Nanovesicle hydrogels and uses thereof

Abstract

Supramolecular hydrogels comprising (i) at least one polymer and (ii) extracellular vesicles; processes for production thereof; and methods of use.

Description

NANO VESICLE HYDROGELS AND USES THEREOF

[0001] Throughout this application, various publications are referenced, including referenced in parenthesis. The disclosures of all publications mentioned in this application in their entireties are hereby incorporated by reference into this application in order to provide additional description of the art to which this invention pertains and of the features in the art which can be employed with this invention.CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 844,908 filed July 16, 2025, U.S. Provisional Application No. 63 / 755,622 filed February 7, 2025, and U.S. Provisional Application No. 63 / 710,411 filed October 22, 2024, the contents of all of which are hereby incorporated by reference.BACKGROUND OF THE INVENTION

[0003] Hydrogels are cross-linked networks of macromolecules that have numerous biomedical applications1. These materials can be made from natural or synthetic components2 4, can regulate the localized release of drugs or bioactive compounds'2 10and resemble biological tissues"14, which have made them useful in the fields of drug delivery and tissue engineering. These capabilities are further enhanced by recent developments with supramolecular hydrogels, a subtype formed through noncovalent interactions, that exhibit dynamic and reversible crosslinking2'3 15which makes these materials injectable346 19for minimally invasive drug delivery and tissue engineering20 22.

[0004] Despite these useful properties, supramolecular hydrogels are often comprised of synthetic materials without intrinsic bioactivity22 26. Synthetic hydrogels can be chemically modified to present biological signals18, but current technical capabilities are limited by the number of signaling ligands that can feasibly be introduced, falling far short of the scope of signals displayed from natural tissues. This hinders the functionality of hydrogels in tissue engineering and regenerative medicine applications, where they are used to drive complex processes that depend on numerous signals1. While previous strategies have used natural extracellular matrix-derived hydrogels to achieve elegant bioactivity, the incorporation of EVs offers a distinct mode of modular biofunctionalization by integrating a form of intercellular communication27.

[0005] EVs are nano-sized vesicles (-30-4,000 nm)28secreted by most cell types and carry diverse signaling molecules such as proteins, lipids, nucleic acids, and glycans that modulate the activity of cells that encounter them29. Upon release, they can navigate through the body by either binding to nearby cells, the extracellular matrix, or disseminating through body fluids29,30. Because EVs are so effective at delivering messages, cells use them to communicate with one another across short and long distances31. Unlike synthetic nanoparticles, which generally present fewer than 10 signaling molecules, EVs present hundreds of signals just on their surface, enabling them to convey much more sophisticated messages than what synthetic technology currently achieves31 33.

[0006] There is growing interest in stably integrating extracellular vesicles into biomaterials due to their inherent bioactivity and signaling potential. However, most existing approaches rely on covalent crosslinking, which often requires chemical modification of EVs and harsh conditions to form networks. While covalent systems offer mechanical stability, they often lack injectability and dynamic reversibility, limiting their clinical utility34-37. In contrast, supramolecular hydrogels support minimally invasive injection, conform to complex tissue geometries, and re-solidify in situ. Their reversible crosslinking enables shear-thinning during delivery and self-healing afterward, making them particularly well-suited for delivery of shear-sensitive EVs.BRIEF SUMMARY OF THE INVENTION

[0007] A supramolecular hydrogel comprising (i) at least one polymer and (ii) extracellular vesicles.

[0008] A method of producing a supramolecular hydrogel comprising (i) at least one polymer and (ii) extracellular vesicles, the method comprising admixing at least one polymer and extracellular vesicles under conditions permitting creation of the supramolecular hydrogel.

[0009] A method of isolating EVs from a biological sample or food product containing EVs, comprising subjecting the biological sample or food product containing EVs to tangential flow filtration (TFF) comprising (a) processing the biological sample or food product containing EVs through a hollow fiber with an pore size that removes cell debris so as to produce a permeate, and (b) passing the permeate through a hollow fiber with a 100 kDa cut-off, so as to thereby retain isolated EVs in the retentate while eliminating smaller soluble molecules.

[0010] A method of injecting a supramolecular hydrogel described herein into an animal, comprising injecting an amount of the supramolecular hydrogel into the animal.BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Fig. 1: Isolation and concentration of yogurt EVs for use in supram olecul ar hydrogel synthesis shows uniform size, charge, and preservation of properties during tangential flow filtration and lyophilization, (a) Representative Nanoparticle Tracking Analysis (NTA) size distribution of yogurt EVs and lyophilized yogurt EVs indicates size remains consistent after lyophilization (b) Summary of yogurt EVs and lyophilized yogurt EVs’ size and uniformity measured by NTA and dynamic light scattering (DLS) (c) Zeta potential of yogurt and lyophilized yogurt EVs demonstrates charge is preserved (d) qualitative immunoblot showing that CD81 surface antigen, a canonical marker for EVs, is maintained after lyophilization of yogurt EVs (e) Morphology of yogurt EVs is maintained after lyophilization, as assessed by Transmission Electron Microscopy (TEM) and Cryo Electron Microscopy (CryoEM) imaging. Scale bars: 100 nm. Bars indicate mean and error bars represent standard error of the mean (SEM). Statistical comparisons were made using paired student’s t-tests. See also Figures 11-13.

[0013] Fig. 2: Hydrophobic modification of HPMC plays a significant role in yogurt EV hydrogel formation and mechanical properties, (a) Less yogurt EVs are required for hydrogel formation with HPMCs modified with longer alkyl chains (forHPMC-C16, gelation is additionally defined as increased stiffness compared to polymer without EVs). (b) Increasing the alkyl chain length in HPMC modification leads to stiffer and (c) more solid-like hydrogels. Increasing degree of functionalization (DoF) of HPMC-C16 enhances both (d) hydrogel stiffness and (e) solid-like behavior. Bars indicate mean and error bars represent SEM. All bar graphs report G’ and G” values at 10 rads / s. Statistical comparisons were made using a one-way ANOVA, and the false discovery rate (FDR) was controlled at 5% using the Benjamini, Krieger, and Yekutieli method in panel b (q<0.05*) and an unpaired t-test in panel d (p<0.05*). See also Figures 20-26, 28-32.

[0014] Fig. 3: Polymer concentration and EVs contribute nonlinearly to hydrogel stiffness, (a-b) Increasing polymer (HPMC-C16) wt% in yogurt EV hydrogels increases hydrogel stiffness, and yogurt EVs contribute to hydrogel stiffness regardless of polymer wt%. Polymer and yogurt EV interactions show a nonlinear enhancement of hydrogel mechanical properties, illustrated by both (c) differing patterns in hydrogel stiffness increase (at 10 rad / s) and (d) deviations from polymer-like Rouse behavior upon EV addition (e) hydrodynamic diameter measurements reveal significant increases when yogurt EVs are incubated with hydrophobicallymodified polymers, especially HPMC-C16, consistent with polymer-EV complexation (f) intensity-weighted size distributions show a clear shift toward larger unimodal particle size distributions upon EV mixing with HPMC-C12 and HPMC-C16, supporting the formation of EV-polymer supramolecular assemblies. Bars indicate mean and error bars represent SEM. Statistical comparisons were made using a one-way ANOVA, and the false discovery rate (FDR) was controlled at 5% using the Benjamini, Krieger, and Yekutieli method (q<0.05*, q<0.01**). See also Figure 27.

[0015] Fig. 4: Hydrogel stiffness and stability are tunable by EV number, (a) Frequency sweep of 3 wt% HPMC-C16 alone (left) and yogurt EV hydrogels incorporating 3wt% HPMC with (middle) and without (right) hydrophobic hexadecyl (Cl 6) modification. Materials exhibit solid-like mechanical properties when G’ > G” and Tan8 < 1.0. Hydrogels should be solid-like across the tested frequency range (0.1-100 rad*sec-1) for both (a) 5.33 mol% modified HPMC-C16 and (b) 2.93 mol% modified HPMC-C16 (c) yogurt EV hydrogel stiffness (G' and G" at 10 rad / s) increases with EV number, demonstrating that mechanical properties can be tuned by EV concentration. Statistical comparisons made by simple linear regression (p<0.01*, p<0.001**) (d) Images of (i) a yogurt EV hydrogel being injected through a 26G needle, (ii) resolidifying, and (iii) resisting gravity (e) Yogurt EV hydrogel injectability exhibited by shear rheology measurements showing that viscosity decreases with increasing shear rate (shear thinning) (f) Self-healing properties of yogurt EV hydrogel demonstrated by step-shear rheological measurements: hydrogels can repeatedly regain their original viscosity after exposure to high shear force (g) Strain-controlled oscillatory shear rheology assessing the linear viscoelastic regime and yielding behavior of yogurt EV hydrogels. See also Figure 31.

[0016] Fig. 5: EV hydrogel platform shows versatility and robust gel formation across nanovesicle types, (a) Summary of EV and artificial cell derived vesicle (ACDV) size and uniformity measured by Nanoparticle Tracking Analysis (NTA) and Dynamic Light Scattering (DLS) (b) Nanovesicle surface charge determined by laser-Doppler electrophoresis (c) Representative Western Blot showing the presence of extracellular vesicle marker TSG101 for B16F10 melanoma ACDVs. Frequency sweep, step-shear rheological measurements, amplitude sweep assessing viscoelastic regime, and flow sweep demonstrating shear-thinning behavior of (d) E.coli ACDV hydrogels and (e) Mammalian ACDV hydrogels (f— g) G', G", and (h) tan(8) valuesfor hydrogels crosslinked with yogurt EVs, E. coli ACDVs, and mammalian ACDVs show similar mechanical enhancement compared to polymer-only controls across a wide frequency range. See also Figures 33-34.

[0017] Fig. 6: Yogurt EV hydrogels are remodeled by infiltrating cells and induce blood vessel formation in vivo, (a) Yogurt EV hydrogels were injected subcutaneously into Balb / cJ mice and mice were monitored for a week for signs of adverse effects prior to explantation of the hydrogels to assess angiogenic effects, (b) Visible vascularization in explanted yogurt EV hydrogel one week after administration, (c-d) Histological analysis of all explanted EV hydrogels using H&E and trichrome staining reveal extensive cellular infiltration, ECM deposition, and formation of morphologically advanced vasculature. Scale bars denote 100 pm for leftmost panel and 50 pm for remaining panels. See also Figures 35-38.

[0018] Fig. 7: Yogurt EV hydrogels promote both neovascularization and vascular maturation through endothelial and cytoskeletal remodeling. Immunofluorescence staining reveals evidence of newly forming vasculature alongside maturing blood vessels, as well as cytoskeletal reorganization within the hydrogel. Representative images of explanted hydrogels stained for F-actin (cyan) highlight endothelial cell migration and dynamic cytoskeletal remodeling, while CD31 (yellow) marks mature endothelial cells and CD34 (magenta) identifies progenitor and immature endothelial cells, indicating ongoing vascular development. Scale bars denote 100 pm for top panels and 200 pm for bottom panels. See also Figures 39-46.

[0019] Fig. 8: F-actin, CD31, and CD34 co-localization indicate dynamic cytoskeletal remodeling and blood vessel formation, (a) Representative images of co-localization of angiogenic markers (F-actin (cyan) and mature (CD31, yellow) and progenitor (CD34, magenta)) in hydrogel explants. Scale bars denote 200 pm for all panels, (b) Mander’s colocalization coefficient. Statistical comparisons were made using a Kruskal-Wallis test with Dunn correction for multiple comparisons (n = 5 explant regions, p<0.05*). See also Figures 39-46.

[0020] Fig. 9: Yogurt EV hydrogels cultivate a distinct immune cell niche compared to synthetic liposomal controls, (a) Representative images of explanted EV and liposomal nanoparticle (LNP) hydrogels at day 7 show visible differences in gel size and vascularization. Scale bar: 5 mm. (b) Total live cell counts were comparable between groups, (c) Representative flow cytometry plot showing gating of CD45+immune cells, (d) EV hydrogels exhibitedsignificantly greater infiltration of CD45+immune cells, while liposomal hydrogels contained more CD45“ cells, (e-g) EV gels showed significantly increased recruitment of myeloid cells, including neutrophils (CD1 lb+Ly6G+), dendritic cells (CDllc+), and macrophages (CD1 lb+F4 / 80+). (h) Immunofluorescence staining confirmed dense macrophage (CD68+) and dendritic cell (CDllc+) infiltration in EV gels. Scale bar: 100 pm. (i-k) EV gels also showed significantly increased infiltration of adaptive immune cells, including total T cells (CD3+) and CD4+and CD8+T cell subsets. (1) Stacked bar plot showing average abundance of immune cell populations within each hydrogel group, (m) Immunofluorescence imaging revealed colocalization of CD3+T cells and FOXP3+regulatory T cells within EV gels, supporting recruitment of anti-inflammatory immune populations. Scale bar: 20 pm. Statistical comparisons were performed using two-tailed unpaired t-tests. Bars indicate mean; error bars represent SEM (p<0.05*, p<0.01**). See also Figures 47-52.

[0021] Fig. 10: Overview of the design and optimization of extracellular vesicle (EV)-crosslinked injectable hydrogels. EVs are incorporated with alkyl-modified cellulose-based polymers to form hydrogels with tunable mechanical properties. The design space is optimized by adjusting polymer alkyl length, degree of functionalization, polymer concentration, as well as EV number and source (yogurt, bacteria, cancer cells), demonstrating broad versatility of the system. In vivo bioactivity validation demonstrates that EV hydrogels promote immune infiltration, remodeling, and angiogenesis, highlighting their potential utility for regenerative medicine and biomedical applications.

[0022] Fig. 11: Mean protein concentration of yogurt EVs (pre / post-lyophilization). The mean protein concentration found through BCA assay indicated a ca. 10-fold concentration after lyophilization. Statistical comparisons were made using a paired t-test (p<0.05).

[0023] Fig. 12: Cryo Electron Microscopy (CryoEM) of yogurt EVs. Cryo Electron Microscopy (CryoEM) imaging of unprocessed yogurt EVs allowing visualization of protein corona.

[0024] Fig. 13: Cryo Electron Microscopy (CryoEM) of lyophilized yogurt EVs. Cryo Electron Microscopy (CryoEM) imaging of lyophilized yogurt EVs allowing visualization of protein corona.

[0025] Fig. 14: Linear regression of yogurt EV number tuning tand. Yogurt EV hydrogel (HPMC-C160.52 mmol / g, 3 wt%) solid-like behavior (tan5 at 10 rad / s) is tuned by the number of yogurt EVs (nanoparticle crosslinkers) in the hydrogel (Statistical comparisons made by simple linear regression).

[0026] Fig. 15: HPMC-C12 (0.52 mmol / g) NMR characterization. HPMC-C12 (052 mmol / g) characterization and modification validation. 'H-NMR (500 MHz, DMSO-d6, 298 K) of (a) dodecyl isocyanate, (b) unmodified hypromellose (HPMC), and (c) HPMC-C12, which shows the carbon chain (dodecyl) terminal methyl group at 0.86 ppm. The mol% modification of HPMC was calculated by taking the ratio of the carbon chain-related peak integration at 0.86 ppm to the HPMC-related peak1. We determined a 6.33 mol% modification of ETPMC-C12.

[0027] Fig. 16: HPMC-C14 (0.52 mmol / g) NMR characterization. HPMC-C14 (052 mmol / g) characterization and modification validation. 'H-NMR (500 MHz, DMSO-d6, 298 K) of (a) tetradecyl isocyanate, (b) unmodified hypromellose (HPMC), and (c) HPMC-C14, which shows the carbon chain (tetradecyl) terminal methyl group at 0.86 ppm. The mol% modification of HPMC was calculated by taking the ratio of the carbon chain-related peak integration at 0.86 ppm to the HPMC-related peak1. We determined a 9.33 mol% modification of HPMC-C14.

[0028] Fig. 17: HPMC-C14 (low DoF, 0.4 mmol / g) NMR characterization. HPMC-C14 (low DoF, 0.4 mmol / g) characterization and modification validation. 'H-NMR (500 MHz, DMSO-d6, 298 K) of (a) tetradecyl isocyanate, (b) unmodified hypromellose (HPMC), and (c) HPMC-C14, which shows the carbon chain (tetradecyl) terminal methyl group at 0.86 ppm. The mol% modification of HPMC was calculated by taking the ratio of the carbon chain-related peak integration at 0.86 ppm to the HPMC-related peak'. We determined a 7.33 mol% modification of HPMC-C14 low DoF.

[0029] Fig. 18: HPMC-C16 (0.52 mmol / g) NMR characterization. HPMC-C16 (052 mmol / g) characterization and modification validation. 'H-NMR (500 MHz, DMSO-d6, 298 K) of (a) hexadecyl isocyanate, (b) unmodified hypromellose (HPMC), and (c) HPMC-C16, which shows the carbon chain (hexadecyl) terminal methyl group at 0.86 ppm. The mol% modification of HPMC was calculated by taking the ratio of the carbon chain-related peak integration at 0.86 ppm to the HPMC-related peak'. We determined a 6.33 mol% modification of HPMC-C16.

[0030] Fig. 19: HPMC-C16 (low DoF, 0.32 mmol / g) NMR characterization. HPMC-C 16 (low DoF, 0.32 mmol / g) characterization and modification validation. 'H-NMR (500 MHz, DMSO-d6, 298 K) of (a) hexadecyl isocyanate, (b) unmodified hypromellose (HPMC), and (c) HPMC-C16, which shows the carbon chain (hexadecyl) terminal methyl group at 0.86 ppm. The mol% modification of HPMC was calculated by taking the ratio of the carbon chain-related peak integration at 0.86 ppm to the HPMC-related peak1. We determined a 2.93 mol% modification of HPMC-C16 low DoF.

[0031] Fig. 20: Yogurt EV and HPMC-C12 hydrogel rheology (1012EVs). Yogurt EV and HPMC-C12 hydrogel rheology (1012EVs) reveals the formation of a solid-like hydrogel across the studied frequency range, but defective self-healing behaviors, (a) Frequency sweep of 3 wt% HPMC-C12 and 1012yogurt EV hydrogel, with no visible crossover between G’ and G” in the tested frequency range (0.1-100 rad*sec-1). (b) Shear thinning abilities of yogurt EV and HPMC-C12 hydrogel (viscosity decreases with increasing shear rate), (c) Hydrogel step-shear rheological measurements to characterize self-healing (d) Strain-controlled oscillatory shear rheology (amplitude sweep assessing the linear viscoelastic regime) of yogurt EV hydrogels (e) Frequency sweep of 3 wt% HPMC-C12 alone (left) and yogurt EV hydrogels incorporating 3wt% HPMC with (middle) and without (right) hydrophobic dodecyl (Cl 2) modification.

[0032] Fig. 21: Yogurt EV and HPMC-C12 hydrogel rheology (1011EVs). Yogurt EV and HPMC-C 12 hydrogel rheology (1011EVs) reveals the formation of a material a cross-over between G’ and G” in the studied frequency range (low-frequency), but optimal self-healing behaviors, (a) Frequency sweep of 3 wt% HPMC-C12 and 1011yogurt EV hydrogel shows crossover between G’ and G’ ’ in the tested frequency range (0.1-100 rad*sec-1). (b) Shear thinning abilities of yogurt EV and HPMC-C12 hydrogel (viscosity decreases with increasing shear rate), (c) Hydrogel stepshear rheological measurements to characterize self-healing (d) Strain-controlled oscillatory shear rheology (amplitude sweep assessing the linear viscoelastic regime) of yogurt EV hydrogels (e) Frequency sweep of 3 wt% HPMC-C 12 alone (left) and yogurt EV hydrogels incorporating 3wt% HPMC with (middle) and without (right) hydrophobic dodecyl (C12) modification.

[0033] Fig. 22: Yogurt EV and HPMC-C14 hydrogel rheology. Yogurt EV and HPMC-C14 hydrogel rheology shows solid-like behavior across the studied frequency range as well as robust self-healing behavior, (a) Frequency sweep of 3 wt% HPMC-C14 and 1011yogurt EVhydrogel, (b) Shear thinning abilities of yogurt EV and HPMC-C14 hydrogel (viscosity decreases with increasing shear rate), (c) Hydrogel step-shear rheological measurements to characterize self-healing (d) Strain-controlled oscillatory shear rheology (amplitude sweep assessing the linear viscoelastic regime) of yogurt EV hydrogels (e) Frequency sweep of 3 wt% HPMC-C14 alone (left) and yogurt EV hydrogels incorporating 3wt% HPMC with (middle) and without (right) hydrophobic dodecyl (Cl 4) modification.

[0034] Fig. 23: Yogurt EV hydrogels: polymer type vs. self-healing behavior. Yogurt-derived EV hydrogels formed with HPMC -C 12, HPMC -C 14, or HPMC -Cl 6 were subjected to alternating low shear (1 s ’, white background) and high shear (10 s ', yellow background) to evaluate the kinetics of network disruption and recovery.

[0035] Fig. 24: Self-healing behavior vs. yogurt EV number (HPMC-C16). HPMC-C16 forms self-healing gels at both high and low EV loadings (Hydrogels are formulated with 3wt% HPMC-C16 and 1010 / 1012 yogurt EVs ). %Recovery was calculated by comparing the final viscosity values at the end of the third and second low-shear recovery phases. Specifically, the viscosity at the end of the third recovery phase was divided by the value at the end of the second recovery phase to account for differences in overshoot and best reflect sustained self-healing behavior.

[0036] Fig. 25: Transient viscosity overshoot during step-shear recovery is enhanced by EV addition in C16 hydrogels, (a) Initial recovery curve from step-shear testing of 3 wt% HPMC-C16 hydrogels with increasing EV concentrations, showing a transient peak in viscosity immediately following the cessation of shear. Dashed line illustrates the steady-state plateau used to quantify overshoot magnitude (b) Quantification of overshoot magnitude (Pa*s) across groups shows increase with EV number.

[0037] Fig. 26: Yogurt EV and 2 wt% HPMC-C16 hydrogel rheology. Yogurt EV and 2 wt% HPMC-C16 hydrogel rheology shows solid-like behavior across the studied frequency range as well as robust self-healing behavior, (a) Frequency sweep of 2 wt% HPMC-C16 and 1012yogurt EV hydrogel, (b) Shear thinning abilities of yogurt EV and HPMC-C16 hydrogel (viscosity decreases with increasing shear rate), (c) Hydrogel step-shear rheological measurements to characterize self-healing (d) Strain-controlled oscillatory shear rheology (amplitude sweep assessing the linear viscoelastic regime) of yogurt EV hydrogels

[0038] Fig. 27: G' (observed vs. expected) vs. HPMC-C16 wt% and yogurt EV number.Increasing polymer wt% and adding yogurt EVs to HPMC-C16 hydrogels leads to a material with a higher storage modulus G’ than the expected (additive) G’ across the entire tested frequency range (0.1-100 rad / s).

