Methods and applications of oral delivery of proteases by recombinant probiotics using the TOSS secretion system

Abstract

This invention relates to a recombinant *Escherichia coli* containing at least one therapeutically active ingredient, preferably a therapeutic enzyme, within its outer membrane vesicles (OMV). The invention also relates to the use of this recombinant *EcN* to deliver the therapeutically active ingredient (preferably a therapeutic enzyme) into circulation via the OMV, penetrating the intact intestinal epithelial barrier. This OMV containing the therapeutic enzyme is capable of stably catalyzing various detoxification reactions in circulation, particularly for the clinical treatment of metabolic disorders such as hyperuricemia.

Description

Technical Field

[0001] This invention belongs to the field of genetic engineering. More specifically, this invention relates to an engineered bacterial TOSS system and its use in the oral delivery of therapeutically active ingredients (such as proteases) to treat diseases. Background Technology

[0002] Among various drug delivery methods, oral administration has advantages in terms of non-invasiveness, patient compliance, and convenience. However, due to its susceptibility to degradation, fragile tertiary structure, and poor permeability to the gastrointestinal epithelium, few therapeutic proteins have been approved for oral administration in clinical practice. There is an urgent need to develop oral protein delivery systems that can overcome various physiological barriers in the gastrointestinal tract, including irritating acids and proteases in the stomach, a viscous mucus layer, and the tightly packed epithelial cell barrier of the intestine. To overcome these absorption barriers, nanoparticle (NP)-based drug delivery systems have been developed for oral protein delivery; however, NP systems face complex manufacturing processes and inefficient cargo loading and delivery. Therefore, developing convenient and efficient oral protein delivery systems is of great significance for the clinical use of therapeutic proteins.

[0003] Genetically engineered symbiotic bacteria are promising live drugs; however, the therapeutic molecules they produce are often confined to the area around their gut colonization region. The Phase 3 clinical failure of the engineered probiotic SYNB1934 in the treatment of phenylketonuria may be attributed to its limited efficacy in reducing toxic blood metabolites solely by degrading metabolites in the gut (Adolfsen, KJ et al. Improvement of a synthetic live bacterial therapeutic for phenylketonuria with biosensor-enabled enzyme engineering. Nat. Commun. 12, 6215 (2021); Puurunen, MK et al. Safety and pharmacodynamics of an engineered E. coli Nissle for the treatment of phenylketonuria: a first-in-human phase 1 / 2 study. Nat. Metab. 3, 1125-1132 (2021); Vockley, J. et al. Efficacy and safety of a synthetic biotic for treatment of phenylketonuria: a phase 2 clinical trial. Nat. Metab. 5, 1685-1690 (2023)). The main challenge in achieving oral symbiotic bacterial system delivery is how to deliver stable therapeutic active ingredients (such as proteins) in the digestive tract and further cross the intestinal barrier into circulation.

[0004] Outer membrane vesicles (OMVs) are considered to be the type zero secretion system (TOSS) of bacteria, mediating microbe-host or microbe-microbe interactions in the gut (Guerrero-Mandujano, A., Hernández-Cortez, C., Ibarra, JA & Castro-Escarpulli, G. The outer membrane vesicles: secretion system type zero. Traffic 18, 425-432 (2017)). OMVs are formed by budding from the outer membrane and subsequently separate from the cell body. In Gram-negative bacteria, the contents of the OMV are separated from the cytoplasm by the inner membrane. Escherichia coli Nissle 1917 (EcN) is a classic Gram-negative probiotic chassis for synthetic biology and has been approved by the FDA as GRAS (Generally Recognized As Safe) (Sonnenborn, U. & Schulze, J. The non-pathogenic Escherichia coli strain Nissle1917 - features of a versatile probiotic. Microbial. Ecology in Health and Disease 21, 122-158 (2009)). Summary of the Invention

[0005] This invention provides a technique for oral delivery of proteins via outer membrane vesicles (OMV) using an engineered bacterial type-zero secretion system (TOSS). Figure 1 ).

[0006] This invention is based on the unexpected discovery that OMVs produced in situ in the intestine by *E. coli*, such as Nissle 1917 (EcN), can penetrate the intact intestinal epithelial barrier and enter circulation. This penetration process involves pinocytosis and actin-dependent endocytosis. EcNs were engineered to load various enzymes into OMVs. These secreted, enzyme-encapsulated OMVs stably catalyze various detoxification reactions in the presence of digestive fluids and enter circulation. This TOSS-based protein secretion system is compatible with various protein payloads, allowing different protein cargoes to be encapsulated in OMVs, making protein-loaded OMVs a promising biocatalyst for various detoxification reactions. Using hyperuricemic mice and uricase OMVs as examples, we demonstrated that engineered EcNs equipped with this modified TOSS exhibit superior therapeutic efficacy compared to directly secreted protein systems. The enzyme-encapsulated OMVs also effectively detoxified human serum samples, highlighting the clinical therapeutic potential of this system for metabolic disorders.

[0007] Therefore, according to one aspect of the invention, a recombinant Escherichia coli such as Nissle 1917 (EcN) is provided, which contains at least one (endogenous or exogenous) therapeutic active ingredient in its outer membrane vesicles (OMV), wherein the therapeutic active ingredient is selected from therapeutic peptides, therapeutic proteins, antigens, therapeutic nucleic acids and small molecule compounds, preferably the (endogenous or exogenous) therapeutic active ingredient is a therapeutic peptide or therapeutic protein encapsulated in the OMV.

[0008] In one embodiment, the therapeutic active ingredient is a therapeutic enzyme, preferably a therapeutic enzyme for (preferably in the circulatory system) metabolic diseases, genetic diseases, cardiovascular diseases, gastrointestinal diseases, inflammatory diseases, or cancer, more preferably wherein:

[0009] 1) The metabolic disease is selected from abnormal glucose metabolism, abnormal lipid metabolism, abnormal calcium and phosphorus metabolism, abnormal bone metabolism, and abnormal purine metabolism. Most preferably, the metabolic disease is hyperuricemia and / or the therapeutic enzyme is uricase, the metabolic disease is peroxisome disease and / or the therapeutic enzyme is catalase, or the metabolic disease is hyperlactatemia and / or the therapeutic enzyme is lactate oxidase, or the metabolic disease is hyperalcoholemia and / or the therapeutic enzyme is ethanol oxidase and acetaldehyde oxidase.

[0010] 2) The hereditary disease is selected from phenylketonuria, mucopolysaccharidosis, Pompe disease, Gaucher disease, Fabry disease, adenosine deaminase deficiency with severe combined immunodeficiency, hypophosphatase disease, and lysosomal acid lipase deficiency. Preferably, the therapeutic enzyme is selected from phenylalanine deaminase, α-L-iduronase, iduronate sulfate, α-glucosidase, vidarabine, β-galactosidase, adenosine deaminase, alkaline phosphatase, and lysosomal acid lipase. Most preferably, the metabolic disease is phenylketonuria and / or the therapeutic enzyme is phenylalanine deaminase.

[0011] 3) The cardiovascular diseases mentioned are selected from coronary heart disease, peripheral artery disease, cerebrovascular disease, rheumatic heart disease, congenital heart disease, deep vein thrombosis and pulmonary embolism. Preferably, the therapeutic enzymes are selected from streptokinase, urokinase, tissue plasminogen activator (t-PA), snake venom defibrase, hirudin, lumbrokinase, nattokinase and deampicase.

[0012] 4) The gastrointestinal diseases are selected from indigestion, insufficient pancreatic juice secretion and gallbladder fibrosis, and the therapeutic enzymes are preferably selected from protease, amylase, lipase, cellulase and pancreatic enzymes;

[0013] 5) The inflammatory disease is selected from AIDS, endocarditis, osteomyelitis, pneumonia, toxic shock syndrome, food poisoning, skin infections (such as folliculitis, abscesses, and mastitis), and burns. Preferably, the therapeutic enzyme is selected from lysozyme (e.g., lysostaphylococcal lysozyme), trypsin, chymotrypsin, bromelain, collagenase, and chondroitin sulfate; and / or

[0014] 6) The cancer is selected from acute lymphoblastic leukemia, lymphosarcoma cell leukemia, granulocytic leukemia, melanoma, lung cancer, prostate cancer, hepatocellular carcinoma, pancreatic cancer, glioblastoma multiforme, and thoracic vertebral carcinoma. Preferably, the therapeutic enzyme is a metabolic enzyme for amino acids required by the tumor and a prodrug-converting enzyme. More preferably, it is selected from L-asparaginase, arginine deiminase, glutaminase, glucose oxidase, kynurenase, carboxypeptidase A, and β-glucuronidase.

[0015] In another embodiment, the therapeutic nucleic acid is selected from small interfering nucleic acids (siRNA), microRNA (miRNA), antisense nucleotide (ASO) messenger RNA (mRNA), and nucleic acid aptamers.

[0016] In another preferred embodiment, the recombinant Escherichia coli, such as Nissle 1917 (EcN), lacks the nlpI gene. OMV biosynthesis is increased by genomic deletion of the outer membrane lipoprotein-encoding gene nlpI. In one preferred embodiment, the nucleotide sequence of the nlpI gene is shown in SEQ ID No. 1.

[0017] In another preferred embodiment, the therapeutic peptide or therapeutic protein is fused with a signal peptide secreted to the OMV, preferably selected from Sec signal peptide, Tat signal peptide, and Srp signal peptide. In one preferred embodiment, the amino acid sequences of the Sec signal peptide, Tat signal peptide, and Srp signal peptide are shown in SEQ ID No. 2, SEQ ID No. 3, and SEQ ID No. 4, respectively.

[0018] In another preferred embodiment, the peptide or therapeutic protein is selected from uricase, catalase, lactate oxidase, phenylalanine deaminase, ethanol oxidase, acetaldehyde oxidase, or combinations thereof.

[0019] In another aspect of the invention, an outer membrane vesicle (OMV) derived from recombinant Escherichia coli such as Nissle 1917 (EcN) or prepared therefrom is provided, wherein the OMV contains the at least one therapeutically active ingredient, preferably the OMV is unmodified.

[0020] In another aspect of the invention, a method for preparing the OMV according to the invention is provided, the method comprising the step of expressing a therapeutic peptide or therapeutic protein in the periplasm of Escherichia coli such as Nissle 1917 (EcN).

[0021] In one specific embodiment, a therapeutic peptide or therapeutic protein is expressed in the periplasm of the EcN using an expression vector, the expression vector comprising a nucleic acid sequence encoding the therapeutic peptide or therapeutic protein operatively linked to a signal sequence of the periplasmic protein.

[0022] Targeting of therapeutic peptides or proteins can be achieved by fusing the signal sequence of proteins naturally found in the periplasm and / or OMV with a therapeutic peptide or protein. Protein translocation across the inner membrane and into the periplasm can occur, for example, via one of three pathways: SecB-dependent (SEC), signal recognition particle (SRP), or trans-arginine esterification (TAT).

[0023] In one specific embodiment, the therapeutic peptide or therapeutic protein is selected from uricase, catalase, lactate oxidase, phenylalanine deaminase, ethanol oxidase, acetaldehyde oxidase, or combinations thereof. Preferably, the expression of the therapeutic peptide or therapeutic protein is controlled by an inducible or constitutive promoter (i.e., operatively linked to an inducible or constitutive promoter). More preferably, the promoter is selected from the TAC promoter, FNRS promoter, NA promoter, NA618 promoter, and J23117 promoter. In a preferred embodiment, the nucleotide sequences of the TAC promoter, FNRS promoter, NA promoter, NA618 promoter, and J23117 promoter are shown in SEQ ID No. 30, SEQ ID No. 26, SEQ ID No. 27, SEQ ID No. 28, and SEQ ID No. 29, respectively.

[0024] In one specific embodiment, the catalase is derived from Escherichia coli.

[0025] In one specific embodiment, the uricase is derived from C. utilis or S. auratus AGR0001.

[0026] In one specific embodiment, the phenylalanine deaminase is derived from P. luminescens or Y. enterocolitica.

[0027] In one specific embodiment, the lactate oxidase is derived from A. viridans.

