Enzyme and nanozyme loaded microparticles

Abstract

This invention describes microcapsule compositions and methods to encapsulate gas modulating enzymes or nanozymes at unprecedented concentrations and stability, that can be used to regulate the local gas concentrations. Microcapsules can be comprised of gelatin with a polydopamine coating. This system can replace expensive incubators / chambers and is the first system that can create spatiotemporal gradients of gases within a finite volume. The microcapsules have a broad range of applications in tissue culture, bioprinting, in vivo, and industrial uses.

Description

[0001]ENZYME AND NANOZYME LOADED MICROPARTICLES Statement of Federally Sponsored Research This work has been funded by the National Cancer Institute of the National Institutes of Health # R01CA251443 and the Australian Research Council # DP210103654.The government has certain rights in the invention. Technical Field The invention relates to microcapsules loaded with active substances, such as enzymes or nanozymes, that allow diffusion of small molecules into and out of the microcapsule. The microcapsule layer protects the payload, preventing the entry of proteases and other degradative agents. This system is inexpensive and simple to produce with unprecedented loading capacity, and can be used in multiple formats, including deposition in hydrogels, as granular solids for 3D bioprinting, and as injectable biomaterials. Background Oxygen is one of the most important molecules that drives life on earth. For humans and other mammals, oxygen tension plays a critical role in aiding tissue development and maintaining normal tissue function, as well as pathological processes like tumour progression. Low levels of oxygen in tissue is termed hypoxia, and hypoxic conditions mediate the transcriptional activity of hypoxia inducible factors (HIF) which regulate gene expression associated with multiple functional activities including angiogenesis, cell migration, differentiation, metabolism, and apoptosis. These HIF proteins are sensitive to the local oxygen tension which varies greatly within and between different tissues of the body. While air has a saturation of around 20% oxygen, circulating blood levels sit around 5%, with various tissues such as bone marrow and cartilage reaching as low as 1%. These values contrast significantly from the 18.6% oxygen found in cell culture incubators, leading to cellular hyperoxia in vitro. For decades, creating an easy and effective way to fine tune the oxygen levels for in vitro cultures has remained a challenge. The current standard for mimicking low oxygen environments in vitro is the use of hypoxic incubators and hypoxic chambers. These devices control atmospheric oxygen levels and thus limit the amount of diffusible oxygen within the cell culture environment. However, there is little granularity with these tools. They are expensive, inaccessible to many researchers, and their oxygen levels reset when opened. To overcome these issues, researchers have explored the use of chemical and enzymatic methods to create hypoxic environments. While existing systems enable greater control of local oxygen levels, the methodology is specialized and can limit selection to a more localized hydrogel network. To broaden this concept’s versatility, an approach was sought that could apply to any hydrogel system while utilizing the benefits afforded via enzymatic oxygen control. This would enable long-term high control over the oxygen levels in of any 2D or 3D cell model without compromising cell- matrix interactions. To this end, it was aimed to entrap the enzymes within microcapsules. Since the substrates of glucose oxidases and catalase are small molecules, a polymer membrane that is large enough to enable small molecule diffusion, but too large to accommodate enzyme diffusion is needed. The capsules would require gentle processing such that a high quantity of active enzymes could be integrated, and the presence of capsules should not disrupt bulk network properties of the hydrogel or resident cells. Currently, there are no microencapsulation approaches that satisfy these criteria. Controlled gas environments are generally provided by chambers connected to gas tanks. Alternatively enzymes or nanozymes can be distributed into solutions to attenuate gas levels, which can control gas levels without chambers. As explained above, neither of these approaches can precisely control gases spatially and temporally. There exists a need for new approaches to control gas levels which can conveniently mimic biological environments. It is an object of the present invention to overcome or ameliorate at least one of the disadvantages of the prior art, or to provide a useful alternative. Any discussion of the prior art throughout the specification should in no way be considered as an admission that such prior art is widely known or forms part of common general knowledge in the field. Summary of the Invention In a first aspect, the invention relates to a population of microcapsules, wherein the microcapsules comprise gelatin coated with a nanoscale layer of polydopamine; are 1 um to 10 mm in size; and encapsulate a payload. In some embodiments the size is 1-100 um, 100-250 um, 250-500 um, 500 um-1 mm, 1-5 mm, or 5-10 mm. In some embodiments, the payload is at least one enzyme. In some embodiments, the at least one enzyme is glucose oxidase and catalase. In some embodiments, the payload is nanoceria. In some embodiments, the microcapsules have an average pore size of about 1 nm. In some embodiments, the microcapsules exclude proteases. In some embodiments, the microcapsules are in a yield-stress fluid, or jammed state. In some embodiments the microcapsules are in a jammed state alone or with other microscale materials, such as microscale gel. In some embodiments the jammed particles are used for 3D biofabrication or bioprinting, for example, on or into a matrix. In some embodiments, the 3D bioprinting can include extrusion (eg, through a 3D bioprinter) or ink jet printing of the microcapsules. In some embodiments, the microcapsules can be diluted and deposited drop-on-demand, eg, through the nozzle of a 3D ink droplet printer. In some embodiments, the microcapsules can withstand a shear stress of at least about 10 kPa. In some embodiments, the microcapsules can withstand shear stress of 10-100 kPa, 100-500 kPa, 500-1000 kPa, 1000-104, 104-105, or 105-106. In some embodiments, the population is lyophilized or cryopreserved. Microcapsules can slip and flow around each other at shear stress of 10-50 Pa, but the upper limit for rupture of microcapsules is extremely high. In some embodiments, the population is in a hydrogel. In some embodiment, the population is on a solid or semisolid matrix. In some embodiments, the population is in a pharmaceutical composition. In some embodiments, the population is in an environment. In some embodiments, the environment is in vitro or in situ cell culture (e.g., in a bioreactor) in vivo (e.g., localized to a specific organ or tumour), a bandage (to aid in wound healing), on a polymer matrix for coating surfaces; a contact lens (e.g., as a therapeutic) or in a chamber containing oxygen-sensitive liquid (e.g. a 3D or ink printer). In a second aspect, the invention provides a hydrogel comprising the population of microcapsules described above. In a third aspect, the invention provides a pharmaceutical composition comprising the population of microcapsules described above. In some embodiments, the pharmaceutical composition includes a suitable pharmaceutically acceptable carrier which are known to those skilled in the art. In some embodiments, the pharmaceutical composition is manufactured for parenteral, intratumoural, or otherwise localized administration. In a fourth aspect, the invention provides a solid or semi-solid matrix comprising the population of microcapsules above. In a fifth aspect, the invention provides a method of preparing a population of microcapsules comprising i) adding gelatin and payload to an oil bath to form an emulsion comprising loaded microcapsules; ii) washing the loaded microcapsules; iii) adding dopamine hydrochloride to form a polydopamine coating on the loaded microcapsules; iv) washing the coated, loaded microcapsules. In some embodiments, the method comprises agitating or stirring the mixture in steps i) and / or iii). In some embodiments, the method is carried out at pH 7.5-9. In some embodiments, the method is carried out at pH 8.5. In a sixth aspect, the invention provides a method of bioprinting comprising printing the population of microcapsules described above. In some embodiments, the population is in a jammed state. In some embodiments the population is combined with a microgel. In some embodiments, printing is by extrusion or by ink drop on demand. In a seventh aspect, the invention provides a method of reducing oxygen levels in an environment comprising adding the population of microcapsules described above to the environment. In some embodiments, the environment is an in vitro cell culture. In some embodiments, the environment is a sample comprising anaerobic bacteria. In some embodiments, the environment is an ex vivo tissue sample or organoid. For example, the microcapsules can be used in tissue engineering to artificially lower oxygen thereby helping angiogenesis / neovascularisation in the sample. In some embodiments, the environment is a chamber containing oxygen-sensitive liquid. In some embodiments, the environment is in vivo and the population is used to stimulate angiogenesis and neovascularisation in a tissue. (See e.g. Figure 6). In an eighth aspect, the invention provides a method for reducing the size of a tumour in a subject comprising administering the population of microcapsules described above to the subject. In some embodiments, the invention provides the population of the microcapsules described above for use in reducing the size of a tumour in a subject, wherein the population is administered to the subject. In some embodiments, the invention provides use of the population of microcapsules described above in the manufacture of a medicament for reducing the size of a tumour in a subject. In some embodiments, the administering is localized to the area of the tumour. In a ninth aspect, the invention provides a method of reducing reactive oxygen species in an environment comprising adding the population of microcapsules described above to the environment. In some embodiments, the environment is an in vitro cell culture. In a tenth aspect, the invention provides a method of reducing reactive oxygen species in a subject comprising administering the population of microcapsules described above to the subject. In some embodiments, the invention provides the population of microcapsules described above for use in reducing reactive oxygen species in a subject, wherein the population is administered to the subject. In some embodiments, the invention provides use of the population of microcapsules described above in the manufacture of a medicament for reducing reactive oxygen species in a subject. In some embodiments, the subject has been exposed to radiation. In some embodiments, the administration is localized to the irradiated area. Description of the Drawings Figure 1. Schematic of an exemplary capsule synthesis procedure. First, 37 °C gelatin is loaded into a stirred warm oil bath. After cooling and an addition of acetone, the particles are washed in hexane and air dried. Then are subsequently placed in a Tris-HCL buffer (pH 8.5 50mM) with 10 mg / mL of dopamine and shaken at 300 rpm for 24 hours before 4 PBS washes. Figure 