Concentrically Encapsulated Dual-Enzyme Capsules for Synergistic Metabolic Disorder Redressing and Cytotoxic Intermediates Scavenging

Enzyme therapy has important implications for the treatment of metabolic disorders and biological detoxification. It remains challenging to prepare enzymatic nanoreactors with high therapeutic efficiency and low emission of cytotoxic reaction intermediates. Here, we propose a novel strategy for the preparation of enzymes-loaded polypeptide microcapsules (EPM) with concentrically encapsulated enzymes to achieve higher cascade reaction rates and minimal emission of cytotoxic intermediates. Mesoporous silica spheres (MSS) are used as a highly porous matrix to efficiently load a therapeutic enzyme (glucose oxidase, GOx), and a layer-by-layer (LbL) assembly strategy is employed to assemble the scavenging enzyme (catalase) and polyelectrolyte multilayers on the MSS surface. After removal of the MSS, a concentrically encapsulated EPM is obtained with the therapeutic enzyme encapsulated inside the capsule, and the scavenging enzyme immobilized in the polypeptide multilayer shell. Performance of the concentrically encapsulated GOx-catalase capsules is investigated for synergistic glucose metabolism disturbance correction and cytotoxic intermediate H2O2 clearance. The results show that the EPM can simultaneously achieve 99% H2O2 clearance and doubled glucose consumption rate. This strategy can be extended to the preparation of other dual- or multi-enzyme therapeutic nanoreactors, showing great promise in the treatment of metabolic disorders.


Introduction
Enzyme therapy is attracting considerable interest in addressing metabolism disorders and biodetoxification, such as phenylketonuria, diabetes, hyperlipidemia, and hyperuricemia [1][2][3][4][5][6]. Currently reported enzyme therapy systems generally involve single types of enzymes such as uricase [7], glucose oxidase (GOx) [8,9], alcohol oxidase [10,11], superoxide dismutase [12], and phenylalanine ammonia lyase [13]. However, toxic intermediates released from the catalyzed reaction of the therapeutic enzyme (e.g., H 2 O 2 released from the catalyzed reaction of oxidases) in a single enzyme system severely hinder the medical application [14][15][16]. In living organisms, metabolism involves an enzyme complex and physical tunnel to channel the enzymatic reaction intermediate, forming cascade reactions and avoiding the release of toxic by-products [17]. Inspired by living organisms, therapeutic nanoreactors of multienzyme systems have been proposed to redress metabolic disorders and decompose toxic intermediate molecules [18]. To improve the performance, the differently functionalized enzymes can be immobilized in a multicompartment system with a desirable spatial arrangement to minimize the free diffusion of intermediates, catalyze the cascade reactions synergistically and reduce toxic by-products release.
To date, various multienzyme nanoreactors with different spatial arrangements have been developed. These nanoreactors are mainly designed based on liposome or polymerencapsulated enzymes including polymersomes and layer-by-layer (LbL) microcapsules [19][20][21][22]. To perform enzymatic reactions in a continuous mode, therapeutic nanoreactors are required to have an efficient membrane diffusion of substrates and products [23]. To overcome the low permeability of liposomes and polymersomes, trapping channel proteins in the polymeric bilayer or preparing permeable polymersomes with an intrinsically porous shell have been proposed [24,25]. For instance, Hest's group has constructed a nanoreactor with positionally assembled multienzymes in porous polymersomes [26,27]. Multicompartments of liposomes, namely capsosomes, were also reported by Caruso's group [28][29][30]. The capsosomes with concentric multicompartments could be LbL assembled by enzymesloaded liposomes and polymer separation layers. Further, glutathione reductase or βlactamase could be loaded in liposomal subcompartments to realize responsive release [31]. In comparison with liposomes and polymersomes, the LbL assembled microcapsules possessed more tunable permeability for the transportation of substrates. With the mediation of porous particles as sacrificial templates, LbL microcapsules can encapsulate a high payload of cargos with controllable size, permeability, degradability, and surface functionality [32][33][34]. For instance, Baumler's group designed a multienzyme nanoreactor with magnetic biopolymer/CaCO 3 cores and CaCO 3 separation layers, in which the cascade reaction starts from β-glucosidase on the outer compartment to GOx in the first peripheral, and to horse radish peroxidase (HRP) in the central [35]. However, these therapeutic nanoreactors of multienzyme systems remained challenging to achieve both high therapeutic efficiency and low emission of cytotoxic intermediates.
