EP4665754A1 - Glycosylated polypeptides - Google Patents

Abstract

The present invention provides a molecule which comprises one or more copies of the following glycosylation amino acid sequence: X₁ Thr Pro X₂ X_3, wherein X₁, X₂, and X₃ are any amino acid; and the Threonine (Thr) amino acid residue is O-glycosylated with a sialylated sugar. The sialylated sugar optionally comprises a chemical group that can be conjugated to a moiety. In some embodiments, the sialylated sugar is conjugated to a moiety. The present invention also provides for uses and methods employing such molecules, particularly in generating conjugates of a desired moiety via the use of the glycosylation sequences.

Description

GLYCOSYLATED POLYPEPTIDES

REFERENCE TO A SEQUENCE LISTING

[0001] The present application is being filed along with a Sequence Listing in ST.26 XML format. The Sequence Listing is provided as a file titled “22760_WO_000 sequence listing ST26” created 12-Feb-2024 and is 15 kilobytes in size. The Sequence Listing information in the ST.26 XML format is incorporated herein by reference in its entirety.

FIELD OF THE INVENTION

[0002] The present invention relates to molecules comprising one or more copies of a glycosylation amino acid sequence for O-glycosylation which provides a convenient way to conjugate the glycosylation amino acid sequence to a desired moiety. The present invention further relates to cell lines for producing such molecules. The present invention also relates to conjugation methods to generate molecules of the invention. The invention also relates to various uses of the molecules and methods employing them, including for therapy and diagnosis.

BACKGROUND OF THE INVENTION

[0003] For more information regarding the headings of the specification, please see MPEP 608.01(a) / 37 CFR 1.77. O-glycosylation is the covalent attachment of sugars to an oxygen group of serine or threonine. Organisms utilize O- glycosylation of proteins for many important biological functions including protection from protease degradation, modulating serum half-life, functional modulation, intracellular trafficking, cell adhesion, and self-versus foreign recognition during an immune response. The most abundant form of O-linked glycosylation in higher eukaryotes, termed “mucin-type,” is characterized by a N-acetylgalactosamine (GalNAc) attached to the hydroxyl group of a serine or threonine side chain. (Hang et al (2005) Bioorg Med CAem.,13(17):5021-5034). There is no known primary amino acid consensus sequence (i.e. template) for mucin-type O-glycosylation.

[0004] Protein glycosylation heterogeneity is a universal feature of life. There are two codified types of variability in glycosylation, namely macroheterogeneity and site occupancy. Macroheterogeneity is the observation that a known glycosylation site in a protein may have a variable occurrence of glycans (i.e. site occupancy). Mucin-type O-linked glycans typically occur clustered together in ‘mucin domains’. (Thanka et al (2001) Biophys J., 80(2):952-960). Site occupancy is difficult to determine due to the lack of a recognized consensus sequence, as well as the lack of a universal enzyme for O-glycan removal. (Jensen et al (2010) I EBS. , 277(1):81-94). Contributing to this complexity is the potential importance of primary, secondary, tertiary, and quaternary protein structures to the efficiency of O-glycosylation of that protein (Wang et al (1993) J Biol Chem., 268(31): 22979-22983). Microheterogeneity is the observation of variation in the glycan structures at the same site of a protein (Galleguillos et al (2017) Comput Struct Biotechnol J. , 15 :212-221 ) .

[0005] There is an ongoing need for providing ways to join a given biological molecule to a desired moiety. The present invention employs a glycosylation sequence to conjugate the two together in a convenient and highly versatile way. It also provides other aspects based on the glycosylation sequence of the present invention which are discussed in more detail below.

SUMMARY OF THE INVENTION

[0006] The present invention is based on a glycosylation amino acid sequence that allows O-glycosylation of a threonine in the sequence. Preferably, after glycosylation of the amino acid sequence the resultant O-glycan comprises at least one sialic acid. The sialic acid may optionally further comprise a chemical group for conjugation to a desired moiety or which has been so conjugated to a desired moiety. A glycosylation amino acid sequence of the invention will be typically present in a polypeptide which is, or forms part of, a molecule of the present invention, providing a convenient and highly versatile way to join the polypeptide to a desired moiety via conjugation. The present invention also provides a molecule comprising one or more glycosylation amino acid sequence(s) of the invention that have yet to be glycosylated. The present invention further provides a molecule comprising one or more glycosylation amino acid sequence(s) of the invention that are O-glycosylated, but which do not comprise a chemical group for conjugation. In one preferred embodiment, the present invention therefore provides a molecule comprising one or more glycosylation amino acid sequence(s) of the invention with terminal sialylation of the O-glycan, where the glycan does not comprise a chemical group for conjugation. In a particularly preferred embodiment, though the present invention therefore provides a molecule comprising one or more glycosylation amino acid sequence(s) of the invention with terminal sialylation of the O-glycan, where the glycan does comprise a chemical group that allows conjugation to a desired moiety. In one preferred embodiment, such conjugation has taken place and the molecule therefore comprises the moiety. In another embodiment, the conjugation has yet to take place, but the chemical group is capable of such conjugation.

[0007] Advantages of the invention include that the glycosylation, and hence the site of any conjugation, is highly specific and can be chosen. The approach of the invention also may help avoid unwanted side-reactions. The conjugation approach of the invention also typically has the advantage of being bioorthogonal, that is the conjugation can take place under physiological conditions and is not toxic to cells. The conjugation is typically biologically and chemically stable under physiological conditions and is not readily reversible. The conjugation also typically does not require the addition of a catalyst. The engineered O-glycan recognition sequence of the invention typically elicits full or near to full site occupation in terms of the proportion of sites being O-glycosylated. The high degree of site occupancy, as well as the consistency in terms of the specific O- glycan sugar added to the amino acid site, mean that the present invention provides an efficient way to generate homogenous conjugates making it a particularly good way to produce therapeutics where such consistency is important. The invention can be used in a wide array of different ways. The conjugation may be, for example, performed by a cell itself. Alternatively, it may be that a molecule of the present invention is purified from a cell that has produced it and it is then conjugated to a desired moiety.

[0008] A particularly preferred approach for the conjugation of the present invention is to employ click chemistry. The use of click chemistry allows for robust and high-yielding reactions that proceed quickly, selectively, under mild conditions, and with few to no side products. In such embodiments, the sialylated O-glycan and the moiety that is the desired conjugation partner have complementary click chemistry groups that are reacted with each other to generate one product molecule. In one embodiment, the moiety is, or comprises, a linker which in turn, or which is already, conjugated to a further molecule or group. In a further preferred embodiment, the molecule comprising the at least one glycosylation amino acid sequence of the present invention and the desired moiety are both polypeptides, with the invention therefore providing a convenient way to join at least two polypeptides together.

[0009] The present invention also provides a cell comprising a polypeptide with a glycosylation sequence of the present invention. The present invention also provides a cell which encodes a molecule of the present invention. The cell may be any of the cell types discussed herein. In a particularly preferred embodiment, the cell is a CHO cell line. In another preferred embodiment, the cell is a human cell line. The present invention further provides a cell suitable for producing molecules of the present invention. In a particularly preferred embodiment, the cell is used for producing the glycosylated form of a molecule of the present invention where the one or more glycosylation sequences are O-glycosylated. In a preferred embodiment, the O-glycosyl sugar chains of the molecule are sialylated. In a preferred embodiment sialylation terminates the O-glycan sugar chain. In one embodiment, the cell is one encoding a molecule comprising one or more glycosylation sequence of the present invention. In another embodiment, the cell further comprises a non-functional UDP-N-acetylglucosamine 2-epimerase gene. Unexpectedly, it has been found that both the epimerase and kinase functions of the endogenous UDP-N-acetylglucosamine 2-epimerase-N-acetylmannosamine kinase gene can be eliminated together without a need to introduce a gene to compensate for the loss of the endogenous N-acetylmannosamine kinase activity. The disruption of the UDP-N-acetylglucosamine 2-epimerase activity means that the cell is deficient in sugar synthesis, providing a convenient way to facilitate the incorporation of sialic acid into the O-glycosyl sugar chains at a glycosylation sequence of the present invention by providing the cell with exogenous modified sugars.

[0010] Accordingly, the present invention provides a molecule which comprises one or more copies of the following glycosylation amino acid sequence:

Xi Thr Pro X₂ X₃ wherein:

Xi, X₂, and X3 are any amino acid; and the Threonine (Thr) amino acid residue is O-glycosylated with a sialylated sugar.

[0011] In a preferred embodiment, the present invention provides a molecule which comprises one or more copies of the following glycosylation amino acid sequence:

Xi Thr Pro X2 X3 wherein

Xi, X₂, and X3 are any amino acid; and the Threonine (Thr) amino acid residue is O-glycosylated with a sialylated sugar which optionally comprises a chemical group that either can be, or is, conjugated to a moiety.

[0012] In a particularly preferred embodiment, the sialylated sugar comprises such a chemical group. In one preferred embodiment, the chemical group is capable of being conjugated to a desired moiety but has not actually yet been so conjugated to the moiety. In a further preferred embodiment, the chemical group is conjugated to a desired moiety.

[0013] The present invention further provides a pharmaceutical composition comprising a molecule of the present invention and a pharmaceutically acceptable carrier. [0014] The present invention further provides a molecule of the present invention for use in therapy of the human or animal body.

[0015] The present invention further provides molecule of the present invention for use in treating a condition selected from cancer, heart disease, obesity, an autoimmune condition, an inflammatory condition, diabetes, or a CNS disorder.

[0016] The present invention also provides a method of treating a condition comprising administering an effective amount of a molecule of the present invention to a subject in need thereof. The condition may be any of those mentioned herein.

[0017] The present invention also provides a cell which comprises:

(a) an endogenous UDP-N-acetylglucosamine 2-epimerase - ManNAc Kinase gene wherein the gene is mutated so that at least the UDP-N- acetylglucosamine 2-epimerase function is reduced or eliminated; and

(b) a sequence encoding a polypeptide comprising the amino acid the amino acid sequence:

Xi Thr Pro X₂ X₃ wherein

Xi, X₂, and X₃ are any amino acid; and the Threonine (T) amino acid residue is O-glycosylated with a sialylated sugar.

[0018] The present invention further provides a method of producing a glycosylated polypeptide comprising culturing a cell of the present invention in a medium supplemented with peracetylated ManNAz.

[0019] The present invention also provides a method of introducing a glycosylation site into a polypeptide comprising modifying the sequence of the polypeptide to include the amino acid sequence:

Xi Thr Pro X₂ X₃ wherein

Xi, X₂, and X₃ are any amino acid; and the Threonine (T) amino acid residue is O-glycosylated with a sialylated sugar. [0020] The present invention further provides a method of conjugating a molecule to a moiety, the method comprising:

(a) providing a molecule of the present invention, wherein the sialylated sugar comprises a chemical group that can be conjugated to a desired moiety with a compatible chemical group;

(b) contacting the molecule of (a) with the desired moiety; and

(c) allowing the chemical group of the molecule to undergo conjugation with the desired moiety through the compatible chemical groups.

[0021] The present also provides a method of joining together two molecules comprising:

(a) providing a molecule of the present invention, wherein the sialylated sugar comprises a chemical group that can be conjugated to a desired second molecule which has a compatible chemical group allowing the conjugation;

(b) contacting the molecule of (a) with the desired second molecule; and

(c) allowing conjugation of the first and second molecules via the compatible chemical groups.

[0022] The present invention further provides a method of generating a combinatorial library, wherein the method comprises:

(a) providing a plurality of molecules of the present invention, wherein the molecules differ from each other, but the sialylated sugar of each molecule comprises the same chemical group that can be conjugated to a desired moiety;

(b) providing a second plurality of molecules, wherein the second plurality of molecules differ from each other and the molecules of (a), but the second plurality of molecules comprise a compatible chemical group to the chemical group of the molecules of (a) allowing conjugation; and

(c) contacting the molecules of (a) and (b) under conditions allowing conjugation and hence the generation of the combinatorial library.

[0023] The present invention also provides a method of conjugating an antibody to a desired moiety comprising: [0024] providing a molecule of the present invention, wherein the sialylated sugar of the molecule comprises a chemical group that can be conjugated to a compatible chemical group of the desired moiety, wherein the molecule is either the antibody or is a component part of the antibody; and

[0025] conjugating the desired moiety to the molecule via said chemical group, wherein if the molecule is a component part of an antibody, rather than the antibody itself, the method further comprises assembling the whole antibody.

[0026] The present invention further provides a method of labelling a molecule, wherein the method comprises:

[0027] providing a molecule of the present invention, wherein the sialylated sugar of the molecule comprises a chemical group that can be conjugated to a desired label which comprises a compatible chemical group to allow conjugation;

[0028] providing the label with the compatible group that allows conjugation to said chemical group of the molecule of (a); and

[0029] contacting the molecule of (a) and the label of (b) under suitable conditions to bring about conjugating of the two.

[0030] The present invention also provides the use of a molecule of the present invention as a capture agent for a desired moiety wherein the sialylated sugar of the molecule comprises a chemical group that can be conjugated to a desired moiety and the desired moiety comprises a compatible conjugation group for the conjugation.

[0031] The present invention also provides a cell encoding a polypeptide which comprises one or more copies of the following glycosylation amino acid sequence:

Xi Thr Pro X₂ X₃ wherein

Xi, X₂, and X₃ are any amino acid; and the Threonine (T) amino acid residue at the second position is O-glycosylated with a sialylated sugar.

[0032] The present invention also provides a cell encoding a molecule of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figures 1, 2(a), 2(b), and 2(c) show the structures of potential O-glycosyl sugar chains with potential sialylation and click chemistry groups.

[0034] Figure 3 shows illustrative protein mass spectroscopy results for particular glycosylation sequences (without GNE knockout - 3(a), and with GNE knockout - 3(b)), with the peak of the antibody molecule with the O-glycosyl sugar chain with sialylation and a click chemistry group depicting the highest peak seen in each instance.

[0035] Figure 4 gives the percentage amounts for site occupancy, clickable glycan, non-conjugatable glycan, conjugation yield with DBCO, and final conjugate homogeneity.

[0036] Figure 5(a) shows viable cell density (VCD) for cultures with an initial viable cell density of >10 x 10⁶/mL. Figure 5(b) shows viable cell densities for cultures with an initial viable cell density of >10 x 106/mL in Example 4.

[0037] Figure 6 shows productivities of molecule #1 for cultures with an initial viable cell density of >10 x 10⁶/mL in Example 4.

[0038] Figure 7 shows in Figure 7(a) growth for cultures with an initial viable cell density of ~1 x 10⁶/mL. Figure 7(b) shows viabilities for cultures with an initial viable cell density of ~1 x 10⁶/mL in Example 4.

[0039] Figure 8: shows productivities of molecule #1 for cultures with an initial viable cell density of ~1 x 10⁶/mL in Example 4.

[0040] Figure 9 shows protein mass spectroscopy results for the sugar titration performed in Example 4.

[0041] Figure 10 shows the product distribution obtained in the sugar titration performed in Example 4 with the proportion of molecules which were asialylated, sialylated, and sialylated with a click sugar.

[0042] Figure 11 shows the cell growth for cultures with an initial viable cell density of ~10 x 10⁶/mL for molecule #2 in Example 5.

[0043] Figure 12 shows the viabilities for cultures with an initial viable cell density of ~10 x 10⁶/mL for molecule #2 in Example 5. [0044] Figure 13 shows the productivities for cultures with an initial viable cell density of ~10 x 10⁶/mL for molecule #2 in Example 5.

[0045] Figure 14 shows the product distribution for cultures with an initial viable cell density of ~10 x 10⁶/mL for molecule #2 in Example 5.

[0046] Figure 15 shows CE-MS summaries for day 9 samples showing incorporation of azidosialic acid on to molecule #2 in Example 5.

[0047] Figure 16 illustrates the basic approach of how an antibody with the glycosylation sequence provided can be conjugated to siRNA through the sialylated O-glycosyl sugar chains present using click chemistry sugars either with or without a linker.

[0048] Figure 17 shows an illustrative chromatogram for the results for conjugation of siRNA to azidosialic acid sugar, with incorporated antibody monitored by analytical anion exchange (aAEX) over time as a function of antibody concentration at 1 mg/mL, 5 mg/mL and 10 mg/mL to assess impact of antibody concentration on conjugation. Drug antibody ratio (DAR) was calculated based on peak area % from the aAEX chromatogram. The results obtained for the conjugation kinetics are shown in Figure 18 (18(a) - 10 mg/mL, 18(b) - 5 mg/mL, and 18(c) - 1 mg/mL mAb).

[0049] Figure 19 shows the conjugate profile obtained in Example 7 analyzed by analytical anion exchange.

[0050] Figure 20 shows the stability of the siRNA conjugate generated in Example 7 in cynomolgus monkey plasma and mouse plasma.

[0051] Figure 21 shows the ability of mouse TfR binding antibody-siRNA conjugates either by eCys or glyco-mAb conjugation chemistry (e.g., mTfR2- dsRNA No. 8 conjugate) to knockdown a target gene in Example 7.

[0052] Figure 22 shows a siRNA conjugate aAEX profile as described in Example 7.

[0053] Figures 23 (a and b), 24 (a and b), and 25 (a and b) show the impact of siRNA: antibody conjugate on gene expression as studied in Example 7.

[0054] Figure 26 illustrates the ability of a siRNA: Ab conjugate of the invention to cross the Blood Brain Barrier (BBB) via the antibody portion of the conjugate and then knockdown the expression of a target gene via the siRNA. Results for an isotype control conjugate where the antibody does not target the TfR protein necessary for transport across the BBB are also provided.

[0055] Figure 27 illustrates the use of the conjugation method provided to generate multi-functional antibodies and combinatorial libraries.

[0056] Figures 28 and 29 illustrate the generation and use of a preferred linker of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

[0057] Definitions

[0058] A number of definitions are set out in the present section and elsewhere in the present application. In cases where a term is not defined, explained or illustrated herein it will typically have the same meaning commonly understood to one of ordinary skill in the art to which this invention pertains. Otherwise, certain terms cited herein have the meanings as set in the specification.

[0059] The use of the singular forms such as “a,” “an,” and “the” herein includes plural reference unless the context clearly dictates otherwise.

[0060] Where the term “comprise”, “comprising” and the like are employed herein, the invention also encompasses embodiments “consisting essentially of” and “consisting of’ what is set out. Hence, where an embodiment is set out using “comprise”, or similar language, also specifically disclosed is an embodiment that consists of what is set out.

[0061 ] The term “specific for”, as used herein in relation to a binding domain, refers to the ability of a binding domain of a multispecific binding molecule to bind, associate with and / or modulate a particular target molecule, over background levels of non-target levels. In some embodiments, “specific for” may be demonstrated by affinity (Kd) measures for the target over non-targets. Where two antigen-binding sites have a “different” specificity they will typically bind either two different antigens or two different epitopes of the same antigen.

[0062] The terms “treatment”, “treat” and “treating” as used herein, includes restraining, slowing, stopping, controlling, delaying, or reversing the progression or severity of an existing symptom or disorder, or ameliorating the existing symptom or disorder, but does not necessarily indicate a total elimination of the existing symptom or disorder. Treatment includes administration of a protein or nucleic acid or vector or composition for treatment of a symptom or disorder in a patient, particularly in a human.

