Peptide-polyanion conjugate to activate sting innate immune signaling

Abstract

Compositions are provided comprising a multivalent peptide -polymer conjugate material that contained only 39 amino acid peptide from the STING C -terminal tail that includes interaction motifs for downstream kinase TBK1 and transcription factor IRF3, as well as optimized lipid nanoparticle formulations to deliver this conjugate material, and uses thereof.

Description

[0001]

[0002] MIT 26286 Peptide-Polyanion Conjugate to Activate STING Innate Immune Signaling

[0003] Sequence Listing Statement

[0004] The instant application contains an electronic Sequence Listing that has been submitted electronically and is hereby incorporated by reference in its entirety. Tire Sequence Listing was created on February 18, 2026, is named ‘"25-0059-WO ST26.xml” and is 16,648 bytes in size.

[0005] Background

[0006] Immunotherapy has been successful in the treatment of cancer, but only achieves long-term remission in a small fraction of patients, lire cytosolic DNA sensing stimulator of interferon genes (STING) pathway has garnered interest as a therapeutic target to overcome this by inflaming immunosuppressive tumors that would otherwise respond poorly to traditional immunotherapies. However, clinical trials of STING agonists have only shown modest efficacy

[0007] One challenge of drugging this pathway is that STING signaling is commonly inactivated in human cancer, often by epigenetic silencing of the STING gene. For example, STING is not expressed in the cancer cells of 50% of metastatic melanoma patients and 72% of stage III ovarian cancer patients. This renders STING agonists, the strategy typically used to drug this pathway, ineffective in these cells because they have no protein target to act upon. While downstream signaling through the cytosolic kinase TBK1 and transcription factor IRF3 is often intact, strategies to activate these molecules directly are not available.

[0008] Summary

[0009] In one aspect, the disclosure provides a composition, comprising a conjugate, wherein the conjugate comprises

[0010] (a) a plurality of peptides, wherein the plurality of peptides comprise peptides comprising (i) a binding site for TBK1, and (ii) a binding site for IRF3; and

[0011] (b) a surface to which the plurality of peptides are covalently conjugated.

[0012] In one embodiment, the binding site for TBK1 comprises or consists of the amino acid sequence PxPLR (SEQ ID NO: 3), wherein x is any amino acid, and the binding site forIRF3 comprises or consists of the amino acid sequence pLxlS, where p is a hydrophilic amino acid, and x is any amino acid. In another embodiment, the binding site for TBK1 comprises or consists of the amino acid sequence selected from the group consisting of (i) PLPLR (SEQ ID NO: 5); (ii) (D / E)XPXPLR(S / T)D (SEQ ID NO: 6), wherein X denotes any ammo acid; and (lii) PLPLR(T / S)D (SEQ ID NO: 7).

[0013] In another embodiment, the plurality of peptides comprises:

[0014] (i) a first plurality of peptides that comprise a binding site for TBK1; and (ii) a second plurality of peptides that comprise a binding site for IRF3. In a further embodiment, the first plurality of peptides and the second plurality of peptides are present in the composition in about an equal ratio. In another embodiment, the plurality of peptides comprise or consist of the amino acid sequence selected from (i) RLLISxxxxPLPLRTD (SEQ ID NO: 8), wherein x is any amino acid, and (ii) ELLISxxxxPLPERTD (SEQ ID NO: 9), wherein x is any ammo acid. In another embodiment, the plurality of peptides each comprise or consist of the amino acid sequence selected from (i) RLLISGMDQPLPLRTD (SEQ ID NO: 10), and (ii) ELLISGMEKPLPLRTD (SEQ ID NO: 11). In another embodiment, the plurality of peptides each comprise or consist of the amino acid sequence selected from SEQ ID NO: 1-2. In a further embodiment, each peptide in the plurality of peptides does not comprise or consists of a full length STING protein. In a further embodiment, each peptide in the plurality of peptides is between 5-100 amino acids in length, or between 5-75 amino acids in length, or between 5-50 amino acids in length, or between 5-40 amino acids in length, or between 5-30 amino acids in length, or between 5-20 amino acids in length, or between 10-100 amino acids in length, or between 10-75 amino acids in length, or between 10-50 amino acids in length, or between 10-40 ammo acids in length, or between 10-30 amino acids in length, or between 10-20 amino acids in length.

[0015] In another embodiment, each peptide in the plurality of peptides is covalently conjugated to the surface at its N-terminus. In another embodiment, each peptide in the plurality of peptides comprises an azido-lysine residue at its N-terminus.

[0016] In one embodiment, the surface is selected from the group consisting of a polymer, a nanoparticle, and a protein. In a further embodiment, the surface comprises a polymer. In a further embodiment, the polymer comprises an anionic polymer. In a further embodiment, the anionic polymer is selected from the group consisting of a negatively charged poly-amino acid, a negatively charged polysaccharide, poly (carboxylic acid), poly(sulfate), poly(phosphate), co-polymers thereof, an functionalized versions thereof. In a furtherembodiment, the anionic polymer is anionic polymer is selected from the group consisting of poly(L-glutamic acid), poly(D-glutamic acid), poly(L-aspartic acid), poly(D-aspartic acid), poly(acrylic acid) (PAA), poly(methacrylic acid) (PMAA), poly(ethylacrylic acid) (PEAA), poly(propylacrylic acid) (PPAA), poly(butylacrylic acid) (PBAA), poly(allylamine hydrochloride)-citraconic anhydride (PAH-Cit), hyaluronic acid, and dextran sulfate, co¬ polymers thereof, and functionalized versions thereof.

[0017] In a further embodiment, the anionic polymer comprises poly (L -glutamate). In one embodiment, the polymer comprises about 300 monomeric subunits, in a further embodiment, the polymer is about a 300-unit poly(L-glutamate) backbone.

[0018] In another embodiment, the anionic polymer comprises a plurality of alkyne residues conjugated to the anionic polymer. In a further embodiment, at least 10%, 25%, 50%, 75%, 90%, or ail of the anionic polymer units comprise alkyne residues conjugated to tire anionic polymer. In one embodiment, at least some of the alkyne residues are also conjugated to peptides in the plurality of peptides. In a further embodiment, at least 10%, 25%, 50%, 75%, 90%, or all of the alkyne residues are also conjugated to peptides in the plurality of peptides.

[0019] In one embodiment, the composition comprises bifunctional linkers between the surface and some or all of the plurality of peptides. In a further embodiment, the bifunctional linkers comprise polyethyelene glycol (PEG) linkers, including but not limited to azido- PEG4-NHS ester linkers. In another embodiment, an azido-lysine residue at the N-temiinus of each peptide in the plurality of peptides is conjugated to an alkyne residue on the anionic polymer. In another embodiment, the plurality of peptides comprises at least 3 peptides, or at least 4 peptides, or at least 8 peptides, or at least 25 peptides, or at least 100 peptides.

[0020] In one embodiment, the conjugate is contained within a delivery vehicle that enables delivery’ of the conjugate to the cytosol. In a further embodiment, the delivery’ vehicle comprises a delivery' vehicle for negatively charged cargo. In another embodiment, the delivery vehicle comprises a lipid nanoparticle or a cationic polymer nanoparticle. In a further embodiment, the delivery vehicle comprises a lipid nanoparticle. In a further embodiment, the cationic polymer nanoparticle comprises poly (beta-amino ester), polyethyleneimine (PEI), polyamidoamine (PAMAM), and / or chitosan. In another embodiment, the poly(beta-amino ester) comprises the structure selected from the following, wherein n is 3-100.

[0021]

[0022] In a further embodiment, the cationic polymer nanoparticle comprises polyethyleneimine (PEI).

[0023] In another embodiment, the delivery vehicle comprises a cationic polymer nanoparticle. In a further embodiment, the lipid nanoparticle is modified with targeting ligands to allow cell specific deliver}'-. In a further embodiment, the polymer is non-covalently adsorbed to the delivery surface vehicle.

[0024] In another embodiment, the lipid nanoparticle comprises an ALC-0315:cholesterol: 18:1 (A9-Cis) PE (DOPE): DMG PEG2000 core. In a further embodiment, the ALC-0315 cholesterol: 18: 1 (A9-Cis) PC (DOPC): DMG PEG2000 core is present at a molar ratio of about 43.8:23.5:31.3:1.5. In a further embodiment, the lipid nanoparticle comprises an ALC-0315 cholesterol: 18:1 (A9-Cis) PE (DOPE): DMG PEG2000 core. In a further embodiment, the ALC-0315 cholesterol: 18:1 (A9-Cis) PE (DOPE): DMG PEG2000 core is present at a molar ratio of about 43.8:23.5:31.3:1.5. In a further embodiment, the ALC- 0315 cholesterol: 18:1 (A9-Cis) PE (DOPE) DMG PEG2000 core is present at a molar ratio of about 50:38.5:10: 1.5,

[0025] In another embodiment the composition further comprises a PD-1 inhibitor. In a further embodiment, the PD-1 inhibitor comprises an anti -PD-1 antibody.

[0026] The disclosure also provides a pharmaceutical composition comprising:

[0027] (a) any of the compositions described herein; and

[0028] (b) a pharmaceutically acceptable carrier.

[0029] The disclosure also provides methods for treating cancer by administering a composition or pharmaceutical composition of the disclosure to a subject in need thereof. In another embodiment, the subject has metastatic melanoma or ovarian cancer. In another embodiment, administering is carried out via intraperitoneal (IP) administration. In a further embodiment, the administering comprises administering in at least five doses. In a further embodiment, the administering is performed weekly for about one year. In a further embodiment, the administering is performed monthly for about one year.Brief Description of the Drawings

[0030] Figure 1: Activating STING signaling cascade downstream using STINGATM allows for efficacy in a wider fraction of human cancer cell lines. (A) Schematic of the STING signaling cascade, noting that upstream molecules cGAS and STING are frequently epigenetically silenced in cancer, leaving cytosolic DNA or STING agonists incapable of activating downstream signaling. A STING mimicking therapeutic instead acts directly on downstream TBK1 and IRF3 to enable activity in STING silenced cells. (B) ELISA quantified CXCI / 10 in supernatant of specified cancer cell line treated with 0.083 pM of STINGATM or inactive control STINGATM S365A dimer or 50 pM cGAMP, sampled at 24 h (N = 3). All treatments used TransIT-X2 as vehicle. Replicates with undetectable levels of CXCI / 10 are plotted at the minimum detectable dose (MDD). Two-way ANOVA on log-transformed concentrations with Tukey correction for multiple comparisons, comparing only- treatments within each cell line. Data represented as geometric mean ± standard deviation. **** = P < 0.001, *** = P < 0.001, ** = P < 0.01, * = P < 0.05. (C) Schematic of mouse STING protein, comparing to full STING to cytosolic domain that is used in existing STING mimicking therapeutics and the STING C -terminal tail that was used in this work. Tire TBK1 and IRF3 interaction motifs in the C -terminal tail are highlighted, noting that phosphory lation of S365 is required for IRF3 interaction.

[0031] Figure 2: The STING peptide-polyanion conjugate was designed with a polyanion backbone to promote electrostatic interactions with positively charged delivery vehicles and allow easy delivery- to the cytosol. Multiple copies of the STING C -terminal tail peptide are conjugated to this backbone to mimic high-valency display of the STING C -terminal tail found in the multimerized active state of STING. In the cytosol, the STING peptide-polyanion conjugates mimics the protein-protein interaction motifs of multimerized STING to enable TBK 1 and IRF3 activation, and the resulting transcription of IRF3 controlled genes.

[0032] Figure 2: Synthesis of peptide-polymer conjugate. (A) Tire sequence of the STING C-terminal tail peptide from mSTING(340-378), highlighting motifs known to interact with TBK1 and IRF3. (B) The peptide-polyanion conjugate was synthesized through copper-catalyzed azide-alkyne cycloaddition of an azidolysine-m STING(340-378) peptide and thepolyanion poly(L-glutamate)-graft-alkyne. (C) The sodium dodecyl sulfate - polyacrylamide gel electrophoresis (SDS-PAGE) of a successful reaction showing a presence on SDS-PAGE get when an identical mass of peptide (4.3 kDa) runs off gel and polymer alone did not interact with Coomassie blue stain. (D) An 3-azido-7-hydroxycoumarin assay detected increasing fluorescence as mass is increased for unreacted poly(L-glutamate) but not conjugate, indicating no detectable alkynes are present on conjugate. (E) HPLC chromatogram using size -exclusion column of azidolysine-mSTING(340-378), poly(L-glutamate)-graft-alkyne, and poly(L-glutamate)-graft-mSTlNG(340-378). (F) Circular dichroism spectrum of azidolysine-mSTING(340-378), poly(L-ghitamate)-graft-alkyne, and poly(L-glutaniate)-graft-mSTING(340-378) in lx PBS, (G) shows that the activity of the conjugate material was examined using a HEK293T-derived reporter with luciferase expression under the control of target transcription factor IRF3. IRF3 reporter signal relative to buffer treatment for HEK293T reporter cells treated with 2 uM STING or Scr peptide alone, conjugated to only C -terminus of size-matched poly(L-glutamate) to generate a monovalent peptide-polyanion conjugate control, or conjugated to side chains as specified in (B) to generate a multivalent peptide-polyanion conjugate. All treatments were transfected using TransIT-X2, activity was measured 24 h post treatment (N = 3 biological replicates). Data represented as geometric mean ± SD.

[0033] Figure 3: Peptide-polymer conjugate effectively activates STING signaling and can be delivered intracellularly. (A) The luciferase reporter signal 24 h after treatment of HEK293T-derived IRF reporter cell line with either ST1NGATM protein or STING(340-378)-poly(L-glutamate) conjugate using TransIT-X2 for cytosolic delivery’. The conjugate was both more effective at an equivalent dose of peptide and had reduced toxicity, allowing for a wider therapeutic window.

[0034] Figure 4: Multivalent peptide-polymer conjugate can be delivered intracellularly by translatable nucleic acid delivery methods. FIGS. 5A-5C show that a LNP formulation was generated to deliver STING(340-378)-poly(L-glutamate) conjugate using a 50:38.5:10:1.5 ALC-0315:cholesterol: DOPE: DMG PEG2000 (mol ratio) core, as is or layered with PLD. (A) shows the size by dynamic light scattering (DES); (B) shows the zeta potential; and (C) shows the luciferase reporter signal 24 h after treatment of HEK293T-derived IRF3 reporter cell line with 4 pg / mE conjugate. FIGS. 5D-5F show a polyplexformulation was produced using the PBAE Poly3, with and without poly(L-aspartate) (PLD).

[0035] (D) shows the size by DLS; (E) shows the zeta potential; and (F) shows the luciferase reporter signal 24 h after treatment of HEK293T-derived IRF3 reporter cell line with 8 pg / mL conjugate were measured.

[0036] Figure 6: STING peptide-polyanion conjugate activates innate immune signaling even in STING-defi cient cancer cells. (A) HEK293T reporter cells were treated with 6.0 pg / mL STING or Scr conjugate using TransIT-X2 as a vehicle, cells were imaged by confocal microscopy at 6 h post treatment to examine conjugate colocalization with target TBK1 (representative of N:::3 biological replicates). Scale bar is 10 pm. (B) HEK293T reporter cells were treated with 8.3 pg / mL SUNG or Scr conjugate using TransIT-X2 as a vehicle, Western blot was perfor ed at 6 h post treatment to examine TBK 1 and IRF3 phosphorylation (representative of N = 3 biological replicates). (C) IRF3 reporter signal relative to buffer treatment for HEK293T reporter cells pretreated for 6 h with TBK1 inhibitor MRT67307 (TBKli) and then treated with 8.3 pg / mL STING or Scr conjugate delivered using TransIT-X2, measured 24 h post treatment (N = 3 biological replicates). (D) Western blot of STING and P-actin expression in ovarian cancer cell lines KURAMOCHI and A2780. (E-F) KURAMOCHI and A2780 ovarian cancer cell lines were treated with 5.0 pg / mL STING or Scr conjugate using TransIT-X2 as a vehicle or 50 pM STING agonist ADU-S100. (E) CXCL10 and (F) IFN-P in supernatant was measured by ELISA 24 h post treatment (N:::3 biological replicates). Replicates where analyte was below the limit of detection (LOD) are labeled as not detected (ND), no summary' statistics were computed if any replicates was ND. (G-I) KURAMOCHI cells were treated with 5.0 pg / mL STING or Scr conjugate using TransIT-X2 as a vehicle or 50 pM ADU-S100, mRNA sequencing was performed at 6 h post treatment (N = 4 biological replicates). (G) Plot of Log2 fold change of Buffer vs, STING conjugate or ADU-S100, showing high correlation between treatments. Plot of Log2 fold change of Buffer vs. Scr conjugate or ADU-S100 is displayed below as a control, showing greatly reduced correlation. The coefficient of determination R2for line of best fit is displayed. (H) Gene set enrichment analysis was performed on MSigDB Hallmark gene set, normalized enrichment and adjusted P value are displayed for the 10 gene sets significantly enriched (P <.05) in Scr vs. STING conjugate (I) Heatmap of gene expression for selected genes. Replicates where a given gene was not detected are labeled ND. Data represented as geometric mean ± SD.Figure 7: Designing a LNP carrier for peptide-polyanion conjugate delivery. (A) Ease of peptide -polyanion conjugate delivery was tested using a variety of “off-the-shelf ’ nucleic acid delivery methods: LNPs prepared with the ionizable lipids ALC-0315, MC3, and SM-102 or polymer complexes prepared using PEI, the PBAE “Poly2”, or the transfection reagent TransIT-X2. IRF3 reporter signal relative to buffer treatment for HEK293T reporter cells treated with 4.2 pg / mL STING or Scr conjugate delivered using specified delivery' method, 24 h post treatment (N = 3 biological replicates), P values computed with a two-way ANOVA on log-transformed IRF3 reporter signal followed by Sidak's post-hoc test are displayed above each figure, comparing STING to Scr conjugate for each delivery method. (B) Schematic of LNP formation method, where conjugate in aqueous phase is mixed with lipids in ethanol phase to generate nanoparticles. (C) An ALC-0315-based LNP formulation was optimized by screening a library of LNPs with varied phospholipid components (DOPE, DSPC, DOPC) and molar ratios of ALC-0315, phospholipid, and cholesterol. (D) LNP Z- A verage diameter and polydispersity index measured by DLS (N = 3 technical replicates), (E) Encapsulation efficiency (EE%) measured by native PAGE assay. (F) IRF3 reporter signal relative to buffer treatment for HEK293T reporter cells treated with 3.3 ug / mL conjugate delivery by each LNP formulation or TransIT -X2 control, 24 h post treatment (N = 3 technical replicates), (G) C3 LNP formulation selected for further examination is composed of 43.8: 31.3: 23.5: 1.5 ALC-0315: DOPC: Cholesterol: DMG-PEG2000 (mol ratio). The C3 formulation was loaded with STING or Scr conjugate and tested for (H) LNP Z -Average diameter and polydispersity index measured by' DLS (N = 7 (S TING) or N = 5 (Scr) LNP batches), (I) zeta potential (N = 6 (STING) or N = 4 (Scr) LNP batches), and (J) Encapsulation efficiency by native page assay (N = 6 (STING) or N = 4 (Scr) LNP batches).

[0037] (K) Cry o-TEM image of C3 LNP formulation containing STING conjugate (representative of N = 2 LNP batches). Data represented as mean ± SD for data on linear scale and geometric mean ± SD for data on log-scale.

[0038] Figure 8: STING peptide-polyanion conjugate treatment induces innate immune cytokines production mice. (A-B) Mice were dosed with 20 pg of STING peptide conjugate delivered by LNP IP and serum 'as collected at 0, I, 3, 6, 10, 24, and 50 h. Serum was analyzed to measure (A) STING peptide conjugate concentration by Cy5 fluorescence(showing one phase exponential decay fit to data) and (B) CXCL10 concentration by ELISA (N = 3 mice). Conditions where analyte was below the limit of detection (LOD) are labeled as not detected (ND). (C) Mice were inoculated -with 3x106 BPPNM cells IP and dosed with 20 pg of STING or Scr peptide conjugate (N ==:3 mice) delivered by LNP IP at 14 days after inoculation. (D) Omental tumor, (E) ascites, and (F) serum were collected 6 h after dosing. Concentrations of CXCL10, IFN-P, IL-6, TNF-a, IFN-y were measured by ELISA and are reported relative to total protein concentration in tumor and ascites (D-E) or relative to volume in serum (F). Conditions where analyte was below the LOD are labeled as ND. Data represented as mean ± SD.

