Bioconjugate compositions and methods for producing bioconjugates

Abstract

The present invention relates to improved pharmaceutical compositions comprising one or more bioconjugate compositions. Each bioconjugate composition comprises a mixture of mono- and multi-glycosylated bioconjugates of a specific Escherichia coli (E. coli) O-antigen polysaccharide covalently attached to a Pseudomonas aeruginosa exoprotein A carrier protein. The present invention further relates to methods for producing purified bioconjugates from Gram-negative host cells, and bioconjugates obtainable or obtainable by said methods.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to European Patent Application No. 22211401.9, filed December 05, 2022, the disclosure of which is incorporated herein by reference in its entirety.

[0002] Reference to an electronically submitted sequence listing The contents of the electronic sequence listing (CRU6090EPEPA1.xml; size: 4,785 bytes; creation date: November 28, 2023) are incorporated herein by reference in their entirety. [Background technology]

[0003] In contrast to commensal E. coli, extraintestinal pathogenic Escherichia coli (ExPEC) strains express a wide range of virulence factors that enable them to colonize the gastrointestinal tract and cause a wide range of extraintestinal infections, which are associated with significant healthcare costs due to hospital stays and mortality. Neonates, the elderly, and immunocompromised patients are particularly susceptible to ExPEC infections, including invasive ExPEC disease (IED).

[0004] O-antigens comprise the immunodominant components of cell wall lipopolysaccharides (LPS) in Gram-negative bacteria, including Escherichia coli (E. coli). Currently, >180 serologically unique E. coli O-antigens have been identified, and most ExPEC isolates fall within the range of fewer than 20 O-antigen serotypes. The full-length E. coli O-antigen polysaccharide (O-PS) typically consists of approximately 10–25 repeating sugar units. In wild-type E. coli, such O-PS is attached to a highly conserved LPS core structure. Each of these components, i.e., the O-PS components and the LPS core structure, are synthesized separately by enzymes encoded primarily in the rfb and rfa gene clusters, respectively. Following O-antigen polymerization, the O-PS backbone can be modified, typically by the addition of acetyl or glucose residues.

[0005] Efforts toward developing vaccines to prevent ExPEC infection have focused on multivalent compositions comprising bioconjugates of Escherichia coli (E. coli) O-antigen polysaccharides covalently attached to carrier proteins (see, e.g., Poolman and Wacker, J. Infect. Dis. (2016) v.213(1), pp. 6-13; WO 2015 / 124769; WO 2017 / 035181; WO 2020 / 191082).

[0006] In general, CD4 + Efficient activation of T cells is driven by the endosome-derived generation of antigenic fragments of polysaccharides that are covalently linked to peptides derived from carrier proteins [see, e.g., Avci et al., Nat. Med. (2011), 17:1602-1609]. Although a promising class of vaccines, the production of complex bioconjugates in general, and in particular bioconjugates containing bacterial O-antigen polysaccharides covalently linked to carrier proteins, is challenging. However, several production methods, including purification of such bioconjugates, have been described [see, e.g., WO 2009 / 104074; WO 2020 / 191082; WO 2022 / 214620; van den Dobbelsteen et al., Vacine (2016), 34:4152-4160].

[0007] Generally, O-EPA bioconjugates can be produced by enzymatic conjugation of O-antigen polysaccharide (O-PS) components to carrier proteins within Gram-negative host cells, such as Escherichia coli (E. coli), using, for example, the PglB oligosaccharyltransferase system (see, e.g., WO 2015 / 124769; WO 2020 / 191082; Poolman and Wacker, J. Infect. Dis. (2016) v.213(1), pp. 6-13, and references therein). In this method, coupling of the O-PS components to the carrier protein occurs in the periplasmic space of Gram-negative bacteria, such as E. coli, i.e., the space between the inner and outer cytoplasmic membranes.

[0008] Specifically, WO 2022 / 214620 describes a method for producing O-EPA bioconjugates on a commercial scale. This method is therefore a significant improvement over previous methods for producing O-EPA bioconjugates, which involve a size exclusion chromatography step (SEC), making these methods disadvantageous for large-scale production (e.g., van den Dobbelsteen et al., Vaccine (2016), 34:4152-4160; Burckhardt et al., Vaccine (2019), 37(38):5762-5769; Ravenscroft et al., Glycobiology (2016), 26(1):51-62; WO 2009 / 104074; or WO 2015 / 124769). Thus, WO 2022 / 214620 describes for the first time a method suitable for large-scale production of O-EPA bioconjugates, e.g., in bioreactors having volumes from 100 L to 20,000 L. However, despite its significant advantages over previous methods, this method can be further improved, for example, to prepare bioconjugate compositions having a relatively high proportion of multiglycosylated O-EPA bioconjugates, i.e., EPA carrier proteins in which two or more glycosylation sites are occupied, e.g., EPA carrier proteins attached to two, three, or four O-antigen polysaccharides. Consequently, a higher proportion of multiglycosylated O-EPA bioconjugates results in a lower amount of carrier protein relative to the amount of O-antigen polysaccharide. This is particularly important because a minimal amount of carrier protein reduces the potential risk of carrier protein-induced adverse events or carrier protein-induced immunosuppression (see, e.g., Juergens et al., Hum Vaccin Immunother. 2018; 14(8): 1948-1956; Knuf et al. Vaccine 2011; 29(31): 4881-4890).

[0009] Although generally well tolerated, it is important to develop pharmaceutical compositions that minimize the risk of adverse events, however negligible, which is particularly important in the development of vaccines that are typically administered to healthy individuals.

[0010] Considering the above-described mechanism of T cell action (Avci et al., Nat. Med. (2011), 17:1602-1609), and the goal of developing vaccines that minimize the risk of adverse events, the development of bioconjugate compositions with relatively high multiglycosylation patterns, and methods for producing such bioconjugates, is considered a major improvement over currently known bioconjugate compositions and production methods.

[0011] Thus, there is an unmet medical need to provide improved O-EPA bioconjugate compositions, and correspondingly, improved pharmaceutical compositions comprising one or more O-EPA bioconjugate compositions.

[0012] Furthermore, those skilled in the art will appreciate that efficient manufacturing methods are required to obtain sufficient quantities of safe product for vaccination of large numbers of people in an economically feasible manner. Thus, there is a constant need for improved methods for producing and / or purifying such complex bioconjugates. For example, in addition to improvements in certain attributes of the resulting bioconjugate itself, such as improved degree of glycosylation, the methods described in WO 2022 / 214620 may be further improved with respect to efficient large-scale production of bioconjugates (e.g., culturing host cell cultures in bioreactors having volumes of at least 100 L and up to 20,000 L). Such improvements may relate, inter alia, to yield, purity, and buffer consumption. For example, reduced buffer consumption is desirable for both economic (lower costs) and environmental (less waste) reasons.

[0013] As a result, there remains a need for improved methods for producing purified bioconjugates comprising bacterial O-antigen polysaccharides covalently linked to carrier proteins, particularly Escherichia coli (E. coli) O-antigen-EPA bioconjugates, that address the above-mentioned shortcomings. Summary of the Invention

[0014] The present invention relates to improved pharmaceutical compositions comprising one or more bioconjugate compositions. Each bioconjugate composition comprises a mixture of bioconjugates, each comprising a bacterial O-antigen polysaccharide of a specific Escherichia coli (E. coli) serotype covalently attached to a Pseudomonas aeruginosa exoprotein A carrier protein (EPA; O-EPA bioconjugate). Each O-EPA bioconjugate can be monoglycosylated with one O-antigen polysaccharide attached to the EPA or multiglycosylated with two, three, or four O-antigen polysaccharides attached to the EPA, such that the bioconjugate composition is a mixture of monoglycosylated and multiglycosylated O-EPA bioconjugates. In certain embodiments, the EPA carrier protein has the amino acid sequence of SEQ ID NO: 1. The present invention further relates to methods for producing purified O-EPA bioconjugates from Gram-negative host cells. The method includes providing a bioconjugate in a filtered periplasmic fraction (FPF) of a host cell, and purifying the bioconjugate. The purification includes (i) a first anion exchange chromatography step followed by (ii) a mixed-mode chromatography (MMC) step followed by (iii) hydrophobic interaction chromatography (HIC), and (iv) a second anion exchange chromatography step. The present invention further relates to an O-EPA bioconjugate composition obtainable or obtainable by the method according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0015] The above object is achieved by a pharmaceutical composition according to claim 1, a method for producing O-EPA bioconjugates according to claim 5, and an O-EPA bioconjugate composition obtained or obtainable by said method according to claim 16. Further aspects of the invention are disclosed in the description and independent claims, and preferred embodiments are disclosed in the description and dependent claims.

[0016] The present invention will be described in more detail below. It is understood that the various embodiments, options, and ranges provided / disclosed in this specification can be arbitrarily combined. Furthermore, depending on the specific embodiment, the selected definition, embodiment, or range may not apply. Unless otherwise specified, the following definitions shall apply herein.

[0017] As used herein, the terms "a," "an," "the," and similar terms used in the context of this invention (particularly in the context of the claims) are to be construed to cover the plural as well as the singular, unless otherwise specified herein or clearly contradicted by context.

[0018] As used herein, the terms "including," "containing," and "comprising" are used herein in their open, non-limiting sense. It is understood that the various embodiments, options, and ranges are combinable at will.

[0019] As used herein, the term "about" when used in conjunction with a number refers to any number within ±10%, preferably ±5% or ±1% of the referenced number.

[0020] A number of abbreviations are used throughout this specification, including the following: AEX1 First anion exchange chromatography AEX2 Second Anion Exchange Chromatography cGE Capillary Gel Electrophoresis cHA Ceramic Hydroxyapatite CPF Centrifuged Periplasmic Fraction CV column volume DOGY Degree of Glycosylation EPA Pseudomonas aeruginosa exoprotein A (also known as exotoxin A) ExPEC extraintestinal pathogenic Escherichia coli (E. coli) FPF Host cell filtered periplasmic fraction HA hydroxyapatite HCP host cell protein HIC Hydrophobic Interaction Chromatography HPLC High-Performance Liquid Chromatography IC-PAD Ion Chromatography with Pulsed Amperometric Detection IED Invasive ExPEC disease MMC Mixed Mode Chromatography MMR Multimodal Chromatography Resin O-EPA: a bioconjugate of EPA carrier protein covalently linked to bacterial O-antigen polysaccharide PF periplasmic fraction RP-HPLC Reversed-phase high-performance liquid chromatography SEC size exclusion chromatography SE-HPLC Size-exclusion high-performance liquid chromatography sEPA short-chain glycosylated EPA TFF Tangential Flow Filtration WFI water for injection

[0021] The term "glycoconjugate" is known in the art and particularly describes a chemical entity covalently bound to one or more polysaccharides. Such glycoconjugates can be obtained by biological conjugation in living cells ("bioconjugates" or "biological conjugates") or by chemical conjugation of polysaccharides ("chemical" or "synthetic" glycoconjugates). In particular, suitable chemical entities are proteins, and the corresponding glycoconjugates are glycoproteins. Specifically, the term glycoconjugate relates to conjugation products in which a polysaccharide (i.e., a glycan) is covalently bound to a carrier protein. When a carrier protein contains two or more glycosylation sites, a mixture of glycoconjugates, particularly bioconjugates, is obtained, such as a mixture of mono-, di-, tri-, and tetra-glycosylated bioconjugates. An example of a carrier protein containing four glycosylation sites is the EPA carrier protein having the amino acid sequence of SEQ ID NO: 1.

[0022] "Bioconjugate compositions" include such mixtures, which may contain a single polysaccharide bioconjugate (e.g., in a drug substance, a glycoconjugate of a carrier protein bound to an O-antigen polysaccharide of a particular bacterial serotype; sometimes referred to as a "monovalent composition"), or may contain two or more polysaccharide bioconjugates (e.g., in a drug product, multiple bacterial serotype O-antigen polysaccharides, each individually bound to a carrier protein, are included in the same composition; sometimes referred to as a "multivalent composition").

[0023] The term glycoprotein includes "glycoconjugate vaccines," in which the emphasis is on the glycan moiety against which an immune response is desired, because the glycan is the relevant antigen, and the protein moiety simply serves as a carrier to generate the desired T cell memory immune response.

[0024] The term "polysaccharide" is known in the art and specifically describes a polymeric hydrocarbon, either linear or branched, consisting of monosaccharide units linked together by glycosidic bonds. Such polysaccharides are characterized by their repeating units, each of which is described by its respective monosaccharide composition. The repeating units contain one or more monosaccharides that may be further chemically modified (e.g., amidated, sulfonated, acetylated, phosphorylated, etc.). The monosaccharides typically found in the repeating units are cyclic or linear monosaccharides containing 3 to 7 carbon atoms. In the specific case of glycoconjugate vaccines, the conjugated polysaccharide is derived from a pathogenic species (e.g., Escherichia coli) bearing the repeating units, as defined by the genetics of the specific pathogen. Thus, the repeating units can be a specific marker / identifier for the pathogen.

[0025] The term "polysaccharide component" therefore refers to one or more glycan chains of a glycoconjugate. Glycans can be monomers or polymers of sugar residues, but typically contain at least three sugars and can be linear or branched. Glycans can contain natural sugar residues (e.g., glucose, N-acetylglucosamine, N-acetylneuraminic acid, galactose, mannose, fucose, arabinose, ribose, xylose, etc.) and / or modified sugars (e.g., 2'-fluororibose, 2'-deoxyribose, phosphomannose, 6'-sulfo N-acetylglucosamine, etc.). The term "glycan" includes homopolymers and heteropolymers of sugar residues. The term "glycan" also includes glycan components of glycoconjugates (e.g., of glycoproteins, glycopeptides).

[0026] As used herein, the term "O-acetylated polysaccharide" refers to a polysaccharide in which one or more monosaccharides of the repeating unit have been modified by acetylation. The monosaccharide has one or more of its present hydroxyl groups in an acetylated state. With respect to pathogen-derived repeating units used in glycoconjugate vaccines, O-acetylation of specific monosaccharides can affect the induction of an immune response against the pathogen. Examples of pathogen-derived polysaccharide components are shown in Table 1.

[0027] The terms "glycan" / "glycan chain" are synonyms for "polysaccharide," as defined below. Thus, in the context of the present invention, "glycan" and the prefix "glyco" refer interchangeably to glycoconjugates, e.g., the carbohydrate moiety of a glycoprotein. Thus, "glycosylation" refers to one or more polysaccharides covalently attached to a protein; thus, the protein is "monoglycosylated" or "multiglycosylated," respectively.

[0028] As used herein, the term "serotype" refers to glycoconjugates having different polysaccharide chains derived from different bacterial serotypes. Examples of glycans from a number of E. coli serotypes are identified in Table 1 below.

[0029] As used herein, the term "load adjustment" refers to the step of adjusting a load of a process intermediate, such as a periplasmic fraction of a host cell containing an O-EPA conjugate, or a fraction containing O-EPA after the first, second, or third purification step, to conditions suitable for applying the process intermediate to a chromatography resin for further purification. Unless otherwise specified or clearly contradicted by the context, "load adjustment" refers to adjusting the conductivity of the load to a target conductivity suitable for the subsequent purification step and / or adjusting the pH of the load to a target pH suitable for the subsequent purification step. Furthermore, "load adjustment" includes adjusting the concentration of a process intermediate, i.e., reducing the processing volume, particularly by TFF.

[0030] The terms "resin" and "media" are used interchangeably herein and refer to a chromatography resin or medium used in the separation of a target protein, i.e., an O-EPA bioconjugate, from impurities. Resins useful in the present invention can be in different formats, e.g., beads, filters (membranes), cartridges, etc., and all should be considered "resins" according to the present invention. In certain embodiments, the resin is in the form of beads that can be used in a column. In certain embodiments, the resin is in the form of a membrane with functional groups. In certain embodiments, the resin is in the form of a cartridge that can be used directly. Resins that can be used according to the present invention are commercially available from vendors such as Cytiva (formerly GE Healthcare), Bio-Rad, and / or others.

[0031] The term "capture step" is known in the art and refers to the first chromatography step, which aims to bind proteins of interest from crude samples and isolate them from harmful contaminants such as proteases and glycosidases. The target protein, e.g., O-EPA bioconjugate, is concentrated and transferred to a buffer that maintains the functional and structural integrity of the O-EPA bioconjugate. Removal of other harmful contaminants can also be achieved by carefully optimizing the binding conditions.

[0032] The focus in optimizing the capture step is on capacity and speed, so it may be acceptable to compromise resolution in order to maximize the capacity and / or speed of the separation in the first step.

[0033] The term "final polishing step" is known in the art and relates to a chromatography step performed as a final chromatography step with the aim of further increasing the purity of a target protein, e.g., an O-EPA bioconjugate.