[0039] Fig. 28: HPMC-C14 DoF variation rheology. Increasing degree of functionalization (DoF) of HPMC-C14 enhances both (a) hydrogel stiffness and (b) solid-like behavior

[0040] Fig. 29: Yogurt EV and HPMC-C18 hydrogel rheology. Representative rheology of hydrogels formulated with yogurt EVs and HPMC-C18. (a) Addition of yogurt EVs destabilizes HPMC-C18 polymerpolymer interactions, leading to decreased stiffness of polymer-only hydrogel. Stiffness subsequently increases as more EVs are added but does not exceed that of the polymer alone, (b) Increasing polymer (HPMC-C18) wt% in yogurt EV hydrogels also increases hydrogel stiffness, but there is no significant contribution by yogurt EVs (~I012) to hydrogel stiffness regardless of polymer wt% (c) Increasing degree of functionalization (DoF) of HPMC-C18 enhances both (c) hydrogel stiffness and (d) solid-like behavior, but there is no significant contribution by yogurt EVs.

[0041] Fig. 30: Representative rheology of HPMC-C16 and HPMC-C18 hydrogel formulations, (a) Addition of yogurt EVs enhances HPMC-C16 hydrogel solid-like properties, marked by a decrease in tan 5 with increasing EV number (b) Addition of yogurt EVs destabilizes HPMC-C18 polymerpolymer interactions, leading to lower solid-like behavior (larger tan8) values compared to polymer-only hydrogel. System becomes more-solid as EV number is increased but not more than polymer alone.

[0042] Fig. 31: Yogurt EVs enable stiffness tuning even with lower HPMC-C16 functionalization. Hydrogels formed with less functionalized HPMC-C16 (2.93 mol%) show a significant correlation between yogurt EV number and hydrogel stiffness (G' and G" at 10 rad / s), similar to trends observed with the higher DoF polymer. Statistical comparisons were made using simple linear regression.

[0043] Fig. 32: Yield strain reveals different gelation mechanisms across polymer systems. Polymers that form gels on their own show decreased yield strain with more EVs, suggesting crowding or partial jamming. In contrast, low DoF polymers rely on EV crosslinking and maintain consistent yield strain across doses.

[0044] Fig. 33: Representative NTA traces and TEM images of lyophilized nanovesicles.Representative NTA traces and transmission electron microscopy (TEM) images of (a) bacterial (E. coli nissle) ACDVs and (b) mammalian (B16F10) melanoma cell-derived ACDVs.

[0045] Fig. 34: Mammalian (B16F10) cells and ACDVs do not express CD81 protein.

[0046] Fig. 35: Yogurt EVs are gradually released from hydrogels and retain structural integrity post-release, (a) Cumulative in vitro release profile of yogurt EVs from hydrogels over two weeks shows sustained release behavior (b) Korsmeyer-Peppas2model fitting of release data demonstrates good agreement across replicates (r2= 0.99), with parameters indicating Non-Fickian transport-dominated release (n = 0.61, K = 0.15) (c) Nanoparticle tracking analysis (NTA) confirms that EVs released from the gel maintain size distribution comparable to pre-gel EVs (d) Representative cryo-TEM images of released EVs show intact vesicle morphology, supporting preservation of EV structural integrity post-release. Scale bars denote 100 nm.

[0047] Fig. 36: Whey EV hydrogels do not cause weight loss in vivo. Stable body weights of hydrogel-treated mice are consistent with body weights of yogurt EV bolus injection treated BALB / cJ mice. Data points and bars represent mean and error bars represent SEM. Statistical comparisons were made using an unpaired t-test of the area under the curve (AUC) data.

[0048] Fig. 37: Whey EV hydrogels do not cause weight loss in vivo. Stable body weights of hydrogel -treated mice are consistent with body weights of yogurt EV bolus injection treated BALB / cJ mice. Data points and bars represent mean and error bars represent SEM. Statistical comparisons were made using an unpaired t-test of the area under the curve (AUC) data.

[0049] Fig. 38: Histological analysis of whole EV hydrogel explants. Histological analysis of explanted yogurt EV hydrogels using (a) H&E and (b) trichrome staining of whole hydrogel explant. Scale bars denote 800 pm.

[0050] Fig. 39: Vascularization of whole yogurt EV hydrogel explant (IF inc. DAPI). EV hydrogel cellular infiltration is illustrated by DAPI (gray) nuclear stain, and EV hydrogel-induced angiogenesis is confirmed by immunofluorescence staining of angiogenic molecular markers (F-actin (magenta) and mature (CD31) and progenitor (CD34) endothelial cell markers). Scale bar denotes 500 pm.

[0051] Fig. 40: Vascularization of whole hydrogel explant (TF). EV hydrogel-induced angiogenesis is confirmed by immunofluorescence staining of angiogenic molecular markers. Explanted hydrogels were stained for F-actin (magenta) and mature (CD31) and progenitor (CD34) endothelial cell markers. Scale bars denote 500 pm all panels.

[0052] Fig. 41: CD34 and CD31 co-localization shows neovascularization and maturing blood vessels. Explanted hydrogels were stained for progenitor (CD34) and mature (CD31) endothelial cell markers. Scale bars denote 100 pm for all panels.

[0053] Fig. 42: F-actin co-localization with CD31 highlights endothelial cell organization and dynamic vascular remodeling. Explanted hydrogels were stained for the actin cytoskeleton (F-actin) and endothelial cell junctions (CD31). Scale bars denote 100 pm for all panels.

[0054] Fig. 43: F-actin co-localization with CD34 highlights endothelial cell organization and dynamic vascular remodeling. Explanted hydrogels were stained for the actin cytoskeleton (F-actin) and progenitor endothelial cells (CD34). Scale bars denote 100 pm for all panels.

[0055] Fig. 44: CD34 and CD31 co-localization shows neovascularization and maturing blood vessels. Explanted hydrogels were stained for progenitor (CD34) and mature (CD31) endothelial cell markers. Scale bars denote 200 pm for all panels.

[0056] Fig. 45: F-actin co-localization with CD31 highlights endothelial cell organization and dynamic vascular remodeling. Explanted hydrogels were stained for the actin cytoskeleton (F-actin) and endothelial cell junctions (CD31). Scale bars denote 200 pm for all panels.

[0057] Fig. 46: F-actin co-localization with CD34 highlights endothelial cell organization and dynamic vascular remodeling. Explanted hydrogels were stained for the actin cytoskeleton (F-actin) and progenitor endothelial cell cells (CD34). Scale bars denote 200 pm for all panels.

[0058] Fig. 47: Vascularization marker staining of liposomal nanoparticle (LNP) hydrogel control (IF). HPMC-C12 (2 wt%) and liposomal nanoparticle (5wt%, ~1012liposomes / mL) hydrogel stained for angiogenesis markers (CD34, CD31, F-actin) shows no significant blood vessel formation.

[0059] Fig. 48: CD3+FOXP3+regulatory T cells localize near vasculature in yogurt EV hydrogels. Representative immunofluorescence images of hydrogel explant stained for CD3 (magenta), FOXP3 (yellow), and CD34 (cyan) show co-localization of regulatory T cells andvascular structures across multiple regions. Top panels highlight the presence of CD34+vessels, while bottom panels emphasize CD3+FOXP3+cells indicative of regulatory T cell infiltration. Scale bars denote 200 pm for main panels.

[0060] Fig. 49: CD3+FOXP3+regulatory T cells localize near vasculature in yogurt EV hydrogels, (a-c) Multicolor immunofluorescence images from three representative explants show spatial proximity of CD3+FOXP3+regulatory T cells (yellow arrowheads) and CD34+vasculature (blue arrowheads), (i-iii) Zoomed-in regions from Explant 1 illustrate regulatory T cell localization near vessels at single-cell resolution. Grayscale single-channel images are shown for each marker. Scale bars denote 200 pm in overview panels and 20 pm in insets.

[0061] Fig. 50: IF confirms marker-specific staining (unstained vs. stained gel), (a) Unstained control hydrogel shows minimal background signal across all channels. Dashed line outlines the explant boundary, (b) The same yogurt EV hydrogel imaged in Supplemental Figures 1 and 2 displays broad and specific staining for CD34 (cyan), CD3 (magenta), and FOXP3 (yellow) across the entire explant. Scale bar denotes 500 pm.

[0062] Fig. 51: Pre-processing strategy to isolate a particular cell class for immunophenotyping.

[0063] Fig. 52: Flow gating strategy for immunophenotyping.DETAILED DESCRIPTION OF THE INVENTION

[0064] A supramolecular hydrogel comprising (i) at least one polymer and (ii) extracellular vesicles.

[0065] A method of producing a supramolecular hydrogel comprising (i) at least one polymer and (ii) extracellular vesicles, the method comprising admixing at least one polymer and extracellular vesicles under conditions permitting creation of the supramolecular hydrogel.

[0066] A method of isolating EVs from a biological sample or food product containing EVs, comprising subjecting the biological sample or food product containing EVs to tangential flow filtration (TFF) comprising (a) processing the biological sample or food product containing EVs through a hollow fiber with an pore size that removes cell debris so as to produce a permeate, and (b) passing the permeate through a hollow fiber with a 100 kDa cut-off, so as to thereby retain isolated EVs in the retentate while eliminating smaller soluble molecules.

[0067] A method of injecting a supramolecular hydrogel described herein into an animal, comprising injecting an amount of the supramolecular hydrogel into the animal.

[0068] A supramolecular hydrogel comprising (i) at least one polymer and (ii) extracellular vesicles.

[0069] In some embodiments of the methods or supramolecular hydrogel described herein, the following can apply:

[0070] In some embodiments, molecules of the at least one polymer are reversibly and / or non-covalently crosslinked by the extracellular vesicles. In some embodiments, the extracellular vesicles are the only crosslinkers of the supramolecular hydrogel. In some embodiments, the extracellular vesicles are the only nanoparticulate crosslinkers in the supramolecular hydrogel. In some embodiments, the supramolecular hydrogel does not contain liposomes. In some embodiments, the EVs include EVs with a above 50nM. In some embodiments, the supramolecular hydrogel does not contain HPMC-C12 or HPMC-C18.

[0071] In some embodiments, the supramolecular hydrogel is self-healing. In some embodiments, the supramolecular hydrogel can return to its original viscosity within 60 seconds of having high shear applied thereto which high shear reduces its viscosity from its original level.In some embodiments, the supram olecul ar hydrogel is injectable, i.e., can be passed through an injection needle if pressure is applied to it.

[0072] In some embodiments, the at least one polymer is a cellulose-based polymer, preferably an alkyl-modified cellulose polymer. In some embodiments, the at least one polymer is a modified hydroxypropyl methylcellulose (HPMC-CX), wherein X is 12, 14, 16, or 18. In some embodiments, the X is 14 or 16. In some embodiments, X is 14. some embodiments, X is 16.

[0073] In some embodiments, the polymer concentration of the hydrogel is 2-4 wt%. In some embodiments, the polymer concentration of the hydrogel is 2 wt%. In some embodiments, the polymer concentration of the hydrogel is 3 wt%. In some embodiments, the polymer concentration of the hydrogel is 4 wt%. In some embodiments, the polymer concentration of the hydrogel is one of 2.0, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8, 2.9, 3.0, 3.1, 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9, or 4.0 wt%.

[0074] In some embodiments, the polymer has a degree of functionalization of about 0.32 -0.52 mmol / g. In some embodiments, the polymer has a degree of functionalization of about 0.20, 0.21, 0.22, 0.23, 0.24, 0.25, 0.26, 0.27, 0.28, 0.29, 0.30, 0.31, 0.32, 0.33, 0.34, 0.35, 0.36, 0.37, 0.38, 0.39, 0.40, 0.41, 0.42, 0.43, 0.44, 0.45, 0.46, 0.47, 0.48, 0.49, 0.50, 0.51, 0.52, 0.53, 0.54, 0.55, 0.56, 0.57, 0.58, 0.59, 0.60, 0.61, 0.62, 0.63, 0.64, 0.65, 0.66, 0.67, 0.68, 0.69, 0.70, 0.71, 0.72, 0.73, 0.74, 0.75, 0.76, 0.77, 0.78, 0.79, or 0.80 mmol / g, or any sub-range thereof.

[0075] In some embodiments, the EVs of the supramolecular hydrogel have a negative surface charge with mean zeta potential of about -32 to -40 mV. In some embodiments, the EVs of the supramolecular hydrogel have a negative surface charge with mean zeta potential of about -36.09 mV. In some embodiments, the EVs of the supramolecular hydrogel have a negative surface charge with mean zeta potential of about -26 to -31 mV. In some embodiments, the EVs of the supramolecular hydrogel have a negative surface charge with mean zeta potential of about -28.84 mV.

[0076] In some embodiments, the EVs of the supramolecular hydrogel have a mode diameter of about 130 to 150 nM for yogurt EVs. In some embodiments, the EVs of the supramolecular hydrogel have a mode diameter of about 139.2 nm. In some embodiments, the EVs of the supramolecular hydrogel have a mode diameter of about 150 to 165 nM for lyophilized yogurt EVs. In some embodiments, the EVs of the supramolecular hydrogel have a mode diameter ofabout 158.0 nm for lyophilized yogurt EVs. In some embodiments, the EVs of the supram olecul ar hydrogel have aZ-average diameter of about 140 to 160 nm for yogurt EVs. In some embodiments, the EVs of the supramolecular hydrogel have a Z-average diameter of about 148.1 nm for yogurt EVs. In some embodiments, the EVs of the supramolecular hydrogel have a Z-average diameter of about 160 to 180 nm for lyophilized yogurt EVs. In some embodiments, the EVs of the supramolecular hydrogel have a Z-average diameter of about 170.19 nm for lyophilized yogurt EVs.

[0077] In some embodiments, the supramolecular hydrogel comprises cross links between molecules of at least one polymer and one or more extracellular vesicles.

[0078] In some embodiments, the extracellular vesicles are derived from an agricultural source, bovine milk, milk whey isolated from yogurt, cells from a cultured cell line, human cells, murine cells, cancer cells, bacteria, or microalgae. In some embodiments, the extracellular vesicles are derived from yogurt. In some embodiments, the extracellular vesicles are derived from yogurt made from bovine milk.

[0079] In some embodiments, the extracellular vesicles are isolated by tangential flow fdtration. In some embodiments, SEC-HPLC is further used to purify the extracellular vesicles.

[0080] In some embodiments, the extracellular vesicles are lyophilized. In some embodiments, the extracellular vesicles are reconstituted after lyophilization.

[0081] In some embodiments, the final extracellular vesicle weight percentage in the supramolecular hydrogel ranges from 1-10% by weight.

[0082] In some embodiments, the supramolecular hydrogel comprises 1010-1013EVs / mL, preferably IxlO11to 5xl012EVs / mL.

[0083] In some embodiments, the supramolecular hydrogel has a storage modulus of about 200-700 Pa and / or a loss modulus of about 150-250 Pa. In some embodiments, the supramolecular hydrogel comprises HPMC-C16 at 2 wt% and has a storage modulus (G’) of about 120 to 180 Pa. In some embodiments, the supramolecular hydrogel comprises HPMC-C16 at 2 wt% and has a storage modulus (G’) of about 150 Pa. In some embodiments, the supramolecular hydrogel comprises HPMC-C16 at 3 wt% and has a storage modulus (G’) of about 630 to 710 Pa. In some embodiments, the supramolecular hydrogel comprises HPMC-C16 at 3 wt% and has a storagemodulus (G’) of about 669 Pa. In some embodiments, the supram olecul ar hydrogel comprises HPMC-C16 at 3 wt% and ~1O10'12EVs per m . In some embodiments, the supramolecular hydrogel comprises HPMC-C16 at 3 wt% and ~1O10 12yoghurt EVs per mL. In some embodiments, the supramolecular hydrogel comprises HPMC-C16 at 3 wt% and ~1012yoghurt EVs per mL. In some embodiments, the supramolecular hydrogel comprises HPMC-C16 at 3 wt% and -IO11yoghurt EVs per mL.

[0084] In some embodiments, the supramolecular hydrogel demonstrates shear-thinning and self-healing upon removal of the shear, preferably returning to the original viscosity within 60 seconds of the shear removal.

[0085] In some embodiments, the hydrogel has a storage modulus of about 80-100%.

[0086] In some embodiments, the extracellular vesicles are artificial cell-derived vesicles (ACDVs) or are secreted from a cell. In some embodiments, the extracellular vesicles are artificial cell-derived vesicles (ACDVs).

[0087] In some embodiments, the extracellular vesicles are secreted from, or manufactured from, a bacterial, yeast, or a mammalian cell. In some embodiments, the extracellular vesicles are secreted from, or manufactured from, a yeast. In some embodiments, the extracellular vesicles are secreted from, or manufactured from, a mammalian cell line. In some embodiments, the extracellular vesicles are isolated from a T-cell or a cancer cell. In some embodiments, the extracellular vesicles are isolated from bacteria and are 175-350 nm in diameter, have a Z-average of 230-325nm, and / or have a mean zeta potential of -40-60 mV. In some embodiments, the extracellular vesicles are isolated from cancer cells and are 75-210 nm in diameter, have a Z-average of 150-200nm, and / or have a mean zeta potential of -30-45 mV. In some embodiments, the extracellular vesicles are artificial cell-derived vesicles (ACDVs) and the yield strain for the hydrogel is about 200%, the storage modulus at lOrad / s is about 525-575 Pa, and / or the loss modulus at lOrad / s is about 200-225 Pa.

[0088] In some embodiments, the extracellular vesicles are 30 to 4,000 nm in diameter. In some embodiments, the extracellular vesicles are above 50nM; 150-200nm; or 230-325nm in diameter. In some embodiments, the extracellular vesicles are 30-150 nm in diameter. In some embodiments, the extracellular vesicles are microvesicles. In some embodiments, the extracellular vesicles are 100 nm - 1 uM in diameter. In some embodiments, the extracellular vesicles areexosomes. In some embodiments, the extracellular vesicles are isolated from plasma, urine, semen, saliva, bronchial fluid, cerebral spinal fluid (CSF), breast milk, serum, amniotic fluid, synovial fluid, tears, lymph, bile, or gastric acid.

[0089] In some embodiments, the extracellular vesicles are secreted from or isolated from cells. In some embodiments, the extracellular vesicles comprise cytosolic and / or plasma membrane associated proteins. Such proteins can, in some embodiments, include proteins known to cluster at the plasma membrane surface, such as tetraspanins. In some embodiments, the EVs comprise CD81.

[0090] In some embodiments, the extracellular vesicles are derived from endothelial cells, mesenchymal stem cells, stromal cells, platelets, leukocytes, erythrocytes, T-cells or cancer cells. In some embodiments, the extracellular vesicles of the supramolecular hydrogel are derived from same species as the supramolecular hydrogel is administered to. In some embodiments, the extracellular vesicles are not derived from mesenchymal stem cells.

[0091] In some embodiments, HPMC-C12, C14, Cl 6, or C18 polymer is modified using the respective isocyanate. In some embodiments 0.52 mmol of the respective isocyanate is used per gram of HPMC. In some embodiments, the supramolecular hydrogel further comprises water or further comprises phosphate-buffered saline.

[0092] In some embodiments, a cargo molecule has been incorporated into the EVs of the supramolecular hydrogel. In some embodiments, no exogenous cargo molecules have been incorporated into the EVs of the supramolecular hydrogel.

[0093] A method of producing a supramolecular hydrogel comprising (i) at least one polymer and (ii) extracellular vesicles, the method comprising admixing at least one polymer and extracellular vesicles under conditions permitting creation of the supramolecular hydrogel.

[0094] In some embodiments, at least one polymer is a cellulose-based polymer, preferably an alkyl-modified cellulose polymer. In some embodiments, at least one polymer is a modified hydroxypropyl methylcellulose (HPMC-CX), wherein X is 12, 14, 16, or 18. In some embodiments, at least one polymer is a modified hydroxypropyl methylcellulose (HPMC-CX), wherein X is 14. In some embodiments, at least one polymer is a modified hydroxypropyl methylcellulose (HPMC-CX), wherein X is 16.

[0095] A method of isolating EVs from a biological sample or food product containing EVs, comprising subjecting the biological sample or food product containing EVs to tangential flow filtration (TFF) comprising (a) processing the biological sample or food product containing EVs through a hollow fiber with an pore size that removes cell debris so as to produce a permeate, and (b) passing the permeate through a hollow fiber with a 100 kDa cut-off, so as to thereby retain isolated EVs in the retentate while eliminating smaller soluble molecules.

[0096] In some embodiments, the biological sample or food product containing EVs comprises or is obtained from bovine milk yoghurt. In some embodiments, the biological sample or food product containing EVs comprises whey is obtained from bovine milk yoghurt. In some embodiments, no ultracentrifugation is used in the method. In some embodiments, an average size pore size of the hollow fiber is 650 nm.

[0097] A method of injecting a supram olecular hydrogel described herein into an animal, comprising injecting an amount of the supramolecular hydrogel into the animal.

[0098] In some embodiments, the amount of the supramolecular hydrogel is injected into a tissue or organ of the animal. In some embodiments, no ECM forms around the injected microgel within 7 days post-injection. In some embodiments, cells from the subject injected infiltrate the injected supramolecular hydrogel. In some embodiments, cells from the subject injected infiltrate the injected supramolecular hydrogel within 7 days post-injection. In some embodiments, the cells are CD34 and / or CD31 positive. In some embodiments, the cells are CD3 positive. In some embodiments, the cells are CD3 express FOXP3 with nuclear localization. In some embodiments, the cells are endothelial cells. In some embodiments, the cells are endothelial cells are innate immune cells. In some embodiments, the cells are endothelial cells are myeloid. In some embodiments, the EVs are derived from cell of same species as is being injected. In some embodiments, the EVs are derived from an organ or tissue of same type that the hydrogel is inj ected into.

[0099] In some embodiments, the extracellular vesicles of the supramolecular hydrogel contain a nucleic acid. In some embodiments, the extracellular vesicles of the supramolecular hydrogel contain a protein or peptide. In some embodiments, the extracellular vesicles of the supramolecular hydrogel contain an antibody or antibody fragment. In some embodiments, the extracellular vesicles of the supramolecular hydrogel contain a pro-angiogenic factor. In someembodiments, the pro-angiogenic factor is vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), or a microRNAs. In some embodiments, the microRNA is miR-126 or miR-214.