[0063] A “molecule” of the present invention will comprise at least one glycosylation amino acid sequence of the present invention. The glycosylation amino acid sequence is O-glycosylated. The glycosylation may further comprise a chemical group for conjugation. Where such a chemical group is present, it may be conjugated already, or be capable of such conjugation, to a desired moiety. In cases where such conjugation has already taken place, the molecule may be said to also comprises the conjugated desired moiety. A molecule may be, or may comprise, a peptide, polypeptide, or protein comprising at least one glycosylation amino acid sequence of the present invention. A preferred molecule is a polypeptide. A molecule though is not limited to being a single polypeptide. For example, an IgG antibody comprises four polypeptide chains and represents an IgG molecule. A molecule may itself comprise smaller molecules as constituent parts. The term “molecule” in the present application therefore encompasses higher order structures, for example where polypeptides and/or other moieties have been joined together, such as via the present invention.

[0064] A “polypeptide” is a linear polymer of amino acids connected by amide bonds, specifically peptide bonds. As a glycosylation amino acid sequence of the present invention is at least five amino acids in length a polypeptide of the present invention will be at least that length. The term “peptide” denotes a short polypeptide, for example from 5 to 50 amino acids, such as from 10 to 50 amino acids. The term “polypeptide” as used herein encompasses “peptide” sequences. It may though denote a longer sequence, for example one of at least 50 amino acids in length.

[0065] As used herein, a “protein” comprises one or more polypeptides that are covalently linked or noncovalently associated. Proteins typically have a more defined tertiary' and quaternary' structure and may have homogenous or heterogeneous post-translational modifications. Examples of proteins include but are not limited to antibodies, enzymes, and cy tokines. A molecule of the present invention may be, or comprise, a protein comprising one or more glycosylation amino acid sequences of the invention.

[0066] A “binding molecule” may be thought of as any molecule which binds, and preferably specifically binds, a target. Particularly preferred examples of binding molecules include antibodies, but the term is not limited to antibodies, encompassing any polypeptide, protein or nucleic acid which specifically binds a target. Binding molecules may be protein binding molecules. They may be nucleic acid binding molecules. Examples of the latter include anti-sense nucleic acid molecules, as well as siRNA molecules, and aptamers.

[0067] The term “antibody,” as used herein, refers to an immunoglobulin molecule that binds an antigen. Embodiments of an antibody include a monoclonal antibody, polyclonal antibody, human antibody, humanized antibody, chimeric antibody, bispecific or multispecific antibody, or conjugated antibody. The antibodies can be of any class (e.g., IgG, IgE, IgM, IgD, IgA), and any subclass (e.g., IgGl, IgG2, IgG3, IgG4).

[0068] Embodiments of the present invention also include antibody fragments or antigen-binding fragments that, as used herein, comprise at least a portion of an antibody retaining the ability to specifically interact with an antigen or an epitope of the antigen, such as Fab, Fab’, F(ab’)2, Fv fragments, scFv antibody fragments, scFab, disulfide-linked Fvs (sdFv), a Fd fragment. It includes linear antibodies, which may be for example, fused to a Fc region or an IgG heavy chain constant region, such as an scFv-CH3 minibody, an scFv-Fc antibody, an scFv-zipper antibody, a Fab2 bispecific, a bis-scFv, a sdAb, a tetrabody, a triabody, a diabody, or a Fabs trispecific antibody.

[0069] The term “antibody” includes single chain antibody formats. It includes heavy chain only antibodies, such as VH and VHH domain antibodies. It includes other types of antigen-binding molecules such as antibody analogues like DARPins (designed ankyrin repeat proteins). It also includes artificially constructed formats of antibody, such as naturally occurring antibody format molecules, but with further antigen-binding sites added, such as at the C-terminus of the light and/or heavy chains.

[0070] The term “antigen binding domain”, as used herein, refers to a portion of a binding molecule, antibody, antibody fragment, bispecific antibody, multispecific binding protein, etc. that binds an antigen or an epitope of the antigen

[0071] The term “bispecific”, as used herein, refers to a molecule that comprises two distinct antigen-binding domains. A bispecific binding molecule can bind two different antigens or two different epitopes of the same antigen. Exemplary embodiments of bispecific molecules include the bispecific antibodies disclosed herein.

[0072] The term “multispecific”, as used herein, refers to a molecule that comprises two or more distinct antigen-binding domains. A multispecific binding molecule can bind two or more different antigens, or two or more different epitopes of the same antigen. Exemplary embodiments of multispecific binding molecules include bispecific, trispecific or tetraspecific binding molecules known in the field, as well as single-chain multispecific binding molecules such as diabodies, tandem scFvs, tandem VHHs, or tandem scFabs.

[0073] The term “agonize”, as used herein, refers to the ability of an antibody, antibody fragment, or a binding molecule to induce or increase one or more activities or functions associated with an antigen. In one embodiment, a molecule of the present invention may be, or comprise, an agonist.

[0074] The term “antagonize”, as used herein, refers to the ability of a molecule, antibody, antibody fragment, or a binding molecule to decrease or eliminate one or more activities or functions associated with an antigen. In one embodiment, a molecule of the present invention may be, or comprise, an antagonist.

[0075] The term “neutralize”, as used herein, refers to the ability of a molecule, antibody, antibody fragment or a binding molecule to counteract or render inactive or ineffective at least one activity or function of an antigen or other target.

[0076] The terms “bind” and “binds” as used herein are intended to mean, unless indicated otherwise, the ability of a protein or molecule to form a chemical bond or attractive interaction with another protein or molecule, which results in proximity of the two proteins or molecules as determined by common methods known in the art. The term “conjugation” entails the joining of a molecule to a desired moiety via a chemical bond.

[0077] As referred to herein, the term “epitope” refers to the amino acid residues, of an antigen, that are bound by an antibody. An epitope can be a linear epitope, a conformational epitope, or a hybrid epitope.

[0078] The term “ structural epitope” may be used to describe the region of an antigen which is covered by an antibody (e.g., an antibody’s footprint when bound to the antigen). In some embodiments, a structural epitope may describe the amino acid residues of the antigen that are within a specified proximity (e.g., within a specified number of Angstroms) of an amino acid residue of the antibody.

[0079] The term “functional epitope” may also be used to describe amino acid residues of the antigen that interact with amino acid residues of the antibody in a manner contributing to the binding energy between the antigen and the antibody.

[0080] An epitope can be determined according to different experimental techniques, also called “epitope mapping techniques.” It is understood that the determination of an epitope may vary based on the different epitope mapping techniques used and may also vary with the different experimental conditions used, e.g., due to the conformational changes or cleavages of the antigen induced by specific experimental conditions. Epitope mapping techniques are known in the art (e.g., Rockberg and Nilvebrant, Epitope Mapping Protocols: Methods in Molecular Biology, Humana Press, 3^rd ed. 2018), including but not limited to, X- ray crystallography, nuclear magnetic resonance (NMR) spectroscopy, site- directed mutagenesis, species swap mutagenesis, alanine-scanning mutagenesis, hydrogen-deuterium exchange (HDX) and cross-blocking assays.

[0081] As used herein, the term “competes for binding” or “competes with”, refers to two antibodies which cross-compete (i.e., compete against each other) for binding to the same antigen. In some embodiments, two antibodies may compete for binding to the same antigen where they bind to spatially overlapping regions of the same antigen. In some embodiments, two antibodies may compete for binding to a same antigen where the antibodies bind to non-overlapping regions of the antigen, but the binding of one antibody blocks binding by the other antibody, for example, due to steric hindrance or conformational changes of the antigen induced by the first antibody.

[0082] The term “paratope”, as used herein, refers to the amino acid residues of an antibody that bind the antigen. Amino acid residues of a paratope can be identified based on a specified proximity (e.g., within a specified number of Angstroms) from an amino acid residue of the antigen, for example, as may be determined by X-ray crystallography. Amino acid residues of a paratope may also be identified based on contributing to the binding energy between the antigen and the antibody. For example, such amino acid residues of a paratope may be determined by examining protein binding in functional binding assays of the antibody to the antigen, where the antibody is mutated at different sites of the paratope.

[0083] “Specifically binds” indicates binding to a target in preference to the nontarget. It may be that binding is to the target, but not at any significant level to non-targets. It may be the affinity of binding is at least 5, 10, 50, 100, 1000-fold, or greater for the target than the non-target.

[0084] A “monoclonal antibody” is an antibody produced by a single clone of cells or a cell line and consisting of identical antibody molecules.

[0085] A “chemical group for conjugation” represents a chemical group that can be conjugated to a complementary chemical group, for instance, but not exclusively, by click chemistry.

[0086] A “glycosylation amino acid sequence” as used herein refers to the amino acid sequence Xi Thr Pro X2 X3, wherein Xi, Xi, and Xi may be any amino acid and the sequence can be glycosylated at the Thr residue. Such a sequence is referred to as a glycosylation amino acid sequence or glycosylation sequence of the present invention. The introduced O-glycan typically includes a sialic acid sugar which may also include a group that allows for conjugation to a desired moiety. Particularly preferred is for the O-glycan to comprise a click chemistry group and the molecule it is desired to conjugate to also comprising a compatible click chemistry group. The two click chemistry groups can react, conjugating the polypeptide and desired moiety together.

[0087] A “moiety” as used herein is any entity that can be conjugated so that it forms part of a molecule of the present invention. A moiety may itself be a molecule but can be conjugated to form a larger molecule. Hence, for instance, a molecule of an invention typically comprises at least one glycosylation sequence which allows conjugation to a moiety to form a molecule which is larger, but still represents a molecule of the present invention. The moiety that it is desired to be conjugated to, comprises a complementary group for reaction with the conjugatable sialic acid analog comprising the chemical group for conjugation. A “molecule” encompasses a molecule which itself comprises constituent molecules which are joined or are associated together, for instance the term molecule includes molecules such an IgG molecule which typically comprises four polypeptide chains.

[0088] The numbering of the amino acid residues in antibody sequences set out herein is based on the EU index as in Kabat. Kabat et al, Sequences of Proteins of Immunological Interest, 5th edition, Bethesda, MD: U.S. Dept, of Health and Human 20 Services, Public Health Service, National Institutes of Health (1991). The term EU Index numbering or EU numbering is used interchangeably herein. Assignment of amino acid residues to the CDRs may be done according to the well-known schemes, including those described in Kabat, Chothia, North or IMGT.

[0089] Molecules and glycosylation sequences

[0090] A glycosylation sequence of the present invention comprises the amino acid sequence Xi Thr Pro X2 X3. In particular, the present invention provides a molecule comprising a glycosylation sequence which comprises an amino acid sequence: Xi Thr Pro X2 X3, wherein: (i) Xi, X2, and X3 are any amino acid; and (ii) the threonine (T) amino acid residue is O-glycosylated with a sialylated sugar. In a particularly preferred embodiment, the sialylated sugar comprises a chemical group thatcan beconjugated to a desired moiety. In a further particularly preferred embodiment, the sialylated sugar comprises a chemical group that actually is conjugated to a desired moiety As used herein “glycosylation sequence” relates to such an amino acid sequence unless otherwise stated.

[0091] The Threonine in the amino acid sequence Xi Thr Pro X2 X3 in a glycosylation sequence of the present invention is glycosylated. In another aspect, the molecule may be glycosylated at other sites and in other ways, but, as a minimum, will comprise one or more glycosylation sequence(s) of the present invention which are glycosylated. In one embodiment, the only glycosylation of a molecule of the present invention is that involving the one or more glycosylation sequences of the present invention. In another embodiment, it is not the only glycosylation.

[0092] In one embodiment, in the glycosylation sequence of the invention: (i) Xi is Pro; (ii) X3 is Pro; (iii) Xi and X3 are both Pro; (iv) X2 is Ala; (iv) one or both of Xi and X3 are Pro and X2 is Ala; or (v) Xi and X3 are both Pro and X2 is Ala. In one embodiment, the glycosylation sequence comprises the amino acid sequence Xi Thr Pro X2 X3, wherein at least one of Xi and X3 is Pro. In one embodiment, both Xi and X3 are Pro. In one preferred embodiment X2 is Ala. In one embodiment, Xi is Pro and X2 is Ala. In another embodiment, X2 is Ala and X3 is Pro. In another preferred embodiment Xi and X3 are Pro and X2 is Ala.

[0093] In an especially preferred embodiment, the glycosylation sequence comprises the amino acid sequence Pro Thr Pro Ala Pro (SEQ ID NO: 1). In any of the embodiments described herein, a glycosylation sequence may be, for instance, any of the glycosylation sequences of the invention set out herein. However, in an especially preferred embodiment, the one or more glycosylation sequence(s) in any of the embodiments set out herein comprises the amino acid sequence Pro Thr Pro Ala Pro.

[0094] In one preferred embodiment, any of the glycosylation sequences set out above may be flanked on the N-terminal side by an Alanine (Ala) residue. In one preferred embodiment, any of the glycosylation sequences set out above may be flanked on the C-terminal side by an Ala residue. [0095] In one preferred embodiment, any of the glycosylation sequences set out above may be flanked on the N-terminal and C-terminal sides by an Ala residue. In a preferred embodiment, the glycosylation sequence comprises, or consist of, Ala Pro Thr Pro Ala Pro Ala. In a preferred embodiment, the glycosylation sequence comprises, or consists of, Ala Ala Pro Thr Pro Ala Pro. In one preferred embodiment, the glycosylation sequence comprises, or consist of, Ala Ala Pro Thr Pro Ala Pro Ala.

[0096] In another preferred embodiment, the glycosylation sequence comprises, or consists of, Ala Ala Ala Thr Pro Ala Pro (SEQ ID NO: 2).

[0097] The glycosylation amino acid sequences of the invention are typically present in one or more polypeptides forming part of the molecule of the present invention. In one embodiment, a molecule of the present invention comprises a polypeptide comprising at least one glycosylation amino acid sequence of the present invention. In one embodiment, a molecule of the present invention is such a polypeptide. In one embodiment, a molecule of the present invention comprises, or consists of, a polypeptide comprising a glycosylation amino acid sequence which is glycosylated and comprises a chemical group for conjugation.

[0098] Preferably a glycosylation sequence of the invention is O-glycosylated at the threonine residue of the glycosylation sequence. The invention also provides the molecules before they have been glycosylated and such intermediates form part of the invention. In one preferred embodiment, the O-glycan comprises a sialic acid group. In a further preferred embodiment, the sialylated sugar comprises a chemical group allowing conjugation to a desired moiety. As discussed further below, in an especially preferred embodiment the sialic acid group comprises a click chemistry group allowing for conjugation of that click chemistry group with a second compatible click chemistry group on the moiety that it is desired to conjugate to.

[0099] One advantage of the invention is that the glycosylation sequence can be simply introduced into the amino acid sequence of a given polypeptide at the location it is desired to conjugate a moiety to. That means that the invention can be applied to any desired molecule which comprises an amino acid sequence which can be modified to be a glycosylation amino acid sequence of the present invention. An amino acid sequence may have been modified by changing the existing amino acid sequence to include the glycosylation sequence or by inserting the glycosylation sequence into the amino acid sequence. In another embodiment, a region of the amino acid sequence may be replaced with a glycosylation sequence of the present invention, for example a region of the same length may be replaced with a glycosylation sequence of the present invention.

[0100] In one embodiment, a molecule of the present invention comprises at least one glycosylation sequence of the present invention. In one embodiment, a molecule of the present invention comprises only one glycosylation sequence of the present invention. In another embodiment, it comprises at least two such glycosylation sequences. In another embodiment, it comprises only two such glycosylation sequences. In one embodiment, a molecule of the present invention comprises at least four glycosylation sequences. In one embodiment, it comprises only four such glycosylation sequences. In one embodiment, a molecule of the present invention comprises one, two, three, four, five, six, seven or more glycosylation sequences of the present invention. In another embodiment, it comprises from one to seven, for example from two to six, such as two such glycosylation sequences. In one embodiment, the molecule comprises an even number of such glycosylation sequences. In one embodiment, it comprises, two, four, six, eight, or ten glycosylation sequences. In another it comprises at least such numbers or a range formed by such numbers for example from two to ten such sequences. The presence of a plurality of glycosylation sequences may be used to generate molecules that comprise a higher order structure, such as dimers or multimers, by using the O-glycosylation to join polypeptides. In one embodiment, two glycosylation amino acid sequences of the present invention are present in a molecule of the present invention with the two conjugated to each other. In another embodiment, a molecule of the present invention comprises at least one such pair of conjugated glycosylation sequences. In one embodiment, it comprises two, three, four, five, or six such pairs of conjugated amino acid sequences of the present invention. [0101] In one embodiment, there may be more than one glycosylation amino acid sequence present in a molecule of the invention. It may be that the presence of O- glycans near to a glycosylation site of the invention helps promote O- glycosylation. Hence, in one embodiment a molecule of the invention may comprise two or more glycosylation sequences of the present invention in the same polypeptide within 100 amino acids of each other, for instance within 75 amino acids of each other, preferably within 50 amino acids of each other. In one embodiment, glycosylation sequences of the invention are present at regular intervals in an amino acid sequence present in a molecule of the present invention, for example to allow conjugation to a plurality of moieties or to form multiple bridges between two molecules and hence strengthen their joining to each other.

[0102] The invention may be used to conjugate a desired moiety to a given protein to form a desired molecule of the present invention. In another embodiment, the invention may be used to form a bridge between two polypeptides present in the molecule, for instance to help stabilize the overall molecule. In one embodiment, the invention is used to form a molecule which is, or comprises, a “locked” polypeptide where conjugated polypeptides are covalently bonded via the O-glycosyl bridge and so do not readily disassociate. In one embodiment, the invention is used to form molecules which are, or comprise, locked dimers or multimers where the individual monomers are covalently bonded via the glycosyl bridges. The invention may also be used to generate a ligand that becomes covalently bound to its receptor once the ligand has bound the receptor via the conjugation approach of the invention so forming a molecule of the present invention which is a complex of the ligand and receptor covalently bound together.

[0103] A molecule of the present invention may be, or comprise, any suitable polypeptide. The molecule may be, or comprise, a therapeutic protein. In one preferred embodiment, a molecule of the present invention is, or comprises a cytokine. Hence, at least one polypeptide chain of the cytokine may comprise one or more glycosylation sequence of the present invention. In one embodiment, all the polypeptide chains of the cytokine may do so. In one embodiment, the cytokine is an Interleukin. In another embodiment, the cytokine is an interferon. Examples of cytokines that the present invention may be applied to include TNF- a, IL-1, IL-10, IL-12, INF-a, or INF-y. In another embodiment, the cytokine is IL- 2, IL-4, IL-5, TGF-P, or INF- . In a further embodiment, the cytokine is a colony stimulating factor (CSF). In one embodiment, the cytokine is GM-CSF. Other preferred proteins include, for instance, growth factors, hormones, blood clotting factors, tumor necrosis factors, interferons, and cytokines. The protein may be, or comprise, EPO (erythropoietin) or an EPO analog. The protein may be an enzyme.

[0104] In one especially preferred embodiment, the molecule of the present invention may comprise, or be, an antibody, as discussed further below.

[0105] In a preferred embodiment, a molecule of the present invention, may be or comprise a vertebrate protein. The protein may be, for instance, a mammalian protein. In an especially preferred embodiment, the protein may be a human protein. The protein may be an animal protein, for instance, a mouse, rat, or monkey protein. The protein may be a cow, sheep, dog, or cat protein.

[0106] In one embodiment, a molecule of the present invention may be secreted. In one preferred embodiment, the polypeptide with one or more glycosylation sequence of the present invention also comprises a secretion signal. In another embodiment, the polypeptide comprises a signal which results in it being displayed on the cell surface.