Mesoporous silica spheres (MSS) have been widely used in biomedical applications due to their many distinct characteristics, such as highly tunable particle and pore size, high loading capacity, and easy surface chemical modification [36]. In this manuscript, we proposed a novel strategy to prepare enzymes-loaded polypeptide microcapsules (EPM) with concentric multicompartments through MSS-mediated enzyme loading and LbL shell assembly (Scheme 1). As a proof-of-concept, concentrically encapsulating the therapeutic enzyme (glucose oxidase, GOx) in the hollow core and scavenging enzyme (catalase, Cat) in the shell were investigated for realizing synergistic glucose metabolism disturbance correction and cytotoxic intermediate H 2 O 2 clearance since these enzymes have been frequently employed in therapeutic nanoreactor for blood glucose redressing [10,18,37]. Owing to the concentrically encapsulated GOx and catalase and their deliberately arranged locations, the H 2 O 2 scavenging (99% clearance rate) in the EPM system is much more efficient than previously reported multienzyme nanoreactors, such as the co-localized multienzyme nanocomplex (about 85%) [10] and the supramolecular hydrogel (about 88%) [37]. The concentrically compartmented EPM holds enormous promise in developing more multienzyme therapeutic nanoreactors for addressing metabolic disorders and scavenging toxic intermediates. Nanomaterials 2022, 12, x FOR PEER REVIEW 3 of 12 Scheme 1. Schematic illustration of the preparation of EPMGOx-Cat through MSS-mediated immobilization of the therapeutic enzyme (GOx) and concentric assembly of the scavenging catalase, followed with LbL assembly of a protecting polypeptide multilayer shell and removal of the MSS template.

Preparation of EPM
Briefly, the preparation of GOx-loaded EPM (EPMGOx) included three steps. (1) GOx loading: 5.0 mg of amine-modified MSS (dispersed in ethanol, supporting information) was washed with HAc buffer (50 mM, pH 5.0) twice to remove ethanol. Then 2.0 mL of GOx solution (1.0 mg mL −1 ) was added to the MSS suspension. After incubation at room temperature for 12 h, GOx-loaded MSS were collected by centrifugation (1500 g, 5 min). To crosslink the amino groups in GOx with the amine-modified MSS, 1.0 mL of fresh GA solution (0.2 wt.%, in HAc buffer, pH 5) was added to the GOx-loaded MSS suspension and incubated at room temperature for 1 h. After that, the particles were centrifuged and Scheme 1. Schematic illustration of the preparation of EPM GOx-Cat through MSS-mediated immobilization of the therapeutic enzyme (GOx) and concentric assembly of the scavenging catalase, followed with LbL assembly of a protecting polypeptide multilayer shell and removal of the MSS template.

Materials
Catalase from bovine liver (C100, 45,573 U mg −1 ), glucose oxidase from Aspergillus niger (Type X-S, 127 U mg Horseradish peroxidase (HRP, 200 U mg −1 ) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC) were purchased from Aladdin Reagent Company (Shanghai, China). All chemicals were of at least analytical grade and used without further purification. Fluorochrome-labeled enzymes and PLL (GOx labeled with green FITC, catalase labeled with blue DCCA, and PLL labeled with red SRB) were prepared by adding 50 µL of dye (10 mg mL −1 ) to 1.0 mL of enzyme or PLL solution (10 mg mL −1 ) in 0.1 M sodium bicarbonate buffer at pH 9.0. After stirring at room temperature for 1 h, the mixture was sufficiently dialyzed against deionized water to remove free dyes.