[0100] In some embodiments, the extracellular vesicles of the supram olecul ar hydrogel after injection contains a higher percentage of CD45+immune cells than an equivalent liposomal hydrogel does. In some embodiments, the extracellular vesicles of the supramolecular hydrogel after injection contains a higher percentage of neutrophils ([F4 / 8O'CD1 lb+Ly6G+]), dendritic cells ([F4 / 80'Ly6G'CDl lc+]), and / or macrophages ([CD1 lb+F4 / 80+]) than an equivalent liposomal hydrogel does. In some embodiments, the extracellular vesicles of the supramolecular hydrogel after injection contains a higher percentage in T cell (CD1 lb'CD3+) than an equivalent liposomal hydrogel does.

[0101] A method of eliciting angiogenesis in a subject is provided comprising administering to the subject a supramolecular hydrogel described herein, wherein the extracellular vesicles of the supramolecular hydrogel contain a pro-angiogenic factor, effective to elicit angiogenesis in a subject. In some embodiments, the extracellular vesicles of the supramolecular hydrogel contain a pro-angiogenic factor. In some embodiments, the pro-angiogenic factor is vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), or a microRNAs. In some embodiments, the microRNA is miR-126 or miR-214. In some embodiments, the supramolecular hydrogel comprises HPMC-C16 (3 wt%) and -1011yogurt EVs.Definitions

[0102] Unless otherwise defined, all technical and / or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention pertains. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of embodiments of the invention, exemplary methods and / or materials are described below. In case of conflict, the patent specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and are not intended to be necessarily limiting.

[0103] In the discussion unless otherwise stated, adjectives such as “substantially” and “about” modifying a condition or relationship characteristic of a feature or features of an embodiment of the invention, are understood to mean that the condition or characteristic is defined to within tolerances that are acceptable for operation of the embodiment for an application for which it is intended. In embodiments, about means within a standard deviation using measurements generally acceptable in the art. In embodiments, about means a range extending to + / - 10% of the specified value. In embodiments, about includes the specified value. Unless otherwise indicated, the word “or” in the specification and claims is considered to be the inclusive “or” rather than the exclusive or, and indicates at least one of and any combination of items it conjoins.

[0104] It should be understood that the terms “a” and “an” as used above and elsewhere herein refer to “one or more” of the enumerated components. It will be clear to one of ordinary skill in the art that the use of the singular includes the plural unless specifically stated otherwise. Therefore, the terms “a,” “an” and “at least one” are used interchangeably in this application.

[0105] For purposes of better understanding the present teachings and in no way limiting the scope of the teachings, unless otherwise indicated, all numbers expressing quantities, percentages or proportions, and other numerical values used in the specification and claims, are to be understood as being modified in all instances by the term “about.” Accordingly, unless indicated to the contrary, the numerical parameters set forth in the following specification and attached claims are approximations that may vary depending upon the desired properties sought to be obtained. At the very least, each numerical parameter should at least be construed in light of the number of reported significant digits and by applying ordinary rounding techniques.

[0106] In the description and claims of the present application, each of the verbs, “comprise,” “include” and “have” and conjugates thereof, are used to indicate that the object or objects of the verb are not necessarily a complete listing of components, elements or parts of the subject or subjects of the verb. Other terms as used herein are meant to be defined by their well-known meanings in the art.General

[0107] For the foregoing embodiments, each embodiment disclosed herein is contemplated as being applicable to each of the other disclosed embodiments.

[0108] As used herein, all headings are simply for organization and are not intended to limit the disclosure in any manner. The content of any individual section may be equally applicable to all sections. All combinations of the various elements disclosed herein are within the scope of the invention.

[0109] Additional obj ects, advantages, and novel features of the present invention will become apparent to one ordinarily skilled in the art upon examination of the following examples, which are not intended to be limiting. Additionally, each of the various embodiments and aspects of the present invention as delineated hereinabove and as claimed in the claims section below finds experimental support in the following examples.

[0110] It is appreciated that certain features of the invention, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the invention, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable subcombination or as suitable in any other described embodiment of the invention. Certain features described in the context of various embodiments are not to be considered essential features of those embodiments, unless the embodiment is inoperative without those elements.

[0111] Examples are provided below to facilitate a more complete understanding of the invention. The following examples illustrate the exemplary modes of making and practicing the invention. However, the scope of the invention is not limited to specific embodiments disclosed in these Examples, which are for purposes of illustration only.EXAMPLES

[0112] Yogurt production All yogurt production was performed using a 10-quart Instant Pot Duo Nova model. The Instant Pot, stainless steel insert, and sealing ring were thoroughly cleaned with hot soapy water, rinsed, and dried before use. Two gallons of whole milk (Cream O Land Dairy, 529 Cedar Lane Florence, NJ 08518) were heated in the Instant Pot using the "Yogurt" setting on "boil" mode for approximately 90 minutes to reach 80 °C. The milk was then cooled at room temperature for about 2 hours until it reached 40 °C (thermometer used to measure). For the initial batch, 1 / 2 cup of whole-milk yogurt containing live cultures was used as a starter (The Icelandic Milk and Skyr Corporation, NY, Siggi’s); for subsequent batches, 1 cup of whey fromthe previous batch was utilized. The starter was thoroughly mixed with the warm milk using a whisk and the yogurt was then cultured in the Instant Pot for 10 hours using the "Yogurt" setting. Following incubation, the yogurt was refrigerated overnight in the Instant Pot insert covered with plastic wrap. Straining was performed using a Hatrino 1 -gallon stainless steel yogurt strainer. The chilled yogurt was divided into two batches and strained for a minimum of 12 hours each to separate the whey. All utensils, including measuring cups, spoons, ladles, and mixing bowls, were cleaned and sanitized before use.

[0113] Yogurt production All yogurt production was performed using a 10-quart Instant Pot Duo Nova model. The Instant Pot, stainless steel insert, and sealing ring were thoroughly cleaned with hot soapy water, rinsed, and dried before use. Two gallons of whole milk (Cream O Land Dairy, 529 Cedar Lane Florence, NJ 08518) were heated in the Instant Pot using the "Yogurt" setting on "boil" mode for approximately 90 minutes to reach 80 °C. The milk was then cooled at room temperature for about 2 hours until it reached 40 °C (thermometer used to measure). For the initial batch, 1 / 2 cup of whole-milk yogurt containing live cultures was used as a starter (The Icelandic Milk and Skyr Corporation, NY, Siggi’s); for subsequent batches, 1 cup of whey from the previous batch was utilized. The starter was thoroughly mixed with the warm milk using a whisk and the yogurt was then cultured in the Instant Pot for 10 hours using the "Yogurt" setting. Following incubation, the yogurt was refrigerated overnight in the Instant Pot insert covered with plastic wrap. Straining was performed using a Hatrino 1 -gallon stainless steel yogurt strainer. The chilled yogurt was divided into two batches and strained for a minimum of 12 hours each to separate the whey. All utensils, including measuring cups, spoons, ladles, and mixing bowls, were cleaned and sanitized before use.

[0114] Yogurt EVs Bovine milk whey was obtained from yogurt made from whole milk (Isolation protocol above). Pre-processing methods consisted of subsequent centrifugation steps at 1000 g for 10 minutes at 4°C to eliminate dead cells, followed by a centrifugation at 5000 g for 20 minutes at 4°C to eliminate cells debris. The collected milk whey was used fresh or stored at 4°C for less than 48 hours. The KrosFlo KR2i TFF System from Repligen (Spectrum Labs, Los Angeles, CA, USA) was used to isolate EVs. Firstly, whey (1.0-1.5 L) was clarified by microfiltration using a modifed Polyethersulfone (mPES) hollow fiber (HF) module with a pore size of 650 nm (D02-E65U-07-N, Spectrum Labs) with an active membrane area of 0.01 m2. TheTFF system operated in Constant Feed Concentration / Diafiltration mode with an input flow rate of 89 mL / min to keep the shear force of the feed stream below 2000 s'1.52Both feed and permeate were kept on ice. The EV-containing suspension recovered from the permeate was then used for the subsequent ultrafiltration step. EVs were isolated using a smaller mPES 100-kDa cut-off TFF module (D06-E100-05-N, Spectrum Labs) with an active membrane area of 0.04 m2. The TFF system operated in Constant Feed Concentration / Diafiltration / Concentration mode with a feed flow of 53 ml / min. Both feed and retentate were kept on ice. Final samples were concentrated to 100 m and diafiltrated at least 5 times with 0.22 pm filtered PBS. Final EV samples are stored at -80 °C or frozen at -80°C and then lyophilized using the FreeZone Benchtop Freeze Dryer (Labconco). After lyophilization, EV samples were reconstituted in MilliQ water. Further details regarding the

[0115] The automated TFF system used in the present invention consists of a digital peristaltic main pump and auxiliary pumps, scales, digital readouts of pressure values with automated shutoff controls, and backpressure valves that add automatic transmembrane pressure (TMP) control to the TFF process, reducing operator hands-on-time, improving process reproducibility and enhancing system automation. HF modules were employed to enhance sample concentration, fractionation, and washing while preventing membrane fouling and maximizing product recovery. Specifically, the TFF system operated in Constant Feed Concentration / Diafiltration / Concentration (CFC / D / C) mode.

[0116] E. coli ACDVs 10 mb of E. Coli Nissle 1917 was cultured overnight (16-20 hours) and then harvested; optical density (OD) was confirmed using a plate reader and the culture was diluted to 1 OD600=8xl08cells / mL (BioTek Synergy Neo2). 1 L of lysogeny broth (LB) was prepared with 100 pg / mL of Erythromycin. The LB broth was split (500 mL was placed in two Erlenmeyer flasks), and 5 mL of the E. Coli overnight culture was added to each. The cultures were grown overnight for 16-20 hours. The OD was then measured again to assess growth, and the cultures were centrifuged to allow the A. Coli culture to pellet. The supernatant was then collected for TFF, as described for yogurt EVs. Briefly, E. Coli culture (1 L) was clarified by microfiltration using a mPES HF module with a pore size of 650 nm (D02-E65U-07-N, Spectrum Labs, flow rate = 89 mL / min).The E. Coli vesicle-containing suspension recovered from the permeate of the 650 nm TFF HF modules then underwent an ultrafiltration step using a 100-kDa cut-off TFF module (D06-E100-05-N, Spectrum Labs, flow rate = 53 mL / min). Final samples were concentrated to 25 mL. Amicon centrifugal filters (Amicon Ultra Centrifugal Filter, 10 kDa MWCO; regenerated cellulose membrane, 15 mL) were used to concentrate the final TFF retentate to 1-2 mL, centrifuging at 4200 g for 45 mins at 4°C (three 15 min intervals), as per the manufacturer’s instructions. Following concentration, the nanovesicles were subjected to size reduction via extrusion, using an Avanti mini extruder with 1 pm, 400 nm, 200 nm, and 100 nm membranes. ACDVs were then lyophilized using the FreeZone Benchtop Freeze Dryer (Labconco). After lyophilization, E. Coli ACDV samples were reconstituted in lx PBS.

[0117] Mammalian ACDVs B16-F10 melanoma cells (ATCC CRL-6475) were plated at 100,000 cells / mL in 50 mL (T75 tissue culture flasks). Once desired confluency to yield an adequate number of ACDVs to generate hydrogels was reached (approximately 170 M cells total, 7.02xl03ACDVs per cell), cells were centrifuged at 1000 g for 5 minutes and resuspended in 20 M cells / mL lysis buffer (10 mM Tris-HCl). The mixture was vortexed and cells were incubated in the lysis buffer for 60 minutes at 4°C. The lysed cells were loaded in a syringe and passed through a Confined Impinging Jet (CIJ) mixer, depressing the syringe as quickly as possible. This step was repeated twice, and then nuclease was added (25 units / mL of cell solution or 0.1 pL nuclease / mL cell solution). This solution was mixed by inverting lOx and incubated for 15 minutes on ice. Large cell debris was removed by passing the suspension through a 0.45 pm filter, after which the sample was concentrated using an Amicon centrifugal filter unit (Amicon Ultra-4 100 kDa, 4 mL, cellulose membrane) at 4000 g for 10-15 minutes, as per the manufacturer’s instructions. The cell solution was then extruded using an Avanti mini extruder with 1 pm, 400 nm, 200 nm, and 100 nm membranes to form ACDVs through mechanical disruption. ACDVs were then lyophilized using the FreeZone Benchtop Freeze Dryer (Labconco). After lyophilization, mammalian ACDV samples were reconstituted in lx PBS.

[0118] Yield and concentration information for yogurt EVs and E. coli and mammalian cell-derived ACDVs can be found in Table 1.

[0119] Table 1: Yield and final post-lyophilization concentration comparison of Yogurt EV and bacterial (E. coli nissle) ACDVs and mammalian (B16F10) melanoma cell-derived ACDVs.Average concentration achieved postNanovesicle Type YieldlyophilizationYogurt EVs IO10particles / mL feed 1.0 x io13± 3.3 x io12particles / mL E. coli ACDV s 109parti cles / mL feed 1.9 x 1011± 1.0 x 1011particles / mL Mammalian108parti cles / mL feed 2.6 x 1011± 8.8 x io10particles / mL (B16F10) ACDVsYield is normalized by feed volume. For yogurt EVs, feed volume is defined as pre-processed milk whey (typically around IL). For bacterial and mammalian samples, it refers to the total culture volume prior to nanovesicle induction and isolation — specifically, typically 1 L of E. coli culture at ODeoo ~ 1.70, and 212.5 mb of media collected from melanoma cell flasks, which was concentrated to 10 mL at 1.76 x 106cells / mL for ACDV isolation.

[0120] Nanoparticle tracking analysis Measurement of nanoparticle size distribution and concentration was performed using Nanoparticle Tracking Analysis (NT A) using either NanoSight NS300 (Malvern Panalytical, United Kingdom) or ZetaView (Particle Metrix, USA). For Nanosight, EVs samples were diluted in 0.22 pm filtered PBS to reach 20-120 particles per frame. The analysis of the samples was executed using the NanoSight Software NTA 3.4 Build 3.4.003 (camera level 15-16, syringe pump speed 30-50) acquiring five videos of 60-s duration. The frame analysis was carried out setting a detection threshold (max 6) so that the observed particles are marked (red crosses in the software) and no more than five particles are rejected (blue crosses). For ZetaView analysis, EV samples were diluted in 0.22 pm filtered PBS and loaded into the cell. The instrument measured each sample at 11 distinct positions within the cell, performing two cycles of readings at each position. After automated analysis of all 11 positions and removal of any outliers, the ZetaView 8.02.28 software calculated the mean, median, and mode particle sizes (indicated as diameter), along with the concentration of the sample. The pre-acqui sition parameters were set to a temperature of 23 °C, sensitivity of 80, frame rate of 30 fps, shutter speed of 100, and a laser pulse duration matching the shutter duration. Post-acquisition settings included a minimum brightness of 25, maximum particle size of 200 pixels, and minimum size of 5 pixels. For both Nanosight and ZetaView measurements, three replicates were taken per sample, with the mode selected as the representative value for particle size. Results were expressed as the mean ± SEM.

[0121] Dynamic Light Scattering and Zeta Potential Dynamic Light Scattering (DLS) and Zeta Potential measurements were performed with a Zetasizer nano (Malvern Instruments, United Kingdom). EVs were diluted from 100 to 1000-fold in 0.22 pm filtered milliQ water. Size measurements were performed in disposable plastic cuvettes, while zeta potential measurements in DTS1070 cuvettes. Three measurements per sample were recorded and results were expressed as mean ± SEM

[0122] Bicinchoninic Acid Assay The sample protein content was quantified using the colorimetric BCA protein assay (Thermo Fisher Scientific, Rockford, IL, USA). Briefly, a standard curve of Human Serum Albumin standards in MilliQ water (0.1- 0.7 mg / mL concentration range) were prepared to estimate the total protein content of the samples. EV samples were diluted 10 folds in MilliQ water to perform the assay. After 30 minutes incubation at 37°C, the absorbance of the BCA soluble compound was measured at 562 nm, according to the manufacturer’s instructions, using Agilent BioTek Synergy Neo2 Multi-Mode Microplate Reader (Agilent).

[0123] Immunoblotting For the identification of CD81 EV marker in yogurt-EVs and TSG101 in mammalian ACDVs, samples were mixed with reducing sample buffer (480 mM Tris at pH 6.8, 12% sodium dodecyl sulfate, 45% glycerine, 0.06% bromophenol blue, and 12% 2-mercaptoethanol) and heated for 10 minutes at 70° C. The denatured samples were loaded alongside 5 pl of Protein Standard markers (1610374, BioRad) on polyacrylamide gels (4-20%; Mini Protean TGX Gels, BioRad) that were then inserted into an electrophoresis chamber filled with SDS Running Buffer (1610732, BioRad). The proteins were fractionated by size at 200V for approximately 30 minutes. After separation, a semi-dry blotting technique was used for the transfer of proteins from the gels to a PVDF membrane. Resolving gels were placed on top of the PVDF membranes and sandwiched between Trans-Blot Turbo Mini Transfer Packs (1704156, BioRad) and placed in a Trans-Blot Turbo System (BioRad). After the protein transfer step, membranes were blocked with Blotting-Grade Blocker (1706404, BioRad) and incubated with primary antibody anti CD81 with Bovine reactivity for yogurt-EVs (1:500 Rabbit NBP1-77039; Novus Bio) and with primary antibody anti TSG101 (1:1000 Rabbit anti-TSGlOl, ab30871, Abeam) overnight at 4°C. After washing, membranes were incubated for 1 hour with secondary antibody HRP (1:5000 Goat anti -Rabbit, 31460, Invitrogen) according to the manufacturer’s instructions.Immunoblots were revealed using Clarity Western Blot substrate (BioRad) using ChemiDoc Imaging System (BioRad).

[0124] Transmission Electron Microscopy EVs were fixed in 4% paraformaldehyde (1 : 1) at room temperature (2 minutes), placed (2 pL) on carbon-coated copper grids (300 Mesh, Electron Microscopy Sciences), and blotted. Water (2 pL) was added to the samples that were then blotted, and 2% aqueous uranyl acetate (2 pL) was placed on the grid (2 minutes) and blotted. The grids were examined with a Transmission Electron Microscope (Hitachi HT7800, Hitachi) at the Simons Electron Microscopy Center (SEMC, New York).

[0125] Cryo-Electron Microscopy EVs were diluted to obtain a final concentration of 1011parti cles / mL in PBS. Approximately 3.5 pL of the sample was placed on a holey carbon-coated copper grid Quantifoil® R 1.2 / 1.3 (300 Mesh, Electron Microscopy Sciences). The unstained sample was vitrified with the aid of a robotic accessory (Vitrobot; FEI, Hillsboro, OR, USA) used to plunge-freeze the aqueous sample into liquid ethane maintained at the temperature of liquid nitrogen. Once vitrified, the samples were stored in liquid nitrogen. Prior to image acquisition, the stored samples were transferred to a cryo-holder, which maintained their temperature at approximately - 180 °C during imaging. The cryo-TEM images were obtained at 200 kV using a Glacios 2 Cryo-Transmission Electron Microscope (Thermofisher) at the SEMC.

[0126] Synthesis of HPMC-Cx polymers To ensure consistency with prior studies3,38, HPMC-C12, C14, C16, and C18 polymers were modified using 0.52 mmol of the respective isocyanate per gram of HPMC. This degree of functionalization (DoF) was selected based on established protocols and allows for direct comparisons across different alkyl chain lengths. The volumes of isocyanate reagents added during the modification reaction were carefully calculated by accounting for the molecular weights and densities of the different molecules, maintaining uniformity in reaction conditions. Additionally, to explore the effects of reduced modification, lower DoF variants of HPMC-C14, C16, and C18 were synthesized. These formulations were prepared using 0.40 mmol / g, 0.32 mmol / g, and 0.26 mmol / g of the respective isocyanates, allowing for systematic investigation of the relationship between alkyl chain length and functionalization extent (Figures 28-29). To precisely quantify the degree of modification, 'H NMR spectroscopy was used to calculate mol% functionalization based on the integration ofcarbon chain proton peaks relative to the polymer backbone, ensuring accurate reporting and reproducibility (Figures 14-18).

[0127] Modified HPMC was generated using a protocol adapted from Correa, et al38. HPMC (1.0 g, Sigma-Aldrich) was dissolved in N-methylpyrrolidone (NMP; 40 ml, Sigma-Aldrich) by stirring overnight. Following, 1 -dodecylisocyanate (0.5 mmol, Sigma-Aldrich), 1-tetradecylisocyanate (0.5 mmol, Sigma-Aldrich), 1 -hexadecylisocyanate (0.5 mmol, Sigma-Aldrich), or 1 -octadecylisocyanate (0.5 mmol, Sigma- Aldrich) were dissolved in NMP (5 ml) and added to the reaction mixture along with a stochiometric amount of Hunig’s Catalyst (NN-Diisopropylethylamine, Sigma-Aldrich) while stirring at 50 °C. The reaction mixture was then stirred at room temperature for 24 hours, after which it was precipitated from acetone, recovered via centrifugation, and purified via dialysis (3.5 kDa molecular weight cutoff [MWCO]) over 4 days at room temperature. Pure HPMC-Cx was then lyophilized and dissolved in lx sterile PBS to yield a 6 wt % solution, which was stored at 4 °C until used.

[0128] Nuclear magnetic resonance spectroscopy characterization 1H nuclear magnetic resonance (NMR) spectra were obtained and recorded on a Bruker 500-MHz NMR spectrometer at 298 K. 1HNMR spectra were referenced to residual proton resonances in the deuterated solvents and chemical shifts (8) are given in parts per million. Deuterated DMSO was purchased from Cambridge Isotope Laboratories. Taking the ratio of the carbon chain-related peak integration at 0.86 ppm to the HPMC-related peak integration at 1 ppm112, we determined the mol% modification of HPMC-Cx polymers. NMR characterization results can be found in Table 2 and Figures 15-19.