[0107] In one embodiment, the invention may be applied to dimers or multimers, for instance to form covalent bridges from one subunit of the dimer or multimer to another. The invention may be used to form covalent bridges joining together at least two of the individual subunits of a multimer. In one embodiment, the invention may be used to form locked dimers or multimers where at least two subunits are joined together via the glycosylation sequence or sequences of the present invention forming a bridge. In one embodiment, cytokine dimers may be generated where the individual cytokine subunits comprise at least one glycosylation sequence of the present invention allowing addition of an O-glycan sugar chain and subsequent conjugation of one cytokine monomer to another. Such an approach may be used, for instance, to form locked cytokine dimers. The generation of such cytokine dimers may be used to generate, for instance, low affinity cytokine dimers.

[0108] In an alternative embodiment of the present invention, the molecule of the invention comprises one or more glycosylation sequences of the present invention which are O-glycosylated with sialylation terminating the O-glycan sugar, but where the O-glycan sugar does not comprise a group for conjugation. In a preferred embodiment, the presence of the O-glycosylation is used to modify the properties of the molecule, such as the physical properties of the molecule. In another embodiment, the O-glycosylation is used to promote immunotolerance. In another embodiment, the O-glycosylation is used to mask another site in the polypeptide. In a further alternative embodiment, the invention further provides a molecule where the one or more glycosylation sequences of the invention are O- glycosylated, but without sialylation.

[0109] In one especially preferred embodiment, a molecule of present invention comprises a polypeptide comprising the glycosylation sequence, wherein the polypeptide is at least 10 amino acids in length. In another preferred embodiment, the polypeptide is at least 20 amino acids in length. In another embodiment, the polypeptide is at least 50 amino acids in length. In a further preferred embodiment, the polypeptide is at least 100 amino acids in length.

[0110] In a further particularly preferred embodiment, a molecule of the invention comprising a glycosylation amino acid sequence is present in a cell. In another embodiment, it is expressed in a cell. In one preferred embodiment, it is present or expressed in a mammalian cell. Examples of preferred cells include human cells. They also include rodent cells, for example CHO cells are particularly preferred.

[0111] Sugar chains

[0112] Preferably the threonine residue of a glycosylation sequence in a molecule of the present invention is O-glycosylated. Hence, the threonine has a sugar chain attached to it. In one particularly preferred embodiment, a glycosylation sequence of the present invention is O-glycosylated with a sugar chain comprising N-acetyl hexosamine and hexose. In one preferred embodiment, the O-glycan sugar chain is a glycan comprising N-acetyl hexosamine and hexose. In a particularly preferred embodiment, the O-glycosylation comprises a sialic acid group. Typically, the sialylation terminates the O-glycan sugar chain, with either the group for conjugation forming part of the sialic acid group or joined to it. In one preferred embodiment, the O-glycan sugar chain comprises, or is, a glycan which has at least one sialic acid at its terminus. In one embodiment, the glycan may have one or two sialic acids.

[0113] In one particularly preferred embodiment, the O-glycosylation of the Threonine comprises: Threonine - N-acetyl hexosamine - hexose - sialic acid - conjugation group.

[0114] In one preferred embodiment, at least two glycosylation sequences are so glycosylated, where the conjugation groups are compatible with each other, and then conjugated together in a molecule of the present invention via the compatible conjugation groups. In another embodiment, one glycosylation sequence is so glycosylated and then conjugated to a moiety with the appropriate compatible conjugation group, but without the moiety needing or having an O-glycosyl sugar chain, as the presence of a glycan on the moiety it is desired to conjugate to is unnecessary for the conjugation reaction to occur in such embodiments. In one preferred embodiment though the polypeptide and the moiety it is being conjugated to both comprise at least one glycosylation sequence of the present invention, so that both are O-glycosylated with sialylated sugar chains that have compatible groups for conjugation allowing the polypeptide and moiety to be conjugated to each other so joining them covalently together. In one embodiment, the desired moiety for conjugation is, or comprises, a linker which can be conjugated to the chemical group of the sialylated sugar chain at a glycosylation amino acid sequence of the invention. In one embodiment, a molecule of the invention comprises a glycosylation amino acid sequence which has been so conjugated to a linker. In another embodiment, a molecule of the invention comprises a glycosylation amino acid sequence conjugated to a desired moiety comprising a linker which itself is conjugated to a second molecule forming a further part of the desired moiety. [0115] Advantages of the present invention may include that for a molecule comprising a glycosylation sequence of the present invention there is typically a very high proportion of the molecules which will have the glycosylation sequence glycosylated. This may be referred to as “site occupancy”. In one embodiment, the proportion of the glycosylation sites of the present invention which are glycosylated will be at least 60%, at least 70%, preferably at least 80%, and more preferably at least 90%. In one embodiment, the proportion which are glycosylated will be at least 95%. In another embodiment, the proportion will be at least 99%. A further advantage of the invention is that the sugar chain in each molecule for the O-glycosylation will be usually the same. Such consistency is an advantage for drug production. In one embodiment, at least 60%, at least 70%, preferably at least 80%, and more preferably at least 90% of the molecules in a sample of a molecule of the present invention will have the same sugar chain at the glycosylation sequence or sequences of the present invention that are present in the molecule. In another embodiment, at least 95% of the sugar chains will be the same. In another embodiment, at least 99% will be the same.

[0116] In one embodiment, the present invention may be used to stabilize a molecule. For example, in one embodiment one or more pairs of glycosylation amino acid sequences of the present invention are conjugated to each other with that stabilizing the molecule. In another embodiment, the present invention may be used to prevent disassociation, for example between two polypeptides.

[0117] Conjugation groups and conjugation

[0118] In a particularly preferred embodiment, the O-glycosyl sugar chain at a glycosylation sequence of the present invention comprises a chemical group which acts as a conjugation group. The first conjugation group can be conjugated to a second compatible conjugation group allowing a means to join a desired moiety to the polypeptide. The conjugation is typically covalent. Typically, the first and second conjugation groups are not identical but can be conjugated to each other. The conjugation chemistry employed is typically bioorthogonal. Preferably the conjugation can therefore take place under physiological conditions. In one embodiment, the conjugation can take place without needing an external catalyst, for instance without needing the addition of exogenous copper. In one embodiment, the conjugation can take place in a cell without needing the addition of any further reagent to bring about the conjugation.

[0119] An especially preferred means for conjugation is click chemistry. As used herein, the term “click chemistry group” or “click chemistry handle” refers to a reactant or a reactive group that can partake in a click chemistry reaction. A click reaction group may be a moiety that is rarely, or never, found in naturally occurring biomolecules and is chemically inert towards biomolecules. In one embodiment, the click chemistry groups are azide-reactive or alkyne- reactive groups. Such groups can react efficiently under biologically relevant conditions, for example in cell culture conditions, without requiring excess heat or harsh reactants. In one embodiment, the click chemistry reaction takes place in a cell. In one embodiment, it takes place in vitro. In another embodiment, it takes place ex vivo. In another embodiment, the conjugation takes place in vivo. In another embodiment, it may take place ex vivo. In one embodiment, the invention comprises a transgenic animal that encodes and expresses a molecule of the present invention. In one embodiment, the animal is any of those mentioned herein. In one embodiment, the entities to be conjugated are simply mixed, for instance in isolated form.

[0120] In general, click chemistry reactions require at least two molecules comprising click reaction partners that can react with each other. Such click reaction partners that are reactive with each other are sometimes referred to as click chemistry handle pairs or click chemistry pairs. In some embodiments, the click reaction partners are reactive alkenes or alkynes and suitable tetrazines. For example, trans-cyclooctene, norbomene, or bicyclononyne can be paired with a suitable tetrazine as a click reaction pair. In other embodiments, tetrazoles can act as latent sources of nitrile imines, which can pair with unactivated alkenes in the presence of ultraviolet light to create a click reaction pair, termed a “photo-click” reaction pair. In some embodiments, the click reaction partners are an azide and an alkyne, in particular a strained alkyne, e.g. a cyclooctyne, or any other alkyne. [0121] Other suitable click chemistry handles are known to those of skill in the art (see, for instance Spicer et al., 2014, Nature Communications, 5: page 4740). In other embodiments, the click reaction partners are Staudinger ligation components, such as phosphine and azide. In other embodiments, the click reaction partners are Diels- Alder reaction components, such as dienes, such as tetrazine, and alkenes, such as trans-cyclooctene (TCO) or norbornene. Exemplary click reaction partners are described, for instance, in US2013/0266512 and in WO2015/073746, both of which are incorporated by reference in their entirety as well as specifically in relation to the relevant description on click reaction partners in both of which are incorporated by reference herein. According to preferred embodiments, one of the first and second click reaction partners comprises an alkyne group, and the other click reaction partner comprises an azide. According to other preferred embodiments, one of the first and second click reaction partners comprises an alkene group, and the other click reaction partner comprises a diene.

[0122] As used herein, the term “alkyne”, “alkyne group” or “alkyne moiety” refers to a functional group comprising a carbon-carbon triple bond. Alkyne moieties include terminal alkynes and cyclic alkynes, preferably terminal alkynes and cyclic alkynes that are reactive with azide groups. A terminal alkyne has at least one hydrogen atom bonded to a triple-bonded carbon atom. A cyclic alkyne is a cycloalkyl ring comprising one or more triple bonds. Examples of cyclic alkynes include, but are not limited to, cyclooctyne and cyclooctyne derivatives, such as bicyclononyne (BCN), difluorinated cyclooctyne (DIFO), dibenzocyclooctyne (DIBO/DBCO), keto- DIBO, biarylazacyclooctynone (BARAC), dibenzoazacyclooctyne (DIBAC), dimethoxyazacyclooctyne (DIMAC), difluorobenzocyclooctyne (DIFBO), monobenzocyclooctyne (MOBO), and tetramethoxy DIBO (TMDIBO).

[0123] In a particularly preferred embodiment, one of the first and second click reaction partners comprises a cyclic alkyne, preferably DBCO. DBCO is a particularly preferred conjugation group. According to preferred embodiments, the other click reaction partner comprises an azide, Hence, a particularly preferred “click pair” is DBCO with an azide. [0124] As used herein, the term “diene” refers to a compound having two carbon- to-carbon double bonds where these double bonds are conjugated in the Imposition. The double bonds of the diene can be either cis or trans. Examples of dienes include, but are not limited to, a tetrazine or a tetrazole group.

[0125] As used herein, the term “alkene”, “alkene group” or “alkene moiety” refers to an unsaturated hydrocarbon molecule that includes a carbon-carbon double bond. In one embodiment, an alkene can include from 2 to 100 carbon atoms. Examples of alkenes include, but are not limited to, norbornene and transcyclooctene (TCO). According to other preferred embodiments, one of the first and second click reaction partners comprises an alkene group, preferably norbomene or TCO. According to preferred embodiments, the other click reaction partner comprises a diene, preferably a tetrazine or tetrazole group.

[0126] As used herein, “proteinogenic amino acids” are amino acids that are incorporated biosynthetically into proteins during translation and preferably the genetically encoded (proteinogenic) amino acids, 20 in the standard genetic code and an additional 2 (selenocysteine and pyrrolysine) that can be incorporated by special translation mechanisms.

[0127] As used herein, the term “covalently linked” means that the molecule is attached to the first click functional group via at least one covalent linkage, and that the conjugation partner is attached to the second click functional group via at least one covalent linkage. The linkage can be direct, i.e. without a linker, or indirect, i.e. via a linker.

[0128] One advantage of the present invention is that it will typically allow the entities being conjugated to retain their activity or not suffer a significant reduction in activity. In one embodiment, the location of the glycosylation site or sites is chosen to help avoid any loss of activity, or at least any significant loss, in the entities being conjugated to each other. Hence, preferably the moiety that it is desired to conjugate to will still retain activity after the conjugation. For example, where the molecule is, or comprises, an antibody and is conjugated to a desired moiety using the present invention the antibody will typically retain antigenbinding activity. [0129] Linkers

[0130] In one embodiment, the moiety may be, or comprise, a linker, so that the O-glycosyl sugar chain is conjugated to a linker as, or as part of, the moiety. The linker may then be, or may already be, conjugated to a further component. Hence, in any of the embodiments set out herein there may be a linker which is a bridge between the O-glycosyl sugar and a further molecule with the linker and further molecule representing the desired moiety. Alternatively, in other embodiments such a linker may be absent. A linker may be used as the bridge between an O- glycosylated amino acid sequence of the present invention and a further molecule. Hence, for instance, the linker may be joined to the sialylated sugar and the other end of the linker joined to the chosen molecule. Hence, in any of the embodiments described herein, such a linker may be used.

[0131] An example of a preferred linker is, or one which comprises, DBCO.

[0132] Any suitable linker may be used that is capable of forming a bond with the

O-glycosylated sugar. In one preferred embodiment, a linker comprising a polyarginine sequence may be employed, for instance a linker with from 2 to 15, preferably from 3 to 10, and more preferably from 4 to 8 consecutive arginine residues. In a particularly preferred embodiment, the linker comprises six consecutive arginine residues. In an especially preferred embodiment, the polyarginine sequence is flanked on each side by a lysine residue. In an especially preferred embodiment, the sequence is acetylated on the N-terminus and amidated on the C-terminus. In one preferred embodiment, the linker comprises the amino acid sequence Lys-Arg-Arg-Arg-Arg-Arg-Arg-Lys with an N-terminal acetyl group and C-terminal amidation, in other words Ac-Lys-(Arg)6-Lys-NH2. In one particularly preferred embodiment, the linker has the structure shown in Figure 28(c). Hence, in one embodiment, the linker comprises a Lys-(Arg)6-Lys sequence with both lysine side chains modified by PEG5-DBCO, so Ac- Lys(PEG5-DBCO)-(Arg)6-Lys(PEG5-DBCO)-NH2. In one preferred embodiment, the linker comprises, or is, that depicted in Figure 28(c). [0133] In one particularly preferred embodiment, two molecules of the invention are joined via their O glycosylation sites via a linker and in particular that described above to form a larger molecule of the invention.

[0134] Moieties

[0135] In a preferred embodiment, a molecule of the present invention comprises a glycosylation sequence of the present invention with a sialylated O-glycosyl sugar chain that is conjugated to a desired moiety. In some embodiments, the desired moiety may be, or comprise, a linker.

[0136] Hence, the present invention may be used to join any desired entities using one or more glycosylation sequences of the present invention which are O- glycosylated with a sialylated sugar chain comprising a suitable conjugation group. The invention may be, for instance, used to join polypeptides, which themselves represent molecules of the invention, to each other to form a larger molecule of the present invention. In one embodiment, a polypeptide refers to the segment that may ultimately form part of a molecule of the invention when it is has been conjugated to the polypeptide comprising one or more glycosylation sequence(s) of the present invention. For instance, the ability to conjugate together two polypeptides may be used to join polypeptide chains within a larger molecule. As discussed further below, the ability to readily join two molecules via the invention may be used in multiplexing to generate permutations of different molecules being joined together and then screen them for a desired property. The invention is therefore a highly versatile way to join-together a molecule and a desired moiety. Particular moieties are described below but reference to them should not be seen as limiting and particular moieties are identified simply as illustrative examples.

[0137] In one embodiment, the moiety is selected from Fc, PEG, fluorophore, radioactive tracer, a fatty acid, glycan, peptide, nucleic acid, enzyme, and steroid.

[0138] In one embodiment, a molecule of the present invention is conjugated to, or comprises via conjugation, a nucleic acid. In one embodiment, the nucleic acid is single-stranded. In another embodiment, it is double-stranded. In one embodiment, the nucleic acid is DNA. In another embodiment, the nucleic acid is RNA. In one embodiment, the nucleic acid is an anti-sense nucleic acid. In one embodiment, the nucleic acid is an anti-sense RNA. In an especially preferred embodiment, the nucleic acid is an siRNA. In one particularly preferred embodiment, a conjugate of the invention comprises an antibody which is conjugated to a nucleic acid molecule via the invention. In one preferred embodiment, the nucleic acid may inhibit expression of a target gene. In a preferred embodiment, the invention provides a conjugate which is a conjugate of an antibody and an siRNA using the conjugation approach of the present invention. Hence, in such embodiments, the antibody will comprise glycosylation sequences of the present invention allowing conjugation to a nucleic acid of choice.

[0139] In one embodiment, the molecule may be, or may comprise, a viral or microbial polypeptide. In one embodiment, a virus may be produced such that it comprises one or more glycosylation sequences of the present invention in a capsid polypeptide of the virus. In another embodiment, a molecule of the invention may be one on the surface of a cell. In another embodiment, the molecule is on the surface of a pathogen. In one embodiment, the invention may be then used to conjugate a desired molecule onto the surface of the virus, pathogen, or cell. The invention may be used to bring together different cells where both cells have compatible conjugation groups with at least one of the cells having such groups introduced via use of glycosylation sequences of the present invention. The invention may also be used to help target a virus to a desired cell. The invention may also be used to target to cancer cells. In one embodiment, a molecule of the present invention may be first targeted to a desired cell, with the molecule having a compatible conjugation group to then allow conjugation to a target on the cell. In another embodiment, cells, viruses, liposomes, or other means for delivery may have a sialylated O-glycosyl sugar of the invention on their surface, with that then being used for conjugation to a desired targeting molecule. The targeting molecule may be chosen for its specificity, with it being possible to swap to a different targeting molecule to change the specificity. In one embodiment, the targeting molecule is an antibody specific for a target cell. Hence, in preferred embodiment, a molecule of the present invention may comprise a portion of the molecule that targets it to a specific target.

[0140] In another embodiment, the molecule may be, or comprise, an antigen. In one embodiment, the antigen it is desired to elicit an immune response against has been modified to comprise one or more glycosylation sequence(s) of the present invention, and the invention is used to conjugate the antigen to a desired moiety. In one embodiment, the moiety is a carrier. In one embodiment, the invention is used to conjugate an antigen to a carrier such as diphtheria toxoid (DT), tetanus toxoid (TT), CRM197, Haemophilus protein D (PD), or the outer membrane protein complex of serogroup B meningococcus (OMPC). The present invention also provides a conjugate vaccine comprising an antigen and a protein carrier joined using the conjugation approach of the present invention. In one embodiment, the polypeptide comprising one or more glycosylation sequences of the present invention is the carrier and the moiety conjugated to it using the invention is a glycan. Hence, for instance, the invention provides a convenient way to conjugate glycans to a polypeptide carrier. In another embodiment, the invention may be used to conjugate alpha gal to a polypeptide to help promote an immune response to the polypeptide with one or more glycosylation sequence(s) of the present invention. Thus, the invention may be used to promote the efficacy of a vaccine.

[0141] In one embodiment, a molecule of the present invention comprises one or more glycosylation sequence(s) of the invention which is, or are, conjugated to a label. In one embodiment, the molecule comprises a fluorophore conjugated to a glycosylation sequence of the present invention. In one embodiment, the label is a fluorescent protein. In another embodiment, the fluorescent protein is GFP. In one embodiment, the polypeptide sequence is conjugated to a dye using the glycosylation sequence of the invention. In another embodiment, the polypeptide is conjugated to biotin via the O-glycosylation. In another embodiment, the polypeptide is conjugated to a steroid. In one embodiment, the steroid is a steroid hormone. In another embodiment, it is an androgen. In another embodiment, it is a cortisone. In another embodiment, it is an anabolic steroid. In a further embodiment, the molecule is conjugated to a moiety which influences the stability and/or half-life of the molecule. For instance, in one embodiment, the glycosylation amino acid sequence is conjugated to BSA, for instance to alter the stability of the molecule in circulation. In one embodiment, the molecule is conjugated via a glycosylation sequence of the present invention to PEG (polyethylene glycol). The invention may also be used to change the isoelectric point (pl) of a molecule of the invention, for instance as the addition of sialic acid should result in a lower pl. That may be, for instance, used to increase the half-life of a molecule of the present invention.