[0028] In one specific embodiment, the acetaldehyde oxidase is derived from P. blakesleeanus NRRL1555 or C. reinhardtii or C. cardunculus var. scolymus or L. sativa or S. punctatus DAOMBR117.

[0029] In one specific implementation, recombinant Escherichia coli, such as Nissle 1917 (EcN), has the outer membrane lipoprotein encoding gene nlpI deleted from its genome to increase OMV biosynthesis.

[0030] In another aspect of the invention, there is provided a pharmaceutical composition for oral systemic administration to a subject, comprising: 1) a recombinant *Escherichia coli* such as Nissle 1917 (EcN) or an outer membrane vesicle (OMV) according to the invention; and 2) a pharmaceutically acceptable adjuvant, wherein the pharmaceutical composition, the recombinant *Escherichia coli* such as Nissle 1917 (EcN) or the OMV are formulated for and / or ( wholly / only) dependent on the OMV to penetrate the intestinal epithelial barrier, thereby delivering the therapeutically active ingredient to the circulatory system of the subject, achieving systemic administration.

[0031] In another aspect of the invention, a pharmaceutical composition for systemic administration of a drug to a subject via the intestinal mucosa is provided, comprising: 1) a recombinant *Escherichia coli* such as Nissle 1917 (EcN) or an outer membrane vesicle (OMV) according to the invention; and 2) a pharmaceutically acceptable adjuvant, wherein the pharmaceutical composition, the recombinant *Escherichia coli* such as Nissle 1917 (EcN) or the OMV are formulated for and / or ( wholly / only) dependent on the OMV to penetrate the intestinal epithelial barrier, thereby delivering the therapeutically active ingredient to the circulatory system of the subject, achieving systemic administration.

[0032] In another aspect of the invention, the use of recombinant Escherichia coli such as Nissle 1917 (EcN) or outer membrane vesicles (OMV) according to the invention is provided in the preparation of a medicament for oral or intestinal mucosal administration (to a subject), wherein the medicament, recombinant Escherichia coli such as Nissle 1917 (EcN) or OMV is formulated for and / or ( wholly / only) dependent on the OMV to penetrate the intestinal epithelial barrier, thereby delivering the therapeutically active ingredient to the circulatory system (of the subject) for systemic administration.

[0033] In another aspect of the invention, the use of recombinant Escherichia coli such as Nissle 1917 (EcN) or outer membrane vesicles (OMV) according to the invention is provided in the preparation of a medicament for oral or intestinal mucosal administration (to a subject), wherein the medicament, recombinant Escherichia coli such as Nissle 1917 (EcN) or OMV is formulated for and / or ( wholly / only) dependent on the OMV to penetrate the intestinal epithelial barrier, thereby delivering the therapeutically active ingredient to the circulatory system (of the subject) for systemic administration.

[0034] In a preferred embodiment, the therapeutic active ingredient is uricase and / or the subject has hyperuricemia, or the therapeutic active ingredient is catalase and / or the subject has peroxisomal disease, or the therapeutic active ingredient is lactate oxidase and / or the subject has hyperlactatemia, or the therapeutic active ingredient is phenylalanine deaminase and / or the subject has phenylketonuria, or the therapeutic active ingredient is ethanol oxidase and / or acetaldehyde oxidase and / or the subject has hyperalcoholemia.

[0035] In a preferred embodiment, the therapeutic enzyme is derived from microorganisms, plants, or animals, such as mammals, and more preferably from microorganisms.

[0036] In a preferred embodiment, the amino acid sequence of the uricase is as shown in SEQ ID No. 5 or SEQ ID No. 31.

[0037] In a preferred embodiment, the amino acid sequence of the catalase is shown in SEQ ID No. 6.

[0038] In a preferred embodiment, the amino acid sequence of the lactate oxidase is shown in SEQ ID No. 7.

[0039] In a preferred embodiment, the amino acid sequence of the phenylalanine deaminase is shown in SEQ ID No. 8, SEQ ID No. 32 or SEQ ID No. 33.

[0040] In a preferred embodiment, the amino acid sequence of the acetaldehyde oxidase is shown in SEQ ID No. 34, SEQ ID No. 35, SEQ ID No. 36, SEQ ID No. 37, or SEQ ID No. 38.

[0041] In a preferred embodiment, the therapeutic enzyme is expressed in fusion with a marker protein or tag protein. Preferably, the marker protein or tag protein is selected from GFP protein (SEQ ID No. 22), RFP protein (SEQ ID No. 23), His tag protein (SEQ ID No. 24), and HA tag protein (SEQ ID No. 25). Attached Figure Description

[0042] The above features and advantages of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, wherein:

[0043] Figure 1 A schematic diagram illustrating the oral administration of therapeutic proteins by symbiotic bacteria, detoxifying blood metabolites via a modified TOSS. The EcN is engineered to enhance OMV biosynthesis through genomic deletion of the outer membrane lipoprotein-encoding gene nlpI, heterologously overexpressing the protein of interest (POI) on a stable plasmid in the absence of antibiotic selection (by deleting the essential gene thyA from the genome and compensating for it on the plasmid), and endogenously loading the POI into the OMV under the guidance of a periplasmic targeted secretion signal peptide. Following oral administration of our engineered EcN with modified TOSS, the protein-loaded OMV generated in situ by the engineered EcN is resistant to proteolysis, penetrates the intact intestinal epithelial barrier, and enters the circulation. Pinocytosis and actin-dependent pathways are involved in this transcellular transport. Engineered EcNs equipped with modified TOSS can deliver therapeutic proteins into circulation and detoxify blood metabolites, providing an oral protein delivery strategy.

[0044] Figure 2 1. Analysis of OMV yield and morphology derived from wild-type EcN and recombinant EcNΔnlpI. a. OMV yield from wild-type EcN and recombinant EcNΔnlpI was determined by BCA protein quantification (n=3). Data are mean ± SD. b. Transmission electron microscopy images of OMV from EcN and EcNΔnlpI, showing similar spherical morphology. Scale bar = 200 nm. c. Zeta potentials of OMV from EcN and EcNΔnlpI (n=3). Data are mean ± SD. d. Size distribution of OMV from wild-type EcN and recombinant EcNΔnlpI was determined by nanoparticle tracking analysis (NTA).

[0045] Figure 3Engineered EcNs based on the TOSS secretion system can secrete protein-encapsulating OMVs and package various different "cargoes" within a single OMV. a. Western blot analysis was performed on the GFP abundance in OMV samples from EcNΔnlpI-OG (abbreviated as "OG"), EcNΔnlpI-GR (abbreviated as "OGR"), and EcNΔnlpI (abbreviated as "NC") using an anti-GFP antibody. b. After modification with the TOSS and endogenous loading system, the engineered EcN was able to secrete protein-encapsulating OMVs with an encapsulation efficiency of 97.9%. c. Nanoflow cytometry was used to detect the GFP encapsulation ratio of OMVs, calculated as the ratio of GFP-loaded OMVs to the total OMVs derived from EcNΔnlpI-OG. EVMembrane Red was used to stain the OMV membranes. d. Western blot analysis of RFP abundance was performed on OMV samples containing EcNΔnlpI-OR (abbreviated as "OR"), EcNΔnlpI-GR (abbreviated as "OGR"), and EcNΔnlpI (abbreviated as "NC") using an anti-RFP antibody. e. Engineered EcN-synthesized OMVs can simultaneously encapsulate multiple protein payloads within a single OMV vector, making the synthesized OMVs a promising cascade biocatalyst. f. Polar-SIM super-resolution microscopy revealed that engineered OMVs can simultaneously encapsulate GFP and RFP within a single OMV. Scale bar = 2 μm.

[0046] Figure 4 Images of GFP-loaded OMVs and RFP-loaded OMVs were observed using Polar-SIM super-resolution microscopy. GFP-loaded OMVs are denoted as "OMV". GFP The OMV loaded by the RFP is represented as "OMV". RFP ". Scale bar = 2μm.

[0047] Figure 5The TOSS-based protein secretion system is compatible with various enzymes and can perform various detoxification reactions. A. Western blot analysis was performed using anti-His antibodies to determine the abundance of StlA (S) in OMV secreted by EcNΔnlpI-OS (OS), Lox (L) in OMV secreted by EcNΔnlpI-OL (OL), and Uox (U) in OMV secreted by EcNΔnlpI-OU (OU). OMV secreted by EcNΔnlpI was designated as "NC". B. Western blot analysis was performed using anti-HA antibodies to determine the abundance of KatE (K) in OMV secreted by EcNΔnlpI-OK (OK) and EcNΔnlpI (NC). CF. OMVs from EcNΔnlpI-OS / EcNΔnlpI-OP1 / EcNΔnlpI-OP2 (C), EcNΔnlpI-OK (D), EcNΔnlpI-OL (E), EcNΔnlpI-OU / EcNΔnlpI-OSaU (F), and EcNΔnlpI-OIsA / EcNΔnlpI-OCrA / EcNΔnlpI-OSpA / EcNΔnlpI-OPbA / EcNΔnlpI-OCvA (G) demonstrated the ability to degrade phenylalanine, H2O2, lactic acid, UA, and acetaldehyde in vitro (n=3). OMVs secreted by EcNΔnlpI were designated as "NC". Data are mean ± SD. H. Enzyme-loaded OMVs effectively catalyzed the detoxification reaction. These reactions fall into two groups: decomposition reactions and oxidation reactions uncoupled from NADH. OMVs loaded with StlA and OMVs loaded with KatE can catalyze the decomposition reactions. Oxidases (Uox and Lox) are coupled with endogenous catalases in the OMV, and catalysis is dependent on the H₂O₂ / O₂ cycle rather than NAD. + Oxidation reaction of NADH.

[0048] Figure 6Protein payloads in OMVs offer advantages in protein stability and catalytic activity in simulated intestinal fluid. a. Schematic diagram of Uox secretion methods, involving a T1SS-based direct protein secretion system or a modified T0SS-based protein secretion system. b. Western blot analysis of Uox abundance in intracellular, supernatant, and OMV samples of EcNΔnlpIΔthyA-SU and EcNΔthyA-SUT using an anti-His antibody. c. EcNΔnlpIΔthyA-SU and the T0SS-based EcNΔthyA-SUT showed similar uric acid degradation rates (n=3) in basal medium supplemented with 0.5 mM uric acid under microaerobic conditions. d. EcNΔnlpIΔthyA-SU and EcNΔthyA-SUT showed higher uric acid degradation efficiency (n=3) in simulated intestinal fluid supplemented with 0.5 mM uric acid under microaerobic conditions. e. Comparison of enzyme activity of purified protein (T1SS-Uox) or OMV loaded with protein (OMV-Uox) (derived from EcNΔthyA-SUT or EcNΔnlpIΔthyA-SU, respectively) in PBS at different time points at 37°C (n=3). f, g. Stability of T1SS-Uox and OMV-Uox was assessed by monitoring enzyme activity (f) and Uox protein levels (g) in simulated intestinal fluid at 37°C (n=3). Data are mean ± SD. P-values ​​were determined by two-way ANOVA using a Tukey multiple comparison test (* indicates P < 0.05).

[0049] Figure 7 In the absence of antibiotic options, a stable plasmid system was established in engineered EcN. a. Growth curves of the EcN strain cultured in M9 basal medium, using OD... 600Monitoring (n=3). The growth of thymine-dystrophic EcN strains (EcNΔthyA and EcNΔnlpIΔthyA) in M9 basal medium was assessed in the presence (labeled "+thymine") or absence of exogenous thymine. Furthermore, the growth of thymine-dystrophic EcN strains containing complementary plasmids (EcNΔthyA-SUT and EcNΔnlpIΔthyA-SU) in the absence of exogenous thymine was assessed in M9 basal medium. Data are mean ± SD. b. Plasmid retention of EcN strains was assessed in antibiotic-free M9 basal medium, and cell viability was calculated by CFU count on LB agar plates containing kanamycin (n=3). Plasmid retention rates in non-thymine-dystrophic EcN strains (EcN-UT2 and EcNΔnlpI-OU) and thymine-dystrophic EcN strains (EcNΔthyA-SUT and EcNΔnlpIΔthyA-SU) with exogenous thymine (labeled "+thymine") decreased significantly within 14 days. Complementary plasmids in thymine-dystrophic EcN strains (EcNΔthyA-SUT and EcNΔnlpIΔthyA-SU) remained stable in the absence of exogenous thymine. Data are mean ± SD.