[0129] Table 2: HPMC-Cx modificationFormulation Subformulation Theoretical Mol %modification modificationamount (via NMR)HPMC-C12 0.52 mmol / g 6.33 mol% HPMC-C14 Low DoF 0.40 mmol / g 7.33 mol%High DoF 0.52 mmol / g 9.33 mol%HPMC-C16 Low DoF 0.32 mmol / g 2.93 mol%High DoF 0.52 mmol / g 6.33 mol%

[0130] Yogurt EV polymer corona DLS experiments Dynamic Light Scattering (DLS) measurements were performed with a Zetasizer nano (Malvern Instruments, United Kingdom). While stirring, yogurt EVs were added to a ~0.005 mg mL1solution of HPMC-Cx to achieve a dilute solution with a 3 : 4 wt% HPMC-Cx: yogurt EV ratio, which corresponded to most yogurt EV hydrogel formulations. N = 3 samples were made for each formulation and results were expressed as mean ± SEM66.

[0131] Yogurt EV hydrogel formation Hydrogels were formed by simple mixture of yogurt EVs either in an Eppendorf tube or using a dual-syringe mixer. In an Eppendorf tube, the desired amounts of EVs and HPMC-Cx (both measured by weight) were added and lx PBS was added for any remaining volume. A metal spatula was used to mix the solution and formulate the hydrogel, usually for 5-10 minutes. When using a dual-syringe to mix, the desired amounts of EVs and HPMC-Cx (both measured by weight) were loaded into two separate 1-mL luer-lock syringes, which were connected to each other using a luer-lock elbow connector and mixed for 5- 10 minutes by alternating depression, as described previously18. Hydrogel formulations generally consisted of 3 wt% HPMC-Cx and 4 wt% yogurt EVs. For experiments varying weight percentages of EVs, either 0.1%, 1%, 4%, or 6 wt% of EVs were used. These wt% amounts consistently corresponded to 1010-1012yogurt EVs, and EV number should be measured empirically by NTA rather than taking wt% approximations to determine the amount of EVs in hydrogels. Details on hydrogel formulations can be found in Table 3.

[0132] Table 3Formulation Polymer Type Polymer EV Type EV wt° / o EV Number Hydrogel Figures wt% (# of EVs per (i.e. no G7G” Shown mL of crossover inhydrogel) frequencyrange)?HPMC-Cx HPMC-C16 3 wt% N / A N / A N / A YES Figures 2e, 3a- Polymer only 0.52 mmol / g d, 4a, 5f-h, S16,S17,S15 HPMC-C16 3 wt% N / A N / A N / A NO Figures 2e, 4b 0.32 mmol / gHPMC-C16 2 wt% N / A N / A N / A YES Figures 3a-d, 0.52 mmol / g S16-17 HPMC-C18 3 wt% N / A N / A N / A YES Figures S19- 0.52 mmol / g 20 HPMC-C18 2 wt% N / A N / A N / A YES Figures S19- 0.52 mmol / g 20 HPMC-C18 3 wt% N / A N / A N / A YES Figures S19- 0.26 mmol / g 20 Unmodified HPMC 3 wt% Yogu rt 4 wt% 6.50x1011NO Figures 4a-b HPMC andYogurt EVs 4 wt% 1.95x1012Formulation Polymer Type Polymer EV Type EV wt% EV Number Hydrogel Figures wt% (# of EVs per (i.e. no G7G” Shown ml of crossover inhydrogel) frequencyrange)?HPMC-Cx HPMC-C16 3 wt% Yogu rt 4 wt% 1.85x1012YES Figures 2a-4„ and Yogurt 0.52 mmol / g 3a-d, 4a, 4c-g, EVs 4 wt% 1.95x10125f-h, 6, 7, 8, 9,4 wt% 2.60x1012S4, S13-15, 4 wt% 4.19x10” S17, S20, S22, 4 wt% 4.19x1011S25-241 4 wt% 2.55x10111 wt% 1.36x10”4 wt% 1.24x10”4 wt% 2.55x1010HPMC-C16 3 wt% Yogu rt 4 wt% 1.85x1012YES Figures 2d-e, 0.32 mmol / g 4b, S21-224 wt% 1.95x10124 wt% 5.06x10”HPMC-C12 3 wt° / o Yogu rt 4 wt% 1.35x10” NO Figures 2a-c, 0.52 mmol / g 4 wt% 2.55x10” NO S10-11.S134 wt% 7.35x1011NO4 wt% 3.80x1012YESHPMC-C14 3 wt% Yogu rt 0.1 wt° / o 1.O5X1O10NO Figures 2a-c, 0.52 mmol / g 4 wt% 1.00x10” NO S12-131 wt% 1.36x10” NO4 wt% 4.19x10” NO4 wt% 4.19x1011NO4 wt% 5.44x1011YES6 wt% 6.29x10” YES6 wt% 8.16x10” YES4 wt% 2.06x1012YESHPMC-C16 2 wt% Yogu rt 4 wt% 2.04x10” YES Figures 3a-d, 0.52 mmol / g S164 wt% 1.94x10124 wt° / o 3.80x1012HPMC-C14 3 wt% Yogu rt 4 wt% 4.19x10” YES Figure S18 0.40 mmol / gHPMC-C18 3 wt% Yogu rt 4 wt% 0 YES Figures S19- 0.52 mmol / g 4 wt% 8.00x1010206 wt% 1.20x10”4 wt% 5.44x10”4 wt% 2.83x1012HPMC-C18 2 wt% Yogu rt 4 wt% 1.94x1012YES Figure S19 0.52 mmol / gHPMC-C18 3 wt% Yogu rt 4 wt% 2.06x1012YES Figure S19 0.26 mmol / gHPMC-Cx HPMC-C16 3 wt% Bacterial (E. coli) 4 wt% 2.63x1010YES Figures 5d, 5f- and 0.52 mmol / g ACDV gh BacterialACDVsHPMC-Cx HPMC-C16 3 wt% Mammalian 4 wt% 1.50x10” YES Figures 5e, 5f- and 0.52 mmol / g (B16F10 cancer h Mammalian cell) ACDVACDVs

[0133] Rheological characterization of EV hydrogels (shear rheology) Rheological testing was performed using a 20-mm-diameter serrated parallel plate at a 700-mm gap on a stress-controlled TA Instruments DHR-2 rheometer at 25 C equipped with a humidity chamber, unless otherwise specified. About 700 pL of the EV hydrogel was applied to the plate.

[0134] Frequency sweeps were performed at a strain of 1% from 0.1 to 100 rad / s. Amplitude sweeps were performed at frequency of 10 rads / s. Flow sweeps were performed from high to low shear rates with steady-state sensing. A 120 s recovery time was used for self-healing testing. TRIOS software (TA Instruments) was used for instrument management, experiment setup and execution and data collection. A solvent trap cover (TA Instrument) was used to create a thermally stable vapor barrier, virtually eliminating any solvent loss during rheological experiments and improving temperature uniformity.

[0135] Fluorescent Labeling of EVs Yogurt EVs were diluted to a concentration of 1012particles / mL in PBS. Concentrated AZDye 555 NHS ester in DMSO (Vector Laboratories, FP-1166) was then added the EVs to bring the concentration of dye to 200uM. This reaction mix was incubated at 25C for 2 hours with 500rpm shaking followed by 4 C incubation overnight. Trisbase was then added to a final concentration of lOmM and incubated at 25 C for 20 minutes to quench the reaction. Finally, unbound dye was separated from the sample by centrifugation through a 40 kDa MWCO Zeba spin desalting column (ThermoFisher Scientific, A57764), following the manufacturer’s protocol. EV labeling and unbound dye removal were confirmed via imaging cytometric single-particle fluorescence analysis (Cytek Biosciences ImageStream MKII).

[0136] In vitro EV release studies 3 wt% HPMC-C16 and -1011yogurt EV hydrogels were prepared with AZDye 555-labeled yogurt EVs. Approximately 100 pL of hydrogel (n = 3) was injected into the bottom of a 5 mm Wilmad NMR sample tube (Sigma Aldrich, Z272027). 300 pL of sterile PBS were added to the tube without disturbing the hydrogel surface, and the specimens were then incubated at 37 C in the dark. At the indicated time points, the supernatant was collected and replaced with fresh PBS (t = 2h, 4h, 6h, 8h, 24h, 48h, 3d, 4d, 5d, 6d, 8d, 12d). The collected supernatants were transferred to a black 96-well plate and analyzed using an Agilent BioTek Synergy Neo2 Multi-Mode Microplate Reader (Agilent)18.

[0137] In vivo hydrogel biocompatibility studies BALB / cJ mice were used to assess possible adverse effects of the yogurt EV hydrogels in vivo. Mice were subcutaneously injected in the hind flank with a 50 pL volume of either a 3 wt% HPMC-C16 and ~10n / mL yogurt EVs hydrogel(1010EVs per injection) or a bolus EV injection (1010yogurt EVs-same as hydrogel concentration). Mice were monitored daily for signs of skin irritation and weight loss or any signs of morbidity.

[0138] Day 7 explanted hydrogels were embedded in OCT compound (Tissue-Tek) and cryopreserved by rapid freezing in pre-cooled isopentane within a liquid nitrogen bath. Samples were stored at -80°C until cryosectioning. Frozen specimens were sectioned at 12 pm thickness and mounted on glass slides. Slides were brought to room temperature (RT) for 30 minutes and hydrophobic barriers were drawn around samples using a PAP pen. Sections were fixed in 4% paraformaldehyde for 20 minutes at RT, then washed twice with PBS-Tween. Fc receptor blocking was performed using BlockAid (Invitrogen, Bl 0710) for 2 hours at RT or overnight at 4°C in a humidified chamber.

[0139] For intracellular staining, slides were permeabilized with 0.3% Triton X-100 in PBS for 30 min at RT, followed by a 30-minute block with 5% FBS in PBS. For sections stained with biotinylated antibodies, endogenous biotin was blocked using a streptavidin-biotin blocking kit (Invitrogen, R37628) according to the manufacturer’s protocol. Primary antibodies were diluted in 5% FBS in PBS and applied to slides for 2 hours at RT or overnight at 4°C in a dark humidified chamber. For FoxP3 staining, 0.1% Triton X-100 was included in the primary antibody cocktail. Slides were then washed with PBS-Tween, followed by incubation with secondary detection reagents diluted in 5% FBS in PBS for 1 hour at RT in the dark. All antibodies, fluorophores, clones, and working dilutions are listed in Table 4. After washing, slides were briefly rinsed in PBS and mounted using ProLong Gold antifade reagent (with or without DAPI; Invitrogen, P10144 / P36941), then sealed with glass coverslips. Slides were cured overnight at RT in the dark and stored at 4°C until imaging. Confocal imaging was performed using a Nikon AXR resonant scanning confocal microscope at 20* magnification.

[0140] Table 4. Antibody panel used for immunofluorescenceTarget Fluorophore Clone Type Vendor Catalog Final # Dilution CD31 AF647 390 Primary BioLegend 102415 1:200CD34 - / Biotin MEC14 Primary BioLegend 119303 1:100 .7Biotin DyLight488 — Secondary BioLegend 405129 1:100 F-actin TRITC — Primary Invitrogen R37112 1 drop CD3 AF647 17A2 Primary BioLegend 100209 1:200 FoxP3 - / Rat IgG FJK- Primary Invitrogen 1457738 1:10016s 2Rat IgG Cy3 — Secondary Jackson 7121651 1:400 I.R. 53CDllc AF594 N418 Primary BioLegend 117356 1:100 CD68 AF647 FA-11 Primary BioLegend 137003 1:200 DNA DAPI — Mounting Invitrogen P36941 1 drop Media

[0141] For histological analysis, 12 gm cryosections were sent to the Molecular Pathology Core at Columbia University Irving Medical Center (CUIMC) for hematoxylin and eosin (H&E) staining using standard protocols.

[0142] Flow Cytometry Sample Preparation and Staining Day 7 explanted tissues were processed for spectral flow cytometry. Tissues were first mechanically dissociated using Kimble BioMasher II tissue grinders (749625-0020), followed by enzymatic digestion in DMEM containing DNase I (lOOpg / mL; Sigma-Aldrich, 10104159901) and Collagenase / Dispase (Img / mL; Sigma- Aldrich, 10269638001). Digestion was performed at 37°C for 40 minutes with constant agitation on a shaker.

[0143] Cell suspensions were filtered through a 70pm strainer into a 5 mb polystyrene roundbottom tubes (Falcon, 352235) and centrifuged at 4°C for 10 minutes. For spleen samples, red blood cells were additionally lysed with ACK lysis buffer for 10 minutes at room temperature. Supernatants were discarded, and cell pellets were transferred to a U-bottom 96-well plate.

[0144] Cells were stained with Zombie NIR viability dye (BioLegend, 423105; lOOpL at 1 : 1000 in PBS) for 20 minutes at RT, followed by one half-volume (100 pL) and one full volume (200 pL) wash with FACS buffer (PBS with 2% FBS). Fc receptors were then blocked using 50 pL of Fc Block diluted 1:100 in FACS buffer (Biolegend TrueStain FcX, 423105) for 20minutes at 4°C . Cells were then stained with an antibody cocktail, with each antibody diluted 1:100 in FACS buffer. A total volume of 50 pL was added per well, resulting in a final antibody dilution of 1:200. Staining was performed for 30 minutes at 4°C. After one half-volume and full-volume wash, cells were fixed in 4% paraformaldehyde for 10 minutes at RT, washed twice, and resuspended in FACS buffer. Samples were analyzed the following day using a Sony ID7000 Spectral Flow Cytometer. A complete list of antibodies is provided in Table 5.

[0145] Table 5. Antibody panel used for spectral flow cytometryTarget Fluorophore Clone Vendor Catalog # Final Dilution Live / Dead NIR — BioLegend 423105 1:1000 CD45 BV510 30-F11 BioLegend 103137 1:200 CD3 PE 17A2 BioLegend 100205 1:200 CD4 AF700 GK1.5 BioLegend 100429 1:200 CD8 BUV395 53-6.7 BioLegend 104011 1:200 CD19 BV421 6D5 BioLegend 115537 1:200 CDllc APC N418 BioLegend 117309 1:200 CDllb AF488 MI / 70 BioLegend 101219 1:200 F4 / 80 RB705 T45-2342 BDBios 570289 1:200 Ly6G APC-Cy7 1A8 BioLegend 127623 1:200

[0146] Immunofluorescence (IF) The 12 pm hydrogel sections were prepared by thawing slides at room temperature (RT) for 30 minutes and drawing hydrophobic barriers around the samples using a PAP pen. Slides were fixed in 4% paraformaldehyde at RT for 20 minutes and washed twice in wash buffer (PBS-Tween). For blocking, FC block was applied to the samplesand incubated at RT for 2 hours or overnight at 4°C, ensuring the samples remained hydrated. A streptavidin-biotin blocking kit was used to minimize background from endogenous biotin. Slides were incubated with streptavidin solution for 15 minutes, rinsed and incubated with biotin solution for 15 minutes, and then rinsed again with wash buffer. Primary antibody staining was performed by preparing a cocktail at a 1:200 dilution of: Alexa Fluor® 647 anti-mouse CD31, Biotin antimouse CD34, ActinRed 555 ReadyProbes (Invitrogen) in 5% FBS FACS buffer. 100 microliters of the cocktail was applied to each section, and slides were incubated in a dark, humidified chamber for 2 hours at RT. Slides were washed and the anti-CD34 secondary antibody was prepared at a 1:100 dilution and applied to the sections, followed by incubation in a dark, humidified chamber for 1 hour at RT. Slides were washed and quickly rinsed with PBS. Excess PBS was gently removed without disturbing the tissue. Mounted tissue sections (with DAPI) were allowed to dry in the dark at RT, then stored at 4°C until imaging. Stained slides were imaged using a Nikon AXR resonant scanning confocal microscope at a 20x magnification. This process was repeated for immune cell staining, but staining was instead done with anti-mouse CD68 (Biolegend).

[0147] Statistical methods Statistical comparisons were made in the GraphPad PRISM statistical software using either paired T tests, or a one-way or two-way ANOVA, depending on the nature of the dataset. In the case of multiple comparisons, the false discovery rate was controlled at 5% for multiple comparisons using the Benjamini, Krieger, and Yekutieli method113. For non-parametric analysis (Mander’s co-localization coefficient), comparisons were made using a Kruskal-Wallis test with Dunn correction for multiple comparisons. The frequency dependence of G' was determined by performing a nonlinear regression fit to the experimental data, where G' was modeled as a power law function of frequency (G1~ co11), and the exponent (n) was extracted from the slope across the measured frequency range; a 95% confidence interval for the slope is reported.RESULTS

[0148] Isolation and concentration of yogurt EVs for use in supramolecular hydrogel synthesis

[0149] A significant challenge in developing supramolecular EV hydrogels lies in the large quantity of EVs required (~10n-1012EVs per mL of hydrogel). Traditional, cell-derived,approaches to isolating EVs often yield insufficient amounts (108-l O10EVs per 106cells52), making it difficult to produce and optimize EV biomaterials or explore their potential applications. To address this, the present disclosure turned to an alternative, scalable source: bovine milk yogurt-derived EVs (yogurt EVs), which can be obtained in large quantities from milk whey, a byproduct of yogurt production46.

[0150] Yogurt EVs were isolated from whey obtained from strained yogurt via tangential flow filtration (TFF) using a two-step protocol. TFF was used instead of the conventional ultracentrifugation techniques used in EV isolation29because TFF achieves higher yields, induces less mechanical stress on EVs, which preserves their morphology, and scales from bench to industrial processes29'52 54. In the first step of isolation, whey volume was processed through a hollow fiber with an average pore size of 650 nm to remove cell debris. The permeate from this step was then passed through a hollow fiber with a 100 kDa cut-off, which retained EVs in the retentate while eliminating smaller soluble molecules such as proteins. The yield for yogurt EVs, determined by calculating the number of particles per unit volume of the initially processed whey, was approximately 10'° parti cles / mL feed (with around 1 L for a typical feed volume), compared to cell culture-derived EVs, which yield approximately 108EVs / mL of cell culture media using TFF as isolation method52, highlighting a 100-fold scalability advantage of whey as a source.

[0151] Following their isolation, EVs were lyophilized and concentrated by approximately 10-fold for incorporation into supramolecular hydrogels. Generating supramolecular polymer-nanoparticle hydrogels requires large amounts of concentrated nanoparticles, typically achieving final nanoparticle weight percentages in the hydrogel ranging from 1% to 10% by weight3,18,38. For reference, making a 4 wt% liposome hydrogel (1 m ) requires approximately IxlO12liposomes based on NTA analysis. This poses specific challenges when working with EVs due to the high concentrations required. In this study, lyophilization was the primary method to concentrate vesicles for incorporation into materials. This approach also facilitated long-term storage and facile mass determination of yogurt EVs for their incorporation into supramolecular hydrogels.

[0152] Both immediately after isolation and concentration, yogurt EVs were characterized through nanoparticle tracking analysis (NTA), dynamic light scattering (DLS), transmission electron microscopy (TEM), Cryo-electron microscopy (CryoEM), bicinchoninic acid (BCA) assays, and western blot, in order to confirm the nanovesicles’ properties and protein markerexpression according to the minimal criteria defined by the International Society for Extracellular Vesicles (ISEV)29.

[0153] The TFF isolation protocol yielded yogurt EVs with a uniform size and charge distribution, and lyophilization proved to be an effective method for concentrating the EVs while maintaining their properties. NTA showed the mean particle concentration to be 1.9 * 1011± 4.4 x IO10parti cles / mL for yogurt EVs and 1.0 x 1013± 3.3 x 1012parti cles / mL for lyophilized yogurt EVs (p = 0.0061) (Figure la). The mean protein concentration found through BCA assay was 4.1 ± 3.0 mg / mL for yogurt EVs and 31.6 ± 16.1 mg / mL for lyophilized yogurt EVs, also indicating a 10-fold concentration (Figure 11). NTA results showed a mode diameter of 139.2 ± 8.1 nm for yogurt EVs and 158.0 ± 5.8 nm for lyophilized yogurt EVs (p = 0.1113) (Figures la-b), and DLS measurements indicated a Z-average diameter of 148.1 ± 9.0 nm (n =9, 0.22 ± 0.02) for yogurt EVs and 170.19 ± 9.58 nm (n =9, PD1 = 0.27± 0.02) for lyophilized yogurt EVs (p = 0.0967) (Figure lb). Both yogurt and lyophilized yogurt EVs exhibited a negative surface charge, with mean zeta potentials of -36.09 ± 2.25 mV and -28.84 ± 1.45 mV, respectively (Figure 1c) indicating that the anionic charge characteristic of EVs remains after lyophilization, and that overall colloidal stability is preserved. To confirm that EVs retain biofunctional molecules during and after processing, a qualitative immunoblot was performed to confirm retention of CD81 (Figure Id), which is a tetraspanin enriched in EVs and is recognized by the ISEV as an EV marker29.

[0154] EVs were then characterized using both transmission electron microscopy (TEM) and Cryo-electron microscopy (CryoEM) to ensure EV morphology was preserved during isolation and after concentration. TEM images confirmed the characteristic cup-shaped morphology of both yogurt EVs and lyophilized yogurt EVs (Figure le), and Cryo-EM enabled visualization of a halo around vesicles, suggesting the presence of the EV protein corona29in both samples (Figure le, Figures 12-13). These findings indicate that the isolation and lyophilization processes maintain EV structural integrity, ensuring their suitability for downstream applications.

[0155] These results confirm the effectiveness of TFF for EV isolation and demonstrate that lyophilization preserves yogurt EV stability, yielding a product that aligns with reported EV properties29, including uniform size distribution, negative surface charge, and vesicular morphology. This is consistent with prior studies demonstrating TFF’s utility for isolating EVsfrom diverse sources, including human and murine cell lines55, engineered cell lines56, bovine milk46, and microalgae57. While ultracentrifugation remains a gold standard, it is limited by low yield, scalability, and time consumption. In contrast, TFF is scalable, flexible, and suited for continuous operations52. These findings further support that TFF is an effective method to recover naturally released EVs with uniform size and charge from bovine milk whey at large scales, thereby addressing limits in EV isolation for biomaterials development. While TFF enables high-throughput isolation and preserves vesicle morphology, it may not offer the highest possible purity. Combination of TFF with SEC-HPLC can provide improved resolution of vesicle populations.

[0156] Overall, it was shown that yogurt EVs are well preserved by lyophilization, aligning with prior reports that milk EVs are stable after freeze-thaw cycles50,58. However, there remains some disagreement as to whether lyophilization can destabilize EVs by changing their morphology, functionality, particle size, or concentration59. The best methods for lyophilizing EVs also remain under debate, with ongoing discussions about the use of cryoprotectants60. In the present disclosure, lyophilized EVs showed a minor reduction in zeta potential and a slight increase in diameter, indicating lower surface charge and possible changes in electrostatic stabilization. Cryo-EM also revealed minor membrane disruption, though these did not appear to significantly alter particle counts by NTA, suggesting this affects only a subset of vesicles, as no distinct population of small fragments or aggregates was observed. These data indicate that lyophilized EVs can be used quickly after reconstitution but that repeated lyophilization cycles should be avoided.