[0039] Figure 9: STING peptide-polyanion conjugate treatment shrinks tumors and prolongs survival in metastatic ovarian cancer models. (A) Mice were inoculated with 3 - 106BPPNM cells IP and dosed with 20 pg of S UNG or Scr peptide conjugate delivered by LNP IP at 10, 13, and 16 days after inoculation. Groups included N = 6 (PBS, Scr LNP) or N = 5 (STING LNP) mice. (B) Survival plot w'ith P values were determined by log(rank) (Mantel- Cox) test. (C-D) Tumor burden measured by IVIS, displaying (C) geometric mean ± SD bioluminescent intensity over the treatment period and (D) Individual mouse bioluminescent intensity values. (E) Mice w7ere inoculated with 106KPCA. C cells IP and dosed with 20 pg of STING or Scr peptide conjugate delivered by LNP IP at 7, 10, 13, and 16 days after inoculation. Groups included N = 5 (PBS), or N = 4 (STING LNP, Scr LNP) mice. (F) Survival plot with P values were determined by log(rank) (Mantel-Cox) test, (G-H) Tumor burden measured by IVIS, displaying (G) geometric mean ± SD bioluminescent intensity- over the treatment period and (H) Individual mouse bioluminescent intensity values.

[0040] Figure 10: STING peptide-polyanion conjugate treatment repolarizes ovarian tumor microenvironment to improve response to PD-1 checkpoint blockade. (A) Mice were inoculated w ith 3 * 106BPPNM cells IP and dosed with 20 pg of STING or Scr peptide conjugate delivered by LNP IP at 10, 13, 16, and 19 days after inoculation. Omental tumor was collected on day 20 for analysis by flow cytometry. Groups included N = 5 mice. (B) Cell populations in tumor after treatment, displaying percentage of CD45+ cells made up by each T cells (CD3+), CD8+ T cells, CD4-J- T cells, NK cells (CD3- NK1. I-J-), B cells (CD19+), Macrophages (F4 / 80+), DCs (CDllc+ MHCII+), MDSCs (CDllb+ Gr-1+).Percentage of CD45- cells made up by BPPNM cancer cells (GFP+) are also displayed. (C-D) Polarization of Macrophages (F4 / 80+), showing MFI and representative distributions of (C) CD86 and (D) CD206. (E) Activation of DCs (GDI lc+ MHCII+), showing MFI and representative distributions of CD86. (F) Expression of PD-1 on CD8+ T cells, showing percentage of cells that are PD-1+ and representative distributions of PD-1. P values computed with a one-way ANOVA followed by Tukey’s post-hoc test are displayed above each figure. All flow data is represented as mean ± SD. (G) Mice were inoculated with 3x]06BPPNM cells IP and dosed with 20 ug of STING or Scr peptide conjugate delivered by LNP IP at 10, 13, 16, 19, and 22 days after inoculation. A subset of groups were additionally treated with 100 pg aPD-1 antibody at 11 and 17 days after inoculation. Groups included N = 6 (PBS, STING LNP, Scr LNP), N = 5 (STING LNP + aPD-1 ), or N = 4 (PBS + aPD-1, Scr LNP + aPD-1) mice. (H) Survival plot with P values were determined by log(rank) (Mantel- Cox) test. (I- J) Tumor burden measured by IVIS, displaying (I) geometric mean ± SD bioluminescent intensity over the treatment period and (J) Individual mouse bioluminescent intensity values.

[0041] Figure 11: Peptide-polyanion conjugate has consistent bioactivity across batches. IRF3 reporter signal relative to buffer treatment for HEK293T reporter cells treated with 8.3 pg / mL STING or Scr conjugate delivered using TransIT-X2, measured 24 h post treatment. N = 4 independently synthesized batches of conjugate, displaying geometric mean ± SD of N = 3 technical replicate activity measurements for each batch.

[0042] Figure 12: STING peptide-polyanion conjugate activates IRF3 and NF-KB transcription factors in THP1 monocytes. (A) 1RF3 reporter signal relative to buffer treatment and (B) relative NF-KB reporter signal for THP 1 dual reporter cells treated with S UNG or Scr conjugate delivered at specified dose of peptide using TransIT-X2, or STING agonist ADU-S100, measured 24 h post treatment (N = 3 biological replicates). A variable slope Hill equation was fit to data, with the EC50 and Top reported underneath each plot for the active treatments STING conjugate and ADU-SI00.Figure 13: Optimization of LNP formulation for multivalent peptide-polyanion conjugate delivery. (A) The identity of the phospholipid component and ratios of ionizable lipid, phospholipid, and cholesterol were varied to optimize LNP stability, encapsulation, and size. The ionizable lipid ALC-0315 was used for all formulations, the mole percentage of PEG lipid was held at 1.5%, and the mass ratio of total lipid to peptide-polyanion conjugate was held at 20: 1. (B) Mole percentage of each lipid in five compositions examined. (C) IRF3 reporter signal and (D) viability (resazurin assay) relative to buffer treatment for HEK293T reporter cells treated with 3.3 pg / mL conjugate delivery by each LNP formulation or TransIT-X2 control, 24 h post treatment (N = 3 technical replicates). LNP Z -A verage diameter and polydispersity index measured by DLS (E) immediately after particle formation (F) after purification by centrifugal filtration (N = 3 technical replicates). (G) LNP zeta potential measured in water (N = 3 technical replicates). (H) Encapsulation efficiency of peptide-polyanion conjugate in LNP measured by native PAGE assay. Data represented as mean ± SD for data on linear scale and geometric mean ± SD for data on log-scale.

[0043] Figure 14: Characterization of optimized C3 LNP formation process. (A) Z-average diameter, polydispersity index, and (B) zeta potential measurements (N = 4-8 independent LNP batches) were taken during the process of generating LNPs loaded with STING or Ser peptide-conjugate. The “Initial” measurement was taken immediately upon LNP formation, the “Purified” measurement tvas taken after purification and concentration of LNPs by centrifugal filtration into water, and the “Final” measurement was taken after dilution to a final conjugate concentration of 100 pg / mL in the lx PBS buffer used for dosing. (C) Conjugate yield after purification and concentration of LNPs by centrifugal filtration (N = 6-8 independent LNP batches). Data represented as mean ± SD.

[0044] Figure 15: TNS titration curve of C3 LNPs loaded with STING peptide-polyanion conjugate. Apparent pKa was determined by conducting a four-parameter logistic regression in GraphPad Prism and determining pH at 50% normalized TNS fluorescence. (N = 3 independent LNP batches).Figure 16: Biodistribution STING peptide-polyanion conjugate in ovarian cancer model. (A) Mice were inoculated with 3×106BPPNM cells IP and dosed with 20 pg of STING peptide conjugate delivered either by LNP or unencapsulated (“free”) (N = 4 mice) or PBS vehicle (N:::3 mice) delivered by LNP IP at 15 days after inoculation. Organs were collected 4 h after dosing for ex vivo IVIS. (B) Tumor bioluminescence measured as average radiance in each organ, pooled for N = 11 mice in all groups. (C-D) Conjugate fluorescence measured as (C) average radiant efficiency and (D) as a fraction of total fluorescence signal recovered from all organs. Data is represented as mean ± SD.

[0045] Figure 17: Cellular biodistribution of peptide-polyanion conjugate LNP in tumor and ascites. (A) Mice were inoculated with 3×106BPPNM cells IP and dosed with 20 pg of STING or Scr peptide conjugate delivered by LNP IP at 10, 13, 16, and 19 days after inoculation. Omental tumor and ascites (peritoneal fluid) was collected on day 20, 12 h after the final dose, for analysis by flow cytometry’. Groups included N = 5 (All tumors, PBS ascites) or N = 4 (STING LNP ascites, Scr LNP ascites) mice. (B-C) The percentage of each listed cell type that is positive for Cy5-labeled peptide-polyanion conjugate was measured in the (B) ascites and (C) tumor. Cell populations examined include BPPNM cancer cells (CD45- GFP+), stromal cells (CD45- GFP-), myeloid cells (CDl lb+), CD8+ and CD4+ T cells (CD3+), NK cells (CD3- NK1.1+), and B cells (CD19+). Specific myeloid cell populations were examined in the tumor including M1 (CD86hiCD206low) and M2 (CD86lowCD206hi) polarized macrophages (F4 / 80+) and DCs (CDl lc+ MHCII+), Data is represented as mean ± SD.

[0046] Figure 18: Change in mouse mass during monotherapy tumor efficacy studies. (A) Mice were inoculated with 3×106BPPNM cells IP and dosed with 20 pg of STING or Scr peptide conjugate delivered by LNP IP at 10, 13, and 16 days after inoculation. Groups included N = 6 (PBS, Scr LNP) or N = 5 (STING LNP) mice. (B-C) Change in mouse mass was measured, displaying (B) mean ± SD over the treatment period and (C) individual mouse changes in mass, (D) Mice were inoculated with 106KPCA. C cells IP and dosed with 20 pg of STING or Scr peptide conjugate delivered by LNP IP at 7, 10, 13, and 16 days after inoculation. Groups included N = 5 (PBS), or N = 4 (STING LNP, Scr LNP) mice. (E-F) Change inmouse mass was measured, displaying (E) mean ± SD over the treatment period and (F) individual mouse change in mass.

[0047] Figure 19: Acute toxicity study hematology measurements. Mice were inoculated with 3 x 106BPPNM cells IP and dosed with 20 pg of STING or Scr peptide conjugate delivered by LNP IP at 14 days after inoculation. Groups included N = 4 (STING LNP, Scr LNP) or N = 3 (PBS) mice. 24 h after treatment, blood was collected for complete blood counts, P values computed with a one-way ANOVA followed by Tukey’s post-hoc test are displayed above each figure. Data represented as mean ± SD.

[0048] Figure 20: Acute toxicity study serum chemistry measurements. Mice were inoculated with 3 x IO6BPPNM cells IP and dosed with 20 pg of STING or Scr peptide conjugate delivered by LNP IP at 14 days after inoculation. Groups included N = 4 (STING LNP, Scr LNP) or N = 3 (PBS) mice. 24 h after treatment, blood was collected for analysis of serum chemistry. P values computed with a one-way ANOVA followed by Tukey’s post-hoc test are displayed above each figure. Data represented as mean ± SD.

[0049] Figure 21: STING peptide-pol anion conjugate treatment impact on ascites cell populations. Mice were inoculated with 3×106BPPNM cells IP and dosed with 20 pg of STING or Scr peptide conjugate delivered by LNP IP at 10, 13, 16, and 19 days after inoculation. Ascites (peritoneal fluid) was collected on day 20 for analysis by flow cytometry. Groups included N = 5 (PBS) or N = 4 (STING LNP, Scr LNP) mice. Cell populations in tumor after treatment, displaying percentage of CD45+ cells made up by each T cells (CD3+), CD8+ T cells, CD4+ T cells, NK cells (CD3- NK1.1+), B cells (CD19+), or Myeloid cells (CD1 lb+). Percentage of CD45- cells made up by BPPNM cancer cells (GFP+) are also displayed. P values computed with a one-way ANOVA followed by Tukey’s post-hoc test are displayed above each figure. Data represented as mean ± SD.

[0050] Figure 22: Change in mouse mass during combination therapy tumor efficacy studies. (A) Mice were inoculated with 3×106BPPNM cells IP and dosed with 20 µg of STING or Scrpeptide conjugate delivered by LNP IP at 10, 13, 16, 19, and 22 days after inoculation. A subset of groups were additionally treated with 100 pg aPD-1 antibody at 11 and 17 days after inoculation. Groups included N = 6 (PBS, STING LNP, Scr LNP), N = 5 (STING LNP + aPD-1 ), or N = 4 (PBS + aPD-1, Scr LNP + aPD-1) mice. (B-C) Change in mouse mass was measured daily, displaying (B) mean ± SD over the treatment period and (C) individual mouse change in mass.

[0051] Detailed Description

[0052] As used herein and unless otherwise indicated, the terms “a” and “an” are taken to mean “one”, “at least one” or “one or more”. Unless otherwise required by context, singular terms used herein shall include pluralities and plural terms shall include the singular.

[0053] Unless the context clearly requires otherwise, throughout the description and the claims, the words ‘comprise’, ‘comprising’, and the like are to be construed in an inclusive sense as opposed to an exclusive or exhaustive sense; that is to say, in the sense of “including, but not limited to”. Words using the singular or plural number also include the plural or singular number, respectively. Additionally, the words “herein,” “above” and "below" and words of similar import, when used in this application, shall refer to this application as a whole and not to any particular portions of this application.

[0054] All embodiments of any aspect of the disclosure can be used in combination, unless the context clearly dictates otherwise.

[0055] As used herein, "about" means plus or minus 5% of the particular value.

[0056] In one aspect, the disclosure provides compositions, comprising a conjugate, wherein the conjugate comprises

[0057] (a) a plurality of peptides, wherein the plurality of peptides comprise peptides comprising (i) a binding site for TANK -binding kinase 1 (TBK1), and (ii) a binding site for interferon regulatory factor 3 (IRF3); and

[0058] (b) a surface to which the plurality of peptides are covalently conjugated.

[0059] The conjugate mimics activated STING when multivalently displayed, letting it act directly on downstream proteins TBK1 and IRF3. The composition can thus be delivered to the cytosol and interact with cytosol resident proteins to activate the desired TBK1 and IRF3 cellular signaling pathw ays for therapeutic purposes, such as treating cancer.In some embodiments the binding site for TBK1 comprises or consists of the amino acid sequence PxPLR, wherein x is any amino acid, and (ii) the binding site for IRF3 comprises or consists of the amino acid sequence pLxIS, where p is a hydrophilic amino acid, and x is any amino acid. In other embodiments, the binding site for TBK1 comprises or consists of the amino acid sequence selected from the group consisting of (i) PLPLR (SEQ ID NO: 5); (ii) (D / E)XPXPLR(S / T)D (SEQ ID NO: 6), wherein X denotes any amino acid; and (iii) PLPLR(T / S)D (SEQ ID NO: 7). See, for example, Nature volume 567, pages 394- 398 (2019); Molecular Cell Volume 84, issue 13 p2436-2454. elO July 11, 2024; and Nature volume 569, pages 718-722 (2019).

[0060] The peptide sequences used in the examples are exemplary; the sequences outside of the TBK1 and IRF3 binding motifs serve as a flexible linker that lets the binding sites have additional conformational flexibility. The shorter the peptides, the less expensive they are to synthesize at high purity, and will enable a higher negative charge to peptide ratio, which make the conjugate easier to encapsulate in a delivery vehicle. The 39 amino acid versions (SEQ ID NO: 1 or 2) are intrinsically disordered, such that they are resistant to denaturing or disrupting their folded structure during processing to conjugate it to the polymer (or encapsulation in a delivery' vehicle). Longer sequences are also effective, but will lower the negative charge to mass ratio of the conjugate.

[0061] In one embodiment, the plurality of peptides comprises:

[0062] (i) a first plurality of peptides that comprise a binding site for TBK1; and (ii) a second plurality of peptides that comprise a binding site for IRF3.

[0063] In this embodiment, the binding sites are on different peptides, and thus the plurality of peptides comprises two separate plurality of peptides. The first and second plurality of peptides may be in any ratio suitable for an intended purpose of the composition. In one embodiment, the first plurality of peptides and the second plurality of peptides are present in the composition in about an equal (50:50) molar ratio.

[0064] In other embodiments, the TBK1 and IRF3 binding sites are on the same peptides. In some such embodiments, the plurality of peptides comprise or consist of the amino acid sequence selected from (i) RLLISxxxxPLPLRTD (SEQ ID NO: 8), wherein x is any amino acid, and (ii) ELLISxxxxPLPLRTD (SEQ ID NO: 9), wherein x is any amino acid. In another embodiment, the plurality of peptides each comprise or consist of the amino acid sequence selected from (i) RLLISGMDQPLPLRTD (SEQ ID NO: 10), and (ii)ELL1SGMEKPLPLRTD (SEQ ID NO: 11). In another embodiment, the plurality of peptides each comprise or consist of the amino acid sequence selected from SEQ ID NO: 1-2.

[0065] VTMNAPMTSVAPPPSVLSQEPRLLISGMDQPLPLRTDLI (SEQ ID NO: 1; Mouse STING C-terminal peptide) VTVGSLKTSAVPSTSTMSQEPELLISGMEKPLPLRTDFS (SEQ ID NO: 2; Human STING C-terminal peptide)

[0066] Full length

[0067] Mouse STING

[0068] MPYSNLHPAI PRPRGHRSKY VALIFLVASL MILWVAKDPP NHTLKYLALH LASHELGLLL KNLCCLAEEL CHVQSRYQGS YWKAVRACLG CPIHCMAMIL LSSYFYFLQN TADIYLSWMF GLLVLYKSLS MLLGLQSLTP AEVSAVCEEK KLNVAHGLAW SYYIGYLRLI LPGLQARIRM FNQLHNNMLS GAGSRRLYIL FPLDCGVPDN LSVVDPNIRF RDMLPQQNID RAGIKNRVYS NSVYEILENG QPAGVCILEY ATPLQTLFAM SQDAKAGFSR EDRLEQAKLF CRTLEEILED VPESRNNCRL IVYQEPTDGN SFSLSQEVLR HIRQEEKEE VTMNAPMTSVA PPPSVLSQEP RLLISGMDQP LPLRTDLI (SEQ ID NO: 12)

[0069] Human STING

[0070] MPHSSLHIPSI PCPRGHGAQK AALVLLSACL VTLWGLGEPP EHTLRYLVLH LASLQLGLLL NGVCSLAEEL RH II-ISRYRGS YWRTVRACLG CPLRRGALLL LSIYFYYSLP NAVGPPFTWM LALLGLSQAL NILLGLKGLA PAEISAVCEK GNFNVAHGLA WSYYIGYLRL ILPELQARIR TYNQHYNNLL RGAVSQRLYI LLPLDCGVPD NLSMADPNI R FLDKLPQQTG DRAGI KDRVY SNS I YELLEN GQRAGTCVLE YATPLQTLFA MSQYSQAGFS REDRLEQAKL FCRTLEDILA DAPESQNNCR LIAYQEPADD SSFSLSQEVL RHLRQEEKEE VTVGSLKTSA VPSTSTMSQE PELLISGMEK PLPLRTDFS (SEQ ID NO: 13)In one embodiment, each peptide in the plurality of peptides does not comprise or consists of a full length STING protein. In other embodiments, each peptide in the plurality of peptides is between 5-100 amino acids in length, or between 5-75 amino acids in length, or between 5-50 amino acids in length, or between 5-40 amino acids in length, or between 5-30 amino acids in length, or between 5-20 amino acids in length, or between 10-100 amino acids in length, or between 10-75 amino acids in length, or between 10-50 amino acids in length, or between 10-40 amino acids in length, or between 10-30 amino acids in length, or between 10- 20 amino acids in length.

[0071] The peptides can be conjugated to the surface by any means, as suitable for an intended purpose. In one embodiment, each peptide in the plurality of peptides is covalently conjugated to the surface at its N-terminus. In one such embodiment, each peptide in the plurality of peptides comprises an azido-lysine residue at its N-terminus.

[0072] The composition may compri se any surface suitable for an intended use, so long as the surface comprises a material that can be delivered into the cell cytosol. In non-limiting embodiments, the surface is selected from the group consisting of a polymer, a nanoparticle, and a protein. In one embodiment, the surface comprises a polymer; in another embodiment, the polymer comprises an anionic polymer. Any anionic polymer may be used as appropriate for an intended use. In various embodiments, the anionic polymer is selected from the group consisting of a negatively charged poly-amino acid, a negatively charged polysaccharide, poly(carboxylic acid), poly(sulfate), poly(phosphate), co-polymers thereof, and functionalized versions thereof. In other embodiments, the anionic polymer is selected from the group consisting of poly(L-glutamic acid), poly (D -glutamic acid), poly(L-aspartic acid), poly(D-aspartic acid), polyfacrylic acid) (PAA), poly(methacrylic acid) (PMAA), poly(ethylacrylic acid) (PEAA), poly(propylacrylic acid) (PPAA), poly(butylacrylic acid) (PBAA), poly(allylamine hydrochloride)-citraconic anhydride (PAH-Cit), hyaluronic acid, and dextran sulfate, co-polymers thereof, and functionalized versions thereof. In a specific embodiment, the anionic polymer comprises poly(L-glutamate).