Preparation of EPM
Briefly, the preparation of GOx-loaded EPM (EPM GOx ) included three steps. (1) GOx loading: 5.0 mg of amine-modified MSS (dispersed in ethanol, supporting information) was washed with HAc buffer (50 mM, pH 5.0) twice to remove ethanol. Then 2.0 mL of GOx solution (1.0 mg mL −1 ) was added to the MSS suspension. After incubation at room temperature for 12 h, GOx-loaded MSS were collected by centrifugation (1500 g, 5 min). To crosslink the amino groups in GOx with the amine-modified MSS, 1.0 mL of fresh GA solution (0.2 wt.%, in HAc buffer, pH 5) was added to the GOx-loaded MSS suspension and incubated at room temperature for 1 h. After that, the particles were centrifuged and washed with HAc buffer twice. (2) PLL infiltrating: 1 mL of s-PLL (5 mg mL −1 in PBS buffer, pH 7.4) was used to infiltrate into the porous particles for 12 h. After the infiltration of s-PLL and removal of excess s-PLL, 2.0 mL of freshly prepared EDC solution (4 mg mL −1 in MES buffer, pH 5.5) was added to crosslink the residual carboxyl groups on GOx and amino groups on s-PLL. After 2 h of incubation, the obtained materials were washed with MES buffer twice. (3) LbL shell assembling: Three PLL/PGA bilayers were deposited on the MSS surface by mixing 2.0 mL of alternative PGA/PLL (1 mg mL −1 in MES buffer) and MSS for 10 min. Twice washing with MES buffer was applied to remove excess polypeptide after each step of polypeptide deposition. After that, EDC was added to crosslink the assembled enzymes and polypeptides. The EPM GOx was obtained after removal of the MSS template with 0.3 mL of HF/NH 4 F solution (2 M HF: 8 M NH 4 F) for 1 min. The loading and leaching of Gox during EPM Gox preparation were monitored by UV-vis spectrophotometry at the band of 276 nm ( Figure S1). The concentrically encapsulated dual enzymes capsule (i.e., EPM Gox-Cat ) was prepared by the same procedure for EPM Gox preparation except that the second enzyme (catalase) was sequentially assembled on the Gox-loaded MSSs before the PLL/PGA multilayer coating. After incubating at 37 • C for 30 min, 10 µL of H 2 SO 4 solution (2.0 M) was added to each well and the absorbance at 450 nm was recorded. The concentration of H 2 O 2 was calculated based on the color reaction in absence of GOx compared to standard H 2 O 2 , while that of glucose was calculated by the difference value in the presence and absence of GOx.

Blood Glucose Reduction Assay
To assess the efficiency of EPM GOx-Cat on catalytic reduction of blood glucose, whole blood was extracted from an adult mouse with the addition of heparin and extra glucose. Then, 0.5, 1.0, or 2.0 µL of EPM GOx-Cat suspension was added into 200 µL of whole blood in a water bath at 37 • C. Then, 10 µL of blood was withdrawn from the reaction system at fixed time intervals for immediate detection of the blood glucose concentration.

Characterization
Transmission electron microscope (TEM) images of EPM were collected from the Hitachi H-600 microscope (Tokyo, Japan). The high-resolution TEM images of MSS were Nanomaterials 2022, 12, 625 5 of 13 obtained from Tecnai G2 F20 S-Twin (Hillsboro, OR, USA). Scanning electron microscope (SEM) images were collected from a field emission SEM (Zeiss Ultra 55, Oberkochen, Germany). Fourier-transform infrared (FTIR) spectra were measured on a Nicolet 6700 Spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). The fluorescence images were obtained from a Leica DM4000 B microscope (Wetzlar, HE, Germany). The UV-vis spectra were recorded on a Shimadzu, UV-2600 spectrophotometer (Shimadzu, Kyoto, Japan). The blood glucose concentration was monitored by a one-touch blood glucose meter (Lifescan Inc., Seattle, WA, USA).

Results and Discussion
The MSS with an average particle size of 800 nm was synthesized according to a method reported earlier [38]. High resolution TEM and SEM images revealed that the MSS has a dual pore distribution (~3 nm and~20 nm) (Figure 1a,b). GOx, molecular size of 6-7 nm, was used as the therapeutic enzyme for glucose oxidation. Hence, the larger mesopores in the MSS are important for achieving high content of GOx loading inside the particles. To prepare EPM with concentrically encapsulated dual enzymes, catalase and PLL was sequentially assembled on the GOx-loaded MSSs. By using this strategy, various enzymes with tunable content could be facilely assembled on the particle surface to construct concentric compartments. (Scheme 1). It is worth noting that the MSS has dual roles, one is to effectively load enzymes inside the highly porous particle, and the other is to provide a spherical template for LbL assembling the polymeric capsule. The enzyme capsules would not form without using the MSS template, since aggregated enzyme and polyelectrolyte complex would be generated when the negatively charged enzymes were mixed with the positively charged PLL. The TEM images indicated a significantly lower density of the particles due to the removal of the inorganic MSS template (Figure 1c,d). The SEM images of EPMGOx-Cat showed a bulging morphology (Figure 1e,f), primarily due to their high enzyme loading and structural robustness of the EPM. As the GOx molecules were adsorbed into the mesopores through electrostatic interaction, changing of the solution environments (e.g., pH, salt concentration, or in the presence of competitive adsorption molecules) could cause leakage of the physically adsorbed enzyme molecules. It was found that up to 90% of the loaded GOx could be released during the procedure to assemble a PLL/PGA multilayer shell on the MSS to encapsulate the loaded enzyme. To minimize the leakage of loaded GOx, 0.2 wt.% GA solution was used to crosslink the enzyme molecules to form a covalently connected enzyme network inside As GOx has an isoelectric point (IEP) of 4.2, the amino-modification of the MSS is essential to obtain a high enzyme loading through the electrostatic interaction of the Nanomaterials 2022, 12, 625 6 of 13 negatively-charged GOx and the positively-charged MSS surface. The loading of GOx in the MSS particles could be adjusted by changing the mass ratio of MSS to GOx in the incubation solution. When the MSS/GOx mass ratio was set at 2.5, 250 mg of GOx could be loaded in one gram of MSS particles (i.e., 25 wt.% GOx loading), which corresponded to an enzyme immobilization rate of 62.5%. The GOx loading could be increased to 35 wt.% when the MSS/GOx mass ratio was set at 1, however, the enzyme immobilization rate was decreased to~36%. Thereafter, the mass ratio of MSS to GOx was fixed at 2.5 in following experiments.