[0157] While the present disclosure indicates lyophilization to be useful for EV hydrogel preparation for certain types of EVs, more work is needed to understand how EV identity may dictate optimal processing. Overall, using TFF for isolation and lyophilization for concentration, the present disclosure developed a scalable method to produce highly concentrated EV solutions, enabling a systematic optimization of design principles for supramolecular EV hydrogels18,38.

[0158] Polymer identity determines yogurt EV hydrogel gelation and mechanical properties

[0159] Using scalable yogurt EVs to carry out high throughput EV hydrogel synthesis, the present disclosure conducted a systematic study to define the role of polymer identity on EV hydrogel mechanical properties. Previous work on liposomal hydrogels used dodecyl modifiedhydroxypropyl methylcellulose (HPMC-C12)38, thus the present disclosure focused on varying the alkyl chain length (HPMC-Cx: C12, C14, and C16) as well as the degree of functionalization (0.52 mmol / g, which is consistent with previous studies3,38, or 0.32 mmol / g). Overall, the present disclosure determined these parameters were important for efficiently generating a robust supram olecular hydrogel that exhibits both consistent solid-like and self-healing properties.

[0160] To determine the minimum number of EVs required for gelation, the present disclosure prepared a series of mixtures with each polymer using iteratively less EV (3 wt% and 0.52 mmol / g modified HPMC-Cx; Figure 2a). Successful gelation was defined as the lack of a crossover of storage modulus (G') and loss modulus (G") over the studied frequency range of 0.1 to 100 rad / s. Unlike previously reported liposomal hydrogels, which exhibited robust mechanical properties when mixed with HPMC-C1238, the present disclosure found that longer alkyl chains (i.e. HPMC-C14 or HPMC-C16) were needed to form EV hydrogels with consistent solid-like properties and robust self-healing behavior. Notably, EV hydrogels prepared with HPMC-C12 required an order of magnitude increase in the number of EVs ( 1012EVs / mL of hydrogel) to form a hydrogel that does not exhibit a crossover (Figures 2a-c, Figures 20-21). However, although it resulted in hydrogel formation, using very high numbers of EVs in hydrogels made with HPMC-C12 resulted in deteriorating self-healing behaviors as shown by viscosity plateaus decaying over multiple stepshear cycles between high and low shear (Figures 20, 23). Specifically, viscosity decreased by approximately 50% after the first two cycles, and 30% after the third, illustrating a progressive and substantial loss in the regeneration of the hydrogel mesh network at these timescales. Mixing HPMC-C12 and 1011EVs did not form stable hydrogels, marked by low-frequency crossovers between their storage and loss moduli (Figure 21). In contrast, HPMC with longer alkyl groups (C14 and C16) formed stable hydrogels using 7-fold and 50-fold fewer EVs, for C14 and C16 respectively (Figures 2a-c, Figure 22).

[0161] These findings suggest that crosslinking is driven not only by the presence of hydrophobic groups but also by the kinetic stability of their interactions with the EV membrane. Prior studies in polymer-nanoparticle (PNP) systems support that increasing hydrophobicity enhances binding strength, stabilizes loop and bridge formation, and slows the dissociation of polymer-nanoparticle interactions. In this context, HPMC -C 12 likely forms contacts that are too transient to re-form efficiently during recovery window, resulting in poor self-healing. The poorgelation efficiency and the progressive loss of viscosity in step-shear tests further support a kinetic limitation for C12. By contrast, C14 and C16 chains provide stronger and more persistent interactions, enabling dynamic but stable network reformation after shear. The enhanced gelation efficiency with longer alkyl chain polymers suggests that optimizing polymer structure can reduce the quantity of EVs required for hydrogel formation, which may be particularly advantageous when working with limited or high-value EV sources.

[0162] Longer carbon chains also resulted in stiffer and more solid-like hydrogels, which is likely due to more stable alkyl chain insertion into the EV membrane, thus strengthening polymer-EV interactions. Rheology data for hydrogels containing 3 wt% HPMC-Cx and ~1012EVs / mL supports this hypothesis, as hydrogel storage moduli significantly increase between HPMC-C12 and HPMC-C16 (from 267.62 ± 37.38 Pa to 668.72 ± 31.47 Pa; q<0.0001) as well as between HPMC-C14 and HPMC-C16 (from 311.45 + 27.18 Pa to 668.72 + 31.47 Pa; qO.OOOl) (Figure 2b). The present disclosure observed no significant increase in stiffness when increasing the alkyl chain length from 12 to 14 carbons (q = 0.0574). Similar trends were observed when comparing solid-like properties of the EV hydrogels, as tan 8 at 10 rad / s is significantly lower (i.e., more solid) for hydrogels formed with EVs and HPMC-C16 than both HPMC-C12 (0.362 ± 0.006 vs. 0.651 ± 0.032; q = 0.0002) and HPMC-C14 (0.643 + 0.034 vs. 0.651 + 0.032; q = 0.0002) (Figure 2c).Although there is no significant difference between the land values of HPMC-C12 and HPMC-C14 hydrogels, utilizing HPMC-C14 consistently produced gel-like materials (e.g., no phase transitions in the tested frequency range) with robust self-healing behavior (Figure 22), while HPMC-C12 did not (Figures 2a-c, Figures 20-21).

[0163] The present disclosure formulated EV hydrogels using 3 wt% HPMC-C18 (0.52 mmol / g modification), which revealed an upper limit to the benefit of alkyl chain length (Figures 29-30). While C18 led to overall stronger hydrogels, closer inspection of polymer-only control samples revealed that HPMC-C18 forms a robust hydrogel by itself, likely due to efficient hydrophobic interactions between the longer C18 chains. However, unlike HPMC-C14 or HPMC-C16, the mechanical properties of the system are not greatly enhanced upon addition of high numbers of EVs (1012 / mL). In fact, at lower levels (1010to 1011), the addition of EVs drastically decreases in the storage and loss moduli of the material by as much as 65% (Figure 29a). Theseresults suggest that EVs are disruptive of the stronger C18-C18 interactions of this hydrogel and indicate a narrow optimal window of alkyl chain length for EV hydrogels.

[0164] Building on the observation that hydrophobic interactions drive EV gel crosslinking, the present disclosure hypothesized that hydrogel stiffness could also be tuned by varying the degree of hydrophobic modification of HPMC. The present disclosure examined this by comparing hydrogels formulated with 0.32 mmol / g modified HPMC-C16 to 0.52 mmol / g modified HPMC-C16. As expected, hydrogel stiffness increased with greater functionalization, as indicated by higher storage moduli for 0.32 mmol / g vs. 0.52 mmol / g HPMC-C16 (G’ = 363.01 ± 6.53 vs.668.72 ± 31.47; p = 0.0007) (Figure 2d). However, higher functionalization did not significantly affect solid-like behavior, with no significant difference in loss modulus (G” = 174.96 ± 24.81 vs.242.02 ± 10.34; p = 0.0671) and therefore tano (0.480 ± 0.061 vs. 0.362 ± 0.006; p = 0.1281) (Figure 2e). The greater effect on stiffness than viscous dissipation aligns with previous findings on liposomal hydrogels38and provides a strategy for controlling the storage modulus (G) in EV hydrogels with minimal impact on the loss modulus (G").

[0165] Unlike liposomal hydrogels, which exhibit robust gelation across a range of polymer formulations38, EV hydrogels display a greater sensitivity to polymer identity, likely due to differences in EV membrane composition relative to synthetic vesicles. Unlike synthetic liposomes, which consist of uniform lipid bilayers, EV membranes incorporate a diverse array of proteins, glycans, and lipid species29that can hinder their interactions with amphiphilic polymers. These features may impact crosslinking efficiency, steric interactions, and membrane stability, necessitating more precise polymer engineering to achieve optimal hydrogel properties. For example, while liposomal hydrogels readily formed stable networks with HPMC-C12, EV hydrogels required longer alkyl chain modifications (HPMC-C16) to achieve comparable mechanical properties. This suggests that EV-polymer interactions are more dependent on hydrophobic insertion mechanics and membrane fluidity.

[0166] The development of supramolecular hydrogels crosslinked by EVs has been met with conflicting reports, highlighting the need for a fundamental understanding of the material properties governing their formation. Prior efforts have either failed to generate hydrogels solely crosslinked by EVs or reported EV hydrogel formation under conditions suggest such systemswould be unstable39,40. The results in the present disclosure help to overcome these issues by design parameters necessary for EV-mediated hydrogel formation and stability.

[0167] For example, a prior report attempted to synthesize supramolecular EV hydrogels using HPMC-C12 and HPMC-C18, but found that EVs alone could never effectively crosslink their system39. Achieving stable hydrogels required liposomes as co-crosslinking agents. This inability in the art to form an exclusively EV-crosslinked hydrogel likely stems from two key limitations. First, this approach relied on mesenchymal stem cell-derived EVs, which are more challenging to isolate in high yield, preventing study of materials with more than IO10EVs / mL hydrogel. This highlights an advantage of the present disclosure’s approach, which leverages low-cost, scalable yogurt EVs, allowing for the systematic exploration of EV hydrogel design principles. Second, the findings indicate that low levels of EVs can destabilize C18 systems, which may explain why liposomes were required to restore stability even when working with a longer alkyl chain.

[0168] In contrast, a recent report described successful EV hydrogel formation using HPMC-C1240, which contrasts with the present disclosure’s findings that C12-modified polymers are in fact insufficient for robust EV-based gelation. It was suspected that this discrepancy arises from the extrusion of EVs through 50 nm membranes, a process that not only reduces particle size, but also significantly increases the number of individual EV particles. Although the previous study did not report the total EV concentration in their system, the present disclosure’s findings suggest that such an approach could yield EV numbers as high as 1012, a regime where gel-like behavior during frequency sweep rheology was also observed. However, the present disclosure showed HPMC-C12 EV hydrogels in facxt show undesirable self-healing properties (Figures 20, 21, 23), revealing that is critical to assess multiple rheological properties of these materials when choosing their design space.

[0169] Another important consideration that remains to be addressed by the field is the contribution of the EV surface. It is possible that extrusion, which is used in some of the prior reports, may disrupt key molecular features of the EV surface, particularly the protein corona, which may play a role in steric interactions and crosslinking. If extrusion removes sterically hindering surface proteins, the resulting EVs may behave more similarly to liposomes, potentially facilitating hydrogel formation in HPMC-C12 systems where non-extruded EVs fail. This suggeststhat EV surface modifications — intentional or unintentional — can strongly influence hydrogel assembly and should be carefully characterized when comparing across studies.

[0170] The findings in the present disclosure demonstrate that polymer identity is a key parameter for tuning the mechanical properties of EV hydrogels, enabling formulations that range from C14 systems (G' -219 Pa, G" -156 Pa) to HPMC-C16 systems (G -668.72 Pa, G" -242.02 Pa). This tunability is critical for biomedical applications, where hydrogel mechanics influence cell behavior, tissue integration, and therapeutic performance. In tissue engineering, stiffness affects cell proliferation and differentiation61 64. while in drug delivery, it modulates release kinetics65. For cell-based therapies, viscoelastic properties regulate cell migration and retention62’63. By leveraging polymer structure as a design tool, the present disclosure establishes a framework for optimizing EV hydrogels for regenerative medicine, controlled drug release, and engineered cell scaffolds.

[0171] Polymer concentration can synergistically improve EV hydrogel mechanical properties

[0172] To further investigate the mechanical properties of EV hydrogels, the present disclosure evaluated how polymer concentration influenced the storage modulus (G) of HPMC-C16-based hydrogels at 10 rad / s, a frequency commonly used in rheological characterization of biomaterials. The present disclosure formulated hydrogels with either 2 wt% (Figure 26) or 3 wt% HPMC-C16, both with and without -1012yogurt EVs per mL. As expected, increasing polymer concentration from 2 wt% to 3 wt% HPMC-C16 resulted in a stiffer hydrogel due to increased polymer-polymer interactions (2 wt%: G = 89.96 ± 15.79 Pa; 3 wt%: G = 230.95 ± 23.56 Pa). Similarly, adding EVs to 2 wt% HPMC-C 16 increased stiffness (2 wt% + EV: G' = 150.07 ± 24.81 Pa), consistent with increased EV-polymer interactions and crosslinking.

[0173] However, when EVs were added to the 3 wt% HPMC-C16, an unexpectedly large increase in (668.72 ± 31.47 Pa) was observed — exceeding what would be expected from the additive contributions (G’eXpected=291.06 ± 37.68 Pa) of increasing polymer concentration and adding EVs separately (Figure 3a-b). This was consistent across all frequencies tested (Figure 27). Furthermore, examining these stiffness values using an interaction plot (Figure 3c), it was observed that when increasing polymer wt%, G’ at 10 rad / s increases at a different rate dependent on EV presence, suggesting that EVs enhance polymer-polymer interactions (Figure 3d). Thisnonlinear increase in stiffness indicates a cooperative relationship between polymer concentration and EVs in stabilizing the hydrogel network.

[0174] Table 6: Addition of yogurt EVs to HPMC-C16 hydrogels results in lower G’ frequency dependence and deviations from the Rouse behavior of pure polymer matrices. Hydrogel Formulation Frequency Dependence (G* ~ 95% Confidence Interval (Dx)2 wt%HPMC-C16 0.43 (0.39, 0.48)2 wt% HPMC-C16 and 0.33 (0.30, 0.36)EVs3 wt%HPMC-C16 0.31 (0.29, 0.33)3 wt% HPMC-C16 and 0.25 (0.24. 0.26)EVs

[0175] Frequency-dependent rheology further supports the observed nonlinear mechanical trends and suggests that multiple physical factors may contribute to the enhanced stiffness (Table 6). Hydrogels formed with 2 wt% HPMC-C16 exhibit frequency-dependent behavior (G’ ~ to043), similar to Rouse-like polymer dynamics (G’ ~ OJ°3), indicative of a weakly connected network dominated by polymer segment relaxation66,67. The addition of either polymer (1 wt% HPMC-C16) or EVs (IO11’12EVs) reduces frequency dependence (G’ ~ co031and G’ ~ co033, respectively), consistent with increased network connectivity. The combination of both polymer and EVs further reduces frequency dependence (G' ~ co023), suggesting a transition toward a more elastic, solidlike network. While these changes are consistent with increased supramolecular crosslinking, the progressive reduction in frequency dependence with increasing polymer and EV content may also reflect other physical influences on hydrogel network formation, like restricted chain mobility or steric confinement, that emerge at higher formulation densities. These observations raise the possibility that physical crowding effects, in addition to polymer-EV interactions, may contribute to mechanical reinforcement in specific formulations.

[0176] To further probe the mechanism underlying EV-polymer interactions, the present disclosure conducted DLS-based polymer corona experiments66(Figures 3e-f). When yogurt EVs were incubated with HPMC-C12 or HPMC-C16 under dilute conditions, significant increases in hydrodynamic diameter was observed, indicating the formation of larger EV-polymer assemblies.In contrast, unmodified HPMC did not induce a comparable shift, resulting in a broad bimodal distribution and suggesting that alkyl modification is required for this interaction. Notably, the hydrodynamic diameter of EVs increased more substantially with HPMC-C16 (251.6 ± 23.26 nm increase, q < 0.001) than with HPMC-C12 (142.94 ± 44.65 nm increase, q = 0.0014), consistent with enhanced association driven by longer hydrophobic side chains. These data support the hypothesis that polymers with alkyl chains interact with the EV lipid bilayer to form supramolecular complexes. This behavior is consistent with prior reports of hydrophobically driven polymer-nanoparticle corona formation66. Because these interactions occur under dilute, non-gelling conditions, they are unlikely to result from crowding. Instead, the data suggest that hydrophobic post-insertion or membrane adsorption may contribute to polymer-EV association and crosslinking within the hydrogel.

[0177] Overall, the results in the present disclosure in context with prior work highlight the importance of both hydrophobic polymer modification and scalable EV production in formulating EV and HPMC-Cx hydrogels with robust and tunable mechanical properties. These systematic studies reveal an optimal window for formulating EV hydrogels using linear alkyl modified HPMC, demonstrating how the parameters of alkyl chain length, degree of functionalization, and polymer concentration each contribute to mechanical properties of this system. These insights indicate that alkyl length is perhaps most important for improving the efficiency of gelation (e.g., stable gel formation using fewer EVs), while degree of functionalization and polymer concentration are better parameters for regulating the stiffness and viscoelastic behaviors of the system.

[0178] EVs contribute mechanically to hydrogels through crosslinking and crowding effects

[0179] Having defined the polymer properties needed to form EV hydrogels, the present disclosure then set out to further confirm the central hypothesis that EVs behave as a dynamic crosslinker in these systems. Despite rheological differences seen between EV and liposomal hydrogels, the present disclosure reasoned that the presence of a lipid bilayer in both species enables reversible insertion of the alkyl groups to form a type of physical crosslink38,68’69.

[0180] To test this hypothesis, the present disclosure evaluated the change in mechanical properties caused by the addition of EVs to a solution containing HPMC-C16. As anticipated, the addition of EVs to HPMC-C16 generated stronger hydrogels consistent with the formation of newcrosslinks in the network. Using a solution of HPMC-C16 that is sufficiently modified to form a weak hydrogel (3 wt% of 0.52 mmol / g modified polymer), it was observed that addition of ~1012yogurt EVs enhanced the storage modulus G’ ~3-fold (from 230.95 ± 23.56 Pa to 668.72 ± 31.47 Pa; q = <0.0001; Figure 4a) and loss modulus by ~2-fold (from 112.92 ± 11.48 Pa to 242.02 ± 10.34 Pa; q = <0.0001). Notably, EVs fail to form hydrogels when mixed with unmodified HPMC. These data indicate that the EVs strengthened the existing hydrogel network, consistent with the hypothesis that EVs form crosslinks through hydrophobic interactions with C16.

[0181] To more rigorously test whether EVs become structural elements of the hydrogel, the present disclosure examined their ability to crosslink a polymer solution that lacks an existing stable gel structure. The present disclosure achieved this by reducing the degree of C16 modification in HPMC-C16 (0.32 mmol / g modified, 3 wt% solution) to weaken polymer-polymer interactions, creating a system that does not form a stable hydrogel on its own, as confirmed by a crossover observed in the frequency sweep (Figure 4b). Consistent with the hypothesis, addition of ~1012yogurt EVs stabilized this weaker network, eliminating the crossover over the tested range and producing a hydrogel. Again, addition of EVs to unmodified HPMC failed to generate a hydrogel. Combined, these experiments provide evidence that EVs, like liposomes, can crosslink certain hydrophobically modified polymers to improve their stiffness and solid-like properties over a range of frequencies.

[0182] To further validate the role of EVs as crosslinkers, the present disclosure varied the number of EVs in the hydrogel and evaluated the effect on hydrogel stiffness. The present disclosure found that increasing the number of nanoparticles per mL of hydrogel, from approximately 1010to 1012EVs / mL, linearly increased both the storage modulus (r2= 0.78, p<0.0034) and loss modulus (r2= 0.62, p<0.0206) of the resulting hydrogels (Figure 4c). The effect size was much greater on the storage modulus compared to the loss modulus. This likely reflects EVs' stronger reinforcement of the hydrogel’s elastic crosslinking network (G1) compared to viscous dissipation (G"), which depends more on polymer mobility70. In addition to the increase in stiffness, the present disclosure saw a significant shift to more solid-like properties with increasing EV number, characterized by a decrease in tanb = G” / G’ (r2= 0.61, p = 0.0215, Figure 14). In previously reported liposome system, increasing the number of nanoparticle crosslinkers significantly increased both G’ and G”, which is consistent with these findings38. Increasing EVnumber also yielded a dose-dependent increase in G' and G" at 10 rad / s in the lower DoF HPMC-C16, despite the polymer not forming a gel on its own (Figure 31).

[0183] Injectable hydrogels must demonstrate dynamic crosslinking behavior, disassembling under shear forces (shear-thinning) and reassembling rapidly once shear is removed (self-healing)71. To demonstrate the injectability of EV hydrogels, the present disclosure conducted step shear and steady shear rheology. These measurements revealed that EV hydrogels exhibited shearthinning behavior, with viscosity decreasing ~3 orders of magnitude as shear rate increased (Figure 4e). EV hydrogels also exhibited self-healing behavior, returning to their original viscosity within 60 seconds after applying high shear and maintaining this over multiple cycles (Figure 4f, Figures 24-25). This behavior was observed when injecting the hydrogel through a 1-mL syringe and 26G needle, after which the gel promptly resolidified (Figure 4d). The robust mechanical properties observed in the frequency response were also consistent with strain-dependent oscillatory rheology data showing that G and G" remain constant in the linear viscoelastic regime but begin to decrease in the nonlinear regime during yielding, the point at which the hydrogel begins to shift into a liquid state. The present disclosure observed yield strains around 80-100% (100.59 ± 13.61%) (Figure 4g), which are consistent with previously reported supramolecular systems3,38,72.

[0184] These findings demonstrate that yogurt EVs form dynamic and reversable crosslinks with C16 modified cellulose, formulating a robust injectable hydrogel. Because these hydrogels utilize EVs as the sole nanoparticle crosslinker instead of a synthetic nanomaterial, these findings have important implications in generating hydrogels featuring the numerous signaling molecules intrinsic to EVs.

[0185] While data support a role for EVs as dynamic crosslinkers via hydrophobic membrane association with alkyl-modified polymers, the present disclosure also observed evidence of additional mechanisms contributing to gelation, such as crowding-related stiffening and potential EV jamming-like behavior in more polymer-rich formulations. For instance, in HPMC-C16 systems capable of gelation on their own, increasing EV number led to both an increase in stiffness and a reduction in yield strain (Figure 32), a hallmark of j mming-driven stiffening73. In contrast, in lower DoF HPMC-C 16 formulations that do not gel independently, increasing EV concentration increased stiffness without reducing yield strain (Figures 31-32), which is more consistent with acrosslinking-driven mechanism. However, rheological behavior across all tested systems remained frequency-dependent and did not exhibit the flat G' responses which would be expected in nanoparticle jamming-dominated materials73. The present disclosure also did not observe uncaging behavior, which would be indicated by a G” overshoot during the yield transition. This behavior is typically associated with jammed networks, where jammed polymer-nanoparticle units locally yield and dissipate stress66,74,75. This absence may reflect either a lower proportion of jammed particles or indicate that while jammed particles are present, yielding in materials is dominated by more gradual mechanisms, such as dynamic polymer-EV bond dissociation, that mask or dampen the emergence of a distinct G" peak. This interpretation is further supported by the present disclosure’s polymer corona experiments, which demonstrated increased hydrodynamic diameter upon mixing EVs with hydrophobically modified HPMC, indicating polymer adsorption to the vesicle surface (Figure 3e-f). These findings suggest that supramolecular EV-polymer interactions likely play an important role in hydrogel formation and crosslinking kinetics. However, the present disclosure also see formulation-dependent signs of physical crowding, particularly in stiffer and polymer-rich systems, highlighting the likelihood that a combination of interactions underlies the assembly and mechanics of these hydrogels.