[0073] The polymers may be of any length as suitable for an intended purpose. In various embodiments, the polymers may comprise at least 10 monomeric units. The number of monomer units will depend on the polymer used and the intended use. In some embodiments, the polymer may comprise at least 10, 25, 50, 75, 100, 200, 250, 300, or more monomeric units. In other embodiments, the polymer may comprise between 20-5000 monomeric units, or between 20-2500 monomeric units, or between 20-1000 monomericunits, or between 20-800 monomeric units. In a further embodiment, the polymer may comprise about 300 monomeric subunits.

[0074] Tire peptides may be conjugated to the surface (such as a polymer) using any chemistry as suitable for a specific surface. In various non-limiting embodiments, the following linker groups can be added to the peptides or surface, to facilitate conjugation:

[0075] (a) N-Hydroxysuccinimide (NHS)-ester to primary amine;

[0076] (b) Maleimide to thiol;

[0077] (c) Azide to alkyne;

[0078] (d) Azide to dibenzocyclooctyne (DBCO); or

[0079] (e) Tetrazine to trans-cyclooctene (TCO).

[0080] In one embodiment, the anionic polymer comprises a plurality of alkyne residues conjugated to the anionic polymer. It is not required that all monomeric units of the anionic polymer be functionalized for peptide binding. In various embodiments, at least 10%, 25%, 50%, 75%, 90%, or all of the anionic polymer units comprise alkyne residues conjugated to the anionic polymer. In another embodiment, at least some of the alkyne residues are also conjugated to peptides in the plurality of peptides. Similarly, it is not required that all alkyne residues are also conjugated to peptides. In various embodiments, at least 10%, 25%, 50%, 75%, 90%, or all of the alkyne residues are also conjugated to peptides in the plurality of peptides.

[0081] The peptides may be conjugated to the surface, such as polymers, at any positions on the surface as appropriate for an intended use. By way of non-limiting example, the peptides may be conjugated to side chains of a polymer (such as a linear polymer), or may be conjugated at the ends of a polymer with non-linear geometry (such as a dendrimer).

[0082] Functionalization of the peptide may be done at any position on the peptides so long as it does not interfere with the TBK1 and IRF3 binding sites. In one embodiment, an azide moiety (including but not limited to an azido-lysine residue) is present at the N-temiinus of each of each peptide in the plurality of peptides, and is conjugated to an alkyne residue on the anionic polymer.

[0083] In another embodiment, the composition comprises bifunctional linkers between the surface and some or all of the plurality of peptides. Any bifunctional linker may be used as appropriate for an intended purpose. Such linkers can, for example, provide more flexibility in the conjugates. In some embodiments, the bifunctional linkers comprises polyethylene glycol (PEG) linkers, including but not limited to azido-PEG4-NHS ester linkers.The valency of peptides in the conjugate may be any valency as deemed appropriate for an intended use. In various embodiments, the plurality of peptides comprises present in the conjugate comprises at least 3 peptides, or at least 4 peptides, or at least 8 peptides, or at least 25 peptides, or at least 100 peptides.

[0084] For therapeutic use, the conjugate may contained within a delivery vehicle that enables delivery’ of the conjugate to the cytosol. Any such delivery' vehicle may be used. In one such embodiment, the delivery vehicle comprises a delivery vehicle for negatively charged cargo. In another embodiment, the delivery vehicle comprises a lipid nanoparticle or a cationic polymer nanoparticle. In one embodiment, the cationic polymer nanoparticle comprises poly(beta-amino ester), polyethylene imine (PEI), polyamidoamine (PAMAM), and / or chitosan. In another embodiment, the poly(beta-amino ester) comprises the structure selected from the following, wherein n is 3-100

[0085]

[0086] In a specific embodiment, the cationic polymer nanoparticle comprises polyethyleneimine (PEI).

[0087] In another embodiment, the delivery vehicle comprises a lipid nanoparticle. Any lipid nanoparticle may be used as appropriate for an intended purpose. In one embodiment, the lipid nanoparticle comprises an ALC-0315:cholesterol: 18: 1 (A9-Cis) PE (DOPE): DMG PEG2000 core. In various embodiments, the components may be present in a range of molar ratios such as:

[0088] • Ionizable lipid (ALC-0315): 30-60%

[0089] • Sterol (cholesterol): 20-75%

[0090] • Phospholipid (DOPE): 5-30%

[0091] • PEG-lipid (DMG PEG2000): 0.1-5%

[0092] In another embodiment, the lipid nanoparticle comprises an about 50:38.5:10:1.5 molar ratio of ALC-0315: cholesterol: 18:1 (A9-Cis) PE (DOPE): DMG PEG2000 core. In another embodiment, the lipid nanoparticle comprises an about 43.8:23.5:31.3: 1.5 molar ratioof ALC-0315:cholesterol: 18:1 (A9-Cis) PE (DOPE): DMG PEG2000. In another embodiment, the lipid nanoparticle comprises an about 50:38.5: 10: 1.5 molar ratio of ALC-0315:cholesterol: 18:1 (A9-Cis) PE (DOPE): DMG PEG2000, In another embodiment, the lipid nanoparticle may comprise different ionizable lipids (DLin-MC3-DMA, DLin-KC2-DMA, C 12-200, C-KK-E12, SM-102) or cationic lipids (l,2-dioleoyl-3 -trimethylammonium propane (DOTAP)) or phospholipids (18:0 PC (DSPC), 18:1 d9-Cis PC (DOPC)), in the same ratios as listed above. In another embodiment, the lipid nanoparticle comprises an ALC- 0315: cholesterol: 18:1 (A9-Cis) PC (DOPC): DMG PEG2000 core. In various embodiments, the components may be present in a range of molar ratios such as:

[0093] • Ionizable lipid (ALC-0315): 20-60%

[0094] • Sterol (cholesterol): 10-75%

[0095] • Phospholipid (DOPC): 5-60%

[0096] • PEG -lipid (DMG PEG2000): 0.1 -5 %

[0097] In another embodiment, the lipid nanoparticle comprises an about 43.8:23.5:31.3:1.5 molar ratio of ALC-0315:cholesterol: DOPC: DMG PEG2000. The delivery vehicles can be modified as appropriate for an intended use. In some embodiments, the delivery vehicles may be functionalized with targeting ligands (such as proteins) that allow cell-specific delivery, where the targeting ligands bind to cell surface proteins on a cell that the composition is to be delivered to.

[0098] The polymers can be incorporated into the delivery’ vehicles by any means appropriate, depending on specific of the delivery vehicle. In one embodiment, the polymer is non-covalently adsorbed to the delivery surface vehicle. An exemplary such technique is disclosed in ACS Nano 2020, 14, 2, 2224-2237.

[0099] The compositions can be combined as appropriate with other molecules, such as therapeutic small molecules, peptides, proteins, or other drugs, as needed. All compositions described herein, such as the conjugates, polymers, and delivery vehicles, are contemplated as compositions further comprising an additional therapeutic molecule or multiple additional therapeutic molecules. In one embodiment, the additional therapeutic molecule comprises an immune checkpoint inhibitor. Any immune checkpoint inhibitor is contemplated.

[0100] In another embodiment, the immune checkpoint inhibitor comprises a PD-1 inhibitor. Contemplated PD-1 inhibitors include, but are not limited to, a-PD-1 antibodies, a-PD-1 antibody conjugates, or a-PD-1 antibody fragments. Non-limiting examples of PD-1 inhibitors include: camrelizumab, cemiplimab, cetrelimab, nivolumab, pembrolizumab,penpulimab, pidilizumab, retifanlimab, sintilimab, spartalizumab, sugemalimab, and / or toripalimab.

[0101] The disclosure also provides pharmaceutical compositions comprising:

[0102] (a) the composition of any embodiment or combination of embodiments herein; and

[0103] (b) a pharmaceutically acceptable carrier.

[0104] The pharmaceutical compositions of the disclosure can be used, for example, in the methods of the disclosure described herein. The pharmaceutical composition may further comprise (a) a lyoprotectant; (b) a surfactant; (c) a bulking agent; (d) tonicity adjusting agent; (e) a stabilizer; (f) a preservative and / or (g) a buffer.

[0105] In some embodiments, the buffer in the pharmaceutical composition is a Tris buffer, a histidine buffer, a phosphate buffer, a citrate buffer or an acetate buffer. The pharmaceutical composition may also include a lyoprotectant, e.g. sucrose, sorbitol or trehalose. In certain embodiments, the pharmaceutical composition includes a preservative e.g. benzalkonium chloride, benzethonium, chlorohexidine, phenol, m-cresol, benzyl alcohol, methylparaben, propylparaben, chlorobutanol, o-cresol, p-cresol, chlorocresol, phenylmercuric nitrate, thimerosal, benzoic acid, and various mixtures thereof. In other embodiments, the pharmaceutical composition includes a bulking agent, like glycine. In yet other embodiments, the pharmaceutical composition includes a surfactant e.g., polysorbate-20, polysorbate-40, polysorbate- 60, polysorbate-65, polysorbate-80 polysorbate-85, poloxamer-188, sorbitan monolaurate, sorbitan monopalmitate, sorbitan monostearate, sorbitan monooleate, sorbitan trilaurate, sorbitan tristearate, sorbitan trioleaste, or a combination thereof. The pharmaceutical composition may also include a tonicity adjusting agent, e.g., a compound that renders the formulation substantially isotonic or isoosmotic with human blood.

[0106] Exemplary tonicity adjusting agents include sucrose, sorbitol, glycine, methionine, mannitol, dextrose, inositol, sodium chloride, arginine and arginine hydrochloride. In other embodiments, the pharmaceutical composition additionally includes a stabilizer, e.g., a molecule which, when combined with a protein of interest substantially pre vents or reduces chemical and / or physical instability of the protein of interest in lyophilized or liquid form. Exemplary stabilizers include sucrose, sorbitol, glycine, inositol, sodium chloride, methionine, arginine, and arginine hydrochloride.The compositions may be the sole active agent in the pharmaceutical composition, or the composition may further comprise one or more other active agents suitable for an intended use.

[0107] In a further aspect, the disclosure provides methods for treating cancer, comprising administering to a subject in need thereof the composition or pharmaceutical composition of any embodiment or combination of embodiments herein. As used herein, "treat" or "treating" means accomplishing one or more of the following: (a) reducing the severity of the disorder; (b) limiting or preventing development of symptoms characteristic of the disorder(s) being treated; (c) inhibiting worsening of symptoms characteristic of the disorder(s) being treated; (d) limiting or preventing recurrence of the disorder(s) in patients that have previously had the disorder(s); and (e) limiting or preventing recurrence of symptoms in patients that were previously symptomatic for the disorder! s).

[0108] The subject may be any subject that has a relevant disorder. In one embodiment, the subject is a mammal, including but not limited to humans, dogs, cats, horses, cattle, etc.

[0109] In one embodiment, the subject has metastatic melanoma or ovarian cancer.

[0110] In various embodiments of the method, the composition or pharmaceutical composition may be administered in any form known in the art. Non-limiting examples of administering the composition or pharmaceutical composition include: intraperitoneally, intravenously, intra-arterially, subcutaneously, intramuscularly, intranasally, orally, or topically. In one embodiment of the method, the composition or pharmaceutical composition is carried out via intraperitoneal (IP) administration.

[0111] Contemplated dosing regiments may also include single deliveries, or repeated deliveries (for example: 2, 3, 4, 5, or more administrations). Repeated doses may be delivered via the same or different delivery method, concentration, or with or without similar agents. The administering may be performed daily, weekly, bi-weekly, monthly, bi-monthly, or annually, for any total length of time. In some embodiments, the administering is performed weekly for about one year. In other embodiments, the administering is performed monthly for about one year.

[0112] Examples

[0113] Immunotherapy has been successful in the treatment of cancer, but only achieves long-term remission in a small fraction of patients. The cytosolic DNA sensing stimulator of interferon genes (STING) pathway has garnered interest as a therapeutic target to overcomethis by inflaming immunosuppressive tumors that would otherwise respond poorly to traditional immunotherapies. Activated by the leakage of DNA to the cytosol that has been observed in cancer cells, this innate immune pathway is central to the spontaneous immune mediated elimination of early tumors by initiating a type I interferon (1FN) response.

[0114] A less explored challenge relates to the fundamental biology of STING signaling, which is frequently blunted or inactivated entirely in human cancer, presumably as a mechanism to escape from immune surveillance. One challenge of drugging this pathway is that STING signaling is commonly inactivated in human cancer, often by epigenetic silencing of the STING gene. For example, STING is not expressed in the cancer cells of 50% of metastatic melanoma patients and 72% of stage III ovarian cancer patients. This renders STING agonists, the strategy typically used to drug this pathway, ineffective in these cells because they have no protein target to act upon.

[0115] STING signaling (Figure 1 A) begins when cytosolic DNA activates cGAMP synthase (cGAS), which catalyzes the production of the cyclic dinucleotide 2’, 3 ’-cGAMP.

[0116] 2 ’,3 ’-cGAMP binds to a pocket in the cytosolic domain of the endoplasmic reticulum (ER) resident STING protein, inducing conformational changes that initiate STING trafficking, multimerization, and the freeing of STING’s disordered C-temiinal tail. Activated STING multimers act as a scaffold presenting many copies of the C-terminal tail, which contains a PLPLRT / SD (SEQ ID NO: 7) motif that recruits the kinase TANK-binding kinase 1 (TBK1). Close proximity of multiple TBKI molecules on this scaffold allows for its transautophosphorylation and activation. TBK1 then phosphorylates pLxIS motif on the C-terminal tail of a neighboring STING molecule. The phosphory lated pLxIS motif recruits the transcription factor interferon regulatory factor 3 (IRF3), which itself is phosphory lated by TBKI and converted into its dimeric active state. Active IRF3 enters the nucleus and drives the transcription of Type I IFNs and interferon stimulated genes, that ultimately drive the downstream innate immune response. A wide range of strategies to disrupt STING signaling have been observed in cancer, with examples including degradation of cytosolic DN A by TREX1 upregulation, inhibition of STING trafficking by PCSK9 upregulation, and inhibition of TBK1 by NF2 mutants. While downstream signaling through the cytosolic kinase TBK1 and transcription factor IRF3 is often intact, strategies to activate these molecules directly were not available. One of the most common mechanisms of inactivation is the epigenetic silencing of upstream proteins cGAS and STING (Figure 1 A). Loss of STING expression in particular is observed in 50% of patients with metastatic melanoma and over 70% of patents with metastatic ovarian cancer.Two main approaches have been developed to overcome STING dysregulation in cancer, epigenetic reprogramming and STING mimicking therapeutics. Existing epigenetic reprogramming strategies utilize inhibitors of proteins involved in the epigenetic silencing of STING, and have been shown to reactivate STING expression and improve the efficacy of STING agonists. However, current epigenetic reprogramming strategies risks off target effects on cancer and host cell phenotypes, and still requires subsequent treatment with a STING agonist, complicating treatment strategies. Instead, a more direct strategy is to bypass silenced STING by “replacing” its function with mimics to directly activate TBK1 and IRF3 downstream of STING, bypassing the need for the cancer cell’s endogenous STING entirely. Existing STING mimicking therapeutic strategies deliver an activated version of the STING protein to the cytosol where it can interact with endogenous TBK1 and IRF3. Tire STING protein, or a nucleic acid encoding for the STING protein, must first be converted into an active state by premixing with a ligand, inducing protein aggregation by interactions with a delivery vehicle, introduction of mutations that drive constitutive multimerization, or fusion to a self-oligomerizing protein domain.

[0117] We previously demonstrated that delivery of exogenous the STING protein fragment STINGΔTM complexed with its ligand cGAMP is able to directly activate downstream signaling molecules TBK1 and IRF3 in the cytosol (Error! Reference source not found. A). This bypasses tire common roadblock in cancer, allowing for signaling activation even in cells that do not express endogenous STING and thus do not respond to traditional STING agonists (Error! Reference source not found. B).

[0118] This system demonstrated efficacy in many mouse models of cancer. Despite its potential, delivery of this protein complex is challenging and the magnitude of STING activation in is weak in many types of cells, so methods to improve its potency and delivery are desirable. Multimerization is required for activation, but is often limited to lower valency tetramerization or left to uncontrolled processes inside the cell. Further, existing designs use the full STING protein or the entire cytosolic domain that contains the folded ligand-binding domain (Figure 1C). Mechanistically, this ligand binding domain controls STING multimerization but does not contain the TBK1 and 1RF3 interaction motifs necessary to activate downstream signaling, so it should not be required for activity when alternative means of inducing multimerization are used. This has been confirmed in prior biochemical studies, where the STING C-terminal tail alone (Figure 1C) was able to activate TBK1 and IRF3 in an in vitro reconstitution system without the ligand binding domain. We observed that STING mimics retain their ability to activate TBK1 and IRF3 in cells even after thermaldenaturation was used to unfold the STING ligand binding domain, providing further evidence that the ligand binding domain is not required in a STING mimicking therapeutic. Ultimately, we believe that the design of STING mimics could be simplified while developing more effective therapeutics. This disclosure describes a new strategy for activating STING signaling in cancer cells that improves upon STINGATM, with increased potency and greater ease to delivery’ to the cytosol.

[0119] Technology

[0120] We hypothesized that a molecule presenting multiple copies of only the STING C-terminal tail would contain the protein interaction motifs necessary’ to engage and activate downstream proteins TBK1 and IRF3 (Figure 2). We describe the development of a peptidepolyanion conjugate that can be delivered to the cytosol and interact with cytosol resident proteins to activate a desired cellular signaling pathway for therapeutic purposes. The peptide-polyanion conjugate is generated through the covalent attachment of multiple copies of a peptide containing protein-protein interaction domains to a polyanion (Error! Reference source not found.). The negative charge of the polyanion allows for efficient encapsulation and intracellular delivery’ using repurposed delivery' vehicles originally designed for the delivery’ of negatively charged nucleic acids, such as polymer or lipid nanoparticles. Peptides are selected to contain the protein-protein interaction domains that recruit and activate the desired signaling proteins, multivalent display on a single polyanion can replicate the multivalent active state of many intracellular signaling proteins. We built this molecule on a negatively charged polymer, or poly anion, scaffold to promote electrostatic interactions that we hypothesized would enable easy’ encapsulation and delivery’ using existing nanocarriers designed to deliver negatively charged nucleic acids. When delivered to the cytosol, this multivalent peptide-polyanion conjugate was observed to phosphorylate and activate TBK1 and IRF3, resulting in the TBK1 -dependent expression of IRF3 controlled genes. Peptide-polyanion conjugate treatment initiated signaling comparable to a high dose of the STING agonist ADU-S100 in a cancer cell line with functional STING signaling, and also retained activity’ even in a STING-silenced cancer cell line that showed no response to ADU-S100. To enable application as a therapeutic, we developed a lipid nanoparticle formulation (LNP) capable of effectively delivering this conjugate material to the cytosol. As a proof of concept, we used conjugate loaded LNP for the immunotherapy of ovarian cancer, a disease that has responded poorly to traditional immunotherapies and has an extremely’ high frequency ofSTING inactivation that may limit existing STING agonists. In mice, administration of conjugate-loaded LNP resulted in great increases in IRF3 controlled cytokine levels, and repolarized the tumor microenvironment towards an inflamed state. Ultimately, STING peptide -polyanion conjugate therapy was able to shrink tumors and prolong survival in multiple mouse models of metastatic ovarian cancer.