As the GOx molecules were adsorbed into the mesopores through electrostatic interaction, changing of the solution environments (e.g., pH, salt concentration, or in the presence of competitive adsorption molecules) could cause leakage of the physically adsorbed enzyme molecules. It was found that up to 90% of the loaded GOx could be released during the procedure to assemble a PLL/PGA multilayer shell on the MSS to encapsulate the loaded enzyme. To minimize the leakage of loaded GOx, 0.2 wt.% GA solution was used to crosslink the enzyme molecules to form a covalently connected enzyme network inside the mesopores. There is only ca. 20% of GOx released during the GA crosslinking and the subsequent PLL infiltrating steps ( Figure 2). As the GOx molecules were adsorbed into the mesopores through electrostat action, changing of the solution environments (e.g., pH, salt concentration, or in th ence of competitive adsorption molecules) could cause leakage of the physically ad enzyme molecules. It was found that up to 90% of the loaded GOx could be releas ing the procedure to assemble a PLL/PGA multilayer shell on the MSS to encapsu loaded enzyme. To minimize the leakage of loaded GOx, 0.2 wt.% GA solution w to crosslink the enzyme molecules to form a covalently connected enzyme network the mesopores. There is only ca. 20% of GOx released during the GA crosslinking subsequent PLL infiltrating steps ( Figure 2). GOx was immobilized steadily in the MSS after GA crosslinking, however, the immobilized enzymes would be released provided the MSS template was re GOx was immobilized steadily in the MSS after GA crosslinking, however, 80% of the immobilized enzymes would be released provided the MSS template was removed. Inspired by previous work that enzymes could be stably immobilized inside nanoporous polymer particles by bridging proteins with infiltrated polymer [39,40], we combined polymer infiltrating and LbL-based multienzyme assembling with a crosslinking process to couple the amine groups and carboxyl groups under the catalyst of EDC. In this process, the s-PLL with less steric effect was used to infiltrate into the GOx-loaded mesopores and bridging GOx (Scheme 1). Then, a protective PLL/PGA bilayer shell was LbL assembled on the particle surface to encapsulate the enzymes. Finally, the MSS template was dissolved by adding an HF buffer. About 97% of the catalytic activity of enzymes was retained after HF treatment, implying the successful enzyme immobilization through the s-PLL infiltrating and EDC crosslinking process. The stable immobilization of GOx in the crosslinked PLL/PGA multilayer capsule could be further demonstrated from the leaching tests, in which 93-95% of GOx activity was retained in the capsules after shaking the EPM GOx in PBS at 37 • C for 24 h (Figure 2). The final product, EPM GOx-Cat , contained about 430 µg mL −1 of GOx (55 U mL −1 ) and 58 µg mL −1 of catalase (2651 U mL −1 ). In a similar leaching test in PBS buffer, about 98% of catalase was retained in EPM GOx-Cat . The enzyme retention rate in the capsules was directly related to the structural stability of EPM. The high retaining rate of GOx and catalase in the EPM clearly suggested the structural stability of the crosslinked EPM.