[0142] In another preferred embodiment, a glycosylation amino acid sequence of the invention may be conjugated to an anti-cancer agent. In one embodiment, the anti-cancer agent is Monomethyl auristatin E (MMAE). In one embodiment, the glycosylation amino acid sequence is conjugated to a chemotherapeutic agent. Examples of chemotherapeutic agents which may be conjugated include: Lenalidomide (REVLIMID®, Celgene), Vorinostat (ZOLINZA®, Merck), Panobinostat (FARYDAK®, Novartis), Mocetinostat (MGCD0103), Everolimus (ZORTRESS®, CERTICAN®, Novartis), Bendamustine (TREAKISYM®, RIBOMUSTIN®, LEV ACT®, TREANDA®, Mundipharma International), erlotinib (TARCEVA®, Genentech/OSI Pharm.), docetaxel (TAXOTERE®, Sanofi- Aventis), 5-FU (fluorouracil, 5 -fluorouracil, CAS No. 51-21-8), gemcitabine (GEMZAR®, Lilly), PD-0325901 (CAS No. 391210-10-9, Pfizer), cisplatin (cis-diamine, dichloroplatinum(ll), CAS No. 15663-27-1), carboplatin (CAS No. 41575-94-4), paclitaxel (TAXOL®, Bristol-Myers Squibb Oncology, Princeton, N.J.), trastuzumab (HERCEPTIN®, Genentech), temozolomide (4- methyl-5-oxo- 2,3,4,6,8-pentazabicyclo [4.3.0] nona-2,7,9- triene- 9-carboxamide, CAS No. 85622-93-1, TEMODAR®, TEMODAL®, Schering Plough), tamoxifen ((Z)-2-[4-(l ,2-diphenylbut-l-enyl)phenoxy]-/V,/V-dimethyl ethanamine, NOLVADEX®, ISTUBAL®, VALODEX®), and doxorubicin (ADRIAMYCIN®), Akti-1/2, HPPD, and rapamycin. In another embodiment, the invention may be used to form a molecule of the present invention comprising a radionucleotide, for example for use in cell killing or alternatively in labelling or imaging.

[0143] In one embodiment, a molecule of the invention may comprise a label, for example the label may be conjugated to the molecule via a glycosylation amino acid sequence of the invention. Such labels may be useful, for instance, in diagnostics and/or imaging. In one embodiment, the label is a fluorochrome. In another embodiment, it is a radiolabel.

[0144] In one embodiment, the moiety is a polypeptide or peptide. The invention may be used to bring together any suitable polypeptides. In one embodiment, the conjugation helps stabilize a molecule comprising at least two polypeptides. In another embodiment, it is used to help stabilize a dimeric or multimeric protein via conjugation between glycosylation sequences of the present invention location in different polypeptides. In another embodiment, the conjugation joins together subunits of a protein to make it more stable. In one embodiment, it is used to join a ligand to its receptor, for instance after binding of the ligand to its receptor.

[0145] In one embodiment, the invention is used to provide a way to recover a molecule of the invention by virtue of its ability to conjugate to a compatible conjugation group. In one embodiment, the compatible conjugation group is provided by a support, for instance such as a bead or plate. In another embodiment, the invention is used to localize a molecule at the site of the cell producing it, for example where the molecule is labelled and conjugation to the support provides a localized region of labelled molecule or simply a localized region of the molecule that can be detected using a secondary label. In another embodiment, the invention may be used in techniques such as ELISA to immobilize a molecule of interest to a support prior to detection.

[0146] Nucleic acid sequences and vectors

[0147] Also provided is a nucleic acid sequence or sequences at least encoding a molecule of the present invention or a polypeptide part of it. Hence, the invention provides a nucleic acid molecule encoding a polypeptide comprising one or more glycosylation sequences of the present invention. In one preferred embodiment, the present invention provides a nucleic acid molecule or molecules encoding a molecule of the present invention which is, or comprises, an antibody of the present invention. The nucleic acid will typically further comprise regulatory elements for expression of the polypeptide or protein of the present invention such as, for instance, a promoter and polyadenylation addition signal.

[0148] The present invention also provides a vector comprising such a nucleic acid or nucleic acids of the present invention. The vector may be, for instance, an expression vector. The vector may be a cloning vector. The invention also provides a sequence for introducing a nucleic acid sequence of the present invention into the genome of a cell via CRISPR.

[0149] Binding molecules and antibodies

[0150] The present invention is highly versatile and may be used to join any desired molecules provided at least one of them comprises one or more glycosylation sequences of the present invention. The glycosylation sequence(s) may be either inserted into the amino acid sequence forming part of a molecule of the invention, such as a polypeptide, or used to replace part of the original sequence. In one particularly preferred embodiment, the invention is applied to binding molecules and in particular antibodies. The present invention provides a binding molecule comprising one or more glycosylation sequences of the present invention. In particular, the present invention provides such a molecule where one or more glycosylation sequence is O-glycosylated at the Threonine amino acid residue, where the O-glycosylation comprises a sialylated sugar with a group allowing conjugation to a further moiety. An especially preferred binding molecule of the present invention is, or comprises, an antibody.

[0151] A particularly preferred molecule of the present invention is an antibody conjugate. Hence, the present invention provides an antibody wherein at least one polypeptide of the antibody comprises one or more glycosylation sequences of the present invention. In particular, the present invention provides such an antibody comprising at least one glycosylation sequence where the Threonine residue is O- glycosylated. In a preferred embodiment, the O-glycosylated sugar comprises a sialylated sugar group. In a preferred embodiment, the O-glycan comprises a sialylated sugar with a chemical group which is functional for conjugation. In one preferred embodiment, the antibody comprises at least one glycosylation sequence of the present invention where the threonine residue is O-glycosylated, with the O-glycosyl sugar group being conjugated to another molecule. For instance, such an antibody can be represented by the formula: Ab-O-DM, where Ab denotes the antibody, O the O-glycosyl sugar chain and DM the desired moiety conjugated to the antibody.

[0152] In one embodiment, a linker (L) may be used to join an O-glycan and a chosen molecule (the linker itself may be thought of as part of the desired moiety). In one embodiment, the molecule may have the format Ab-O-DM, wherein the DM comprises a linker and a second molecule. In one embodiment, the molecule may have more than one glycosylation sequence of the invention. For example, it may have two such sequences and take the format DM-O-Ab-O-DM. Again, the DM may, or may not, comprise a linker or be a linker. In one embodiment, the binding molecule may be symmetrical, for instance it may have two glycosylation sites at the same site in the heavy chains of the antibody and/or in the same locations of the light chains of the antibody.

[0153] The present invention may be applied to any antibody format which can comprise a glycosylation sequence of the present invention. Hence, a molecule of the present invention may be, or comprise, an antibody. In one preferred embodiment, the antibody is one comprising two light chains and two heavy chains. In one embodiment, at least the light chain variable regions of the two light chains are identical to each other. In one embodiment, at least the two heavy chain variable regions of the two heavy chains are identical to each other. In one embodiment, the light chain variable regions of the two light chains are identical to each other and the heavy chain variable regions of the two heavy chains are identical to each other. The invention though may be used to bring together antigen binding sites of different specificities as well, such as a light and heavy chain for a first specificity and a different heavy and light chain pair for a second specificity. In some embodiments the light chains in both pairs are the same, but the heavy chains differ leading to the different specificity. That is discussed further below.

[0154] In one embodiment, the antibody comprises at least two glycosylation sequences of the present invention. In one embodiment, the antibody comprises at least four such sequences. In one embodiment the antibody comprises two or four glycosylation sequences of the present invention. In one embodiment, the antibody comprises at least two glycosylation sequences of the present invention where one of the sequences is present in each heavy chain. In one embodiment, the antibody comprises at least two glycosylation sequences of the present invention, where one such amino acid sequence is present in each light chain. In one embodiment, an antibody comprises a glycosylation sequence of the present invention in each of the two light chains and two heavy chains. In one embodiment, the glycosylation sequence or sequences of the present invention is, or are, present in the antibody constant region. In one embodiment, the glycosylation sequences of the present invention are only present in the antibody constant regions. In another embodiment a glycosylation sequence of the present invention is present in the light or heavy chain variable regions of the antibody, but in the frame-work regions of the antibody. In one embodiment, the glycosylation sequences present are in the C-terminal half of the light chain. In another embodiment, the glycosylation sequences present are in the C-terminal half of the heavy chain. In another embodiment, the glycosylation sequences present are in the N-terminal half of the heavy chain. In another embodiment, the glycosylation sequences present are in the N-terminal half of the light chain. In one embodiment, the glycosylation sequences present are at the C-termini of the light chains. In one embodiment, the glycosylation sequences present are at the C- termini of the heavy chains. In one embodiment, the glycosylation sequences present are at the C-termini of the light and heavy chains of the antibody. In one particularly preferred embodiment, the glycosylation sequence, or sequences, of the present invention are present in the hinge regions of the heavy chains of the antibody. [0155] An exemplary antibody is an immunoglobulin G (IgG) type antibody comprised of four polypeptide chains: two heavy chains (HC) and two light chains (LC) that are cross-linked via inter-chain disulfide bonds. The amino-terminal portion of each of the four polypeptide chains includes a variable region of about 100-125 or more amino acids primarily responsible for antigen recognition. The carboxyl-terminal portion of each of the four polypeptide chains contains a constant region primarily responsible for effector function. Each heavy chain is comprised of a heavy chain variable region (VH) and a heavy chain constant region. The heavy chain constant region refers to a region of an antibody, which comprises the Fc region and CHI domain of the antibody heavy chain. Each light chain is comprised of a light chain variable region (VL) and a light chain constant region. The IgG isotype may be further divided into subclasses (e.g., IgGl, IgG2, IgG3, and IgG4). The numbering of the amino acid residues in the constant region is based on the EU index as in Kabat. Kabat et al, Sequences of Proteins of Immunological Interest, 5th edition, Bethesda, MD: U.S. Dept, of Health and Human 20 Services, Public Health Service, National Institutes of Health (1991). The term EU Index numbering or EU numbering is used interchangeably herein.

[0156] The VH and VL regions can be further subdivided into regions of hypervariability, termed complementarity determining regions (CDRs), interspersed with regions that are more conserved, termed framework regions (FR). The CDRs are exposed on the surface of the protein and are important regions of the antibody for antigen binding specificity. Each VH and VL is composed of three CDRs and four FRs, arranged from amino-terminus to carboxyl-terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. Herein, the three CDRs of the heavy chain are referred to as “HCDR1, HCDR2, and HCDR3” and the three CDRs of the light chain are referred to as “LCDR1, 30 LCDR2 and LCDR3”. The CDRs contain most of the residues that form specific interactions with the antigen. Assignment of amino acid residues to the CDRs may be done according to the well-known schemes, including those described in Kabat (Kabat et al., “Sequences of Proteins of Immunological Interest,” National Institutes of Health, Bethesda, Md. (1991)), Chothia (Chothia et al., “Canonical structures for the hypervariable regions of immunoglobulins”, Journal of Molecular Biology, 196, 901-917 (1987)); Al-Lazikani et al., “Standard conformations for the canonical structures of 5 immunoglobulins”, Journal of Molecular Biology, 273, 927-948 (1997)), North (North et al., “A New Clustering of Antibody CDR Loop Conformations”, Journal of Molecular Biology, 406, 228-256 (2011)), or IMGT (the international ImMunoGeneTics database available on at www.imgt.org; see Lefranc et al., Nucleic Acids Res. 1999; 27:209-212). The North CDR definitions are used for the exemplified anti-human CD33 antibodies as described herein.

[0157] The present invention may also be applied to other antibody formats, for example the antibody may be a heavy chain only antibody. The antibody may be a VHH format antibody. The antibody may be a camelid antibody. The antibody may be from a cartilaginous fish, for example be an IgNAR format antibody. The antibody may be a heavy chain only antibody that comprises heavy chains lacking CH3 regions. The antibody can be in a tandem configuration with other VHH of the same or of different identity, e.g. VHH comprising of a half-life extender that binds to human serum albumin (HSA). The invention may be applied to antibodylike molecules, for instance DARPins and affibodies.

[0158] Given that the present invention provides a way to join two molecules which each comprise the glycosylation sequence of the present invention, the present invention may be used to join separate antigen-binding sites. In one embodiment, the present invention is employed to join two antibodies together to produce an antibody with a higher valency, i.e., with a higher number of antigenbinding sites. The joining may be direct or it may be that two molecules each comprising antigen-binding sites are brought together with the molecules comprising other scaffold sequences as well as the antigen binding sites. The “valency” of a binding molecule is typically the number of binding sites present in the binding molecule. For example, in the case of an antibody the “valency” of an antibody indicates the number of antigen-binding sites that the antibody has. Reference to an “antibody” herein includes a structure comprising individual antibodies joined together with the “O” glycosyl groups acting as a “bridge” joining the antibodies. For instance, the present invention provides an antibody where a further antibody is conjugated to each of the two light chains. The present invention also provides an antibody where a further antibody is conjugated to each of the two heavy chains. In one embodiment, the further antibody is conjugated at, or near to, the C-terminal end of the polypeptide chain they are joined to. In one embodiment, the further antibodies conjugated to the antibody molecule are scFv antibodies or heavy chain only antibodies. So, for instance, the present invention provides an Ig molecule where the valency of the antibody has been increased from two to four by conjugating an antibody to each of the two light or two heavy chains. In one embodiment, a molecule of the present invention is a multi-specific antibody. In a preferred embodiment, it is a bispecific antibody. The present invention also provides for the use of a molecule of the present invention in the generation of a multi-specific antibody, particularly a bispecific antibody.

[0159] In an alternative embodiment, rather than j oining a further antibody to the initial antibody, the invention is used to join a ligand to the antibody, so the ligand effectively represents a moiety. In one embodiment, the ligand is specific for the same target as the antigen-binding sites of the antibody. The present invention also provides a method of joining two antigen-binding sites together by employing the glycosylation sequence of the present invention to form a bridge between the two antigen-binding sites.

[0160] The present invention also provides a method of modifying a known antibody by introducing one or more glycosylation sequences of the present invention into the primary sequence of the antibody. The glycosylation sequence of the present invention can be introduced in any of the locations discussed above. For example, known antibodies such as Humira®, Herceptin®, Avastin®, Keytruda®, Rituximab®, Remicade®, Stelara®, Enbrel®, Imbruvica®, Opdivo®, Cosentrx®, Ocrevus®, and other known antibodies may be modified to introduce one or more glycosylation sequences of the present invention and hence provide a convenient way to conjugate desired moieties to those antibody drugs.

[0161] The present invention also provides a convenient way to produce antibody dimers or multimers by providing a way to join antibodies via glycosylation sequences in each. In one embodiment, the present invention provides an antibody dimer where the two antibodies each have one or more glycosylation sequences of the present invention allowing for conjugation of the two individual antibodies. In another embodiment, the invention is used to form antibody trimers. In another embodiment, it is used to form antibody hexamers. It may be that all the individual antibodies within a dimer or multimer are identical apart from the conjugation groups. For example, it is possible to produce two batches of the same antibody only differing in the conjugation group in the O-glycosyl sugar chain at the glycosylation sequence of the present invention, with the two batches having different but compatible conjugation groups. Antibodies from the two batches can then be conjugated together to form dimers. In another embodiment, antibodies with different specificities are joined together in dimers or higher order structures using the invention.

[0162] In one particularly preferred embodiment, the antibody binds to a target that allows for transport across the blood brain barrier (BBB). In one embodiment, such an antibody conjugate may be used to deliver a moiety to the CNS and in particular the brain. In one embodiment, the moiety may be for inhibiting expression of a target gene. In one particularly preferred embodiment, the moiety may be for inhibiting expression of a target gene in the brain. In an especially preferred embodiment, the antibody specific for a target that allows for transport across the blood brain barrier (BBB) may be conjugated using the invention to a moiety which is an siRNA. Antibodies may also be conjugated to a chemotherapeutic agent or other anti-cancer agent which otherwise would not be able to cross the BBB readily. Such conjugates may be, for instance, used to treat cancer.

[0163] Bispecific and multi-specific binding molecules

[0164] A binding molecule of the present invention, and in particular an antibody, may have, for instance, a valency of one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, or more. It may have a valency in the range of those values. It may have a valency of at least those values. It may have a valency up to, or including, those values. In one preferred embodiment, an antibody may have a valency of two. In one embodiment, all the antigen-binding sites of the binding molecule, and in particular, antibody may have the same specificity. In an alternative embodiment, an antibody may comprise at least two different antigenbinding sites which have a different specificity. The antigen-binding sites may have a different specificity in the sense that the two bind different antigens. In another embodiment, they may have a different specificity in the sense that they bind two different epitopes on the same antigen.

[0165] In one embodiment, the present invention provides a bispecific antibody comprising one or more glycosylation sequences of the present invention. In a preferred embodiment, the bispecific antibody comprises at least two glycosylation sequences of the present invention which are O-glycosylated at the threonine, with the sugar chains being conjugated through the sialic acid groups of the O-glycans, forming a bridge between two different antigen-binding sites. In one embodiment, the bispecific antibody comprises an Ig antibody comprising two heavy and two light chains, wherein each of the light chain constant regions comprise a glycosylation sequence of the present invention which is conjugated to a Fab comprising an antigen-binding site with different specificity to that of the antigen-binding sites of the Ig antibody. In such an embodiment, typically each of the Fab fragments conjugated to the Ig is the same and has the same antigenspecificity, but that specificity is different to that of the antigen-binding sites of the Ig. In another embodiment, the Fab fragments are conjugated to the heavy chains.

[0166] In one particularly preferred embodiment, a bispecific antibody is provided comprising two heavy and two light chains, where each heavy chain comprises a glycosylation sequence of the present invention which is O-glycosylated, with the two sugar chains conjugated to each other, with the antigen-binding site formed by the first light chain and first heavy chain having a different specificity to that formed by the second light chain and second heavy chain. In one preferred embodiment, the glycosylation sequence of the present invention is substituted for the cysteines present in the heavy chain hinge. [0167] In one preferred embodiment a bispecific antibody of the present invention is generated via antibody arm exchange. A well -recognized problem with bispecific antibodies is that if a bispecific antibody is formed of two different light chains and two different heavy chains, expressing all four antibody chains in the same cell results not only in the bispecific antibody, but also in a much larger proportion of unwanted monospecific antibodies and other unwanted antibody species. In one embodiment, the present invention provides a solution to promote antibody arm exchange between the two “parent” monospecific antibody species to generate bispecific antibodies.

[0168] Hence, the present invention provides a method for the generation of a bispecific antibody, the method comprising:

(a) contacting a first antibody with a first specificity and a second antibody with a second specificity, wherein the first and second antibody each comprise one or more glycosylation sequences of the present invention which is O-glycosylated with a sialylated sugar chain with a conjugation group, with the first and second antibodies comprising compatible conjugation groups that can conjugate to each other; and

(b) allowing the first and second antibodies to undergo antibody arm exchange and conjugation between the complementary functionalized sialylated sugar chains to generate a bispecific antibody, wherein the resultant bispecific antibody comprises an antigen-binding site from the first antibody and an antigen-binding site from the second antibody and hence two antigen binding sites with two different specificities.