[0186] Artificial cell-derived vesicles as supramolecular hydrogel crosslinkers

[0187] Following optimization of a supramolecular hydrogel formulation with yogurt EV model system, the present disclosure sought to determine the versatility of these design principles by exploring incorporation of different types of cell-derived nanovesicles. As previously mentioned, EVs can be useful in therapeutics but are usually difficult to produce in high yields. However, recent research indicates that artificial cell-derived vesicles (ACDVs), a type of EV mimetic produced artificially by mechanical cell disruption, can be a scalable solution29. For instance, use of a lipid extruder can generate over 20 times more ACDVs from an equivalent cell quantity while retaining a substantial array of parent cell-derived proteins and miRNAs76. Although these vesicles are not naturally secreted by cells, their similar structural properties to EVs and bioactivity29,77make them another promising scalable approach to obtain nanovesicles. The present disclosure generated ACDVs from E. coli Nissle, a probiotic bacterial strain, and B16F10 melanoma cells to determine if the present disclosure’s approach with yogurt EV hydrogels can extend to other biomedically relevant ACDVs. Gram negative bacterial-derivedvesicles78have potential in therapeutic and immunostimulatory applications due to their inherent immunostimulatory properties79,80, efficient protein display, and high delivery efficiency of cargo such as drugs or RNA molecules81 83, while mammalian cell-derived vesicles are being explored for their role in regenerative medicine and targeted therapies due to their compatibility and bioactivity in human systems44,84,85.

[0188] E. coli and melanoma ACDVs were isolated using TFF and extrusion or turbulent mixing and extrusion, respectively. Characterization of size, charge, and yields indicated that ACDVs were under 300 nm, negatively charged and generated at sufficient scales for hydrogel production (Figure 33). NTA mode diameters of 265.3 ± 67.7 nm for A. coli ACDVs and 140.2 ± 65.7 nm for melanoma ACDVs were obtained (Figure 5a). DLS measurements indicated a Z-average of 279.5 ± 40.4 nm (PDI = 0.2) forE. coli ACDVs and 174.8 ± 11.8 nm (PDI = 0.4) for melanoma ACDVs, and showed negative surface charges, with mean zeta potentials of -49.1 ± 4.9 mV and -38.9 ± 4.0 mV, respectively (Figure 5b). Due to the lack of standardized markers and validated reagents for bacterial vesicles, characterization of E. coli ACDVs was limited to size distribution measurements. To validate that melanoma ACDVs retain EV-specific protein, the present disclosure performed immunoblot analysis to confirm the presence of TSG101 in these ACDVs (Figure 5c), as this cell line does not express the CD81 protein (Figure 34). TSG101 is a key component of the Endosomal Sorting Complex Required for Transport (ESCRT-I)29and is an alternative marker commonly used to characterize EVs.

[0189] Supramolecular hydrogels were then prepared using analogous formulations to the yogurt EV model system: ACDVs were manually mixed with 5.33 mol% modified HPMC-C16 (3 wt% HPMC-C16, 4 wt% / ~1010’11ACDVs). Both types of ACDV hydrogels exhibit solid-like behavior with a consistently greater storage modulus (G’) than loss modulus (G’ ’) across the entire frequency domain, with no crossover and tan5<l (Figures 5d-e). ACDV hydrogels also exhibit robust self-healing properties during step-shear rheology, returning to baseline viscosity upon cessation of shear. During testing, the present disclosure observed a transient viscosity overshoot, a phenomenon that has been previously reported in supramolecular hydrogels86. This overshoot likely reflects temporary structural rearrangements upon shear cessation, such as polymer-nanoparticle reorganization, before the network fully stabilizes. Importantly, this behavior did not impair the hydrogel’s ability to recover its original mechanical properties. Similarly to the yogurtEV hydrogels, the present disclosure also saw a clear linear viscoelastic region for both ACDV hydrogels during strain sweep tests (yield strain -200% for both types of ACDV hydrogels).

[0190] Both artificially derived vesicle types also formed hydrogels comparable in stiffness to the yogurt EV hydrogels. The storage and loss moduli at 10 rad / s were G’ = 574.87 Pa and G” = 208.50 Pafor / • / coli ACDVs and G’ = 531.67 Pa and G” = 211.15 Pa for mammalian cell ACDVs (Figure 5d-g). This is consistent with similar yogurt EV formulations, which ranged in stiffness (G’ at 10 rad / s) from 230 to 726 Pa at 3 wt% HPMC-C16. Solid-like properties were similar to those of yogurt EVs as well, and both the bacterial and mammalian derived ACDV hydrogels were also greater in stiffness and more solid-like than the polymer alone (G’ = 230.95, G” = 112.92) (Figures 5f-h). These data are consistent with the present disclosure’s hypothesis that ACDVs contribute to network formation in the system, potentially through polymer-nanoparticle hydrophobic interactions in combination with physical crowding, in a similar manner to the yogurt EVs.

[0191] The findings of the present disclosure therefore indicate injectable hydrogels can also be formed using artificial cell-derived vesicles in addition to naturally secreted yogurt EVs, highlighting the versatility of the supram olecul ar EV hydrogel platform and its potential in a wider variety of biomedical applications through altering the identity of the EV crosslinker.

[0192] While the present disclosure’s system displayed important evidence of versatility, further investigation into the differences in crosslinking capabilities of nanovesicles from different sources would provide further insights into the design considerations for these hydrogels. The data indicate relatively consistent rheological properties, but it is possible that subtle yet important differences in hydrogel properties may exist due to changes in vesicle identity. These more subtle differences will require larger studies, which may be constrained by current yield limitations in EV manufacturing. Yogurt-derived EVs yielded -10- 100-fold more particles per mb than bacterial or mammalian ACDVs (Table 1), highlighting the practical advantages of agricultural sources for formulation development and higher-throughput studies such as these.

[0193] Supramolecular yogurt EV hydrogels drive angiogenesis

[0194] Rheological analysis confirmed the successful development of a yogurt EV-based injectable hydrogel. The present disclosure then sought to examine the material’s biocompatibility and bioactivity. Previous studies have demonstrated that EVs can mediate cell-to-cellcommunication and act as therapeutic agents in vivo30-31, and bovine yogurt EVs have been shown to be biocompatible drug delivery vehicles88. This led us to hypothesize that the present disclosure’s yogurt EV hydrogels would not induce adverse effects in vivo. Moreover, the present disclosure anticipated that the hydrogel’s EV-mediated bioactivity would promote cellular infdtration, but unlike previously reported synthetic hydrogels, also drive dynamic biological processes. Specifically, given the proangiogenic potential of EVs demonstrated in previous studies40,89it was expected that hydrogel in the present disclosure to locally stimulate blood vessel formation in vivo.

[0195] Before evaluating bioactivity in vivo, the present disclosure first assessed the in vitro release profile and structural integrity of yogurt EVs following encapsulation in 3 wt% HPMC-C16 hydrogels. EVs were gradually released over two weeks, reaching approximately 65% cumulative release by day 12 (Figure 35a). Korsmeyer-Peppas modeling yielded n = 0.61 and K = 0.15, (Figure 35b), consistent with anomalous (non-Fickian) transport. This suggests that release is governed not only by diffusion but also by matrix relaxation, as expected in erosion of nanoparticle crosslinked systems. Nanoparticle tracking analysis and TEM confirmed that released EVs maintained size distribution and morphology comparable to pre-gel samples (Figure 35c-d), supporting preservation of vesicle integrity post-release.

[0196] Having confirmed sustained EV release and integrity in vitro, the present disclosure next evaluated the hydrogels’ in vivo performance. To determine biocompatibility in vivo, 50 pL of yogurt EV hydrogel (HPMC-C 16 (3 wt%) and -1011yogurt EVs) were injected subcutaneously into the hind flank of BALB / cJ mice. A control group of BALB / cJ was injected with a 50 pL bolus injection of 1011yogurt EVs. The present disclosure monitored the mice for signs of acute toxicity by tracking body weight for seven days. The body weights of the mice remained stable after hydrogel injection, with no significant differences (p = 0.3294) between the bolus and hydrogel injected groups (Figure 36). In addition, the mice exhibited no visible signs of illness or discomfort. These data indicate that yogurt EVs, whether administered alone or within a hydrogel, do not trigger significant inflammatory responses in vivo, such as cytokine storms, which would typically manifest in ruffled fur, hunched posture, lethargy, and acute weight loss.

[0197] The present disclosure explanted the EV hydrogel on day seven and visually observed signs of blood vessel formation (Figure 6b). Necropsy of bolus-injected mice did not reveal visiblesigns of new blood vessels in the injection site. The present disclosure then analyzed the explanted hydrogels via histology and immunofluorescence (Figure 6c-d, Figure 7), which confirmed angiogenesis within the hydrogels. Blood vessels are clearly identifiable in H&E-stained hydrogel sections, appearing as circular structures surrounded by densely stained ECM and cellular infiltrates (Figure 6c). In the Masson’s tri chrome stained sections (Figure 6d), blood vessels can also be identified throughout the hydrogel as circular regions with an outer collagen layer (stained blue) and inner endothelial cell nuclei (stained dark purple). This pattern is distinct from foreign body encapsulation, where vasculature is typically excluded from the biomaterial and restricted to the periphery of the fibrotic capsule90,91. Instead, the presence of morphologically advanced vasculature throughout the hydrogel suggests a pro-regenerative microenvironment, where endothelial cells and matrix-producing cells are actively engaged in tissue repair rather than immune-mediated rejection.

[0198] In addition to evidence of angiogenesis, histological analysis of the explanted EV hydrogels revealed extensive cellular infiltration and active tissue remodeling, with little to no signs of a foreign body response. H&E staining showed extensive cellular ingress throughout the material, with infdtrating cells engaging in ECM remodeling rather than forming fibrotic encapsulation (Figure 6c, Figure 38). Masson’s trichrome staining further confirmed significant collagen deposition within the hydrogel (Figure 6d, Figure 38), suggesting that rather than being walled off, the hydrogel is actively being remodeled by infiltrating cells. The organization of infiltrating cells (nuclei stained dark purple) within the deposited ECM includes large numbers of lymphocyte-like cells with a higher ratio of nucleus to cytoplasm. These data suggest that the EV hydrogel promotes constructive tissue remodeling, with ECM deposition occurring within the hydrogel rather than forming a fibrotic barrier, distinguishing it from materials that trigger chronic inflammation and a foreign body response. Similar cellular infiltration and ECM deposition were previously observed in liposome-based hydrogels38; however, a key distinction from previous liposome hydrogels is that, in addition to active ECM remodeling, the EV hydrogels also drive vascularization.

[0199] Following these observations of angiogenesis and cellular infiltration, the present disclosure sought to further characterize the developmental stage of vessel formation by performing angiogenesis-specific immunofluorescence (IF) studies. Hydrogel sections werestained for: CD31 (yellow) and CD34 (magenta), F-actin (cyan) (Figures 7-8, Figure 39- inc. DAPI nuclear stain (gray), Figures 40-46). CD31 is specific to mature endothelial cells lining blood vessels while CD34 is indicative of endothelial progenitor cells, immature endothelial cells, and hematopoietic stem cells (HSCs)92. CD31 and CD34 staining revealed areas of both neovascularization and maturing blood vessels (Figures 7-8, Figures 41, 44). At seven days, the present disclosure observed the presence of both new (magenta) and maturing (yellow and magenta) vessels.

[0200] Angiogenesis is further supported by the presence of F-actin-rich endothelial structures, indicative of active cell migration and vascular remodeling (Figures 7-8, Figures 39-46). Areas of strong F-actin and CD31 co-localization suggest regions of dynamic endothelial remodeling, where cytoskeletal reorganization facilitates lumen formation and vessel stabilization93. Conversely, the present disclosure also observed CD31+ vessels with minimal F-actin signal, which may represent more mature, stabilized vasculature with reduced cytoskeletal remodeling activity (Figures 7-8, Figures 42, 45). CD34+ vessels were more likely to be colocalized with F-actin (Figure 8, Figures 43, 46), which is consistent with their role in rearrangement and sprouting of new vessels. This spatial variation in F-actin expression suggests the presence of vessels at different stages of maturation, reinforcing the role of EV hydrogels in supporting both early neovascularization and the stabilization of developing vascular networks.

[0201] The presence of both endothelial progenitors and mature endothelial cells at day 7 parallels vascularization kinetics observed in both natural and biofunctionalized synthetic hydrogels94 96. However, while these materials typically require encapsulated angiogenic cues to drive vascularization, yogurt EV hydrogels appear to induce this response intrinsically. This ability to promote vascularization, in contrast to liposomal hydrogels38, suggests that EV-based materials retain bio-instructive signals inherent to the EVs used in their formulation. Various types of EVs, including those derived from endothelial cells, mesenchymal stem cells97,98, stromal cells, platelets, leukocytes, and erythrocytes99, have been implicated in angiogenesis, largely through the transfer of growth factors and regulatory microRNAs. EVs mediate these effects by carrying pro-angiogenic factors like vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF), as well as microRNAs such as miR-126 and miR-214, which enhance endothelial proliferation, migration, and lumen formation". Moreover, EVs derived from endothelial cellshave been shown to promote angiogenesis by directly interacting with endothelial cells, facilitating the formation of new blood vessels100. Additionally, proteins associated with the EV corona have been shown to contain significant amounts of pro-angiogenic factors like VEGF and epidermal growth factor (EGF), and prior work has demonstrated that corona-enriched EVs promote angiogenesis and skin regeneration89.

[0202] Taken together, the results in the present disclosure indicate that the EV hydrogel platform enables localized and sustained delivery of bioactive EV components, facilitating angiogenesis without observable toxicity. The ability to elicit blood vessel formation within the hydrogel suggests that EV hydrogels can function as bio-instructive scaffolds that preserve and localize the pro-regenerative functions of EVs for tissue engineering applications. This capability differentiates EV hydrogels from traditional synthetic biomaterials, which require additional functionalization or payloads to promote vascularization. By leveraging the intrinsic bioactivity of yogurt EVs, this hydrogel offers a versatile and biologically inspired platform for guiding tissue repair and regeneration.

[0203] Yogurt EV hydrogels support distinct immune cell niche compared to synthetic liposomal controls

[0204] Since EVs can modulate immune responses and several immune cell types are known to influence angiogenesis, it was hypothesized that differences in immune infiltrates could provide insights into the angiogenic activity of yogurt EV hydrogels. To test this hypothesis, the present disclosure performed multiparameter flow cytometry on explants from BALB / c mice injected with either yogurt EV hydrogels (3% HPMC-C16, ~1010 11EVs) or previously developed synthetic liposomal nanoparticle (LNP) hydrogels38. After 7 days, EV gels were visibly larger and showed more vascularization (Figure 9a, Figure 47). Flow cytometry revealed similar total live cell counts (Figure 9b), but EV gels contained a significantly higher percentage of CD45+immune cells: 87.0% ± 3.9% compared to 37.1% ± 5.0% in liposomal gels (p = 0.0001) (Figure 9c-d)

[0205] Further analysis showed that EV gels recruited significantly more innate immune cells, including neutrophils ([F4 / 8O‘CD1 lb+Ly6G+], p = 0.0374), dendritic cells ([F4 / 80'Ly6G‘ CD1 lc+], p = 0.0010), and macrophages ([CD1 lb+F4 / 80+], p = 0.0106) (Figure 9e-g). These cell types are commonly associated with clearance of biological material and have established roles in promoting tissue repair and angiogenesis101 103. To further characterize the location of theseinvading myeloid cells in the EV hydrogel, the present disclosure performed immunofluorescence staining for macrophages ([CD68+]) and dendritic cells ([CD1 lc+]) which confirmed their robust infiltration and co-localization within EV hydrogels (Figure 9h, Figure 37). These findings suggest that, while both liposomal and EV hydrogels are infiltrated by innate immune cells, EV hydrogels accumulate a larger and more diverse myeloid population. Given this robust infiltration and prior reports that certain myeloid subsets — particularly macrophages — can adopt pro-angiogenic phenotypes104’06, it was speculated that these cells may contribute to the vascularization observed in EV hydrogels. Early neutrophil accumulation has also been observed in regenerative settings, where neutrophils rapidly localize to injury sites and contribute to early ECM remodeling prior to macrophage-driven tissue repair101. These observations highlight the potential for coordinated innate immune activity to shape regenerative outcomes in EV hydrogels; however, additional studies will be necessary to clarify their role.

[0206] EV hydrogels also elicit a distinct lymphoid response that may contribute to a pro-regenerative immune environment. While B cell (CD1 lb’CD19+) levels were comparable between groups (Figure 9i), the present disclosure observed a striking increase in T cell (CDllb'CD3+) infiltration in EV hydrogels (p = 0.0003) (Figure 9j), driven by an ~8-fold enrichment in CD4+T cells (p = 0.0003) and a ~6-fold increase in CD8+T cells (p = 0.0147) (Figure 9k). Given the established role of lymphoid populations — particularly CD4 T cells — in modulating myeloid cell function and coordinating tissue remodeling107l0 / 2l / 2025 11:10:00 AM, this shift in cellular composition may play an important role in shaping the immune niche associated with angiogenesis.

[0207] To explore the functional identity of the expanded CD4+T cell population, the present disclosure assessed FOXP3 expression to determine whether these cells included regulatory T cells (Tregs), a subset known to promote angiogenesis and tissue repair107’108. Tregs exert their effects in part through IL-10-mediated reprogramming of macrophages toward pro-angiogenic phenotypes107,109. Immunofluorescence staining confirmed the presence of abundant CD3+T cells throughout the hydrogel, including the material core. Most CD3+cells also expressed FOXP3 with clear nuclear localization, confirming their identity as Tregs (Figure 9m; Figures 48-52). Given the known roles of Tregs in promoting angiogenesis and coordinating tissue repair, theirenrichment in EV hydrogels presents a compelling candidate mechanism for the vascular remodeling observed in vivo.

[0208] The data in the present disclosure indicates that yogurt EV hydrogels generate a distinct immune microenvironment enriched in myeloid and regulatory T cell populations, suggesting the establishment of an anti-inflammatory, pro-regenerative niche. This immune profile may underlie the angiogenesis observed in vivo and represents a key distinction from synthetic liposomal controls. These findings are consistent with previous studies showing that EV-containing biomaterials can modulate immune responses to support tissue remodeling and vascular growth39,40. However, a notable distinction in the present disclosure is the use of a low-cost, highly scalable yogurt-derived EV product, which exhibits comparable biological activity without the limitations of stem cell EV sourcing. This highlights the potential for developing regenerative biomaterials that are both effective and accessible. While the results in the present disclosure point to a meaningful link between EV-driven immune modulation and angiogenesis, further investigation is needed to elucidate the mechanisms that connect these observations and to determine how they can be harnessed in therapeutic contexts.CONCLUSION

[0209] Hydrogels are multifunctional biomaterials but are often composed of synthetic building blocks that do not inherently present biological signals to cells. Extracellular vesicles (EVs) offer unique bioactivity, but stably incorporating them into hydrogels remains a challenge. Here, we define the design principles for supram olecul ar hydrogels crosslinked by EVs. Bovine-derived yogurt EVs are used as a scalable source of bioactive EVs for systematic hydrogel development. Mixing EVs with optimally modified cellulose-based polymers yields injectable hydrogels with tunable mechanical properties. Following optimization with yogurt EVs, this platform’s versatility is demonstrated by formulating hydrogels with artificial microbial and mammalian nanovesicles. In vivo studies show EV hydrogel biocompatibility, intrinsic angiogenic activity, and emergence of an immune niche with broad immune cell engagement, highlighting their potential in regenerative medicine. These findings establish a framework for designing EV-crosslinked supramolecular hydrogels that integrate the natural bioactivity of EVs with the biomedical potential of injectable hydrogel technology.

[0210] The present disclosure defines a functional design space for a class of hydrogels crosslinked with extracellular vesicles (EVs), engineered to harness EV biosignaling capabilities and support the development of next-generation regenerative materials.It should be possible to create a supramolecular EV hydrogel using a strategy similar to previously developed liposomal hydrogels38, given the structural similarities between EVs and liposomes. However, direct substitution of liposomes with EVs using prior methods has led to mixed results in the literature. Some studies report that EVs alone are insufficient for gelation39, while others describe successful EV hydrogel formation40. These conflicting outcomes suggested that, despite structural similarities, the physicochemical properties of EVs differ enough from synthetic liposomes to alter their integration into supramolecular systems. Critically, no study to date has systematically examined how these interactions vary across EV and polymer types. As a result, the field lacks a clear design framework for reliably forming EV-crosslinked supramolecular hydrogels. The present invention define that framework by identifying the key polymer and EV parameters that govern robust, tunable hydrogel formation using EVs as the sole nanoparticulate crosslinker.

[0211] To systematically study the design principles for EV hydrogels, the present disclosure first develop a scalable manufacturing strategy to generate high quantities of EVs (e.g., IxlO11'12EVs) for adequate crosslinking. EVs suffer from labor-intensive isolation procedures and remarkably low yields30,41-44and these limitations have broadly hindered adoption of EVs into biomaterials research. To overcome this barrier, the present disclosure turned to agriculturally sourced EVs to optimize the formulation of EV hydrogels, utilizing yogurt EVs derived from bovine milk. Among agricultural sources, bovine milk EVs are the most extensively studied due to their potential for mass production and drug delivery45. They are biocompatible and naturally home to specific tissues such as the liver, lungs, and intestinal mucosa30,46,47, and may cross the intestinal and blood brain barriers (BBB)46,48. Bovine milk EVs can also modulate physiological processes that regulate immune responses and promote tissue repair49-51. Using milk whey isolated from yogurt, the present disclosure optimized a method for isolating and concentrating yogurt EVs through tangential flow filtration and lyophilization, generating ~1013of high-quality yogurt EVs from a single isolation.