[0050] Figure 8 Exogenous T1SS-modified EcNs can effectively secrete Uox into the surrounding environment and degrade UA. a. The membrane proteins HlyB and HlyD of *E. coli* J96 were added as heterologous T1SS to engineered EcNs to achieve direct secretion of the protein load. EcN-U1 and EcN-U2 contain synthetic Uox-secreting T1SS under the control of a medium-strength promoter (fnrs promoter) (SEQ ID No. 26) and a low-strength promoter (nar promoter) (SEQ ID No. 27), respectively. The degradation rates of UA in the culture medium by wild-type EcN, recombinant EcN-ut1, and EcN-ut2 were compared (n=3). Data are mean ± SD. b. Western blot analysis was performed on the abundance of Uox in the supernatant of engineered EcNs loaded with T1SS using an anti-His antibody.

[0051] Figure 9Quantification of Uox secreted by T0SS or T1SS. a. Coomassie-stained SDS-PAGE gel of purified His-Uox-HlyAsp fusion protein by His-tag-based affinity purification. To obtain sufficient levels of secreted Uox by T1SS, Uox was overexpressed in EcN via the HlyA signal peptide (abbreviated as "sp") and the His tag (His-Uox-HlyAsp) under the control of the IPTG-induced promoter tac promoter (SEQ ID No. 30). Channel 1: Cell extract without IPTG addition; Channel 2: IPTG-induced His-Uox-HlyAsp expression; Channel 3: His-Uox-HlyAsp purified after his-tag-based affinity purification. Channel M indicates the standard protein label. b. Quantification of the amount of Uox encapsulated in OMV (abbreviated as "OMV-Uox") and purified protein His-Uox-HlyAsp (abbreviated as "T1SS-Uox") by Western blotting.

[0052] Figure 10 OMV from engineered EcN can penetrate the intact intestinal barrier and enter circulation. A. Representative images of colon sections after injection of Cy5.5-labeled OMV (abbreviated as "Cy5.5-OMV") or PBS (abbreviated as control) into the colonic lumen of healthy mice. Nuclei are reverse stained with DAPI (blue). Scale bar = 50 μm. B. Representative images of different tissues (colon, liver, and kidney) from healthy mice after oral administration of EcNΔnlpIΔthyA-SK or PBS. OMV was detected using anti-HA antibody (pink). Nuclei are reverse stained with DAPI (blue). Scale bar = 50 μm.

[0053] Figure 11 Representative images of liver and kidney sections from healthy mice after intracolonic injection of Cy5.5-labeled OMV (Cy5.5-OMV) or PBS (control). Cell nuclei were reverse-stained with DAPI (blue). Scale bar = 50 μm.

[0054] Figure 12 Representative images of different tissues (colon, liver, and kidney) from healthy mice orally administered EcNΔnlpIΔthyA-SU or PBS. OMV was detected using anti-His antibody (pink). Cell nuclei were reverse stained with DAPI (blue). Scale bar = 50 μm.

[0055] Figure 13Proteomic profiles of OMVs from wild-type or engineered EcN. OMVs were isolated from wild-type and engineered EcN and analyzed by SDS-PAGE. Channel 1: OMV from EcN; Channel 2: OMV from EcNΔnlp; Channel 3: OMV from EcNΔnlpIΔthyA; Channel 4: OMV from EcNΔnlpIΔthyA-OG. Channel M indicates a standard protein label.

[0056] Figure 14OMV derived from engineered EcN can penetrate the intestinal epithelial barrier via pinocytosis and dynamically mediated endocytosis. a. The mechanism by which EcN OMV penetrates the intestinal epithelial cell barrier was investigated using an in vitro intestinal epithelial model of a Caco-2 monolayer grown in a Transwell culture system. Various inhibitors targeting different endocytic pathways and RFP-loaded OMV were sequentially added to the upper chamber. b. To determine whether epithelial transcytosis of OMV occurred via an active or passive mechanism, RFP-loaded OMV was incubated for 4 hours in the upper chamber of a Transwell chamber at 37°C or 4°C. Fluorescent signals were detected in the lower chamber and further normalized to the 37°C group (n=3). c. By focusing on the first endocytic step of transcellular poisoning, the mechanism of OMV transepithelial endocytosis was examined using various chemical inhibitors at different concentrations (low, medium, and high) for different endocytic pathways, including a dimethyl sulfoxide (DMSO) group as a negative control. Inhibitors used included chlorpromazine (inhibiting clathrin-dependent endocytosis), cytochalasin D (inhibiting membrane fusion), wortmannin (inhibiting phagocytosis), amiloride (inhibiting pinocytosis), or nystatin (inhibiting caveolin-mediated endocytosis and lipid raft formation). Low concentrations: 0.05 μM wortmannin; 0.25 μM cytochalasin D; 10 μM chlorpromazine; 20 μM amiloride; 0.1 μM nystatin and 20 μM dynasore; Medium concentrations: 0.1 μM wortmannin; 0.5 μM cytochalasin D; 20 μM chlorpromazine; 100 μM amiloride; 0.2 μM nystatin and 100 μM dynasore; High concentrations: 1 μM wortmannin; 2 μM cytochalasin D; 50 μM chloropromazine; 200 μM amiloride; 1 μM nystatin and 200 μM dynasore. Fluorescence signals detected in each lower chamber were normalized using the DMSO group data as a standard (n=3). d. Permeability of epithelial monolayers after incubation with OMV and DMSO or inhibitors was determined using Lucifer Yellow (n=3). Data are mean ± SD. P-values ​​were determined by one-way ANOVA using the Tukey multiple comparison test (* indicates P<0.05).

[0057] Figure 15The in vivo safety of the modified TOSS engineered EcN was evaluated by hematological and blood biochemical analyses. Healthy mice were administered saline (labeled "Control"), EcN (labeled "EcN"), or EcNΔnlpIΔthyA-SU (labeled "EcNΔnlpIΔthyA-SU") every two days for one week. White blood cell (WBC) count (a), red blood cell (RBC) count (b), platelet (PLT) count (c), hemoglobin (HGB) level (d), serum alanine aminotransferase (ALT) level (e), creatinine (CRE) level (f), and blood urea nitrogen (BUN) level (g) were measured. Data are presented as mean ± SEM, n = 3.

[0058] Figure 16 The in vivo safety of the modified TOSS engineered EcN was evaluated through histopathological analysis of major organs. Normal saline (represented as "control group"), EcN, or EcNΔnlpIΔthyA-SU was administered every two days for one week. Representative histopathological images of heart, liver, spleen, lung, and kidney sections from healthy mice were obtained. Sections were stained with hematoxylin and eosin (H&E). Scale bar = 100 mm.

[0059] Figure 17Engineered EcN equipped with improved TOSS demonstrated superior therapeutic effects in treating hyperuricemic mice and detoxifying human serum samples. Experimental design and treatment procedures for UA and PO-induced hyperuricemic mouse models. Mice fed a normal diet and intraperitoneally injected with excipients were denoted as “Con”. Hyperuricemic mice were induced by dietary supplementation with UA (2% UA in the diet) and daily intraperitoneal injection of 250 mg / kg PO for 4 weeks. A. At the start of hyperuricemia induction, hyperuricemic mice were administered physiological saline (abbreviated as “UA”), EcN (abbreviated as “UA+E”), EcNΔthyA-SUT (abbreviated as “UA+T”), or EcNΔnlpIΔthyA-SU (abbreviated as “UA+O”) via tube feeding (n=6). B to E: Serum UA (B), urinary UA (C), serum CRE (D), and IL-1β (E) levels. (F) Representative histopathological images of colon slices obtained from the five groups of mice. Sections were stained with heme and eosin (H&E). Black arrows indicate infiltration of immune cells, and red arrows indicate exfoliated epithelial cells. Scale bar = 50 μm. G and H: Effect of Uox-loaded OMV extracted from EcNΔnlpIΔthyA-SU on UA(G) and glucose(H) degradation in serum samples from patients with hyperuricemia in response to OMV treatment for 30 minutes (n=9). I and J: Lox degradation effect of Lox-loaded OMV extracted from EcNΔnlpI-OL (n=7) in serum samples from patients with lung cancer (n=7) after 30 minutes of OMV treatment. Data are presented as mean ± SD. P-values ​​were determined by one-way ANOVA with Tukey's multiple comparison test (BE) or paired t-test (GJ) (* indicates P<0.05, ** indicates P<0.01 and *** indicates P<0.001). Detailed Implementation

[0060] Unless otherwise indicated, the terms used herein have their general technical meanings as understood by those skilled in the art. For definitions and terms in this field, those skilled in the art are particularly recommended to refer to Sambrook et al., Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor, Plainsview, New York (1989); and Ausubel et al., Current Protocols in Molecular Biology (Supplement 47), John Wiley & Sons, New York (1999).

[0061] This invention provides a recombinant Escherichia coli such as Nissle 1917 (EcN) containing at least one (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10) therapeutic active ingredient in its outer membrane vesicles (OMV). When multiple therapeutic active ingredients are included, these active ingredients may be of the same type or different types.

[0062] In this invention, the term "therapeutic active ingredient" refers to any substance that has pharmacological activity or effect and can be used to treat diseases. Therapeutic active ingredients may include therapeutic peptides, therapeutic proteins, antigens, therapeutic nucleic acids, and small molecule compounds. Therapeutic peptides, polypeptides, proteins, and antigens that can be loaded into the OMV of Gram-negative bacteria have been described and enumerated in CN108779482A. Therapeutic nucleic acids (such as siRNA) that can be loaded into the OMV of Escherichia coli have been described in CN114958891B. CN 118497094A describes small molecule compounds such as radionuclides that can be delivered via bacterial vesicles.

[0063] In a preferred embodiment of the invention, the therapeutic active ingredient is a therapeutic enzyme, particularly a therapeutic enzyme for metabolic diseases, genetic diseases, cardiovascular diseases, gastrointestinal diseases, inflammatory diseases, or cancer.

[0064] In this invention, the term "therapeutic enzyme" refers to a biocatalytic protein used to treat diseases and improve medical conditions. Enzymes, as drugs, have two prominent characteristics: (1) enzymes have high affinity and specificity for their substrates; and (2) enzymes can convert substrates into desired products with minimal side effects. These two characteristics distinguish enzymes from all other types of drugs and make them valuable therapeutic tools. Therapeutic enzymes can be widely derived from animals, plants, and microorganisms, or obtained through recombinant DNA technology. Therapeutic enzymes have been widely used in the treatment of genetic diseases, cardiovascular diseases, gastrointestinal diseases, cancer, and other diseases (Shahriari M, Zahiri M, Abnous K, et al. Enzyme responsive drug delivery systems in cancer treatment[J]. Journal of Controlled Release, 2019, 308: 172-189; Gurung N, Ray S, Bose S, et al. A broader view: microbial enzymes and their relevance in industries, medicine, and beyond[J]. BioMed Research International, 2013, 2013: 329121. DOI: 10.1155 / 2013 / 329121; Baldo BA. Enzymes approved for human therapy: indications, mechanisms and adverse effects). effects[J].BioDrugs,2015,29(1):31-55.DOI:10.1007 / s40259-015-0116-7;Kurosawa Y,Nirengi S,Homma T,et al.Asingle-dose of oral nattokinase potentiates thrombolysis and anti-coagulationprofiles[J].Scientific Reports,2015,5:11601.DOI:10.1038 / srep11601).

[0065] The application of therapeutic enzymes in cancer involves two main types: enzymes that metabolize amino acids required by tumors and enzymes that convert prodrugs into drugs. Metabolic enzymes are used to consume the amino acids necessary for tumor cells, thereby inhibiting tumor growth; converting enzymes are used to convert prodrugs into cytotoxic drugs within tumor cells, thereby killing the tumor cells.