2. Schematic of hypoxic capsules in microgels. A) Reaction scheme of glucose oxidase and catalase. B) Schematic representation of glucose oxidase enzyme within a capsule. The enzymes are entrapped within the capsule walls, but the wall pore sizes are small enough to enable the exchange or enzyme substrates (glucose and oxygen). (Right) Schematic representation of oxygen levels around an individual capsules (top) and a capsule laden hydrogel in cell culture medium. Figure 3. Characterization of PDA gelatin microcapsules. A) (i) Bright field microscopy images of uncoated gelatin microgels. Scanning electron microscope images of uncoated (ii) and PD coated (iv) gelatin capsules. (iii) Bright field microscopy image polydopamine (PD) coated gelatin microgels. B) Size distribution histogram of uncoated gelatin microgels. C) Normalized permeability of capsules based on fluorescence. D) Confocal microscopy images of capsules (coated for 0, 4, or 24 hours) sitting in FITC-Dextrans for 24 hours (37 °C). E) (Left) Confocal microscopy image of 24-hour PD coated microcapsules loaded with 250 kDa FITC-Dextran (2 mg / mL). (Right) Plot of encapsulation efficiencies calculated for 10, 40, and 20kDa dextrans. F) Plot of normalized fluorescence intensity of FITC-BSA in microcapsules after 28 days. G) Plot of the final concentration for 5 batches of BSA loaded microcapsules. H) Viscosity vs shear strain rate plot from shear strain sweeps (log ramp from 0.02-10 s-1). I) a photograph of a printed continuous line filament of jammed microcapsules. J) Bright field microscopy images of jammed microcapsule printed lines with various nozzle sizes (23, 22, and 18G). K) A plot of analyzed microcapsule line diameters with varied nozzle dimeters. Red lines indicate the inner diameter of the nozzles used. Scale bars: 300 µm. Scale bars: 10 µm (A (ii and iv)), 50 µm (A (i and iii), E), 300 µm (J). **P < 0.01, ***P < 0.001, ANOVA. Error bars represent s.d. Figure 4. Characterization glucose oxidase activity in capsules. A) Glucose oxidase (GOx) activity overtime for glucose oxidase loaded microcapsules (1 U / mL) measured with a glucose oxidase activity kit on plate reader with 517 nm emission. B) Long term activity of glucose oxidase capsules stored at 4 °C in between measurements. C) Long term activity of glucose oxidase capsules and free enzyme stored at 37 °C in between measurements. D) Plot of the dissolved oxygen partial pressures for varied concentrations of glucose oxidase (with a 1:60 ratio with catalase) in cell media at room temperature. E) plot of the saturation half time versus glucose oxidase concentration. The two x’s correspond to data points that were not used for the linear fit. F) A plot of the dissolved oxygen partial pressures for capsules at varied concentrations (5, 12.5, or 25 mg / mL of glucose oxidase loaded capsules (1000 U / mL, 60,000 U / mL catalase)). ***P < 0.001, ANOVA. Error bars represent s.d. Figure 5. Effect of Hypoxia on ADSCs and tumour cells. A) Confocal z-stack projections of ADSC cells (1 × 106cells per mL) stained with Hoechst (Nuclei, blue), phalloidin (Actin, not shown), and HIF1-alpha (red) after 1 or 3 days in culture in microgel suspensions (10 wt%, 1% filler) with or without 2 vol% of glucose oxidase capsules (1000 U / mL, 60,000 U / mL of catalase). B) Cell counts per replicate region of HIF1-alpha positive cells for each of the glucose oxidase concentrations (capsules loaded with 0, 100, 250, and 1000 U / mL). Concentrations (C), size plots (D), and average diameters (E) of EVs isolated from ADSCs in microgel suspensions with capsules loaded with varied glucose oxidase concentrations. F) Confocal z-stack projections of A375-P cells stained with Hoechst (Nuclei, blue), phalloidin (Actin, red), and HIF1-alpha (green) after 1 or 7 days in culture in microgel suspensions (Medium, 10 wt%, 1% filler) with or without 2 vol% of glucose oxidase capsules (100 U / mL, 6,000 U / mL of catalase). G) Plots of average A375-P aggregate size per replicate overtime. H) Optical images of MCF7 breast cancer spheroids embedded in Geltrex over 6 days with no capsules or with capsules loaded with an equivalent of 10 U / mL. I) Growth curves of breast cancer spheroids diameters. Scale bars: 50 µm (A, F), 200 µm (H). *P < 0.05, **P < 0.01, ANOVA. Error bars represent s.d. Figure 6. Effect of hypoxic capsules on endothelial cells. Plots of relative expression of VEGFA (A) and ANGPT-1 (B) for HUVECs in microgel suspensions with varied amounts of hypoxic capsules for 24 hours. Plots normalized to GAPDH. C) Confocal z-stack projections of HUVEC (5 × 106cells per mL) cells stained with Hoechst (Nuclei, blue), and phalloidin (Actin, red), after 5 days in culture in microgel suspensions (10 wt%, 0.5% filler) with or without 2 vol% of glucose oxidase capsules (100 U / mL, 6,000 U / mL of catalase). D) Plots of master segment length (right) and number (left) and lengths for control and glucose oxidase capsule conditions calculated using the angiogenesis analyzer plugin for ImageJ. E) Bright field images of HUVEC tubulogenesis over the course of 24 hours. F) Analysis of master segment length from the tubulogenesis assay. G) Dissecting microscope images of representative CAM assay egg embryos after 4 days of incubation with or without microgel suspensions loaded with glucose oxidase capsules. H) Quantification of the number interacting branching points from the CAM assay. Scale bars: 50 µm (C), 100 µm (E), 500 μm. (G). *P < 0.05, **P < 0.01, ***P < 0.001, ANOVA. Error bars represent s.d. Figure 7. Nanoceria rods loaded into microcapsules. A) Reaction scheme of nanoceria’s catalase activity (Top), schematic of nanoceria in capsules (left), and TEM image of nanoceria rods (Right). B) Confocal z-stack projection of live ADSCs stained with Hoechst (Nuclei, blue) and MitoSOX (superoxide species, green) in either control microgel suspensions (Medium, 10 wt%, 1% filler) or in suspensions with 2 vol% ceria loaded microcapsules (5 mg / mL). C) Normalize intensity plots of cellular MitoSOX levels take from the fluorescence images. D) High resolution image of MitoSOX and Hoechst stained ADSCs. E) Confocal microscopy z- stack projections of live dead stained ADSCs (1 × 106cells per mL) after 24 hours of exposure to H2O2 (0, 100, 1000 µM) in microgel suspensions (Medium, 10 wt%, 0% filler) with either no capsules, 2 vol% ceria loaded capsules (5 mg / mL), or 2 vol% catalase loaded capsules (60,000 U / mL). F) Cell viability plots of the data captured on confocal. Scale bars: 200 nm (A), 5 µm (D), and 50 µm (B, E). *P < 0.05, **P < 0.01, ***P < 0.0001, ANOVA. Error bars represent s.d. Figure 8. Spatial control of oxygen directs organoid morphogenesis. A) Blank GelMA C hydrogel is printed as a disc with 0.5 cm internal and 2 cm external radius. Same hydrogel with suspended glucose oxidase microcapsules was dispensed into the centre to generate a source-sink disc pattern. Scale = 1 cm. B) Bioprinted disc with excess culture media in cell culture incubator. Oxygen levels measured along the radial direction, r, from the disc centre at 5 mm intervals for up to 1 day. C) Computational simulations (COMSOL) modeling of hydrogels in media with loaded capsules after 1 hour (top) or 24 hours (bottom). D) Plot of computationally modelled oxygen partial pressures in solution that mimics geometry of the bioprinted disc. E) Oxygen distributions of bioprinted disc incubated in 30 mL of cell culture media for 4 days. No media change was performed. F) Oxygen distribution of bioprinted disc incubated in 30 mL cell culture media with daily culture media change. G) Schematic of capsules loaded in a gel droplet near two Matrigel domes at varied distances. H) Optical images intestinal organoids in Matrigel domes that are exposed to varied levels of hypoxia (top), and fluorescent images of intestinal organoids stained with a hypoxia IT dye (bottom). I) plot of organoid diameters as a function of which Matrigel dome and oxygen conditions they were in. Scale bars: 1 cm (A), 500 µm (H). Data are presented as average of n > 3 repeats with S.D. ***P < 0.001, ANOVA. Error bars represent s.d. Description The present description provides the first microscale gas-modulating system with spatiotemporal control. In this case, control over oxygen tension in cell culture was demonstrated. The novel microencapsulation system uses a gelatin microgel template with a polydopamine coating, where tuning the coating polymerization alters the permeability and stability of capsules. This approach provides unprecedented concentrations of enzyme, up to 40 mg / mL within the microcapsules, with stable activity maintained for over a month in cell culture conditions. A dose-response in oxygen levels with enzyme concentration was achieved and demonstrated functional outcomes through reproducible control over stem cell and cancer cell activities. Local attenuation of oxygen levels was shown to increase neovascularization in 2D, 3D, and in ovo models, providing scope for these materials to be used in promoting in vivo angiogenesis. As a demonstration of versatility, nanozymes were added to capsules as an alternative approach to modulate the gas balance, demonstrating reduction in active oxygen species to protect adipose derived stem cells in a tissue irradiation model. Overall, the present disclosure demonstrates a versatile system where gas-modulating capsules can be integrated with virtually any soft material system to locally control gas levels. The present invention provides a method to encapsulate gas modulating enzymes or nanozymes at unprecedented concentrations and stability. These can be used to regulate the local gas concentrations. Therefore, this system can replace expensive incubators / chambers and is the first system that can create spatiotemporal gradients of gases within a finite volume. Definitions The term “comprise” and variants of the term such as “comprises” or “comprising” are used herein to denote the inclusion of a stated integer or stated integers but not to exclude any other integers unless in the context or usage an exclusive interpretation of the term is required. As used herein, the term “microcapsule” refers to a small capsule encapsulating a payload with a specified pore size that allows small molecules to diffuse in an out of the microcapsule. The size of the microcapsule can range from about 1 nm to about 10 mm, depending on the encapsulation technique and microcapsule components. When referring to the size of the microcapsules, the size given is the D50 value, as determined by the diameter of generally spherical capsules. A “population of microcapsules” refers to multiple microcapsules, generally with similar characteristics and payload. Reference to the microcapsules in a population is general, and does not indicate that all of the microcapsules have the exact features described (dimensions, payload concentration, pore size, etc.). The term “payload” refers to the substance(s) loaded and contained within the microcapsule. It must be larger than the pore size of the microcapsule (about 1 nm). Exemplary payloads are enzymes and nanoceria. “Nanoceria” are cerium oxide nanoparticles or metal oxide nanoparticles that possess outstanding reactive oxygen species (ROS) scavenging activities. They include inherent catalase, SOD, oxidase, peroxidase and phosphatase mimetic properties. See, e.g., Saifi et al. (2021) J Controlled Release 338:164. A “hydrogel” is a three-dimensional network of colloidal dimensions with interpenetrating solid and aqueous phases. Hydrogels can absorb large amounts of water. They are particularly effective in the context of the present invention, as they