[0121] Results

[0122] Designing a multivalent peptide-polyanion conjugate material

[0123] Despite prior biochemical work demonstrating that the STING C -terminal tail alone is capable of activating TBKI and IRF3, it is still challenging to apply this strategy therapeutically. This demonstration occurred in an in vitro reconstitution system where STING C-terminal tail was added directly to solution containing TBKI and IRF3, whereas a therapeutic would need to cross the cell membrane before it could interact with TBKI and IRF3 in the cytosol. Additionally, only the high molecular weight aggregates of the STING C-terminal tail and not individual soluble molecules are active, these larger aggregates would presumably be even more difficult to deliver into the cytosol. In our prior work, we attempted to deliver truncations of the STING protein into the cytosol of live cells using a commercial transfection reagent. Truncations containing only C-terminal tail and a short segment from the crystallizable domain were able to activate IRF3, supporting our hypothesis that the crystallizable domain of STING was not necessary in a STING mimicking therapeutic. However, this fragment activated IRF3 only weakly compared to the entire STING cytosolic domain, and tire C-terminal tail segment that was active in biochemical work showed no detectable activity in live cells. This means that the STING C-terminal tail fails or be delivered into the cytosol effectively, or to multimerize into an active form, or both. We developed a multivalent STING C-terminal tail peptide-polyanion conjugate to overcome both of these challenges, by designing this conjugate material for easier delivery to the cytosol and high valency.

[0124] To activate STING signaling, a peptide taken from the STING protein’s C-terminal tail was selected. This peptide contains a PxPLR motif to recruit TBK1 and a pLxIS motif to recruit IRF3 (Error! Reference source not found. A). Multivalent display of this proteinprotein interaction motifs mimics the multivalent activated state of STING, acting as a scaffold for TBKI and IRF3 interaction and activation, leading to the transcription ofdownstream genes like Type I IFNs that are useful for generating an anti-cancer immune response.

[0125] Section 1: Synthesis and Characterization of STING activating conjugate

[0126] We built the conjugate material on a poly(L-glutamate) backbone, as this polyanion has a high negative charge density, is biocompatible, and a controlled number of carboxylate side chains can be modified to generate a reactive alkyne handle. A 300-repeat unit polymer with alkyne groups grafted onto 10% of side chains was selected, to enable high valency display of 30 peptides on each molecule while retaining a strong negative charge. The STING C -terminal tail, mouse STING(340-378), is only 39 amino acids, allowing solid phase synthesis (Figure 3A). An azido-lysine was added as a reactive handle during solid phase synthesis, on the N-terminus of the peptide to ensure that reaction with the polymer doesn’t sterically interfere w ith tire known protein interaction motifs on the C-terminus of the peptide. Peptide -polyanion conjugates were synthesized by copper catalyzed azide-alkyne cycloaddition with excess azide-peptide (Figure 3B).

[0127] Poly (L-glutam ate) was selected as the polymer backbone due to its strong negative charge density, biodegradability, biocompatibility’, a controlled number of carboxylate side chains can be modified to generate a reactive alkyne handle, and commercial availability with alkynes grafted on to various percentages of side chains. Tire STING C -terminal tail, mouse STlNG(340-378), is only 39 amino acids, allowing solid phase synthesis (Figure 3A). The 39 amino acid segment from the C-temiinal tail of mouse STING was generated through solidphase synthesis with an azido-lysine at its N-terminus; azido-lysine was added as a reactive handle during solid phase synthesis, on the N-terminus of the peptide to ensure that reaction with the polymer doesn’t sterically interfere with the known protein interaction motifs on the C-terminus of the peptide. Alkynes and azides were joined together using copper-catalyzed click chemistry to generate the final conjugate (Error! Reference source not found. B). SDS-PAGE was used to verify that this reaction occurs successfully (Error! Reference source not found. C) and no remaining alkynes were detected on the polymer using an alkyne quantification assay (Error! Reference source not found. D), indicating complete conversion alkynes on the polymer backbone to peptides.

[0128] Complete conversion was further confirmed by the disappearance of detectable alkynes via NMR The removal of excess peptide after purification was confirmed using sizeexclusion chromatography, which also demonstrated the expected increase in size of the peptide-polyanion product compared to peptide and polyanion starting materials Figure 3E).

[0129] Tire mSTING(340-378) peptide showed a random coil secondary structure as measured by circular dichroism, both alone and after conjugation to a polyanion (Figure 3F), as expected for the disordered STING C-terminal tail.

[0130] To measure that the STING C-terminal tail peptide-polyanion conjugate material was capable of activating STING signaling, the commercial transfection reagent TransIT-X2 was used to deliver this material to the cytosol of the HEK293T-derived reporter cell line. This reporter expresses luciferase under the control of the target transcription factor IRF3, allowing IRF3 activation to be quantified. Treatment with a multivalent peptide-polyanion conjugate containing 30-copies of the mSTING(340-378) peptide resulted in strong IRF3 activation, nearly 1000-fold over basal levels (Figure 3G). No increase in IRF3 activation was observed after treatment with a matched conjugate using a scramble control peptide, indicating that this activity depends on the mSTING(340-378) peptide sequence and is not a result of non-specific effects of the transduction reagent or introduction of exogenous material into the cytosol. A monovalent conjugate with only a single copy of

[0131] the mSTING(340-378) peptide conjugated to the C -terminus of a size-matched 300 repeat unit poly(L-glutamate) w as synthesized to control for the valency of peptide display. At an equal molar dose of peptide, tire monovalent conjugate show ed no increase in IRF3 activity, confirming that multivalent display is required for activity (Figure 3G), Multiple independently synthesized batches of multivalent peptide-polyanion conjugate showed consistent IRF3 activity in HEK293T (Figure 11). This conjugate material appears resistant to thermal denaturation, as it maintains consistent IRF3 activity in HEK293T after exposure a 95 °C thermal stress of up to 10 minutes. Overall, these results confirm that multivalent display of tire STING C-terminal tail sequence on a poly anion can activate tire target transcription factor IRF3, so tlie multivalent conjugate was examined further.

[0132] Section 2: Biological Activity and Delivery of STING activating conjugate

[0133] Using a HEK293T-derived IRF reporter cell line, we examined the ability of this conjugate to activate the transcription factor IRF3 downstream of STING activation. Using a commercial transfection reagent TransIT-X2, we demonstrate that the STING peptide-polyanion conjugate described above was able to activate the IRF3 transcription factor (Error! Reference source not found. A), and that it both had higher activity at the same dose of STTNGATM and could be delivered at higher doses before being limited by toxicity, widening the therapeutic window.

[0134] The effects of the STING peptide -polyanion conjugate was examined at the cellular level, using TransIT-X2 for delivery’ with the cytosol. On target activation of TBK1 and IRF3 downstream in the STING signaling cascade w as demonstrated in HEK293T. Confocal microscopy show ed puncta of TBK1 co-localized w ith puncta of the peptide conjugate, providing evidence that the conjugate is capable of recruiting TBK1 in the cytosol (Figure 6A) Phosphorylation of TBK1 and IRF3 was observed by Western blot, confirming on target activation of these molecules ( Figure 6B). Pretreatment of cells w ith the specific TBK1 inhibitor MRT67307 completely' eliminated IRF3 activation after peptide conjugate treatment (Figure 6C), confirming that TBK1 is required for activity of the peptide conjugate. Overall, these results provide evidence delivery' of the multivalent peptide-polyanion conjugate to the cytosol is capable to mimicking activated STING to activate TBK1 and IRF3.

[0135] We then examined if the peptide-polyanion conjugate could activate IRF3 in cancer cells w ith dysfunctional STING, We compared treatment w ith the clinically advanced STING agonist ADU-S100 to treatment with peptide-polyanion conjugate in ovarian cancer cell lines: KURAMOCHI cells that express functional S TING and A2780 cells that have lost STING expression (Figure 6D). As expected, the STING agonist ADU-S100 only induced secretion of the IRF3 controlled cytokines CXCL10 and IFN-β in STING-proficient KURAMOCHI cells, while no detectable cytokines were observed after treatment of ST1NG-deficient A2780 cells (Figure 6E-6F). Treatment with the STING peptide-polyanion conjugate resulted in secretion of CXCL10 and IFN-P in both KURAMOCHI and A2780 (Figure 6E-6F). This demonstrates that the STING peptide-polyanion conjugate is capable of activating desired innate immune signaling even in STING-deficient cancer cells that do not response to STING agonists.

[0136] The cellular response to treatment with the STING peptide-polyanion conjugate w as examined in greater detail using mRNA sequencing. Treatment with the STING-peptide polyanion conjugate was compared to treatment with the S TING agonist ADU-S100 in the STING-proficient cell line KURAMOCHI. At 6 h after treatment, over 500 differentially expressed genes were observed for both STING-peptide polyanion conjugate and ADU-S100 treatment compared to a buffer control. Transcriptional changes were highly similar betw een the STING agonist and STING mimic treatment. Compared to a buffer treated control,differential gene expression following STING peptide-polyanion conjugate was well correlated with gene expression following ADU-S100 treatment (line of best fit R2= 0.81), with no obvious outliers that showed large changes in expression with one treatment but not the other (Figure 6G). To contrast this, differential gene expression following Scr peptide- polyanion conjugate was only weakly correlated with gene expression following ADU-S100 treatment (line of best fit R2= 0.13), with genes that tvere strongly upregulated by ADU-S100 seeing no notable change after treatment with the control conjugate (Figure 611). Gene set enrichment analysis (GSEA) using the MSigDB hallmark gene sets showed 10 gene sets significantly enriched (adjusted P <.05) when comparing Scr vs. S TING peptide-polyanion conjugate treatment. These 10 enriched gene sets that were the same as the top 10 gene sets enriched (P <.05) by ADU-S100 treatment (Figure 611). Combined, these results provide evidence that STING peptide-polyanion conjugate treatment activates on-target STING signaling without any major induction off-target signaling. Gene sets strongly enriched after STING peptide-polyanion conjugate treatment are consistent with activation of IRF3 transcription factor and downstream signaling such as Interferon Alpha Response and Interferon Gamma Response, but also suggest strong activation of the NF-KB transcription factor with enrichment in TNFA Signaling VIA NFKB and IL6 JAK STAT3 Signaling (Figure 611). NF-κB is known to be following STING activation, downstream of TBK1 but independent of canonical 1RF3 activation. Strong increases in specific IRF3 controlled genes such as RSAD2, ISG15, and ISG20 and NF-KB controlled genes such as IL 6, CCL5, and RIGI are observed following both STING peptide-polyanion conjugate and ADU-S100 treatment (Figure 61). Activation of both IRF3 and NF-KB transcription factors was confirmed in a THP1 monocyte dual transcription factor reporter cell line, where treatment with the STING peptide-polyanion conjugate resulted in dose-dependent activation of IRF3 and NF-KB, at a similar ratio to treatment w ith ADU-S100 (Figure 12A-12B). These results demonstrate that the STING peptide-polyanion conjugate effectively engages therapeutically relevant innate immune signaling dow nstream of STING, including both IRF3 and NF-KB transcription factors.

[0137] The therapeutic effect of the STING peptide-polyanion conjugate requires successful delivery to the cytosol for the conjugate to engage its targets TBK1 and IRF3. Development of an effective delivery vehicle is crucial for the STING peptide-polyanion conjugate to be applied as a therapeutic. A w ide variety of delivery vehicles have been developed to encapsulate and deliver nucleic acids intracellularly, often by leveraging electrostatic interactions between the negatively charged nucleic acid cargo and a positively chargednanocarrier. We hypothesized that designing the STING peptide-polyanion conjugate material with a strongly negatively charged backbone would mimic the physicochemical properties of nucleic acids to allow for encapsulation and delivery using existing positively charged nanocarriers designed for nucleic acid delivery. To test this hypothesis, we attempted to deliver the STING peptide-polyanion conjugate using a panel of nucleic acid carriers, using IRF3 activation measured in a HEK293T-based reporter cell line as a functional measure of successful intracellular delivery. We tested lipid nanoparticle (LNP) formulations using ionizable lipids ALC-0315, MC3, and SM-102 from three clinically approved LNP therapeutics, using 50:10:38.5:1.5 (mol ratio) ionizable lipid: DOPE:cholesterol: DMG-PEG2000 to reflect the composition of clinically-approved LNP formulations. We also examined cationic polymer-based delivery' vehicles including linear poly(ethyleneimine) (PEI), the commonly used poly(p-amino ester) (PBAE) “Poly2"’, and the commercial transfection reagent TransIT-X2. All vehicles showed demonstrated functional delivery’ of the conjugate, demonstrated by a significant increase in IRF3 activation compared to a Scr control delivered using the same vehicle (Figure 7A). All vehicles showed a >200-fold increase in IRF3 activation relative to buffer treated controls, with the exception of PEI which showed a weaker 40-fold increase in IRF3 activation. The SM-102 LNP formulations showed approximately 10-fold increase in IRF3 activation, even when loaded with the inactive Scr peptide-polyanion conjugate. This is expected as LNPs themselves are known to activate innate immune signaling through interaction of the ionizable lipid with pattern recognitional receptors and damage caused during endosomal escape, with stronger innate immune activation being observed with SM-102 in prior literature. Overall, these results demonstrate that the STING peptide-polyanion conjugate can be effectively delivered to the cytosol using a variety of “off-the-shelf ’ nucleic acid delivery’ vehicles.

[0138] We decided to advance a LNP formulation for peptide-polyanion conjugate delivery (Figure 7B) due to this vehicle’s widespread use in the clinic. ALC-0315 was selected for its strong functional delivery ability and low off-target IRF3 activation of the LNP itself compared to SM-102 (Figure 7 A). The initial 50: 10:38.5: 1.5 (mol ratio) ALC- 0315: DOPE:cholesterol: DMG-PEG2000 formulation (labeled formulation A2) showed strong functional delivery in HEK293T in vitro, however, LNP size was larger than typical nucleic acid formulations at 200 nm, size increased during the centrifugal filtration process used to purify and concentrate LNPs, and the encapsulation efficiency (EE) was lower than ideal at 74% (Figure 13). We optimized the LNP formulation aiming to decrease size and increase encapsulation efficiency. To this end, we screened a library' of LNPs with varyingphospholipid components and compositions (Figure 7C). Ultimately, LNPs composed of DOPE or DOPC with high phospholipid and reduced cholesterol and ALC-0315 showed the most promise (labeled formulation A3 and C3, respectively), decreasing LNP diameter to approximately 100 nm, polydispersity index < 0.2 (Figure 7D), and EE to >85% (Figure 7E). All DOPE and DOPC based formulations performed similarly in the HEK293T functional delivery’ assay, leading to a >100-fold increase in IRF3 activation (Figure 7F), The inclusion of DSPC on the other hand rendered LNPs ineffective at delivering the conjugate to the cytosol, leading to only 2- to 10- fold increases in IRF3 activation (Figure 7F). We moved forward with the C3 formulation over the A3 formulation because the A3 formulation showed signs of instability during purification by centrifugal filtration, increasing in diameter by 20 nm (Figure 13).

[0139] The final C3 LNP formulation is prepared at a 43.8:31.3:23.5: 1.5 (mol ratio) ALC-0315: DOPE:cholesterol: DMG-PEG2000 (Figure 7G, Figure 14). Validation experiments generating LNPs loaded with both STING and Scr peptide-polyanion conjugate showed that this formulation produces 80-100 nm diameter LNPs with a polydispersity index < 0.2 (Figure 7H), a near neutral zeta potential (Figure 7I), and an EE of 85% (Figure 7 J). LNPs showed an apparent of pKa of 6.4 that is consistent with other ALC-0315 LNP formulations loaded with mRNA (Figure 15). CryoTEM showed that STING peptide-polyanion conjugate loaded C3 LNPs showed an electron-dense core typical of nucleic acid cargos, with a high degree of blebbing (Figure 7K). Ultimately, the C3 LNP formulation consistently generated sub- 100 nm nanoparticles with high peptide-polyanion conjugate encapsulation and effective cytosolic delivery in vitro.

[0140] We then examined more translationally relevant nucleic acid nanocarriers including lipid nanoparticles (LNPs) in addition to degradable poly (beta-amino ester) nanoparticles. We were able to successfully form LNPs containing the conjugate with an ALC-0315 based- formulation similar to the Pfizer™-BioNTech™ COVID-19 vaccine. We were able to form -100 nm nanoparticles and adsorb the ovarian cancer targeting polyanion poly(L-aspartate) (PIT)) to the surface to invert the charge (Error! Reference source not found. A-5C), demonstrating that cancer targeting chemistry can be incorporated in this formulation. These nanoparticles were able to deliver functional conjugate into the cytosol, measured by STING IRF3 transcription factor activation in a HEK293T reporter cell line (Error! Reference sourcenot found. C). We were also able to successfully form polyplexes containing the conjugate using the PBAE Poly3. We were able to form ~200 nm nanoparticle by incorporating the cancer targeting polymer PLD (Error! Reference source not found. D-5F), and were also able to deliver functional conjugate into the cytosol in HEK293T (Error! Reference source not found. F).

[0141] Advantages

[0142] Effective activation of STING signaling, with stronger activation and wider therapeutic window than previously developed therapy STING ATM.

[0143] Acts directly on downstream signaling proteins TBK 1 and IRF3, allowing for activation independent of exogenous STING expression in the target cell, allowing for activity in STING-deficient cancer cells that do not respond to traditional STING agonist drugs. Easy delivery into the cytosol using a variety of existing delivery vehicles, limiting delivery vehicle development required compared to protein STINGATM.

[0144] Potential to be a platform technology, STING peptide may be swapped for peptides containing other protein-protein interaction motifs to activate other cellular signaling pathways.

[0145] Conjugate treatment activates on-target innate immune response in vivo

[0146] We selected ovarian cancer as a proof of concept to apply the STING peptidepolyanion conjugate LNP as a therapeutic. There is a great clinical need to improve treatments for ovarian cancer, as a majority of patients are diagnosed after metastasis when existing therapeutics have limited benefit. T cell directed immunotherapies like checkpoint blockade have enabled durable tumor control in other metastatic cancers but have thus far shown limited efficacy in ovarian cancer patients, correlated with poor T cell infiltration and immunosuppressive microenvironment typical of ovarian cancer, STING signaling shows promise to repolarize immunosuppressive ovarian cancer tumors and promote anti-tumor immune responses. An extremely high rate of STING expression loss in over 70% of patents with metastatic disease suggests that there may be a benefit of using STING mimicking therapy. We selected intraperitoneal (IP) administration for STING peptide-polyanion conjugate LNP therapy. This route of administration is used clinically for ovarian cancertreatment, and has been observed to significantly improve accumulation in abdominal tumors when compared to intravenous administration.

[0147] We first examined pharmacokinetics of IP administered LNP in the BPPNM syngeneic mouse model of ovarian cancer. BPPNM mimics common mutations found in patients with homologous recombination-deficient high-grade serous ovarian carcinoma

[0148]

[0149] (Brca1− / − Trp53− / −R172H Pten− / − This model recapitulates immunosuppressive tumor microenvironment, resistance to checkpoint blockade therapies, and the abdominal metastasis pattern commonly observed in human ovarian cancer patients. Tracking the fluorescence of Cyanine5-labled peptide-polyanion conjugate, this therapy had a 2.5 h serum half-life that is in line with other LNP delivered nucleic acid therapeutics (Figure 8A). Biodistribution was measured at 4 h post IP administration with ex vivo IVIS (Figure 16A). BPPNM tumor luminescence showed largest tumors were present on the omentum, with tumor nodules also on the upper genital tract (UGT) and intestines (Figure 16B) The majority of recovered peptide-polyanion conjugate fluorescence was observed in highly tumored organs, with 25%, 13%, and 25% in the omentum, UGT, and intestines, respectively (Figure 16C-16D). Primary sites of off-target accumulation were the liver and kidneys, with 14% and 10% of recovered conjugate fluorescence, and there was minimal (< 2%) accumulation in the spleen, lungs, or heart (Figure 16C-16D). Cellular level distribution of conjugate was examined in the omental tumor and ascites (peritoneal fluid) using flow cytometry (Figure 17A). Conjugate association was expectedly high in the ascites as this fluid is the site of injection, with association observed in 56% of BPPNM cancer cells and 74% of stromal cells (Figure 17B). Conjugate association in the tumor was highest BPPNM cancer cells (15%), stromal cells (47%), DCs (14%), and macrophages, with higher accumulation in Ml-polarized (35%) compared to M2-polarized (9.1 %) macrophages (Figure 17C). Conjugate association with lymphoid cells was low overall, with the exception of B cells that showed moderate association in both the ascites (17%) and tumor (5.4%) (Figure 17B-17C).