[0066] For example, L-asparaginase hydrolyzes L-asparaginase into L-aspartic acid, thereby preventing L-asparaginase from nourishing tumor cells. L-asparaginase has shown good efficacy in treating acute lymphoblastic leukemia, lymphosarcoma cell leukemia, and granulocytic leukemia in clinical practice, and also has a certain effect on melanoma (Meghavarnam AK, Salah M, Sreepriya M, et al. Growth inhibitory and proapoptotic effects of L-asparaginase from Fusarium culmorum ASP-87 on human leukemia cells (Jurkat)[J]. Fundamental & Clinical Pharmacology, 2017, 31(3):292-300).

[0067] Arginine deiminase (ADI) hydrolyzes L-arginine into L-citrulline, thereby preventing L-arginine from nourishing tumor cells. Under normal circumstances, arginine in the body can be synthesized by the cell's own urea cycle enzymes, namely arginine deiminase synthase and arginine deiminase lyase (Somani RR, Chaskar PK. Arginine deiminase enzyme evolving as a potential antitumor agent[J]. Mini-Reviews in Medicinal Chemistry,2018,18(4):363-368.DOI:10.2174 / 1389557516666160817102701). However, some malignant tumors with metabolic defects, such as melanoma, lung cancer, prostate cancer and hepatocellular carcinoma, often lack these enzymes (Xiong LF, Teng JLL, Botelho MG, et al. Arginine metabolism in bacterial pathogenesis and cancer therapy[J]. International Journal of Molecular Sciences,2016,17(3):363.DOI:10.3390 / ijms17030363;Wu LN, Li L,Meng SC,et al.Expression of argininosuccinatesynthetase in patients with hepatocellular carcinoma[J].Journal ofGastroenterology and Hepatology,2013,28(2):365-368;Ott PA,Carvajal RD,Pandit-Taskar N,et al.PhaseⅠ / Ⅱstudy of pegylated arginine deiminase(ADI-PEG 20)inpatients with advanced melanoma[J].Investigational New Drugs, 2013, 31(2):425-434.Since the growth of these tumor cells depends on arginine in the environment, arginine deiminase can be used to treat these arginine auxotrophic tumors. There are also related reports that ADI-PEG-20 is used as an adjunct to radiotherapy for pancreatic cancer, and can also be used to treat glioblastoma multiforme and thoracic spine cancer (mesothelioma and non-small cell lung cancer) (Beddowes E, Spicer J, Chan PY, et al. Phase 1 dose-escalation study of pegylated arginine deiminase, cisplatin, and pemetrexed in patients with argininosuccinate synthetase 1-deficient thoracic cancers[J]. Journal of Clinical Oncology, 2017, 35(16): 1778-1785; Singh PK, Deorukhkar AA, Venkatesulu BP, et al. Exploiting arginine auxotrophy with pegylated arginine deiminase (ADI-PEG20) to sensitize pancreatic cancer to radiotherapy via metabolic dysregulation[J].MolecularCancer Therapeutics,2019,18(12):2381-2393.DOI:10.1158 / 1535-7163.MCT-18-0708; Han RZ,Xu GC,Dong JJ,et al.Arginine deiminase:recent advances in discovery,crystal structure,and protein engineering for improved properties as an anti-tumor drug[J].Applied Microbiology and Biotechnology,2016,100(11):4747-4760.DOI:10.1007 / s00253-016-7490-z; Maletzki C,Rosche Y,Riess C,etal.Deciphering molecular mechanisms of arginine deiminase-based therapy-comparative response analysis in paired human primary and recurrentglioblastomas[J].Chemico-Biological Interactions,2017,278:179-188;SzlosarekPW,Steele JP,Nolan L,et al.Arginine deprivation with pegylated argininedeiminase in patients with argininosuccinate synthetase 1-deficient malignantpleural mesothelioma:a randomized clinical trial[J].JAMAOncol,2017,3(1):58-66;Covini D,Tardito S,Bussolati O,et al.Expanding targets for a metabolictherapy of cancer:L-asparaginase[J].Recent Patents on Anti-Cancer DrugDiscovery,2012,7(1):4-13.DOI:10.2174 / 157489212798358001)。.

[0068] Glutaminase hydrolyzes L-glutamine into L-glutamate, thereby preventing L-glutamine from nourishing tumor cells. Glutaminase isolated from Pseudomonas 7A, when used in combination with asparaginase, can treat asparaginase-resistant lymphomas and also has some efficacy against various leukemias (Li Q, Wang YT, Wu WT. Research on the application of enzymes as therapeutic drugs [A] / / China Bio-Pharmaceutical Innovation Research Forum [C]. Deyang: Chinese Pharmaceutical Association, 2012: 51-62).

[0069] Glucose oxidase oxidizes glucose into gluconic acid and H2O2, thereby killing tumor cells (Fu LH, Qi C, Lin J, et al. Catalytic chemistry of glucose oxidase in cancer diagnosis and treatment[J]. Chemical Society Reviews, 2018, 47(17):6454-6472.DOI:10.1039 / c7cs00891k).

[0070] Kynurenase hydrolyzes kynurenine into L-alanine and anthranilic acid, thereby stimulating the human immune system to attack tumor cells. Studies have shown that in vivo PEG-kynurenase treatment can clear kynurenine, and when used in combination with immune checkpoint inhibitors and tumor vaccines, it has shown good therapeutic effects on melanoma, breast cancer and colon cancer (Triplett TA, Garrison KC, Marshall N, et al. Reversal of indoleamine 2,3-dioxygenase-mediated cancer immune suppression by systemic kynurenine depletion with a therapeutic enzyme[J]. Nature Biotechnology, 2018, 36(8):758-764).

[0071] Carboxypeptidase A hydrolyzes the methotrexate-phenylalanine prodrug into methotrexate and L-phenylalanine, restoring the cytotoxicity of methotrexate. Modified carboxypeptidase A can activate the activity of the prodrug methotrexate-phenylalanine and be used for the treatment of colorectal cancer (Deckert PM, Bornmann WG, Ritter G, et al. Specific tumor localization of ahuA33 antibody-carboxypeptidase A conjugate and activation of methotrexate-phenylalanine[J]. International Journal of Oncology, 2004, 24(5):1289-1295.DOI:10.3892 / ijo.24.5.1289). Human β-glucuronidase is also an attractive enzyme that can activate the prodrug glucuronide (Chen KC, Wu CH, Chang CY, et al. Directed evolution of alysosomalenzyme with enhanced activity at neutral pH by mammalian cell-surface display[J]. Chemistry & Biology, 2008, 15(12): 1277-1286. DOI: 10.1016 / j.chembiol.2008.10.008).

[0072] The therapeutic enzyme can also be used to treat metabolic disorders in the subject's body (circulatory system). The metabolic disorders are selected from abnormalities in glucose metabolism, lipid metabolism, calcium and phosphorus metabolism, bone metabolism, and purine metabolism. Most preferably, the metabolic disorder is hyperuricemia and the therapeutic enzyme is uricase; the metabolic disorder is peroxisome disease and the therapeutic enzyme is catalase; or the metabolic disorder is hyperlactatemia and the therapeutic enzyme is lactate oxidase.

[0073] Therapeutic enzymes can also be used to treat genetic diseases in the subject's body (within the circulatory system).

[0074] Phenyleneketonuria (PKU) is a common genetic disorder characterized by the accumulation of phenylalanine in the body. This is caused by low or absent activity of phenylalanine hydroxylase, which cannot catalyze the conversion of phenylalanine to tyrosine, leading to the accumulation of phenylalanine and its keto acids. Phenylalanine ammonia lyase (PAL) can catalyze the conversion of phenylalanine to non-toxic trans-cinnamic acid and ammonia. Trans-cinnamic acid is further metabolized and eventually excreted in the urine as hippuric acid. Because the reaction catalyzed by PAL produces non-toxic compounds, PAL is considered a drug for the treatment of PKU (Wang L, Gamez A, Sarkissian CN, et al. Structure-based chemical modification strategy for enzyme replacement treatment of phenylketonuria[J]. Molecular Genetics and Metabolism, 2005, 86(1 / 2):134-140).

[0075] Mucopolysaccharidosis (MPSS) is an autosomal recessive genetic disorder caused by mutations in the α-L-iduronidase gene. α-L-iduronidase and iduronidase sulfate have been approved for the treatment of MPSS. CM, Giugliani R. Evaluation of galsulfase for the treatment of mucopolysaccharidosis VI (Maroteaux-Lamy syndrome) [J]. Expert Opinion on Orphan Drugs, 2014, 2(4): 407-417).

[0076] Pompe disease is a rare disorder caused by a defect or dysfunction in the lysosomal acid alpha-glucosidase (GAA) gene. The FDA approved alpha-glucosidase for the treatment of Pompe disease in 2010.

[0077] Gaucher disease (GD) is a condition caused by a deficiency of glucocerebrosidase or glucocerebrosidase, leading to the accumulation of glucocerebrosides in cells. Agarosidase was the first drug used to treat GD.

[0078] Fabry's disease is an X-linked genetic disorder caused by a deficiency of galactosidase A, leading to the accumulation of acylsphingolipid trihexose in the body. Beta-galactosidase (fabrazyme) was approved for the treatment of Fabry's disease in 2003.

[0079] Adenosine deaminase severe combined immunodeficiency (ADA-SCID) is an extremely rare genetic disorder caused by a deficiency of adenosine deaminase (ADA). In 2018, the FDA approved recombinant adenosine deaminase for the treatment of ADA-SCID in children and adults (Lv Xunlei, Lin Kuaile, Guo Linlin, et al. A brief introduction to new drugs approved by the US FDA in 2018 [J]. China Pharmaceutical Industry Journal, 2019, 50(1):1-33).

[0080] Hypophosphatasia (HPP) is a rare, inherited systemic disorder caused by a loss-of-function mutation in the alkaline phosphatase gene. Alkaline phosphatase was approved by the FDA in 2015 as a drug treatment for HPP.

[0081] Lysosomal acid lipase deficiency (LALD) is a rare autosomal recessive inherited lysosomal storage disorder in which a gene mutation results in the absence or reduction of lysosomal acid lipase function. In 2015, the FDA approved human lysosomal acid lipase as a drug treatment for LALD.

[0082] Therapeutic enzymes can also be used to treat cardiovascular diseases in subjects (within the circulatory system). Cardiovascular diseases are a class of diseases comprised of important conditions such as coronary heart disease, peripheral artery disease, cerebrovascular disease, rheumatic heart disease, congenital heart disease, deep vein thrombosis, and pulmonary embolism. One of the causes of these diseases is the formation of thrombi within blood vessels.

[0083] Streptokinase (SK) has been used to treat peripheral artery occlusive disease. Urokinase has been used to treat embolic diseases. Tissue plasminogen activator (t-PA) was approved by the US FDA and approved for the treatment of myocardial infarction in 1987. Snake venom defibrase, which acts directly on fibrin or fibrinogen, has good efficacy in the treatment of cerebral thrombosis and myocardial infarction. Hirudin is an anticoagulant and antithrombotic agent that acts on thrombin. It is a specific inhibitor of thrombin and can effectively inhibit the formation of thrombi (Li Qian, Wang Youtong, Wu Wutong, Research on the Application of Enzymes as Therapeutic Drugs [A] / / Proceedings of the China Biopharmaceutical Innovation Research Forum [C], Deyang: Chinese Pharmaceutical Association, 2012:51-62). Lumbrokinases (LKs) are a group of proteases isolated from different species of earthworms. They can reduce blood viscosity, increase blood oxygen saturation, and improve blood circulation, and are an effective treatment for cerebral infarction (Wang XM, Fan SC, Chen Y, et al. Earthworm protease in anti-thrombosis and anti-fibrosis[J]. Biochimica et BiophysicaActa(BBA)-General Subjects,2019,1863(2):379-383.DOI:10.1016 / j.bbagen.2018.11.006). Natto kinase (NK) is a protease with strong thrombolytic function produced by Bacillus subtilis natto (Duan CR, Feng YF, Zhou H, et al. Optimization offermentation condition of man-made bee-bread by response surface methodology [A] / / Zhang TC, Nakajima M. Advances in Applied Biotechnology [M]. Berlin, Heidelberg: Springer, 2015:353-363 https: / / link.springer.com / chapter / 10.1007%2F978-3-662-46318-5_38). Desmoteplase is a thrombolytic drug with high fibrin specificity and no neurotoxicity.It was isolated from the saliva of the vampire bat Desmodus rotundus in 1974 and can be used to reduce thrombus formation (Mican J, Toul M, Bednar D, et al. Structural biology and protein engineering of thrombolytics[J]. Computational and Structural Biotechnology Journal, 2019, 17: 917-938. DOI: 10.1016 / j.csbj.2019.06.023).