[0034] The term "bind-elute mode" is known in the art and refers to a mode of separation that operates by first binding sample components, particularly proteins / bioconjugates of interest, to a chromatography resin. Once the sample components are bound, the resin is washed with a buffer to remove unbound materials. The bound materials are then eluted. This mode of separation is in contrast to flow-through mode, in which the pH / ionic strength of the sample and buffer are selected so that proteins flow through the column rather than bind, leaving most or certain impurities bound.

[0035] In an embodiment of the present invention, in steps ii) to v), the conditions are adjusted to first allow binding of the O-EPA bioconjugate to the chromatography medium and then to allow elution of the O-EPA bioconjugate from the medium, i.e., each chromatography step is performed in bind-elute mode.

[0036] Viable and non-viable particles: Viable particles are particles that contain one or more living microorganisms, e.g., bacteria. They can affect the sterility of a pharmaceutical product. Non-viable particles are particles that do not contain living microorganisms but can act as a delivery vehicle for viable particles. Viable and non-viable particles generally range in size from about 0.2 μm to 30 μm, typically from about 0.2 μm to 5 μm.

[0037] As used herein, the term "drug substance" refers to a bulk product of an individual bioconjugate (e.g., an E. coli O-antigen polysaccharide covalently linked to an EPA carrier protein, e.g., an E. coli O25B O-antigen covalently linked to EPA) at a higher concentration than the product ultimately administered to a subject in need. The drug substance may be produced after purification of the bioconjugate. The drug substance may be stored, for example, in a more concentrated form in an appropriate formulation buffer (see, e.g., WO 2018 / 077853, WO 2020 / 191082), for example, under frozen conditions, e.g., at -70°C.

[0038] As used herein, the term "pharmaceutical product" refers to a formulation of a bioconjugate, particularly an E. coli O-antigen polysaccharide individually conjugated to an EPA carrier protein, in its final form for administration to a subject in need thereof. As used herein, the term "pharmaceutical product" particularly relates to a multivalent vaccine composition, e.g., a tetravalent ExPEC glycoconjugate vaccine composition comprising E. coli O-antigen polysaccharides O25B, O1A, O2, and O6A, each individually conjugated to an EPA carrier protein. Further non-limiting examples of multivalent glycoconjugate vaccine compositions include, for example, a nonavalent glycoconjugate vaccine composition comprising E. coli O-antigen polysaccharides O1A, O2, O4, O6A, O15, O16, O18A, O25B, and O75, and a decavalent glycoconjugate vaccine composition comprising, for example, E. coli O-antigen polysaccharides O1A, O2, O4, O6A, O8, O15, O16, O18A, O25B, and O75, wherein each E. coli O-antigen polysaccharide is individually conjugated to an EPA carrier protein (see, e.g., WO 2020 / 191082 and WO 2022 / 058945 for examples of nonavalent and decavalent glycoconjugate vaccine compositions, respectively). Pharmaceutical products can typically be prepared by mixing the drug substances of each glycoconjugate and, if necessary, diluting with an appropriate formulation buffer to produce the target dose of the vaccine (see, e.g., WO 2018 / 077853, WO 2020 / 191082, WO 2022 / 058945).

[0039] In a first aspect, the present invention relates to a method for producing a purified bioconjugate from a Gram-negative bacterial host cell, said bioconjugate in particular comprising a bacterial O-antigen polysaccharide covalently linked to a carrier protein, said carrier protein being Pseudomonas aeruginosa exoprotein A (O-EPA bioconjugate). It should be understood that the Pseudomonas aeruginosa exoprotein A (EPA) carrier protein may be a detoxified or recombinant version thereof. The method includes providing a filtered periplasmic fraction (FPF) of host cells expressing a bioconjugate (step i), the FPF containing the bioconjugate. The method further includes several chromatographic purification steps (referred to herein as steps ii-v). Step ii involves subjecting the optionally adjusted load of the FPF to a first anion exchange chromatography (AEX1) step to obtain a first AEX eluate (AEX1) containing the bioconjugate. This step is the capture step. Step iii comprises "subjecting the optionally adjusted load of AEX1 eluate to a mixed-mode chromatography (MMC) step using a multimodal resin (MMR) to obtain an MMR eluate, said MMR comprising both anion exchange functionality and hydrophobic functionality, to obtain an MMR eluate comprising the bioconjugate." Step iv involves "subjecting the conditioned load of MMR eluate to a hydrophobic interaction chromatography (HIC) step to obtain an HIC eluate containing the bioconjugate." Step v involves subjecting the conditioned HIC eluate load to a second anion exchange chromatography (AEX2) step to obtain the AEX2 eluate containing the bioconjugate as product, which is the final purification step.

[0040] In purification steps ii-v, i.e., all chromatography steps, conditions are adjusted to first allow binding of the bioconjugate to the chromatography medium and subsequently to allow elution of the bioconjugate from said chromatography medium (bind-elute mode).

[0041] It should be understood that steps i-v are performed in the order shown. According to this purification scheme, the eluate after each chromatographic step is enriched with respect to the O-EPA bioconjugate relative to the eluate of the previous step, i.e., the purity of the O-EPA bioconjugate increases after each chromatographic step relative to the previous purity. The individual production steps outlined above, taken individually, are known per se. However, it has been found that a particular combination of steps i-v above is particularly suitable for the production of a variety of different O-EPA bioconjugates and is amenable to large-scale production. The inventive process typically results in an O-EPA bioconjugate with a purity of ≧90%, e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.5%, 99.7%, 99.9%, or 100%, as measured by SE-HPLC, or ≧95%, e.g., 95%, 96%, 97%, 98%, 99%, 99.5%, 99.7%, 99.9%, or 100%, as measured by RP-HPLC. Typically, the purity of the O-EPA bioconjugate is ≧95%, preferably ≧98%, e.g., 98%, 99%, 99.5%, 99.7%, 99.9%, or 100%, as measured by either SE-HPLC or RP-HPLC. A purity of at least 90% or greater, e.g., at least 95%, or at least 98%, as measured by SE-HPLC, is advantageous for components of pharmaceuticals intended for human administration, e.g., multivalent bioconjugate vaccines containing multiple O-EPA conjugates.

[0042] Furthermore, following the above protocol, the overall yield of the O-EPA bioconjugate is acceptable for a large-scale process for producing a pharmaceutical preparation of the O-EPA bioconjugate (e.g., at least a 5% overall yield, preferably at least a 10% overall yield, e.g., about a 5-35% overall yield [relative to the O-EPA bioconjugate present in the filtered periplasmic fraction] from at least a 150 L bioreactor, with a purity of at least 90% as measured by SE-HPLC). Typically, the purity is at least 95% as measured by either SE-HPLC or RP-HPLC.

[0043] Importantly and surprisingly, the particular combination of steps i-v typically results in an O-EPA bioconjugate composition having a relatively high proportion of multiply glycosylated O-EPA bioconjugates.

[0044] Further details regarding each of steps i-v are provided below.

[0045] Step i: In the first step, FPFs of Gram-negative bacterial host cells expressing the O-EPA bioconjugate are provided. As described above, the FPFs contain the O-EPA bioconjugate.

[0046] The preparation of host cell filtered periplasmic fractions containing O-EPA bioconjugates is known per se (see, for example, WO 2009 / 104074, WO 2015 / 124769 or WO 2020 / 191082). To obtain the periplasmic fraction by osmotic shock treatment, cells are first incubated in a buffer with a relatively high osmolarity (hypertonic) and then in a buffer with a relatively low osmolarity (hypotonic). Such osmotic shock treatment results in at least partial removal of the cell wall and the production of spheroplasts (i.e., cells, particularly Gram-negative cells, whose cell walls have been at least partially removed). Thus, the majority of host cell proteins remain in the spheroplasts, while periplasmic proteins are released into the suspension medium. Typically, Escherichia coli (E. coli) host cells containing genetic information encoding PglB, EPA, and the corresponding enzymes for biosynthesis of the O-antigen polysaccharide are cultured in a bioreactor having a volume of, for example, about 100 L to 20,000 L, e.g., 150 L to 5,000 L, e.g., 200 L, 800 L, or 2,000 L, and the cells produce the O-EPA bioconjugate. Typically, once the cells enter stationary phase, the culture is cooled to below 20°C and the cells are harvested via centrifugation, e.g., continuous centrifugation using a disc stack centrifuge. The cells are resuspended in an appropriate liquid, e.g., 0.9% NaCl solution or Tris-buffered saline (TBS), and subjected to an osmotic shock using a solution preferably at about 2 to 15°C, preferably about 6 to 10°C. Osmotic shock can be performed, for example, by adding a sucrose solution (e.g., 60% sucrose, pH 8, 480 mM Tris-HCl, 24 mM EDTA) to the cells to a target sucrose concentration of about 20-30%, e.g., 25%, and incubating the mixture with mixing at about 2-15°C, preferably about 6-10°C, for about 15 minutes to 4 hours, e.g., about 1 hour. After incubation with sucrose, the cell / sucrose solution is mixed with a solution of lower osmolarity, e.g., about 4x volume of 10 mM Tris-HCl (pH 8.0), at 6-10°C. Mixing can be performed, for example, with a static mixer. The O-EPA product is released from the periplasmic space into the supernatant (referred to as the periplasmic fraction, PF), which is then collected. The material is then clarified by removing cellular debris in the PF, e.g., by separation in a disc stack centrifuge, and the supernatant (referred to as the centrifugal periplasmic fraction, or CPF) is collected. To remove remaining cellular debris, the CPF is preferably further filtered, e.g., through a depth bioburden reduction filter (e.g., a membrane with a pore size of approximately 0.2 μm), and the resulting material is collected as the filtered periplasmic fraction (FPF). This FPF can then be used in the chromatographic purification process of the invention described herein [starting with step ii].

[0047] Thus, in one embodiment, step i further comprises: (i-1) incubation of the Gram-negative bacterial host cells in a bioreactor having a volume of between 100 L and 20,000 L, e.g., 150 L to 5,000 L, e.g., 200 L or 800 L or 2,000 L, at a temperature of between 34°C and 36°C, e.g., 35°C, for growth to stationary phase prior to harvesting; followed by (i-2) recovering the Gram-negative bacterial host cells, which comprises a continuous flow centrifugation step to obtain recovered Gram-negative bacterial host cells containing the O-EPA bioconjugate in the periplasm.

[0048] Thus, in another embodiment, step i further comprises (i-3) osmotic shock treatment of the Gram-negative bacterial host cells to obtain a periplasmic fraction of the host cells containing the O-EPA bioconjugate. It will be apparent to one skilled in the art that step (i-3) follows step (i-2). In certain embodiments, the osmotic shock treatment in step (i-3) comprises adding a sucrose solution, preferably further containing EDTA, to the cells to a target sucrose concentration of about 25%, incubating the mixture at about 6-10°C for about 15 minutes to 2 hours with mixing, followed by adding a solution with a low osmolarity (e.g., 10 mM Tris-HCl pH 8) to the cell / sucrose solution at about 6-10°C to reduce the osmolarity by at least four-fold (compared to a composition containing 25% sucrose) (wherein the O-EPA bioconjugates are released from the periplasm into the supernatant), and collecting the supernatant (periplasmic fraction), followed by filtering the periplasmic fraction to obtain a filtered periplasmic fraction.

[0049] In a further embodiment, step i further comprises step (i-4) filtration to obtain a filtered periplasmic fraction (FPF) containing the O-EPA bioconjugate. As outlined above, the filtration step is used to remove remaining cellular debris, for example, by filtration through a depth bioburden reduction filter (e.g., a membrane with a pore size of about 0.2 μm).

[0050] In one embodiment, step (i) comprises: (i-1) Incubating the host cells in a bioreactor having a volume of between 100 L and 20,000 L, e.g., 150 L to 5,000 L, at a temperature of between 34°C and 36°C to grow to stationary phase prior to harvesting; (i-2) harvesting the host cells by a continuous flow centrifugation step to obtain harvested host cells containing the bioconjugate; and (i-3) osmotic shock treatment of the host cells to obtain a periplasmic fraction of the host cells containing the bioconjugate; and (i-4) Filtration to obtain the filtered periplasmic fraction (FPF) containing the O-EPA bioconjugate Includes:

[0051] Host Cell: In a preferred embodiment, the Gram-negative bacterial host cell from which the O-EPA bioconjugate is obtained contains (a) a bacterial O-antigen polysaccharide, and (b) a recombinant Pseudomonas aeruginosa exoprotein A (EPA) containing at least one glycosylation site, and (c) genetic information encoding the metabolic machinery for N-glycosylation of EPA by the O-antigen polysaccharide, thereby producing the O-EPA conjugate in vivo in the periplasm of the Gram-negative bacterial host cell. In a preferred embodiment, the Gram-negative bacterial host cell is an Escherichia coli host cell, more preferably an E. coli K-12 host cell, such as E. coli K-12 strain W3110. Details of this aspect are described in more detail below and in the art, for example, in International Publication Nos. 2009 / 104074, 2020 / 191082, and 2020 / 191088, which are incorporated herein by reference in their entireties.

[0052] In a preferred embodiment, the waaL gene is deleted from or functionally inactivated in the genome of the host cell of the present invention. The terms "waaL" and "waaL gene" refer to an O antigen ligase gene that encodes a membrane-bound enzyme with an active site located in the periplasm. The enzyme encoded by the waaL gene transfers undecaprenyl phosphate (UPP)-linked O antigens to the lipid A core to form lipopolysaccharide. Deletion or disruption of the endogenous waaL gene (e.g., a ΔwaaL strain) can disrupt the transfer of O antigens to lipid A and instead enhance the transfer of O antigens to other available biomolecules, such as carrier proteins expressed in the host cell of the present invention. In one embodiment of the host cell of the present invention, the E. coli gtrABS gene, responsible for glucosylation of the O16 O antigen, is deleted from the genome of the host cell or functionally inactivated in the genome. In a preferred embodiment, the E. coli gtrABS gene is deleted from the genome of an E. coli W3110 host cell or functionally inactivated in the genome. The gtrA and gtrB genes in different serotypes are highly homologous and interchangeable, while the gtrS gene encodes a serotype-specific O antigen glycosyltransferase. The GtrS of E. coli W3110 can transfer a glucose (Glc) residue to the GlcNAc sugar in the aL-Rha-(1→3)-D-GlcNAc motif of the E. coli O16 O antigen.

[0053] O-antigen polysaccharide: In a preferred embodiment, the O-antigen polysaccharide is specific for Gram-negative bacteria selected from the list consisting of Escherichia and Shigella, preferably E. coli.

[0054] In one embodiment, the O-antigen polysaccharide is selected from the group consisting of E. coli O-antigen polysaccharides O1A, O2, O4, O8, O6A, O15, O16, O18A, O25B and O75. In one embodiment, the O-antigen polysaccharide is selected from the group consisting of E. coli O-antigen polysaccharides O1A, O2, O4, O6A, O15, O16, O18A, O25B and O75. In one embodiment, the O-antigen polysaccharide is selected from the group consisting of E. coli O-antigen polysaccharides O1A, O2, O6A, O16, O18A, O25B and O75. In one embodiment, the O-antigen polysaccharide is selected from the group consisting of E. coli O-antigen polysaccharides O1A, O2 and O25B. In one embodiment, the O-antigen polysaccharide is E. coli O-antigen polysaccharide O1A. In one embodiment, the O-antigen polysaccharide is E. coli O-antigen polysaccharide O2. In one embodiment, the O-antigen polysaccharide is E. coli O-antigen polysaccharide O25B.