[0150] On target biological activity was validated in vivo by measuring the appearance of IRF3 and NF-KB controlled cytokines. Serum levels of IRF3 controlled chemokine CXCL10 are below the ELISA assay the limit of detection (LOD) at baseline, but rise to detectable levels between 3 and 10 h, hitting a maximum at 6 h after treatment with STING peptide- polyanion conjugate LNP (Figure 8B). Cytokine levels in an omental tumor, ascites (peritoneal fluid) that fills the abdomen, and serum were examined at more detail 6 h post treatment with S TING or Scr peptide-polyanion conjugate (Figure 8C). BPPNM tumors hadelevated baseline cytokine levels as observed in prior work, treatment resulted in significant increases in CXCL10 (9.6-fold), IFN-P (2.6-fold), and IL-6 (2.5-fold) (Figure 8D). In ascites, treatment resulted in large increases in CXCL10 (>42-fold), IFN-P (>150-fold), IL-6 (>14- fold), TNF-a (> 16-fold), and IFN-y (>3.7 fold) levels above the scramble control, where cytokine levels were below the LOD for many replicates or low otherwise (Figure 8E). Similar cytokine increases were observed in the serum to a lesser magnitude, with CXCL10 (> 12-fold), IFN- (>9.8-fold), IL-6 (>7.7-fold) and TNF-a (>3.1-fold) levels increasing above the Scr control where most cytokines were below' the LOD (Figure 8F).

[0151] Conjugate shows therapeutic efficacy in metastatic ovarian cancer

[0152] With evidence that STING peptide-polyanion conjugate LNP leads to innate immune activation, we examined therapeutic efficacy in the BPPNM model of metastatic ovarian cancer (Figure 9A). Treatment with STING peptide-polyanion conjugate LNP resulted in significant survival improvement compared to controls (P =.006 vs PBS, P =.006 vs Scr LNP), increasing median survival to 37 days compared to 31 or 34 days for PBS or Scr controls, respectively (Figure 9B). Treatment resulted in a partial response with BPPNM luminescence shrinking over an order of magnitude, however, tumor growth resumed after treatment (Figure 9C-9D). To examine the generalizability of this treatment, 'e examined efficacy in the KPCA. C syngeneic mouse model of metastatic ovarian cancer (Figure 9E), which mimics mutations found in different subset of patients, homologous recombination-proficient high-grade serous ovarian carcinoma (KrasG12D / +Trp53− / R172HCcne1OEPik3caOE). The KPCA. C model has a highly immunosuppressive tumor microenvironment, and is more resistant to immune checkpoint blockade than the BPPNM model. Treatment with STING peptide-polyanion conjugate LNP resulted in significant survival improvement compared to controls (P =.02 vs PBS, P =.03 vs Scr LNP), increasing median survival to 33 days compared to 25 days for both PBS or Scr controls (Figure 9F). Treatment resulted in a partial response with KPCA. C luminescence shrinking over an order of magnitude, however, tumor growth resumed after treatment (Figure 9G-9H). Observing therapeutic efficacy in two challenging to treat and immunosuppressive ovarian cancer models demonstrates the promise of STING peptide-polyanion conjugate as an immunotherapy, even as only a monotherapy.

[0153] Administration of STING peptide-polyanion conjugate LNP was well tolerated over the course of treatment of both BPPNM and KPCA models. Mice lost an average of 5-10% initial body mass in the day following treatment, but this effect w'as transient and mice recovered to the initial mass within 1-2 days of each dose (Figure 18). Acute toxicity wasexamined by performing complete blood counts and serum chemistry panel 24 h after a single dose of S TING peptide-polyanion conjugate LNP or controls. Results were consistent with the known toxicity profiles of innate immune activation observed in previously examined STING activating therapies. A drop in total white blood cell counts was observed, driven by drops in blood lymphocytes and monocytes (Figure 19). A minor (<20%) but significant drop in platelet counts was also observed (Figure 19), however, platelet counts remained within a normal range for mice and changes were small compared to larger drops observed with systemic administration of small molecule STING agonists. No significant changes in red blood cell counts or quality were observed with S TING peptide-polyanion conjugate treatment (Figure 19). We also did not observe any signs of major liver or kidney damage, with no significant changes in liver enzymes, total protein, blood urea nitrogen, or creatinine (Figure 20).

[0154] Conjugate repolarizes the tumor microenvironment to improve response to checkpoint blockade

[0155] We evaluated the tumor immune response to the STING peptide-polyanion conjugate LNP in greater detail, focusing on the BPPNM model where there was more room to improve therapeutic response. Four doses of STING peptide-polyanion conjugate LNP (Figure 10A) resulted in major changes in the immune composition of the tumor compared to PBS and Scr controls, as measured by flow cytometry (Figure 10B) STING peptide-polyanion conjugate LNP treatment decreased the fraction of macrophages in the tumor, while increasing CD86 expression (Figure 10C) and decreasing CD206 expression (Figure 10D), indicating a shift from an immunosuppressive M2 -phenotype to an inflammatory M1-phenotype. Treatment decreased the fraction of DCs in the tumor (Figure 10B), while increasing CD86 expression (Figure 10E), indicating DC activation. Treatment resulted in an increase in the fraction of CD4+ and CD8 + T cells present in the tumor (Figure 10B), but also increased PD-1 expression on both types of T cell (Figure 10F), demonstrating successful T cell recruitment but suggesting that tire PD-1 immune checkpoint may limit T cell effector function. Overall, these results demonstrate that STING peptide-polyanion conjugate LNP treatment can shift the BPPNM tumor from an immunosuppressive to inflamed phenotype, both activating innate immune cells and recruiting T cells. We also observed changes that appeared to be caused by the presence of the LNP vehicle itself independently of its therapeutic cargo, that were observed with STING and Scr Conjugate LNP treatment but not PBS control. LNP vehicle effects included increase in the fraction of myeloid cells (CD1 lb+) in the ascites (Figure 21)and MDSC recruitment to the tumor (Figure 10B). Myeloid recruitment is a previously demonstrated outcome of LNP administration, caused by the immunogenic nature of the LNP ionizable lipid and endosomal escape. The presence of lipid particles has also previously been shown to induce MDSC-like cell populations.

[0156] Combination therapy with PD-1 checkpoint blockade was examined to leverage the T cell inflamed tumor microenvironment and overcome PD-1 expression induced by STING peptide-polyanion conjugate LNP treatment, using five doses of LNP with anti -PD-1 antibody (aPD-1) administered IP 1 day after the first and third dose (Figure 10G). As observed in prior work in this model, aPD-1 treatment resulted in a modest survival improvement compared to PBS control (P = 0.03), increasing median survival to 33 days compared to 27 days for the PBS control (Figure 10H). Combination of STING peptide- polyanion conjugate LNP with anti -PD-1 treatment resulted in significant survival improvement compared to controls (P =.02 vs PBS + aPD-1, P =.003 vs Scr LNP + aPD-1), increasing median survival to 42 days compared to 33 or 29 days for PBS + aPD-1 or Scr + aPD-1 controls, respectively (Figure 10H). Combination treatment resulted in a partial response with BPPNM luminescence shrinking and remaining low over the course of treatment, however, tumor growth resumed after treatment (Figure 101-1 OJ). Combination therapy was also well tolerated, with mice experiencing only small (<5- 10%) and transient (1-2 days) drops in mass after each treatment that were similar to STING peptide-polyanion conjugate monotherapy (Figure 22), These results show that STING peptide-polyanion conjugate LNP treatment complements PD-1 checkpoint blockade to further improve therapeutic outcomes beyond either treatment as a monotherapy.

[0157] Discussion

[0158] This work demonstrates that the design of STING mimicking therapeutics can be greatly simplified. High valency display of ligands with affinity for TBK1 and IRF3 in the cytosol is all that is required to activate IRF3. We were able to achieve this with a 39-amino acid peptide that contains short linear TBK1 and IRF3 interaction motifs, a peptide that is only 10% the length of the 378 amino acid STING protein or only 16% the length of the approximately 241 amino acid S TING cytosolic domain commonly used in other STING mimicking therapeutics. The STING peptide-polyanion conjugate activated IRF3 and NF-KB transcription factors at a similar ratio to a traditional STING agonist.

[0159] We designed our therapeutic molecule from the ground up with nanocarrier integration in mind, using a polyanionic backbone to promote encapsulation by existingnanocarriers that use cationic moieties to encapsulate negatively charged cargos. This greatly accelerated the development of our delivery system compared to prior work delivering a STING mimicking protein therapeutic STINGATM to the cytosol. Using the same HEK293T reporter-based functional delivery screen, STINGATM protein delivery' was only effective with a fraction of tested vehicles and had poor functional delivery efficacy, while peptide-polyanion conjugate delivery in this work resulted in strong functional delivery’ with all vehicles tested. We only needed to perform a single round of LNP optimization to develop a formulation of small, monodisperse nanoparticles with high encapsulation efficiency and cytosolic delivery efficiency, that were ultimately effective for in vivo delivery of this novel peptide-polyanion conjugate cargo in vivo. Still, this screen revealed some differences in LNP formulation design rules when delivering peptide-polyanion conjugate compared to mRNA or siRNA, likely because of the high mass of hydrophilic and neutrally charged peptide that must also be encapsulated. Notably, higher than typical proportion of phospholipid and lower cholesterol were required to achieve a formulation that retained a small and stable size as well as high encapsulation efficiency.

[0160] We clearly demonstrated that STING peptide-polyanion conjugate has promise as a therapeutic, initiating an on target immune response in the tumor and improving survival in multiple challenging to treat models of metastatic ovarian cancer. This therapeutic efficacy was achieved with a proof-of-concept for a first-generation peptide-conjugate STING mimic. There is certainly evidence that STING activation in cancer cells is beneficial to strengthen an anti -cancer immune response.

[0161] Conclusion

[0162] We demonstrated that the signaling biology of the activated STING-TBK1-IRF3 complex can be mimicked by display of short, linear protein interaction motifs on multivalent polymer conjugate. This allowed us to therapeutically activate the innate immune transcription factors IRF3 and NF-KB, even in cell lines that are unresponsive to STING agonists. The peptide-polyanion conjugate could easily be loaded in a LNP carrier that allowed for its application as an immunotherapy and demonstrated efficacy in multiple mouse models of metastatic ovarian cancer. This work provides further evidence that polyanion conjugates are a versatile tool to enable the delivery’ of therapeutic cargo into cells.

[0163] Materials and MethodsPoly(L-glutamic acid sodium salt, MW ≈ 45,000 Da) 10% graft alkyne (PLEsoo-g-AKio%, #000-E300-g-AK010) and (poly-L-glutamic acid sodium salt, MW = 45,000 Da) with an alkyne function on the C-terminal (AK-PLE300, #000-AKE300) were purchased from Alamanda Polymers. Molecular weights and alkyne substitution from the lot specific specification sheet were used for all calculations. PLE3oo-g-AKio% Lot #000-E300-g-AK010-101: number average molecular weight (Mn) by NMR = 47,600 Da, number average degree of polymerization by NMR = 312, poly dispersity index (PDI) by GPC = 1.02, alkyne functionalization by NMR = 11%. AK-PLE300 Lot #000-AKE300-101: Mnby NMR = 48,600 Da, DPn by NMR = 320, PDI by GPC = 1.01, alkynyl substitution by NMR = 100%.

[0164] l,2-Dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE), 1,2-distearoyl-sn-glycero-3 -phosphocholine (DSPC), l,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC), cholesterol (ovine), and 6-((2-hexyldecanoyl)oxy)-N-(6-((2-hexyldecanoyl)oxy)hexyl)-N -(4-hydroxybutyl)hexan-l-aminium (ALC-0315) were purchased from Avanti Research, 8-[(2-hydroxyethyl)[6-oxo-6-(undecyloxy)hexyl]amino]-octanoic acid, 1 -octylnonyl ester (SM- 102) and l,2-dimyristoyl-rac-glycero-3 -methoxypoly ethylene gly col-2000 (DMG-PEG2000) were purchased from Cayman Chemical Company. Dilinoleylmethyl-4-dimethylaminobutyrate (MC3) and 2'3’-c-di-AM(PS)2 (Rp, Rp) disodium salt (ADU-S100 disodium salt) were purchased from MedChemExpress. MRT67307 was purchased from InvivoGen. Sulfo-Cyanine5 azide was purchased from Lumiprobe. Copper(II) sulfate pentahydrate, (+)-sodium L-ascorbate, aminoguanidine hydrochloride, glycerol, 3M sodium acetate solution (pH 5.2), Tween 20, Triton X-100, 0.4% Trypan Blue solution, 0.01% poly-I- lysine solution, bovine serum albumin (BSA), Collagenase from Clostridium histolyticum Type V, DNase I, and Hyaluronidase Type I-S were purchased from Sigma- Aldrich. Sodium chloride and absolute (200 proof) ethanol were purchased from Fisher Scientific. Tris-hydroxypropyltriazolylmethylamine (THPTA) was purchased from Vector Laboratories. Tris-HCl was purchased from G-Biosciences. IM sodium acetate solution (pH 4.5 ) and 16% Formaldehyde (w / v), methanol-free was purchased from Thermo Scientific. Opti-MEM, Fetal bovine serum (FBS), 5000 U / mL penicillin and 5000 pg / mL streptomycin solution, 0.25% Trypsin-EDTA, TrypLE Express Enzyme (lx), and ACK Lysis Buffer were purchased from Gibco. Dulbecco's Modified Eagle's Medium (DMEM), Roswell Park Memorial Institute (RPMI) 1640, and lx Phosphate Buffered Saline (lx PBS) was purchased from Coming. DMEM / F12 1:1 Media was purchased from Cytiva. 0.5 M ethylenediaminetetraacetic acid (EDTA) (pH 8.0) was purchased from Growcells. Type I Ultrapure water generated with aMilli-Q IQ 7000 Ultrapure Lab Water System equipped with a Biopak polisher (Millipore) was used for all experiments.

[0165] Conjugate Synthesis and Characterization

[0166] Peptide Design and Synthesis

[0167] The STING C-terminal tail peptide was based on mouse STING(340-378) (Uniprot Q3TBT3): VTMNAPMTSV APPPSVLSQE PRLLISGMDQ PLPLRTDLI (SEQ ID NO: 1). The scramble (Ser) control peptide was randomly rearranged: LLRNRPVTSG ESMPVPISQA PADLLLPPLQ MMPVSTDTI (SEQ ID NO: 14). An azidolysine residue was included the N-temiinus of both peptides. Peptides were custom synthesized and characterized by GenScript Biotech. Peptide molecular weight was consistent with the theoretical molecule weight of 4341.13 Da by electrospray ionization (ESI) mass spectrometry (MS). Purity was >98% by HPLC. Endotoxin levels were <10 EU / mg by Limulus Amebocyte Lysate (LAL) assay.

[0168] Fluorescent labeling of poly(L-glutamate) graft alkyne

[0169] Poly(L-glutamate)-graft-alkyne was modified with sulfo-cyanine5 azide using copper catalyzed azide-alkyne cycloaddition. The reaction was performed in water by adding sulfo-cyanine5 azide and poly(L-glutamate)-graft-alkyne to a mixture with a final concentration of 0.1 mM Poly(L-glutamate)-graft-alkyne, 10 mM copper (II) sulfate, 100 mM sodium ascorbate, and 10 mM THPTA. Sulfo-cyanine5 azide was added at 2 molar equivalents to Poly(L-glutamate)-graft-alkyne for all experiments except for confocal microscopy, where a highly-labeled batch was prepared with 10 molar equivalents of sulfo-cyanine 5 azide. The reaction was carried out at room temperature (RT) protected from light for 1 hour and then at 4°C overnight. Tire reaction mixture was dialyzed (3.5 kDaMWCO Slide-A-Lyzer™ Dialysis Cassette) against 5 mM EDTA, followed by 50 mM sodium chloride, followed by water and then lyophilized (Labconco FreeZone Freeze Dryer, <-85cC and <0,3 mbar) to generate a blue solid.

[0170] The removal of unreacted fluorophore was confirmed with thin layer chromatography (TLC) using silica gel 60 F254 (Sigma) as the stationary phase and methanol as the mobile phase. Unreacted sulfo-cyanine5 azide migrated near the solvent front, while the purified sulfo-cyanine5 azide-labeled poly(L-glutamate)-graft-alkyne did not migrate and showed no signs of unreacted dye. Fluorescent tag incorporation was quantified by comparing sulfo-cyanine5 absorbance at 646 nm (save = 271,000 M~lenr’, reported by vendor) to poly(L-glutamate)-graft-alkyne absorbance at 205 nm (εPLE300-g-AK10%:::1,020,000 M−1cm−1measured for stock solution), correcting for dye absorbance at 205 nm using the correction factor (CF205 = 0.14, measured for stock solution), using a NanoDropTMOneCultraviolet- visible (UV - vis) spectrophotometer (Thermo Fisher).

[0171] Multivalent peptide poly(L-glutamate) conjugate synthesis

[0172] Sulfo-cyanine5-labeled poly(L-glutamate)-graft-alkyne was modified with azidolysine containing peptides using copper catalyzed azide -alkyne cycloaddition. The reaction was performed in water by adding azidolysine containing peptide and poly(L-glutamate)-graft-alkyne to a mixture with a final concentration of 1 mM alkyne, 1.2 mM azide, 10 mM copper (II) sulfate, 100 mM sodium ascorbate, 10 mM THPTA, and 100 mM aminoguanidine. The reaction was carried out at RT protected from light for 1 hour and then at 4°C overnight. The conjugate was purified by repeated centrifugal filtration (10 kDa Amicon Ultra Centrifugal FilterTM) at 1500xg, washing first with 5 mM EDTA, followed by 50 mM sodium chloride, and then water to reach a total dilution factor of >1,000,000.

[0173] Conjugate solution in water was sterilized with a 0.2 μm cellulose acetate filter and stored at -80°C until use.

[0174] Conjugate was quantified by measuring sulfo-cyanine5 absorbance of the labeled poly(L-glutamic acid) backbone at 646 nm (εdye= 271,000 M−1cm−1) using a NanoDrop™ Onecultraviolet-visible (UV-vis) spectrophotometer. Complete reaction was confirmed by measuring the disappearance of detectable alkynes using a 3-azido-7-hydroxycoumarin-based alkyne quantification kit (ProteinMods™) following manufacturer directions. NMR (Bruker AVANCE™, 500 MHz1H, D2O) was used to confirm the disappearance of the poly(L-glutamate)-graft-alkyne alkyne-adjacent methylene δ 3.95 (-CH2-C≡CH) protons. Successful reaction was also confirmed by the appearance a triazole δ 7.89 proton in the conjugate. Removal of unreacted excess peptide was confirmed by size-exclusion chromatography (Zorbax™ GF-250 column (Agilent™) with 1 x PBS mobile phase). Endotoxin levels were quantified with HEK-Blue mTLR4 reporter cell assay (InvivoGen) following manufacturer instructions, with all conjugates having <2 EU / mg.

[0175] Monovalent peptide poly(L-glutamate) conjugate synthesis

[0176] End-functionalized alkyne-poly(L-glutamate) w as modified with azidolysine containing peptides using copper catalyzed azide -alkyne cycloaddition. The reaction was performed in water by adding azidolysine containing peptide and alkyne-poly(L-glutamate)to a mixture with a final concentration of 0.1 mM alkyne. 0.12 mM azide, 10 mM copper (II) sulfate, 100 mM sodium ascorbate, 10 mM THPTA, and 100 mM aminoguanidine. The reaction was carried out at RT protected from light for 1 hour and then at 4°C overnight. The conjugate was purified by repeated centrifugal filtration (3 kDa Amicon™ Ultra Centrifugal Filter) at 12,000×g, washing first with 5 mM EDTA, followed by 50 mM sodium chloride, and then water to reach a total dilution factor of >1,000,000. Conjugate solution in water was sterilized with a 0.2 μm cellulose acetate filter and stored at -80°C until use. Conjugate was quantified by measuring peptide backbone absorbance at 205 nm (ε = 1,000,000 M−1cm−1, estimated based on amino acid composition86) using a NanoDrop™ Onecultraviolet-visible (UV-vis) spectrophotometer. NMR (Bruker AVANCE™, 500 MHz1H, D2O) was used to confirm the disappearance of the alkyne-poly(L-glutamate) alkyne -adjacent methylene δ 4.24 (-CH2-C≡CH) protons.