[0212] Yogurt EVs enabled a systematic study of gelation behaviors between EVs and hydrophobically modified cellulose polymers, which identified an optimal polymer for producing supramolecular EV hydrogels. Although EVs are structurally similar to liposomes, EV-based hydrogels required re-engineering the polymer component to achieve robust mechanical properties. To determine if the present disclosure’s approach was compatible with other types of biomedically-relevant EVs, the present disclosure also validated this approach with mammalian and bacterial nanovesicles. To demonstrate the safety and bioactivity of EV hydrogels, EV hydrogels were injected into immunocompetent mice where no adverse reactions were observed, but spontaneous angiogenesis was observed.

[0213] In vivo analysis also revealed that EV hydrogels support the emergence of a distinct immune niche marked by diverse immune cell infdtration, including myeloid cells and regulatory T cells, suggesting that their bioactivity may be mediated in part through immune-driven tissue remodeling. Together, these findings define the critical parameters governing supramolecular EV hydrogel formation and establish a versatile platform for integrating extracellular vesicle bioactivity with injectable hydrogel technology.

[0214] The present disclosure establishes a fundamental framework for designing EV-crosslinked supramolecular hydrogels, addressing key uncertainties in their formation and material properties. Through systematic evaluation of polymer hydrophobicity, degree of functionalization, and polymer concentration, the present disclosure defines a design space that achieves stable, injectable, and tunable EV hydrogels. The present disclosure’s findings reveal that precise polymer engineering is essential for supramolecular hydrogel formation, allowing EVs to contribute to hydrogel mechanics rather than destabilize the network. Within this optimized design space, EVs enhance mechanical integrity, increasing the storage modulus while maintaining dynamic, shearthinning and self-healing behavior critical for injectability. These results not only confirm that EVs participate in hydrogel formation but also establish the polymer parameters necessary for their successful integration, addressing previous inconsistencies in the literature.

[0215] Beyond defining these principles, the present disclosure demonstrates the versatility of its approach across multiple EV sources, including yogurt whey, E. coll and melanoma cell-derived vesicles. The ability to harness the cell signaling capabilities of EVs in an injectable hydrogel has broad ranging potential for biomedical applications. The bioactivity and therapeuticpotential of these hydrogels is illustrated by angiogenesis in vivo following injection of a yogurt EV hydrogel. The vascularization observed within the hydrogel is advantageous as it enables the gel to contribute to a wide variety of biological processes by promoting a well-perfused environment that can support long term delivery of EVs or other cargo, including possibly the delivery of therapeutic cells. These capabilities have direct relevance to applications in tissue engineering, wound healing, and regenerative medicine12’62,110’111. While robust blood vessel formation and formation of an immune niche seven days after injection was reported, more extensive studies will be conducted moving forward to investigate the kinetics of long-term vascular remodeling and stabilization, as well as to further characterize the material’s interactions with the surrounding tissue. Future work will also expand upon additional biomedical applications through further diversifying the types of EVs used to formulate these hydrogels, as source-specific differences in vesicle composition and signaling properties are expected to influence functional outcomes and may be leveraged to tailor hydrogels for distinct therapeutic contexts.

[0216] While the present disclosure establishes key design principles for supramolecular EV hydrogels, the present disclosure’s findings are limited to HPMC modified with linear alkyl chains, leaving open the possibility that alternative hydrophobic modifications or entirely different polymer backbones could further refine hydrogel formation. Beyond polymer engineering, a key challenge in EV hydrogel development is the complexity of EV surface chemistry, particularly the possible role of the protein corona in mediating EV-polymer interactions. Unlike synthetic nanoparticles, EVs present a heterogeneous and dynamic surface composition that varies depending on cell source, isolation method, and processing conditions.89This variability complicates predictions of how EVs structurally behave within the hydrogel network, as differences in lipid composition, glycoprotein content, or membrane rigidity may significantly impact hydrogel mechanics. Future studies should carefully characterize EV surface properties, particularly when using extrusion, ultrafiltration, or enzymatic processing, to assess how these modifications influence crosslinking efficiency and hydrogel stability. Establishing best practices for corona characterization and EV formulation will be essential for translating EV hydrogels into standardized, reproducible biomaterials with well-defined mechanical and bioactive properties.

[0217] To ensure progress in the development of supramolecular EV hydrogels, it was recommended to adopt standardized methodologies that facilitate reproducibility and meaningfulcomparisons across studies. Including polymer-only hydrogels as rheological controls is essential to determine EV-specific contributions to the rheological properties and ensure that observed mechanical effects stem from EV-polymer interactions rather than polymer self-assembly. Additionally, reporting EV number and size rather than bulk concentration will allow for more precise formulation comparisons and avoid discrepancies arising from variations in EV yield and size distribution across different sources. Given the dynamic nature of supramolecular hydrogels, flow rheology should complement frequency sweep analysis to fully characterize shear-thinning and self-healing properties, which are critical for injectability and practical application.