[0055] In E. coli, gene products involved in O-antigen polysaccharide biosynthesis are encoded by the rfb locus. Accordingly, the host cells provided herein further comprise a nucleotide sequence of the E. coli rfb locus corresponding to an E. coli serotype, preferably selected from the list consisting of O1A, O2, O4, O6A, O8, O15, O16, O18A, O25B, and O75. As used herein, "O-antigen rfb locus" and "O-antigen rfb gene cluster" refer to a locus in a Gram-negative bacterial genome that comprises a cluster of genes that together encode the enzymatic machinery capable of synthesizing the O-antigen polysaccharide structure. The term rfb locus preferably refers to a genomic locus from the genus Escherichia, particularly E. coli. In certain embodiments, the O antigen rfb locus is heterologous to the host cell, e.g., has been introduced into, and preferably integrated into, the genome of a progenitor cell of the host cell. Preferably, the original rfb gene cluster, if present in the progenitor cell, is replaced in the host cell by the O antigen rfb gene cluster to allow for the production of bioconjugates of O antigen polysaccharides, preferably selected from the list consisting of E. coli O1A, O2, O4, O6A, O8, O15, O16, O18A, O25B, and O75 antigenic polysaccharides. Exemplary sequences of the rfb gene cluster (rfb locus) that can be used in production strains for bioconjugates with O-antigen polysaccharides of the E. coli serotypes listed above are shown below. In one exemplary and non-limiting embodiment, the rfb locus encoding the O-antigen polysaccharide of E. coli serotype O1A has the sequence of SEQ ID NO: 11 of WO 2020 / 191082. In one exemplary and non-limiting embodiment, the rfb locus encoding the O-antigen polysaccharide of E. coli serotype O2 has the sequence of SEQ ID NO: 12 of WO 2020 / 191082. In one exemplary and non-limiting embodiment, the rfb locus encoding the O-antigen polysaccharide of E. coli serotype O4 has the sequence of SEQ ID NO: 9 of WO 2020 / 191082. In one exemplary and non-limiting embodiment, the production strain comprising the rfb locus encoding the O-antigen polysaccharide of E. coli serotype O4 further comprises a nucleic acid encoding an E. coli O4-specific GtrS protein, e.g., having SEQ ID NO: 4 of WO 2020 / 191082. In one exemplary and non-limiting embodiment, the rfb locus encoding the O-antigen polysaccharide of E. coli serotype O6A has the sequence of SEQ ID NO: 13 of WO 2020 / 191082. In one exemplary and non-limiting embodiment, the rfb locus encoding the O-antigen polysaccharide of E. coli serotype O8 has the sequence of SEQ ID NO: 14 of WO 2020 / 191082. In one exemplary and non-limiting embodiment, the rfb locus encoding the O-antigen polysaccharide of E. coli serotype O15 has the sequence of SEQ ID NO: 15 of WO 2020 / 191082. In one exemplary and non-limiting embodiment, the rfb locus encoding the O-antigen polysaccharide of E. coli serotype O16 has the sequence of SEQ ID NO: 16 of WO 2020 / 191082. In one exemplary and non-limiting embodiment, the rfb locus encoding the O-antigen polysaccharide of Escherichia coli (E. coli) serotype O18A has the sequence of SEQ ID NO: 5 of WO 2022 / 208430, except that the Wzy O-antigen polymerase encoded in the rfb locus preferably contains an isoleucine at position corresponding to 199, a lysine at position corresponding to 377, and an alanine at position corresponding to 395 in the amino acid sequence of SEQ ID NO: 1 of WO 2022 / 208430. In one exemplary and non-limiting embodiment, the rfb locus encoding the O-antigen polysaccharide of E. coli serotype O25B has the sequence of SEQ ID NO: 18 of WO 2020 / 191082. In one exemplary and non-limiting embodiment, the rfb locus encoding the O-antigen polysaccharide of E. coli serotype O75 has the sequence of SEQ ID NO: 19 of WO 2020 / 191082.

[0056] The structures of the repeating units of the E. coli O-antigen polysaccharides O1A, O2, O4, O6A, O8, O15, O16, O18A, O25B, and O75 are shown in Table 1.

[0057] In some cases, the chain length of the O antigen can be manipulated by manipulating the native Wzz O antigen chain length regulator machinery, e.g., by overexpressing or supplementing with a Wzz O antigen chain regulator, e.g., by replacing the E. coli wzzB gene with one of its counterparts from a species within the Salmonella and Shigella range, or from P. aeruginosa, e.g., the Salmonella enterica counterpart, fepE, or another wzz homolog (see, e.g., U.S. Patent Application Publication No. 2018 / 0099038; WO 2020 / 039359), although, e.g., the number of repeat units of the O antigen can be increased with or without additional overexpression of the Wzy protein. The action of wzz-like genes is described in the literature for those species. However, neither is required to obtain bioconjugates with good immunogenicity, and in preferred embodiments, bioconjugates according to the present invention are prepared without manipulation of the Wzz chain length regulator machinery.

[0058] Carrier protein: The carrier protein is detoxified exoprotein A (EPA) of P. aeruginosa, also called recombinant EPA.

[0059] Regarding EPA, various detoxified protein variants have been described in the literature and can be used as carrier proteins. In certain embodiments, the EPA carrier protein used in the bioconjugates of the present invention is modified to reduce protein toxicity and / or increase susceptibility to glycosylation. For example, detoxification can be achieved by mutation and deletion of catalytically essential residues, L552V and ΔE553 (see, e.g., Lukac et al., 1988, Infect Immun, 56: 3095-3098; Ho et al., 2006, Hum Vaccin, 2:89-98). In a specific embodiment, the carrier protein used in producing the bioconjugates of the present invention is modified so that the number of glycosylation sites in the carrier protein is optimized in a manner that allows for a lower concentration of the administered protein, for example, in its bioconjugate form in an immunogenic composition. In a particular embodiment, the host cell encodes an EPA comprising 1 to 10, preferably 2 to 4, more preferably 4 glycosylation sites comprising the glycosylation consensus sequence Asn-X-Ser(Thr), where X can be any amino acid except Pro, more preferably the glycosylation consensus sequence Asp(Glu)-X-Asn-Z-Ser(Thr), where X and Z are independently selected from any amino acid except Pro. Thus, in a preferred embodiment, the carrier protein is a recombinant EPA comprising four N-glycosylation sites comprising the glycosylation consensus sequence Asn-X-Ser(Thr), preferably the glycosylation consensus sequence Asp(Glu)-X-Asn-Z-Ser(Thr). In one embodiment, the carrier protein comprises an amino acid sequence having at least 80, 85, 90, 95, 96, 97, 98, or 99% sequence identity to SEQ ID NO: 1 and comprises 1 to 10, preferably 2 to 4, preferably 4 glycosylation sites comprising the glycosylation consensus sequence Asn-X-Ser(Thr), more preferably the glycosylation consensus sequence Asp(Glu)-X-Asn-Z-Ser(Thr). In a preferred embodiment, the carrier protein is recombinant EPA comprising the amino acid sequence of SEQ ID NO: 1.In a further preferred embodiment, the host cell comprises a nucleotide sequence encoding a carrier protein comprising SEQ ID NO:1.

[0060] Metabolic machinery for N-glycosylation of EPA: Gram-negative bacterial host cells preferably contain an oligosaccharyltransferase (OST) that transfers an oligosaccharide to an N-glycosylation site on a carrier protein. Thus, preferably, the host cells provided herein contain a nucleotide sequence encoding an oligosaccharyltransferase (OST). As used herein, an oligosaccharyltransferase is an enzyme that transfers a lipid-linked oligosaccharide to a residue of a nascent polypeptide chain that contains a glycosylation consensus motif, for example, to an asparagine (Asn, N) residue of a nascent polypeptide chain that contains an N-glycosylation consensus motif. Examples of such N-glycosylation consensus motifs include motifs with the amino acids Asn-X-Ser(Thr) or (Asp(Glu)-X-Asn-Z-Ser(Thr)). Throughout this application, it should be understood that with respect to the N-glycosylation motif Asn-X-Ser(Thr), X can be any amino acid other than proline, and with respect to the N-glycosylation motif Asp(Glu)-X-Asn-Z-Ser(Thr), both X and Z can be any amino acid other than proline. Preferably, such an oligosaccharyltransferase transfers an oligosaccharide to an asparagine residue of the glycosylation consensus sequence Asp(Glu)-X-Asn-Z-Ser(Thr) in the polypeptide chain of a carrier protein described herein. Nucleic acids encoding oligosaccharyltransferases can be native to the host cell or introduced into the host cell using genetic approaches. In a preferred embodiment, the oligosaccharyltransferase is heterologous to the host cell. Because E. coli does not naturally contain oligosaccharyltransferases, when E. coli is used as a host cell for production of bioconjugates, a heterologous oligosaccharyltransferase is included in such host cells, e.g., upon introduction by genetic engineering. The oligosaccharyltransferase may be from any source known in the art in view of the present disclosure. In certain preferred embodiments, the oligosaccharyltransferase is an oligosaccharyltransferase from Campylobacter. For example, in one embodiment, the oligosaccharyltransferase is an oligosaccharyltransferase from Campylobacter jejuni (i.e., PglB; see, e.g., Wacker et al., 2002, Science 298:1790-1793; see also, e.g., NCBI Gene ID: 3231775, UniProt Accession No. O86154). In other embodiments, the oligosaccharyltransferase is an oligosaccharyltransferase from Campylobacter lari (see, e.g., NCBI Gene ID: 7410986). In specific embodiments, the oligosaccharyltransferase is PglB from Campylobacter jejuni, including the native (wild-type) protein or any variant thereof, such as those described in WO 2016 / 107818 and WO 2016 / 107819. PglB can transfer lipid-linked oligosaccharides to asparagine residues in the consensus sequences Asn-X-Ser(Thr) and Asp(Glu)-X-Asn-Z-Ser(Thr). In certain embodiments, the PglB oligosaccharyltransferase is a polypeptide having oligosaccharyltransferase activity as defined herein, wherein the polypeptide comprises an amino acid sequence having at least 80, 85, 90, 95, 96, 97, 98, or 99% sequence identity to SEQ ID NO:2. In a preferred embodiment, the PglB oligosaccharyltransferase comprises an amino acid sequence identical to SEQ ID NO:2, or an amino acid sequence identical to SEQ ID NO:2 and containing designated amino acid changes specific for optimizing expression of the bioconjugate for a particular E. coli O-antigen serotype (see, e.g., WO 2020 / 191088 and further examples hereinbelow). Thus, in a preferred embodiment, the host cell comprises a nucleotide sequence encoding an oligosaccharyltransferase comprising SEQ ID NO:2, or a variant thereof described herein. In certain embodiments, one or more endogenous glycosylation consensus sequences in wild-type PglB have been mutated to avoid PglB autoglycosylation, e.g., SEQ ID NO:2 containing the mutation N534Q.Examples of variant PglB suitable for use in the host cells provided herein include the PglB of SEQ ID NO: 2 (e.g., for EPA bioconjugates comprising E. coli O-antigens of serotype O2, O8, O18A, or O25B); the PglB of SEQ ID NO: 2 with the mutation N311V; the PglB of SEQ ID NO: 2 with the mutations N311V, K482R, D384H, and A669V (e.g., for EPA bioconjugates comprising E. coli O-antigens of serotype O1A or O15); the PglB of SEQ ID NO: 2 with the mutations Y77H and N311V (e.g., for EPA bioconjugates comprising E. coli O-antigens of serotype O4); the PglB of SEQ ID NO: 2 with the mutations Y77H, S80R, Q287P, K289R, and N311V (e.g., for EPA bioconjugates comprising E. coli O-antigens of serotype O6A or O16). and PglB of SEQ ID NO: 2 containing mutations Y77R and N311V (e.g., for a bioconjugate of EPA containing an Escherichia coli (E. coli) O-antigen of serotype O75).

[0061] As will be understood by those skilled in the art, the polypeptides of the present invention contained in the host cells of the present invention, e.g., carrier proteins, oligosaccharyltransferases, and enzymes of the rfb gene cluster, including polypeptides with Wzy O antigen polymerase activity, as provided, for example, in WO 2020 / 191088 or WO 2022 / 208430, are conveniently provided to the host cells in the form of nucleic acids encoding the polypeptides of the present invention. The nucleic acids encoding the polypeptides of the present invention are preferably nucleic acid constructs, specifically expression constructs comprising one or more expression cassettes for expressing the polypeptides of the present invention, provided that the nucleotide sequence encoding the polypeptide of the present invention is operably linked to expression control sequences for expression of the polypeptide in the host cells of the present invention. Suitable expression control sequences typically include at least a promoter and may include additional sequence elements, such as, in particular, a Shine-Dalgarno sequence and a transcription termination sequence. The promoter may be a constitutive promoter or a promoter whose activity can be regulated, e.g., repressed or induced in response to certain conditions, such as temperature changes or the presence of certain chemicals or proteins in the cell, any of which are themselves well known in the art. Non-limiting examples of suitable promoters are the ara, phoA, tac, tet, trc, trp, PBAD, λPL, T5, or T7 promoter. The nucleic acid construct can be extrachromosomal, for example, a plasmid or other vector, or the nucleic acid construct can be integrated into the genome of the host cell of the present invention. Molecular biology methods for constructing and / or synthesizing nucleic acid constructs for expression of polypeptides in the host cells of the present invention are generally well known in the art. In certain embodiments, the rfb locus is integrated into the genome of the host cell. In certain embodiments, the nucleic acid encoding the oligosaccharyltransferase and the nucleic acid encoding the EPA carrier protein are present on a plasmid in the host cell.

[0062] Step ii: In the second step, the O-EPA bioconjugate is purified by anion exchange chromatography (AEX1; step ii). This step is used to reduce the processing volume and process-related impurities derived from the fermentation process, i.e., incubation of host cells and extraction of the periplasmic fraction containing the O-EPA bioconjugate. The main task of AEX1 is to remove most DNA, endotoxins, and host cell proteins (HCPs). In particular, impurities that may lead to degradation of the O-EPA conjugate, such as proteases, peptidases, and glycosidases, are removed. Therefore, the AEX1 step is specifically a capture step.

[0063] In certain embodiments, AEX1 is performed using a strong anion exchange resin. In certain embodiments, the resin is, for example, a ceramic resin that has rigid, incompressible characteristics and a high dynamic binding capacity at high flow rates. One non-limiting example of a resin that is particularly suitable for the AEX1 step is Q Ceramic HyperD F resin. In certain embodiments, the AEX1 step is performed in bind-elute mode.

[0064] Step ii comprises subjecting the optionally adjusted FPF load to a first anion exchange chromatography (AEX1) step to obtain a first AEX eluate (AEX1). This step is known per se. Typically, adjustment of the filtered periplasmic fraction load prior to AEX1 is not required. Instead, the filtered periplasmic fraction is typically directly exposed to AEX1, as long as the conditions, particularly the conductivity, of the filtered periplasmic fraction are suitable for exposure to AEX1. Therefore, adjustment of the FPF load prior to AEX1 is optional. In certain embodiments, step ii further comprises one or more of the following steps (ii-1) to (ii-3), preferably in the order shown:

[0065] In one embodiment, step ii comprises (ii-1) contacting the optionally adjusted, filtered periplasmic fraction load with AEX1 medium and washing the medium containing the bound O-EPA bioconjugate with a wash buffer. Typically, the wash buffer has a relatively low salt concentration and a relatively low conductivity. An example of a suitable low-salt, low-conductivity wash buffer is a buffer containing 10 mM Tris and 50 mM NaCl at pH 8.5.

[0066] In one embodiment, step ii comprises (ii-2) eluting the O-EPA bioconjugate with an elution buffer, typically having a relatively high salt concentration and a relatively high conductivity. An example of a suitable high-salt, high-conductivity buffer is a buffer containing 10 mM or 50 mM Tris and 1 M NaCl at pH 8.5. Typically, elution is performed using a step gradient.

[0067] For each of these steps, the nature of the AEX1 resin, AEX1 column or membrane or cassette format, the exact buffer components, pH and / or salt concentration, or conductivity, and gradients can obviously be varied and would be known to one skilled in the art based on this disclosure.

[0068] In one embodiment, step ii includes (ii-3) optionally pooling fractions containing enriched O-EPA bioconjugate content to obtain an AEX1 eluate. Typically, fractions containing enriched O-EPA conjugate content are identified by SDS-PAGE analysis, or any other analysis suitable for determining the amount or relative purity of the desired O-EPA conjugate. Performing SDS-PAGE analysis to identify fractions containing the intended product, i.e., O-EPA bioconjugate, is well known to those skilled in the art. Instead of pooling fractions based on their content, it is also possible to simply collect larger fractions known to contain the product during elution, especially if the process steps are established and knowledge of the product's elution profile is available.

[0069] Step iii: In the third step, the O-EPA bioconjugate is further purified by subjecting the optionally adjusted AEX1 eluate load to a mixed-mode chromatography (MMC) step using a multimodal resin (MMR), which contains both anion exchange and hydrophobic functionality, to obtain an MMR eluate. This chromatography step further removes process-related impurities, particularly endotoxins and host cell proteins (HCPs). Surprisingly, MMR containing both anion exchange and hydrophobic functionality has been found to be particularly suitable for the separation of the bioconjugates described herein. Importantly, as shown herein, the use of this step made it possible to obtain O-EPA bioconjugate compositions with increased levels of multiple glycosylation compared to previously preferred methods. Furthermore, the use of this MMR reduces the overall buffer consumption of the entire purification process, which is considered beneficial due to reduced costs and reduced environmental impact (less waste). This is particularly important for large-scale production methods, e.g., incubation of host cells in bioreactors having volumes between 100 L and 20,000 L, e.g., 150 L to 5,000 L, e.g., 200 L, 800 L, or 2,000 L. A variety of MMRs containing anion exchange functionality and hydrophobic functionality (i.e., containing cationic and hydrophobic groups) are commercially available; non-limiting examples include PPA HyperCel, HEA HyperCel, MEP HyperCel, and Capto adhere. In a specific embodiment, the MMR is Capto adhere.