To monitor component changes of the samples, FTIR spectra of relative intermediates including MSSs, (PGA/PLL) 3 (Catalase/PLL) 2 @GOx-loaded MSSs (EMPM), and EPM GOx-Cat were detected (Figure 3). The FTIR spectra of MSSs and EMPM both exhibited peaks at the bands attributed to Si−O−Si bending vibration (463 cm −1 ), Si−O−Si symmetric stretching (801 cm −1 ), surface Si−OH groups (957 cm −1 ), and Si−O−Si asymmetric stretching (1000-1200 cm −1 ) [41]. These peaks were totally disappeared in the EPM, indicating the efficient removal of the MSS template. The absorption band at 1638-1653 cm −1 displayed in the MSSs and EMPM was mainly attributed to the overlap of the amino group (1630 cm −1 ) and water molecules (1652 cm −1 ) [41,42]. After enzyme loading and PLL/PGA polypeptide multilayer assembling, the FTIR spectra of EMPM showed a series of additional peaks that could be attributed to COO − symmetric stretching vibration (1401 cm −1 ), CH 3 /CH 2 in-plane bending vibration (1450 cm −1 ), COO − antisymmetric stretching vibration and NH 3 + symmetric stretching vibration (1548 cm −1 ) [23]. These peaks were retained in the EPM, suggesting that the enzymes and polypeptides were well preserved after the removal of the MSS template.
μg mL −1 of GOx (55 U mL −1 ) and 58 μg mL −1 of catalase (2651 U mL −1 ). In a sim test in PBS buffer, about 98% of catalase was retained in EPMGOx-Cat. The enzy rate in the capsules was directly related to the structural stability of EPM. The ing rate of GOx and catalase in the EPM clearly suggested the structural st crosslinked EPM.
To monitor component changes of the samples, FTIR spectra of relative in including MSSs, (PGA/PLL)3(Catalase/PLL)2@GOx-loaded MSSs (EMPM), an were detected (Figure 3). The FTIR spectra of MSSs and EMPM both exhibi the bands attributed to Si−O−Si bending vibration (463 cm −1 ), Si−O−Si symm ing (801 cm −1 ), surface Si−OH groups (957 cm −1 ), and Si−O−Si asymmetric stret 1200 cm −1 ) [41]. These peaks were totally disappeared in the EPM, indicating removal of the MSS template. The absorption band at 1638-1653 cm −1 disp MSSs and EMPM was mainly attributed to the overlap of the amino group (16 water molecules (1652 cm −1 ) [41,42]. After enzyme loading and PLL/PGA multilayer assembling, the FTIR spectra of EMPM showed a series of addi that could be attributed to COO − symmetric stretching vibration (1401 cm −1 ), plane bending vibration (1450 cm −1 ), COO − antisymmetric stretching vibrati symmetric stretching vibration (1548 cm −1 ) [23]. These peaks were retained suggesting that the enzymes and polypeptides were well preserved after th the MSS template. Homogeneity of the enzyme-loaded EPM was studied via fluorescence images. For distinguishing different species in the EPM, three fluorescent dy tinguishing excitation and emission wavelengths, including FITC (green), D Homogeneity of the enzyme-loaded EPM was studied via fluorescence microscopy images. For distinguishing different species in the EPM, three fluorescent dyes with distinguishing excitation and emission wavelengths, including FITC (green), DCCA (blue), and SRB (red), were used to label GOx, catalase, and PLL, respectively. The brightly colored particles seen in the images (Figure 4a and SRB (red), were used to label GOx, catalase, and PLL, respectively. The brightly col ored particles seen in the images (Figure 4a-c) reflected the homogeneous loading of each component in the capsules. The fluorescent microscope images also suggested the mono dispersity of the EPMGOx-Cat in solution. The merged fluorescent image shown in Figure 4d clearly demonstrates the colocalization of GOx, catalase, and PLL in EPMGOx-Cat. Roles of the concentric compartment on scavenging the toxic reaction intermediat H2O2 and cycling the cascade reaction were studied. To