[0084] Therapeutic enzymes can also be used to treat gastrointestinal diseases in the body (circulatory system). The earliest application of therapeutic enzymes was as digestive aids. These enzymes hydrolyze and digest food components such as proteins, carbohydrates, and lipids, and are commonly used as digestive aids to promote digestion. They are often complex preparations containing proteases, amylases, lipases, and cellulases. Pancreatic enzyme preparations are complex enzyme preparations containing lipases, proteases, and amylases, used to treat malabsorption and insufficient pancreatic juice secretion, especially in patients with gallbladder fibrosis.

[0085] Therapeutic enzymes can also be used to treat inflammatory diseases in the body (circulatory system). Lysozyme can selectively degrade viral RNA and has anti-HIV activity (Lee-Huang S, Huang PL, Sun YT, et al. Lysozyme and RNases as anti-HIV components in β-core preparations of human chorionic gonadotropin[J]. Proceedings of the National Academy of Sciences of the United States of America, 1999, 96(6):2678-2681). Lysostaphin can be used to treat endocarditis, osteomyelitis, pneumonia, toxic shock syndrome, food poisoning and various skin infections (such as folliculitis, abscess and mastitis) caused by Staphylococcus aureus infection. Clinically, trypsin, chymotrypsin, bromelain and other enzymes are often used to treat inflammation and edema to remove necrotic tissue, increase tissue permeability, inhibit edema, promote the drainage of tissue fluid near the lesion and inhibit the formation of granulation tissue. In addition, collagenase can be used to clean burn sores and is effective in removing scabs; chondroitin sulfate can be used to treat spinal injuries and can effectively promote the repair of spinal cord injuries.

[0086] In addition to delivering therapeutic proteases, the recombinant Escherichia coli of the present invention, such as Nissle 1917 (EcN), can also be used for oral delivery of therapeutic nucleic acids, such as small interfering nucleic acids (siRNA), microRNA (miRNA), antisense nucleotide (ASO) messenger RNA (mRNA), or nucleic acid aptamers.

[0087] Oral protein delivery offers patients a convenient and highly compliant route of administration. However, oral delivery efficiency is significantly limited by the physiological barriers of the gastrointestinal tract, including irritating acids and proteases in the stomach, the intestinal mucus layer, and the intestinal epithelial barrier. To overcome these physiological obstacles, NP-based drug delivery systems have made significant progress in improving oral protein delivery efficiency through materials engineering, surface modification, and other methods (Mitchell, MJ et al. Engineering precision nanoparticles for drug delivery. Nat. Rev. Drug Discov. 20, 101-124 (2021); Zhang, F. et al. Lipid-based intelligent vehicle capabilitized with physical and physiological activation. Research (Wash DC) 2022, 9808429 (2022)). Despite these findings, NP-based drug delivery systems still suffer from complex manufacturing processes and inefficient cargo loading and delivery. In our work, we have developed an oral protein delivery technology that can be achieved by orally administering genetically engineered EcNs equipped with modified TOSS. The EcN chassis exhibits tolerance to the digestive tract, and in situ secreted OMV has been found to penetrate the intact intestinal barrier and enter circulation. Therefore, the protein payload is sequentially escorted by the engineered EcN and the secreted OMV carrier for efficient and programmable oral protein delivery. The protein payload is encapsulated in the OMV in vivo using an endogenous protein loading strategy, achieving an encapsulation ratio of 97.9%, significantly higher than the exogenous protein loading methods of EV and NP. Thus, our engineered EcN equipped with a modified TOSS provides a self-programmable and convenient oral protein delivery technology.

[0088] We unexpectedly discovered that unmodified EcN OMVs can cross the intact intestinal barrier into circulation, demonstrating the physiological role of bacterial outer membrane vesicles in host-microbe interactions outside the gut. Based on this important finding, EcNs were engineered to improve OMV synthesis and endogenously load proteases of interest into the OMVs. The OMV carriers possess the advantage of high protein stability, resistance to proteolysis, and the ability to recycle. The TOSS-based protein secretion system is compatible with a wide range of protein payloads and can self-package multiple different protein cargoes within a single OMV carrier for cascade reactions. Furthermore, EcN OMVs can contain endogenous oxidoreductases, including peroxidases, making OMVs ideal biocatalysts for detoxification, including degradation, cascade reactions, and NADH-independent oxidation. Our engineered EcNs for TOSS-based drug secretion demonstrate exceptional performance in detoxifying blood metabolites, thanks to the high stability of OMV-encapsulated proteins and their greater delivery range, providing a genetically engineered symbiotic-based oral protein delivery system for the clinical treatment of various metabolic disorders.

[0089] The invention is further illustrated in the following examples. These examples are for illustrative purposes only and are not intended to limit the scope of the invention. All chemicals used in the following reactions are commercially available products unless otherwise specified.

[0090] Materials and methods

[0091] Material

[0092] DAPI (BL739B) was obtained from Bioharp Life Science. Cell culture plates, 96-well plates, kanamycin (GC301008-5g), IPTG (GC205011-1g), chloramphenicol (GC301018-5g), penicillin-streptomycin (G4003-100ML), DMEM (G4525-500mL), anti-HA-tag antibody (GB151252-100, for Western blotting), anti-GFP antibody (GB15603-100), anti-RFP antibody (GB125053-100), and L-phenylalanine (GC304010-10g) were purchased from ServicebioBIOTechnology. Cy5.5-NHS ester (A8103) was purchased from APExBIOTechnology. Anti-HA tag antibody (26183, for immunofluorescence), anti-His tag antibody (MA121315), goat anti-IgG (H+L) secondary antibody, combined with Alexa Fluor 647 (A32728TR), and fetal bovine serum (FBS; A5669401) were purchased from Fisher Thermo Scientific. Mouse IL-1β ELISA kit (KE10003), goat anti-mouse IgG (H&L)-HRP (PR30012) and goat anti-rabbit IgG (H&L)-HRP (PR30011) for Western blotting were purchased from Protentech. EVMembrane Red Stain (NEPU-638-50T) was purchased from NanoFCM. Sodium carboxymethyl cellulose (IS9000), uric acid assay kit (BC1365), creatinine assay kit (BC4915), and blood urea nitrogen assay kit (BC1535) were obtained from Solarbio Science & Technology. Cytochalasin D (GC13440), chlorpromazine HCl (GC14216), dynasore (GC10395), nystatin (GC10090), amiloride HCl (GC17853), and wortmannin (GC12338) were purchased from Glpbio Technology. Uric acid (U0018) and potassium oxalate (O0164) were purchased from Tokyo Chemical Industry Co., Ltd. Sodium lactate (S108838) was purchased from Aladdin Scientific Co., Ltd.

[0093] Animals and cells

[0094] Male Kunming mice (4-5 weeks old) were obtained from Vital Rivers (China) in Beijing. All animal experiments were approved by the Institutional Animal Use and Animal Experimentation Ethics Committee of Tsinghua University. Mice were housed under specific pathogen-free (SPF) conditions with a 12-hour light / dark cycle, free access to water and food. Caco-2 cells (ATCC HTB-37) were cultured in DMEM at 37°C, supplemented with 20% FBS, 100 U / ml penicillin, and 100 μg / ml streptomycin, with 5% CO2.

[0095] Bacterial strains and culture conditions

[0096] Table 1 provides a list of bacterial strains used in this invention. *Escherichia coli* S17-1 strain (Simon R, Priefer U, Puhler A. 1983. A broad host range mobilization system for in vivo genetic engineering-transposon mutagenesis in Gram-negative bacteria. Bio-Technol. 1:784-791) was used for plasmid construction, and *Escherichia coli* WM3064 strain (Saltikov CW, Newman DK. 2003. Genetic identification of a respiratory arsenate reductase. Proc. Natl. Acad. Sci. USA 100:10983-10988) served as the donor strain for conjugation. The gene encoding catalase (SEQ ID No. 6) (katE gene) was amplified from the genomic DNA of *Escherichia coli* Nissle 1917 (GenBank CP007799.1). Heterologous genes, namely the uricase (SEQ ID No. 5) encoding gene cu_uox from C. utilis (UniProt Entry P78609) or the uricase oxidase (SEQ ID No. 31) encoding gene sa_uox from S. auratus AGR0001, the hlyBD-hlyA (hemolysin gene) from Escherichia coli J96 (Tzschaschel, BD, Guzmán, CA, Timmis, KN ​​& de Lorenzo, V. An Escherichia coli hemolysin transport system-based vector for the export of polypeptides: export of Shiga-like toxin IIeB subunit by Salmonella typhimurium aroA. Nat. Biotechnol. 14, 765-769 (1996)), and the phenylalanine deaminase (SEQ ID No. 5) from P. luminescens TT0147. The gene encoding stlA (SEQ ID No. 8) or ye_pal1 (SEQ ID No. 32) of phenylalanine deaminase from Y. enterocolitica, and the gene encoding ye_pal2 (SEQ ID No. 33) of phenylalanine deaminase from Y. enterocolitica, are derived from A.The lactate oxidase (SEQ ID No. 7) encoding gene lox from *Viridans* (UniProt Entry Q44467), the aldehyde oxidase (SEQ ID No. 34) encoding gene pb_alod from *P. blakesleeanus* NRRL 1555, the aldehyde oxidase (SEQ ID No. 35) encoding gene cr_alod from *C. reinhardtii*, the aldehyde oxidase (SEQ ID No. 36) encoding gene cv_alod from *C. cardunculus* var. *scolymus*, the aldehyde oxidase (SEQ ID No. 37) encoding gene ls_alod from *L. sativa*, and the aldehyde oxidase (SEQ ID No. 38) encoding gene sp_alod from *S. punctatus* DAOM BR117 were synthesized and cloned by GenScript (Nanjing, China) and inserted into the pYYD plasmid (Yang, Y. et al. Enhancing bidirectional electron transfer of *Shewanella oneidensis* by a synthetic flavin pathway). Synth. Biol. 4, 815-823 (2015)). Subsequently, the plasmid was transformed into *E. coli* strain WM3064 and then conjugated with EcN.

[0097] The suicide plasmid pRE112 (Edwards, RA, Keller, LH & Schifferli, DMI improved allelic exchange vectors and their use to analyze 987P fimbria gene expression. Gene 207, 149-157 (1998)) was used to knock out the nlpI (SEQ ID No. 1) and thyA (SEQ ID No. 9) genes in EcN. The plasmid was constructed using Gibson assembly from E. coli S17-1, subsequently transformed into E. coli WM3064, and then conjugated to EcN. Single exchange mutants were selected on LB agar supplemented with chloramphenicol and validated by colony PCR. The resulting double exchange mutants were selected on LB agar supplemented with 10% sucrose and 3 mM thymidine (for the ΔthyA mutant) and further confirmed by colony PCR and sequencing. EcN and *E. coli* S17-1 were cultured in LB medium (10 g peptone, 10 g sodium chloride, 5 g yeast extract per liter) at 37°C. *E. coli* WM3064 was cultured in LB medium supplemented with 50 μg / mL 2,6-dicarboxylic acid at 37°C. Antibiotic selection involved the use of 50 μg / mL kanamycin (for pYYD-derived plasmids) and 25 μg / mL chloramphenicol (for pRE112-derived plasmids). The strains were cryopreserved in 15% glycerol at -80°C. The PCR primers used are shown in Table 2.

[0098] Table 1. Strain Genotypes

[0099]

[0100]

[0101] Note: mbp in genotype sp This indicates the Sec signal peptide (SEQ ID No. 2); torA sp This indicates Tat signal peptide (SEQ ID No. 3); dsbA sp This refers to the Srp signal peptide (SEQ ID No. 4); P nar P fnrs P nar618 and P J23117 These represent the nar promoter (SEQ ID No. 27), fnrs promoter (SEQ ID No. 26), nar618 promoter (SEQ ID No. 28), and J23117 promoter (SEQ ID No. 29), respectively.