[0070] In one preferred embodiment, the MMR is of formula (I):

[0071] [ka] (In the formula, RES represents the stationary phase resin; R 1 represents C1-C4 alkyl, preferably methyl; R2 represents C1-C4 alkyl substituted with phenyl, tolyl or xylyl, preferably benzyl; R 3 represents a C1-C4 alkyl substituted with a hydroxyl group or a thiol group, preferably 2-hydroxyethyl) The ligands include: An example of a suitable ligand of formula (I) is N-benzyl-N-methylethanolamine. In a preferred embodiment, the MMR is a compressible resin. The compressible resin described herein has the advantage of being relatively resistant to attrition during slurry preparation (mixing the resin with an elution buffer or wash buffer) compared to rigid resins. Higher resistance to attrition results in higher reusability of the resin. Therefore, the compressible MMR described herein is economically advantageous. Suitable compressible MMR containing both anion exchange and hydrophobic functionality are commercially available, for example Capto adhere. Adjustment of the AEX1 eluate load prior to MMC is optional. Typically, adjustment of the AEX1 eluate load prior to the MMC step is not necessary. Instead, the AEX1 eluate is typically exposed directly to MMC. In an alternative embodiment, the AEX1 eluate load is adjusted. For example, the pH and conductivity are adjusted to predetermined values ​​by the addition of an appropriate buffer, e.g., to pH 7.2±0.2 using a buffer containing 50 mM BisTris, pH 6.0, followed in a subsequent step by adjustment to a target conductivity of 7.5-9 mS / cm using a buffer containing 10 mM BisTris, pH 7.0. In a preferred embodiment, the MMC step is performed in bind-elute mode. Step iii is performed after step ii.

[0072] In certain embodiments, step iii further comprises one or more of the following steps (iii-1) to (iii-4), preferably in the order shown:

[0073] In one embodiment, step iii comprises (iii-1) performing particle reduction filtration 1. This step results in the removal of viable and non-viable particles.

[0074] In one embodiment, step iii comprises (iii-2) contacting the optionally conditioned AEX1 eluate with an MMR and subjecting the MMR containing the bound O-EPA bioconjugate to several washing steps. Typically, in the first washing step, the MMR is washed with a buffer having a relatively low conductivity. An example of a suitable low-conductivity washing buffer is a buffer containing 50 mM BisTris and 50 mM NaCl at a pH of 6.0. Typically, in the second washing step, the MMR is washed with a buffer having a higher conductivity and a lower pH than the first washing buffer. An example of a suitable second washing buffer is a buffer containing 50 mM acetate and 250 mM NaCl at a pH of 4.7. Typically, in the third wash step, the MMR is washed again with a buffer having a relatively low conductivity, e.g., the low-conductivity wash buffer used in the first wash step, e.g., a buffer containing 50 mM BisTris, 50 mM NaCl, pH 6.0.

[0075] In one embodiment, step iii comprises eluting the (iii-3) O-EPA bioconjugate with a buffer having a relatively high conductivity. An example of a suitable high-conductivity elution buffer is an elution buffer comprising 50 mM BisTris, 2 M NaCl, pH 6.0.

[0076] In one embodiment, step iii comprises (iii-4) pooling fractions containing enriched O-EPA bioconjugate content to obtain an MMR eluate. In a preferred embodiment, to obtain an MMR eluate containing enriched O-EPA bioconjugate content, fixed volumes of fractions are pooled rather than pooling based on SDS-PAGE analysis of individual fractions. Of course, rather than pooling fractions, it is instead possible to simply collect a fraction known to contain product during elution, especially if process steps are established and knowledge of the product elution profile is available.

[0077] For each of these steps, the nature of the MMR, the column format, the exact buffer composition, pH and / or salt concentration, or conductivity, and gradients are clearly variable and known to those skilled in the art based on this disclosure.

[0078] Step iv: In the fourth step, the O-EPA bioconjugate is further purified by subjecting the conditioned MMR eluate load to a hydrophobic interaction chromatography (HIC) step to obtain a bioconjugate-containing HIC eluate. The HIC step is used to further remove HCPs, particularly non-glycosylated EPA and various medium-sized HCPs (approximately 40-60 kDa).

[0079] In certain embodiments, step iv "subjecting the adjusted load of the MMR eluate obtained in step (iii) to a hydrophobic interaction chromatography (HIC) step to obtain a HIC eluate" further comprises one or more of the following steps (iv-1) to (iv-5), preferably in the order shown:

[0080] In one embodiment, step iv includes (iv-1) adjusting the conductivity of the MMR eluate load. The conductivity is adjusted using a loading buffer that is suitable for binding to the HIC medium and typically has a relatively high conductivity. One example of a suitable loading buffer is a 2 M potassium phosphate buffer at pH 7.0. Typically, the MMR eluate load is adjusted by adding a weight of loading buffer that is three times the weight of the MMR eluate. A non-limiting example of a hydrophobic interaction medium that can be used in the HIC of the present invention is Sartobind Phenyl Adsorbent, e.g., Sartobind Phenyl Jumbo 5L Capsules.

[0081] In one embodiment, step iv includes (iv-2) performing particle reduction filtration 2 to remove viable and non-viable particles.

[0082] In one embodiment, step iv comprises (iv-3) contacting the conditioned MMR eluate with an HIC medium and washing the medium containing the bound O-EPA bioconjugate with an appropriate buffer, typically a buffer having a relatively high conductivity, for example, a buffer containing 2 M potassium phosphate, pH 7.0, optionally followed by a second washing step with a buffer of reduced conductivity (for example, a mixture containing 70% of a buffer containing 2 M potassium phosphate, pH 7.0, and 30% WFI water).

[0083] In one embodiment, step iv comprises (iv-4) eluting the O-EPA bioconjugate using a suitable elution buffer, typically an elution buffer with a relatively low conductivity. Preferably, elution is performed by a step gradient using a suitable elution buffer. An example of a suitable buffer for eluting the O-EPA bioconjugate is a mixture containing 30% 2 M potassium phosphate buffer at pH 7.0 and 70% WFI water.

[0084] For each of these steps, the nature of the HIC resin, column or membrane or cassette format, the exact buffer composition, pH and / or salt concentration, or conductivity, and gradient can obviously be varied and would be known to one of skill in the art based on this disclosure.

[0085] In one embodiment, step iv comprises (iv-5) pooling fractions containing enriched O-EPA conjugate content to obtain an HIC eluate. It will be apparent to those skilled in the art that instead of pooling fractions, it is also possible to simply collect larger fractions known to contain product during elution, especially if process steps are established and knowledge of the product elution profile is available.

[0086] Step v: In the fifth step, the O-EPA bioconjugate is further purified by subjecting the adjusted load of the HIC eluate obtained in step (iv) to a second anion exchange chromatography (AEX2) step to obtain a second AEX eluate as the product. The AEX2 step is used to further remove process-related protein impurities, including unglycosylated EPA, short-chain glycosylated EPA (sEPA; i.e., an EPA carrier protein with a polysaccharide chain containing only about 1-3 repeating units), and, in particular, Escherichia coli (E. coli) transaldolase B (approximately 37 kDa). Thus, in one aspect, the present invention provides the use of a final purification step by anion exchange chromatography to reduce the amount of transaldolase B in preparations containing the O-EPA bioconjugate. In certain embodiments, the O-EPA bioconjugate undergoes a prior purification step that includes anion exchange, MMR, and hydrophobic interaction chromatography, in that order.

[0087] AEX media suitable for the AEX2 step are commercially available; one non-limiting example is Source Q resin.

[0088] In certain embodiments, step (v) further comprises one or more of the following steps (v-1) to (v-4), preferably in the order shown:

[0089] In one embodiment, step (v) includes (v-1) performing tangential flow filtration 1 (TFF1) of the HIC eluate load to reduce the conductivity and pH. The TFF1 step is a diafiltration step. In this step, new buffer is added to the feed at the same rate as the permeate flow rate, i.e., the rate at which the feed passes through the membrane. The HIC eluate is then obtained in a suitable buffer, and the volume of the HIC eluate is kept constant (buffer exchange). A typical suitable diafiltration volume is, for example, 5 to 6 times the volume of the HIC eluate, but it can be adjusted if necessary by those skilled in the art, for example, by routine testing of the conductivity and pH of the retentate. A suitable buffer for such buffer exchange is, for example, a buffer containing 10 mM BisTris and 50 mM NaCl, pH 6.0.

[0090] In one embodiment, step (v) includes (v-2) performing particle reduction filtration 3 to remove viable and non-viable particles.

[0091] In one embodiment, step (v) comprises contacting the (v-3) conditioned, filtered HIC eluate load with AEX2 medium and eluting the O-EPA bioconjugate with an elution buffer. In certain embodiments, elution is performed by a step gradient, particularly a linear gradient of increasing salt concentration. Typically, the elution buffer has a relatively high salt concentration and high conductivity. A non-limiting example of a suitable process for eluting the O-EPA bioconjugate is applying 21% of a buffer (Buffer V) containing 10 mM BisTris, 200 mM NaCl, pH 6.0 in Buffer U (Buffer U containing 10 mM BisTris, 50 mM NaCl, pH 6.0) over 7.5 CV (step gradient), followed by a linear gradient of 21 to 56% of Buffer V in Buffer U over 7 to 8, e.g., 7.5 CV.

[0092] In one embodiment, step (v) comprises (v-4) pooling fractions containing enriched O-EPA bioconjugate content to obtain AEX2 eluate as product.

[0093] In one embodiment, during step (v), the O-EPA bioconjugate binds to the AEX2 matrix and is eluted with a step gradient preceded by a linear gradient of increasing salt concentration, as described above, to obtain O-EPA bioconjugate with a purity of at least 90%, preferably at least 95%, more preferably at least 98%, or at least 99%.

[0094] For each of these steps, the nature of the AEX2 resin, AEX2 column or membrane or cassette format, the exact buffer components, pH and / or salt concentrations, or conductivity, gradients, and fraction pooling strategies are clearly variable and known to those skilled in the art based on this disclosure.

[0095] As outlined above: In certain embodiments, AEX1 is performed in bind-elute mode. In certain embodiments, MMC is performed in bind-elute mode. In certain embodiments, HIC is performed in bind-elute mode. In certain embodiments, AEX2 is performed in bind-elute mode. In certain embodiments, AEX1, MMC, HIC and AEX2 are performed in bind-elute mode.

[0096] According to the above protocol, in each of the second, third, fourth and fifth steps, i.e., in each of the chromatography steps (steps ii-v), the relative amount of O-EPA conjugates to total protein is higher in the collected eluate than in the load.

[0097] In certain embodiments, the production method further comprises an additional step (vi), in which the load of the AEX2 eluate is adjusted to a pharmaceutically acceptable buffer and concentration, whereupon purified O-EPA conjugate is obtained as a pharmaceutical drug substance.

[0098] In certain embodiments, step (vi) further comprises one or more of the following steps (vi-1) to (vi-3), preferably in the order shown:

[0099] In one embodiment, step (vi) comprises (vi-1) performing tangential flow filtration 2 (TFF2) of the AEX2 eluate to convert it into a pharmaceutically acceptable buffer and concentration of O-EPA bioconjugate. In one embodiment, TFF2 is performed with an excipient buffer containing 6.19 mM KH2PO4, 3.81 mM Na2HPO4, 5% (w / w) sorbitol, 10 mM methionine, pH 7.0. In one embodiment, following TFF2, polysorbate-80 is added at a concentration of 0.02% (w / w) to obtain purified O-EPA bioconjugate in a pharmaceutically acceptable buffer. Thus, in one embodiment, a pharmaceutically acceptable buffer comprises 6.19 mM KH2PO4, 3.81 mM Na2HPO4, 5% (w / w) sorbitol, 10 mM methionine, 0.02% (w / w) polysorbate-80 at pH 7.0. A typically suitable diafiltration volume for TFF2 is, for example, 5-6 times the volume of the AEX2 eluate, although it can be adjusted if necessary by one of skill in the art following, for example, routine testing of the conductivity and pH of the retentate.

[0100] Optionally, the pH of the AEX2 eluate load is adjusted prior to TFF2. A suitable pH for TFF2 is, for example, 6.5±0.2, and one example of a suitable buffer is 100 mM NaHPO buffer (Buffer W). Typically and preferably, the pH of the AEX2 eluate load is not adjusted prior to TFF2 (in which case the pH is typically about 6.0±0.2).

[0101] In one embodiment, step (vi) comprises (vi-2) performing bioburden filtration of the purified O-EPA bioconjugate to obtain a purified O-EPA bioconjugate drug substance. Such bioburden filtration can be performed, for example, using a filter having a PES membrane with a 0.45+0.2 μm cutoff, such as a Sartopore2 capsule size 9 filter.

[0102] In one embodiment, step (vi) comprises (vi-3) distributing and freezing the purified O-EPA bioconjugate drug substance to obtain a bulk drug substance.

[0103] In certain embodiments, the production method further comprises an additional step (vii) in which several purified O-EPA bioconjugate drug substances are combined to obtain a multivalent drug product. In certain embodiments, the multivalent pharmaceutical product comprises at least four, five, six, seven, eight, nine, or ten purified O-EPA bioconjugates comprising O-antigen polysaccharides selected from the list consisting of E. coli O serotypes O1A, O2, O4, O6A, O8, O15, O16, O18A, O25B, and O75. In certain embodiments, the multivalent pharmaceutical product comprises purified O-EPA bioconjugates comprising O-antigen polysaccharides selected from the list consisting of E. coli O serotypes O1A, O2, and O25B. In certain embodiments, the multivalent pharmaceutical comprises at least a purified O-EPA bioconjugate comprising the O-antigen polysaccharide of Escherichia coli (E. coli) O serotype O1A. In certain embodiments, the multivalent pharmaceutical comprises at least a purified O-EPA bioconjugate comprising the O-antigen polysaccharide of Escherichia coli (E. coli) O serotype O2. In certain embodiments, the multivalent pharmaceutical comprises at least a purified O-EPA bioconjugate comprising the O-antigen polysaccharide of Escherichia coli (E. coli) O serotype O25B. In certain embodiments, the multivalent pharmaceutical comprises at least purified O-EPA bioconjugates comprising O-antigen polysaccharides of Escherichia coli (E. coli) O serotypes O1A, O2, O6A, and O25B. In certain embodiments, the multivalent pharmaceutical product comprises at least four, preferably at least eight, and more preferably nine purified O-EPA bioconjugates comprising O-antigen polysaccharides selected from the list consisting of E. coli O serotypes O1A, O2, O4, O6A, O8, O15, O16, O18A, O25B, and O75, preferably E. coli O serotypes O1A, O2, O4, O6A, O15, O16, O18A, O25B, and O75. In further embodiments, conjugates (i.e., O-antigen polysaccharides covalently linked to a carrier protein) of different E. coli serotypes can be added to obtain a multivalent pharmaceutical product comprising, for example, 10 to 20 conjugates, e.g., O-EPA conjugates. Such conjugates of different E. coli serotypes can also be bioconjugates and may also be purified by the methods described herein. In a specific embodiment, a nonavalent pharmaceutical product is provided that includes up to nine purified O-EPA bioconjugates according to the invention, wherein the O antigen polysaccharides in the nonavalent pharmaceutical product consist of Escherichia coli (E. coli) O serotypes O1A, O2, O4, O6A, O15, O16, O18A, O25B, and O75. In a specific embodiment, a decavalent pharmaceutical product is provided that includes up to ten purified O-EPA bioconjugates according to the invention, wherein the O antigen polysaccharides in the decavalent pharmaceutical product consist of Escherichia coli (E. coli) O serotypes O1A, O2, O4, O6A, O8, O15, O16, O18A, O25B, and O75.

[0104] In any embodiment, additional filtration steps, such as tangential flow filtration, ultrafiltration / diafiltration (variants of tangential flow filtration), dead-end filtration, and / or sterile filtration, as well as optional concentration and / or dilution and / or buffer exchange steps of the O-EPA bioconjugate, can be added at certain points in the method, but preferably no additional chromatography steps are included, and thus the method preferably does not include more than four chromatography steps. Preferably, the method according to the present invention does not include a size-exclusion or hydroxyapatite chromatography step for purification of the bioconjugate. Of course, additional purification steps, such as chromatography steps, can be added to the method of the present invention to obtain even higher purity, but this is undesirable as it would inevitably reduce the overall yield and increase the complexity and material consumption of the method, making the method economically unfeasible for large-scale production of biopharmaceuticals. However, the method of the present invention already provides the purity level desired for pharmaceuticals suitable for human administration at yields that make the method of the present invention economically and practically feasible.

[0105] Further details regarding the process steps and terminology used are provided below.