provide support for enzymes within the microcapsule and greatly increase activity and stability. See, e.g., Peppas et al. Plymer Science: A comprehensive reference, 2012. A “matrix”, e.g. solid or semisolid matrix, is a surface or environment that can hold or support the presently described microcapsules. Hydrogels are a type of matrix. Microcapsules in a “jammed” state are in a yield-stress fluid. In this state, the jammed microparticles are very tightly packed and behave as a solid until they are sheared. The concentration and density of the jammed microcapsules depends on the size of the microcapsules. Other materials can be printed within the jammed suspension. The microcapsules can be in a jammed state alone or with other microscale materials, such as microscale gel. The term “reactive oxygen species” (ROS) refers to highly reactive variants of diatomic oxygen, and include superoxide (*O2-). The terms “reducing oxygen levels” or “reducing reactive oxygen species” refer to a reduction in the level of the component after a given treatment compared to the level before the treatment. The reduction can be about a 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or higher amount. In some embodiments, the reduction can be determined by comparison to a population of comparable environments, cell cultures, or individuals that are not treated. The term “inducing angiogenesis” or “inducing neovascularisation” refers to stimulating formation of new blood vessels and / or capillaries in response to biological signaling, for example hypoxia signaling in an in vitro, ex vivo, or in vivo tissue. The term can be interpreted as relative to the level of angiogenesis before treatment with the presently described microcapsules. The term “reducing the size of a tumour” refers to a reduction in the level of an existing tumour. The reduction can be about a 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or higher amount, or an elimination of the tumour to an undetectable size. The tumour can be a primary tumour or a secondary (or higher level) tumour resulting from metastasis. Overview An aqueous solution of enzymes or nanozymes can be mixed at high concentration (>20 mg / mL) with gelatin and oil to form an emulsion. The enzyme-gelatin droplets can be coated with a nanoscale layer of polydopamine to entrap the enzymes within the particles. Modifying the emulsion technique can yield particles from 1 um to 10 mm. This can be carried out by varying the agitation frequency and strength, adding surfactants or salts, varying temperature, and / or using microfluidics. For example, span surfactants, tween surfactants, lipids, or plurionic surfactants can be used. Distributing the particles within solution or hydrogels allows small molecule substrates to diffuse through the membrane while keeping large macromolecules like the enzymes entrapped within. The results provided herein show the highest reported concentration of enzymes encapsulated within a microscale capsule, which is key to the success of the technique. While the approach is general to virtually any biomolecule / nanoparticle, the present inventors demonstrated proof of concept using a combination of glucose oxidase and catalase. The ability of this system to make an oxygen gradient within a finite volume is a powerful aspect of the invention. When glucose is added to the solution the two enzymes convert 1 glucose molecule and 2 oxygen atoms into gluconic acid and 1 oxygen atom. In this way the environment can be made hypoxic. Oxygen levels in tissue range from 1-14%; however, the oxygen in the atmosphere is 21%. In order to recreate the hypoxic environment, researchers need specialised chambers, which are expensive and can only provide one gas concentration at a given time. The presently described microcapsules allow gas levels to be finely tuned through controlling the amount of encapsulated enzyme, the number and localization of microcapsules, and the concentration of glucose in solution. Furthermore, the dynamic activity of the enzyme reactions provides a spatial gradient as a function of distance from the capsules. This emulates the gradients observed in living tissue. Controlling oxygen and hypoxia in culture without the need for specialised chambers will be highly beneficial to cell biology, tissue engineering and body-on-a-chip technologies. There is broad scope for the invention to impact numerous markets where control of gases or other enzyme products benefits from spatiotemporal control. This can include various chemical industries where enzymes are employed in reactors or products. In the case of oxygen modulating microcapsules, the invention can be used in 3D cell culture as the capsules could be marketed as reagents to directly control oxygen in cell culture without needing specialised chambers. The presently described oxygen-modulating microcapsules can also be used in the biofabrication / 3D bioprinting industry, where microcapsules can be precisely positioned during printing to spatially control oxygen content in a tissue scaffold. Jammed microcapsules, alone or in combination with other microscale materials (e.g., gels) are useful for 3D biofabrication or bioprinting, for example, on or into a matrix. The 3D bioprinting can include extrusion (e.g., through a 3D bioprinter) or ink jet printing of the microcapsules. In some embodiments, the microcapsules are diluted and deposited drop-on- demand, e.g., through the nozzle of a 3D ink droplet printer. This technology can also be used for microbial therapeutics. The ability to control oxygen is highly beneficial for the collection of stool samples where beneficial anaerobic bacteria require low oxygen to survive. Furthermore, cell cultures where a gradient of oxygen can be established will allow new assays for microbiome research and development. Examples 1. Materials and methods 1.1 Gelatin microcapsules synthesis For empty capsules, type B Gelatin (Bloom strength 300) was dissolved at 10 wt% in 1X PBS in a 40 °C water bath (75 mL of oil). For loaded capsules, a 20 wt% solution of gelatin was dissolved, and it was mixed 1:1 with a solution containing the cargo. The gelatin solution (2 mL) was pushed through a 0.45 µm filter into an oil bath (olive oil) under constant stirring at 38 °C. After 10 minutes of equilibration, the oil bath was cooled to 10 °C, using a surrounding ice bath, and left to stabilize for 30 minutes. Hexane (25 mL) was subsequently added and left to stir for 10 minutes before the microgels were decanted off into centrifuge tubes. The particles were left to settle for 30 minutes prior to 4 washing steps with hexane. The particles were agitated in the hexane solution and quickly pipetted onto a 25 µm nylon filter over Kim wipes in a fume hood to let the hexane dry off. A spatula was used to spread out the particles on the filter to ensure all hexane was evaporated. Once hexane free, the particles were placed into a centrifuge tube again and resuspended in a Tris-HCl buffer (0.1M pH 8.5, 10 mL). Dopamine hydrochloride was added (final concentration of 10 mg / mL) and the tube was covered with a Kim wipe to allow oxygen exchange. The tube was then placed on an orbital shaker at 300 rpm for the specified time. The particles were then washed with 1X PBS four times and stored until further use. For size characterization, uncoated particles were placed onto a glass slide and imaged under an optical microscope. One hundred particles were imaged across 7 images, and the diameters were measured using ImageJ. 1.2 Encapsulation efficiency and Maximal loading To quantify encapsulation efficiency, the capsule synthesis procedure was followed (loaded with 10, 40, or 250 kDa FITC-dextran cargo (20 µg / mL)) until just before adding in the dopamine. The particles were then placed on the orbital shaker for 4 hours to simulate the time prior to a stable coating forming on the surface of the microgels. Microgels were then weighted out and dissolved at 37 °C in 1X PBS to extract the contents. The contents were then measured using a fluoresce spectrophotometer with a laser excitation at 488 nm (Cary eclipse). The emission at 517 nm was plugged into a line of best fit from a standard curve to calculate the concentration of Dextrans in solution. For maximal BSA loading, the same procedure was followed expect the gelatin particles were loading with 100 mg / mL of Bovine Serum Albumin (BSA) and 125 µg / mL of fluorescent FITC- labelled BSA. A separate standard curve for FITC-BSA was produced to create a line of best fit. 1.3 Cell Culture and Maintenance All cell culture was performed using aseptic techniques inside a sterile class II biological safety cabinet. All cells were cultured inside an incubator at 37 °C with a 5% CO2 and 100% humidity atmosphere. All cells used in this study were between passages 2-13. Media changes were performed every 2 days. The adipose derived stem cells (ADSC), B16F10, MCF-7, MDA-MB-231 and A375-P cells were all passaged when a confluency of 80- 90% was reached. Human umbilical vascular endothelial cells (HUVECs) were passaged at 70-80% confluency. For all cell lines, 0.25% Trypsin-EDTA was used for detachment. 1.4 Immunofluorescence Staining and Tissue Clearing Clearing solutions were prepared as described previously with minor modifications (Molley et al. (2021) Biomater Sci 10:1039 / D1BM00574J. Cubic solution 2 was prepared by mixing 50 wt% sucrose, 25 wt% urea, 10 wt% triethanolamine with DI water at 55°C until also fully dissolved. Microgel suspensions were fixed using 4 wt% Paraformaldehyde (PFA) for 1- 4 days at room temperature to ensure full penetration of PFA into thick constructs. The gels were then rinsed with 1X PBS followed by 31X PBS washes at 2-4-hour intervals. Afterwards, the gels were incubated in 0.5 wt% triton overnight before repeating the same wash procedure. The staining for all gels was performed using the dyes and antibodies in Table S2. The gels were washed with 1X PBS three times after both primary and secondary staining before the addition of the Cubic 2 clearing solution for 1-4 days. All confocal imaging was performed with a Zeiss LSM 800. A 10x objective with a 2.5 mm working distance was used to see deeper into the samples. Samples were coated with clearing 2 solutions throughout the duration for the imaging to prevent drying. 1.5 Glucose oxidase activity characterization Glucose oxidase activity was characterizing by using the glucose oxidase activity assay kit according to the manufacturer’s protocol. Briefly, glucose oxidase was first dissolved to 20 mg / mL (2000 U / mL) in a 50mM sodium acetate buffer at a pH of 5.0. The enzyme was mixed 1:1 with a 20 wt% Gelatin solution to make microcapsules that were coated for 24 hours. Two other microcapsules were then synthesized: one with 20% as much glucose oxidase, and one with no glucose oxidase. The capsules were then place on 25 µm nylon filters, to remove excess water, and were weighed into an Eppendorf tube. The particles were then resuspended to 20 mg / mL in the GOx buffer solutions from the kit. Then, 50 µL of the capsule buffer solution was added to 4 different wells in 96 well plate, along 4 wells for blanks, and 5 wells for the standard curve. The plate was then placed into a plate reader (CLARIO Star Plus) set to 37 °C to warm up. Finally, 50 µl of the kit working solution was added to each well and measurements were taken every 3 minutes for 45 minutes total. Unused microcapsules were then stored in 4 °C for 1 month, with new measurements taken on days 2, 7, 14, and 21. 