this purpose, EPMGOx-Cat was used to oxidize glucose in PBS solution in an enclosed environment. The concentration of glu cose decreased by EPMGOx and EPMGOx-Cat showed approximately linear reduction (Figur 5a), with a total glucose consumption of 3.9 mM and 7.2 mM, respectively, in 6 h. Im portantly, the reaction rates ofEPMGOx (10.8 μM min −1 ) and EPMGOx-Cat (20.0 μM min −1 were significantly different, exhibiting a synergistic effect of EPMGOx-Cat. Catalase cata lyzed the decomposition of cytotoxic intermediates H2O2 and generates O2, which could be further reused in the GOx catalyzed glucose oxidation reaction. The cascade reaction was performed circularly based on Equations (1) and (2), in which the cycle ratio could b significantly enhanced by the rapid H2O2 scavenging and the concentrically arranged bi oreactor compartments in a microenvironment.  (Figure 5b). On contrary, only 74 μM H2O2 was found in sur rounding media after decomposing 7.2 mM glucose by EPMGOx-Cat under the same reaction conditions. In other words, 99% H2O2 generated in the EPMGOx-Cat system was successfully eliminated by downstream scavenging enzyme-catalase. In the case of the protease pre treated EPMGox, the content of the H2O2 produced was very close to that of H2O2 gener ated from the untreated EPMGox, demonstrating that the capsule can protect the loaded enzymes from protease degradation. To the best of our knowledge, the H2O2 scavenging in the EPMGOx-Cat system is much more efficient than other multienzyme nanoreactors Roles of the concentric compartment on scavenging the toxic reaction intermediate H 2 O 2 and cycling the cascade reaction were studied. To this purpose, EPM GOx-Cat was used to oxidize glucose in PBS solution in an enclosed environment. The concentration of glucose decreased by EPM GOx and EPM GOx-Cat showed approximately linear reduction (Figure 5a), with a total glucose consumption of 3.9 mM and 7.2 mM, respectively, in 6 h. Importantly, the reaction rates ofEPM GOx (10.8 µM min −1 ) and EPM GOx-Cat (20.0 µM min −1 ) were significantly different, exhibiting a synergistic effect of EPM GOx-Cat . Catalase catalyzed the decomposition of cytotoxic intermediates H 2 O 2 and generates O 2 , which could be further reused in the GOx catalyzed glucose oxidation reaction. The cascade reaction was performed circularly based on Equations (1) and (2), in which the cycle ratio could be significantly enhanced by the rapid H 2 O 2 scavenging and the concentrically arranged bioreactor compartments in a microenvironment.
H 2 O 2 generated from glucose decomposition by EPM GOx exhibited a linear increase reaching up to 4 mM in 6 h (Figure 5b). On contrary, only 74 µM H 2 O 2 was found in surrounding media after decomposing 7.2 mM glucose by EPM GOx-Cat under the same reaction conditions. In other words, 99% H 2 O 2 generated in the EPM GOx-Cat system was successfully eliminated by downstream scavenging enzyme-catalase. In the case of the protease pre-treated EPMGox, the content of the H 2 O 2 produced was very close to that of H 2 O 2 generated from the untreated EPMGox, demonstrating that the capsule can protect the loaded enzymes from protease degradation. To the best of our knowledge, the H 2 O 2 scavenging in the EPM GOx-Cat system is much more efficient than other multienzyme nanoreactors, such as co-localized multienzyme nanocomplex (about 85%) [10] or supramolecular hydrogel (about 88%) [37]. The culture media of HeLa cells was further used to demonstrate the cytotoxicity of H 2 O 2 generated during the glucose oxidation (Figure 5c). After 3 h of incubation, the generated H 2 O 2 from EPM GOx resulted in 71% of mortality, while that from EPM GOx-Cat resulted in only 2% of mortality, very close to the peptide-multilayered microcapsules (PM), in which no H 2 O 2 was generated.