[0106] Anion Exchange Chromatography (AEX): Generally, in ion exchange chromatography, binding is based on electrostatic charge. In AEX, the resin has positively charged functional groups, so that sample components with negatively charged functional groups bind to the resin. As known to those skilled in the art, AEX can be performed using weak or strong anion exchangers. In certain embodiments of the present invention, strong anion exchange resins are used in the AEX steps (AEX1 and AEX2). A non-limiting example of a functional group suitable for AEX resins is a quaternary ammonium group. Such resins are commercially available, including, for example, Q Ceramic HyperD F resin and Source 15Q resin. Based on this disclosure and general knowledge, those skilled in the art will know how to diversify the different available AEX resins for use in the methods of the present invention. As a non-limiting example, AEX1 can be suitably performed using Q Ceramic HyperD F resin, and AEX2 can be suitably performed using Source 15Q resin. With increasing salt concentration, salt ions in the elution buffer compete with the resin-bound material for binding, displacing and eluting the bound material. Alternatively, as the pH is changed, the bound protein is titrated and eventually becomes uncharged or has the same charge as the functional groups on the resin, resulting in the repulsion and elution of the bound protein.

[0107] Mixed-mode chromatography: The term mixed-mode chromatography (MMC) is known in the art and is used herein synonymously with multimodal chromatography. The abbreviation MMC should not be confused with the commercially available product "Capto MMC Multimodal Chromatography Resin" (Cytiva). This term refers to a chromatographic method that utilizes two or more types of interactions between the stationary phase and the target protein, e.g., bioconjugates, to achieve separation from impurities. Thus, in MMC, binding to the multimodal resin (MMR) and separation of proteins depend on at least two different types of interactions. In some cases, e.g., MMRs containing both anion-exchange and hydrophobic functionality, binding is determined by a combination of electrostatic and hydrophobic interactions. A non-limiting example of such a resin is an MMR containing a ligand of Formula I, e.g., N-benzyl-N-methylethanolamine. Such resins may be particularly compressible resins and are commercially available, e.g., Capto Adhere multimodal resin. Additionally, hydrogen bonding may contribute to binding. However, this type of MMC is distinctly different from hydroxyapatite chromatography (HA), which is sometimes considered a specific form of mixed-mode chromatography, such as ceramic HA (cHA). With HA (or cHA), separation is achieved, inter alia, through a combination of ionic and metal affinity interactions (e.g., CHT ceramic hydroxyapatite XT resin). Proteins typically bind to HA via HA-phosphoryl (cation exchange) or HA-calcium interactions (metal affinity). Small, basic proteins usually bind to HA via phosphoryl cation exchange, while acidic proteins typically interact primarily through calcium affinity. Large proteins frequently bind to HA using both mechanisms. However, despite being a suitable stationary phase for a wide range of applications, including the purification of bioconjugates, HA (or cHA) does not provide the excellent results for this application of bioconjugate purification obtained using MMR according to the present invention. For example, the use of HA results in high costs, and furthermore, the MMR of the present invention importantly and unexpectedly results in improved multiglycosylation yields in bioconjugate compositions prepared by the methods of the present invention compared to similar methods in which cHA is used instead of MMR [see WO 2022 / 214620].

[0108] Hydrophobic Interaction Chromatography (HIC): In HIC, a target protein, e.g., an O-EPA bioconjugate, is separated from impurities based on its hydrophobicity. A target protein containing hydrophobic and hydrophilic regions is typically applied to an HIC column in a high-salt buffer. The salt in the buffer reduces the solvation of the target protein. Reduced solvation allows the exposed hydrophobic regions to adsorb to the HIC resin. The more hydrophobic the molecule, the less salt is required to promote binding. Typically, a decreasing salt gradient is used to elute samples from the column in order of decreasing hydrophobicity. Sample elution can be further assisted by the addition of additional components, such as surfactants, to the elution buffer. A non-limiting example of an HIC capsule suitable for performing the HIC step described above is the Sartobind Phenyl Jumbo 5L Capsule. Those skilled in the art can vary HIC resins, and various commercially available HIC resins can be used according to the present invention and based on this disclosure.

[0109] Bioconjugate: This term is discussed above. Specifically, bioconjugates are glycoconjugates prepared in host cells, where the host cell machinery produces a glycan and a carrier protein and attaches the glycan to the carrier protein, for example, via an asparagine or arginine N-linkage. A particularly preferred host cell for producing bioconjugates is Escherichia coli (E. coli), preferably containing nucleic acids encoding (i) a carrier protein, (ii) an oligosaccharyltransferase, such as C. jejuni PglB, capable of covalently attaching an O-antigen polysaccharide to an asparagine (Asn) residue in the glycosylation consensus sequence (Asn-X-Ser(Thr), where X can be any amino acid except Pro) in the carrier protein via N-linked glycosylation, and (iii) an rfb gene cluster encoding an enzyme responsible for producing the O-antigen polysaccharide of the desired serotype. By creating host cells with different rfb loci, different bioconjugates can be prepared, for example, containing O-antigen polysaccharides from different E. coli or Shigella serotypes. Cultivation of such host cells produces bioconjugates in the periplasm of the host cells, containing carrier proteins to which O-antigen polysaccharides encoded by the rfb locus are covalently attached. More detailed descriptions of the production of bioconjugates in such host cells can be found, for example, in WO 2009 / 104074, WO 2015 / 124769, WO 2017 / 035181, or WO 2020 / 191082. Optimized variants of PglB oligosaccharyltransferase for the production of bioconjugates of specific E. coli O-antigens are described in WO 2020 / 191088. The present invention addresses new and improved methods for purifying bioconjugates produced from such host cells, as well as bioconjugate compositions obtainable by or obtained from such host cells. Host cells for the production of bioconjugates are typically bacterial cells, preferably Gram-negative bacterial cells; in a preferred embodiment, the host cell is Escherichia coli (E. coli).O-EPA bioconjugates therefore need to be purified from E. coli host cell proteins. One example of a host cell protein is transaldolase B, particularly E. coli transaldolase B, and the methods described in this invention can yield O-EPA preparations that contain very small amounts of host cell proteins, surprisingly including very small amounts of transaldolase B, which appears to be one of the most abundant remaining host cell proteins that is difficult to remove from O-EPA bioconjugates. Host cells are typically engineered to express bioconjugates in the periplasm; therefore, a convenient starting point for purifying O-EPA bioconjugates is the periplasmic fraction of host cells, e.g., from Gram-negative host cells, e.g., E. coli host cells. Particularly useful bioconjugates include carrier proteins having one or more polysaccharides attached thereto. Such bioconjugates are used, for example, as the active ingredient of specific vaccines aimed at inducing a functional immune response against the polysaccharides of the bioconjugate. In an embodiment of the invention, the bioconjugate comprises a carrier protein and one or more polysaccharides covalently attached to the carrier protein, preferably 1 to 4 polysaccharides covalently attached to the carrier protein. In an embodiment of the invention, the bioconjugate is a conjugation product containing an Escherichia coli (E. coli) O-antigen polysaccharide covalently linked to a carrier protein. In an embodiment of the invention, the bioconjugate is a conjugation product containing a Shigella O-antigen polysaccharide covalently linked to a carrier protein. The term O-antigen is known in the art, is used in its normal context, and should not be confused with O-linkage. In an exemplary embodiment, the O-antigen polysaccharide is N-linked to the carrier protein. The term O-antigen polysaccharide generally refers to the repeating glycan polymer contained within the LPS of bacteria, e.g., E. coli. The E. coli O-antigen is a polymer of immunogenic repeating oligosaccharides (typically 1-40 repeating units, e.g., 5-30 repeating units) and is typically used for serotyping and glycoconjugate vaccine production.

[0110] Carrier protein: This term has been discussed above. A particularly suitable carrier protein in the context of the present invention is detoxified Pseudomonas aeruginosa exotoxin A (EPA) (the terms P. aeruginosa exotoxin A and exoprotein A, or EPA, are used interchangeably). In certain embodiments, the carrier protein is detoxified P. aeruginosa exotoxin A. With regard to EPA, various detoxified protein variants have been described in the literature and can be used as carrier proteins. For example, detoxification can be achieved by mutation and deletion of catalytically essential residues, L552V and ΔE553. Preferably, the EPA carrier protein comprises 1 to 20, preferably 1 to 10, preferably 2 to 4 glycosylation sites. In certain embodiments, EPA contains four glycosylation sites having the amino acid sequence Asn-X-Ser(Thr), preferably Asp(Glu)-X-Asn-Z-Ser(Thr). For descriptions of examples of bioconjugation of Escherichia coli (E. coli) O-antigen polysaccharide to an EPA carrier protein, see, e.g., WO 2015 / 124769, WO 2017 / 035181, or WO 2020 / 191082, or for descriptions of examples of bioconjugation of Shigella dysenteriae (Shigella) O-antigen polysaccharide to an EPA carrier protein, see, e.g., WO 2009 / 104074. In a non-limiting preferred embodiment, the carrier protein of a bioconjugate according to the present invention comprises SEQ ID NO: 1. In certain embodiments, the carrier protein contains a signal sequence that targets the carrier protein to the periplasmic space during expression in a host cell. A variety of signal sequences may be used. In one non-limiting embodiment, the signal sequence comprises SEQ ID NO: 3. The signal sequence may be cleaved after translocation of the protein into the periplasm and therefore may be absent in the final carrier protein of the bioconjugate.

[0111] Polysaccharide: This term is discussed above. Suitable polysaccharides include repeating units n of 1 to 100, e.g., 1 to 50, 1 to 40, 1 to 30, 1 to 20, and 1 to 10, 3 to 50, 3 to 40, e.g., at least 5, e.g., 5 to 40, e.g., 7 to 30, e.g., 7 to 25, e.g., 5 to 20, e.g., 10 to 20. Such repeating units contain, i.e., comprise or consist of, (i) unmodified monosaccharides and / or (ii) modified monosaccharides. The term "modified monosaccharide" in non-limiting embodiments includes N-acetylation, O-acetylation, amidation, and / or amination of monosaccharides. Such modified monosaccharides can include zero, one, or more modifications, e.g., zero, one, two, or three of the aforementioned modifications, in the same monosaccharide. In certain embodiments, the modified monosaccharide is an O-acetylated and / or N-acetylated monosaccharide, particularly a monosaccharide comprising one O-acetylation or N-acetylation.

[0112] In an embodiment of the invention, suitable repeating units comprise monosaccharides selected from the group consisting of mannose, rhamnose, glucose, fucose, galactose, modified mannose, modified rhamnose, modified glucose, modified fucose, and modified galactose. In an embodiment of the invention, the O polysaccharide is specific for Gram-negative bacteria selected from the list of Escherichia and Shigella, preferably E. coli. Non-limiting, exemplary structures of E. coli O-antigen polysaccharides are shown in Table 1 below. A single repeating unit of each E. coli O-antigen polysaccharide is shown. In this table, each n is independently an integer between 1 and 100, e.g., between 1 and 50, 1 and 40, 1 and 30, 1 and 20, and 1 and 10, 3 and 50, 3 and 40, e.g., at least 5, e.g., between 5 and 40, e.g., between 7 and 30, e.g., between 7 and 25, e.g., between 5 and 20, e.g., between 10 and 20, but may in some cases be 1 or 2. In certain preferred embodiments of bioconjugate compositions of E. coli O-antigen polysaccharide purified by the methods of the present invention, n is, on average, approximately between 5 and 30, preferably between 10 and 25, and preferably between 10 and 20.

[0113] [Table 1-1]

[0114] [Table 1-2]

[0115] The structures of other E. coli or Shigella or other bacterial O-antigen polysaccharides from various serotypes are known and can be found in the art.

[0116] Bioconjugate composition: This term is known in the art and discussed above. When a carrier protein carries multiple glycosylation sites, for example, the EPA carrier protein having SEQ ID NO: 1, production of the bioconjugate typically results in a mixture of bioconjugates (referred to herein as a bioconjugate composition) rather than a single bioconjugate product.

[0117] The term "bioconjugate composition," as used in certain embodiments herein, therefore refers to a mixture of bioconjugates, wherein each bioconjugate comprises the same carrier protein and the same O-antigen polysaccharide, but the individual bioconjugates differ in the number of occupied glycosylation sites (e.g., a "monovalent" composition). In the case of an EPA carrier protein containing four N-glycosylation sites, e.g., an EPA carrier protein containing four occurrences of the glycosylation consensus sequence Asp(Glu)-X-Asn-Z-Ser(Thr), e.g., an EPA carrier protein having SEQ ID NO: 1, a monovalent bioconjugate composition typically comprises several individual bioconjugates, each having one to four glycans attached to the EPA carrier protein, i.e., mono-, di-, tri-, or tetra-glycosylated bioconjugates. Depending on the particular bioconjugate, particularly the O-EPA bioconjugates defined above, not all particular bioconjugates may be present, e.g., an O-EPA bioconjugate composition may include mono-, di-, and tri-glycosylated O-EPA bioconjugates but not tetra-glycosylated O-EPA bioconjugates. Nevertheless, the O-EPA bioconjugate composition defined above may include all four glycosylated forms, i.e., mono-, di-, tri-, and tetra-glycosylated O-EPA bioconjugates. In other embodiments, the bioconjugate composition can be a mixture of several monovalent compositions, i.e., various drug substances each comprising a carrier protein to which an O-antigen polysaccharide of a particular bacterial serotype has been individually covalently bioconjugated, but with the total composition (the "multivalent" composition) including different monovalent compositions, the multivalent composition includes different O-antigen polysaccharides of different bacterial serotypes, each individually covalently bound to the carrier protein. Such multivalent compositions, in certain embodiments, can also be referred to as "drugs," because they can be used as vaccine compositions for administration to subjects to induce an immune response against the bacterial serotypes whose O-antigen polysaccharides are included in the composition. Each of the individual serotype bioconjugate drug substances present in such a multivalent composition can also be considered part of such a monovalent composition; for example, if the EPA carrier protein contains four N-linked glycosylation sites, then for each serotype, mono-, di-, tri-, and / or tetra-glycosylated bioconjugates can be present in the multivalent composition.

[0118] A bioconjugate composition comprising a carrier protein (EPA) having SEQ ID NO: 1 having an O-antigen polysaccharide of a particular Escherichia coli (E. coli) serotype covalently attached thereto (O-EPA bioconjugate) therefore comprises a mixture of bioconjugates, except that the O-EPA bioconjugates can be either monoglycosylated with one O-antigen polysaccharide attached to the EPA or multiglycosylated with two, three, or four O-antigen polysaccharides attached to the EPA, and thus the bioconjugate composition is a mixture of mono- and multiglycosylated forms of the O-EPA bioconjugate.

[0119] Multiglycosylated bioconjugate form: As used herein, the term "multiglycosylated bioconjugate form" refers to a bioconjugate comprising a carrier protein having two or more covalently attached O-antigen polysaccharides of the same serotype, i.e., a carrier protein in which two or more N-glycosylation sites are occupied. As noted above, an O-EPA bioconjugate comprising an EPA carrier protein having four N-glycosylation sites, e.g., an EPA carrier protein having SEQ ID NO: 1, can be mono-, di-, tri-, or tetra-glycosylated. Thus, in this case, the term "multiglycosylated form" refers to the sum of the di-, tri-, and tetra-glycosylated forms of the bioconjugate.

[0120] Comparison with Previous Methods for Producing O-EPA Bioconjugates: As noted above, WO 2022 / 214620 describes methods for producing O-EPA bioconjugates that are suitable for large-scale production of bioconjugates. However, previously described methods for producing O-EPA bioconjugates comprising an EPA carrier protein covalently attached to an O-antigen polysaccharide of a particular Escherichia coli (E. coli) serotype result in bioconjugate compositions with a relatively low proportion of multiglycosylated forms of the O-EPA bioconjugate, i.e., EPA carrier protein attached to two, three, or four O-antigen polysaccharides. Therefore, pharmaceutical compositions comprising the specific O-EPA bioconjugates necessarily contain compositions of the specific O-EPA bioconjugates with a relatively low proportion of multiglycosylated forms of the O-EPA bioconjugate.

[0121] The degree of glycosylation (DOGY) of O-EPA bioconjugates can be measured by different methods. An established and particularly suitable method is capillary gel electrophoresis (cGE). In cGE, charged molecules are separated based on their size in a capillary filled with a porous gel matrix. CGE is commonly used to separate large biomolecules, such as proteins or glycoproteins. For example, in the case of O-EPA bioconjugates, the presence of covalently bound E. coli O-antigen polysaccharides leads to a change in the mobility of the EPA carrier protein in the capillary. Advantageously, DOGY is measured according to the following protocol: The O-EPA bioconjugates are concentrated to a target concentration of 3 mg / mL and separated in a bare fused silica capillary (total capillary length 30.2 cm and inner diameter 50 μm) containing SDS gel buffer (pH 8.0, 0.2% SDS, 1% PEG6000) using a voltage of −15 kV. Advantageously, for cGE, the absorbance of the bioconjugates is measured at 220 nm. To determine the percentage of glycosylated forms, the relative amounts (peak area percentages) of mono-, di-, tri-, and tetra-glycosylated O-EPA bioconjugate forms are determined by integrating the corresponding peak areas in the electropherogram. Comparisons between the abundance of mono- and multi-glycosylated O-EPA bioconjugates can be established based on the percentages of mono-, di-, tri-, and tetra-glycosylated O-EPA bioconjugate forms.