1.6 Seeding cells with glucose oxidase and catalase microcapsules For cancer cell response, microgel suspensions were loaded glucose oxidase capsule laden microgels with A375-P (1 × 106cells per mL) and cultured for 1 and 7 days. Cells were subsequently fixed and stained with Hoechst, Phalloidin, and HIF1-alpha antibody. The images were subsequently taken on a Zeiss LSM 800 confocal and imaging (100 µm z-stack with 2 µm slices). Aggregate size analyses were performed in ImageJ. Briefly, the actin cytoskeleton of aggregates was thresholded and masked and the Analyze particles plugin was used to calculate the average aggregate size per replicate. For the ADSC cell response, microgel suspensions (medium, 10 wt%, 1% filler) were loaded with 2 vol% microcapsules (0, 10, 25, or 100 U / mL glucose oxidase with a 1:60 ratio of catalase) and 1 × 106ADSCs per mL. For images, 100 µm z-stacks with 2 µm slices were taken. To calculate mean grey value intensities, the phalloidin channel was used to create regions of interest (ROIs) of cell boundaries, and then these ROIs were applied to the HIF1- alpha channel to achieve the mean grey of hif1-alpha staining per replicate region. For cell counting, the nuclei were thresholding and counted via the analyze particle’s function, and HIF1alpha expressing cells were counted manually. 1.7 Irradiation protection study Two sets of microcapsules were prepared; one with 5 mg / mL of ceria nanoparticles loaded (dissolved in DI water), and one with 6 mg / mL of catalase loaded (dissolved in 1X PBS). Three separate microgel suspensions were made using medium-sized 10 wt% GelMa particles, 0.5 wt% filler, and 0.05 wt% of LAP: one with 20 mg / mL of nanoceria capsules, one with 20 mg / mL of catalase capsules, and one with no capsules. ADSCs were loaded into each suspension at 1 × 106cells per mL, and the suspensions were UV crosslinked in plastic molds. The crosslinked suspensions were placed in 24 wells plates with 1 mL of media and placed in an incubator for 2 hours. Then, the media was replaced with fresh media containing either 0, 10, 100, or 1000 µM of H2O2 and cultured for 24 hours. The media was then removed, and the sample were washed once with 1X PBS. A 1X PBS solution of Calcein AM (2 µM) and Ethidium Homodimer-1(4 µM) was added (500 µL) to each well and the gels were incubated at 37 °C for 40 minutes. Afterwards, the samples were washed with 1X PBS, placed back in the incubator for 5 minutes, and washed again before imaged. Quantification was performed using the same method as described in the supplemental methods. 1.8 Statistical analysis All whiskers in box plots are one standard deviation (s.d.). For particle size study, 100 different particles were used. Four replicates were used for the permeability study, and 3 replicates used per condition for the encapsulation efficiency study. Five replicates were used for the BSA capsule stability study, as well as the maximal BSA loading test. For 3D printed Line fidelities, 8 separate lines were used, with the diameter averaged across 6 different points along the line. Quadruplicates were used for the Glucose oxidase activity studies, and triplicates for the dissolved oxygen tests for the capsules. Triplicates were used for the live dead studies. Quadruplicates were used for all ADSC studies, triplicates were used for the A375 studies, and six replicates were used for the breast cancer studies. No replicates were performed for the western blots. Five replicates were used for all HUVEC studies, and 3 eggs were used for each replicate in the CAM assay. Triplicates were used in the MitoSOX study and the irradiation model test. Statistical significance was determined using a one-way ANOVA with Tukey’s Post Hoc HSD analysis. Differences were considered significant when P < 0.05. Statistical significance was highlighted in figures with the following convention: * = P < 0.05; ** = P < 0.01; *** = P < 0.001, **** = P < 0.001. 2. Results and discussion 2.1 Characterization of polydopamine coated gelatin microparticles The method of oxygen reduction relies upon the dual reaction of glucose oxidase paired with catalase. The glucose oxidase enzyme first oxidizes a glucose molecule, forming a gluconic acid and hydrogen peroxide molecule. Catalase in turn breaks two hydrogen peroxide down into two water molecules and an oxygen molecule. Together, this reaction reduces one dissolved oxygen molecule for every two glucose molecules (Figure 2A). Previous use of this enzyme pair in vitro has been shown effective at creating hypoxia in cell culture conditions, but the free enzymes had poor stability and lost function quickly over time. Without wishing to be bound by theory, the present inventors have hypothesized that microencapsulation of these two enzymes in a stable hydrogel environment would provide a simple yet effective method to prolong enzyme life and stability while maintaining efficacy. For the capsule design, gelatin was chosen as the capsule core due its low thermal gelation temperature which enables easy microgel formation while preserving payload bioactivity. Polydopamine (PDA) was chosen as the capsule wall since it coats most materials with ease and has a mesh size that is smaller than the enzymes but larger than glucose (Pore diameters: PDA ~ 1 nm; glucose oxidase radius ~ 6 nm). This enables diffusion of analytes in and out of the capsules while preventing enzymes from escaping or proteases from entering and degrading the enzymes. Once loaded, these capsules function as oxygen sinks, with ease of dispersion and spatial localization for control over oxygen tension (Figure 2B). The kinetics of oxygen reduction can subsequently be tuned by varying either the number of embedded capsules or the concentration of enzyme within each capsule. Spatial arrangement of capsules with different enzyme content is also possible to provide gradients of gas concentration. The microcapsule synthesis scheme using a water in oil emulsion method is shown in Figure 1. To maximize loading potential, cargo is added to the gelatin solution at high concentration prior to addition to the oil bath. Importantly, previous methods to generate microgels of gelatin or gelatin methacryloyl (GelMa) with oil emulsions use acetone to dehydrate the particles prior to washing and collection. However, acetone will render loaded enzymes inactive. Therefore, the particles require a non-water-miscible solvent to wash away excess oil. Hexane was chosen as it readily dissolves food oils while remaining immiscible with water. Optical images verify the formation of stable gelatin microgel particles without acetone dehydration (Figure 3Ai); scanning electron micrographs show smooth morphologies on the uncoated particle surfaces (Figure 3Aii). For the coating, dopamine hydrochloride (10 mg / mL) was added to a dilute suspension of gelatin microgels (10 w / v% particles in Tris-HCL Buffer (pH 8.5)) and shaken in a capless centrifuge tube to enable gas exchange. After 24 hours, the particles adopted a brown / black appearance (Figure 3Aiii), corresponding to the color of polydopamine (PD). SEM showed a slightly rougher surface on coated particles (Figure 3Aiv), and size characterization revealed an average particle size of 60 ± 24 µm (Figure 3B). To determine the barrier functionality of the PDA coating, a permeation study was performed with gelatin microparticles coated for either 0, 2, 4, or 24 hours. The capsules were placed in a PBS solution with FITC labelled dextrans (250 kDa) for 24 hours and subsequently imaged using confocal microscopy to measure capsule fluorescence at 20 and 37 °C. There was a noticeable decrease in capsule fluorescence as coating time increases for the 20 °C condition, though minimal. When assessing the 37 °C conditions, it became clear that stable coatings did not occur until after 4 hours of coating (Figure 3C-D). Even after 4 hours, faint silhouettes of capsules can be seen with a significant difference in fluorescence when compared to the room temperature condition (P < 0.0001), suggesting a leaky coating. However, after coating for 24 hours, there is a slight difference in permeability between the two temperature conditions. Therefore, 24-hour coatings were subsequently used for the rest of the work. It was next sought to verify that the capsules of the present invention can encapsulate cargo. To measure encapsulation efficiency, loaded capsules (2 mg / mL of 10, 40, or 250 kDa FITC- Dextran) were synthesized identically but without any dopamine added to the coating solution. The particles were then left to shake for 4 hours without dopamine for only four hours to simulate the time until a stable coating would form and prevent further leakage (Figure 3D). Confocal microscopy images of capsules loaded with 250 kDa dextrans were first used to confirm loading (Figure 3E). The uncoated capsules were then collected, washed, and melted. A fluorescent spectrophotometer was used to measure the amount of FITC-dextran with an emission maximum of 517 nm. To quantify the dextran concentration, standard curves of FITC- dextrans at varied concentrations were made for each size (10, 40, and 250 kDa). Confocal microscopy images of capsules loaded with 250 kDa dextrans were first used to confirm loading (Figure 3E). The encapsulation efficiencies for 10, 40 and 250 kDa dextrans were 32 ± 2%, 32 ± 4%, and 30 ± 0.5% respectively (Figure 3E). Material loss most likely occurs during the coating process before a sufficiently thick enough coating is formed. To measure the long-term stability of the coatings, FITC labelled bovine serum albumin (BSA) molecules were loaded into capsules and stored at 4 °C. These capsules were subsequently imaged under a confocal microscope over 28 days. To combat the gradual photobleaching of FITC, the internal particle fluorescence was normalized to a solution of free FITC-BSA stored under the same conditions. A 94 ± 1% retention in fluorescence is seen over the 28 days (Figures 3F). To test the maximal loading capacity of the capsules of the present invention, 10 wt% BSA (with 0.1% being FITC-BSA) was added to the gelatin precursor solution. A similar procedure to the encapsulation efficiency was performed on supernatants of dissolved non- coated capsules after 4 hours. A standard curve was made for the FITC-BSA which gave a final BSA concentration of 39 ± 2 mg / mL of BSA (39% BSA, wet weight) (Figure 3G). As far as the present inventors are aware, this is the highest loading of bioactive molecules in a microcapsule to date. Microscale hydrogels have been widely used as delivery vehicles, scaffolds and support matrices for biofabrication. It was next aimed to determine if a jammed suspension of capsules would behave similarly to standard microgel suspensions. To induce jamming, capsules were placed on 25 µm nylon filters to drain away excess water. Parallel plate rheology on jammed microcapsules revealed similar shear thinning behaviour (Figure 3H) with a yield stress value of 40 Pa. To demonstrate reversibility of their jammed state, 2-minute intervals of high (200%) and low (0.2%) strain were applied cyclically. Here, the jammed capsules can be seen to recover full mechanical stiffness near instantaneously for multiple continuous cycles. To assess injectability, jammed microgels were loaded into 1 mL syringes to be extruded. Notably, when the particles have not drained enough water, the ink does not print as a single jammed line. If too wet, the ink forms small droplets at the nozzle tip. As the suspension loses more water, the droplet begins to shrink in size and small fragments of jammed particles can be printed. However once adequately dry and jammed, the suspension prints as a single unbroken line (Figure 3I). These findings correspond with those found in recent work where unjammed microgel inks will expel water print poorly until a critical microgel concentration is reached. Finally, a fidelity assessment of 3D printed lines demonstrates high control of line with widths varying from 300 – 800 µm, with average capsule line width diameter less than 5% from of the nozzle’s inner diameter (Figure 3J-K). These studies demonstrate how the microcapsules can be used in multiple formats from single particles to high density suspensions for injection. 