[0169] In a preferred embodiment, the first and second antibody comprise one or more glycosylation sequences of the present invention in the heavy chain constant regions of the antibodies. A glycosylation sequence of the present invention is typically at a location in the heavy chains that will allow antibody arm exchange. In one preferred embodiment, the glycosylation sequences of the present invention are present in place of a section of the sequence containing the cysteines usually present at the heavy chain hinge. In a different embodiment, the cysteines usually present at the heavy chain hinge are replaced with serine residues and one or more glycosylation sequences of the present invention are present elsewhere, for example at the C-terminal part of the heavy chains.

[0170] In one embodiment, the method may further comprise purifying the bispecific antibody. For instance, in one embodiment, the method may comprise the further steps of:

(c) allowing the antibody to bind to a support comprising the antigen or epitope for one of the specificities of the bispecific antibody, removing the unbound antibody, then recovering the antibody that has bound to the antigen or epitope present on the support; and

(d) allowing the antibody to bind to a support comprising the antigen or epitope for the other of the specificities of the bispecific antibody, removing the unbound antibody, then recovering the antibody that has bound to the antigen or epitope present on the support.

[0171] In one embodiment, the first and second antibodies differ not only in the heavy chain variable regions, but also in the heavy chain constant regions. For example, the heavy chain constant regions of the first and second antibodies may comprise amino acid sequence differences that either promote the formation of bispecific antibody (heterodimer) over monospecific antibodies (monomer) or which help the purification of the bispecific antibodies from the monospecific/unwanted species of antibody. In one preferred embodiment, the heavy chains have differences in their heavy chain constant regions that promote bispecific antibody formation over monospecific antibody. For example, the heavy chain constant regions may have charge and/or shape modifications that promote bispecific formation over monospecific antibody. An example of heavy chain modifications that promote the formation of bispecific antibody over monospecific antibody are the “knob-and-hole” amino acid modifications. Hence, in one embodiment the first or second antibody has the heavy chain constant region “knob” modification and the other of the first and second antibody has the “hole” modification.

[0172] Another example of a modification that may be employed is one that alters affinity for a purification agent. For instance, one parental monospecific may have a heavy chain constant region modification that changes the affinity of the antibody for purification agent Protein A, for instance eliminating binding to Protein A. The other parental monospecific may have heavy chains that lack that modification and so bind Protein A normally. The bispecific antibody may therefore comprise one heavy chain which binds to Protein A and the other that either does not bind to Protein A or has reduced affinity for Protein A. Overall, that means the bispecific antibody has an intermediate affinity for Protein A meaning it can be separated from the parental monospecific antibodies on that basis.

[0173] Cells and cell lines

[0174] In a further preferred embodiment, the present invention provides a cell line where the ability of the cell to produce sialic acid has been disrupted so that the cell has reduced, or no ability, to naturally produce sialic acid. Such cell lines may also be employed in the present invention. Such cell lines can be then supplemented with synthetic compounds that include a functional group for conjugation which are preferentially incorporated when the polypeptides comprising one or more glycosylation sequence of the present invention is O- glycosylated, in particular they may be supplemented with a peracetylated ManNAz compound that gets taken up, de-acetylated, and converted to sialic acid which includes a click chemistry group. This provides a convenient way to ensure that the click chemistry groups are incorporated as part of the O-glycosylation of the one or more glycosylation sequences of the present invention.

[0175] The present invention therefore provides a cell line that is unable itself to produce sialic acid, unless supplemented, in particular where the cell line is unable to produce sialic acid unless supplemented with peracetylated ManNAz that can be taken up, de-acetylated, and converted to sialic acid. It is possible to disrupt the endogenous synthesis of sialic acid in a eukaryotic cell by mutating the UDP- GlcNAc-2-epimerase/ManAc kinase (GNE) encoding gene (SEQ ID NO: 6, NCBI accession number NM_001246709). Unexpectedly, it has been found that the epimerase function of the gene can be reduced without the need to either retain the kinase function or introduce a sequence encoding a kinase to compensate for its loss. In one embodiment, the sequence encoding the UDP-N-acetylglucosamine 2-epimerase portion of the GNE is mutated so that the UDP-N-acetylglucosamine 2-epimerase function is either reduced or eliminated. In one embodiment, the N- acetylmannosamine kinase function is also reduced or eliminated. The finding that the N-acetylmannosamine kinase function does not need to be retained or reintroduced means that producing the cell line is simpler. In one embodiment, whilst UDP-N-acetylglucosamine 2-epimerase is knocked-out, the N- acetylmannosamine kinase encoding sequence is not modified and thus the kinase is still active. The cell will typically comprise a polynucleotide encoding a polypeptide comprising one or more glycosylation sequences of the present invention. The present invention also provides a method of producing such a cell comprising disrupting the endogenous UDP-N-acetylglucosamine 2-epimerase/N- acetylmannosamine kinase (GNE) encoding gene and either before, or afterward, introducing a sequence encoding a polypeptide with one or more glycosylation sequences of the present invention.

[0176] The disruption of the UDP-N-acetylglucosamine 2-epimerase/N- acetylmannosamine kinase (GNE) encoding gene may be performed by any suitable means and in one embodiment is performed via CRISPR.

[0177] In one embodiment, the invention also provides a cell culture comprising a cell line of the present application where the cell culture medium is supplemented to allow it to produce sialic acid, in particular it is supplemented with peracetylated ManNAz that gets taken up, de-acetylated, and converted to sialic acid. In one preferred embodiment, the peracetylated ManNAz comprises a conjugation group. In one embodiment, the synthetic sialic acid precursor comprises a click chemistry group. In another embodiment, the synthetic precursor is mannosamine with an azide group. In one preferred embodiment, the synthetic compound is N-azidoacetylmannosamine (ManNAz).

[0178] Any suitable cell type that can O-glycosylate and sialylate a glycosylation sequence of the invention may be employed. Typically, the cell type employed is eukaryotic. In a preferred embodiment, the cell line is mammalian. An especially preferred cell line is CHO. Further examples of mammalian cell lines include HeLa cells, HEK293 cells, WI-38 cells, MRC-5 cells, and HepG2 cells. In one embodiment, the cell line is a HEK cell line. Examples of rodent cells which may be employed include 3T3, L929, and BHK-21 cells. In one embodiment, the cell line is a stem cell line. In one embodiment, the cell line is an Embryonic Stem (ES) cell line.

[0179] The present invention also provides a transgenic animal that expresses a polypeptide with one or more glycosylation sequences of the present invention. In one embodiment, the transgenic animal also comprises sialic acid biosynthesis engineering. In one embodiment, the animal may be fed artificial sugars to lead to sialylation of the sugar chains at the glycosylation sequence of the present invention. In one embodiment, the animal may be given food or water supplemented with peracetylated ManNAz that gets taken up, de-acetylated, and converted to sialic acid. In one embodiment, the animal is used to produce a molecule of the invention. In another embodiment, the animal is used as a model. In one embodiment, the desired moiety with a compatible conjugation group is given to the animal, for instance to target the moiety to a specific location or cell type. Such animals are typically non-human.

[0180] Combination with other conjugation methods

[0181] In one preferred embodiment, the use of a glycosylation sequence of the invention to bring about conjugation to a desired moiety may be combined with a different conjugation approach so that both are employed. In one particularly preferred embodiment, the glycosylation sequence of the invention may be used to conjugate a first moiety to a molecule, with a second conjugation approach used to conjugate a second and different moiety to the molecule. The use of such a combined approach may be, for instance, used to introduce two different functionalities to a molecule of the present invention via conjugation of the first and second moieties to it.

[0182] In a particularly preferred embodiment, the second conjugation approach employed may be the use of cysteine amino acids present in the molecule to provide a means to conjugate to the second moiety. In one embodiment, the cysteines have been engineered into the primary amino acid sequence of a polypeptide in the molecule, for instance in the same polypeptide which comprises a glycosylation sequence of the present invention (the approach can be referred to as eCys). The approach of introducing cysteines as a means for conjugation is described in WO 2018/232088 which is both incorporated by reference in its entirety and incorporated specifically in relation to conjugation via cysteine residues.

[0183] In one embodiment, where the combination of the glycosylation sequence of the invention and introduced cysteine residues are employed the molecule is an antibody. In one embodiment, an antibody of the invention comprises an IgG heavy chain constant region and light chain constant region wherein:

• at least one polypeptide of the antibody comprises at least one glycosylation sequence of the formula Xi Thr Pro X2 X3, wherein the Threonine (T) amino acid residue at the second position is O-glycosylated with a sialylated sugar which comprises a chemical group that either can be, or is, conjugated to a moiety Y; and

• said antibody comprises a cysteine at least one of the following residues: residue 124 in the CHI domain, residue 157 in the CHI domain, residue 162 in the CHI domain, residue 262 in the CH2 domain, , residue 378 in the CH3 domain, residue 397 in the CH3 domain, residue 415 in the CH3 domain, residue 156 in the Ckappa domain, residue 171 in the Ckappa domain, residue 191 in the Ckappa domain, residue 193 in the Ckappa domain, residue 202 in the Ckappa domain, or residue 208 in the Ckappa domain.

[0184] In one embodiment, the antibody comprises a cysteine at residue 124 in the CHI domain and further comprises a cysteine at one, but not all, of residue 157 and 162 in the CHI domain, residue 262 in the C2 domain and residues 378 and 415 in the CH3 domain. In one embodiment, the antibody comprises a cysteine at residue 157 in the CHI domain. In one embodiment, the antibody comprises a cysteine at residue 378 in the CH3 domain. In one embodiment, the antibody comprises a cysteine at residue 415 in the CH3 domain.

[0185] In one embodiment, the IgG heavy chain constant region is a human, mouse, rat, or rabbit IgG constant region. In one embodiment, the IgG heavy chain constant region of the antibody is a human IgGl or human lgG4 or human IgG2 isotype. In one embodiment, the IgG heavy chain constant region is a human IgGl constant region.

[0186] In a preferred embodiment, the IgGl heavy chain constant region of the antibody further comprises an isoleucine substituted at residue 247, a glutamine substituted at residue 339, and optionally a glutamic acid substituted at residue 332. In another preferred embodiment, the IgG heavy chain constant region is a human lgG4 constant region. In a further preferred embodiment, the lgG4 heavy chain constant region of the antibody further comprises a proline substituted at residue 228, an alanine substituted at residue 234, and an alanine substituted at residue 235 and a glutamine substituted at residue 339.

[0187] In one embodiment, the antibody is one comprising two heavy chains and two light chains, wherein each heavy chain comprises an IgG heavy chain constant region comprising a cysteine at one of the following residues: residue 124 in the CHI domain, residue 378 in the CH3 domain, and residue 397in the CH3 domain.

[0188] In one embodiment, the antibody comprises a cysteine at residue 124 in the CHI domain of each heavy chain and further comprises a cysteine at one, but not all, of residues 378 and 397 in the CH3 domain, and residue 157 in the CHI domain, of each heavy chain. In one embodiment, said antibody comprises a cysteine at residue 378 in the CH3 domain of each heavy chain. In a further embodiment, said antibody comprises a cysteine at residue 397 in the CH3 domain.

[0189] In one embodiment, the introduced cysteine replaces a native serine, valine, alanine, glutamine, asparagine, threonine, or glycine. In one embodiment, the total number of engineered cysteines is from two to six. [0190] Any of the embodiments outlined herein concerned with cysteine residues as a means for conjugation may be combined with any of the embodiments outlined herein employing a glycosylation sequence or sequences of the present invention.

[0191] Methods

[0192] In one embodiment, the present invention provides a method of producing a molecule of the present invention comprising: (a) culturing a cell line which expresses a polypeptide comprising one or more glycosylation sequences of the present invention under conditions that allow for the O-glycosylation of the glycosylation sequence or sequences of the present invention with an O-glycan comprising a sialic acid group and a group for conjugation; and (b) harvesting the O-glycosylated polypeptide. The O-glycosylated polypeptide may be harvested by any suitable means, for instance the polypeptide may also have a sequence for binding Protein A.

[0193] In one embodiment, step (a) is performed with the cell being cultured in a medium comprising an artificial sugar, for instance to promote incorporation of the modified sialic acid. In one preferred embodiment the artificial sugar is N- azidoacetylmannosamine (ManNAz). In one preferred embodiment, the sugar is peracetylated ManNAz. In one embodiment, the cell line used is one that cannot itself synthesize de novo sialic acid. In one preferred embodiment, the cell line is a cell line of the present invention which lacks UDP-N-acetylglucosamine 2- epimerase activity. In one embodiment, the cell lacks both the UDP-N- acetylglucosamine 2-epimerase and N-acetylmannosamine kinase activities. In one embodiment, it is a cell line where the endogenous GNE gene has been disrupted. In one particularly preferred embodiment, step (a) comprises both employing such a cell line and the medium comprising such an artificial sugar.

[0194] In one embodiment, the method further comprises step (c) where the purified polypeptide is then conjugated to a desired moiety. Hence, in one preferred embodiment, the present invention provides a method comprising: (a) culturing a cell line which expresses a polypeptide comprising one or more glycosylation sequences of the present invention under conditions that allow for the O-glycosylation of the glycosylation sequence with an O- glycan comprising a sialic acid group and a group for conjugation, wherein the cell line has at least the UDP-N-acetylglucosamine 2-epimerase activity of the GNE gene disrupted and is cultured in a medium comprising an artificial sugar promoting incorporation of modified sialic acid, preferably wherein the sugar comprises peracetylated ManNAz that gets taken up, de-acetylated, and converted to sialic acid;

(b) harvesting the O-glycosylated polypeptide with an O-glycan comprising sialic acid and a group for conjugation; and

(c) conjugating the harvested polypeptide with a desired moiety that has a compatible group for conjugation to the conjugation group of the polypeptide.

[0195] The present invention also provides a method comprising just step (c) without the earlier steps. Hence, the present invention provides a method comprising contacting:

(a) a molecule comprising one or more glycosylation amino acid sequences of the present invention which is, or are O-glycosylated with a sugar chain comprising sialic acid and a group for conjugation; and

(b) a desired moiety that comprises a conjugation group that is compatible with that of the molecule, allowing conjugation to take place.

[0196] Screening and multiplexing methods

[0197] One advantage of the present invention is that it allows combinatorial screening to identify combinations with a desired property. For example, the present invention provides a method comprising:

(a) providing a Pool A which comprises a plurality of molecules of the present invention each comprising a one or more glycosylation sequence of the present invention which is O-glycosylated with an O-glycosyl sugar chain comprising a sialic acid sugar and a chemical group for conjugation; and

(b) providing a Pool B which comprises a plurality of moieties that each comprise a chemical group for conjugation which is compatible with the conjugation group of the members of Pool A; and

(c) testing different pairwise combinations of a molecule from Pool A conjugated with a moiety from Pool B for a desired property.

[0198] In one embodiment, the molecules of Pool B are also glycosylated molecules of the present invention comprising a compatible chemical group for conjugation group to that of the molecules of Pool A.

[0199] In one embodiment, each pool has at least five, ten, twenty, one hundred, five hundred or more different molecules of the invention. In one embodiment, the members of a Pool are not physically present together but are classified as a pool because they all have the same conjugation group which is compatible with the conjugation group used in the other pool.

[0200] Such methods may be used to screen combinations of any desired molecules, provided that one of the conjugation partners comprises one or more glycosylation sequences of the present invention, which are O-glycosylated with terminal sialylation and a group allowing conjugation to a compatible conjugation group. Whilst assessing pairwise combinations of antibodies represents a particularly preferred embodiment, the invention is not limited to that. So, for instance, any combinations may be screened, such as screening possible combinations of an antibody with different drugs or different permutations of cytokines. In one embodiment, one or more molecules screened may be variants of a molecule and the screening is performed to see which variant form works best or has a particular desired property.

[0201] In one preferred embodiment, the method is used to assess different combinations of antigen-binding sites. In one embodiment, both the molecules of Pool A and Pool B are antibodies. In one embodiment, the antibodies in Pool A target one antigen and those in Pool B target a different antigen. In one embodiment, the molecules of Pool A and B are monospecific antibodies. In one embodiment, the antibodies in the pools are antibodies that can undergo arm exchange via the conjugation using the glycosylation sequences of the present invention. In one embodiment, the property screened for is the ability to bind both specificities, i.e. to act as a bispecific antibody. In one embodiment, rather than needing arm exchange, Pool A and Pool B both comprise antibodies with one or more glycosylation sequences of the present invention which is/are O- glycosylated with a sugar chain with sialic acid and a conjugation group, with the conjugation group for Pool A being compatible with that of Pool B, and with the antibodies of Pool A and B having different specificities, for instance against different antigens or different epitopes of the same antigen.

[0202] A screening method of the invention may include screening for any desired property. For example, in one embodiment, the method may assess the ability of different combinations to kill cancer cells. In another embodiment, the method may assess binding ability. In another embodiment, the method may screen for the ability of combinations to bind, and optionally activate, a receptor. In another embodiment, the method may assess the ability of different combinations to stimulate a cell to release a molecule.

[0203] The present invention also provides a library comprising of molecules of the present invention. The present invention further provides a combinatorial library comprising molecules of the present invention, where the library is split into at least two portions which have compatible conjugation groups to each other allowing for screening of pairwise combinations.

[0204] Pharmaceutical compositions and therapy

[0205] Also provided is a pharmaceutical composition comprising a molecule of the present invention and a pharmaceutically acceptable carrier or pharmaceutically acceptable excipient. The invention also provides a pharmaceutical composition comprising a nucleic acid or vector of the present invention and a pharmaceutically acceptable carrier or pharmaceutically acceptable excipient. Pharmaceutically acceptable carriers in therapeutic compositions may include, or be, liquids such as water, saline, glycerol, and ethanol. Additionally, auxiliary substances, such as wetting or emulsifying agents or pH buffering substances may be present in such compositions.

Pharmaceutically acceptable carriers include: sugars, such as lactose, glucose and sucrose; starches, such as corn starch and potato starch; cellulose and its derivatives, such as sodium carboxy methyl cellulose, ethyl cellulose and cellulose acetate; gelatin; talc; waxes; oils, such as peanut oil, cottonseed oil, safflower oil, sesame oil, olive oil, com oil and soybean oil; glycols, such as ethylene glycol and propylene glycol; polyols, such as glycerin, sorbitol, mannitol and polyethylene glycol, esters, such as ethyl oleate and ethyl laurate; agar; buffering agents; water; isotonic saline; pH buffered solutions; and other non-toxic compatible substances employed in pharmaceutical formulations. The pharmaceutically acceptable carrier may also include a manufacturing aid (e.g., lubricant, talc magnesium, calcium or zinc stearate, or stearic acid), a solvent, or encapsulating material. If desired, certain sweetening and/or flavoring and/or coloring agents may be added. Such carriers enable the pharmaceutical compositions to be formulated in formats such as tablets, pills, dragees, capsules, liquids, gels, syrups, slurries, and suspensions, for ingestion by the patient. A pharmaceutical composition may be formulated to take into account the nature of the active agent. Compositions comprising pharmaceutically acceptable carriers are formulated by well-known conventional methods (see, for example, Remington's Pharmaceutical Sciences, 18^th edition, A. Gennaro, ed., Mack Publishing Co., Easton, PA, 1990; and Remington, The Science and Practice of Pharmacy, 20th Ed., Mack Publishing, 2000 which is incorporated by reference in its entirety).