such as co-localized multienzyme nanocomplex (about 85%) [10] or supramolec drogel (about 88%) [37]. The culture media of HeLa cells was further used to demo the cytotoxicity of H2O2 generated during the glucose oxidation (Figure 5c). Afte incubation, the generated H2O2 from EPMGOx resulted in 71% of mortality, while th EPMGOx-Cat resulted in only 2% of mortality, very close to the peptide-multilayered capsules (PM), in which no H2O2 was generated. Although catalase is commonly considered as an efficient H2O2 scavenging for therapeutic purposes, the synergistic effect of GOx and catalase on nanoreacto has been rarely reported [10,37,43]. In living organisms, a physical tunnel of enzym plex is developed to limit and channel the intermediates essentially [17]. It m pointed out that current co-loading or multicompartments of multienzymes in a actor fail to limit the free diffusion of reaction intermediates. In the case of a G catalase dual-enzyme system, the generated H2O2 and regenerative O2 will keep fusion and only part of these is involved in the cascade enzymatic reaction, resu compromised H2O2 scavenging. Herein, we designed the concentric compartmen MGOx-Cat to limit the release of toxic H2O2 and further utilized it as a useful intermed O2 generation. On the whole, during the enzymatic reaction, all substrates and p form a certain concentration gradient (Figure 5d), imposing mass transfer restric the reaction substrates (outside-in approach) and the products (inside-out approa oxidation of one glucose molecule catalyzed by GOx consumes one O2 and genera H2O2 molecule, which could be decomposed by catalase to produce half O2 for re use. The high-efficiency clearance of H2O2 was achieved by arranging catalase downstream position of the H2O2 concentration gradient (Figure 5b). Significantly generated O2 from catalase decomposition of H2O2 will diffuse inward to EPM du O2 partial pressure gradient (Figure 5d). The iterative cycle of H2O2 doubled the uti ratio of original O2 from 50% to 100% and further enhanced the oxidation rate of Although catalase is commonly considered as an efficient H 2 O 2 scavenging enzyme for therapeutic purposes, the synergistic effect of GOx and catalase on nanoreactor design has been rarely reported [10,37,43]. In living organisms, a physical tunnel of enzyme complex is developed to limit and channel the intermediates essentially [17]. It must be pointed out that current co-loading or multicompartments of multienzymes in a nanoreactor fail to limit the free diffusion of reaction intermediates. In the case of a GOx and catalase dual-enzyme system, the generated H 2 O 2 and regenerative O 2 will keep free diffusion and only part of these is involved in the cascade enzymatic reaction, resulting in compromised H 2 O 2 scavenging. Herein, we designed the concentric compartment of EPM GOx-Cat to limit the release of toxic H 2 O 2 and further utilized it as a useful intermediate for O 2 generation. On the whole, during the enzymatic reaction, all substrates and products form a certain concentration gradient (Figure 5d), imposing mass transfer restriction on the reaction substrates (outside-in approach) and the products (inside-out approach). The oxidation of one glucose molecule catalyzed by GOx consumes one O 2 and generates one H 2 O 2 molecule, which could be decomposed by catalase to produce half O 2 for recycling use. The high-efficiency clearance of H 2 O 2 was achieved by arranging catalase on the downstream position of the H 2 O 2 concentration gradient (Figure 5b). Significantly, the regenerated O 2 from catalase decomposition of H 2 O 2 will diffuse inward to EPM due to the O 2 partial pressure gradient (Figure 5d). The iterative cycle of H 2 O 2 doubled the utilization ratio of original O 2 from 50% to 100% and further enhanced the oxidation rate of glucose compared with EPM GOx in the same reaction condition (Figure 5a). The low emission of H 2 O 2 and synergistic effects were simultaneously achieved by concentric encapsulation of dual enzymes. Furthermore, the roles of the concentric compartment and created microenvironment are not influenced by diluting or fast diffusing liquid such as blood, which brought extra potentials for therapeutic nanoreactor application.
The therapeutic application of nanoreactors on metabolic disorder redressing was demonstrated through fast blood glucose oxidation by EPM GOx-Cat . Different doses of EPM GOx-Cat were added into mouse whole blood to decrease blood glucose. As shown in Figure 6, the initial glucose consumption rate of EPM GOx-Cat in blood was about 400 µM min −1 , much higher than that of 20 µM min −1 obtained in PBS solution (Figure 5a). This enhancement was primarily ascribed to the increase of the total blood oxygen capacity induced by hemoglobin in red blood cells. Meanwhile, the glucose oxidation rate by EPM was dose-dependent. For instance, the EPM GOx-Cat with a dosage of 1%, 0.5%, and 0.25% (v/v) required a treatment time of around 15, 30, and 60 min, respectively, to redress the high blood glucose (10.5 mM) to normal blood glucose (~5 mM). ure 6, the initial glucose consumption rate of EPMGOx-Cat in blood was about 400 μM min −1 , much higher than that of 20 μM min −1 obtained in PBS solution (Figure 5a). This enhancement was primarily ascribed to the increase of the total blood oxygen capacity induced by hemoglobin in red blood cells. Meanwhile, the glucose oxidation rate by EPM was dosedependent. For instance, the EPMGOx-Cat with a dosage of 1%, 0.5%, and 0.25% (v/v) required a treatment time of around 15, 30, and 60 min, respectively, to redress the high blood glucose (10.5 mM) to normal blood glucose (~5 mM).