2.2 Glucose oxidase loaded capsules reduce dissolved oxygen levels It was next evaluated the ability to load functional enzyme into the microcapsule system of the present invention. First, glucose oxidase was loaded into microgel capsules (1 U / mL). A glucose oxidase activity kit with a fluorescent reporter was used to measure the hydrogen peroxide production activity of enzyme loaded microcapsules. Capsules (20 mg / mL) were added to wells in a plate reader and the fluorescent output (515 nm) was measured every 3 minutes for 45 minutes. When measuring the activity around the capsules over time, the enzyme activity reached a plateau after 3-6 minutes (Figure 4A). This is likely due to initial diffusion constraints as glucose substrates and peroxides need to freely diffuse into and out of capsules. After 45 minutes, capsules loaded with 1 U / mL of glucose oxidase had an activity of 46 mU / mL, providing a functional encapsulation efficiency of 5% (Figure 4B). This discrepancy between these values and the 30% found for loaded BSA may be due to loss of functionality of some of the encapsulated enzymes during encapsulation and coating at elevated pH (~ 8.5). The substrates and products of the reaction also need to diffuse through the capsule wall measurement which can limit the reaction rate and the perceived activity. In addition, immobilized enzyme activity can decrease by up to 80% when compared to free enzyme. This underscores the importance of an encapsulation system with high cargo loading capacity. Glucose oxidase is known for having high shelf life under proper storage conditions. To measure the shelf life of the enzyme loaded capsules at 4 °C in PBS, a batch of capsules had activity measured at 1, 3, 7, 14, and 21 days after synthesis. No noticeable difference in glucose oxidase activity was found after 21 days in storage demonstrating high stability (Figure 4C). Next, glucose oxidase loaded capsules were stored in cell culture media (DMEM) at 37 °C for five weeks, along with a solution of free enzyme as control. After 1 week, the capsules lost only 11 ± 0.3% enzyme activity (P = 0.0094), whereas the free enzyme had a reduction of activity of 84.4 ± 0.1% (P < 0.0001). After 5 weeks, the capsules lost only 34 ± 0.4% enzyme activity (P < 0.0001), whereas the free enzyme had lost 99 ± 0.2% enzyme activity (P < 0.0001). Together, these results demonstrate how the microencapsulation system of the present invention yields excellent enzyme stability providing a tool that could maintain hypoxic cell cultures for weeks. With confirmation that capsules loaded with glucose oxidase can generate hydrogen peroxide, it was next aimed to verify that encapsulated glucose oxidase, along with catalase, can reduce oxygen levels. For all tests, catalase was loaded at a functional activity of 60 times that of glucose oxidase. This ratio is to ensure that hydrogen peroxide produced by glucose oxidase is immediately extinguished by the catalase. This ratio is also consistent with previous work using enzymes to create hypoxic cell culture environments. To best simulate the oxygen level conditions for cell culture conditions, a dissolved oxygen (DO) probe was placed 7 mm deep (the depth of the hydrogels used in the present invention in a 24 well plate with 1 mL of liquid) into a 12 well plate loaded with 4 mL of cell culture media. A 6 mm stir bar was added to the well and rotated at 250 rpm to ensure enough fluid flow at the probe interface for accurate measurement while preventing disruption of the air-liquid interface. All measurements were performed at 21 °C. When testing at room temperature and pressure, the probe baseline read 9.07 mg / L of oxygen which, when using Henry’s law, gives a partial pressure of 21.0% oxygen. The dissolved oxygen levels were measured for glucose oxidase concentrations of 0.03, 0.1, 0.3, 1, 3, and 10 U / mL. As expected, the dissolved oxygen levels were found to decrease as a function of the glucose oxidase concentration (Figure 4D). At each concentration, the oxygen saturation versus time relationship follows a sigmoidal curve shape. As the added enzymes first begin removing oxygen, there is a slight delay before the probe can register the reduction. And once the oxygen levels begin reaching lower saturation levels (< 1%), the enzymes begin to compete with the oxygen diffusion from the air-liquid interface. This is most apparent when comparing the 10 U / mL sample vs the 0.03 U / mL. Since some solutions do not reach 0% saturation, a different metric is needed for comparison between concentrations. Saturation half time is defined as the time required for the dissolved oxygen levels in solution to reach half (10.5% pO2) of atmospheric oxygen concentration (a pO2 of 21%). The half times for each solution was plotted against the concentration of glucose oxidase to create a standard curve. The half time-activity relationship followed a power law model with high accuracy (R- square = 0.995) for activities of 1 U / mL and below (Figure 4E). Notably, the 3 and 10 U / mL solutions did not decay fast enough due to the limitation in speed of the oxygen probe used. Therefore, they were not used for the best fit equation. To measure the effect of catalase on the solution, pure glucose oxidase (0.3 U / mL activity) without catalase was added to the microcapsules. Here, the sample without glucose oxidase shifts the activity curve to the left, suggesting faster oxygen depletion. Calculation of the saturation half time gives 5.8 and 3.5 minutes, a 40% reduction, for the catalase and catalase free conditions respectively. This is as expected, as catalase should lead to a reduction of oxygen depletion by 50%. Glucose oxidase and catalase loaded microcapsules of varied activities (1000 U / mL of glucose oxidase and 60,000 U / mL of catalase; 5, 12.5, and 25 mg / mL of capsules) were subsequently added to the same DO probe setup to measure the effect of encapsulation on activity. These capsules all reached hypoxic levels of oxygen tension (< 5% pO2 oxygen) by 15 minutes, with a half saturation time of 1.9 ± 0.4, 3.0 ± 0.3, and 5.9 ± 0.6 minutes for the 5, 12.5, and 25 mg / mL conditions respectively (Figure 4F). Finally, a capsule loaded GelMa microgel suspensions (2 vol% capsules in the gel, 0.4mL of hydrogel volume; final of 2 mg / mL of capsules) was analyzed with the identical setup. These loaded microgel suspensions had a half saturation time of 23 minutes, reaching a plateau of 2% oxygen saturation after 2 hours. These data suggest that hypoxic conditions can readily be reached with a low total volume fraction of capsules. One caveat with this methodology is that it can only read the oxygen levels in solution. It is well appreciated that both in vitro hydrogels and in vivo tissues can lead to reduced oxygen levels when thick enough. And even slight differences in depth of cell culture medium in flasks and wells can affect the oxygen levels to adherent cells. Therefore, the measurements of liquid at the surface likely underestimate the levels of hypoxia that cells experience within the microgel suspensions. Nonetheless, these measurements provide insight into the upper bound of glucose oxidase needed to establish these conditions. 2.3 Low oxygen environments perturb MSC function and tumour growth With functioning oxygen modulating microcapsules, the effect of hypoxia on cell viability was investigated. Live dead assays were performed on days 1 and 5 for 3 cell lines: A375-P, Adipose derived stromal cells (ADSC), and human umbilical vascular endothelial cells (HUVECs). Capsules loaded with 1000 U / mL of glucose oxidase and 60,000 U / mL of catalase were added to GelMa microgel suspensions, used previously,13(2 vol% of capsules) along with one of the cell lines. Here, 1000 U / mL inside the capsules would give 2 U / mL in total solution in the well of a 24 well plate. With a 5% enzyme functionality, this provides a rough activity of 100 mU of glucose oxidase activity in one well (1 mL of volume). Since all following in vitro studies were performed in 24 well plates with 1 mL of media, all future references of U / mL of enzyme will refer to the initial loading within the capsules for consistency. After 1 day, there was a trend, although not statistically significant, of reduction in viability for the melanoma cell line, from 97 ± 1% to 65 ± 20% when hypoxic capsules were added. Meanwhile, the ADSCs and HUVECs had little difference. By 5 days, nearly all A375-P cells were dead (16 ± 12%, P = 0.000017), while HUVEC cells had a drop in viability to 36 ± 20 % (P = 0.0016). Surprisingly, the ADSCs showed no decrease in viability even after 5 days (93.7 ± 0.3 vs 92.5 ± 5.5%, P = 0.999). However, the ADSCs, along with the HUVECs, showed spreading in the control condition, but no noticeable cell spreading in the glucose oxidase condition. Data from the dissolve oxygen probe showed these capsules exhibit an activity around 0.3 U / mL which gives a < 1% oxygen saturation after 20 minutes at the surface of the gel. Thus, while some cells may survive at this level of hypoxia, cell function is significantly impaired. To explore the effect of hypoxia on intracellular HIF1-alpha, microgel suspensions were loaded with ADSCs (1 × 106cells per mL) with either no capsules, or capsules with 100, 250, or 1000 U / mL of glucose oxidase (2 vol%; 6,000, 24,000, or 60,000 U / mL of catalase respectively). At both days 1 and 3, significant expression of HIF1-alpha can be seen in in the capsule conditions but not the control (Figure 5A). Here, the control sample had 7.5 ± 3.5% of cells with expression, while the 1000U / mL glucose had 73 ± 11% (P = 0.0013) (Figure 5B). Hypoxia has been shown to have an effect on the extracellular vesicle (EV) output of mesenchymal stem cells (MSCs). To measure if capsule induced hypoxia had any effect on ADSC extracellular vesicle output, ADSCs were embedded into 3D microgel suspensions with capsules containing 0, 30, or 250 U / mL of glucose oxidase (1 × 106cells per mL, 2 vol% capsules in hydrogel). After 72 hours in serum free media, the supernatants were collected, and the EVs were isolated and analyzed. No differences were found in the concentration of exosomes across all three conditions (Figure 5C). However, there was a significant increase in mean vesicle diameter for the hypoxic condition compared to the control (P = 0.010) (Figure 5D-E). This correlates well with previously reported work (Zhu et al. 2018 Theranostics 8:6163). The presence of CD9, a known surface marker for exosomes, in the EVs was also confirmed via western blots. Overall, the results indicate that the ADSCs may indeed be experiencing and responding to the low oxygen environment, leading to changes in the secretion of extracellular vesicles. When sufficiently large enough, all cancerous tumours exhibit hypoxic cores. Aberrant tumour paracrine signaling networks disrupt nearby vasculature, leading to leaky vessels and poor nutrient and oxygen delivery. This disruption can generate hypoxic cores with a tumour volume as they continue to grow. These hypoxic cores have been implicated in tumour progression, including higher cancer survival, aggressiveness, and more frequent incidence