[0206] A pharmaceutical composition of the invention may be administered to a subject by any suitable route. Administration may be, for example, via oral, intravenous, intramuscular, intra-arterial, intramedullary, intrathecal, intraventricular, transdermal, transcutaneous, subcutaneous, intraperitoneal, intranasal, enteral, topical, sublingual, intravaginal or rectal routes. Administration may be, for example, via parental administration. Administration may be, for example, by injection or infusion. Examples of such administration include bolus injection or continuous injection. A pharmaceutical composition may be provided, for instance, in a form that is suitable for such administration routes. For example, a preferred route of administration is via injection, hence the pharmaceutical composition may be provided as in a format suitable for injection, for instance in a format suitable for intravenous injection.

[0207] A pharmaceutical composition will provide a therapeutically effective dose of the active agent. For example, a pharmaceutical composition may provide a dosage of about 0.01 mg/kg to about 50 mg/kg of the active agent, for example 0.05 mg/kg to 50 mg/kg, for instance about 0.10 mg/kg to about 5 mg/kg of the active agent, or about 0.10 mg/kg to about 0.50 mg/kg. The dosage may be, for example, adjusted depending on the molecular weight of the molecule, for example a lower mg/kg dosage may in some embodiments be given where the molecular weight of the molecule is small, such as in the case of a peptide. The dosage may be chosen by a physician as a suitable dose for a given condition. The present invention also provides a unit dosage form of the pharmaceutical composition. Further provided are various formats that facilitate administration to the subject. For example, an autoinjector or pen delivery device loaded with a pharmaceutical composition of the invention is also provided. Also provided is an intravenous drip bag loaded with a pharmaceutical composition of the present invention. Also provided is a pharmaceutical composition in lyophilized form that can be reconstituted and then administered to the subject.

[0208] A pharmaceutical composition of the present invention may be administered to any suitable subject. As used herein, the term "subject" refers to a mammal, including, but are not limited to, a human, chimpanzee, ape, monkey, cattle, horse, sheep, goat, swine, rabbit, dog, cat, rat, mouse, guinea pig, and the like. In a preferred embodiment, the subject is a mammalian subject. In an especially preferred embodiment, the subject is human.

[0209] A pharmaceutical composition of the present invention may be used to treat a subject. The present invention provides a method of treating a condition in a subject comprising administering a therapeutically effective amount of the composition to a subject in need thereof. Also provided is a pharmaceutical composition of the present invention for use in a method of treatment of the human or animal body. The present invention also provides a pharmaceutical composition of the present invention for use in a method of treating a condition. A pharmaceutical composition may further comprise another therapeutic agent in addition to a conjugate of the present invention to allow both to be given to a subject at the same time. The two though may be separately administered to a subject. For instance, where administered separately, the two may be administered simultaneously, sequentially, or separately to a subject.

[0210] The present invention also provides any of the active molecules of the present invention for use in the manufacture of a medicament, for instance to treat any of the conditions mentioned here.

[0211] The versatility of the present invention means that it may be used to treat any suitable condition. In one embodiment, it is used to treat a condition selected from cancer, heart disease, infection, an autoimmune disorder, a respiratory disorder, diabetes, dementia, pain, neurodegeneration and liver disease. In one embodiment, the cancer is selected from bladder cancer, breast cancer, colon or rectal cancer, endometrial cancer, kidney cancer, leukemia, liver cancer, lung cancer, skin cancer (e.g. melanoma), non-Hodgkin lymphoma, pancreatic cancer, prostate cancer, and thyroid cancer. In one embodiment, the invention may be used to induce tolerance to treat an autoimmune disorder. For instance, tolerance may be elicited via SIGLEC engagement with sialic acid on an engineered O- glycan without any conjugation. Alternatively, click chemistry may be utilized to conjugate a glycan known to induce immunotolerance. The invention may be used to treat autoimmune diseases and to treat allergies, for example allergies to serious allergens like peanut proteins. Therapeutic proteins with immunogenic regions may be engineered with a nearby glycosylation sequence of the invention to promote immunotolerance.

[0212] In one embodiment, the invention is used to help promote targeting. For instance, the moiety joined to the polypeptide with a glycosylation sequence of the present invention may target the molecule of the invention to a particular location. In one embodiment, the moiety that is conjugated is specific to a particular organ or cell type. In one embodiment, the targeting is to the liver. In one embodiment, the moiety is, or comprises a LYTAC (lysosomal targeting chimera). LYTACs typically bind to the ASGPR receptor and hence provide targeted delivery to the liver. By analogy molecules able to target the heart, lungs, brain, kidneys, or other organ may be employed. In one embodiment, the targeting is to a particular cell type. In a further embodiment, the conjugation of the moiety to the molecule leads to it being targeted to a particular compartment in the cell.

[0213] In a preferred embodiment, targeting is to the CNS and in particular the brain. In a preferred embodiment, a binding molecule of the invention binds a target that means it will be transported across the blood brain barrier (BBB). In one embodiment, it binds to the transferrin receptor.

[0214] The present invention may also be used in diagnosis. For example, the conjugated moiety may be a label allowing for the location of the molecule of the invention to be identified. For instance, the moiety may be a radiolabel which allows for the molecule to be located. Such labelled molecules may, for instance, be used in ADME (absorption, distribution, metabolism, and excretion) studies. Such methods may be in vivo, for example a molecule of the invention may be administered to a subject and then its location identified via the conjugated label. The invention may be used in imaging techniques such as MRI and PET imaging.

[0215] Further uses

[0216] In one embodiment, the desired moiety that is conjugated to the molecule is one that results in the targeting of the molecule to a particular cell or cell receptor. In one embodiment, the invention is used to modulate protection from protease degradation, modulating serum half-life, functional modulation, intracellular trafficking, cell adhesion, and self- versus foreign recognition during an immune response.

[0217] The invention may be used for a wide variety of purposes. In one embodiment, the invention is used to increase the stability of a molecule. The presence of glycosylation at a glycosylation amino acid sequence of the present invention may prevent, reduce, or slow down the degradation of the molecule. In one embodiment, the glycosylation sequence is introduced at a site, or sites, such that its glycosylation reduces the ability of a second entity to bind to the molecule. That may be the case with the glycosylation before any conjugation or the conjugated moiety may result in protection of the molecule. In one embodiment, the one or more glycosylation sequences are introduced near the cleavage site for a protease.

[0218] In one embodiment, the invention results in a reduction of the ability of an enzyme to bind to the molecule. In another, it results in a reduction of the ability of a molecule to bind to a receptor. In another embodiment, the invention may alter the serum half-life of the molecule. Hence, the molecule may have, for instance, a longer serum half-life once it is conjugated to the desired moiety. In one embodiment, the desired moiety is PEG and conjugation results in an increase in serum half-life. In another embodiment the moiety is serum albumin.

[0219] In one embodiment, the conjugated moiety facilitates purification of a molecule of the present invention. For example, the desired moiety may be conjugated to biotin to allow for rapid purification. The invention may also be used to immobilize a desired molecule onto a surface via the conjugation. Hence, the invention may be used to produce a protein immobilized for techniques such as ELISA. The invention may also be used to immobilize a given protein onto beads via the conjugation.

[0220] The invention may also be used to replace one or more natural glycosylation sites in a polypeptide with one or more glycosylation sequences of the present invention.

[0221] In another embodiment, the invention may be used to conjugate moi eties with a desired activity to a given protein, so, for instance, enzymes like hyaluronidase (e.g. for tumor penetration), reporters (e.g., luciferase, dyes), affinity tags (e.g. biotin, FLAG, HA), small molecules (e.g., solubilized steroids), albumin binding lipids, or membrane binding lipids.

[0222] The present invention may also be used to alter the solubility of a molecule. In one embodiment, that may be done by conjugating a desired moiety with a particular solubility. So, for instance, water soluble groups may be conjugated to organic groups to promote their solubility. For example, in one embodiment cyclodextrin or HPMA, loaded with a steroid, may be conjugated to an antibody for targeted delivery.

[0223] Kits

[0224] The present invention also provides a kit comprising a molecule of the present invention. In one embodiment, the kit comprises a library of the present invention. In another embodiment, the kit may further comprise a cell line of the invention. In one embodiment, the kit may comprise instructions for use.

[0225] All patents, published patent applications and publications cited herein are incorporated by reference as if set forth fully herein.

[0226] EXAMPLES

[0227] Example 1: O-glycosylation sequence optimization

[0228] Introduction

[0229] Fully homogenous O-linked glycosylation of proteins is unprecedented and highly desirable. As such, a suitable test system was employed to assess the ability of test sequences to act as motifs for O-glycosylation in cells. It was known that immunoglobulins such as human IgAl or IgG3 are O-glycosylated at the hinge to an extent and hence the hinge region was used as a model system to investigate possible O-glycosylation sequences.

[0230] To mimic the location of O-glycosylation in wild type human IgG3 and IgAl, putative O-glycosylation motifs were inserted into the heavy chain of human IgGl given its preponderance in antibody-based drugs.

[0231] The following motifs were inserted into the hinge:

• PTPSP (O-glycosylation motifs in IgAl)

• DTPPP (O-glycosylation motifs in IgG3)

• ATPAP (variant of above sequence)

• PTPAP (variant of above sequence) [0232] Proteins were expressed in HEK 293 cells, isolated with Protein A and characterized by mass spectrometry to study the level of O-glycosylation.

[0233]

[0234] Materials & Methods

[0235] Transient protein expression in HEK 293 cells

[0236] Human Embryonic Kidney cells (aka HEK 293) were maintained in log phase of growth in DMEM/F-12 media as a serum free suspension culture in shaking flasks. Cultures were grown at 37 °C, with 7% CO2, humidified, 160RPM, 50mm orbital. On the day of transfection, HEK 293 cells were diluted to 2xl0⁶ cells per ml in DMEM-F-12. A DNA carrier complex was then prepared by mixing lOpg of the DNA expression vector into 1ml of growth media. 20pL polyethylenimine (25k MW free base, Img/mL in water) was added to ImL of the DNA solution and gently mixed by inversion. The mixture was allowed to incubate at room temperature for approximately 20 minutes to allow the complex to form. 10% volume/volume of DNA carrier complex was added to the prepared HEK culture. The transfected culture was incubated for 5 days at 37 °C, with 7% CO2, humidified, 160RPM, 50mm orbital.

[0237] Protein Isolation

[0238] Antibodies were isolated with Protein A affinity resin as set out below.

[0239] Running Buffers:

■ Buffer A - IX PBS pH 7.2.

■ Buffer B - 20 mM Acetic Acid + 5 mM Citric Acid pH 2.7.

■ Neutralization Buffer - IM Tris pH 8.0.

■ Resin: mAb Select SuRe (GE Healthcare Cat: 17543801).

[0240] Resin preparation:

The following protocol was employed:

■ 50 mL mAb Select SuRe resin slurry was added to a Falcon tube;

■ the tube was centrifuged at 3500rpm for 10 mins;

■ the supernatant was poured off; ■ the resin was washed with 50% v:v of IX PBS pH 7.2, repeating centrifugation for a total of 6X; and

■ the resin was equilibrated with Buffer A- IX PBS pH 7.2, creating a 50% slurry.

[0241] Batch Procedure:

[0242] To harvest from a 100 mL culture (1 g/L titer), 4 mL of resin was added for binding (assuming 25 mg/mL) and the culture mixed for ~1 hour with shaking. The supernatant/resin was mixed to 10 mL poly-prep column (Bio Rad Cat # 7311550) and the flow-through collected. The resin was then washed with 10X resin bed volumes of Buffer A (IX PBS + pH 7.4) and the wash again collected. 10% of the final elution volume of Neutralization Buffer - IM Tris pH 8.0 - was pipetted into a fresh collection tube. The column was transferred to a collection tube and protein eluted with 5X resin bed volumes of Buffer B - 20 mM Acetic Acid + 5 mM Citric Acid.

[0243] Protein Mass Spectrometry

[0244] Product was characterized by mass spectrometry after reduction with di thioerythritol. An Agilent G7100A CE System was used for analysis. The CE capillary was a neutral coated, 60 cm long, 360 pm OD x 50 um ID capillary (Agilent, PVA coated,) with a custom tapered end. The running buffer was 1% formic acid. The sheath liquid was 12% methanol, 0.1% formic acid. The spray tip was pulled in house from uncoated glass rods to an ~ 10 pm tip opening. The sample was injected by pressure (50 mbar for 20 sec). The separation was run at 30,000 V for 15 minutes.

[0245] An Agilent 6550 Q-TOF mass spectrometer was used for analysis with a custom nanosheath flow CEMS interface with 3000V for the nanosheath spray. The mass spectrum was collected in the Q-TOF at 1 scan/sec from 600-3200 m/z. All spectra were collected in profile and centroid mode. Data was analyzed in the Agilent Qualitative analysis software vB08.

[0246] Results [0247] Figures 1 and 2 show potential O-glycosylation sugar chain structures depending on whether the sugar chain has been sialylated and a click sugar incorporated. Figure 3 provides an illustration of the protein mass spectroscopy results obtained showing the different peaks for products. The overall results obtained are summarized in Table 1 below.

[0248] Table 1: Level of glycosylation and sialylation with test sequences

[0249] Example 2: GNE Knockout

[0250] CHO cells were generated where the UDP-N-acetylglucosamine 2- epimerase/N-acetylmannosamine kinase (GNE) gene was modified to disrupt both the epimerase and kinase activities. CRISPR technology was used to introduce a genetic modification into CHO cells to cause a translational frame shift in all GNE gene alleles of the CHO cell line. Lentivirus was used to deliver CRISPR genes to CHO cells, generating a stable CHO pool after puromycin selection (lOug/ml). The constitutive expression of the CRISPR genes ensured that GNE activity was abrogated. The constructs utilized in the preparation of the cell line were hamster Gne-l-pLentiCRISPRv2(T15257), Gne-2-pLentiCRISPRv2(T15258), Gne-3- pLentiCRISPRv2(T 15259), Gne-4-pLentiCRISPRv2(T 15260) and Gne-5- pLentiCRISPRv2(T 15261), whose sequences are provided as SEQ ID Nos: 4 to 8 respectively, with the actual CRISPR sequences provided as SEQ ID Nos: 9 to 13. [0251] Anti-poly sialic acid antibody binding, followed by secondary fluorescent antibody in flow cytometry was used to determine the level of sialylation in the modified CHO cells compared to parental cells (positive control) and parental cells treated with neuraminidase (negative control). Reduced fluorescence on the cell surface confirmed the lack of sialic acid generation in all five CRISPR modified cell lines with the results obtained set out in Table 2 below.

[0252] Table 2: Demonstration of loss of sialylation activity in GNE modified cell lines

[0253] Example 3: Transient Expression: mAb with O-Link on Light chain

C-terminus

[0254] Introduction [0255] A therapeutic antibody (Molecule #1) was engineered to have the optimal invented O-linked glycan recognition site (AAAPTPAPAAA) at the C-terminus of the light chain. A DNA vector, expressing this glycan-engineered LC and the unmodified HC, was transiently transfected into CHO cells defective in de novo sialic acid synthesis, as set out below. The level of glycosylation and sialylation was then analyzed.

[0256] Materials & methods

[0257] Cell culture and transient expression of secreted proteins

[0258] CHO cells were grown at 37°C as a suspension in a perfusion bioreactor in a completely defined medium that allowed cell densities of up to 5xl0⁷ cells/ml. Cells were centrifuged and then resuspended at 2xl0⁷ cell/ml in fresh media in disposable shaking flasks. Expression vector DNA was added to the cells at 14.5 mg/L, immediately followed by polyethylenamine (PEI, 25k MW, linear) at 27 mg/L. The transfected culture was then temperature shifted for increased protein production in a 32°C incubator, 6% CO2, 160 RPM with a 50mm orbital. Nutrient feeding allowed a seven-day production, with average antibody titers of around lg/L.

[0259] Stable cell generation and expression of secreted proteins

[0260] Recombinant stable CHO pools were generated by transfection of CHO cells with one or more DNA vectors containing transposon elements. A DNA vector expressing an appropriate transposase enzyme was co-transfected, eliciting recombination of the transposable DNA vector(s) with chromosomal DNA. CHO stable pools with one or more recombinant genes were grown in suspension at 37°C until they reached a cell density of 5X10⁶. The culture was centrifuged at 500Xg and resuspended at 2X10⁷ cell/mL. The culture was immediately temperature shifted for increased protein production in a 32°C incubator, 6% CO2, 160 RPM with a 50mm orbital. Nutrient feeding allowed fourteen-day production, with average antibody titers of around 5 g/L. [0261 ] Feeding of click chemistry sugars

[0262] Artificial sugars such as peracetylated ManNAz were formulated in DMSO, typically around 500mM for stock solution, which was stored at -20°C. Acetylated sugar was used to allow robust passive diffusion into cells. On the day of transfection, the click sugar solution was allowed to warm to room temperature and was added to the culture at a final concentration of 800pM. On each subsequent day of the culture click sugar was added to a concentration of 400pM. It was possible to skip feeding of the culture for one day by adding the click sugar to 800pM on the day prior to skipping a feed. Click chemistry sugar concentration could be adjusted commensurate with cell density to avoid toxicity.

[0263] After a 5-day production with peracetylated ManNAz feeds, the antibody was purified using protein A affinity chromatography. A280 spectrophotometry indicated that the volumetric expression level was 0.8g/L.

[0264] Post-expression processing (purification, click chemistry)

[0265] Owing to the bioorthogonal nature of click chemistry, a simple one step purification of the engineered glycoprotein was sufficient to prevent conjugation with host cell proteins in the raw supernatant. Click chemistry was typically performed in physiological conditions by adding a complementary click conjugate. Copper catalyst was included for those reactions requiring copper ions (e.g. alkyne with azide).

[0266] Results

[0267] Figures 1 and 2 show the various O glycosyl chains resulting from O glycosylation, sialylation of the O-glycosyl sugar chain, and the incorporation of a click chemistry group. The results of the mass spectrometry showed the proportion of O-glycosylation sites O-glycosylated, (“occupancy”) as well as what proportion have been sialylated with azidosialic acid incorporated in the O glycosyl chain. The mass spectrometry results obtained showed that glycan occupancy was >99% with >95% azido-sialic acid. Without GNE knockout, cells incorporated natural sialic acid, reducing the content of azidosialic acid to <90%. Figure 3 shows illustrative protein mass spectroscopy results for particular glycosylation sequences (without GNE knockout - 3(a), and with GNE knockout - 3(b)), with the peak of the antibody molecule with the O-glycosyl sugar chain with sialylation and a click chemistry group depicting the highest peak seen in each instance. The results obtained illustrated the usefulness of the GNE knockout generated and also the advantages of the glycosylation sequence provided to achieve high site occupancy.

[0268] Example 4: Stable cell generation and protein expression of an IgGl mAb with O-Link on Light chain C-terminus

[0269] Introduction

[0270] As discussed above, a therapeutic IgGl monoclonal antibody (Molecule #1) was engineered to have the preferred O-linked glycan recognition sites (AAAPTPAPAAA) at the C-terminus of the light chain. A stable cell line was generated expressing Molecule #1 and the ability of the cell line to produce Molecule #1 was studied further.

[0271] Materials & Methods

[0272] A DNA vector, expressing this glycan engineered light chain (LC) and the unmodified heavy chain (HC), with transposon elements flanking the antibody operon, was transfected into CHO cells defective in de novo sialic acid synthesis (see Stable cell generation and expression of secreted proteins),' a second vector expressing an appropriate transposase was co-transfected with the transposable DNA vector (without transposable elements to allow only transient expression of the transposase).