[0177] Circular Dichroism

[0178] Samples of azidolysine-mSTING(340-378) peptide (25 pg / mL), poly(L-glutamate)-graft-alkyne polymer (50 pg / mL), and a multivalent poly(L-glutamate)-graft-mSTING(340-378) conjugate (50 pg / mL) were prepared in 1× PBS. Circular dichroism (CD) spectra were obtained with a Jasco™ J-1500 CD Spectropolarimeter in the Biophysical Instrumentation Facility at MIT. Samples (250 pL) were loaded into a 1 mm quartz cuvette and analyzed at ambient temperature. Spectra were collected from 190 and 250 nm with a scan speed of 50 nm / min, a digital integration time of 4 s, a data pitch of 0.5 nm, and a bandwidth of 1 nm. Reported spectra are the average of 3 measurements. To process data, signal from a lx PBS blank was subtracted and then raw millidegrees (mdeg) of rotation (m°) were converted to mean residual ellipticity (MRE) using the equation MRE = m° × M / (10 × L × C × R), where M is the molar mass of the construct in g / mol, L is the pathlength of the cuvette in cm, C is the concentration of the construct in mg / mL, and R is the average number of amino acid residues in each construct.

[0179] LNP Formation and Characterization

[0180] LNP Formation

[0181] DOPC was dried from a chloroform stock under nitrogen and then dissolved in 100% ethanol. ALC-0305, MC3, DOPE, DSPC, Cholesterol, and DMG-PEG2000 were directly dissolved in 100% ethanol.LNPs were formed by bulk mixing using a previously described method. LNPs were prepared with varying lipids components and compositions as described in Table 1. Peptide¬ polyanion conjugate was diluted into 25 mM sodium acetate at pH 4.5 in a glass scintillation vial charged with a polytetrafluoroethylene (PTFE) stir bar. To 4 volumes of peptide-polyanion conjugate under magnetic stirring (700 rpm), 1 volume of lipid mixture in 100% ethanol was pipetted in rapidly. The solution was allowed to mix for 10 s, then rested without stirring for 5 min. Magnetic stirring was resumed (700 rpm), the solution was diluted with 5 volumes of water and allowed to mix for 10 s. LNPs were purified and concentrated by repeated centrifugal filtration (100 kDa Amicon™ Ultra Centrifugal Filter) at 2,650xg, washing with water. LNPs were diluted to reach a final concentration of 1× PBS, LNP solutions W'ere stored at 4 °C until use.

[0182] Table 1: Compositions of LNP formulations examined

[0183] Formulation Total lipid Ionizable Phospholipid Cholesterol PEG lipid to cargo lipid (mol %) (mol %) (moi %) (mass ratio) (mol %)

[0184] ALC-0315 16.8 ALC-0315 DOPE Cholesterol DMG- (50.0) (10.0) (38.5) PEG2000

[0185] (1.5) MC3 18.2 DLin-MC3- DOPE Cholesterol DMG- DMA (10.0) (38.5) PEG2000 (50.0) (1.5) SM-102 17.3 SM-102 DOPE Cholesterol DMG- (50.0) (10.0) (38.5) PEG2000

[0186] (1.5) Al 20.0 ALC-0315 DOPE Cholesterol DMG- (35.0) (25.0) (38.5) PEG2000

[0187] (1.5) A2 20.0 ALC-0315 DOPE Cholesterol DMG- (50.0) (10.0) (38.5) PEG2000

[0188] (1.5)A3 20.0 ALC-0315 DOPE Cholesterol DMG- (43.8) (31.3) (23.5) PEG2000

[0189] (1.5) A4 20.0 ALC-0315 DOPE Cholesterol DMG- (62.5) (12.5) (23.5) PEG2000

[0190] (1.5) A5 20.0 ALC-0315 DOPE Cholesterol DMG- (52.8) (16.5) (29.2) PEG2000

[0191] (1.5) Bl 20.0 ALC-0315 DSPC Cholesterol DMG- (35.0) (25.0) (38.5) PEG2000

[0192] (1-5) B2 20.0 ALC-0315 DSPC Cholesterol DMG- (50.0) (10.0) (38.5) PEG2000

[0193] (1-5) B3 20.0 ALC-0315 DSPC Cholesterol DMG- (43.8) (31.3) (23.5) PEG2000

[0194] (1-5) B4 20.0 ALC-0315 DSPC Cholesterol DMG- (62.5) (12.5) (23.5) PEG2000

[0195] (1.5) B5 20.0 ALC-0315 DSPC Cholesterol DMG- (52.8) (16.5) (29.2) PEG2000

[0196] (1-5) Cl 20.0 ALC-0315 DOPC Cholesterol DMG- (35.0) (25.0) (38.5) PEG2000

[0197] (1.5) C2 20.0 ALC-0315 DOPC Cholesterol DMG- (50.0) (10.0) (38.5) PEG2000

[0198] (1.5)C3 20.0 ALC-0315 DOPC Cholesterol DMG- (43.8) (31.3) (23.5) PEG2000

[0199] (1.5) C4 20.0 ALC-0315 DOPC Cholesterol DMG- (62.5) (12.5) (23.5) PEG2000

[0200] (1.5) C5 20.0 ALC-0315 DOPC Cholesterol DMG- (52.8) (16.5) (29.2) PEG2000

[0201] (1.5)

[0202] Dynamic Light Scattering and Zeta Potential Measurement

[0203] The Z -average diameter and polydispersity index (PDI) of the nanoparticles were characterized by dynamic light scattering (DLS) and the zeta potential by laser doppler anemometry’, using a MalvernTMZetasizer Pro Red (Malvern Panalytical). The measurements were performed at 25°C with a red laser (wavelength = 633 nm) and a detection angle of 173°. Each data point reported is the mean of technical replicates for an independently assembled nanoparticle formulation.

[0204] Conjugate Quantification in LNP

[0205] Peptide-polyanion conjugate concentration in purified LNP samples was quantified by measuring the fluorescence intensity of the Cyanine5-labeled polyanion backbone compared to a peptide-polyanion conjugate standard at a known concentration. LNPs were lysed by¬ adding 1 volume of 4% (v / v) Triton X-100 in water to 1 volume of LNP or standard sample, followed by 10 min incubation at 37°C with orbital shaking. Fluorescence intensity was read with an Infinite™ MPlex Plate Reader (Tecan), with excitation at 640 nm and emission at 680 nm.

[0206] Encapsulation Efficiency Assay

[0207] Peptide-polyanion conjugate encapsulation in LNPs was measured using native polyacrylamide gel electrophoresis (PAGE), where soluble conjugate can migrate freely into the gel but encapsulated conjugate remains trapped in the well. Each LNP sample wasdivided into two containers to allow for direct comparison of intact to lysed LNPs. To 1 volume of LNP sample, 1 volume of water (intact condition) or 1 volume of 2% (v / v) Triton X-100 (lysed condition) was added. Samples were incubated at 37°C for 10 min with orbital shaking. Each sample was then mixed with 2 additional volumes of Native PAGE Sample Buffer (62.5 mM Tris-HCl, pH 6.8, 40% (v / v) glycerol), mixed thoroughly by pipette, and loaded onto a gel (Mini-PROTEAN TGX™ Precast Protein gel, Bio-Rad). Native PAGE was run in 25 mM Tris, 192 mM glycine, pH 8.3 buffer at 100 V for 1 h using a Mini-PROTEAN Tetra Cell™ (Bio-Rad) and a PowerPac™ HC High-Current Power Supply (Bio-Rad). Gels were imaged using the Cyanine5 method to detect Cyanine5 -labeled polyanion backbone using a ChemiDoc™ MP imaging system (Bio-Rad). Image Lab software (Bio-Rad) was used to quantify the Cyanine5 fluorescence signal from free peptide-polyanion conjugate in each lane using local background subtraction. Encapsulation efficiency (EE%) was calculated by comparing fluorescence signal of free conjugate in lysed LNPs (Flysed) to intact LNPs (Fintact) using the following formula:

[0208] EE% = x 100%

[0209]

[0210] Fintact

[0211] Effective pKa Assay

[0212] Solutions of 20 mM sodium phosphate, 20 mM ammonium acetate, 25 mM citrate, and 150 mM sodium chloride were titrated to pH values of 2.0, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, 8.0, 8.5, 9.5, 10.0, and 11.0. 2-(p-Toluidinonaphthalene-6-sulfonic acid) (TNS) was dissolved in DMSO to a working concentration of 300 pM. In a black 96-well plate, LNPs containing ALC-0315 were incubated in the buffer solutions described above. TNS was added to each well such that the final concentration of LNP ionizable lipid was 20 pM and TNS was 6 pM at a total volume of 100 pL. The plate was covered and placed on an orbital shaker at 300 rpm for 1 min, then fluorescence intensity was read with an Infinite MPlex Plate Reader, with excitation at 322 nm and emission at 431 nm. The pKa was determined by fitting a four-parameter logistic curve to the data using GraphPad Prism.

[0213] Transmission Electron Microscopy

[0214] For cryo-transmission electron microscopy (Cryo-TEM), 3 μL of the nanoparticle sample (exchanged into water using 100 kDa MWCO Amicon™ centrifugal filter at 2650 g) was applied to copper grids coated with a continuous carbon film. The grids were pre-treated with oxygen plasma using a Solarus 950 Gatan ™ Advanced Plasma System. Excess sampleon the grid was gently blotted using the Gatan™ Cryo Plunge III, followed by rapid plunging into liquid ethane to vitrify the sample. The grid was then mounted on a Gatan 626 single tilt cryo-holder, which was subsequently inserted into the TEM column. Both the specimen and the holder tip were maintained at cryogenic temperatures with liquid nitrogen to ensure preservation throughout the transfer and imaging process. For both negative stain TEM and Cryo-TEM, imaging was performed on a JEOL 2100 FEG microscope using a minimum dose method to reduce electron beam damage to the sample. The microscope was operated at 200 kV, with magnifications between 10,000× and 60,000× to evaluate particle size and distribution. All images were captured using a Gatan™ 2k x 2k UltraScan CCD camera.

[0215] In Vitro Experiments

[0216] Cell Lines

[0217] HEK293T pGL4.45 cells are a reporter derivative of HEK293T (ATCC CRL-3216) generated previously in our lab. Briefly, HEK293T were stably transfected with the pGL4.45[luc2P / ISRE / Hygro] vector (Promega) to allow for luciferase induction in response to IRF3 activation. THPl-Dual (thpd-nfis) and HEK-Blue mTLR4 (hkb-mtlr4) reporter cells were purchased from InvivoGen. KURAMOCHI cells were a gift from the laboratory’ of Ronny Drapkin (University of Pennsylvania). A2780 cells were a gift from the laboratory of Stephen Howell (UC San Diego). BPPNM and KPCA. C cells were a gift from the laboratory’ of Robert Weinberg (MIT). BBPNM cells stably expressing luciferase, a gift from the laboratory of Stefani Spranger (MIT), were generated by lentiviral transduction of a plasmid containing the sequence of firefly luciferase cloned into the backbone vector pLV-EFla-IRES-Blast (Addgene™ plasmid #85133) and selected using 15 pg / mL blasticidin. KPCA. C cells stably expressing mCherry-SIY and luciferase, a gift from the laboratory of Stefani Spranger (MIT), were generated by lentiviral transduction of a plasmid containing the sequence mCherry-SIY -P2A-Luciferase (mCherry with the SIY antigen fused to its C-terminus, followed by the “self-cleaving” peptide P2A, then firefly luciferase) was cloned into the backbone vector pLV-Efla-IRES-Puro (Addgene plasmid #85132) and selected using 2.5 pg / mL puromycin.

[0218] Cell Maintenance

[0219] HEK293T pGL4.45 were cultured in DMEM supplemented with 10% FBS, 50 U / mL penicillin, and 50 pg / mL streptomycin (P / S). HEK-Blue mTLR4 cells were cultured in DMEM supplemented with 10% heat inactivated (56°C for 30 min) FBS and P / S. A2780cells were cultured in RPMI 1640 supplemented with 10% FBS and P / S. THPl-Dual were cultured in RPMI 1640 supplemented with 10% heat inactivated (56°C for 30 min) FBS and P / S. KURAMOCHI were cultured in DMEM / F12 1:1 supplemented with 10% FBS and P / S. BPPNM and KPCA. C cells w'ere cultured in DMEM supplemented with 4% heat inactivated FBS (Millipore Sigma #F4135), 1% insulin-transferrin-selenium (Thermo Fisher #41400045), and P / S, Adherent cells were dissociated by tapping the flask (HEK293T pGL4.45 and HEK-Blue mTLR4) or by incubation in 0.25% Trypsin-EDTA ( KURAMOCHI, A2780, BPPNM, and KPCA. C). All cells were incubated at 37°C at 100% humidity and 5% CO2 atmosphere for maintenance and during all assay incubation periods unless otherwise specified. Cells were used at less than 20 passages. Live cells were counted using a Trypan Blue stain with a Cellometer Auto 1000 (Nexcelom). Cells lines were tested upon receipt and routinely during culture for mycoplasma at the Koch Institute’s Preclinical Modeling Core Facility (MIT) using the MycoAlert PLUS™ Assay (Lonza Biosciences). All cell lines tested negative. All cell lines were tested using short tandem repeat (STR) profiling service at American Type Culture Collection, with the exception of commercial reporter cell lines (HEK-Blue mTLR4, THPl-Dual) which were used as received from vendors.

[0220] Transfection Complex Preparation

[0221] TransIT-X2™ Dynamic Delivery System (Minis Bio) is a polymer-based delivery system that was used for routine transfection of peptide-polyanion conjugate following manufacturer directions. TransIT-X2 was warmed to RT and vortexed gently, then added to the cargo in Opti-MEM at a ratio of 1 μL TransIT-X2 per μg of cargo, w ith thorough pipet mixing. The mixture was incubated at RT for 15 to 30 min to allow for complex formation and then added to cells. Transfection grade, linear, MW 25,000, polyethylenimine (PEI) (Polysciences™) is a polymeric transfection reagent. PEI was dissolved in 25 mM sodium acetate at pH 5.2 at a concentration to allow 2 μg of PEI per μg of cargo. The cargo was also diluted in 25 mM sodium acetate at pH 5.2, and then mixed with the PEI solution at a 1: 1 volume ratio. The mixture was incubated at RT for 15 min, then added to cells. The PBAE Poly 2 is a polymeric transfection reagent, using the same batch of polymer synthesized and characterized in previous work. Poly 2 was dissolved in 25 mM sodium acetate at pH 5.2 at a concentration to allow 8 μg of Poly2 per μg of cargo. The cargo was also diluted in 25 mM sodium acetate at pH 5.2, and then mixed with the Poly2 solution at a 1: 1 volume ratio. The mixture was incubated at RT for 15 min, then added to cells.Reporter Cell Assays

[0222] Assays with HEK293T pGL4.45 reporter cells were performed as described previously. Briefly, 3 × 105cells / mL were added to a 96-well plate in 100 μL of DMEM with 10% FBS and P / S, then allowed to incubate for 24 h. To each well, 20 pL of treatment specified in each figure caption was added, then allowed to incubate for 24 h. Luciferase was quantified using the Firefly Luciferase Assay Kit (Biotium), The supernatant was aspirated, 25 pL of Lysis Buffer was added to each well, and then the plate was allowed to incubate for 15 min with orbital shaking. 25 pL of each sample was transferred to an opaque, white, 96- well plate, then 50 μL of Assay Buffer with 0.2 mg / mL D-luciferin was added, and luminescence was measured immediately using an Infinite MPlexTMPlate Reader (Tecan), using an integration time of 1000 ms. Data was analyzed after subtracting the luminescence of a blank (Lysis and Assay Buffer with 0.2 mg / mL d-luciferin in a well with no cells), and reported relative to a buffer-treated control.

[0223] Assays with THP1-Dual reporter cells were performed based on manufacturer instructions. Briefly, 5.6 × 105cell / mL THP1-Dual were added to a 96-well plate in 180 μL RPMI with 10% heat inactivated FBS and P / S cells along with 20 μL of treatment specified in each figure caption, then allowed to incubate for 24 h. To detect IRF3 controlled Lucia luciferase reporter signal, 10 μL of treated cell supernatant was transferred to an opaque, white, 96-well plate, then 50 μL of QUANTI-Luc™ Reagent (InvivoGen) was added and luminescence was measured immediately using an Infinite MPlex™ Plate Reader, using an integration time of 100 ms. Data was analyzed after subtracting the luminescence of a blank (QUANTI-Luc ™ Reagent plus fresh media), and reported relative to a buffer treated control. To detect NF-KB controlled secreted embryonic alkaline phosphatase (SLAP) reporter signal, 20 μL of treated cell supernatant was added to 180 μL QUANTI-Blue Reagent™ (InvivoGen) in a clear 96-well plate. Tire plate was incubated at 37°C with orbital shaking until a visible color difference was observed. Absorbance at 640 nm was measured using an using an Infinite MPlex ™ Plate Reader. Data was analy zed after subtracting the luminescence of a blank (QUANTI-Blue Reagent™ plus fresh media).

[0224] Western Blot

[0225] HEK293T pGL4.45 reporter cells were added to a 6-well plate at 3 x 10’’ cells / mL in 3 ml of DMEM with 10% FBS and P / S, then allowed to incubate for 24 h. Cells were treated with 600 μL of the treatment specified in each figure caption, then allowed to incubate for 6 h. Cells were washed twice with ice-cold 1 x PBS and then lysed in RIP A lysis buffer(Thermo Fisher) supplemented with 1 x Halt™ Protease and Phosphatase Inhibitor Cocktail (Thermo Fisher) and 5 mM EDTA (Thermo Fisher) for 10 min at 4 °C with gentle orbital shaking. Cell lysates were centrifuged at 15,000×g, 4 °C for 15 min, the supernatant was recovered, and the protein concentration was quantified by DC protein assay. The sample was diluted to reach a final composition of 1 x Laemmli Sample Buffer (Bio-Rad) and 25 mM DTT and then heated to 95 °C for 5 min. All lysates were stored at -80 °C until analysis.

[0226] Samples were loaded at 25 μg of total protein. Samples were run on a 4-20% Mini-PROTEAN TGX™ Precast protein gel alongside Precision Plus Protein Dual Color Standards (Bio-Rad) and transferred to a 0.2 μm nitrocellulose membrane (Bio-Rad) using a Trans-Blot™ SD Semi-Dry Transfer Cell (Bio-Rad). The membrane was blocked with 5% (w / v) Nonfat Dry Milk (Cell Signaling) in tris-buffered saline and 0.1% (v / v) Tween 20 (TBST) for 1 h at RT with gentle orbital shaking. The membrane was incubated with primary antibodies anti-TBKl (1:1000, Cell Signaling #3504), antiphospho-TBKl (Serl72) (1:1000, Cell Signaling #5483), anti-IRF3 (1:1000, Cell Signaling #4302), or antiphospho-IRF3 (Ser396) (1: 1000, Cell Signaling #29047) in 5% (w / v) BSA in TBST overnight at 4 °C with gentle orbital shaking. The membrane was incubated with the secondary antibody anti-rabbit IgG, HRP (1:2000, Cell Signaling #7074) in 5% (w / v) Nonfat Dry Milk in TBST for 1 h at RT with gentle orbital shaking. The membrane was incubated with anti-β-actin, HRP (1:25,000, Abeam™ ab49900) in 5% (w / v) Nonfat Dry Milk in TBST for 1 h at RT with gentle orbital shaking. Western Lightning Ultra (Revvity) chemiluminescent substrate was added to the membrane, and it was imaged using a ChemiDoc™ MP imaging system (Bio¬ Rad).