[0122] A comparison of DOGY measured by cGE for several exemplary O-EPA bioconjugate compositions produced by either the method according to WO 2022 / 214620 (Example 2) or the inventive method described herein (Example 1) showed that the inventive method resulted in a higher proportion of multiply glycosylated O-EPA bioconjugates, as shown in Example 3 (Table 4).

[0123] Thus, in a second aspect, the present invention relates to O-EPA bioconjugate compositions obtainable or obtainable by the inventive production methods described herein. In particular, the present invention relates to O-EPA bioconjugate compositions obtainable or obtainable by the methods of the invention, comprising a carrier protein having SEQ ID NO: 1, to which is covalently attached an O-antigen polysaccharide of Escherichia coli (E. coli) selected from serotypes O1A, O2, or O25B. These bioconjugate compositions are particularly suitable for the manufacture of pharmaceutical compositions described herein.

[0124] In certain embodiments, the present invention relates to an O-EPA bioconjugate composition obtainable or obtainable by the inventive methods described herein, wherein each O-EPA bioconjugate comprises an EPA carrier protein comprising four N-glycosylation sites having the amino acid sequence Asn-X-Ser(Thr), preferably having the amino acid sequence Asp(Glu)-X-Asn-Z-Ser(Thr), where X and Z are independently selected from any amino acid except Pro, e.g., the amino acid sequence of SEQ ID NO: 1; An E. coli O-antigen polysaccharide selected from the group consisting of O1A, O2, O4, O8, O6A, O15, O16, O18A, O25B, and O75 comprises a covalently attached EPA carrier protein. In certain embodiments, the E. coli O-antigen polysaccharide is selected from the group consisting of O1A, O2, O6A, O16, O18A, O25B, and O75. In certain embodiments, the E. coli O-antigen polysaccharide is selected from the group consisting of O1A, O2, and O25B.

[0125] In certain embodiments, the present invention provides an O-EPA bioconjugate composition obtainable or obtainable by the inventive methods described herein, comprising: (a) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O1A, wherein the O-EPA bioconjugate is at least 50% multiply glycosylated; (b) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O2, which is at least 44% multiply glycosylated; (c) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O6A that is at least 64% multiply glycosylated; (d) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O16 that is at least 65% multiply glycosylated; (e) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O18A that is at least 70% multiply glycosylated; (f) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O25B that is at least 50% multiply glycosylated; and / or (g) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O75 that is at least 50% multiply glycosylated; is selected from the group consisting of The percentages of multiply glycosylated and monoglycosylated O-EPA, respectively, are for O-EPA bioconjugate compositions as determined by capillary gel electrophoresis.

[0126] It is understood that the definitions, terms, characteristics and parameters disclosed herein (particularly of the first aspect of the invention) apply equally to this second aspect.

[0127] In a third aspect, the present invention relates to certain pharmaceutical compositions comprising one or more bioconjugate compositions as defined herein. This aspect provides, inter alia, pharmaceutical compositions comprising a bioconjugate composition in which an EPA carrier protein is covalently attached to one of the O-antigen polysaccharides of an Escherichia coli (E. coli) serotype selected from the group consisting of O1A, O2, O6A, O15, O16, O18A, O25B, or O75. These pharmaceutical compositions exhibit particularly useful amounts of the multiglycosylated form of the EPA carrier protein. This aspect is described in further detail below. It is understood that the definitions, terms, characteristics, and parameters disclosed herein (specifically, in the first aspect of the invention) apply equally to this third aspect.

[0128] In a preferred embodiment, the pharmaceutical composition comprises one or more bioconjugate compositions, each of which comprises a Pseudomonas aeruginosa exoprotein A carrier protein (EPA) having SEQ ID NO: 1 to which an O-antigen polysaccharide of a specific E. coli serotype is covalently attached (an O-EPA bioconjugate). Each O-EPA bioconjugate can be monoglycosylated with one O-antigen polysaccharide attached to the EPA, or multiglycosylated with two, three, or four O-antigen polysaccharides attached to the EPA, such that the bioconjugate composition is a mixture of mono- and multiglycosylated forms of the O-EPA bioconjugate. In particular, the pharmaceutical composition comprises the following: (a) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O1A, wherein the O-EPA bioconjugate is at least 50% multiply glycosylated; (b) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O2, which is at least 44% multiply glycosylated; (c) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O6A that is at least 64% multiply glycosylated; (d) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O16 that is at least 65% multiply glycosylated; (e) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O18A that is at least 70% multiply glycosylated; (f) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O25B that is at least 50% multiply glycosylated; and / or (g) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O75 that is at least 50% multiply glycosylated; and The percentages of multiply glycosylated and monoglycosylated O-EPA are each determined by cGE.

[0129] Further details regarding bioconjugate compositions (a)-(g) are provided below (DOGY measurements by cGE as described above).

[0130] Ad composition (a): In one embodiment, the bioconjugate composition of an E. coli serotype O1A O-EPA bioconjugate comprises at least 50% multiglycosylated O-EPA and at most 80% multiglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (a) comprises between 50 and 80% multiglycosylated O-EPA, e.g., 50%, 55%, 60%, 65%, 67%, 70%, 75%, or 80% multiglycosylated O-EPA. In one embodiment, bioconjugate composition (a) comprises at least 55% multiglycosylated O-EPA. Furthermore, in certain embodiments, bioconjugate composition (a) comprises at least 20% monoglycosylated O-EPA and at most 50% monoglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (a) comprises between 20 and 50% monoglycosylated O-EPA, e.g., 20%, 25%, 30%, 35%, 40%, 45% or 50% monoglycosylated O-EPA.

[0131] ad composition (b): In one embodiment, the bioconjugate composition of the Escherichia coli (E. coli) serotype O2 O-EPA bioconjugate comprises at least 44% multiglycosylated O-EPA and at most 90% multiglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (b) comprises between 44 and 90% multiglycosylated O-EPA, e.g., 44%, 45%, 50%, 55%, 60%, 65%, 70%, 71%, 75%, 80%, 85%, or 90% multiglycosylated O-EPA. In certain embodiments, bioconjugate composition (b) comprises at least 50% multiglycosylated O-EPA. Furthermore, in certain embodiments, bioconjugate composition (b) comprises at least 10% monoglycosylated O-EPA and at most 56% monoglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (b) comprises between 10 and 56% monoglycosylated O-EPA, e.g., 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55% or 56% monoglycosylated O-EPA.

[0132] ad composition (c): In one embodiment, the bioconjugate composition of the Escherichia coli (E. coli) serotype O6A O-EPA bioconjugate comprises at least 64% multiglycosylated O-EPA and at most 90% multiglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (c) comprises between 64 and 90% multiglycosylated O-EPA, e.g., 64%, 65%, 70%, 72%, 75%, 80%, 85%, or 90% multiglycosylated O-EPA. In certain embodiments, bioconjugate composition (c) comprises at least 70% multiglycosylated O-EPA. Furthermore, in certain embodiments, bioconjugate composition (c) comprises at least 10% monoglycosylated O-EPA and at most 36% monoglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (c) comprises between 10 and 36% monoglycosylated O-EPA, e.g., 10%, 15%, 20%, 25%, 30%, 35% or 36% monoglycosylated O-EPA.

[0133] ad composition (d): In one embodiment, the bioconjugate composition of the Escherichia coli (E. coli) serotype O16 O-EPA bioconjugate comprises at least 65% multiglycosylated O-EPA and at most 90% multiglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (d) comprises between 65 and 90% multiglycosylated O-EPA, e.g., 65%, 70%, 75%, 80%, 81%, 85%, or 90% multiglycosylated O-EPA. In certain embodiments, bioconjugate composition (d) comprises at least 70% multiglycosylated O-EPA. Furthermore, in certain embodiments, bioconjugate composition (d) comprises at least 10% monoglycosylated O-EPA and at most 35% monoglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (d) comprises between 10 and 35% monoglycosylated O-EPA, e.g., 10%, 15%, 20%, 25%, 30% or 35% monoglycosylated O-EPA.

[0134] ad composition (e): In one embodiment, the bioconjugate composition of the Escherichia coli (E. coli) serotype O18A O-EPA bioconjugate comprises at least 70% multiglycosylated O-EPA and at most 95% multiglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (e) comprises between 70 and 95% multiglycosylated O-EPA, e.g., 70%, 75%, 76%, 80%, 85%, 90%, or 95% multiglycosylated O-EPA. In certain embodiments, bioconjugate composition (e) comprises at least 80% multiglycosylated O-EPA. Furthermore, in certain embodiments, bioconjugate composition (e) comprises at least 5% monoglycosylated O-EPA and at most 30% monoglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (e) comprises between 5 and 30% monoglycosylated O-EPA, e.g., 5%, 10%, 15%, 20%, 25% or 30% monoglycosylated O-EPA.

[0135] ad composition (f): In one embodiment, the bioconjugate composition of the Escherichia coli (E. coli) serotype O25B O-EPA bioconjugate comprises at least 50% multiglycosylated O-EPA and at most 90% multiglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (f) comprises between 50 and 90% multiglycosylated O-EPA, e.g., 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 88%, or 90% multiglycosylated O-EPA. In certain embodiments, bioconjugate composition (f) comprises at least 60% multiglycosylated O-EPA. Furthermore, in certain embodiments, bioconjugate composition (f) comprises at least 10% monoglycosylated O-EPA and at most 50% monoglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (f) comprises between 10 and 50% monoglycosylated O-EPA, e.g., 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45% or 50% monoglycosylated O-EPA.

[0136] ad composition (g): In one embodiment, the bioconjugate composition of the Escherichia coli (E. coli) serotype O75 O-EPA bioconjugate comprises at least 50% multiglycosylated O-EPA and at most 90% multiglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (g) comprises between 50 and 90% multiglycosylated O-EPA, e.g., 50%, 55%, 60%, 65%, 66%, 70%, 75%, 80%, 85%, or 90% multiglycosylated O-EPA. In certain embodiments, bioconjugate composition (g) comprises at least 60% multiglycosylated O-EPA. Furthermore, in certain embodiments, bioconjugate composition (g) comprises at least 10% monoglycosylated O-EPA and at most 50% monoglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (g) comprises between 10 and 50% monoglycosylated O-EPA, e.g., 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45% or 50% monoglycosylated O-EPA.

[0137] Preferably, the purity of each of bioconjugate compositions (a)-(g) is at least 90%, e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.5%, 99.7%, 99.9%, or 100% as measured by SE-HPLC, or at least 95%, e.g., 95%, 96%, 97%, 98%, 99%, 99.5%, 99.7%, 99.9%, or 100% as measured by RP-HPLC. Typically, the purity of each of bioconjugate compositions (a)-(g) is at least 95%, e.g., at least 96%, 97%, 98%, 99%, as measured by either SE-HPLC or RP-HPLC. In certain embodiments, one or more of bioconjugate compositions (a)-(g) are in a container and have a volume of between about 0.5 mL and about 100 L, e.g., between about 1 mL and about 10 L, and the bioconjugate concentration is between about 4 μg / mL and about 5000 μg / mL, e.g., between about 15 μg / mL and about 1000 μg / mL, e.g., about 16 or 32 μg / mL (concentrations measured based on the amount of polysaccharide, as is customary in the art; typically, the polysaccharide / protein (w / w) ratio of such bioconjugates is between about 0.1 and about 0.5, e.g., between about 0.15 and about 0.45, e.g., between about 0.2 and about 0.4).

[0138] In one embodiment, the pharmaceutical composition comprises the bioconjugate composition (a).

[0139] In one embodiment, the pharmaceutical composition comprises the bioconjugate composition (b).

[0140] In one embodiment, the pharmaceutical composition comprises the bioconjugate composition (f).

[0141] In one embodiment, the pharmaceutical composition comprises at least four bioconjugate compositions selected from (a)-(g) as defined above, preferably at least (a), (b), (c) and (f).

[0142] In other embodiments, the pharmaceutical composition comprises all seven bioconjugate compositions (a)-(g) defined above.

[0143] In other embodiments, the pharmaceutical composition as defined above further comprises a bioconjugate composition (h) of an O-EPA bioconjugate of E. coli serotype O15. In a particular embodiment, the O15-EPA bioconjugate composition (h) comprises a carrier protein (EPA) having SEQ ID NO: 1, to which is covalently attached an O-antigen polysaccharide of E. coli serotype O15. The O15-EPA bioconjugates can be monoglycosylated with one O-antigen polysaccharide attached to the EPA or multiglycosylated with two, three, or four O-antigen polysaccharides attached to the EPA, such that the bioconjugate composition is a mixture of monoglycosylated and multiglycosylated O15-EPA bioconjugates, provided that at least 50% is multiglycosylated. In one embodiment, bioconjugate composition (h) of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O15 comprises at least 50% multiglycosylated O-EPA and at most 95% multiglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (h) comprises between 50 and 95% multiglycosylated O-EPA, e.g., 50%, 55%, 60%, 65%, 70%, 75%, 80%, 81%, 85%, 90%, or 95% multiglycosylated O-EPA. Further, in certain embodiments, bioconjugate composition (h) comprises at least 5% monoglycosylated O-EPA and at most 50% monoglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (h) comprises between 5 and 50% monoglycosylated O-EPA, e.g., 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, or 50% monoglycosylated O-EPA. The percentage of multiglycosylated and monoglycosylated O-EPA is determined by cGE as described above.

[0144] Preferably, the purity of bioconjugate composition (h) is at least 90%, e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.5%, 99.7%, 99.9%, or 100% as measured by SE-HPLC, or at least 95%, e.g., 95%, 96%, 97%, 98%, 99%, 99.5%, 99.7%, 99.9%, or 100% as measured by RP-HPLC. Typically, the purity of bioconjugate composition (h) is at least 95%, e.g., at least 96%, 97%, 98%, 99%, as measured by either SE-HPLC or RP-HPLC.

[0145] In certain embodiments, the pharmaceutical composition comprises each of bioconjugate compositions (a), (b), (c), (d), (e), (f), (g), and (h), has a volume of between about 0.5 mL and about 10 L, e.g., between about 1 mL and about 1 L, and has a total bioconjugate concentration of between about 90 μg / mL and about 600 μg / mL, e.g., between about 170 μg / mL and about 300 μg / mL (measured based on the amount of polysaccharide as is customary in the art; typically, the polysaccharide / protein (w / w) ratio of the bioconjugate in the composition is between about 0.1 and about 0.5, e.g., between about 0.15 and about 0.45, e.g., between about 0.2 and about 0.4).

[0146] In another embodiment, the pharmaceutical composition as defined above further comprises a bioconjugate composition (i) of an O-EPA bioconjugate of E. coli serotype O4. In a particular embodiment, the O4-EPA bioconjugate composition (i) comprises a carrier protein (EPA) having SEQ ID NO: 1, to which is covalently attached an O-antigen polysaccharide of E. coli serotype O4. Because the O4-EPA bioconjugates can be monoglycosylated with one O-antigen polysaccharide attached to the EPA or multiglycosylated with two, three, or four O-antigen polysaccharides attached to the EPA, the bioconjugate composition is a mixture of monoglycosylated and multiglycosylated forms of the O4-EPA bioconjugate, provided that at least 10% is multiglycosylated. The percentages of multiglycosylated and monoglycosylated O-EPA are determined by cGE as described above. In one embodiment, bioconjugate composition (i) of an E. coli serotype O4 O-EPA bioconjugate comprises at least 10% multiglycosylated O-EPA and at most 60% multiglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (i) comprises between 10 and 60% multiglycosylated O-EPA, e.g., 10%, 15%, 20%, 25%, 30%, 35%, 40%, 44%, 45%, 50%, 55%, or 60% multiglycosylated O-EPA. Furthermore, in certain embodiments, bioconjugate composition (i) comprises at least 40% monoglycosylated O-EPA and at most 90% monoglycosylated O-EPA. Thus, in certain embodiments, bioconjugate composition (i) comprises between 40 and 90% monoglycosylated O-EPA, e.g., 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85% or 90% monoglycosylated O-EPA.

[0147] Preferably, the purity of bioconjugate composition (i) is at least 90%, e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.5%, 99.7%, 99.9%, or 100% as measured by SE-HPLC, or at least 95%, e.g., 95%, 96%, 97%, 98%, 99%, 99.5%, 99.7%, 99.9%, or 100% as measured by RP-HPLC. Typically, the purity of bioconjugate composition (i) is at least 95%, e.g., at least 96%, 97%, 98%, 99%, as measured by either SE-HPLC or RP-HPLC.