[0273] After generating a stable pool, a 7-day production was performed with ManNAz feeds. Fed-batch, shake flask cultures of 100 mL with a starting viable cell density (VCD) of either >10 x 10⁶/mL or ~1 x 10⁶/mL were grown in a chemically defined, proprietary media. A concentrated, chemically defined, proprietary feed was periodically added to each flask during the process. Cultures with an initial VCD of >10 x 10⁶/mL were incubated for 10 days at 32°C, 6% CO2, shaking at 150rpm with a %” throw. Incubation conditions for the cultures with an initial VCD of ~1 x 10⁶/mL were the same, except the incubation temperature started at 36°C, then shifted to 32°C on day 6. The process duration for these cultures was 14 days.

[0274] For all shake flasks, peracetylated N-azidoacetylmannosamine (Ac4ManNAz) was added from a 500mM stock dissolved in dimethyl sulfoxide (DMSO) starting on day 0. Ac4ManNAz was added daily at a rate of 20pM per integrated cell area (ICA) unit until a pre-determined limit was reached. The Ac4ManNAz was then added daily at that limit until the final day of the process. For the cultures that had an initial VCD of >10x 10⁶/mL, Ac4ManNAz addition limits were 200pM, 400pM, 600pM, and 800pM. For cultures that started at a VCD of ~1 x 10⁶/mL, the only limit tested was 400pM. Control cultures were run where only DMSO was added at an equal volume to that of the highest volume of DMSO added to the Ac4ManNAz test cultures. An additional control was run as a normal, fed-batch process. All conditions were run in duplicate.

[0275] Cultures were monitored daily for cell growth utilizing a Vicell XR Viability Analyzer (Beckman Coulter, Indianapolis, IN). At the conclusion of the cell culture processes, cell free media (CFM) samples from each flask were assayed by CE-MS analysis to determine azidosialic acid incorporation. Additional CFM samples from earlier days in the process were assayed for productivity with a Cedex BioHT Analyzer (Roche Diagnostics, Indianapolis, IN).

[0276] Samples from all sets of shake flasks were then analyzed by CE-MS as described in earlier Examples.

[0277] Results

[0278] The results obtained are shown in Figures 5 to 9 as follows:

• Figure 5(a): Growth for cultures with an initial viable cell density of >10 x 10⁶/mL.

• Figure 5(b): Viabilities for cultures with an initial viable cell density of >10 x 10⁶/mL. • Figure 6: Productivities of molecule #1 for cultures with an initial viable cell density of >10 x 10⁶/mL.

• Figure 7(a): Growth for cultures with an initial viable cell density of ~1 x 10⁶/mL.

• Figure 7(b): Viabilities for cultures with an initial viable cell density of ~1 x 10⁶/mL.

• Figure 8: Productivities of molecule #1 for cultures with an initial viable cell density of ~1 x 10⁶/mL.

• Figure 9 shows the results of the sugar titration.

[0279] For the cultures with an initial VCD (viable cell density) of >10 x 10⁶/mL, the 800pM and 600pm Ac4ManNAz limits became toxic at days 8/9 and 9/10, respectively. For lower Ac4ManNAz limits, growth was similar to control levels. There was some detrimental impact by Ac4ManNAz on productivity, except for the 200pM limit, which was at or near control levels (Fig. 2). All data were averages of duplicate shake flasks.

[0280] For the cultures that had an initial VCD of ~1 x 10⁶/mL, the 400pM Ac4ManNAz limit became toxic for one of the two cultures on day 13/14 and was also had some detrimental impact on productivity.

[0281] When samples from all sets of shake flasks were analyzed by CE-MS, all samples had near 100% incorporation of azidosialic acid, regardless of productivity and end-of-process viability with the results obtained shown in Figures 9 and 10.

[0282] Example 5: Stable cell generation and protein expression of an IgG4 mAb with O-Link on Light chain C-terminus

[0283] Introduction

[0284] Next a similar study to that performed in Example 4 was carried out, but with an IgG4 monoclonal antibody, Molecule #2.

[0285] Materials & Methods [0286] A cell line for molecule #2 was generated in the same way as described in Example 4 for Molecule #1. Molecule #2 was then produced in a bulk culture. Fed-batch, shake flask cultures of lOOmL with a starting viable cell density (VCD) of ~10 x 10⁶/mL were grown under identical conditions as >10 x 10⁶/mL cultures for molecule #1.

[0287] As before, Ac4ManNAz was added from a 500mM stock dissolved in DMSO starting on day 0 at a daily rate of 20uM per ICA unit to a pre-determined limit, then added daily at that limit until the final day of the process. Based on the toxicity, productivity, and incorporation results from molecule #1, the Ac4ManNAz addition limits were changed to 50pM, lOOpM, 200pM, and 400pM. Control cultures were run where only DMSO was added at an equal volume to that of the highest volume of DMSO added to the Ac4ManNAz test cultures. An additional control was run as a normal, fed-batch process. All conditions were run in duplicate and monitored daily for growth. Due to extenuating circumstances, all processes were terminated on day 9. Productivities were determined on the BioHT for days 7 and 9. Azidosialic acid incorporation was determined by CE-MS from day 9 samples.

[0288] Results

[0289] The results obtained are shown in:

• Figure 11 : Growth for cultures with an initial viable cell density of ~10 x 10⁶/mL for molecule #2.

• Figure 12: Viabilities for cultures with an initial viable cell density of ~10 x 10⁶/mL for molecule #2.

• Figure 13: Productivities for cultures with an initial viable cell density of ~10 x 10⁶/mL for molecule #2.

• Figure 14: Product distribution for cultures with an initial viable cell density of ~10 x 10⁶/mL for molecule #2.

• Figure 15: CE-MS summaries for day 9 samples showing incorporation of azidosialic acid on to molecule #2 [0290] Since the Ac4ManNAz limits were reduced to lower levels for molecule #2, there was no toxicity observed on growth (Figures 11 and 12). As with molecule #1, there was a detrimental impact on productivity at the 400 pM limit, but no impact at the lower limits (Figure 13). When samples were analyzed by CE-MS, there was a dose response of azidosialic acid incorporation in relation to the addition limits (Figures 14 and 15). The 400pM level was again at or near 100%, and 200pM was approximately 98% incorporation.

[0291 ] Example 6: Transient mAb expression with mAb with O-Link on

Light chain C-terminus and Heavy Chain C-terminu

[0292] Introduction

[0293] A therapeutic antibody, Molecule #3, was engineered to have the optimal O-linked glycan recognition site (AAAPTPAPAAA) at the C-terminus of the light chain (LC) and the C-terminus of the heavy chain (HC). Transient expression of the antibody was studied.

[0294] Materials & Methods

[0295] A DNA vector, expressing the glycan engineered light chain (LC) and the glycan engineered heavy chain (HC), Molecule #3, was transiently transfected into the GNE knockout CHO cells defective in de novo sialic acid synthesis (as described in earlier Examples). After a 5-day production with Ac4ManNAz feeds using the same approach as in earlier Examples, the antibody was purified using protein A affinity chromatography. The expression level as determined by A280 spectrophotometry, glycan occupancy, and azidosialic acid incorporation as determined by mass spectrometry were all recorded.

[0296] Results

[0297] The expression level as determined by A280 spectrophotometry, glycan occupancy, and azidosialic acid incorporation as determined by mass spectrometry are given in Table 3 below. [0298] Table 3: Comparison of O-glycosylation occupancy and level of sialylation

[0299] Example 7: Conjugation of O-glycosylated antibody to siRNA

[0300] Introduction

[0301] Having demonstrated the successful generation of sialylated O- glycosylated sugars with a high degree of occupancy and incorporating click sugars, the antibodies were then used to demonstrate the ease with which the click chemistry groups could be used to conjugate a desired moiety to the antibody. The moiety chosen for conjugation to the antibodies was a siRNA and molecules listed in Table 5 were generated and assessed.

[0302] Materials & Methods

[0303] Conjugation

[0304] Conjugation was performed in two different formats. In the first format, siRNA was functionalized with clickable DBCO group. DBCO allowed direct conjugation of the siRNA duplex to antibody containing the azidosialic acid sugar either at the heavy chain or light chain. In the second format, a DBCO- methyltetrazine bifunctional linker was used, where the linker was first conjugated to the antibody azidosialic acid site. Excess linker was removed by desalting the antibody into IX PBS pH7.2, followed by addition of siRNA duplex that was functionalized with TCO chemistry, to click onto the methyltetrazine of the linker. In both conjugation formats, 4 to 1 molar equivalents of siRNA to antibody were used.

[0305] Figure 16 illustrates the basic approach of how an antibody with the glycosylation sequence provided can be conjugated to siRNA through the sialylated O-glycosyl sugar chains present using click chemistry sugars either with or without a linker.

[0306] Conjugation was monitored using analytical anion exchange chromatography. A ProPac™ SAX- 10 HPLC Column, lOum particle, 4mm diameter, 250mm length was utilized with the following method. Flowrate of 1 mL/min, Buffer A: 20mM TRIS pH 7.0, Buffer B: 20 mM TRIS pH 7.0 + 1.5M NaCl, at 30°C.

[0307] Table 4: HPLC gradient employed

[0308] The drug antibody ratio (DAR) was calculated based on peak area % from the analytical anion exchange (aAEX) chromatogram. An illustrative example of a chromatogram is shown in Figure 17.

[0309] Conjugation kinetics

[0310] Conjugation of siRNA to the azidosialic acid sugar incorporated antibody was monitored by aAEX over time as a function of antibody concentration at 1 mg/mL, 5 mg/mL and 10 mg/mL to assess impact of antibody concentration on conjugation. The results obtained are shown in Figure 18 (18(a) - 10 mg/mL, 18(b) - 5 mg/mL, and 18(c) - 1 mg/mL mAb).

[0311] Purification of antibody-siRNA conjugate:

[0312] Post conjugation of siRNA to the antibody, excess siRNA and unconjugated antibody was removed by further purification. Either preparative size exclusion chromatography (SEC) or preparative anion exchange chromatography was utilized for purification of the final conjugate. Preparative SEC was performed using Cytiva superdex200 in 1XPBS. Alternatively, Cytiva Q-FF was used with starting buffer of 20mM TRIS pH 7.0 and eluting with 20 column volume gradient with a buffer containing 20mM TRIS pH 7.0 and IM NaCl. These resulted in purified antibody-siRNA conjugate devoid of excess siRNA and unconjugated antibody. The resulting conjugate profile was analyzed by analytical anion exchange with Figure 19 illustrating the results obtained and Table 5. After AEX purification, a DAR of 1.86 and 1.72 was obtained for ARC- 180 and ARC-181, respectively.

[0313] Table 5: ARC-180 and ARC-181 conjugate preparations [0314] Ex vivo Plasma Stability

[0315] Antibody-siRNA conjugates were subjected to ex vivo plasma stability assessment to observe stability of the conjugate and any dissociation of the siRNA from the antibody. Antibody-siRNA conjugates were incubated in mouse and/or cynomolgus monkey plasma at 37°C at 0, 24 and 48 hours respectively with rotation at 5rpm. Antibody was immunoprecipitated from the plasma sample using biotinylated goat anti-human IgG. Solution was then incubated with streptavidin beads at room temperature with rotation for 30 minutes. Following multiple washing step with IX PBS, the sample was eluted with 1% formic acid with 20% acetonitrile elution buffer by mixing at 2000rpm for 15 seconds followed by 5- minute static benchtop hold. The eluted sample was then injected into LC-MS for analysis.

[0316] LCMS method:

• Instrument: Sciex I.

• Column: Agilent, PLRP-S 1000A 5pM 50x1.0 MM, PN: PL1312-1502.

• Injection Volume: 20 pL.

• Column Temperature: 80°C.

• Auto sampler Temperature: 5 °C.

• Mobile Phase A: water with 0.05% TFA.

• Mobile Phase B: acetonitrile with 0.05% TFA.

• Gradient:

• MS: m/z 2000-5000

[0317] The results obtained are shown in Figure 20. [0318] In Vitro Target Gene Knockdown Potency Assessment in Murine Cortical Neurons

[0319] Murine primary cortical neurons were isolated from wild type C57BL6 mouse embryos at El 8. Cells were plated in poly-D-lysine coated 96-well plates at a density of 40k cells/well and cultured in NbActivl (BrainBits, LLC) containing 1% Antibiotic/ Antimycotic (Corning) for 7 days at 37°C in a tissue culture incubator in a humidified chamber with 5% CO2. On day 7, half the medium was removed from each well and 2x concentration of one of the following (i) to (iii) in culture media with 2% FBS was added for treatment as CRC and incubated with cells for additional 7 days, with (i) to (iii) being: (i) siRNA (targeting a desired therapeutic target within the CNS); (ii) a control siRNA (non-targeting therapeutic target within the CNS); and (iii) a human IgG4 isotype antibody having a siRNA linked at the C-terminus of the light chain via the azido-sialic acid of either a control untargeted antibody or a CNS-targeted transferrin receptor (TfR) antibody, in culture media with 2% FBS is added for treatment as CRC and incubated with cells for additional 7 days. At the end of siRNA treatment, RT-qPCR was performed to quantify targeted mRNA levels using TaqMan Fast Advanced Cell-to-CT kit. Specifically, cells are lysed, cDNA was generated on Mastercycler X50a (Eppendorf), and qPCR is carried out on QuantStudio 7 Flex Real-Time PCR System (Applied Biosystems). Gene expression levels of the therapeutic target were normalized by P-actin (ThermoFisher Mm02619580_gl) using respective probes.

[0320] Results are provided Figure 26 and Table 6. Results provided in Table 6 demonstrate the exemplified mouse TfR binding protein-siRNA created via either eCys or glyco-mAb conjugation chemistry (e.g., mTfR2-dsRNA No. 8 conjugate) successfully targets mouse gene and provides an order of magnitude greater knockdown than Isotype Ab- siRNA.

[0321] Table 6: Target gene knockdown potency of exemplified mTfR binding protein-siRNA conjugates.

[0322] Multiplexing conjugation using glyco-chemistry and alternate conjugation chemistry

[0323] In order to generate a multi-modal antibody conjugate, glyco-chemistry derived conjugation was combined with orthogonal chemistry such as lysine- or cysteine-based conjugation. Antibodies containing the azidosialic acid sugar were either engineered to have cysteine sites for site-selective conjugation or structural cysteine or lysine was utilized for secondary conjugation. For engineered cysteine conjugation, the antibodies were first reduced with 20 molar equivalent reducing agent, followed by reoxidation with 10 molar equivalent dehydroascorbic acid (DHAA). That was followed by addition of DBCO-functionalized siRNAl and thiol -reactive siRNA2.

[0324] Conjugation of both siRNA was monitored by analytical AEX and resulting conjugate was further purified using the method described above. An example multiplex siRNA conjugate aAEX profile is provided in Figure 21.

[0325] In Vitro Target Gene Knockdown Potency Assessment of Multiple Targets in Human Cells

[0326] Antibody-siRNA conjugates containing siRNAs to two targets (targetl and target2) were assessed for functional target knockdown activity in cultured cells. An ovary cystadenocarcinoma derived EFO-21 cell line was used as exemplary cell line to generate the knockdown data. EFO-21 are seeded at 10,000 per well into 96-well plates and treated with siRNAs to generate concentration response curve(CRC). After 72 hours incubation, cell lysates were prepared for Cell to CT qPCR to examine target gene expression and assess efficacy of siRNAs for knockdown of target gene and compare CRC potency between lipid conjugated siRNA and antibody conjugated siRNA. Cell to CT kit was used to generate the mRNA expression data. Example of the results for target knockdown using antibody-siRNA conjugates are provided in Figures 23(a and b), 24 (a and b), and 25 (a and b).

[0327] Example 8: Use of O-glycosylated antibody conjugated to siRNA for in vivo gene knock-down

[0328] To further illustrate the efficacy of conjugates of the invention, the ability of an antibody-siRNA conjugate of the invention to inhibit expression of a target gene in vivo was studied. An antibody specific for mouse Transferrin Receptor (TfR) was modified to include an O-glycosylation sequence of the present invention at the C-terminus of the light chains of the antibody. The O- glycosylation of the sequence allowed the introduction of a DBCO (dibenzocyclooctyne) linker which was used to generate a conjugate of the antibody and a siRNA specific for a chosen target gene expressed in the brain. The O-glycosylation included an azide group on the terminal sialic acid, allowing for conjugation to the DBCO group of the functionalized siRNA. The antibodysiRNA conjugate (ARC-181) generated therefore includes an antibody specific for TfR to allow the conjugate to cross the blood brain barrier (BBB) and the siRNA then to reduce expression of the target gene in the CNS. A control conjugate, ARC-180, was also generated in the same way differing only in that the antibody used in the conjugate was an isotype control recognizing a different target to TfR meaning that the control should not be transported across the BBB via the TfR. Table 5 provides further details of the ARC-180 and ARC-181 conjugate preparations subsequently used in the study in mice.

[0329] To demonstrate that mouse TfR binding protein (mTBP)-siRNA conjugates (ARC) utilizing glycoconjugate technology crosses the blood-brain- barrier (BBB) and delivers the siRNA cargo to the CNS to reduce target mRNA gene expression, a proof-of-concept study was conducted to assess pharmacodynamic efficacy of the constructs with peripheral delivery in mice. PBS control, Isotype Ab-siRNA conjugate, or mTBP2-siRNA were dosed in 8-week- old FVB mice at 10 mg/kg effective siRNA concentration intravenously as a single dose and sacrificed after 14, 28 and 70 days respectively (see Figure 26) In addition, mouse anti-CD4 antibody (GK1.5) was dosed at 10 mg/kg 2 to 3 days prior to the study to ablate CD4 positive T cells to mitigate undesired pharmacokinetic consequences resulting from spurious anti-drug antibody responses to injected compounds.

[0330] Mice at designated time points were fully anesthetized, then underwent cardiac perfusion with cold PBS (6ml/min for 5 min) until blood was completely removed to collect brain and spinal cord to assess target mRNA levels by RT- qPCR. For RT-qPCR, RNA was isolated by using RNeasy Plus Universal Mini Kit (Qiagen 73404). Briefly hemibrain tissue homogenates were prepared with FastPrep-24 Lysing Matrix D beads to homogenize the tissues with MP Fastprep 24 (MP Biomedical) for at 6m/s for 40 seconds at 4°C, then centrifuging the vials to collect the supernatant. The RNA was then collected. Following determination of RNA quantity with A260/280 ratio with a spectrophotometer, cDNA was generated on Mastercycler X50a (Eppendorf), and qPCR was carried out on QuantStudio 7 Flex Real-Time PCR System (Applied Biosystems). Gene expression levels of the target gene were normalized by P-actin using respective probes (ThermoFisher).

[0331] As shown in Figure 26, single IV administration of mTfR-siRNA conjugate (ARC-181) in mice led to robust selective reduction of target in the brain compared to PBS dosed control, beginning at 14 days following dosing (61% mRNA reduction). Importantly, Isotype Ab-siRNA conjugate (ARC-180) did not elicit significant reduction in target mRNA demonstrating that active TfR mediated transport was required to deliver siRNA cargo to the CNS. Moreover, this provided evidence of successful utilization of present invention to enable blood-brain barrier crossing and delivery of cargo to the brain.