of metastasis. Therefore, model systems that can replicate hypoxic microenvironments in the laboratory are important research tools for fundamental research and drug development. To measure the effect of hypoxia on melanoma cells, A375-P human melanoma cells in GelMa microgel suspensions (between particles) with a filler of 1 wt% GelMa were embedded in the microgel interstitial space. These suspensions were subsequently loaded with microgel capsules that contained 0 or 100 U / mL of glucose oxidase (and 6,000 U / mL of catalase), 10% of the enzyme activity that led to total cell death of the A375-Ps in the live dead study. In the control condition, A375-P cells can be seen to grow in clusters that expand in size by 55% (P = 0.07) over the course of 7 days (Figure 5F). In contrast, the cell clusters in microgel suspensions with hypoxia inducing capsules increased in size by only 23% (P = 0.77) and 61 ± 20% the size of the control condition (P = 0.049) (Figure 5G). Notably, this model used single cells which are less reminiscent of in vivo tumours. Therefore, tumour spheroids were investigated to evaluate if hypoxia would affect tumour outgrowth. Here, MCF-7 breast cancer spheroids were embedded into Geltrex with 0, 10, 30, or 100 U / mL capsules loaded. For the conditions with 30 or 100 U / mL capsules, the spheroids did not proliferate during prolonged culture. For the 10 U / mL condition, there was slight growth through the first two days, after which the spheroids died (Figure 5H-I). These findings were replicated when using a more aggressive breast cancer cell line, MDA-MB-231, where there was little tumour growth for the hypoxic condition over 6 days. Together, the results show that tumour cells within the presently tested in vitro models are sensitive to hypoxia, eventually leading to cell death. This increased susceptibility of tumour cells to apoptosis under hypoxia in vitro has previously been demonstrated, with the level of susceptibility varying from cell line to cell line. The results suggest that it may be difficult to draw direct conclusions of in vivo tumour response using in vitro hypoxia models. 2.4 Hypoxic microcapsules induce angiogenesis in vitro and in vivo It is known that oxygen levels play a role in endothelial cell signaling pathways. Under low oxygen tension, endothelial cells are driven to undergo angiogenesis to create more vessel structures to supply more oxygen. This has also been previously demonstrated using hydrogels with oxygen reducing enzymes tethered to the hydrogel backbone. The effect of hypoxia on vascular endothelial cells was investigated using a variety of methods and materials to demonstrate the platforms versatility. First, HUVECs were seeded in GelMa microgel suspensions (0% Filler, 5 × 106cells per mL) with varied concentrations of glucose oxidase loaded capsules and cultured for 24 hours before snap freezing the gels and collecting cellular mRNA. RT-qPCR analysis showed an upregulation in VEGFa levels with increasing levels of hypoxia, with a 2-fold difference for the 250 U / mL condition compared to the control condition (Figure 6A) (P < 0.0001). There was a simultaneous decrease in ANGPT-1 by 10- fold (Figure 6B) and no statistical difference in MMP1 gene expression. Previous work has demonstrated that hypoxia induces an upregulation in VEGFa and a downregulation in ANGPT-1 in human endometrial cells. This is in contrast to other literature showing an upregulation of ANGPT-1 in vascular cells, and pericytes under hypoxic conditions. Some reports have shown that this upregulation can be acute and transient. The role of ANGPT-1 is primarily vascular protective, preventing endothelial cell death and preserving the structural integrity of vasculature. It is possible that in the first 24 hours the dispersed cells increase VEGFA expression due to hypoxia, but without the paracrine mechanism that is vascular protective. To test whether these endothelial cells would still undergo angiogenesis in this system, the previous experiment was repeated but allowed to proceed for 5 days. After 5 days, HUVECs in enzyme capsule loaded conditions (100 and 250 U / mL of glucose oxidase) began to form interconnected tubes (Figure 6C). For quantification, the angiogenesis analyzer tool in ImageJ was used. When comparing the two conditions, the 100 and 250 U / mL conditions had 430% and 470% increased master segment length (0 – 100: P = 0.019; 0 – 250: P = 0.0070) and 320% and 460% more master segments respectively (0 – 100: P = 0.227; 0 – 250: P = 0.022). However, when measuring average branch length, as well as mesh area, there was a trend suggesting that the 100 U / mL had thicker and healthier vessels, which matched the qualitative observations from the confocal images. To further confirm these angiogenesis results, a tubulogenesis on Matrigel assay was performed using the 0 U / mL and 100 U / mL capsules conditions (Figure 6E). These results demonstrated a temporal response to hypoxia, where the hypoxic condition reached its maximum master segment length (Figure 6F), number of master segments, and number of master junctions 4 hours earlier than the control conditions. These differences abated by 24 hours as is expected with endothelial tubulogenesis experiments on Matrigel. As a final confirmation of the hypoxia in vivo, a Chick Chorioallantoic Membrane (CAM) assay was run, a well-established in vivo angiogenesis assay. Here, microgel suspensions loaded with varied concentrations of 100 U / mL capsules were placed onto the CAM of developing chicken embryos. After 4 days of sample exposure, the number of branches intersecting the edge of the gels was analyzed (Figure 6G). No statistical differences were found between the PBS and the gels loaded with 0, 10, or 25 U / mL enzyme capsules (Figure 6H). However, there was a significant difference between the PBS and the 50 and 100 U / mL conditions (PBS – 50: P = 0.02; PBS – 100: P = 0.0002). Further, the 100 U / mL condition enhanced angiogenesis with 43% and 22% more intersecting vessels relative to the 0 U / mL gels and the VEGF positive control, respectively. Overall, these data suggest that hypoxia induced upregulation of VEGFa leads to angiogenesis in 2D, 3D, and in ovo models. The ability to fine tune the concentration of capsules and enzyme allows for a robust way to investigate subtle changes in the levels of oxygen tension and how this perturbs angiogenesis and associated processes. 2.5 Ceria loaded microcapsules reduce cellular oxidative stress Finally, cargo flexibility the loading of gas modulating nanozymes within the capsules was evaluated. Cerium oxide nanorods (Nanoceria) were chosen as the candidate payload since these nanozymes have found widespread use as in vivo therapeutics by attenuating reactive oxygen species. Nanoceria has catalase (Figure 7A), ATPase, and peroxidase activities depending on its valence state and buffer pH, and it can reduce the oxidative stress on cells in vitro. The enzymatic mimicry of nanoceria make it an attractive cargo as enzymatic substrates could diffuse in and out of the capsule wall while the nanoceria remain entrapped. Nanoceria was synthesized using published methods to produce nanorods with lengths of 50 nm (Figure 7A). To quickly verify the functionality of this nanoceria, the particles were suspended in a McIlvaine buffer solution (0.1M) at varied pH (3, 5, and 7). A solution of an organic dye that changes color when oxidized (2,2'-azino-bis(3-ethylbenzothiazoline-6- sulphonic acid) (ABTS)) by cerium oxide was paired to measure the peroxidase mimicking activity of the nanoceria in solution. To quantify the pH and nanoceria activity relation, ceria and ABTS were added to an array of buffers with varied pH (4, 5, 6, 7, and 8, 2mM ABTS, 0.5 mg / mL ceria) and were left to react overnight. The absorbance spectra showed an order of magnitude higher activity of nanoceria at a pH of 4 versus 6, with a plateau of activity beyond pH 6. This correlates well with the pH response found in the literature, demonstrating that the synthesized nanoceria behave as expected. To verify that nanoceria can be loaded into capsules, fluorophores were conjugated to the ceria particles for visualization. Successful silanization of nanoceria was verified using FTIR where appearance of the 1050-1100 peak for the silanized samples represents the Si-O-Si bond stretching. After particle resuspension in ethanol, a solution of FITC (1 mg / mL) was added for direct conjugation to the terminal amine groups. Confocal microscopy images demonstrate successful fluorophore conjugation as well as successful loading of nanoceria into microgel capsules. Typically, nanoceria loaded in high concentrations above 2-5 µg / mL would be cytotoxic to cells. However, by adding the nanoceria to capsules, one can reach high concentrations that maintain adequate functionality while protecting cell viability. To assess the effect of a high concentration (5 mg / mL) of nanoceria on cell viability, a live dead assay was performed with three cell types: A375, ADSC, and HUVEC Nanoceria loaded microgel capsules were added to GelMa microgel suspensions at 2% of the final volume. Live dead staining at both days 1 and 5 indicate a viability above 90% for all cells lines with ceria capsules. One functional use of nanoceria in cell culture is the reduction of superoxide species (O2•-). Cellular superoxides are typically formed in mitochondria during cellular respiration. If they accumulate too much, they can generate high levels of hydrogen peroxide and hydroxyl radials which disrupt cell activity and led to cell apoptosis. To combat this damage, cells use the enzyme superoxide dismutase (SOD). In humans, up to 99% of SOD is found in the extracellular matrix and blood plasma. Nanoceria has also been shown to reduce superoxide tension in cells in culture and improve cell viabilities. However, studies have only demonstrated super oxide protection when nanoceria particles are endocytosed into cells. Therefore, without wishing to be bound by theory, the present inventors have hypothesized that high concentrations of nanoceria loaded within microcapsules, simulating extracellular SOD enzymes, could also reduce the superoxides within the surrounding matrix. First, ADSCs were seeded in either capsule free microgel suspensions or microgel suspension with ceria-loaded capsules (5 mg / mL of nanoceria, 2 vol% capsules). After 5 days, the cells were stained with Hoechst and MitoSOX, a live cell dye that penetrates cell membranes and fluoresces when exposed to superoxide species (Figure 7B). Analysis of confocal z-stacks showed a 37% reduction in cellular MitoSOX intensity in cultures with ceria-loaded capsules (P = 0.040) (Figure 7C). To confirm correct staining, high resolution images were taken that show a punctate speckled appearance (Figure 7D), matching with mitochondrial stains. Ceria nanoparticles also have potential therapeutic use during cancer treatment. A common cancer therapy is radiation where a focused blast of a photon or particle beam is aimed at the tumour site. However, while these blasts are typically well targeted, they can affect the surrounding matrix, leading to unwanted damage of healthy tissues, including the generation of hydrogen peroxides. This large influx of peroxides can cause significant damage to surrounding tissues via radical hydroxide formation and subsequent DNA damage. Previous work has created in vitro models of this peroxide generation to assess the cytoprotective effect. The cytoprotective effect of nanoceria against peroxides was measured. ADSCs were embedded in microgel suspensions and subjected to 24 hours of culture with varied concentrations of hydrogen peroxide (0, 100, and 1000 µM). For each concentration range, replicates were made with ceria capsules loaded into the suspension (5 mg / mL) at 2% volume fraction. As a negative control, no ceria capsules were added to the suspension; as a positive control, a high concentration of catalase enzyme (1000U / mL) was loaded into microgel capsules and added to the microgel suspension (Figure 7E). When quantifying cell viability, all conditions with equal to or less than 100 µM of H2O2 had cell viabilities above 90% (Figure 7F). When exposed to 1000 µM, the negative control samples had 63% live cells compared to 99.5% (P = 0.00047) and 95% (P = 0.0027) for the catalase and ceria conditions respectively. In addition, the ceria capsule conditions appeared to have less cell spreading than their catalase counterparts. This result indicates that ceria nanoparticles are able to remove hydrogen peroxides from solution and protect cell viability, but potentially at a slower rate than catalase. 