[0102] Table 2. PCR primer sequences

[0103]

[0104] In vitro uric acid (UA) consumption test

[0105] To assess the bacteria's ability to degrade UA, logarithmically grown bacteria in LB medium were resuspended in modified M9 basal medium (6.78 g Na₂HPO₄, 3 g KH₂PO₄, 1 g NH₄Cl, 0.5 g NaCl, 5 g glucose, 1 g yeast extract per liter) and supplemented with 0.5 mM UA for micro-aerobic culture, with a final concentration of 10. 9 Cells / mL. Sampling was performed at different time intervals, and UA concentration was quantified using a uric acid content assay kit.

[0106] Isolation and characterization of outer membrane vesicles

[0107] To isolate and purify OMV, the supernatant of the *E. coli* culture was centrifuged at 4 °C and then filtered through a 0.22 μm syringe to remove large contaminants and bacteria. The resulting solution was concentrated using a 100 kDa cutoff centrifuge filter and further ultracentrifuged at 120,000 × g for 2 h at 4 °C. OMV particles were washed with PBS, ultracentrifuged again, filtered through a 0.22 μm syringe, and resuspended in PBS. The morphology of OMV was characterized using a Hitachi transmission electron microscope (TEM) system. Zeta potentials were measured using a Zetasizer Nano ZS (Malvern, UK). The hydrodynamic diameter of OMV was measured using a ZetaVIEW (Particle Metrix, Germany).

[0108] Nanoflow cell analyzer

[0109] The separated OMV samples were analyzed using a Flow NanoAnalyzer (nanoFCM) instrument equipped with 488 nm and 638 nm lasers. OMV extracted from EcNΔnlpI and EcNΔnlpI-OG were labeled with EVMembraneRed Stain for 30 min at room temperature. Using the separation method described above, the fluorescently labeled samples were washed twice with PBS to remove excess dye. The particles were then resuspended in PBS and measured using a calibrated instrument with 50 mW laser power, 10% SS attenuation, and a 1 kPa sampling pressure. Data analysis was performed using NF Professional 2.0 software. Threshold lines for green and red fluorescence were determined by analyzing the green and red fluorescence signals of the PBS control and empty OMV to reduce the count of nanoscale impurities and background signal (<5%).

[0110] Measurement of metabolites

[0111] UA, creatinine, and urea nitrogen were quantified using urea acid, creatinine, and urea nitrogen content assay kits, respectively. Other metabolites (such as phenylalanine and lactate) in the culture medium or serum were measured by high-performance liquid chromatography (HPLC). Phenylalanine was quantified using an HPLC system equipped with a UV detector and a ZORBAX SB-C18 column. Gradient elution was performed using a mobile phase consisting of acetonitrile and 1.5% acetic acid. Detection was performed at 260 nm with the column temperature maintained at 40 °C. Lactate quantification was performed using an HPLC system equipped with a RID-10A refractive index detector and an Aminex HPX-87H column. The sample was eluted with 5 mM sulfuric acid at 50 °C. Metabolite identification and quantification were accomplished by comparing retention times and peak areas with established standards.

[0112] animal experiments

[0113] Hyperuricemia induction in Kunming mice: Mice were fed 2% UA in their diet and intraperitoneally injected daily with 250 mg / kg potassium oxonate (PO) (dissolved in 0.5% CMC-Na solution) for 4 weeks. At the start of hyperuricemia induction, throughout the 28-day induction period, the four groups of mice were administered EcN (5 × 10⁻⁶) every two days. 9 CFU), EcNΔthyA-SUT (5×10 9 CFU), EcNΔnlpIΔthyA-SU (5×10 9 Mice were administered CFU (carbohydrate, natriuretic peptide, and saline) or an equivalent volume of saline via gavage. On the last day of the study, the mice were euthanized with carbon dioxide. Blood samples were collected via cardiac biopsy, and liver, kidney, and colon samples were collected for histological analysis.

[0114] In vivo safety assessment of engineered EcN equipped with improved TOSS: Three groups of healthy male Kunming mice were administered EcN (5 × 10⁻⁶) to each group. 9 CFU), EcNΔnlpIΔthyA-SU (5×10 9 Mice were administered CFU (carbohydrate, natriuretic peptide, and saline) or an equivalent volume of saline by gavage for one week. These mice were euthanized on the last day of the study. Blood samples were collected via cardiac biopsy for routine blood tests and serum biochemistry assays, and major organs (heart, liver, spleen, lungs, and kidneys) were collected for histological analysis.

[0115] Western blot and immunofluorescence

[0116] Total protein content in the samples was quantified using Bradford assay. Purified OMV or protein was heated at 95°C for 5 minutes, followed by SDS-PAGE and then transferred to a polyvinyl fluoride (PVDF) membrane. After transfer, the membrane was coated with 5% bovine serum albumin (BSA) for 1 hour and incubated overnight at 4°C with gentle shaking and specific primary antibody. Subsequently, the membrane was washed three times with TBST for 10 minutes each time and incubated for another hour with goat anti-mouse IgG (H&L)-HRP or goat anti-rabbit IgG (H&L)-HRP, and the bands were visualized by enhanced chemiluminescence (ECL).

[0117] For immunofluorescence, frozen colon, liver, or kidney tissue embedded in 7-micron OCT cells was placed on a slide. After incubation for 1 hour in 0.5% Triton X-100, it was incubated for 2 hours in 10% sheep serum and then overnight at 4°C with anti-HIS antibody. It was then incubated with an appropriate secondary antibody at room temperature for 2 hours. Images were acquired using an Andor DragonFly confocal imaging system.

[0118] In vitro endocytic transport studies

[0119] Caco-2 cells were seeded in the upper chamber of a Transwell cell line and cultured for 3 weeks. The integrity of the epithelial monolayer was assessed using transepithelial electrical resistance (TEER) measurement with an EVOM TEER meter; only cells with a TEER value greater than 300 Ω × cm⁻¹ were considered intact. 2 The model is used for transcellular transport studies. To study OMV transport using this model, OMV encapsulated with RFP was placed in the upper chamber of a Transwell (~10). 10 (Particles). The effect of temperature was examined by incubating the model at 4 or 37°C. To assess the mechanism of endocytosis transport, intestinal epithelium was incubated at 37°C for 1 hour with a commercially available inhibitor before OMV addition. After 8 hours, the medium was collected from the lower side of the chamber, and the fluorescence signal was quantified using a microplate reader. To assess the effect of OMV on epithelial barrier integrity, Lucifer Yellow (LY) was added to the Transwell insert, and the fluorescence intensity of the apical and basal side liquids was measured using a microplate reader after 1 hour. The percentage of LY rejection was calculated using the following formula: %LY rejection percentage = 100 x (1 - RFU) basolateral / RFU apical ).

[0120] Detection of metabolites in clinical serum samples

[0121] To assess the efficacy of OMVs in degrading metabolites in human blood, 100 μg of purified OMV was incubated with 50 μL of human serum samples for 30 minutes. The concentrations of UA, glucose, or lactate in the samples were then determined as described above. The acquisition and processing of patient serum samples were approved by the Clinical Research Ethics Committee of the China-Japan Friendship Hospital, approval number 2023-KY-007. This study was conducted in accordance with the ethical standards outlined in the 2013 Declaration of Helsinki.

[0122] Statistical data

[0123] Data are expressed as mean ± SD or mean ± SEM. Statistical differences between the control and experimental groups were analyzed using one-way or two-way ANOVA via Tukey's post-hoc test. P < 0.05 was considered significant (*P < 0.05; **P < 0.01; ***P < 0.001). Paired samples were compared using paired t-tests (*P < 0.05; **P < 0.01; ***P < 0.001).

[0124] Example 1

[0125] Developing a modified TOSS for protein secretion

[0126] To construct a TOSS-based protein secretion system in EcN, specific proteins were encapsulated into OMVs in genetically engineered EcN (i.e., recombinant E. coli Nissle 1917 (EcN)) to enhance OMV synthesis. First, nlpI, encoding an outer membrane lipoprotein involved in regulating peptidoglycan dynamics, was deleted in EcN to increase OMV production (McBroom, AJ, Johnson, AP, Vemulapalli, S. & Kuehn, MJ, U.S. Membrane vesicle production by Escherichia coli independent of membrane instability. J. Bacteriol. 188, 5385-5392 (2006)). EcNΔnlpI showed a 2.83 ± 0.24-fold higher OMV production in M9 medium compared to wild-type EcN, exhibiting conserved OMV morphology, zeta membrane potential, and size distribution. Figure 2In Gram-negative bacteria EcN, the natural secretion system transports cytoplasmic proteins carrying signal peptides to the periplasmic space, including general secretions (Sec), diarginine translocation (Tat), and signal recognition granules (Srp)-dependent pathways. The signal peptides are cleaved from secreted proteins by signal peptidases located on the inner membrane (Green, ER & Mecsas, J. Bacterial secretion systems: an overview. Microbiol. Spectr. 4 (2016)). Next, we investigated whether, like other *E. coli* strains (Kesty, NC & Kuehn, MJ, *Incorporation of heterologous outer membrane and periplasmic proteins into *Escherichia coli* outer membrane vesicles. *J. Biol. Chem.* 279, 2069-2076 (2004); Bartolini, E. et al., *Recombinant outer membrane vesicles carrying *Chlamydia muridarum* HtrA induce antibodies that neutralize *chlamydia* linfection in vitro. *J. Extracell. Vesicles 2 (2013)*), it could guide heterologously expressed proteins with signal peptides into the periplasm and further encapsulate them into the OMV of the *EcN*. The recombinant strain *EcNΔnlpI-OG* was obtained by guiding the N-terminus of the fluorescent protein GFP (SEQ ID No. 22) with the Sec signal peptide (SEQ ID No. 2) and overexpressing it in *EcNΔnlpI*. OMVs of EcNΔnlpI-OG were separated by ultracentrifugation, and the presence of GFP in the OMV samples was confirmed by Western blot analysis using anti-GFP antibody and fluorescence images, indicating that heterologous overexpressed proteins labeled with periplasmic targeting signal peptides in engineered EcNs can be successfully packaged into OMVs. Figure 3 a and Figure 4 ).

[0127] Nanoflow cytometry was used to study the protein encapsulation ratio in OMV. The nanoflow cytometry calculated the ratio of OMV encapsulated with GFP to the total OMV from EcNΔnlpI-OG. GFP was used as an example of encapsulated protein, and EVMembraneRed was used to stain the OMV membrane. Figure 3b). Compared to empty OMV from EcNΔnlpI, GFP-loaded OMV from EcNΔnlpI-OG showed a significant increase in GFP fluorescence signal ( Figure 3 (c) The GFP encapsulation ratio of OMV is as high as 97.9%, calculated as the ratio of OMV encapsulating GFP to OMV stained with EVMembrane Red. Traditional exogenous protein loading methods include electroporation, co-incubation, extrusion, freeze-thaw, sonication, and surfactant treatment, with encapsulation efficiencies of approximately 20%-50% (Chen, C. et al. Single-particle assessment of six different drug-loading strategies for incorporating doxorubicin into small extracellular vesicles. Anal. Bioanal. Chem. 415, 1287-1298 (2023)). Compared with exogenous loading strategies, our endogenous loading system outperforms in encapsulation ratio and self-programmed assembly.

[0128] Furthermore, we investigated whether this modified endogenous protein encapsulation system could co-encapsulate multiple different protein payloads into a single OMV. In addition to GFP, RFP (SEQ ID No. 23) fused with the Srp signal peptide was overexpressed in EcNΔnlpI, and the recombinant EcN could encapsulate RFP within the OMV vector. This was confirmed by Western blot analysis of the extracted OMV samples using anti-RFP antibody and fluorescence imaging. Figure 3 d and Figure 4 Then, EcNΔnlpI was designed to co-express two heterologous proteins: GFP fused to the N-terminal Sec signal peptide and RFP fused to the N-terminal Srp signal peptide. The resulting strain, EcNΔnlpI-GR, was isolated and analyzed for OMV. GFP and RFP were detected in the OMV samples by Western blot analysis. Figure 3 (a and d). OMV fluorescence images from EcNΔnlpI-GR show that a single OMV emits both green and red fluorescence signals simultaneously (a and d). Figure 3 (e and f) indicate that a single OMV encapsulates both GFP and RFP payloads. These results demonstrate that a TOSS-based protein secretion system in engineered EcNs can assemble multiple different proteins within a single OMV vector, making engineered OMVs promising cascade biocatalysts.