[0148] In certain embodiments, the pharmaceutical composition comprises each of bioconjugate compositions (a), (b), (c), (d), (e), (f), (g), and (i), or each of bioconjugate compositions (a), (b), (c), (d), (e), (f), (g), (h), and (i), in each case having a volume of between about 0.5 mL and about 10 L, e.g., between about 1 mL and about 1 L, and a total bioconjugate concentration of between about 90 μg / mL and about 600 μg / mL, e.g., between about 170 μg / mL and about 300 μg / mL (measured based on the amount of polysaccharide as is conventional in the art; typically, the polysaccharide / protein (w / w) ratio of the bioconjugate in the composition is between about 0.1 and about 0.5, e.g., between about 0.15 and about 0.45, e.g., between about 0.2 and about 0.4).

[0149] In certain embodiments, the pharmaceutical composition as defined above further comprises a bioconjugate composition (j) of an O-EPA bioconjugate of E. coli serotype O8. In certain embodiments, the O8-EPA bioconjugate composition comprises an O-EPA bioconjugate comprising a carrier protein (EPA) having the amino acid sequence of SEQ ID NO: 1, to which is covalently attached an O-antigen polysaccharide of E. coli serotype O8. Preferably, the purity of bioconjugate composition (j) is at least 90%, e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.5%, 99.7%, 99.9%, or 100% as measured by SE-HPLC, or at least 95%, e.g., 95%, 96%, 97%, 98%, 99%, 99.5%, 99.7%, 99.9%, or 100% as measured by RP-HPLC. Typically, the purity of bioconjugate composition (j) is at least 95%, e.g., at least 96%, 97%, 98%, 99%, as measured by either SE-HPLC or RP-HPLC. Thus, in certain embodiments, the pharmaceutical composition comprises each of bioconjugate compositions (a), (b), (c), (d), (e), (f), (g), (h), (i), and (j), has a volume of between about 0.5 mL and about 10 L, e.g., between about 1 mL and about 1 L, and has a total bioconjugate concentration of between about 90 μg / mL and about 600 μg / mL, e.g., between about 170 μg / mL and about 300 μg / mL (measured based on the amount of polysaccharide as is customary in the art; typically, the polysaccharide / protein (w / w) ratio of the bioconjugate in the composition is between about 0.1 and about 0.5, e.g., between about 0.15 and about 0.45, e.g., between about 0.2 and about 0.4).

[0150] In further embodiments, conjugates of different E. coli serotypes (i.e., O-antigen polysaccharides covalently attached to a carrier protein) can be added to obtain a pharmaceutical composition containing, for example, 10 to 20 conjugates, e.g., O-EPA conjugates. Such conjugates of different E. coli serotypes can also be bioconjugates or bioconjugate compositions, and may also be purified by the methods described herein.

[0151] Array Description SEQ ID NO: 1 (EPA carrier protein containing four N-linked glycosylation consensus sequences)

[0152] [ka]

[0153] SEQ ID NO: 2 (example of PglB oligosaccharyltransferase)

[0154] [ka]

[0155] SEQ ID NO: 3 (example of a signal sequence for an EPA carrier protein) MKKIWLALAG LVLAFSASA

[0156] Exemplary glycosylation consensus sequences Asn-X-Ser(Thr), where X can be any amino acid except Pro.

[0157] Exemplary Optimized Glycosylation Consensus Sequences Asp(Glu)-X-Asn-Z-Ser(Thr), where X and Z are independently selected from any amino acid other than Pro.

[0158] The following examples of the present invention are intended to further illustrate the nature of the present invention, and it is understood that the following examples do not limit the present invention, the scope of which is determined by the appended claims. [Example]

[0159] Example 1: Novel method for producing O-EPA bioconjugates This is an example of an inventive method for producing ExPEC O-EPA bioconjugates.

[0160] Bioconjugate-producing strains The construction of production strains for different bioconjugates has been previously described, for example, in Example 6 of WO 2020 / 191082. For some bioconjugates, previously described strains were used, while for some other bioconjugates, new production strains were constructed to improve yield. The parent cell is E. coli strain W3110. In the production strains, the rfb locus (serotype-specific, the sequence of the respective rfb locus provided in Table 2 herein, with reference to the sequences provided in WO 2020 / 191082 and WO 2022 / 208430) encoding the enzyme responsible for production of the relevant E. coli O-antigen polysaccharide replaces the deleted endogenous W3110 rfb locus; the same EPA carrier protein (having four N-linked glycosylation sequences; the carrier protein having SEQ ID NO: 1) is encoded on a plasmid in each of the production strains; and the oligosaccharyltransferase that transfers the O-antigen to an Asn in the N-linked glycosylation sequences of the carrier protein is also encoded on a plasmid in the production strains and is C. jejuni PglB having SEQ ID NO: 2, or a variant thereof that improves yield of the particular E. coli O-antigen bioconjugate and differs from SEQ ID NO: 2 by the mutations shown in the last column of Table 2 below.

[0161] [Table 2]

[0162] These strains were used for the production of bioconjugates according to the inventive methods described herein.

[0163] The inventive method performed in this example consists of the following steps: AEX1 (Capture AEX Chromatography); Particle Reduction Filtration 1; Mixed-Mode Chromatography; Buffer Preparation for HIC; Particle Reduction Filtration 2; HIC; TFF1; Particle Reduction Filtration 3; AEX2 (Final Purification Chromatography); TFF2; Bioburden Filtration.

[0164] The buffers referred to within this and the following examples, including their compositions, are listed in Table 3 below.

[0165] Step i: Preparation of filtered periplasmic fraction containing O-EPA bioconjugate. The filtered periplasmic fraction containing the O25B-EPA bioconjugate was used as the starting material for the purification process outlined below. Production of the O-EPA bioconjugate has been previously described (see, e.g., WO 2009 / 104074, WO 2015 / 124769, and WO 2020 / 191082) and was performed similarly to those protocols. Specifically, the periplasmic fraction of Escherichia coli (E. coli) host cells containing the O25B-EPA bioconjugate was obtained by osmotic shock treatment after incubation in a 200 L bioreactor (fermentor) as previously described [WO 2020191082]. The filtered periplasmic fraction was obtained essentially as follows. Incubation in a 200 L bioreactor corresponds to 165 L of harvest equivalent (HE; LHE).

[0166] In more detail: For the harvest step, 165 L of culture broth is cooled to below 20°C before starting harvesting in a disc stack centrifuge (DSC) using a separator at a constant flow rate (e.g., about 160 L / h). The biomass is collected and the centrifuged liquid (supernatant) is discarded.

[0167] The next step is osmotic shock, which is performed using a solution at approximately 6-10°C. The cell wet weight (CWW) of the concentrated harvest is determined, and based on the CWW, the cell suspension is diluted with 1 / 3 TBS (pH 7.4) to a target CWW of, for example, approximately 390 g / L, and the total volume is determined. A 60% sucrose solution is added to the diluted harvest to 25% of the target concentration, and the cells are incubated at 6-10°C for 1 h while mixing the solution. After incubation with sucrose, the cell / sucrose solution is mixed with 4x the volume of 10 mM Tris-HCl (pH 8.0) at 6-10°C. Mixing is performed using a static mixer. The product is released from the periplasmic space into the supernatant (referred to as the periplasmic fraction, PF), which is then collected.

[0168] The material is then clarified: the cell debris in the PF is removed by separation in a disc stack centrifuge. A constant flow rate is applied and the flow is adjusted based on the turbidity of the centrate. The supernatant is collected and called the centrifuged periplasmic fraction (CPF).

[0169] The CPF, which still contains cellular debris, is filtered through a depth bioburden-reducing filter and collected as the filtered periplasmic fraction (FPF). The FPF was used for further purification of O-EPA, beginning with the AEX1 step (step ii), as described further below.

[0170] Step ii: Capture AEX1 chromatography ii-1: No adjustment of the conductivity of the filtered fermenter harvest was required. The periplasmic fraction of the osmotic shock-filtered fermenter harvest was loaded onto Q Ceramic HyperD F resin for a first anion exchange chromatography run in bind-elute mode (AEX1; 20 cm bed height, 0.34 L of resin / L equivalent recovery). After equilibration of the column with low-salt buffer A, the filtered periplasmic fraction containing the O-EPA bioconjugate was loaded onto the column. The column was then washed with 3 column volumes (CV) of buffer A.

[0171] ii-2: Elution was achieved with a step gradient using 20% ​​Buffer BV2 in Buffer A. The eluate was fractionated into four fractions of 0.25 CV, with collection of eluate fractions starting when the UV absorption increased with a gradient above 1.0 AU / min.

[0172] ii-3: "AEX1 pool" (AEX1 eluate) was generated by mixing fractions 1 and 2.

[0173] Step iii: Mixed-mode chromatography iii-1: Particle reduction filtration is 0.45 + 0.2 μm cutoff and 0.45 m 2 The experiment was carried out using a PES membrane (Sartopore) with a filter area of ​​165 LHE.

[0174] iii-2: Mixed-mode chromatography (Capto adhere multimodal resin, 20 cm bed height, 6.3 L / 165 LHE (recovery equivalent)) was performed in bind-elute mode. After pre-equilibration with 3 CV of buffer CC, the column was equilibrated with 5 CV of buffer AA and the AEX1 eluate was applied onto the Capto Adhere multimodal resin (bind-elute mode). The column was washed with 5 CV of buffer AA, then 5 CV of buffer BB, and finally 2 CV of buffer AA.

[0175] iii-3: Elution was performed using 3.5 CV of Buffer CC. The eluate was collected after the UV gradient (measured at 280 nm) exceeded 0.05 mAU (MMR eluate).

[0176] iii-4: Two fractions of 1 CV each were collected and pooled without prior analysis by SDS-PAGE.

[0177] Step iv: HIC chromatography iv-1: To the pooled MMR eluates, conditioning buffer (Buffer Q) was added to reach 3 times the initial weight of the pooled MMR eluates.

[0178] iv-2: Particle reduction filtration is 0.45 + 0.2 μm cutoff and 0.45 m 2 The experiment was carried out using a PES membrane (Sartopore) with a filter area of ​​165 LHE.

[0179] iv-3: HIC was performed in bind-elute mode. The HIC capsule (Sartobind Phenyl Jumbo 5L, 0.8 cm bed height) was equilibrated with 3 CV of Buffer Q, and the conditioned MMR eluate load was applied. The HIC capsule was then washed with 1 CV of Buffer Q, followed by 4 CV of a buffer mixture containing 70% Buffer Q and 30% Buffer R.

[0180] The iv-4:O-EPA bioconjugate was eluted by applying 4 CV of a buffer mixture containing 30% Buffer Q and 70% Buffer R. The eluate was fractionated into four 0.8 CV fractions. Collection of eluate fractions began when the UV absorption increased with a gradient greater than 0.5 AU / min and the UV absorption value was greater than 0.05 AU.

[0181] iv-5: “HIC pool” was generated by mixing fractions 1–3 without prior analysis by SDS-PAGE.

[0182] Step v: Final Purification AEX2 Chromatography v-1: Tangential flow filtration (TFF1) was performed to prepare the HIC eluate for the final purification step, i.e., AEX2. The HIC eluate was subjected to TFF (mPES KrosFlo filter module Q, 10 kDa, 1.25 ml) with buffer U until the target diafiltration volume of 5–6 DV was reached. 2 Diafiltration was performed using a 165LHE (transmembrane pressure approximately 0.8 bar).

[0183] v-2: The adjusted HIC eluate was then filtered using a 0.45±0.2 μm cutoff and 0.45 μm 2 The solution was filtered using a Sartopore2 capsule size 0 with a PES membrane of filter area of ​​0.05 mm.

[0184] v-3: Final purified AEX chromatography (AEX2) was performed in bind-elute mode. 3 CV of Buffer U was used to equilibrate the column (Source 15Q, 20 cm bed height, 6.3 L / 165 LHE). The adjusted HIC eluate was applied onto the column. Elution was performed using a 7.5 CV step gradient of 21% Buffer V in Buffer U, followed by a 7.5 CV linear gradient from 21 to 56% Buffer V in Buffer U. Fractionation began 1.2 CV after the start of elution, and 30 fractions of 0.5 CV each were subsequently collected.

[0185] v-4: Individual fractions were analyzed by SDS-PAGE and stained with Coomassie stain. The AEX2 eluate pool was generated by combining fractions starting from the first fraction showing only a strong product band and only a faint to moderately intense impurity band, through the first fraction showing a faint product band. Fractions showing a faint product band frequently show an additional band of increasing intensity, corresponding to unglycosylated EPA. In this case, fractions 11 to 30 were pooled. In this example, the O25B-EPA bioconjugate was obtained with a purity of 99.4% as determined by SE-HPLC.

[0186] Step vi: Adjustment to a pharmaceutically acceptable buffer and concentration To obtain the vi-1:O-EPA bioconjugate at a concentration of OD280 = 0.5 ± 0.1, perform a second TFF (mPES KrosFlo Filter Module Q, 10 kDa, 1.25 ml) using an excipient buffer containing 6.19 mM KH2PO4, 3.81 mM Na2HPO4, 5% (w / w) sorbitol, 10 mM methionine, pH 7.0 (Buffer XV2). 2A 165LHE (1 / 165LHE) was applied. The AEX2 eluate was then diafiltered using excipient buffer over 5-6 diafiltration volumes and then concentrated to a target concentration of OD280 = 1.40 ± 0.15. Polysorbate-80 (Tween-80) was then added to the excipient buffer to a final concentration of 0.02% (w / w) to obtain the O25B-EPA conjugate in a pharmaceutically acceptable buffer (Tween-adjusted buffer XV2) containing 6.19 mM KH2PO4, 3.81 mM Na2HPO4, 5% (w / w) sorbitol, 10 mM methionine, and 0.02% (w / w) polysorbate-80, pH 7.0 (see, e.g., WO 2018 / 077853).

[0187] vi-2: PES membrane (0.45 + 0.2 μm cutoff and 0.2 m 2 Bioburden filtration was applied using a Sartopore2 capsule size 9 with a filter area of ​​1000 sachets.

[0188] vi-3: The resulting product (drug substance) was filled into bottles (drug substance bulk) and frozen in a primary freezer at −70° C. + / − 10° C. After 72 h, the DS bottles were removed from the primary freezer and placed directly into a final storage freezer at −70° C. + / − 10° C. without thawing the DS in between.

[0189] In this example, the O25B-EPA bioconjugate was obtained with a purity of 99.6% as determined by SE-HPLC and a yield of 24 mg PS / L HE, corresponding to an estimated overall yield of approximately 23% (relative to the O-EPA conjugate present in the FPF in step i above).

[0190] Using the same inventive method described above, a wide range of bioconjugates have been purified. In particular, O-EPA bioconjugates were purified for Escherichia coli (E. coli) serotypes O1A, O2, O4, O6A, O15, O16, O18A, O25B, and O75. The purity of each was typically at least 96% as measured by SE-HPLC, and in most cases, the purity was 98-100% as measured by SE-HPLC. The estimated overall yield of this method varied from strain to strain, depending in particular on the starting expression level, ranging from approximately 5 to 35%, with an average of approximately 20%. This demonstrates that the method of the present invention is broadly applicable to a variety of different O-EPA conjugates and is suitable for economical, large-scale production of any O-EPA conjugate to a very high purity sufficient for human administration.

[0191] The E. coli O-antigen bioconjugate so prepared was a drug substance (DS) and showed the following comparable critical quality attributes: (1) The purity (measured by SE-HPLC or RP-HPLC) was greater than 95%. (2) The polysaccharide / protein ratio ranged between about 0.1 and 0.5, mostly between 0.15 and 0.45. (3) Bacterial endotoxin (European Pharmacopoeia 2.2.3) was less than 0.5 EU / μg polysaccharide. (4) The average length of individual polysaccharide chains was typically between approximately 10 and 20 repeating units (measured using high-resolution SDS-PAGE).

[0192] Example 2 (Comparative): Method according to WO 2022 / 214620 This is a comparative example relative to the previously preferred production method for ExPEC O-EPA bioconjugates (see WO 2022 / 214620). Only the differences relative to the inventive method outlined above are described. The most notable change in the inventive method of the present invention compared to previously described methods (e.g., WO 2022 / 214620) concerns the replacement of the hydroxyapatite step with a mixed-mode chromatography step (step iii).

[0193] Step iii: Hydroxyapatite chromatography iii-1: The pooled AEX1 fractions were adjusted to a target pH of 7.2 ± 0.2 using buffer Y, and in a subsequent step to a target conductivity of 7.5-9 mS / cm using buffer Z.