[0227] Immunocytochemistry

[0228] HEK293T pGL4.45 reporter cells were added to a Lab-Tek II Chamber Slides™ (precoated in 0.01% (w / v) poly-l-lysine solution for 5 min) at 3 × 105cells / mL in 400 pL of DMEM with 10% FBS and P / S, then allowed to incubate for 24 h. Cells were treated with 50 μL of the treatment specified in each figure caption, then allowed to incubate for 6 h. Cells were fixed with 4% formaldehyde in 1 x PBS for 10 min, permeabilized by 0.1% (v / v) Triton X-100 for 10 min, blocked with 1% (w / v) BSA and 0.05% (v / v) Tween 20 in lx PBS for 30 min, and stained with a primary rabbit anti-TBK1 antibody (1:66 dilution, Abeam ab235253) at 4 °C overnight. Cells were then stained with secondary Alexa Fluor 488-conjugated donkey anti-rabbit IgG (1:400 dilution, Thermo Fisher A-32790) and simultaneously with Dylight 554 Phalloidin (1:400 dilution, Cell Signaling 13054S) at RT for 1 h. ProLong GoldAntifade Mountant with DNA Stain DAPI (Thermo Fisher P36941) was used to mount a coverslip and was allowed to cure for 24 h at RT. Imaging was performed on a FV4000 confocal laser scanning microscope (Evident) equipped with 405, 488, 561, and 640 nm lasers in the Koch Institute’s Microscopy Core Facility (MIT). Images were acquired with a 100× silicone oil immersion objective. All images were acquired at consistent laser settings. A single Z-slice containing features of interest is displayed. Images were pseudocolored using ImageJ™ software.

[0229] Enzyme-Linked Immunosorbent Assays (ELISA)

[0230] DuoSetTMELISA kits (R&D Systems) for human CXCL10 (#DY266), human IFN-β (#DY814), mouse CXCL10 (#DY466), mouse IFN-β (#DY8234), mouse IL-6 (#DY406), mouse TNF-a (#DY410), and mouse IFN-y (#DY485) were used with 1-Step TMB™ ELISA Substrate Solution (Thermo Scientific) following manufacturer instructions. Cytokine concentrations were calculated using a four-parameter logistic curve-fit of a standard curve, with the bottom parameter constrained at the mean absorbance of the blank (GraphPad PRISM™ software), The limit of detection (LOD) was calculated by adding 3 standard deviations to the mean absorbance of the blank and converting this to a concentration using the standard curve. Conditions where analyte was below the LOD were labeled as not detected (ND).

[0231] In vitro Cytokine Induction

[0232] KURAMOCHI or A2780 cells at 5.6×104cell / mL were added to a 96-well plate in 180 pL of the media used to culture each cell line (described above), then allowed to incubate for 24 h. Cells were treated with 20 pL of the treatment specified in each figure caption, then allowed to incubate for 24 h. Plates were centrifuged at 500xg, then the supernatant was sampled and stored at -20°C until analysis. Human CXCL10 and IFN-P levels were measured by ELISA, as described above.

[0233] mRNA Sequencing

[0234] KURAMOCHI cells were added to a 6-well plate at 1 x 105cells / mL in 3 mL of DMEM / F12 1: 1 with 10% FBS and P / S, then allowed to incubate for 24 h. Cells were treated with 600 μL of the treatment specified in each figure caption, then allowed to incubate for 6 h. Cells were washed 1× PBS and then detached using 100 pL TrypLE Express Enzyme. Cells were diluted with 200 pL DMEM / F12 1:1 with 10% FBS and P / S and then countedwith a Cellometer Auto 1000. The cell suspension was centrifuged at 500*g for 5 min and then the pellet was resuspended in DNA / RNA Shield™ (Zymo Research Corporation) at 2 × 105cells / mL to preserve RNA.

[0235] Preserved cells in DNA / RNA Shield were shipped to Plasmidsaurus for mRNA-sequencing. Briefly, total RNA was extracted, converted to complementary DNA via reverse transcription and second-strand synthesis, followed by tagmentation, library indexing, and amplification. Libraries were sequencing using an Illumina NovaSeq™ instrument. A 3’ end counting approach was used to capture differential gene expression. Sequencing data was processed by Plasmidsaurus. Quality was assessed using FastQC v0.12.1. Reads were quality filtered using fastp v0.24.0 with poly-X tail trimming, 3' quality-based tail trimming, a minimum Phred quality score of 15, and a minimum length of 50 bp. Quality-filtered reads were aligned to the reference genome using STAR aligner v2.7.11 with non-canonical splice junction removal and output of unmapped reads, followed by coordinate sorting using samtools v 1.22.1. PCR and optical duplicates were removed using UMI-based deduplication with UMIcollapse v 1.1.0, resulting in a range of 5.1-15.1 million deduplicated reads per replicate. Gene-level expression quantification was performed using featureCounts (subread package v2.1.1) with strand-specific counting, multi-mapping read fractional assignment, exons and three prime UTR as the feature identifiers, and grouped by gene id. Differential expression was done with edgeR v4.0.16 after filtering for low-expressed genes with edgeR::filterByExprwith default values. Functional enrichment was performed using gene set enrichment analysis with gseapy v0.12 using the MSigDB Hallmark gene set, normalized enrichment score and familywise-error rate adjusted P values are reported.

[0236] Animal Experiments

[0237] Animal Care and Use

[0238] All animal experiments were approved by the MIT Committee on Animal Care (CAC, protocol number 2404000660) and were conducted under the oversight of the Division of Comparative Medicine (DCM). Female C57BL / 6 mice were purchased from Jackson Laboratory and housed at Koch Institute for Integrative Cancer Research at MIT animal facility in cages of no more than five animals with controlled temperature (25 °C), 12 h lightdark cycles, and free access to food and water.

[0239] Tumor ModelsFor the BPPNM model, 3 × 106BPPNM cells (expressing luciferase) were suspended in 200 pL lx PBS and the injected IP in 8-9 week-old female C57BL / 6 mice. For the KPCA. C model, 1 x 106KPCA. C cells (expressing mCherry-SIY and luciferase) were suspended in 200 pL lx PBS and the injected IP in 8-9 week-old female C57BL / 6 mice. Tumor burden was monitored using 2D IVIS bioluminescence imaging. Mice were injected IP with 200 pL of 15 mg / kg D-luciferin, sodium salt (GoldBio, #LUCNA), and luminescence was measured 15 minutes later using the IVIS SpectrumTMIn Vivo Imaging System (Perkin Elmer). Images were analyzed with Living Image software. Prior to each experiment, tumor burden was measured by IVIS bioluminescence and mice were divided into groups with comparable distributions of total tumor bioluminescence signal. Mice with no detectable tumor or poorly dispersed tumor found only at the injection site were excluded from subsequent experiments.

[0240] Biodistribution

[0241] The BPPNM model was initiated as described above, with mice moved onto AIN- 93M™ Maintenance Purified Diet (TestDiet 1810541) at least one week prior to biodistribution measurements. Mice were administered a 20 pg dose of STING peptide-polyanion conjugate either w'ithout a carrier or loaded in the C3 LNP formulation IP in 200 pL of 1 x PBS 15 days after tumor inoculation. 200 pL of 1 x PBS was administered on the same schedule as a control. Mice were injected IP with 200 pL of 15 mg / kg D-luciferin, sodium salt, luminescence was measured 15 minutes later using the IVIS SpectrumTMIn Vivo Imaging System, and then mice were euthanized at 4 hours after initial treatment.

[0242] Immediately following, necropsies were performed to harvest upper genital tract (UGT), omentum, liver, kidneys, spleen, intestines, heart and lungs from each mouse. Organs were immersed in RPMI 1640 media in 24-well plates and placed on ice until imaging. Organs were imaged using bioluminescence to track BPPNM tumors. STING peptide polyanion conjugate was imaged using fluorescence of its sulfo-cyanine5 -label using a 640 excitation and 680 emission filter. Living Image Software w'as used to measure the bioluminescent total flux in [p / s], bioluminescent average radiance in [p / s / cm2 / sr], fluorescent total radiant efficiency in [p / s] / [μW / cm2], fluorescence average radiant efficiency in [p / s / cm2 / sr] / [μW / cm2], and area in [cm2] of each organ. To correct for background fluorescence, average fluorescence of the organ in the vehicle 1 x PBS treated mice w as subtracted from each measurement. The fraction of recovered fluorescence was computed as the total radiantefficiency of the organ divided by sum of the total radiant efficiency for all organs recovered from an individual mouse.

[0243] Acute Toxicity

[0244] The BPPNM model was initiated as described above. Mice were administered a 20 pg dose of STING or Scr peptide polyanion conjugate loaded in the C3 LNP formulation IP in 200 μL of 1 x PBS 14 days after tumor inoculation. 200 pL of 1× PBS was administered on the same schedule as a control. 24 h after treatment, blood was collected via submandibular bleeding into EDTA Microtubes (Sarstedt) for complete blood counts or Serum Gel CAT Microtubes (Sarstedt) for serum chemistry. Analysis was performed by the Division of Comparative Medicine Comparative Pathology Laboratory (MIT), with complete blood counts analyzed using a HemaVet 950FS™ (Drew Scientific) and serum chemistry analyzed using a custom IDEXX panel,

[0245] Pharmacokinetics and pharmacodynamics

[0246] A 20 pg dose of STING peptide polyanion conjugate loaded in the C3 LNP formulation was administered IP in 200 pL of 1× PBS to 11 week old female C57BL / 6 mice. Blood was collected from 3 randomly selected mice out of 5 treated mice at 0 h (immediately), 1 h, 3 h, 6 h, 10 h, 24 h, and 50 h after treatment. At each time point, approximately 50 μL, blood was collected via submandibular bleeding into Serum Gel CAT Microtubes (Sarstedt). Serum Gel CAT Microtubes were centrifuged at 10,000xg for 5 min, serum was collected from the topmost fraction and stored at -20 °C until analysis.

[0247] STING peptide polyanion conjugate was quantified in serum by measuring fluorescence of its sulfo-cyanine5 label. 5 pL of serum was diluted with 20 pL of 2% (v / v) Triton X-100 in water, followed by 10 min incubation at 37°C with orbital shaking. Cyanine5 fluorescence intensity was read in black 384-well flat-bottom plates on an Infinite MPlex™ Plate Reader (Tecan), with excitation at 640 nm and emission at 680 nm. STING peptide polyanion conjugate concentrations were calculated using a four-parameter logistic curve-fit of a standard curve, prepared by diluting LNP stock in serum from an untreated mouse (GraphPad PRISM software). The limit of detection (LOD) was calculated by adding 3 standard deviations to the mean absorbance of the blank and converting this to a concentration using the standard curve. Conditions where analyte was below the LOD were labeled as not detected (ND). Mouse CXCL10 levels in serum were measured by ELISA, as described above.In vivo cytokine induction

[0248] The BPPNM model was initiated as described above. Mice were administered a 20 pg dose of STING or Scr peptide polyanion conjugate loaded in the C3 LNP formulation IP in 200 pL of lx PBS 14 days after tumor inoculation. 6 h after treatment, blood was collected via submandibular bleeding into Serum Gel CAT Microtubes (Sarstedt). Serum Gel CAT Microtubes were centrifuged at 10,000 xg for 5 min, serum was collected from the topmost fraction, and supplemented with 1 x Halt Protease and Phosphatase Inhibitor Cocktail (Thermo Fisher). Immediately after bleeding, mice were euthanized and ascites was collected via peritoneal lavage with 1 mL 1 xPBS. Ascites were centrifuged at 4,000xg for 15 mm at 4 °C, the supernatant was collected and supplemented with I x Halt™ Protease and Phosphatase Inhibitor Cocktail. Tumor nodules located on the omentum were collected and washed with ice-cold 1 x PBS. Approximately 60 mg of tumor (range 53-63 mg) was added to 1 mL RIPA lysis buffer (Thermo Fisher) and supplemented with 1 x HaltTMProtease and Phosphatase Inhibitor Cocktail. Tumor was diced with scissors, homogenized with a syringe plunger, and incubated at 4 °C for 15 minutes with occasional brief sonication. Tissue homogenate was centrifuged at 15,000xg for 15 min at 4 °C, and the supernatant was collected. All samples were stored at -80 °C until analysis. Mouse CXCL10, IFN-β, IL-6, TNF-a, and IFN-y levels were measured by ELISA, as described above. Cytokine levels in ascites and tumor were normalized to total protein levels in each sample, measured by DC (detergent-compatible) protein assay (Bio-Rad) following manufacturer’s instructions.

[0249] Therapeutic Efficacy Studies

[0250] The BPPNM model was initiated as described above. In a monotherapy study, mice were administered a 20 pg dose of STING or Scr peptide polyanion conjugate loaded in the C3 LNP formulation IP in 200 pL of lx PBS on days 10, 13, and 16 after tumor inoculation.

[0251] 200 pL of 1 x PBS was administered on the same schedule as a control. In a combination therapy study, mice were administered a 20 μg dose of STING or Scr peptide polyanion conjugate loaded in the C3 LNP formulation IP in 200 μL of 1× PBS on days 10, 13, 16, 19, and 22 after tumor inoculation. 200 pL of 1 x PBS was administered on the same schedule as a control. In groups specified in figure captions, mice were also administered a 100 pg dose of anti-mouse PD-1 antibody (Bio X Cell #BE0273 InVivoMAb™, Clone 29F.1A12) in 200 pL of 1 x PBS on days 11 and 17 after tumor inoculation. Weight and tumor bioluminescence were measured throughout the study. Mice were monitored and euthanized when bodycondition score dropped below 2, weight loss exceeded 20%, or poor responsiveness was observed.

[0252] The KPCA.C model was initiated as described above. Mice were administered a 20 pg dose of STING or Scr peptide polyanion conjugate loaded in the C3 LNP formulation IP in 200 uL of lx PBS on days 7, 10, 13, and 16 after tumor inoculation. 200 pL of lx PBS was administered on the same schedule as a control. Weight and tumor bioluminescence were measured throughout the study. Mice were monitored and euthanized when body condition score dropped below 2, weight loss exceeded 20%, or poor responsiveness was observed.

[0253] Immunophenotyping

[0254] The BPPNM model was initiated as described above. Mice were administered a 20 μg dose of STING or Scr peptide polyanion conjugate loaded in the C3 LNP formulation IP in 200 μL of 1× PBS on days 10, 13, 16, and 19 after tumor inoculation. 200 pL of lx PBS was administered on the same schedule as a control. On day 20 after tumor inoculation (12 h after the final dose), mice were euthanized to collect organs.

[0255] Tumor, localized on the omentum, was collected and stored in RPMI media on ice. To the gentleMACSTMC Tubes (Miltenyi #130-096-334), 2.5 mL digestion mixture and tumor were added. Tumor was minced using forceps and kept on ice. Tumors were digested using a gentleMACS™ Octo Dissociator (Miltenyi), m_imp_tumor_02 program. Tubes were incubated under shaking (150 rpm) at 37°C for 30 min. Tubes were returned to ice and digested using m_impTumor_03 program. Digested tumors were diluted 1:1 with FACS buffer ( 1% BSA and 2 mM EDTA) to final concentration of 1 mM EDTA and 0.5% BSA as EDTA quenches the enzymatic reaction. Dissociated tumors were filtered through a 70 pm filter (Miltenyi #130-110-916) into new 1 mL tube to obtain a single cell suspension and filter was rinsed with 2 mL FACS buffer to complete the transfer. Cells were collected by spinning 15 mL tubes at 300xg for 5 min at 4°C. Supernatant was decanted and 2 mL ACK lysis buffer was added, tube gently vortexed to resuspend cells and incubated for 2 min to lyse red blood cells. To quench the ACK, 5 mL FACS buffer was added and the tube spun at 300xg for 5 min. This process was repeated until a white cell pellet was obtained.

[0256] Supernatant was removed and cells resuspended in 200 pL of FACS buffer, counted, and left on ice until flow staining.

[0257] Ascites was collected using a 3 mL syringe attached with an 18G needle (Air-Tite #ML3181). Cold RPMI (2.5 mL) was injected into the peritoneal cavity, the peritoneal cavity was gently massaged and then the fluid was collected in the syringe and transferred to 15 mLtube and kept on ice. Tubes were centrifuged at 300g for 5 min at 4°C, supernatant decanted, and cells resuspended in 1 mL of enzyme digestion mixture by gentle vortex. For the digestion, cells were then incubated at 37°C for 30 min on an orbital shaker incubator under agitation (150 rpm). Following enzymatic digestion, 1 mL of FACS buffer was added to quench the enzymatic reaction. Cells were centrifuged at 300g for 5 min at 4°C. Supernatant was decanted and 2 ml, ACK lysis buffer was added, tube gently vortexed to resuspend cells and incubated for 2 min to lyse red blood cells. To quench the ACK, 5 mL FACS buffer was added and the tube spun at 300xg for 5 min. Cells were resuspended in 200 pL FACS buffer by vigorous pipetting and passed through a 70 pm filter into a fresh tube to obtain a single cell suspension. Cells were counted and left on ice until flow staining.

[0258] Based on an average cell count, 106cells of each tumor sample and all of the ascites cells (typically 5xl05-106cells) were transferred to a 96-well V bottom plate for staining. Plate was centrifuged (500xg for 3 min at 4°C for this and subsequent steps), dumped, and cells resuspended in 50 pL Zombie Live / Dead stain (1:100) diluted in PBS and incubated for 10 min at room temperature. Cells were washed with 150 pL FACS buffer then Fc receptors were blocked with anti-CD16 / 32 (Biolegend #101339) at 1 pg per well in 50 pL for 15 min at 4°C, Cells were washed with 150 pL FACS buffer then stained with antibodies in 50 pl,, diluted in FACS buffer and Brilliant Stain Buffer (BD #BDB563794) at specified dilutions in Table 2, and incubated for 30 min at 4°C. Cells were washed with 150 pL FACS buffer then fixed in 200 pL of 2% formaldehyde in PBS for 20 min at 4°C, Cells were centrifuged and resuspended in FACS buffer and stored at 4°C until analyzed. Compensation was calculated with UltraComp eBeads™ Plus Compensation Beads and single stained cells where appropriate. Flow cytometry was performed using a BD Symphony A3™ equipped with HTS (BD Biosciences, New Jersey), and data analyzed in FlowJo VI 0.

[0259] Table 2: Antibody panels used for flow cytometry experiments

[0260] Marker Fluorophore Clone Product Info Dilution

[0261] Lymphoid Panel

[0262] CD45 BUV395 30-F11 BD Biosciences #564279 40 CD11b PE-Dazzle 594 MI / 70 BioLegend #101256 400 NK1.1 APC-Cy7 PK136 BioLegend #108724 40 CD3e PE 17A2 BioLegend #100206 20CD 19 BV711 6D5 BioLegend #115555 80 CD4 BUV563 GK1.5 BD Biosciences #612923 40 CD8 BV605 53-6.7 BioLegend #100744 40 PD-1 PE-Cy7 clone 29F.1A12 BioLegend #135216 60 Zombie Violet BioLegend #423113 100 Myeloid Panel

[0263] CD45 BUV395 30-F11 BD Biosciences #564279 40 MHC-II (I-A / I-E) BUV805 M5 / 114.15.2 BD Biosciences #748844 60 CDllc BV421 N418 BioLegend #117330 20 F4 / 80 BV785 BM8 BioLegend #123141 20 CDllb PE-Dazzle 594 MI / 70 BioLegend #101256 400 Ly6C / Ly6G (Gr-1) BUV661 RB6-8C5 BD BioSciences #741470 400 CD206 BV605 C068C2 BioLegend #141721 20 CD86 PE-Cy7 GL-1 BioLegend #105014 20 CD 19 APC-Cy7 6D5 BioLegend #115530 40 CD3e APC-Cy7 17A2 BioLegend #100222 40 NK1.1 APC-Cy7 PK136 BioLegend #108724 40 Zombie UV BioLegend #423107 100

[0264] Data Analysis and Statistics

[0265] All statistical tests were performed using GraphPad PRISM 10. Comparisons between two groups were performed via unpaired t-tests. Comparisons between multiple groups were performed with one-way ANOVA followed by Tukey's post-hoc test or two-way ANOVA followed by Sidak's post-hoc test. All tests were two-tailed. For survival studies, comparisons were performed using the log(rank) (Mantel-Cox) test. For mRNA sequencing data analysis, results were considered significant at P < 0.05. When an assay determined that an analyte was below the limit of detection (LOD), the result was highlighted as not detected (ND). If any replicates were ND, summary statistics were not computed and statistical comparisons were not performed. Data is presented as mean ± standard deviation (SD) if presented on a linear scale or geometric mean ± geometric SD if presented on a logarithmic scale.

[0266] References(1) Ishikawa, H.; Barber, G. N. STING Is an Endoplasmic Reticulum Adaptor That Facilitates Innate Immune Signalling. Nature 2008, 455 (7213), 674-678.