It must be pointed out that over-consumption of blood glucose will cause hypoglycemia, which is dangerous for health. Even taking account of fast glucose supplement from hepatic glycogen, therapeutic nanoreactor should also consider the problem of overconsumption of glucose or other biomolecules. The ratio of EPMGOx-Cat and blood could be adjusted to avoid the over-consumption of blood glucose, as the dose of EPMGOx-Cat was directly proportional to the amount of glucose consumption ( Figure 6). Interestingly, the glucose oxidation rate of all EPM dosages decreased to a similar level of 0.04-0.06 mM min −1 , when the glucose concentration is approaching the normal blood glucose. For instance, the glucose consumption rate of 1% EPM sample declined around 9 times (from 470 μM min −1 to 50 μM min −1 ) in 60 min, while those of 0.5% and 0.25% EPM declined around 7 times and 4 times, respectively. The reduction of reaction rate was mainly caused by the decreased concentration of substrates. This typical enzymatic reaction feature was further amplified by concentration-dependent diffusion of substances in EPM.

Conclusions
In this work, we have designed and fabricated potent therapeutic capsules by loading GOx inside MSS, assembly of catalase on the shell, and subsequent encapsulating the loaded enzymes with a PLL/PGA multilayer shell. After crosslinking, GOx and catalase were stably trapped inside the concentric multicompartments of EPM with uniform distribution and high enzyme loading. Owing to the concentrically encapsulated GOx and catalase and their deliberately arranged locations, 1% (v/v) EPMGOx-Cat can potently redress hyperglycemia from 10.5 to 5 mM in 15 min with an initial glucose consumption rate of 400 μM min −1 in mouse whole blood. Meanwhile, 99% of the cytotoxic reaction intermediates (H2O2) scavenging ratio and double glucose consumption rate were realized. The highly efficient and concentrically compartmented features of EPM holds enormous promise in developing more multienzyme therapeutic nanoreactors for addressing metabolic disorders and scavenging toxic intermediates. It must be pointed out that over-consumption of blood glucose will cause hypoglycemia, which is dangerous for health. Even taking account of fast glucose supplement from hepatic glycogen, therapeutic nanoreactor should also consider the problem of overconsumption of glucose or other biomolecules. The ratio of EPM GOx-Cat and blood could be adjusted to avoid the over-consumption of blood glucose, as the dose of EPM GOx-Cat was directly proportional to the amount of glucose consumption ( Figure 6). Interestingly, the glucose oxidation rate of all EPM dosages decreased to a similar level of 0.04-0.06 mM min −1 , when the glucose concentration is approaching the normal blood glucose. For instance, the glucose consumption rate of 1% EPM sample declined around 9 times (from 470 µM min −1 to 50 µM min −1 ) in 60 min, while those of 0.5% and 0.25% EPM declined around 7 times and 4 times, respectively. The reduction of reaction rate was mainly caused by the decreased concentration of substrates. This typical enzymatic reaction feature was further amplified by concentration-dependent diffusion of substances in EPM.

Conclusions
In this work, we have designed and fabricated potent therapeutic capsules by loading GOx inside MSS, assembly of catalase on the shell, and subsequent encapsulating the loaded enzymes with a PLL/PGA multilayer shell. After crosslinking, GOx and catalase were stably trapped inside the concentric multicompartments of EPM with uniform distribution and high enzyme loading. Owing to the concentrically encapsulated GOx and catalase and their deliberately arranged locations, 1% (v/v) EPM GOx-Cat can potently redress hyperglycemia from 10.5 to 5 mM in 15 min with an initial glucose consumption rate of 400 µM min −1 in mouse whole blood. Meanwhile, 99% of the cytotoxic reaction intermediates (H 2 O 2 ) scavenging ratio and double glucose consumption rate were realized. The highly efficient and concentrically compartmented features of EPM holds enormous promise in developing more multienzyme therapeutic nanoreactors for addressing metabolic disorders and scavenging toxic intermediates.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/nano12040625/s1, Figure S1: UV-Vis spectra of GOx at different concentrations (a) and the standard calibration curve by using the absorbance at 276 nm (b).
Author Contributions: Data curation, C.D. and X.L.; Funding acquisition, C.D.; Investigation, Q.J. and D.Y.; Methodology, Q.J. and D.Y.; Writing-original draft, C.D. and X.L.; Writing-review and editing, C.D. and Q.J. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Natural Science Foundation of Zhejiang Province (LQ21B030001) and the National Natural Science Foundation of China (22175132).

Institutional Review Board Statement:
The animal study protocol was approved by Ethics Committee of Wenzhou Medical University (protocol code WIUCAS21081601,2021.08.16).

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.

Conflicts of Interest:
The authors declare no conflict of interest.