[0194] iii-2: Particle reduction filtration was performed as described for the inventive method above.

[0195] iii-3: Ceramic hydroxyapatite chromatography (cHA, 20 cm bed height, 0.34 L resin / L recovery equivalent) was performed in bind-elute mode. After pre-equilibration with 6 CV of Buffer N to adjust the pH, the AEX1 eluate load was applied onto the cHA resin, equilibrated and conditioned with 6 CV of Buffer J. The column was washed with 1 CV of low conductivity buffer consisting of 5% Buffer K in Buffer J.

[0196] iii-4: Product elution was performed using a salt gradient, i.e., a gradient of sodium chloride and potassium phosphate. First, a linear gradient of 20 to 45% Buffer K in Buffer J was applied over 2.5 CV, followed by a step gradient to 70% Buffer K in Buffer J over 2.5 CV. Collection of 14 fractions of 15.565 L / 165 L HE was initiated after applying a 1.9 CV linear gradient.

[0197] iii-5: Individual fractions were analyzed by SDS-PAGE. The selection of fractions for generating the "cHA pool" was based on the intensity of the band corresponding to the O-EPA bioconjugate relative to impurities. The selected fractions were then combined to generate the pooled cHA eluate.

[0198] Step iv: HIC iv-1: In this preliminary step, the load of eluate from the previous step was adjusted to reach a target conductivity of 118±2 mS / cm. In the inventive method, adjustment buffer is added until the added weight reaches three times the initial weight of the pooled eluate from the previous chromatography step. Therefore, this step was simplified in the inventive method.

[0199] Step v: Final Purification AEX2 Chromatography v-1: To prepare the HIC eluate for the final purification step, i.e., AEX2, perform tangential flow filtration (TFF1). TFF with Buffer T (mPES KrosFlo filter module Q, 10 kDa, 1.25 ml) 2 The HIC eluate was adjusted (diluted or concentrated) using a 165LHE (transmembrane pressure approximately 0.8 bar) to an OD280 of 1.35 ± 0.15 and then diafiltered with Buffer T until a target conductivity of 5–6 mS / cm was reached.

[0200] In some cases, further adjustment of the pH of the HIC eluate is required prior to the subsequent chromatography step. In that case, buffer Y is used to lower the pH to a target pH of 6.5±0.2. In a further step, buffer V was used to adjust the conductivity to 8.7±0.3 mS / cm. Thus, the adjustment of the HIC eluate involves at least two steps (concentration adjustment and diafiltration), and in some cases three steps (including pH adjustment). In contrast, in the inventive method (Example 1), the adjustment of the HIC eluate involves only one step (diafiltration).

[0201] v-3: The adjusted HIC eluate was applied onto the column (same column as in Example 1), followed by a washing step using 1.5 CV of Buffer U. In the second wash, a linear gradient of 15-21% Buffer V in Buffer U was applied. In contrast, in the inventive method (Example 1), no washing step is required after application of the HIC eluate onto the AEX2 column.

[0202] v-4: Elution was performed using a 3 CV step gradient of 21% Buffer V in Buffer U, followed by a 7.5 CV linear gradient from 21 to 56% Buffer V in Buffer U. Fractionation began 1.25 CV after the start of elution, followed by collection of 20 fractions of 0.4975 CV. The step gradient was therefore extended in the inventive method compared to the previous method (7.5 CV instead of 3 CV).

[0203] v-5: Individual fractions are analyzed by SDS-PAGE and stained with Coomassie stain. The AEX2 eluate pool is generated by combining the first fractions with an OD280 of at least 0.04 AU. Thus, in this comparative method, fraction pooling was based on an OD280 value of at least 0.04 AU, whereas in the inventive method (Example 1), OD measurement is not required. Instead, fraction pooling was based solely on SDS-PAGE analysis.

[0204] In this example, the O25B-EPA bioconjugate was obtained with a purity of 99.7% as determined by SE-HPLC.

[0205] Step 6: TFF2 vi-1: In contrast to the inventive method (Example 1), this comparative method requires adjustment of the pH of the AEX2 eluate load prior to TFF2 (which is only optional and typically not performed in the inventive method).

[0206] [Table 3-1]

[0207] [Table 3-2]

[0208] Example 3: Increased multiglycosylation of O-EPA bioconjugates obtained by the inventive method This example describes the degree of glycosylation (DOGY) of three O-EPA bioconjugate compositions (E. coli O-antigen polysaccharides corresponding to serotypes O1A, O2, and O25B, each individually conjugated to the EPA carrier protein having SEQ ID NO: 1) produced by the inventive method (Example 1) and the comparative method according to WO 2022 / 214620 (Example 2).

[0209] For each pair of O-EPA bioconjugates shown in Table 4, the same production strain (characterized as shown in Table 2 above) and fermentation conditions were used.

[0210] DOGY was determined by cGE according to the following protocol.

[0211] The O-EPA bioconjugate was concentrated to a target concentration of 3 mg / mL and separated in a bare fused-silica capillary (total capillary length 30.2 cm and inner diameter 50 μm) containing SDS gel buffer (pH 8.0, 0.2% SDS, 1% PEG 6000). Sample separation was performed at -15 kV, and the absorbance of the bioconjugate was measured at 220 nm. The relative amounts of mono-, di-, tri-, and tetraglycosylated forms were determined by integration of the corresponding peak portions. For each measurement, the respective O-EPA bioconjugate drug substance from a previous production batch is used as a reference material and analyzed under the same conditions to ensure normal performance of the method (external control).

[0212] [Table 4]

[0213] For each of these three different bioconjugates, the production strains and fermentation conditions were the same between the method according to WO 2022 / 214620 (comparison) and the inventive method described herein; therefore, any differences in the resulting products may be due to differences in the purification steps. The inventive method of the present invention surprisingly resulted in fewer monoglycosylated and more multiglycosylated O-EPA bioconjugates in the final purified product compared to the previously preferred method, the comparative method described in WO 2022 / 214620. This increase in multiglycosylation in the bioconjugate composition is advantageous in that more available glycosylation sites in the EPA carrier protein are likely to be used in the resulting product, potentially resulting in more available immunogenic glycans per carrier protein molecule.

[0214] In addition to improving the percentage of multiply glycosylated O-EPA, the novel method of the present invention (see Example 1) even resulted in reduced levels of transaldolase B (the most abundant remaining host cell protein) in purified O-EPA bioconjugate compositions, generally for some serotypes, compared to the transaldolase B levels obtained by the comparative method (i.e., having a cHA step rather than an MMC step as the second chromatography step; see Example 2).

[0215] Furthermore, for some serotypes, the novel method of the present invention (Example 1) provided a significant reduction in buffer consumption, typically an approximately 81% reduction in buffer consumption for the second column run, compared to the previously preferred method (Example 2).

[0216] Example 4: Degree of Glycosylation of Nine O-EPA Bioconjugate Compositions This example describes the degree of glycosylation (DOGY) of several O-EPA bioconjugate compositions (E. coli O-antigen polysaccharides corresponding to serotypes O1A, O2, O4, O6A, O15, O16, O18A, O25B, and O75, each individually conjugated to the EPA carrier protein having SEQ ID NO: 1) produced by the inventive method (Example 1). Because the bioconjugate production strain for each bioconjugate can also have an effect on DOGY, depending, for example, on the genes at the rfb locus and / or the PglB variant used, these important characteristics of the production strains are identified in Table 2 above. For each bioconjugate composition, several batches with some degree of DOGY variation were produced. Results from material produced in the context of a clinical trial production campaign are shown in Table 5. DOGY was measured by cGE, as described in Example 3.

[0217] [Table 5]

[0218] The sum of mono-, di-, and triglycosylated O-EPA bioconjugate compositions may deviate slightly from 100% due to rounding effects.

[0219] For O4, there was a reduction in multiglycosylation relative to monoglycosylated O-EPA in the resulting composition relative to compositions obtained using the previous production strain (WO 2020 / 191082) and previous production method (Example 2). For O15, the fraction of multiglycosylated forms in the resulting composition was comparable to compositions obtained using the previous production strain (WO 2020 / 191082) and previous production method (Example 2). However, the other of the bioconjugate compositions has a level of multiglycosylation not previously described and not obtained in bioconjugates of the same O serotype conjugated to the same carrier protein but prepared by prior art methods; i.e., in most cases, the novel bioconjugates of the present invention have elevated levels of multiglycosylation.

[0220] Example 5: Immunogenicity of 9-valent pharmaceuticals in rabbits Nine monovalent drug substances, each corresponding to one of the O-EPA bioconjugates comprising an EPA carrier protein having SEQ ID NO: 1 bound to an O-antigen polysaccharide of Escherichia coli (E. coli) serotypes O1A, O2, O4, O6A, O15, O16, O18A, O25B or O75, obtained by the inventive method described herein (Example 1; production strains identified in Table 2), were mixed (step (vii) according to the present invention, counting further from Example 1) to obtain a nonavalent pharmaceutical composition that was suitable for use in humans (see, for example, WO 2022 / 058945).

[0221] As outlined above, the new method generally resulted in even lower levels of transaldolase B in purified O-EPA bioconjugate compositions for some serotypes compared to the transaldolase B levels obtained with the comparative method (i.e., having a cHA step rather than an MMC step as the second chromatography step; see Example 2). Overall, the residual amount of transaldolase B impurity was further reduced by about 2-3 fold in the nonavalent pharmaceutical composition prepared using the new method according to the present invention compared to the decavalent pharmaceutical composition containing the same dose of polysaccharide prepared using the comparative method. This lower level of impurity is advantageous for pharmaceutical products.

[0222] The 9-valent pharmaceutical composition prepared using the novel method of the present invention was administered to rabbits to induce antibodies against the E. coli serotypes (O1A, O2, O4, O6A, O15, O16, O18A, O25B, and O75) whose O antigens were present in the composition. Thus, it was demonstrated that the bioconjugates and compositions according to the present invention are suitable for inducing an immune response against E. coli.

Claims

1. 1. A pharmaceutical composition comprising one or more bioconjugate compositions, each bioconjugate composition comprises a Pseudomonas aeruginosa exoprotein A carrier protein (EPA) having SEQ ID NO: 1 to which an O-antigen polysaccharide of a particular E. coli serotype is covalently attached (O-EPA bioconjugate); wherein the O-EPA bioconjugate may be monoglycosylated with one O-antigen polysaccharide attached to EPA, or may be multiglycosylated with two, three, or four O-antigen polysaccharides attached to EPA, such that each of the one or more bioconjugate compositions is a mixture of mono- and multi-glycosylated forms of the O-EPA bioconjugate; below: (a) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O1A, wherein the O-EPA bioconjugate is at least 50% multiply glycosylated; (b) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O2, wherein the O-EPA bioconjugate is at least 44% multiply glycosylated; (c) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O6A, which is at least 64% multiply glycosylated; (d) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O16, wherein the O-EPA bioconjugate is at least 65% multiply glycosylated; (e) a bioconjugate composition of an O-EPA bioconjugate of Escherichia coli (E. coli) serotype O18A, wherein the O-EPA bioconjugate is at least 70% multiply glycosylated; (f) a bioconjugate composition of an O-EPA bioconjugate of E. coli serotype O25B that is at least 50% multiply glycosylated; and / or (g) a bioconjugate composition of an O-EPA bioconjugate of E. coli serotype O75, wherein the O-EPA bioconjugate is at least 50% multiply glycosylated; and A pharmaceutical composition, wherein the percentages of multiply glycosylated and monoglycosylated O-EPA are each determined by capillary gel electrophoresis.

2. 10. The pharmaceutical composition of claim 1, comprising at least one of bioconjugate compositions (a), (b) and (f).

3. 3. The pharmaceutical composition of claim 1 or 2, comprising all seven bioconjugate compositions (a)-(g).

4. (h) a bioconjugate composition of an O-EPA bioconjugate of E. coli serotype O15, and (i) a bioconjugate composition of an O-EPA bioconjugate of E. coli serotype O4; further including, or (h) a bioconjugate composition of an O-EPA bioconjugate of E. coli serotype O15, (i) a bioconjugate composition of an O-EPA bioconjugate of E. coli serotype O4, and (j) a bioconjugate composition of an O-EPA bioconjugate of E. coli serotype O8. The pharmaceutical composition according to any one of claims 1 to 3, further comprising:

5. 1. A method for producing a purified bioconjugate from a Gram-negative bacterial host cell, preferably from an E. coli host cell, wherein the bioconjugate comprises a bacterial O-antigen polysaccharide (O-EPA) covalently bound to a Pseudomonas aeruginosa exoprotein A carrier protein; i. Providing a filtered periplasmic fraction (FPF) of the host cells expressing the bioconjugate, the filtered periplasmic fraction comprising the bioconjugate; ii. subjecting the optionally adjusted load of FPF to a first anion exchange chromatography (AEX1) step to obtain a first AEX eluate (AEX1) comprising the bioconjugate; iii. subjecting the optionally adjusted load of the AEX1 eluate to a mixed-mode chromatography (MMC) step using a multimodal resin (MMR) containing both anion exchange and hydrophobic functionality to obtain an MMR eluate containing the bioconjugate; iv. subjecting the adjusted load of the MMR eluate to a hydrophobic interaction chromatography (HIC) step to obtain an HIC eluate containing the bioconjugate; and v. subjecting the conditioned load of the HIC eluate to a second anion exchange chromatography (AEX2) step to obtain an AEX2 eluate containing the bioconjugate. Including, A method wherein in purification steps (ii)-(v), conditions are adjusted first to allow binding of the bioconjugate to a chromatographic medium and then to allow elution of the bioconjugate from the chromatographic medium.

6. The MMR is represented by formula (I): 【Chemistry 1】 (In the formula, RES represents the resin, preferably the compressible resin, of said MMR; R １ is C １ ~C ４ represents alkyl, preferably methyl; R ２ is a C substituted with phenyl, tolyl or xylyl １ ~C ４ represents alkyl, preferably benzyl; R ３ is a C substituted with a hydroxyl group or a thiol group １ ~C ４ alkyl, preferably 2-hydroxyethyl) The method of claim 5, comprising a ligand of the formula:

7. 7. The method of claim 6, wherein the ligand of formula (I) is N-benzyl-N-methylethanolamine.

8. 8. The method according to any one of claims 5 to 7, wherein the O polysaccharide is specific for Gram-negative bacteria selected from the list of Escherichia and Shigella, preferably E. coli.

9. the Gram-negative bacterial host cell from which the bioconjugate is obtained (a) Bacterial O antigen polysaccharide; (b) a recombinant Pseudomonas aeruginosa exoprotein A protein (EPA) containing at least one glycosylation site; and (c) a metabolic apparatus that N-glycosylates the EPA with the O-antigen polysaccharide. comprising genetic information encoding The method of any one of claims 5 to 8, thereby producing the bioconjugate in vivo in the periplasm of the host cell.

10. 10. The method according to any one of claims 5 to 9, wherein the recombinant EPA comprises four N-glycosylation sites having the amino acid sequence Asn-X-Ser(Thr), preferably having the amino acid sequence Asp(Glu)-X-Asn-Z-Ser(Thr), where X and Z are independently selected from any amino acid other than Pro.

11. The method of any one of claims 5 to 10, wherein the recombinant EPA comprises the amino acid sequence of SEQ ID NO:

1.

12. 12. The method of any one of claims 5 to 11, wherein the bacterial O-antigen polysaccharide is an E. coli O-antigen polysaccharide, preferably selected from the group consisting of E. coli O-antigen polysaccharides O1A, O2, O4, O8, O6A, O15, O16, O18A, O25B and O75, more preferably selected from the group consisting of E. coli O-antigen polysaccharides O1A, O2, O6A, O16, O18A, O25B and O75, most preferably selected from the group consisting of E. coli O-antigen polysaccharides O1A, O2 and O25B.

13. 13. The method of any one of claims 5 to 12, wherein the host cells are cultivated in a bioreactor having a volume of between 100L and 20,000L, for example 150L to 5,000L.

14. 14. The method of any one of claims 5 to 13, wherein in an additional step (vi), the load of the AEX2 eluate is adjusted to a pharmaceutically acceptable buffer and concentration, whereby a purified O-EPA conjugate is obtained as a pharmaceutical drug substance.

15. In an additional step (vii), several purified bioconjugate drug substances are combined to obtain a multivalent drug product; Optionally, in step (vii), the multivalent pharmaceutical comprises at least four, preferably at least eight O-antigen polysaccharides; and 15. The method of claim 14, wherein the O-antigen polysaccharide is preferably selected from the group consisting of E. coli O-antigen polysaccharides O1A, O2, O4, O6A, O8, O15, O16, O18A, O25B and O75.

16. An O-EPA bioconjugate composition obtainable or obtainable by the method of any one of claims 5 to 15.

Images