[0267] (2) Woo, S.-R.; Fuertes, M. B.; Corrales, L.; Spranger, S.; Furdyna, M. J.; Leung, M. Y. K.; Duggan, R.; Wang, Y.; Barber, G. N.; Fitzgerald, K. A.; Alegre, M.-L.; Gajewski, T. F. STING-Dependent Cytosolic DNA Sensing Mediates Innate Immune Recognition of Immunogenic Tumors. Immunity’ 2014, 41 (5), 830-842.

[0268] (3) Corrales, L.; Glickman, L. H.; McWhirter, S. M.; Kanne, D. B.; Sivick, K. E.;

[0269] Katibah, G. E.; Woo, S.-R.; Lemmens, E.; Banda, T.; Leong, J. J.; Metchette, K.; Dubensky, T. W.; Gajewski, T. F. Direct Activation of STING in the Tumor Microenvironment Leads to Potent and Systemic Tumor Regression and Immunity. Cell Reports 2015, 11 (7), 1018— 1030.

[0270] (4) Sivick, K. E.; Desbien, A. L.; Glickman, L. H.; Reiner, G. L.; Corrales, L.; Surh, N. H.; Hudson, T. E.; Vu, U. T.; Francica, B. J.; Banda, T.; Katibah, G. E.; Kanne, D. B.; Leong, J. J.; Metchette, K.; Bruml, J. R.; Ndubaku, C. O.; McKenna, J. M.; Feng, Y.; Zheng, L.; Bender, S. L.; Cho, C. Y.; Leong, M. L.; Elsas, A. van; Dubensky, T. W.; McWhirter, S. M. Magnitude of Therapeutic STING Activation Determines CD8+ T Cell-Mediated Anti-Tumor Immunity. Cell Reports 2018, 25 (11), 3074-3085.e5.

[0271] (5) Fu, J.; Kanne, D. B.; Leong, M.; Glickman, L. H.; McWhirter, S. M.; Lemmens, E.; Mechette, K.; Leong, J. J.; Lauer, P.; Liu, W.; Sivick, K. E.; Zeng, Q.; Soares, K. C.; Zheng, L.; Portnoy, D. A.; Woodward, J. J.; Pardoll, D. M.; Dubensky, T. W.; Kim, Y. STING Agonist Formulated Cancer Vaccines Can Cure Established Tumors Resistant to PD-1 Blockade. Science Translational Medicine 2015, 7 (283), 283ra52-283ra52.

[0272] (6) Francica, B. J.; Ghasemzadeh, A.; Desbien, A. L.; Theodros, D.; Sivick, K. E.;

[0273] Reiner, G. L.; Hix Glickman, L.; Marciscano, A. E.; Sharabi, A. B.; Leong, M. L.;

[0274] McWhirter, S. M.; Dubensky, T. W., Jr; Pardoll, D. M.; Drake, C. G. TNFa and Radioresistant Stromal Cells Are Essential for Therapeutic Efficacy of Cyclic Dinucleotide STING Agonists in Nonimmunogenic Tumors. Cancer Immunology Research 2018, 6 (4), 422-433.

[0275] (7) Demaria, O.; De Gassart, A.; Coso, S.; Gestermann, N.; Di Domizio, J.; Flatz, L.; Gaide, O.; Michielin, O.; Hwu, P.; Petrova, T. V.; Martinon, F.; Modlin, R. L.; Speiser, D. E.; Gilliet, M. STING Activation of Tumor Endothelial Cells Initiates Spontaneous and Therapeutic Antitumor Immunity. Proceedings of the National Academy of Sciences 2015, 112 (50), 15408-15413.(8) Meric-Bernstam, F.; Sweis, R. F.; Hodi, F. S.; Messersmith, W. A.; Andtbacka, R. H. I.; Ingham, M.; Lewis, N.; Chen, X.; Pelletier, M.; Chen, X.; Wu, J.; Dubensky, T. W.;

[0276] McWhirter, S. M.; Muller, T.; Nair, N.; Luke, J. J. Phase I Dose-Escalation Trial of MIW815 (ADU-S100), an Intratumoral STING Agonist, in Patients with Advanced / Metastatic Solid Tumors or Lymphomas. Clinical Cancer Research 2022, 28 (4), 677-688.

[0277] (9) Meric-Bernstam, F.; Sweis, R F Kasper, S,; Hamid, O.; Bhatia, S.; Dummer, R.; Stradella, A.; Long, G. V.; Spreafico, A.; Shimizu, T.; Steeghs, N.; Luke, J. J.; McWhirter, S. M.; Muller, T.; Nair, N.; Lewis, N.; Chen, X.; Bean, A.; Kattenhorn, L.; Pelletier, M.;

[0278] Sandhu, S. Combination of the STING Agonist MIW815 (ADU-S100) and PD-1 Inhibitor Spartalizumab in Advanced / Metastatic Solid Tumors or Lymphomas: An Open-Label, Multicenter, Phase lb Study. Clinical Cancer Research 2023, 29 (1), 110-121.

[0279] (10) Harrington, K. J.; Brody, J.; Ingham, M.; Strauss, J.; Cemerski, S.; Wang, M.; Tse, A.; Khilnani, A.; Marabelle, A.; Golan, T. LBA15 - Preliminary Results of the First-in-Human (FIH) Study of MK-1454, an Agonist of Stimulator of Interferon Genes (STING), as Monotherapy or in Combination with Pembrolizumab (Pembro) in Patients with Advanced Solid Tumors or Lymphomas. Annals of Oncology 2018, 29, viii712.

[0280] (11) Konno, H.; Yamauchi, S.; Berglund, A.; Putney, R. M.; Mule, J. J.; Barber, G. N. Suppression of STING Signaling through Epigenetic Silencing and Missense Mutation Impedes DNA Damage Mediated Cytokine Production. Oncogene 2018, 37 (15), 2037-2051. (12) Xia, T.; Konno, H.; Barber, G. N. Recurrent Loss of STING Signaling in Melanoma Correlates with Susceptibility to Viral Oncolysis. Cancer Research 2016, 76 (22), 6747- 6759.

[0281] (13) de Queiroz, N. M. G. P.; Xia, T.; Konno, H.; Barber, G. N. Ovarian Cancer Cells Commonly Exhibit Defective STING Signaling Which Affects Sensitivity to Viral Oncolysis. Molecular Cancer Research 2019, 17 (4), 974-986.

[0282] (14) Xia, T.; Konno, H.; Ahn, J.; Barber, G. N. Deregulation of STING Signaling in Colorectal Carcinoma Constrains DNA Damage Responses and Correlates With Tumorigenesis. Cell Reports 2016, 14 (2), 282-297.

[0283] (15) Falahat, R.; Berglund, A.; Perez-Villarroel, P.; Putney, R. M.; Hamaidi, I.; Kim, S.; Pilon-Thomas, S.; Barber, G. N.; Mule, J. J. Epigenetic State Determines the in Vivo Efficacy of STING Agonist Therapy. Nat Commun 2023, 14 (1), 173.

[0284] (16) He, Y.; Hong, C.; Yan, E. Z.; Fletcher, S. J.; Zhu, G.; Yang, M.; Li, Y.; Sun, X.; Irvine, D. J.; Li, J.; Hammond, P. T. Self-Assembled cGAMP-STINGTMSignaling Complex as a Bioinspired Platform for cGAMP Delivery. Sci Adv 2020, 6 (24), eaba7589.(17) He, Y.; Hong, C.: Huang, S.; Kaskow, J. A.; Covarrubias, G.; Pires, I. S.; Sacane, J. C; Hammond, P. T; Belcher, A. M. STING Protein-Based In Situ Vaccine Synergizes CD4+ T, CD8+ T, and NK Cells for Tumor Eradication. Advanced Healthcare Materials 2023, 12 (24), 2300688.

[0285] (18) He, Y.; Hong, C.; Fletcher, S. J.: Berger, A. G.: Sun, X.; Yang, M.; Huang, S.;

[0286] Belcher, A. M.; Irvine, D. J.; Li, J.; Hammond, P. T, Peptide-Based Cancer Vaccine Delivery via the STINGATM-cGAMP Complex. Advanced Healthcare Materials 2022, 11 (15), 2200905.

[0287] (19) Sun, X.; Ni, Y.; He, Y.; Yang, M.; Tani, T.; Kitajima, S.; Barbie, D. A.; Li, J.

[0288] Engineering the Immune Adaptor Protein STING as a Functional Carrier, Advanced Therapeutics 2021, 4 (8), 2100066.

[0289] (20) Zhao, B.; Du, F.; Xu, P.; Shu, C.; Sankaran, B.; Bell, S. L.; Liu, M.; Lei, Y.; Gao, X.; Fu, X.; Zhu, F.; Liu, Y Laganowsky, A.; Zheng, X.; Ji, J.-Y.; West, A. P.; Watson, R, O; Li, P. A Conserved PLPLRT / SD Motif of STING Mediates the Recruitment and Activation of TBK1. Nature 2019, 569 (7758), 718-722. https: / / doi.org / 10.1038 / s41586-019-1228-x. (21) Tanaka, Y.; Chen, Z J. STING Specifies IRF3 Phosphorylation by TBK1 in the Cytosolic DNA Signaling Pathway. Science Signaling 2012, 5 (214), ra20-ra20.

[0290] (22) Boussif, O.; Lezoualc’h, F; Zanta, M. A.; Mergny, M. D.; Scherman, D.; Demeneix, B.; Behr, J. P. A Versatile Vector for Gene and Oligonucleotide Transfer into Cells in Culture and in Vivo: Polyethylenimine. Proceedings of the National Academy of Sciences 1995, 92 (16), 7297-7301.

[0291] (23) Berger, A. G.; DeLorenzo, C.; Vo, C.; Kaskow, J. A.; Nabar, N.; Hammond, P. T. Poly(P-Aminoester) Physicochemical Properties Govern the Delivery of siRNA from Electrostatically Assembled Coatings, Biomacromolecules 2024, 25 (5), 2934-2952.

[0292] (24) Schoenmaker, L.; Witzigmann, D.; Kulkami, J. A.; Verbeke, R.; Kersten, G.; Jiskoot, W.; Crommelin, D. J. A. mRNA-Lipid Nanoparticle COVID-19 Vaccines: Structure and Stability, International Journal of Pharmaceutics 2021, 601, 120586.

[0293] (25) Szebeni, J.; Kiss, B.; Bozo, T.; Turjeman, K.; Levi-Kalisman, Y.; Barenholz, Y.; Kellermayer, M. Insights into the Structure of Comimaty Covid-19 Vaccine: A Theory on Soft, Partially Bilayer-Covered Nanoparticles with Hydrogen Bond-Stabilized mRNA-Lipid Complexes. ACS Nano 2023, 17 (14), 13147–13157.

[0294] (26) Correa, S.; Boehnke, N.; Barberio, A. E.; Deiss-Yehiely, E.; Shi, A.; Oberlton, B.; Smith, S. G.; Zervantonakis, I.; Dreaden, E. C.; Hammond, P. T. Tuning NanoparticleInteractions with Ovarian Cancer through Layer-by-Layer Modification of Surface Chemistry. ACS Nano 2020, 14 (2), 2224-2237.

[0295] (27) Barberio, A. E.; Smith, S. G.; Pires, I. S.; Iyer, S.; Reinhardt, F.; Melo, M. B.; Suh, H.; Weinberg, R. A.; Irvine, D. J.; Hammond, P. T. LAYER-BY-LAYER Interleukin- 12 Nanoparticles Drive a Safe and Effective Response in Ovarian Tumors. Bioengineering & Transla Med 2023, 8 (2), e10453.

Claims

We claim1. A composition, comprising a conjugate, wherein the conjugate comprises(a) a plurality of peptides, wherein the plurality of peptides comprise peptides comprising (i) a binding site for TBK1, and (ii) a binding site for 1RF3; and(b) a surface to which the plurality of peptides are covalently conjugated.

2. The composition of claim 1, wherein the binding site for TBK1 comprises or consists of the amino acid sequence PxPLR (SEQ ID NO: 3), wherein x is any amino acid, and (ii) the binding site for IRF3 comprises or consists of the amino acid sequence pLxIS, where p is a hydrophilic amino acid, and x is any amino acid.

3. The composition of claim 1 or 2, w herein the binding site for TBK1 comprises or consists of the amino acid sequence selected from the group consisting of (i) PLPLR (SEQ ID NO: 5); (ii) (D / 'E)XPXPLR(S / T)D (SEQ ID NO: 6), w'herein X denotes any amino acid; and (iii) PLPLR(T / S)D (SEQ ID NO: 7).

4. The composition of any one of claims 1-3, wherein the plurality of peptides comprises:(i) a first plurality of peptides that comprise a binding site for TBK1; and (ii) a second plurality of peptides that comprise a binding site for IRF3.

5. The composition of claim 4, w herein the first plurality of peptides and the second plurality of peptides are present in the composition in about an equal ratio.

6. Tire composition of any one of claims 1-3, wherein the plurality of peptides comprise or consist of the amino acid sequence selected from (i) RLLISxxxxPLPLRTD (SEQ ID NO: 8), wherein x is any amino acid, and (ii) ELLISxxxxPLPLRTD (SEQ ID NO: 9), wherein x is any amino acid.

7. The composition of claim 1-3 and 6, wherein the plurality of peptides each comprise or consist of the am ino acid sequence selected from (i) RLLISGMDQPLPLRTD (SEQ ID NO: 10), and (ii) ELLISGMEKPLPLRTD (SEQ ID NO: 11).

8. The composition of any one of claims 1-3 and 6-7, wherein the plurality of peptides each comprise or consist of the amino acid sequence selected from SEQ ID NO: 1-2.

9. Tlie composition of any one of claims 1-8, wherein each peptide in the plurality of peptides does not comprise or consists of a full length STING protein.

10. The composition of any one of claims 1-9, wherein each peptide in tire plurality of peptides is between 5-100 amino acids in length, or between 5-75 amino acids in length, or between 5-50 amino acids in length, or between 5-40 amino acids in length, or between 5-30 amino acids in length, or between 5-20 amino acids in length, or between 10-100 amino acids in length, or between 10-75 amino acids in length, or between 10-50 amino acids in length, or between 10-40 amino acids in length, or between 10-30 amino acids in length, or between 10- 20 amino acids in length,11. The composition of any one of claims 1-10, wherein each peptide in the plurality of peptides is covalently conjugated to the surface at its N-terminus.

12. Tire composition of claim 11, wherein each peptide in the plurality of peptides comprises an azido-lysine residue at its N-terminus.

13. The composition of any one of claims 1-12, wherein the surface is selected from the group consisting of a polymer, a nanoparticle, and a protein.

14. Tire composition of any one of claims 1-13, wherein the surface comprises a polymer.

15. Tlie composition of claim 14, wherein the polymer comprises an anionic polymer.

16. The composition of claim 15, -wherein the anionic polymer is selected from the group consisting of a negatively charged poly -amino acid, a negatively charged polysaccharide, polyfcarboxylic acid), poly(sulfate), poly(phosphate), co-polymers thereof, an functionalized versions thereof,17. Tlie composition of any one of claims 15-16, wherein the anionic polymer is selected from the group consisting of poly(L-glutamic acid), poly(D-glutamic acid), polyfL-asparticacid), poly(D-aspartic acid), poly(acrylic acid) (PAA), poly(methacrylic acid) (PMAA), poly(ethylacrylic acid) (PEAA), poly(propylacrylic acid) (PPAA), poly(butylacrylic acid) (PBAA), poly(allylamine hydrochloride)-citraconic anhydride (PAH-Cit), hyaluronic acid, and dextran sulfate, co-polymers thereof, and functionalized versions thereof.

18. The composition of any one of claims 15-17, wherein the anionic polymer compri ses poly (L-glutam ate).

19. The composition of any one of claims 15-18, wherein the anionic polymer comprises a plurality of alkyne residues conjugated to the anionic polymer,20. The composition of claim 19, wherein at least 10%, 25%, 50%, 75%, 90%, or all of the anionic polymer units comprise alkyne residues conjugated to the anionic polymer.

21. The composition of claim 19 or 20, wherein at least some of the alkyne residues are also conjugated to peptides in the plurality of peptides.

22. Tire composition of claim 21, wherein at least 10%, 25%, 50%, 75%, 90%, or all of the alkyne residues are also conjugated to peptides in the plurality of peptides.

23. The composition of any one of claims 1-22, wherein the composition comprises bifunctional linkers between the surface and some or all of the plurality of peptides.

24. The composition of claim 23, wherein the bifunctional linkers comprise polyethyelene glycol (PEG) linkers, including but not limited to azido-PEG4-NHS ester linkers.

25. The composition of any one of claims 15-24, wherein an azido-lysine residue at the N-terminus of each peptide in the plurality of peptides is conjugated to an alkyne residue on the anionic polymer.

26. The composition of any one of claims 1-25, wherein the plurality of peptides comprises at least 3 peptides, or at least 4 peptides, or at least 8 peptides, or at least 25 peptides, or at least 100 peptides.

27. The composition of any one of claims 1-26, wherein the conjugate is contained within a delivery vehicle that enables delivery of the conjugate to the cytosol.

28. Tlie composition of claim 27, wherein the delivery vehicle comprises a delivery vehicle for negatively charged cargo.

29. The composition of claim 27 or 28, wherein the delivery' vehicle comprises a lipid nanoparticle or a cationic polymer nanoparticle.

30. Tire composition of any one of claims 27-29, wherein the delivery' vehicle comprises a cationic polymer nanoparticle.

31. The composition of claim 30, wherein the cationic polymer nanoparticle comprises poly (beta-amino ester), polyethyleneimine (PEI), polyamidoamine (PAMAM), and / or chitosan.

32. The composition of claim 31, wherein the poly (beta-amino ester) comprises the structure selected from the following, wherein n is 3-10033. The composition of claim 30 or 31, wherein the cationic polymer nanoparticle comprises polyethyleneimine (PEI).

34. The composition of claim 27 or 28, wherein the delivery7vehicle comprises a lipid nanoparticle.

35. Tlie composition of claim 34, wherein the lipid nanoparticle comprises an ALC- 0315:cholesterol: 18: 1 (A9-Cis) PE (DOPE): DMG PEG2000 core.

36. The composition of claim 35, wherein the lipid nanoparticle is modified with targeting ligands to allow cell specific deliver}'.

37. The composition of any one of claims 27-36, wherein the polymer is non-covalently adsorbed to the deliver ’ surface vehicle.

38. The composition of any one of claims 14-38, wherein the polymer comprises about 300 monomeric subunits.

39. Tire composition of any one of claims 14-38, wherein the polymer is about a 300-unit poly(L-glutamate) backbone.

40. The composition of any one of claims 35-39, wherein the ALC-0315 cholesterol: 18: 1 (A9-Cis) PC (DOPC): DMG PEG2000 core is present at a molar ratio of about 43.8:23.5:31.3:1.5.

41. The composition of claim 34, wherein the lipid nanoparticle comprises an ALC-0315 cholesterol: 18:1 (A9-Cis) PE (DOPE): DMG PEG2000 core.

42. The composition of claim 34 or 41, wherein the ALC-0315 cholesterol: 18: 1 (A9-Cis) PE (DOPE): DMG PEG2000 core is present at a molar ratio of about 43.8:23.5:31.3:1.5.

43. The composition of claim 34 or 41, wherein the ALC-0315 cholesterol: 18:1 (A9-Cis) PE (DOPE): DMG PEG2000 core is present at a molar ratio of about 50:38.5:10:1.5.

44. The composition of any one of claims 1-43, further comprising a PD-1 inhibitor.

45. The composition of claim 44, wherein the PD-1 inhibitor comprises an anti -PD-1 antibody.

46. A pharmaceutical composition comprising:(a) the composition of any one of claims 1-45; and(b) a pharmaceutically acceptable carrier.

47. A method for treating cancer, comprising administering to a subject in need thereof the composition or pharmaceutical composition of any one of claims 1-38.

48. Tlie method of claim 47, wherein the subject has metastatic melanoma or ovarian cancer.

49. The method of claim 47 or 48, wherein the administering is carried out via intraperitoneal (IP) administration.50, Tire method of any of claims 47-49, wherein in the administering composes administering in at least five doses.

51. The method of any of claims 47-50, wherein the administering is performed weekly for about one year.

52. The method of any of claims 47-50, wherein the administering is performed monthly for about one year.

Images