[0129] Example 2

[0130] The TOSS-based protein secretion system is compatible with various enzymes and can perform a variety of detoxification reactions.

[0131] The compatibility of a range of enzymes with TOSS-based protein secretion systems was tested. The enzymes were expressed by fusion with an N-terminal Tat signal peptide (SEQ ID No. 3) and a protein tag (His or HA) (His: HHHHHH (SEQ ID No. 24); HA: YPYDVPDYA (SEQ ID No. 25)) and overexpressed in EcNΔnlpI. The enzymes included phenylalanine deaminase StlA (SEQ ID No. 8) (from *Photorhabdus luminescens*), phenylalanine deaminase yePAL1 (SEQ ID No. 32) (from *Yersinia enterocolitica*), phenylalanine deaminase yePAL2 (SEQ ID No. 33) (from *Yersinia enterocolitica*), catalase KatE (SEQ ID No. 6) (from EcN), lactate oxidase Lox (SEQ ID No. 7) (from *Aerococcus viridans*), uricase cuUox (SEQ ID No. 5) (from *Candidautilis*), and uricase saUox (SEQ ID No. 1). No. 31 (from *Streptomyces auratus* AGR0001), aldehyde oxidase pbAlod (SEQ ID No. 34) (from *Phycomyces blakesleeanus* NRRL 1555), aldehyde oxidase crAlod (SEQ ID No. 35) (from *Chlamydomonas reinhardtii*), aldehyde oxidase cvAlod (SEQ ID No. 36) (from *Cynara cardunculus* var. *scolymus*), aldehyde oxidase lsAlod (SEQ ID No. 37) (from *Lactuca sativa*), and aldehyde oxidase spAlod (SEQ ID No. 38) (from *Spizellomyces punctatus* DAOM BR117). After isolating OMV samples from engineered EcN strains, their respective anti-tag antibodies were used ( Figure 5The enzymes in the OMVs (A and B) were detected by Western blotting. Compared with the empty OMV containing EcNΔnlpI, the engineered OMVs containing the enzymes exhibited the corresponding enzyme catalytic activities. Specifically, the engineered OMVs containing StlA, yePAL1, and yePAL2 could degrade phenylalanine at rates of 0.38±0.04 μM / h / mg, 0.31±0.02 μM / h / mg, and 0.49±0.02 μM / h / mg, respectively; the OMVs containing KatE could degrade hydrogen peroxide at a rate of 2.92±0.07 μM / h / mg; the OMV containing Lox could degrade lactic acid at a rate of 0.36±0.02 μM / h / mg; and the OMVs containing CuUox and S... aUox's OMV can degrade uric acid (UA) at rates of 0.34±0.03 μM / h / mg and 0.29±0.01 μM / h / mg, respectively. OMVs encapsulating lsAlod, crAlod, spAlod, pbAlod, and cvAlod can degrade acetaldehyde at rates of 15.02±0.93 μg / mL / h / mg, 15.55±1.02 μg / mL / h / mg, 12.51±0.98 μg / mL / h / mg, 13.98±1.13 μg / mL / h / mg, and 17.02±1.07 μg / mL / h / mg, respectively. Figure 5 (C to G). These results demonstrate that engineered EcNs based on engineered TOSS can self-assemble and secrete OMVs encapsulating functional enzymes, and this TOSS-based protein secretion system is compatible with a wide range of proteins.

[0132] Engineered OMV containing various enzymes holds promise for catalyzing a variety of detoxification reactions. Figure 5 Catalytic reactions of OMVs encapsulated with StlA and KatE showed that the synthesized OMVs can catalyze decomposition reactions. Furthermore, empty OMVs from EcNΔnlpI were found to significantly degrade hydrogen peroxide at a rate of 1.61 ± 0.05 μM / h / mg. Proteomic analysis of the OMV samples (provided by Biotech Biotechnology Co., Ltd.) revealed the presence of endogenous peroxidase KatG in EcN OMVs (see Table 3 below). These results confirm that OMVs from EcN contain endogenous catalase, which can degrade hydrogen peroxide into oxygen and water. It should be noted that both Uox and Lox catalyze oxidation reactions and produce hydrogen peroxide (…). Figure 5 (G), while when using OMV loaded with Uox or Lox to catalyze the degradation of UA or lactate, negligible amounts of hydrogen peroxide were detected (data not shown), indicating that the oxidases (Uox and Lox) are coupled with endogenous catalase in the OMV, and the catalysis depends on the H2O2 / O2 cycle rather than NAD. + / NADH oxidation reaction ( Figure 5In summary, these results demonstrate that the OMV synthesized from engineered EcN based on engineered TOSS can catalyze a variety of detoxification reactions, including decomposition and oxidation.

[0133] Table 3. Proteomics identification of enzymes related to mucus degradation function and endogenous catalase in EcN OMV.

[0134]

[0135]

[0136] Example 3

[0137] Proteins encapsulated in OMV exhibit enhanced stability in simulated intestinal fluid.

[0138] Next, using the Uox protein as an example, we evaluated the performance of our TOSS-based secretion system in a simulated intestinal environment by comparing its protein stability and catalytic activity with those of a conventional secretion system. Figure 6 First, to achieve stable expression of heterologous genes in engineered EcN in the intestinal environment, a stable plasmid system was established by deleting the essential growth gene thyA from the EcNΔnlpI genome and then reintroducing thyA using a plasmid. The growth of EcNΔnlpIΔthyA depends on exogenous thymidine ( Figure 7 ,a), while EcNΔnlpIΔthyA carrying a stable plasmid grows well with EcNΔnlpI in the absence of exogenous thymidine ( Figure 7 a) indicates that the stable plasmid was successfully retained in an antibiotic-free microenvironment. Figure 7 (b) The recombinant EcN strain EcNΔnlpIΔthyA-SU overexpressed Uox tagged with Tat signal peptide and His tag on a stable plasmid and was shown to secrete OMV encapsulating Uox. Figure 6b). A genetically engineered EcN strain, EcNΔthyA-SUT, with a heterologous type I secretion system (T1SS) (Tzschaschel, BD, Guzmán, CA, Timmis, KN ​​& de Lorenzo, V. An Escherichia coli hemolysin transport system-based vector for the export of polypeptides: export of Shiga-like toxin IIeB subunit by Salmonella typhimurium aro A. Nat. Biotechnol. 14, 765-769 (1996)), representing a strain with a direct protein secretion strategy, was established. Fusion of Uox and HlyA signal peptide was observed in the cell culture supernatant, and a His tag was also observed. Western blot analysis was performed using an anti-His antibody. Figure 8 (a) and (b) demonstrate that the strain that directly secretes the protein successfully secretes Uox.

[0139] Recombinant EcN strains based on engineered TOSS or T1SS ( Figure 6 (a) EcNΔnlpIΔthyA-SU and EcNΔthyA-SUT, respectively, showed similar UA degradation rates in M9 medium containing 0.5 mM UA, indicating no significant difference in protein secretion efficiency between TOSS and T1SS in engineered EcN strains. However, in simulated intestinal fluid containing trypsin and 0.5 mM UA, EcNΔnlpIΔthyA-SU based on engineered TOSS showed better performance than EcNΔthyA-SUT based on engineered T1SS. Figure 6 The UA degradation efficiency of (c) and (d) was much higher, indicating that Uox secreted by T0SS was more resistant to protein degradation in simulated intestinal fluid than Uox secreted by T1SS. Furthermore, the protein stability and catalytic activity of Uox secreted by T0SS and T1SS were analyzed and compared in simulated intestinal fluid. The OMV of Uox encapsulated in EcNΔnlpIΔthyA-SU (abbreviated as OMV-Uox) and the OMV of Uox fused with the HlyA secretion signal sequence and purified (abbreviated as T1SS-Uox) were extracted and quantified. Figure 9 (a and b). OMV-Uox and T1SS-Uox containing similar amounts of Uox were incubated in PBS or simulated intestinal fluid, and their enzyme activities at different time points were detected. Figure 6 (d and e). It was observed that OMV-Uox and T1SS-Uox exhibited similar enzyme activity and stability in PBS within 2 hours. Figure 6However, the enzyme activity of T1SS-Uox decreased rapidly in simulated intestinal fluid, and after 1 hour it retained 37.39 ± 4.91% of its original enzyme activity. Figure 6 In contrast, the enzyme activity reduction rate of OMV-Uox was significantly lower than that of T1SS-Uox, retaining more than 79.81 ± 2.51% of the original activity after 1 hour of trypsin incubation. Figure 6 These results indicate that the OMV carrier protects protein cargoes from protease degradation, suggesting that protein payloads delivered using an engineered TOSS-based protein secretion strategy are more stable and catalytically active than those delivered via the direct secretion pathway commonly found in the digestive tract.

[0140] Example 4

[0141] EcN OMV, which contains protein, can penetrate the intestinal barrier and enter circulation.

[0142] We investigated whether OMV from EcN could penetrate the intact intestinal barrier. OMV derived from EcNΔnlpIΔthyA was labeled with Cy5.5 and injected into the colonic lumen of healthy mice, followed by colonic ligation. After eight hours of colonic incubation, fluorescent OMV was observed in the basal villi of the colon, as well as in the liver and kidneys of healthy mice incubated with OMV instead of PBS. Figure 10 A and Figure 11 This indicates that EcN-derived OMV can penetrate the intact intestinal epithelial barrier and enter circulation. Furthermore, we explored whether protein-encapsulated OMV generated in situ from engineered EcN strains in the gut could cross the intestinal barrier and enter circulation in healthy mice without intestinal permeability defects. Healthy mice were orally administered EcNΔnlpIΔthyA-SU or EcNΔnlpIΔthyA-OK, and the in vivo distribution of Uox or KatE-encapsulated OMV was tracked at different time points by anti-His or anti-HA immunofluorescence staining (pink). Four hours after gavage administration of EcNΔnlpIΔthyA-SU or EcNΔnlpIΔthyA-OK, Uox or KatE-encapsulated OMV was detected in the mouse colon using anti-His or anti-HA antibodies (pink), respectively. Figure 10 B and Figure 12 OMV was observed in the liver and kidneys, in addition to the colon, at 12 and 24 hours after oral administration of EcNΔnlpIΔthyA-SU or EcNΔnlpIΔthyA-OK. Figure 10 B and Figure 12This indicates that OMV encapsulated in situ by engineered EcN can penetrate the intestinal barrier and enter the circulation of healthy mice. 36 hours after gavage administration of EcNΔnlpIΔthyA-SU or EcNΔnlpIΔthyA-OK, almost no OMV was detected in vivo. Figure 10 B and Figure 12 This indicates that circulating OMV from gut EcN can persist for up to 36 hours after administration of engineered EcN. In contrast, no circulating OMV was detected in healthy mice treated with an excipient carrier (phosphate-buffered saline, PBS). These results collectively suggest that protein-encapsulated OMV produced by engineered EcN in the gut is capable of entering circulation.

[0143] It has been reported that some bacterial OMVs contain degrading enzymes, which may be responsible for accelerating their passage through mucus pores within the mucus layer to intestinal epithelial cells (Schwechheimer, C. & Kuehn, M. J. Uter-membrane vesicles from Gram-negative bacteria: biogenesis and functions. Nat. Rev. Microbiol. 13, 605-619 (2015)). We compared the proteins of wild-type EcN's native OMV and EcNΔnlpIΔthyA-SG's GFP-encapsulated OMV using SDS-PAGE and proteomics analysis of OMV samples. The results showed that the protein distribution of OMVs from wild-type or engineered EcN was similar (…). Figure 13 There were no significant differences in the protein abundance of enzymes (such as proteases, hydrolases, glycosidases, hexosamineases, and sulfatases) between wild-type and engineered EcN OMVs (Table 3). These results indicate that, through the engineered TOSS system of engineered EcNs, heterologous and secreted proteins do not interfere with endogenous proteins in OMVs; therefore, the protein-encapsulated OMVs of engineered EcNs retain their ability to pass through the mucus layer.