[0218] In conclusion, the present disclosure establishes a functional design space for a modular, injectable EV-based supramolecular hydrogel that leverages the bioactivity of EVs for minimally invasive, localized tissue engineering. By revealing the fundamental design principles of EV hydrogel formation, the present disclosure provides a foundational framework to guide the next generation of EV-based biomaterials and foster stronger integration between extracellular vesicle research and biomaterials science.REFERENCES1. Correa, S., Grosskopf, A.K., Lopez Hernandez, H., Chan, D., Yu, A.C., Stapleton, L.M., and Appel, E.A. (2021). Translational Applications of Hydrogels. Chem. Rev. 121, 11385— 11457. https: / / doi.org / 10.1021 / acs.chemrev.0c01177.2. Appel, E.A., Del Barrio, J., Loh, X.J., and Scherman, O.A. (2012). Supramolecular polymeric hydrogels. Chem. Soc. Rev. 41, 6195. https: / / doi.org / 10.1039 / c2cs35264h.3. Appel, E.A., Tibbitt, M.W., Webber, M.J., Mattix, B.A., Veiseh, O., and Langer, R. (2015).Self-assembled hydrogels utilizing polymer-nanoparticle interactions. Nat. Commun. 6, 6295. https: / / doi.org / 10.1038 / ncomms7295.4. Hoffman, A.S. (2002). Hydrogels for biomedical applications. Adv. Drug Deliv. Rev. 54, 3- 12. https : / / doi . org / 10.1016 / SO 169-409X(01 )00239-3.5. Vedadghavami, A., Minooei, E, Mohammadi, M.H., Khetani, S., Rezaei Kolahchi, A., Mashayekhan, S., and Sanati-Nezhad, A. (2017). Manufacturing of hydrogel biomaterials with controlled mechanical properties for tissue engineering applications. Acta Biomater.62, 42-63. https: / / doi.Org / 10.1016 / j.actbio.2017.07.028.6. Ehrbar, M., Schoenmakers, R., Christen, E.H., Fussenegger, M., and Weber, W. (2008). Drugsensing hydrogels for the inducible release of biopharmaceuticals. Nat. Mater. 7, 800-804. https: / / doi.org / 10.1038 / nmat2250.7. Shen, W., Zhang, K., Kornfield, J.A., and Tirrell, D.A. (2006). Tuning the erosion rate of artificial protein hydrogels through control of network topology. Nat. Mater. 5, 153-158. https: / / doi.org / 10.1038 / nmatl573.8. Song, S.Z., Cardinalx, J.R., Kim, S.H., and Kim, S.W. (1981). Progestin Permeation Through Polymer Membranes V: Progesterone Release from Monolithic Hydrogel Devices. J.Pharm. Sci. 70, 216-219. https: / / doi.org / 10.1002 / jps.2600700226.9. Yean, L., Bunel, C., and Vairon, J. (1990). Reversible immobilization of drugs on a hydrogel matrix, 2. Diffusion of free chloramphenicol from poly(2-hydroxyethyl methacrylate) hydrogels. Makromol. Chem. 191, 1119-1129. https: / / doi.org / 10.1002 / macp.1990.021910514.10. Arora, P, and Mukheijee, B. (2002). Design, development, physicochemical, and in vitro and in vivo evaluation of transdermal patches containing diclofenac di ethylammonium salt. J. Pharm. Sci. 91, 2076-2089. https: / / doi.org / 10.1002 / jps.10200.11. Browning, M.B., Russell, B., Rivera, J., Hook, M., and Cosgriff-Hernandez, E.M. (2013).Bioactive Hydrogels with Enhanced Initial and Sustained Cell Interactions.Biomacromolecules 14, 2225-2233. https: / / doi.org / 10.1021 / bm400634j.12. Li, Y, Xu, T„ Tu, Z„ Dai, W, Xue, Y, Tang, C„ Gao, W„ Mao, C., Lei, B., and Lin, C.(2020). Bioactive antibacterial silica-based nanocomposites hydrogel scaffolds with highangiogenesis for promoting diabetic wound healing and skin repair. Theranostics 10, 4929-4943. https: / / doi.org / 10.7150 / thno.41839.Lee, K.Y., and Mooney, D.J. (2001). Hydrogels for Tissue Engineering. Chem. Rev. 101, 1869-1880. https: / / doi.org / 10.1021 / cr000108x.Marklein, R.A., and Burdick, J.A. (2010). Spatially controlled hydrogel mechanics to modulate stem cell interactions. Soft Matter 6, 136-143. https: / / doi.org / 10.1039 / B916933D. Guvendiren, M., Lu, H.D., and Burdick, J.A. (2012). Shear-thinning hydrogels for biomedical applications. Soft Matter 8, 260-272. https: / / doi.org / 10.1039 / ClSM06513K. Kloxin, C.J., and Bowman, C.N. (2013). Covalent adaptable networks: smart, reconfigurable and responsive network systems. Chem Soc Rev 42, 7161-7173. https: / / doi.org / 10.1039 / C3CS60046G.Cromwell, O.R., Chung, J., and Guan, Z. (2015). Malleable and Self-Healing Covalent Polymer Networks through Tunable Dynamic Boronic Ester Bonds. J. Am. Chem. Soc. 137, 6492-6495. https: / / doi .org / 10.1021 / j acs.5b03551.Meis, C.M., Grosskopf, A.K., Correa, S., and Appel, E.A. (2021). Injectable Supramolecular Polymer-Nanoparticle Hydrogels for Cell and Drug Delivery Applications. J. Vis. Exp., 62234. https: / / doi.org / 10.3791 / 62234.Yu, A.C., Lian, H., Kong, X., Lopez Hernandez, H., Qin, J., and Appel, E.A. (2021).Physical networks from entropy-driven non-covalent interactions. Nat. Commun. 12, 746. https : / / doi . org / 10.1038 / s41467-021-21024-7.Piantanida, E., Alonci, G., Bertucci, A., and De Cola, L. (2019). Design of Nanocomposite Injectable Hydrogels for Minimally Invasive Surgery. Acc. Chem. Res. 52, 2101-2112. http s : / / doi . org / 10.1021 / acs . accounts .9b 00114.Alonso, J.M., Andrade Del Olmo, J., Perez Gonzalez, R., and Saez-Martinez, V. (2021). Injectable Hydrogels: From Laboratory to Industrialization. Polymers 13, 650. https: / / doi.org / 10.3390 / polyml3040650.Salzlechner, C., Haghighi, T., Huebscher, I., Walther, A.R., Schell, S., Gardner, A., Undt, G., Da Silva, R.M.P., Dreiss, C.A., Fan, K., et al. (2020). Adhesive Hydrogels for Maxillofacial Tissue Regeneration Using Minimally Invasive Procedures. Adv. Healthc. Mater. 9, 1901134. https: / / doi.org / 10.1002 / adhm.201901134.Kyburz, K.A., and Anseth, K.S. (2015). Synthetic Mimics of the Extracellular Matrix: How Simple is Complex Enough? Ann. Biomed. Eng. 43, 489-500.https : / / doi . org / 10.1007 / s 10439-015- 1297-4.Jeong, S.H., Kim, M., Kim, T.Y, Choi, H., and Hahn, S.K. (2021). Biomimetic Supramolecular Drug Delivery Hydrogels for Accelerated Skin Tissue Regeneration. ACS Biomater. Sci. Eng. 7, 4581-4590. https: / / doi.org / 10.1021 / acsbiomaterials.lc00705.Hoare, T.R., and Kohane, D.S. (2008). Hydrogels in drug delivery: Progress and challenges. Polymer 49, 1993-2007. https: / / doi.org / 10.1016 / j polymer.2008.01.027.Lim, J.Y.C., Lin, Q., Xue, K., and Loh, X.L (2019). Recent advances in supramolecular hydrogels for biomedical applications. Mater. Today Adv. 3, 100021. https: / / doi.Org / 10.1016 / j.mtadv.2019.100021.Martin-Saldana, S., Waeel, M.A., Alsharabasy, A.M., Daly, A., and Pandit, A. (2022). An interdisciplinary framework for the characterization of extracellular matrix-hydrogels for biomedical applications. Matter 5, 3659-3705. https: / / doi.Org / 10.1016 / j.matt.2022.09.020. Brennan, K., Martin, K., FitzGerald, S.P., O’Sullivan, J., Wu, Y, Blanco, A., Richardson, C., and Me Gee, M.M. (2020). A comparison of methods for the isolation and separation of extracellular vesicles from protein and lipid particles in human serum. Sci. Rep. 10, 1039. https: / / doi.org / 10.1038 / s41598-020-57497-7.Welsh, J.A., Goberdhan, D.C.I., O’Driscoll, L., Buzas, E.I., Blenkiron, C., Bussolati, B., Cai, H., Di Vizio, D., Driedonks, T.A.P, Erdbriigger, U., et al. (2024). Minimal information for studies of extracellular vesicles (MISEV2023): From basic to advanced approaches. J. Extracell. Vesicles 13, el2404. https: / / doi.org / 10.1002 / jev2.12404.Herrmann, I.K., Wood, M.J.A., and Fuhrmann, G. (2021). Extracellular vesicles as a nextgeneration drug delivery platform. Nat. Nanotechnol. 16, 748-759. https: / / doi.org / 10.1038 / s41565-021-00931-2.Jahnke, K., and Staufer, O. (2024). Membranes on the move: The functional role of the extracellular vesicle membrane for contact-dependent cellular signalling. J. Extracell.Vesicles 13, el2436. https: / / doi.org / 10.1002 / jev2.12436.Hallal, S., Tuzesi, A., Grau, G.E., Buckland, M.E., and Alexander, K.L. (2022).Understanding the extracellular vesicle surface for clinical molecular biology. J. Extracell. Vesicles 11, el2260. https: / / doi.org / 10.1002 / jev2.12260.Shao, M., Lopes, D., Lopes, J., Yousefiasl, S., Macario-Soares, A., Peixoto, D., Ferreira-Faria, I., Veiga, F., Conde, J., Huang, Y, et al. (2023). Exosome membrane-coated nanosystems: Exploring biomedical applications in cancer diagnosis and therapy. Matter 6, 761-799. https: / / doi.Org / 10.1016 / j.matt.2023.01.012.Wang, C., Wang, M., Xu, T., Zhang, X., Lin, C., Gao, W., Xu, H., Lei, B., and Mao, C. (2019). Engineering Bioactive Self-Healing Antibacterial Exosomes Hydrogel for Promoting Chronic Diabetic Wound Healing and Complete Skin Regeneration.Theranostics 9, 65-76. https: / / doi.org / 10.7150 / thno.29766.Guo, S.-C., Tao, S.-C., Yin, W.-J., Qi, X., Yuan, T., and Zhang, C.-Q. (2017). Exosomes derived from platelet-rich plasma promote the re-epithelization of chronic cutaneous wounds via activation of YAP in a diabetic rat model. Theranostics 7, 81-96.https: / / doi.org / 10.7150 / thno.16803.Shi, L., Jing, Y, Lu, H, Zhao, F., An, M„ Jin, S„ Gao, C , Dai, Y, Zhu, Y, Yang, S., et al. Extracellular Vesicle Crosslinkers Constructing Hydrogels with Stress-Relaxation and Bioactive Protein Modification. Adv. Mater. Interfaces n / a, 2400885.https : / / doi . org / 10.1002 / admi .202400885.Wu, F, Ren, Y, Lv, W., Liu, X., Wang, X., Wang, C., Cao, Z., Liu, J., Wei, J., and Pang, Y. (2024). Generating dual structurally and functionally skin-mimicking hydrogels by crosslinking cell-membrane compartments. Nat. Commun. 15, 802. https: / / doi.org / 10.1038 / s41467-024-45006-7.Correa, S., Grosskopf, A.K., Klich, J.H., Lopez Hernandez, H., and Appel, E.A. (2022). Injectable liposome-based supramolecular hydrogels for the programmable release of multiple protein drugs. Matter 5, 1816-1838. https: / / doi.Org / 10.1016 / j.matt.2022.03.001. Patel, N., Avery, E., and Chung, E.J. (2024). Supramolecular hydrogels for sustained extracellular vesicle delivery. MRS Commun. https: / / doi.org / 10.1557 / s43579-024-00589-6. Zhang, Q., Wang, Y, Zhu, Z., Ahmed, W., Zhou, D., and Chen, L. (2025). Therapeutic Potential of Injectable Supramolecular Hydrogels With Neural Stem Cell Exosomes and Hydroxypropyl Methylcellulose for Post-Stroke Neurological Recovery. Int. J.Nanomedicine 20, 2253-2271. https: / / doi.org / 10.2147 / IJN.S505792.El Andaloussi, S., Mager, I., Breakefield, X.O., and Wood, M.J.A. (2013). Extracellular vesicles: biology and emerging therapeutic opportunities. Nat. Rev. Drug Discov. 12, 347-357. https: / / doi.org / 10.1038 / nrd3978.Syromiatnikova, V, Prokopeva, A., and Gomzikova, M. (2022). Methods of the Large-Scale Production of Extracellular Vesicles. Int. J. Mol. Sci. 23, 10522. https: / / doi.org / 10.3390 / ijms231810522.Cheng, K., and Kalluri, R. (2023). Guidelines for clinical translation and commercialization of extracellular vesicles and exosomes based therapeutics. Extracell. Vesicle 2, 100029. https: / / doi.Org / 10.1016 / j.vesic.2023.100029.Cheng, L., and Hill, A.F. (2022). Therapeutically harnessing extracellular vesicles. Nat. Rev. Drug Discov. 21, 379-399. https: / / doi.org / 10.1038 / s41573-022-00410-w.Turner, N.P (2024). Food-derived extracellular vesicles in the human gastrointestinal tract: Opportunities for personalised nutrition and targeted therapeutics. J. Extracell. Biol. 3, el 54. https: / / doi.org / 10.1002 / jex2.154.Marsh, S.R., and Gourdie, R.G. (2024). Oral delivery of therapeutic peptides by milk-derived extracellular vesicles. Nat. Rev. Bioeng. https: / / doi.org / 10.1038 / s44222-024-00227-9.Prasadani, M., Kodithuwakku, S., Pennarossa, G., Fazeli, A., and Brevini, T.A.L. (2024). Therapeutic Potential of Bovine Milk-Derived Extracellular Vesicles. Int. J. Mol. Sci. 25, 5543. https: / / doi.org / 10.3390 / ijms25105543.Tong, L., Zhang, S., Liu, Q., Huang, C., Hao, H., Tan, M.S., Yu, X., Lou, C.K.L., Huang, R., Zhang, Z., et al. (2023). Milk-derived extracellular vesicles protect intestinal barrier integrity in the gut-liver axis. Sci. Adv. 9, eade5041.http s : / / doi . org / 10.1126 / sci adv. ade5041.Kim, H., Kim, D.E., Han, G., Lim, N.R., Kim, E.H., Jang, Y, Cho, H., Jang, H., Kim, K.H., Kim, S.H., et al. (2022). Harnessing the Natural Healing Power of Colostrum: Bovine Milk-Derived Extracellular Vesicles from Colostrum Facilitating the Transition from Inflammation to Tissue Regeneration for Accelerating Cutaneous Wound Healing. Adv. Healthc. Mater. 11, 2102027. https: / / doi.org / 10.1002 / adhm.202102027.Pieters, B.C.H., Arntz, O.J., Aarts, J., Feitsma, A.L., Van Neerven, R.J.J., Van Der Kraan, P.M., Oliveira, M.C., and Van De Loo, F.A.J. (2022). Bovine Milk-Derived Extracellular Vesicles Inhibit Catabolic and Inflammatory Processes in Cartilage from Osteoarthritis Patients. Mol. Nutr. Food Res. 66, 2100764. https: / / doi.org / 10.1002 / mnfr.202100764.Mecocci, S., Trabalza-Marinucci, M., and Cappelli, K. (2022). Extracellular Vesicles from Animal Milk: Great Potentialities and Critical Issues. Animals 12, 3231. https: / / doi.org / 10.3390 / anil2233231.Busatto, S., Vilanilam, G., Ticer, T., Lin, W.-L., Dickson, D.W., Shapiro, S., Bergese, P., and Wolfram, J. (2018). Tangential Flow Filtration for Highly Efficient Concentration of Extracellular Vesicles from Large Volumes of Fluid. Cells 7, 273. https: / / doi.org / 10.3390 / cells7120273.Patema, A., Rao, E., Adamo, G., Raccosta, S., Picciotto, S., Romancino, D., Noto, R., Touzet, N., Bongiovanni, A., and Manno, M. (2022). Isolation of Extracellular Vesicles From Microalgae: A Renewable and Scalable Bioprocess. Front. Bioeng. Biotechnol. 10, 836747. https: / / doi.org / 10.3389 / fbioe.2022.836747.Correa, S., Choi, K.Y, Dreaden, E.C., Renggli, K., Shi, A., Gu, L., Shopsowitz, K.E., Quadir, M.A., Ben-Akiva, E., and Hammond, PT. (2016). Highly Scalable, Closed-Loop Synthesis of Drug-Loaded, Layer-by-Lay er Nanoparticles. Adv. Funct. Mater. 26, 991— 1003. https: / / doi.org / 10.1002 / adfm.201504385.Paganini, C., Capasso Palmiero, U., Pocsfalvi, G., Touzet, N., Bongiovanni, A., and Arosio, P. (2019). Scalable Production and Isolation of Extracellular Vesicles: Available Sources and Lessons from Current Industrial Bioprocesses. Biotechnol. J. 14, 1800528. https: / / doi.org / 10.1002 / biot.201800528.Kawai-Harada, Y, Nimmagadda, V, and Harada, M. (2024). Scalable isolation of surface-engineered extracellular vesicles and separation of free proteins via tangential flow filtration and size exclusion chromatography (TFF-SEC). BMC Methods 1, 9. https: / / doi.org / 10.1186 / s44330-024-00009-0.Adamo, G., Fierli, D., Romancino, D.P, Picciotto, S., Barone, M.E., Aranyos, A., Bozic, D., Morsbach, S., Raccosta, S., Stanly, C., et al. (2021). Nanoalgosomes: Introducingextracellular vesicles produced by microalgae. J. Extracell. Vesicles 10, el2081.http s : / / doi . org / 10.1002 / j ev2.12081.Pieters, B.C.H., Arntz, O.J., Bennink, MB., Broeren, M.G.A., Van Caam, A.P.M., Koenders, M.I., Van Lent, P.L.E.M., Van Den Berg, W.B., De Vries, M., Van Der Kraan, P.M., et al. (2015). Commercial Cow Milk Contains Physically Stable Extracellular Vesicles Expressing Immunoregulatory TGF-p. PLOS ONE 10, e0121123. https: / / doi.org / 10.1371 / joumal.pone.0121123.Frank, J., Richter, M., De Rossi, C., Lehr, C.-M., Fuhrmann, K., and Fuhrmann, G. (2018). Extracellular vesicles protect glucuronidase model enzymes during freeze-drying. Sci. Rep.8, 12377. https: / / doi.org / 10.1038 / s41598-018-30786-y.Trenkenschuh, E., Richter, M., Heinrich, E., Koch, M., Fuhrmann, G., and Friess, W.(2022). Enhancing the Stabilization Potential of Lyophilization for Extracellular Vesicles. Adv. Healthc. Mater. 11, 2100538. https: / / doi.org / 10.1002 / adhm.202100538.Baker, B.M., and Chen, C.S. (2012). Deconstructing the third dimension - how 3D culture microenvironments alter cellular cues. J. Cell Sci., jcs.079509. https: / / doi.org / 10.1242 / jcs.079509.Rustad, K.C., Wong, V.W., Sorkin, M., Glotzbach, J.P, Major, M.R., Rajadas, J., Longaker, M.T., and Gurtner, G.C. (2012). Enhancement of mesenchymal stem cell angiogenic capacity and sternness by a biomimetic hydrogel scaffold. Biomaterials 33, 80-90. https: / / doi.Org / 10.1016 / j.biomaterials.2011.09.041.Engler, A.J., Sen, S., Sweeney, H.L., and Discher, D.E. (2006). Matrix Elasticity Directs Stem Cell Lineage Specification. Cell 126, 677-689. https: / / doi.Org / 10.1016 / j.cell.2006.06.044.Li, B., Zhang, L., Yin, Y, Chen, A., Seo, B.R., Lou, J., Mooney, D.J., and Weitz, D.A.(2024). Stiff hydrogel encapsulation retains mesenchymal stem cell sternness for regenerative medicine. Matter 7, 3447-3468. https: / / doi.Org / 10.1016 / j.matt.2024.05.041. Li, J., and Mooney, D.J. (2016). Designing hydrogels for controlled drug delivery. Nat. Rev. Mater. 1, 16071. https: / / doi.org / 10.1038 / natrevmats.2016.71.C. Yu, A., A. Smith, A. A., and A. Appel, E. (2020). Structural considerations for physical hydrogels based on polymer-nanoparticle interactions. Mol. Syst. Des. Eng. 5, 401-407. https : / / doi . org / 10.1039 / C9ME00120D .De Gennes, PG. (1976). Dynamics of Entangled Polymer Solutions. I. The Rouse Model. Macromolecules 9, 587-593. https: / / doi.org / 10.1021 / ma60052a011.Allen, T.M., Sapra, P, and Moase, E. (2002). Use of the post-insertion method for the formation of ligand-coupled liposomes. Cell. Mol. Biol. Lett. 7, 217-219.Jang, H., Kim, H., Kim, E.H., Han, G , Jang, Y, Kim, Y, Lee, J.W., Shin, S.C., Kim, E.E., Kim, S.H., et al. (2023). Post-insertion technique to introduce targeting moieties in milk exosomes for targeted drug delivery. Biomater. Res. 27, 124.https: / / doi.org / ! 0.1186 / s40824-023-00456-w.Adibnia, V, Taghavi, S.M., and Hill, R.J. (2017). Roles of chemical and physical crosslinking on the rheological properties of silica-doped polyacrylamide hydrogels. Rheol. Acta 56, 123-134. https: / / doi.org / 10.1007 / s00397-016-0989-5.Loebel, C., Rodell, C.B., Chen, M.H., and Burdick, J. A. (2017). Shear-thinning and self-healing hydrogels as injectable therapeutics and for 3D-printing. Nat. Protoc. 72, 1521— 1541. https: / / doi.org / 10.1038 / nprot.2017.053.Nakamura, K., Kubota, R., Aoyama, T., Urayama, K., and Hamachi, I. (2023). Four distinct network patterns of supramolecular / polymer composite hydrogels controlled by formation kinetics and interfiber interactions. Nat. Commun. 14, 1696.https : / / doi . org / 10.1038 / s41467-023-37412-0.Bovone, G., Guzzi, E.A., Bernhard, S., Weber, T., Dranseikiene, D., and Tibbitt, M.W. (2022). Supramolecular Reinforcement of Polymer-Nanoparticle Hydrogels for Modular Materials Design. Adv. Mater. 34, 2106941. https: / / doi.org / 10.1002 / adma.202106941. Elucidating the G" overshoot in soft materials with a yield transition via a time-resolved experimental strain decomposition https: / / doi.org / 10.1073 / pnas.2003869117.Grosskopf, A.K., Saouaf, O.A., Lopez Hernandez, H., and Appel, E.A. (2021). Gelation and yielding behavior of POLYMER-NANOPARTICLE hydrogels. J. Polym. Sci. 59, 2854-2866. https: / / doi.org / 10.1002 / pol.20210652.Wang, X., Hu, S., Zhu, D., Li, J., Cheng, K., and Liu, G. (2023). Comparison of extruded cell nanovesicles and exosomes in their molecular cargos and regenerative potentials. Nano Res. 76, 7248-7259. https: / / doi.org / 10.1007 / sl2274-023-5374-3.Duan, X., Zhang, R., Feng, H., Zhou, H., Luo, Y, Xiong, W, Li, J., He, Y, and Ye, Q. (2024). Anew subtype of artificial cell-derived vesicles from dental pulp stem cells with the bioequivalence and higher acquisition efficiency compared to extracellular vesicles. J. Extracell. Vesicles 13, el2473. https: / / doi.org / 10.1002 / jev2.12473.Furuyama, N., and Sircili, M.P. (2021). Outer Membrane Vesicles (OMVs) Produced by Gram-Negative Bacteria: Structure, Functions, Biogenesis, and Vaccine Application.BioMed Res. Int. 2021, 1490732. https: / / doi.org / 10.1155 / 2021 / 1490732.Lu, M., Xing, H., Zhao, X., Huang, Y, Zheng, A., and Liang, X.-J. (2024). Engineered extracellular vesicles as a next-generation vaccine platform. Matter 7, 4180-4205.https: / / doi.Org / 10.1016 / j.matt.2024.09.012.Liu, G ., Ma, N„ Cheng, K., Feng, Q., Ma, X., Yue, Y, Li, Y, Zhang, T., Gao, X., Liang, J., et al. (2024). Bacteria-derived nanovesicles enhance tumour vaccination by trained immunity. Nat. Nanotechnol. 79, 387-398. https: / / doi.org / 10.1038 / s41565-023-01553-6. Liu, H., Zhang, Q., Wang, S., Weng, W., Jing, Y, and Su, J. (2022). Bacterial extracellular vesicles as bioactive nanocarriers for drug delivery: Advances and perspectives. Bioact. Mater. 14, 169-181. https: / / doi.Org / 10.1016 / j.bioactmat.2021.12.006.Schwechheimer, C., and Kuehn, M.J. (2015). Outer-membrane vesicles from Gramnegative bacteria: biogenesis and functions. Nat. Rev. Microbiol. 13, 605-619. https: / / doi.org / 10.1038 / nrmicro3525.Fernandez -Rhodes, M., Lorca, C., Lisa, J., Batalla, I., Ramos-Miguel, A., Gallart-Palau, X., and Serra, A. (2024). New Origins of Yeast, Plant and Bacterial -Derived Extracellular Vesicles to Expand and Advance Compound Delivery. Int. J. Mol. Sci. 25, 7151. https: / / doi.org / 10.3390 / ijms25137151.O’Brien, K., Breyne, K., Ughetto, S., Laurent, L.C., and Breakefield, X.O. (2020). RNA delivery by extracellular vesicles in mammalian cells and its applications. Nat. Rev. Mol. Cell Biol. 21, 585-606. https: / / doi.org / 10.1038 / s41580-020-0251-y.Gonzalez-Mel ero, L., Hernandez, R.M., Santos- Vizcaino, E., and Igartua, M. (2023).Tumour-derived extracellular vesicle based vaccines for melanoma treatment. Drug Deliv. Transl. Res. 13, 1520-1542. https: / / doi.org / 10.1007 / sl3346-023-01328-5.Benzi, R., Divoux, T., Barentin, C., Manneville, S., Sbragaglia, M., and Toschi, F. (2021). Stress Overshoots in Simple Yield Stress Fluids. Phys. Rev. Lett. 127, 148003. https: / / doi.org / 10.1103 / PhysRevLett.127.148003.Welsh, J.A., Goberdhan, D.C.I., O’Driscoll, L., Buzas, E.I., Blenkiron, C., Bussolati, B., Cai, H., Di Vizio, D., Driedonks, T.A.P, Erdbriigger, U., et al. (2024). Minimal information for studies of extracellular vesicles (MISEV2023): From basic to advanced approaches. J. Extracell. Vesicles 13, el2404. https: / / doi.org / 10.1002 / jev2.12404.Somiya, M., Yoshioka, Y, and Ochiya, T. (2018). Biocompatibility of highly purified bovine milk-derived extracellular vesicles. J. Extracell. Vesicles 7, 1440132. https: / / doi.org / 10.1080 / 20013078.2018.1440132.Wolf, M., Poupardin, R.W., Ebner-Peking, P, Andrade, A.C., Blochl, C., Obermayer, A., Gomes, F.G., Vari, B., Maeding, N., Eminger, E., et al. (2022). A functional corona around extracellular vesicles enhances angiogenesis, skin regeneration and immunomodulation. J. Extracell. Vesicles 11, el2207. https: / / doi.org / 10.1002 / jev2.12207.Zhang, D., Chen, Q., Bi, Y, Zhang, H., Chen, M., Wan, J., Shi, C., Zhang, W., Zhang, J., Qiao, Z., et al. (2021). Bio-inspired poly-DL-serine materials resist the foreign-body response. Nat. Commun. 12, 5327. https: / / doi.org / 10.1038 / s41467-021-25581-9.Camicer-Lombarte, A., Chen, S.-T., Malliaras, G.G., and Barone, D.G. (2021). Foreign Body Reaction to Implanted Biomaterials and Its Impact in Nerve Neuroprosthetics. Front. Bioeng. Biotechnol. 9. https: / / doi.org / 10.3389 / fbioe.2021.622524.Asante, D.-B., Tierno, D., Woode, M., and Scaggiante, B. (2024). Angiogenesis and Ovarian Cancer: What Potential Do Different Subtypes of Circulating Endothelial Cells Have for Clinical Application? Int. J. Mol. Sci. 25, 6283. https: / / doi.org / 10.3390 / ijms25116283.Yadunandanan Nair, N., Samuel, V, Ramesh, L., Marib, A., David, D.T., and Sundararaman, A. (2022). Actin cytoskeleton in angiogenesis. Biol. Open 11, bio058899. https: / / doi.org / 10.1242 / bio.058899.Lopez-Gutierrez, J., Ramos-Payan, R., Ayala-Ham, A., Romero-Quintana, J.G., Castillo-Ureta, H., Villegas-Mercado, C., Bermudez, M., Sanchez- Schmitz, G., and Aguilar-Medina, M. (2023). Biofunctionalization of hydrogel -based scaffolds for vascular tissue regeneration. Front. Mater. 10. https: / / doi.org / 10.3389 / fmats.2023.1168616.Wei, Z., Lei, M., Wang, Y, Xie, Y, Xie, X., Lan, D., Jia, Y, Liu, J., Ma, Y, Cheng, B., et al. (2023). Hydrogels with tunable mechanical plasticity regulate endothelial cell outgrowth in vasculogenesis and angiogenesis. Nat. Commun. 14, 8307. https: / / doi.org / 10.1038 / s41467-023-43768-0.Xia, Z., Guo, B., Wu, D., Yang, F., and Ding, Y. (2024). Advances of natural hydrogel-based vascularization strategies for soft tissue repair. Front. Mater. 11. https: / / doi.org / 10.3389 / fmats.2024.1446035.Lai, R.C., Yeo, R.W.Y, Tan, K.H., and Lim, S.K. (2013). Mesenchymal Stem Cell Exosome Ameliorates Reperfusion Injury Through Proteomic Complementation. Regen. Med. 8, 197-209. https: / / doi.Org / 10.2217 / rme.13.4.Hu, H., Hu, X., Li, L., Fang, Y, Yang, Y, Gu, J., Xu, J., and Chu, L. (2022). Exosomes Derived from Bone Marrow Mesenchymal Stem Cells Promote Angiogenesis in Ischemic Stroke Mice via Upregulation of MiR-21-5p. Biomolecules 12, 883. https: / / doi.org / 10.3390 / bioml2070883.Todorova, D., Simoncini, S., Lacroix, R., Sabatier, F., and Dignat-George, F. (2017).Extracellular Vesicles in Angiogenesis. Circ. Res. 120, 1658-1673.http s : / / doi . org / 10.1161 / CIRCRE SAHA.117.309681.Chen, P, Pan, K., Song, N., Yang, Y, Gu, C., Zhong, P, Li, L., Li, M., Zhang, Y, Dai, Z., et al. (2023). A natural extracellular matrix hydrogel through selective nutrient restriction for hyperinflammatory starvation therapy. Matter 6, 397-428. https: / / doi.Org / 10.1016 / j.matt.2022.12.002.Anderson, T., Mo, J., Gagarin, E., Sherwood, D., Blumenkrantz, M., Mao, E., Leon, G., Levitz, H., Chen, H.-J., Tseng, K.-C., et al. (2023). Ligament injury in adult zebrafishtriggers ECM remodeling and cell dedifferentiation for scar-free regeneration. Npj Regen. Med. 5, 1-15. https: / / doi.org / 10.1038 / s41536-023-00329-9.Joshi, N., Pohlmeier, L., Ben- Yehuda Greenwald, M., Haertel, E., Hiebert, P, Kopf, M., and Werner, S. (2020). Comprehensive characterization of myeloid cells during wound healing in healthy and healing-impaired diabetic mice. Eur. J. Immunol. 50, 1335-1349. https : / / doi . org / 10.1002 / ej i .201948438.Sinha, M., Sen, C.K., Singh, K., Das, A., Ghatak, S., Rhea, B., Blackstone, B., Powell, H.M., Khanna, S., and Roy, S. (2018). Direct conversion of injury-site myeloid cells to fibroblast-like cells of granulation tissue. Nat. Commun. 9, 936. https: / / doi.org / 10.1038 / s41467-018-03208-w.Eming, S.A., Martin, P, and Tomic-Canic, M. (2014). Wound repair and regeneration: Mechanisms, signaling, and translation. Sci. Transl. Med. 6, 265sr6-265sr6.https : / / doi . org / 10.1126 / scitranslmed.3009337.Gurevich, D.B., Severn, C.E., Twomey, C., Greenhough, A., Cash, J., Toye, A M., Mellor, H., and Martin, P. (2018). Live imaging of wound angiogenesis reveals macrophage orchestrated vessel sprouting and regression. EMBO J. 37, e97786. https: / / doi.org / 10.15252 / embj.201797786.Ramirez-Pedraza, M., and Fernandez, M. (2019). Interplay Between Macrophages and Angiogenesis: A Double-Edged Sword in Liver Disease. Front. Immunol. 10. https: / / doi.org / 10.3389 / fimmu.2019.02882.Luznik, Z., Anchouche, S., Dana, R., and Yin, J. (2020). Regulatory T Cells in Angiogenesis. J. Immunol. 205, 2557-2565. https: / / doi.org / 10.4049 / jimmunol.2000574. D’Alessio, F.R., Zhong, Q., Jenkins, J., Moldobaeva, A., and Wagner, E.M. (2015). Lung Angiogenesis Requires CD4+Forkhead Homeobox Protein-3+ Regulatory T Cells. Am. J. Respir. Cell Mol. Biol. 52, 603-610. https: / / doi.org / 10.1165 / rcmb.2014-02780C.Leung, O.M., Li, J., Li, X., Chan, V.W., Yang, K.Y, Ku, M., Ji, L., Sun, H., Waldmann, H, Tian, X.Y, et al. (2018). Regulatory T Cells Promote Apelin-Mediated Sprouting Angiogenesis in Type 2 Diabetes. Cell Rep. 24, 1610-1626. https: / / doi.Org / 10.1016 / j.celrep.2018.07.019.Rufaihah, A. J., and Seliktar, D. (2016). Hydrogels for therapeutic cardiovascular angiogenesis. Adv. Drug Deliv. Rev. 96, 31-39. https: / / doi.Org / 10.1016 / j.addr.2015.07.003. Peattie, R.A., Nayate, A.P, Firpo, M.A., Shelby, J., Fisher, R.J., and Prestwich, G.D.(2004). Stimulation of in vivo angiogenesis by cytokine-loaded hyaluronic acid hydrogel implants. Biomaterials 25, 2789-2798. https: / / doi.Org / 10.1016 / j.biomaterials.2003.09.054. Grosskopf, A.K., Labanieh, L., Klysz, D.D., Roth, G.A., Xu, P., Adebowale, O., Gale, E.C., Jons, C.K., Klich, J.H., Yan, J., et al. (2022). Delivery of CAR-T cells in a transientinjectable stimulatory hydrogel niche improves treatment of solid tumors. Sci. Adv. 8, eabn8264. https: / / doi.org / 10.1126 / sciadv.abn8264.Glickman, ME., Rao, S.R., and Schultz, M.R. (2014). False discovery rate control is a recommended alternative to Bonferroni-type adjustments in health studies. J. Clin.Epidemiol. 67, 850-857. doi.org / 10.1016 / j.jclinepi.2014.03.012.Grosskopf, A.K., Labanieh, L., Klysz, D.D., Roth, G.A., Xu, P., Adebowale, O., Gale, E.C., Jons, C.K., Klich, J.H., Yan, J., et al. (2022). Delivery of CAR-T cells in a transient injectable stimulatory hydrogel niche improves treatment of solid tumors. Sci. Adv. 5, eabn8264. https: / / doi.org / 10.1126 / sciadv.abn8264.Wu, I.Y., Bala, S., Skalko-Basnet, N., and di Cagno, M.P. (2019). Interpreting nondinear drug diffusion data: Utilizing Korsmeyer-Peppas model to study drug release from liposomes. Eur. J. Pharm. Sci. 138, 105026. https: / / doi.Org / 10.1016 / j.ejps.2019.105026.

Claims

CLAIMS1. A supramolecular hydrogel comprising (i) at least one polymer and (ii) extracellular vesicles.

2. The supramolecular hydrogel of Claim 1, wherein molecules of the at least one polymer are reversibly and / or non-covalently crosslinked by the extracellular vesicles.

3. The supramolecular hydrogel of Claim 1, wherein the extracellular vesicles are the only crosslinkers of the supramolecular hydrogel.

4. The supramolecular hydrogel of Claim 1, wherein the hydrogel does not contain liposomes and / or the supramolecular hydrogel is self-healing.

5. The supramolecular hydrogel of Claim 1, wherein the at least one polymer is a cellulose- based polymer, preferably an alkyl-modified cellulose polymer.

6. The supramolecular hydrogel of Claim 1, wherein the at least one polymer is a modified hydroxypropyl methylcellulose (HPMC-CX), wherein X is 12, 14, 16, or 18, preferably wherein X is 14 or 16.

7. The supramolecular hydrogel of Claim 1, wherein the polymer concentration of the hydrogel is 2-4 wt%.

8. The supramolecular hydrogel of Claim 1, wherein the polymer has a degree of functionalization of about 0.32 - 0.52 mmol / g.

9. The supramolecular hydrogel of Claim 1, wherein the extracellular vesicles are derived from an agricultural source, bovine milk, milk whey isolated from yogurt, cells from a cultured cell line, human cells, murine cells, cancer cells, bacteria, or microalgae.

10. The supramolecular hydrogel of Claim 1, wherein the extracellular vesicles are isolated by tangential flow fdtration.

11. The supramolecular hydrogel of Claim 1, comprising cross links between molecules of at least one polymer and one or more extracellular vesicles.

12. The supram olecul ar hydrogel of Claim 1, wherein the extracellular vesicles are reconstituted after lyophilization.

13. The supramolecular hydrogel of Claim 1, wherein the final extracellular vesicle weight percentage in the hydrogel ranges from 1-10% by weight.

14. The supramolecular hydrogel of any of Claims 1 - 13, wherein the hydrogel comprises 1 Olo-l 013EVs / mL, preferably IxlO11to 5xl012EVs / mL.

15. The supramolecular hydrogel of Claim 1, wherein the hydrogel has a storage modulus of about 200-700 Pa and / or a loss modulus of about 150-250 Pa.

16. A method of producing a supramolecular hydrogel comprising (i) at least one polymer and (ii) extracellular vesicles, the method comprising admixing at least one polymer and extracellular vesicles under conditions permitting creation of the supramolecular hydrogel.

17. The method of Claim 16, wherein at least one polymer is a cellulose-based polymer, preferably an alkyl-modified cellulose polymer.

18. The method of Claim 16 or 17, wherein at least one polymer is a modified hydroxypropyl methylcellulose (HPMC-CX), wherein X is 12, 14, 16, or 18, preferably wherein X is 14 or 16.

19. A method of isolating EVs from a biological sample or food product containing EVs, comprising subjecting the biological sample or food product containing EVs to tangential flow filtration (TFF) comprising (a) processing the biological sample or food product containing EVs through a hollow fiber with an pore size that removes cell debris so as to produce a permeate, and (b) passing the permeate through a hollow fiber with a 100 kDa cut-off, so as to thereby retain isolated EVs in the retentate while eliminating smaller soluble molecules.

20. A method of injecting the supramolecular hydrogel of any of Claims 1-15 into an animal, comprising injecting an amount of the supramolecular hydrogel into the animal.

Images