2.6 Spatial patterning of microcapsules provide oxygen gradient to direct intestinal organoid morphogenesis A key advantage of using microcapsules is the spatial control of oxygen afforded by simply varying placement of the microcapsules. To demonstrate that the microcapsules can be used to establish spatial oxygen gradients, a bioprinter was utilized to control the distribution of glucose oxidase microcapsules. A concentric circular disc construct was bioprinted for this purpose (Figure 8A). The concentric circular disc consisted of a center disc (radius of 0.5 cm) containing 30 U / mL of glucose oxidase microcapsules suspended in a GelMA-based bioink, which served as the oxygen sink. This disc was surrounded by blank GelMA bioink, forming a large annulus with a radius of 2 cm (height = 3 mm). To assess the spatio-temporal patterning of oxygen concentration within the bioprinted disc, the oxygen distribution along the bioprinted construct’s radius was measured at 5 mm intervals with an oxygen probe. First, the temporal dynamics was determined to establish an equilibrium oxygen gradient along the radial axis of the bioprinted disc, which resulted from an influx of oxygen from the surrounding environment (inside a 37oC and 5% CO2incubator) and oxygen scavenging activities by glucose oxidase microcapsules printed in the center of the disc. Excess culture medium (30 mL) was added to ensure that glucose concentration was not limiting the oxygen scavenging activity of the microcapsules. It was observed that the oxygen distribution within the disc achieved equilibrium after ~ 5 mm, consistent with the present inventors’ model (Figure 8B). The equilibrium oxygen concentration gradient in the bioprinted disc ranged from 1.4 ± 0.2% oxygen (oxygen partial pressure in equilibrium with water) at the center of the disc (r = 0 mm) to 19.4 ± 0.5% at the edge (r = 15 mm). The oxygen levels increased along the radius of the disc, where it was noted an equilibrium oxygen level of 5.3 ± 1.0% at r = 5 mm and 14.9 ± 1.3% oxygen at r = 10 mm. This experimental data corroborates the present inventors’ computational simulation model that accounted for diffusion and glucose oxidase reaction kinetics to establish oxygen concentration gradient (Figure 8C-D). Both the experimental data and the simulation model reached ~1% oxygen at the center of the disc after 1 hour (Figure 8C-D). There was also good agreement in the concentration gradient profiles the 1-,3-, and 5-hour time points. However, there was a difference in the 24 hour concentration profile, with the computational model overestimating the oxygen level at 5mm and underestimating at 15mm (Figure 8D). This is likely due to the resolution of the dissolved oxygen probe used. Next, it was evaluated how long the oxygen concentration gradient can be maintained. It was noted that the concentration gradient profile could not be maintained when the bioprinted disc was incubated for 4 days without any medium change. By day 3, the oxygen level at the center of the disc was observed to be at 8.9 ± 0.5 %, which was 6-fold higher than day 1 (1.4 ± 0.2 % oxygen) (Figure 8E). This was likely attributed to a decrease in glucose oxidase activity in the microcapsules over the 4-day period. Given that glucose is a key substrate of glucose oxidase, prolonged incubation without medium change likely caused glucose depletion and consequently resulted in a loss of oxygen scavenging activity. With daily replenishment of culture medium, the oxygen concentration gradient could be consistently maintained (Figure 8F). Therefore, it is shown that by patterning the microcapsules at discrete locations, an oxygen concentration gradient can be readily established under normal atmospheric condition in a standard cell culture incubator. This is not achievable with commonly used hypoxic incubators, which regulate oxygen concentration at a single uniform level. Hence, combining bioprinting with the oxygen scavenging microcapsules can provide a simple and scalable avenue to mimic physiologically-relevant oxygen gradients within a single experiment. Intestinal stem cells are naturally adapted to function in a low oxygen microenvironment (1- 3%), leading to better growth in low oxygen conditions. To evaluate the effect of microcapsule mediated oxygen gradient on the growth of intestinal tissues, intestinal organoids were passaged and seeded in well plates which were maintained under normoxia (in a 5% CO2, 21% O2 incubator), but subjected to an oxygen gradient that was created by depositing a microcapsule disk to one side of the well. The microcapsule disk composition is 2 mg / ml corresponding to 25 U of loaded enzyme. This allowed for the creation of a distinct oxygen gradient within the same well. Next, two 10µl Matrigel / organoid droplets were created within each well. One of the droplets (Dome 1) was positioned near the microcapsule (<5 mm), whereas the second droplet (Dome 2) was positioned at the extreme end of the well, furthest away from the microcapsules (~10 mm) (Figure 8G). This allowed for the comparison of the effect of the oxygen gradient on the growth and morphology of the gut organoids as a function of spatial distance to the hypoxia source. Controls were established by culturing the intestinal organoids in the same setup with a sham disk (no microcapsules) in a 2% O2,hypoxic incubator and a 21% normoxic incubator. After 48 hours, organoid size and proliferative ability (evidence as branching and budding) increased with proximity to the microcapsule disk. Organoids grown in the well with 25 U microcapsule disk had a 630% increase in size compared to untreated controls (P < 0.0001) (Figure 8H-I). The size of the organoids grown in close proximity to the 25 U capsule reservoir (<5 mm; <5% O2) were larger than those cultured far from the capsule reservoir (~10 mm; <15% O2) which was comparable in the size of the positive control organoids grown in the hypoxic incubator (static 2% O2) (Figure 8I). Under microcapsule conditions, organoids in Dome 1 were 300% larger than those in Dome 2 (P<0.001) and formed significant budding and branched structures. When staining the organoids in all conditions with a hypoxic dye (ImageIT), there was elevated signal in the organoids grown in the hypoxic chamber and when cultured with microcapsules (Figure 8H). Additionally, all hypoxic conditions led to enhanced growth compared to normoxic culture conditions and there was no difference in size between Dome 1 and Dome 2 in control organoids grown under hypoxia or under normoxia (blank disk; no microcapsules). Together, these results demonstrate how microcapsules can regulate hypoxia with spatial control, leading to enhanced proliferation with evidence for increased growth characteristics compared to hypoxic chambers. A hypoxic intestinal stem cell microenvironment can have a significant impact on the growth of organoids. The intestinal stem cells are responsible for maintaining the intestinal epithelium and are key for the renewal and repair of the gut lining. These cells are located in a niche within the gut, where they are exposed to low oxygen levels. In hypoxic conditions, the activation of various signaling pathways, such as the HIF-1α pathway, can regulate cellular metabolism and increase the resistance of cells to oxidative stress, leading to enhanced survival and proliferation of the stem cells. This can result in an increased number of stem cells and an improved capacity for self-renewal, which can enhance the growth of the organoids. Furthermore, hypoxia can also influence the expression of various genes involved in cellular metabolism, cellular signaling, and angiogenesis, leading to changes in the cellular behavior of the stem cells that can promote their survival and growth. This can result in an improvement in the quality and size of the organoids, as well as an increased branching pattern, which is indicative of their growth and proliferation capacity. The hypoxic microenvironment of gut has a significant impact on the microbiome as many gut bacteria are anaerobic and require low oxygen levels to grow and thrive. By maintaining a hypoxic environment for the cultivation of gut organoids, the microorganisms that are associated with the gut can be better replicated, which can lead to a more accurate representation of the gut microbiome. This highlights the importance of maintaining a hypoxic environment for the cultivation of organoids to achieve optimal growth and function, and to mimic the native microenvironment of the gut. These results demonstrate how using oxygen modulating microcapsules can mimic gut organoid morphogenesis, comparable to what is achieved using a hypoxic chamber. By spatially controlling the number and activity of capsules, it was possible to locally tune and establish hypoxic gradients, providing the first example of directing different activities spatially within a single experiment. 3. Discussion This disclosure presents a novel way to spatially control hypoxic conditions in any 2D or 3D hydrogel system. The microcapsule system enables high loading of enzymes with prolonged shelf life and maintenance of activity, which allows for control of hypoxia in cell culture for several weeks. Tuning the capsule concentration or initial enzyme concentration allows for fine control over the oxygen levels across a 3D cell culture, and across different samples in a well plate, towards integration in high throughput assays. Hypoxic conditions were shown to attenuate Mesenchymal Stem Cell extracellular vesicle output, as well as diminish tumour growth characteristics. Hypoxic capsules also promoted an upregulation in endothelial VEGFa which led to enhanced angiogenesis in 2D, 3D, and in ovo models. This advance could transform body-on-a-chip devices, where variable oxygen levels can be precisely tuned to different tissues across a single device, something that is not currently possible. This approach can also be adapted to nanozymes, other enzyme systems, or gas releasing hydrogels for use in a broad range of applications. The injectability of the capsules suggests potential use for direct in vivo therapy or as a cytoprotective agent to attenuate damage due to radiation therapy. The flexibility of the microcapsule format provides scope for use as a reservoir for inclusion in topical treatments and gel-based bandages, where gas attenuation may have clinical utility (e.g., wound healing). In addition to these practical applications, it is anticipated this system will provide a basis for future work investigating the role of oxygen deprivation on cell signaling pathways and functional cellular outputs.