[0144] In addition, we used an in vitro intestinal epithelial model of Caco-2 monolayers grown in the Transwell culture system. Figure 14 (a) This study investigated the mechanism behind the ability of protein-encapsulated OMVs to penetrate the intestinal epithelial cell barrier. To determine whether OMVs are transported across the intestinal epithelium via active or passive mechanisms, RFP-encapsulated OMVs were incubated with intestinal epithelium in the upper chamber of the Transwell at 37°C or 4°C for 4 hours. In the lower chamber of the Transwell, the fluorescence signal detected at 37°C was significantly higher than that at 4°C, indicating that the transfer of OMVs across the intestinal epithelial monolayer under physiological conditions involves an active mechanism (a). Figure 14Furthermore, by focusing on endocytosis, the first step in transcellular transport, we explored the mechanism of OMV transepithelial transport using various chemical inhibitors targeting different endocytic pathways (Mulcahy, LA, Pink, RC & Carter, DR Routes and mechanisms of extracellular vesicle uptake. J. Extracell. Vesicles 3 (2014)). The addition of chlorpromazine (inhibiting clathrin-dependent endocytosis), cytochalasin D (inhibiting membrane fusion), woumacil (inhibiting phagocytosis), or nystatin (inhibiting caveolin-mediated endocytosis and lipid raft formation) did not significantly affect the efficiency of RFP-loaded OMV endocytosis in the intestinal epithelium. Figure 14 In contrast, compared with the control group treated with solvent (dimethyl sulfoxide, DMSO), kinin (an inhibitor) and aminolide (an inhibitor of pinocytosis) significantly reduced OMV transcellular transport in a dose-dependent manner. Figure 14 (c) Furthermore, incubation with endocytosis inhibitors and OMV did not alter the permeability of intestinal epithelium to fluorescein. Figure 14 These results indicate that pinocytosis and actin-dependent endocytosis are involved in the transcellular transport of protein-loaded OMVs from engineered EcNs. To our knowledge, this is the first report on the transport of EcN OMVs across the intestinal epithelial barrier.

[0145] Example 5

[0146] In vivo safety assessment of the engineering EcN equipped with modified TOSS

[0147] To test whether oral administration of engineered EcN equipped with modified TOSS is a safe treatment, healthy mice were administered saline, EcN, or EcNΔnlpIΔthyA-SU orally for one week for potential toxicity assessment, followed by histological analysis of major organs and blood and blood biochemical analysis. Compared with mice administered saline, mice treated with EcN or EcNΔnlpIΔthyA-SU showed no significant changes in white blood cell (WBC) count, red blood cell (RBC) count, platelet (PLT) count, or hemoglobin (HGB) level. Figure 15(a to d). These results indicate that neither EcN nor engineered EcN equipped with modified TOSS to produce OMV has adverse effects on the hematopoietic system. Furthermore, biochemical analysis of plasma samples showed that treatment with EcN or EcNΔnlpIΔthyA-SU did not increase levels of alanine aminotransferase (ALT), creatinine (CRE), or blood urea nitrogen (BUN), suggesting that our engineered EcN does not induce liver or kidney toxicity in vivo. Figure 15 (e to g). Furthermore, H&E staining analysis of major organs (including heart, liver, spleen, lungs, and kidneys) showed no signs of organ damage in any group, indicating that taking our engineered EcN does not produce significant histopathological abnormalities. Figure 16 In summary, in vivo safety assessments demonstrate that our engineered EcN equipped with improved T0SS is biocompatible and safe, and has potential for clinical application.

[0148] Example 6

[0149] Engineered EcN oral Uox equipped with improved TOSS showed excellent therapeutic effects in treating hyperuricemia in mice.

[0150] We evaluated the efficacy and advantages of our engineered EcN equipped with a modified TOSS for the treatment of metabolic diseases, compared with engineered EcN equipped with a direct protein secretion device (T1SS), using hyperuricemia as a demonstration. A mouse model of hyperuricemia was induced by UA-supplemented diet and daily intraperitoneal injection of potassium oxyacid (PO) for 28 days, where PO served as a mouse Uox inhibitor (Lu, J. et al. Mouse models for human hyperuricaemia: a critical review. Nat. Rev. Rheumatol. 15, 413-426 (2019)). Healthy mice were used as a control group by gavage with saline and intraperitoneal injection of sodium carboxymethyl cellulose. During the 28-day UA- and PO challenge, hyperuricemic mice were orally administered saline, wild-type EcN, or recombinant EcN strains equipped with TOSS or T1SS, namely EcNΔnlpIΔthyA-SU or EcNΔthyA-SUT, every two days. Figure 17A). In a mouse model of hyperuricemia treated with saline, serum UA concentration increased dramatically to ~599.7 μM, approximately 2.18 times higher than the normal control group (serum UA concentration ~274.7 μM). In hyperuricemic mice treated with wild-type EcN, EcNΔnlpIΔthyA-SU, or EcNΔthyA-SUT doses, treatment with wild-type EcN or recombinant EcN equipped with T1SS reduced serum UA to ~514.3 μM or ~464.8 μM, respectively. Notably, our engineered EcN equipped with T1SS demonstrated therapeutic efficacy in reducing circulating UA concentration, significantly decreasing serum UA concentration from ~599.7 μM to ~279.3 μM, approaching the level of healthy controls (serum UA concentration approximately 274.7 μM). Figure 17 (B). These results demonstrate that our engineered EcN equipped with T0SS is superior to recombinant EcN equipped with T1SS in alleviating hyperuricemia, indicating that OMV-encapsulated Uox secreted by T0SS has a superior effect on degrading circulating UA due to its excellent protein stability and wide delivery range outside the intestine.

[0151] We further evaluated the performance of our engineered EcN equipped with modified TOSS in improving symptoms of hyperuricemia. Hyperuricemia is often accompanied by increased urinary UA levels (Wu, Y. et al. Limosilactobacillus fermentum JL-3 isolated from "Jiangshui" ameliorates hyperuricemia by degrading uric acid. Gut Microbes 13, 1-18 (2021)), which was also observed in a mouse model of hyperuricemia. Figure 17 (C). Compared with the healthy control group, the urinary UA level in hyperuricemic mice treated with saline was increased by approximately 2.25 times (C). Figure 17 Treatment with EcN, EcNΔthyA-SUT, or EcNΔnlpIΔthyA-SU significantly reduced urinary uric acid (UA) in hyperuricemic mice by approximately 21.59%, 41.40%, and 46.70%, respectively. Figure 17 UA is widely considered a damage-associated molecular pattern (DAMP) that can activate the NLRP3 inflammasome and increase the release of inflammatory cytokines, such as IL-1β. G., TO, Klück, V., Popp, RA & Joosten, LABUrate-induced immune programming: Consequences for gouty arthritis and hyperuricemia. Immunol. Rev. 294, 92-105 (2020). Therefore, serum levels of the inflammatory cytokine IL-1β were examined to assess systemic inflammation, and serum creatinine (CRE) levels were measured to assess renal function and the biosafety of drug treatment. Compared with healthy controls, serum IL-1β and serum creatinine (CRE) levels were significantly increased in hyperuricemic mice treated with saline. Figure 17 (D and E). Compared with hyperuricemic mice administered saline, oral administration of EcNΔnlpIΔthyA-SU, instead of EcNΔthyA-SUT or wild-type EcN, significantly reduced serum CRE and serum IL-1β levels by ~15.94% and ~32.91%, respectively. Figure 17 (D and E). Furthermore, histological analysis of colon, liver, and kidney tissues was performed using H&E staining. Figure 17 In hyperuricic mice treated with saline or wild-type EcN, significant immune cell infiltration was observed in the liver and kidneys, while the group treated with EcNΔnlpIΔthyA-SU showed a significant reduction in immune cell infiltration. Figure 17 These results demonstrate that our engineered EcN equipped with improved TOSS exhibits superior degradation of oral Uox and serum UA, thus demonstrating excellent efficacy in alleviating the pathological symptoms of hyperuricemia and with favorable biocompatibility. Using hyperuricemia as an example, we demonstrate that our engineered EcN equipped with improved TOSS is superior in treating metabolic disorders compared to conventional recombinant co-coagulation that secretes therapeutic proteins into the periphery of the gut, potentially benefiting from the superior stability, high endogenous encapsulation rate, and broad systemic delivery range of the therapeutic proteins encapsulated in OMVs.

[0152] Example 7

[0153] Engineering OMV can effectively detoxify clinical samples from patients.

[0154] To further investigate the clinical potential of our engineered EcN equipped with TOSS for treating metabolic diseases, the efficacy of OMV containing protein-encapsulated protein extracted from engineered EcN in treating human serum samples was evaluated. Serum samples were collected from nine patients with hyperuricemia and incubated with OMV loaded with Uox from EcNΔnlpIΔthyA-SU. After 30 minutes of incubation, serum UA levels decreased significantly after OMV treatment, with an average reduction of 42.6%, from 496.9 μM to 285.3 μM. Figure 17 The serum glucose levels of these nine samples were not affected by OMV incubation. Figure 17 These results demonstrate that the synthetic Uox-loaded OMV extracted from our engineered EcN is effective and specific in reducing UA degradation in the blood of patients with hyperuricemia. Furthermore, serum samples were collected from seven lung cancer patients with elevated lactate levels and incubated with Lox-loaded OMV derived from EcNΔnlpI-OL. Incubation with Lox-loaded OMV for 30 minutes resulted in a decrease in serum lactate levels in all seven human samples, with a mean decrease from ~5.67 mM to ~5.50 mM. Figure 17 Similarly, Lox-loaded OMV significantly reduced serum lactate levels but did not affect serum glucose levels. Figure 17 (J). Overall, we demonstrate that the therapeutic proteins delivered by our engineered EcN based on engineered TOSS can effectively detoxify circulating human metabolites, and our engineered EcN-based protein delivery system holds promise for the clinical treatment of a wide range of metabolic disorders.

[0155] Those skilled in the art should understand that although the present invention has been specifically described with reference to the above embodiments, the present invention is not limited to these specific embodiments. Based on the methods and technical solutions taught in this invention, those skilled in the art can make appropriate modifications or improvements without departing from the spirit of the present invention, and the equivalent embodiments obtained therefrom are all within the scope of the present invention.

[0156]

[0157]

[0158]

[0159]

[0160]

[0161]

[0162]

Claims

1. Use of recombinant *Escherichia coli* or its derived outer membrane vesicles (OMV) in the preparation of a medicament for oral or intestinal mucosal administration to a subject, wherein the recombinant *Escherichia coli* comprises at least one therapeutically active ingredient in its outer membrane vesicles (OMV), wherein the therapeutically active ingredient is a therapeutic enzyme, wherein the medicament, recombinant *Escherichia coli*, or OMV is formulated for and / or relies on the OMV to penetrate the intestinal epithelial barrier, thereby delivering the therapeutically active ingredient to the circulatory system of the subject, wherein: The therapeutic enzyme is uricase and the drug is used to treat hyperuricemia; The therapeutic enzyme is catalase and the drug is used to treat peroxisome disorders; The therapeutic enzyme is lactate oxidase and the drug is used to treat hyperlactatemia; The therapeutic enzyme is ethanol oxidase and / or acetaldehyde oxidase, and the drug is used to treat hyperalcoholemia; or The therapeutic enzyme is phenylalanine deaminase and the drug is used to treat phenylketonuria.

2. The use according to claim 1, wherein the recombinant Escherichia coli is recombinant Nissle 1917 (EcN) and / or lacks the nlpI gene.

3. The use according to claim 1, wherein the therapeutic enzyme is fused with a signal peptide secreted to the OMV.

4. The use according to claim 3, wherein the signal peptide is selected from Sec signal peptide, Tat signal peptide and Srp signal peptide.

5. The use according to claim 1, wherein the expression of the therapeutic enzyme is controlled by an inducible or constitutive promoter.

6. The use according to claim 5, wherein the promoter is selected from the TAC promoter, FNRS promoter, NA promoter, NA618 promoter, and J23117 promoter.

Images