FIG. 51), as well as the highest infiltration of tumor-killing lymphocytes (CTLs, CD4+T helper cells, NK cells) in non-irradiated tumors (FIGS. 6e-6h). A strikingly upregulated proportion of GzmB+CTLs was found in non-irradiated tumors from mice with combinational treatment than those receiving SABR alone (FIGS. 6i-6j). Furthermore, the abscopal effects by BSCgai+ SABR were notably diminished when CD8+T cells, CD4+T cells, or NK cells were respectively depleted using anti-CD8, anti-CD4, and anti-NKl.l antibodies (FIG. 52), confirming the mixed contribution of these immune cells in suppressing distant nonirradiated tumors. Collectively, these observations verified that BSCgaienhanced distant tumor regression by SABR through systemically raising tumor-inhibitory leucocytes and promoting tumor-killing capabilities of CTLs.

[0288] Example 2

[0289] Supraclusters delivering OVA mRNA as BMDC vaccine

[0290] Supraclusters were prepared as generally described above with OVA mRNA as the oligonucleotide. Sec Figure 7: (a) Scheme showing supraclusters loaded with OVA mRNA (termed BSCova) and its stimulation on BMDC; (b) Flow cytometry plots showing the surface expression of antigen peptide on BMDCs after co-incubation with BSCova; (c) Flow cytometry plots showing the increased antigen presentation of BMDCs after co-incubation with BSCova.

Claims

STDU2-43706.601CLAIMS1. A supracluster comprising a plurality of gold nanoclusters in a reversibly crosslinked polymer matrix.

2. The supracluster of claim 1, wherein the gold nanoclusters comprise surface amine groups.

3. The supracluster of claim 1 or claim 2, wherein the gold nanoclusters further comprise one or more capping moieties.

4. The supracluster of claim 3, wherein the capping moiety comprises an arginine-rich protein selected from protamine, polyarginine, penetratin, maurocalcine, Pep-1, Tat48-60, polyargininc, and polylysinc.

5. The supracluster of claim 4, wherein the capping moiety comprises protamine.

6. The supracluster of any one of claims 1-5, wherein the gold nanoclusters are complexed with one or more therapeutic nucleic acids.

7. The supracluster of claim 6, wherein the one or more therapeutic nucleic acids are RNA or DNA.

8. The supracluster of claim 6 or claim 7, wherein the one or more therapeutic nucleic acids are selected from small interfering RNAs, antisense oligonucleotides, microRNAs, and PIWLinteracting RNAs.

9. The supracluster of claim 8, wherein the one or more therapeutic nucleic acids is a small interfering RNA (siRNA).

10. The supracluster of claim 9, wherein the siRNA downregulates a cancer-promoting or immunosuppressive target.

11. The supracluster of claim 10, wherein the siRNA is an anti-Gal- 1 siRNA.STDU2-43706.60112. The supracluster of any one of claims 1-11, wherein the crosslinked hydrophilic polymer matrix comprises a hydrophilic polymer selected from polyethylene glycol, chitosan, hyaluronic acid, alginate, poly(acrylic acid), poly(vinyl alcohol), poly(acrylamide), poly(2-hydroxypropyl methacrylamide), poly(vinylpyrrolidone), poly(2V,A- dimethylacrylamide), poly(ethylene imine), and poly(2-methyl-2-oxazoline).

13. The supracluster of any one of claims 1 -12, wherein the crosslinked hydrophilic polymer matrix comprises a polyethylene glycol.

14. The supracluster of any one of claims 1-13, wherein the crosslinked hydrophilic polymer matrix comprises a branched polyethylene glycol.

15. The supraclustcr of claim 14, wherein the branched polyethylene glycol is an 8-armcd branched polyethylene glycol having terminal amine groups.

16. The supracluster of any one of claims 1-15, wherein the crosslinked hydrophilic polymer matrix comprises a crosslinker having a disulfide bond.

17. The supracluster of claim 16, wherein the crosslinker is 3,3’- dithiobis(sulfosuccinimidyl propionate).

18. A pharmaceutical composition comprising a plurality of supraclusters of any one of claims 1-17, and a pharmaceutically acceptable carrier.

19. A method of making a supracluster of claim 1, the method comprising: reacting a gold(III) salt with a reducing agent and a polycationic protein in aqueous solution to generate cationic gold nanoclusters; and mixing the cationic gold nanoclusters with a hydrophilic polymer and a crosslinker comprising a disulfide bond.

20. The method of claim 19, wherein the gold(III) salt is gold(III) chloride trihydrate.

21. The method of claim 19 or claim 20, wherein the reducing agent is glutathione.STDU2-43706.60122. The method of any one of claims 19-21, wherein the polycationic protein is protamine.

23. The method of any one of claims 19-22, wherein the hydrophilic polymer is a branched polyethylene glycol.

24. The method of claim 23, wherein the branched polyethylene glycol is an 8-armed branched polyethylene glycol having terminal amine groups.

25. The method of claim 24, wherein the crosslinker is 3,3’-dithiobis(sulfosuccinimidyl propionate).

26. A method of treating cancer in a subject in need thereof, comprising: administering to the subject an effective amount of a pharmaceutical composition of claim 18, and subjecting the subject to radiotherapy.

27. The method of claim 26, wherein the cancer comprises a solid tumor.

28. The method of claim 27, wherein the cancer is a head and neck cancer, lung cancer, pancreatic cancer, or a glioma.

29. The method of claim 28, wherein the cancer is a head and neck cancer.

30. The method of any one of claims 26-29, wherein the administering step comprises intravenous administration.

31. The method of any one of claims 26-30, wherein the radiotherapy comprises stereotactic ablative radiotherapy.

Images