Claims

CLAIMS What is claimed is:

1. A population of microcapsules, wherein the microcapsules comprise gelatin coated with a nanoscale layer of polydopamine; are 1 um to 10 mm in size; and encapsulate a payload.

2. The population of claim 1, wherein the microcapsules have an average pore size of about 1 nm.

3. The population of claim 1 or 2, wherein the microcapsules exclude proteases.

4. The population of any one of claims 1-3, wherein the population can withstand a shear stress value of at least 10 kPa.

5. The population of any one of claims 1-4, wherein the payload is at least one enzyme.

6. The population of claim 5, wherein the at least one enzyme includes glucose oxidase and catalase.

7. The population of any one of claims 1-4, wherein the payload is nanoceria.

8. A hydrogel comprising the population of any one of claims 1-7.

9. A pharmaceutical composition comprising the population of any one of claims 1-7.

10. A solid or semi-solid matrix comprising the population of any one of claims 1-7.

11. A method of preparing a population of microcapsules comprising i) adding gelatin and payload to an oil bath to form an emulsion comprising loaded microcapsules; ii) washing the loaded microcapsules; iii) adding dopamine hydrochloride to form a polydopamine coating on the loaded microcapsules; iv) washing the coated, loaded microcapsules.

12. The method of claim 11, further comprising agitating or stirring in step i).

13. The method of claim 11 or 12, further comprising agitating or stirring in step iii).

14. A method of bioprinting comprising printing the population of any one of claims 1-7.

15. The method of claim 14, wherein the population is in a jammed state and / or 16. The method of claim 14 or 15, wherein the population is combined with a microgel.

17. A method of reducing oxygen levels in an environment comprising adding the population of any one of claims 1-6 to the environment.

18. The method of claim 17, wherein the environment is an in vitro cell culture.

19. The method of claim 17, wherein the environment is an ex vivo tissue sample or organoid.

20. The method of claim 17, wherein the environment is localised in vivo in a subject.

21. The method of any one of claims 17-20, wherein the method induces angiogenesis.

22. The method of claim 17, wherein the environment is a sample comprising anaerobic bacteria.

23. The method of claim 17, wherein the environment is selected from the group consisting of a chamber containing an oxygen-sensitive liquid, bandage, a bioreactor, a contact lens, or a polymer matrix for coating a surface.

24. A method of creating an oxygen gradient in an environment comprising adding the population of any one of claims 1-6 to the environment.

25. The method of claim 24, wherein the environment is an ex vivo tissue sample or organoid.

26. A method for reducing the size of a tumour in a subject comprising administering the population of any one of claims 1-6 to the subject.

27. The method of claim 26, wherein the administering is localized to the region of the tumour.

28. A method of reducing reactive oxygen species in an environment comprising adding the population of claim 7 to the environment.

29. The method of claim 28, wherein the environment is an in vitro cell culture.

30. A method of reducing reactive oxygen species in a subject exposed to radiation comprising administering the population of claim 7 to the subject.

Images