[0332] Example 9: Use of the conjugation method provided to generate multi-functional antibodies and combinatorial libraries [0333] Figure 27 illustrates how the conjugation method provided can be used to generate multi-functional binding molecules, such as multi-functional antibodies, as well as combinatorial libraries. The conjugation groups introduced as part of the O-glycosyl sugar chains are highly versatile and can be used to join any two molecules desired. The approach may also be used to generate different permutations of panels of molecules so the combination can be assessed for desired properties, such as synergy, or the ability to bind more than one desired epitope.

[0334] Example 10: Further conjugates

[0335] Introduction

[0336] A further linker was prepared and used in generating a conjugate in conjunction with glycosylation sequence of the present invention.

[0337] Materials and Methods

The following were utilized:

• Peptide acetyl -KRRRRRRK-NH2 (5 mg Genscript lot#:U047MGA130- 1/PE1083) depicted in Figure 28(a);

• DBCO-PEG5-NHS (2 vials of 2 mg) Click Chemistry Tools cat#:A102P-2 lot#:3955 depicted in Figure 28(b);

• N-methylpyrrolidone (NMP); and

• N-N’ -Diisopropylethylamine (DIEA) Sigma cat#: 496219 lot#:33496KJ.

[0338] Linker preparation

[0339] A vial of 2 mg DBCO-PEG5-NHS was dissolved in 100 pL of NMP and then transferred to a new vial of DBCO-PEG5-NHS with mixing to dissolve. Once dissolved the contents of the vial were transferred to a peptide vial and 1 pL of DIEA added. The contents of the vial were mixed and the vial incubated at room temperature. [0340] 0.5 pL of the contents of the vial were diluted 1 :500. 20 pL of the dilution were placed in an Agilent micro vial and a CE/MS analysis performed to verify the completeness of the reaction. The linked obtained is shown in Figure 28(c).

[0341] Once the reaction was judged complete by CE/MS, 1 mL of water and 5 pL of 5% acetic acid were added with the pH adjusted to pH ~5.5 as measured by using a pH strip. No precipitation was observed and the preparation was stored at 4 °C.

[0342] Linker Purification and Lyophilization

[0343] The peptide was purified by preparatory HPLC on a Shimadzu LC system (system controller CBM-20A; pump model LC-20AP; oven model CTO-20A; detector SPD-20A; fraction collector FRC-10A):

• Buffer A: 0.05% TFA in water.

• Buffer B: Acetonitrile.

• Gradient: 18% B to 30% B over 39 minutes

• Column: Waters SymmetryPrep C18 7 um, 19x300mm, Part No WAT066245, S/N 01553030811204.

• Column oven: 50 °C

• Flow rate: 20 mL/min

[0344] The purity of the linker fractions was assessed using analytical RP-HPLC (Agilent 1290 Infinity II LC system) and fractions >95% pure were pooled. Subsequent lyophilization of the final main product pool yielded the lyophilized linker TFA salt (3.3 mg, >95% pure). The molecular weight was determined by LC7MS to match the structure shown in Figure 28c (Found: [M+3H]3+=804.1; Calculated [M+3H]3+=804.3; Found MW (avg)=2409.3; Calc. MW (avg): 2409.88).

[0345] Linker Conjugation to a Fab antibody fragment via an O-glycosylated glycosylation sequence

[0346] 0.5 pmol Fab was placed into a 50 mL tube and 1.37 mmol of DBCO linker peptide added. The tube was then mixed by several inversions and placed at 4°C for a weekend. CE/MS was used to assess the extent of reaction, indicating -75% conversion. The material was concentrated with a Millipore Ultracell-15 30kDa MWCO filter 50 mL tube style, then washed three times with PBS buffer. A Final recovery of 12.6 mg was expected, with about 80% conjugation obtained.

[0347] All patents, published patent applications and publications cited herein are incorporated by reference as if set forth fully herein.

SEQUENCES

[0348] SEQ ID NO: 1 provides the sequence of a preferred glycosylation amino acid sequence of the present invention, PTPAP.

[0349] SEQ ID NO: 2 provides the sequence of a further preferred glycosylation amino acid sequence of the present invention, AAAPTPAPAAA.

[0350] SEQ ID NO: 3 provides the sequence of a glycosylation test amino acid sequence employed in the Examples of the present application, AAATPAP.

[0351] SEQ ID NO: 4 provides the sequence of a further glycosylation test amino acid sequence employed in the Examples of the present application, PTPSP.

[0352] SEQ ID NO: 5 provides the sequence of a further glycosylation test amino acid sequence employed in the Examples of the present application, DTPPP.

[0353] SEQ ID NO: 6 provides the sequence of the gene encoding bifunctional UDP-N-acetylglucosamine 2-epimerase/N-acetylmannosamine kinase (also known as UDP-GlcNAc-2-epimerase/ManAc kinase, GNE), ATGGAGAAGAATGGGAATAACCGAAAGCTTCGGGTTTGCGTTGCTACC TGCAACCGTGCAGATTACTCCAAATTGGCCCCGATCATGTTCGGCATT AAGACAGAGCCTGCCTTCTTTGAGCTGGATGTGGTGGTGCTGGGCTCT CACCTCATAGATGACTACGGAAACACATATCGAATGATTGAGCAAGAT GACTTTGACATTAACACCAGGCTACACACGATCGTTAGAGGGGAAGAT GAAGCAGCCATGGTAGAGTCAGTAGGCCTAGCTCTAGTGAAGCTACCA GATGTCCTTAATCGCCTGAAGCCTGACATCATGATTGTTCATGGAGAC CGATTTGATGCCCTTGCCCTGGCTACATCTGCTGCCTTGATGAACATCC GAATCCTTCACATTGAAGGAGGAGAGGTCAGTGGGACTATTGATGACT CTATCAGACATGCCATAACAAAACTGGCTCATTACCACGTGTGCTGCA CCAGAAGTGCAGAACAACACCTGATATCCATGTGTGAGGACCACGAC CGTATCCTTTTGGCAGGCTGCCCTTCCTATGACAAACTGCTCTCAGCCA

AGAATAAAGACTATATGAGCATCATTCGGATGTGGCTAGGTGATGATG

TAAAATGTAAAGATTACATTGTTGCACTGCAGCACCCTGTGACCACTG

ACATTAAGCATTCCATAAAGATGTTTGAATTAACACTGGATGCACTTA

TCTCATTTAACAAGAGGACCCTAGTTCTGTTTCCAAATATTGATGCAGG

GAGCAAGGAGATGGTTCGAGTGATGCGGAAGAAGGGCATTGAGCATC

ATCCCAATTTTCGTGCAGTCAAGCATGTCCCATTTGACCAGTTTATACA

GCTGGTCGCCCATGCTGGTTGTATGATTGGCAATAGCAGCTGTGGAGT

TCGAGAGGTTGGCGCTTTTGGAACACCCGTGATAAACCTGGGAACACG

CCAGATAGGAAGAGAAACAGGGGAGAATGTTCTTCATGTGCGGGATG

CTGACACCCAAGATAAAATATTACAAGCGCTCCACCTTCAGTTCGGTA

AACAGTACCCTTGCTCAAAGATATATGGGGATGGAAATGCTGTTCCAA

GGATTTTAAAGTTTCTCAAATCTATTGACCTCCAAGAGCCACTACAGA

AGAAATTCTGCTTCCCTCCTGTGAAGGAGAACATCTCTCAAGATATTG

ACCATATTCTTGAAACTCTGAGTGCCTTGGCTGTTGATCTTGGAGGAAC

AAACCTCAGAGTGGCAATAGTTAGCATGAAGGGTGAAATAGTTAAGA

AGTACACTCAGTTCAATCCTAAAACCTACGAAGAAAGGATTAGTTTAA

TCCTGCAGATGTGTGTGGAAGCTGCAGCAGAAGCTGTGAAACTGAACT

GCAGAATTCTGGGAGTAGGCATTTCCACAGGTGGCCGTGTGAATCCTC

AGGAAGGAATTGTGCTGCATTCAACCAAACTGATACAGGAATGGAAC

TCTGTGGACCTCAGAACACCCCTCTCTGACACCTTACATCTCCCTGTGT

GGGTGGACAATGATGGCAACTGTGCTGCCATGGCAGAAAGGAAGTTT

GGCCAAGGAAAAGGACAAGAAAACTTTGTGACACTCATTACAGGCAC

AGGGATTGGTGGTGGAATCATTCACCAGCATGAGCTGATCCACGGCAG

CTCTTTCTGTGCGGCAGAGCTTGGCCACCTCGTGGTGTCTCTGGATGGG

CCTGACTGTTCCTGTGGAAGCCATGGCTGTATTGAAGCATATGCTTCTG

GGATGGCCTTGCAGAGGGAAGCGAAGAAGCTCCATGATGAGGATCTG

CTCTTGGTGGAAGGGATGTCAGTGCCAAAAGACGAACCTGTGGGTGCC

CTCCATCTCATCCAGGCCGCCAAACTTGGCAACGTGAAGGCCCAGAAT

ATCCTACGAACAGCTGGAACTGCTTTGGGACTTGGGGTTGTGAACATC

CTCCACACTATGGATCCTTCCCTGGTGATCCTGTCTGGAGTCCTGGCCA GTCACTACATCCACATTGTCAAGGACGTCATCCGCCAGCAAGCCTCGT CCTCCGTGCAGGATGTGGACGTCGTGGTTTCAGACTTGGTTGACCCTG CCCTGCTTGGTGCTGCCAGCATGGTTCTGGACTACACAACCCGCAGGA TCCACTAG

[0354] SEQ ID NO: 7 provides the sequence of the hamster Gne-1- pLentiCRISPRv2(T 15257), ATGGGAATAACCGAAAGCTT.

[0355] SEQ ID NO: 8 provides the sequence of the hamster Gne-2- pLentiCRISPRv2(T 15258), CCGTGCAGATTACTCCAAAT.

[0356] SEQ ID NO: 9 provides the sequence of the hamster Gne-3- pLentiCRISPRv2(T 15259), CCAATTTGGAGTAATCTGCA.

[0357] SEQ ID NO: 10 provides the sequence of the hamster Gne-4- pLentiCRISPRv2(T 15260), CTTAATGCCGAACATGATCG.

[0358] SEQ ID NO: 11 provides the sequence of the hamster Gne-5- pLentiCRISPRv2(T 15261), AC ATCC AGCTC AA AGAAGGC .

[0359] SEQ ID NO: 12 provides the amino acid sequence of the linker,

Ac-K(PEG5-DBCO)-RRRRRR-K(PEG5-DBCO)-NH₂.

Claims

CLAIMS We claim:

1. A molecule which comprises one or more copies of the following glycosylation amino acid sequence:

Xi Thr Pro X₂ X₃ wherein:

Xi, X₂, and X3 are any amino acid; and the Threonine (Thr) amino acid residue is O-glycosylated with a sialylated sugar.

2. The molecule of claim 1, wherein:

Xi is Pro;

X3 is Pro;

Xi and X3 are both Pro; or

X₂is Ala.

3. The molecule of claim 2, wherein Xi and X3 are both Pro.

4. The molecule of any one of the preceding claims, wherein the amino acid sequence Xi Thr Pro X₂ X3 is the amino acid sequence Pro Thr Pro Ala Pro.

5. The molecule of any one of the preceding claims, wherein the sialylated sugar is a sialylated N-acetylhexosamine linked to hexose.

6. The molecule of any one of the preceding claims, wherein the molecule comprises a binding molecule, interleukin, cytokine, chemokine, hormone, or enzyme.

7. The molecule of claim 6, wherein the binding molecule is an antibody.

8. The molecule of claim 7, wherein the antibody is a bispecific antibody.

9. The molecule of any one of the preceding claims, wherein the molecule comprises a nucleic acid molecule.

10. The molecule of claim 9, wherein the nucleic acid molecule is an siRNA molecule.

11. The molecule of any one of the preceding claims, wherein said sialylated sugar comprises:

(a) a chemical group that can be conjugated to a moiety; or

(b) a chemical group that is conjugated to a moiety.

12. The molecule of claim 11, wherein said chemical group comprises a click chemistry group.

13. The molecule of claim 12, wherein the click chemistry group is an azide or alkyne group.

14. The molecule of any one of claims 11 to 13, wherein said sialylated sugar comprises a chemical group that is conjugated to a moiety.

15. The molecule of claim 14, wherein the moiety comprises a linker.

16. The molecule of claim 15, wherein the linker comprises a poly arginine amino acid sequence.

17. The molecule of any one of claims 14 to 16, wherein the moiety itself also comprises at least one copy of the Xi Thr Pro X2 X3 glycosylation amino acid sequence.

18. The molecule of claim 17, wherein a Xi Thr Pro X2 X3 glycosylation amino acid sequence present in the moiety is used to conjugate the moiety to a Xi Thr Pro X2 X3 glycosylation amino acid sequence elsewhere in the molecule.

19. The molecule of any one of claims 14 to 18, wherein the molecule is a bispecific antibody, wherein:

(a) a copy of the Xi Thr Pro X2 X3 glycosylation amino acid sequence in the molecule is present in an antibody heavy chain and is conjugated to a moiety which comprises an antibody light chain, so that the antibody heavy chain and antibody light chain form an antigen-binding site with a first specificity; and

(b) a second copy of the Xi Thr Pro X2 X3 glycosylation amino acid sequence in the molecule is present in a second antibody heavy chain and is conjugated to the moiety which comprises a second antibody light chain, so that the second antibody heavy chain and second antibody light chain form an antigen-binding site with a second specificity, wherein the first and second specificities are not the same.

20. The molecule of any one of the preceding claims, wherein the molecule comprises two Xi Thr Pro X2 X3 glycosylation amino acid sequences, wherein the Thr of each of the sequences is O-glycosylated with a sugar chain comprising a sialylated sugar comprising a chemical group allowing conjugation between the two Xi Thr Pro X2 X3 glycosylation amino acid sequences.

21. The molecule of claim 20, wherein the conjugation acts as a bridge joining a first and second binding molecule to form the overall molecule or joining scaffolds each comprising antigen-binding sites to form the overall molecule.

22. The molecule of any one of the preceding claims, wherein the sialylated sugar comprises a chemical group allowing conjugation to a moiety, with the molecule further comprising a different means for conjugation at a separate site or sites in the molecule.

23. The molecule of claim 22, wherein the different means for conjugation is one or more cysteine residues.

24. The molecule of claim 23, wherein the molecule comprises:

(a) a Xi Thr Pro X2 X3 glycosylation amino acid sequence which is conjugated to a first moiety; and

(b) a cysteine residue which is conjugated to a second moiety which is different to the first moiety.

25. The molecule of claim 24, wherein the Xi Thr Pro X2 X3 glycosylation amino acid sequence and the cysteine residue are present together in the same polypeptide in the molecule.

26. The molecule of any one of the preceding claims, wherein the molecule comprises from two to ten of the Xi Thr Pro X2 X3 glycosylation amino acid sequences.

27. A pharmaceutical composition comprising the molecule of any one of claims 1 to 26 and a pharmaceutically acceptable carrier.

28. The molecule of any one of claims 1 to 26 for use in therapy of the human or animal body.

29. The molecule of any one of claims 1 to 26 for use in treating a condition selected from cancer, heart disease, obesity, an autoimmune condition, an inflammatory condition, diabetes, or a CNS disorder.

30. The molecule for the use of claim 29, wherein the molecule is administered to, or targeted to, the CNS.

31. The molecule for the use of claim 30, wherein: (a) the molecule comprises an antibody specific for a protein allowing delivery across the blood brain barrier; and/or

(b) the molecule comprises a moiety conjugated to the Xi Thr Pro X2 X3 glycosylation amino acid sequence, wherein the moiety comprises a nucleic acid molecule capable of inhibiting expression of a target gene.

32. The molecule for the use of any one of claims 28 to 31, wherein the molecule comprises an antibody specific for a Transferrin Receptor (TfR) allowing transport across the blood brain barrier.

33. A method of treating a condition comprising administering an effective amount of a molecule according to any one of claims 1 to 26 to a subject in need thereof.

34. The method of claim 33, wherein the condition is selected from cancer, heart disease, obesity, an autoimmune condition, an inflammatory condition, diabetes, or a CNS disorder.

35. The method of claim 34, wherein:

(a) the molecule comprises an antibody specific for a protein allowing delivery of the molecule across the blood brain barrier; and/or

(b) the molecule comprises a moiety which is a nucleic acid molecule capable of inhibiting expression of a target gene.

36. The method of claim 35, where the molecule comprises an antibody specific for a Transferrin Receptor (TfR) allowing transport across the blood brain barrier.

37. A cell which comprises:

(a) an endogenous UDP-N-acetylglucosamine 2-epimerase - ManNAc Kinase gene wherein the gene is mutated so that at least the UDP-N- acetylglucosamine 2-epimerase function is reduced or eliminated; and

(b) a sequence encoding a polypeptide comprising the amino acid the amino acid sequence:

Xi Thr Pro X2 X3 wherein

Xi, X2, and X3 are any amino acid; and the Threonine (T) amino acid residue is O-glycosylated with a sialylated sugar.

38. The cell of claim 37, wherein the sialylated sugar comprises a chemical group, wherein the chemical group:

(a) can be conjugated to a moiety; or

(b) is conjugated to a moiety.

39. The cell of claim 37 or 38, wherein the UDP-N-acetylglucosamine 2-epimerase - ManNAc Kinase gene is mutated so that both the epimerase and kinase functions are reduced or eliminated.

40. The cell of claim 39, wherein both the epimerase and kinase functions are knocked-out.

41. A method of producing a glycosylated polypeptide comprising culturing the cell of any one of claims 37 to 40, in a medium supplemented with peracetylated ManNAz .

42. A method of introducing a glycosylation site into a polypeptide comprising modifying the sequence of the polypeptide to include the amino acid sequence:

Xi Thr Pro X₂ X₃ wherein

Xi, X₂, and X₃ are any amino acid; and the Threonine (T) amino acid residue is O-glycosylated with a sialylated sugar.

43. The method of claim 42, wherein the sialylated sugar comprises a chemical group wherein the chemical group:

(a) can be conjugated to a moiety; or

(b) is conjugated to a moiety.

44. A method of conjugating a molecule to a moiety, the method comprising:

(a) providing a molecule of any one of claims 1 to 26, wherein the sialylated sugar comprises a chemical group that can be conjugated to a desired moiety with a compatible chemical group;

(b) contacting the molecule of (a) with the desired moiety; and

(c) allowing the chemical group of the molecule to undergo conjugation with the desired moiety through the compatible chemical groups.

45. The method of claim 44, wherein the conjugation of the molecule and moiety is via click chemistry.

46. The method of claim 44 or 45, wherein the contacting is in vitro, in vivo, or ex vivo.

47. A method of joining together two molecules comprising:

(a) providing a molecule of any one of claims 1 to 26, wherein the sialylated sugar comprises a chemical group that can be conjugated to a desired second molecule which has a compatible chemical group allowing the conjugation;

(b) contacting the molecule of (a) with the desired second molecule; and

(c) allowing conjugation of the first and second molecules via the compatible chemical groups.

48. A method of generating a combinatorial library, wherein the method comprises:

(a) providing a plurality of molecules of any one of claims 1 to 26, wherein the molecules differ from each other, but the sialylated sugar of each molecule comprises the same chemical group that can be conjugated to a desired moiety;

(b) providing a second plurality of molecules, wherein the second plurality of molecules differ from each other and the molecules of (a), but the second plurality of molecules comprise a compatible chemical group to the chemical group of the molecules of (a) allowing conjugation; and

(c) contacting the molecules of (a) and (b) under conditions allowing conjugation and hence the generation of the combinatorial library.

49. A method of conjugating an antibody to a desired moiety comprising:

(a) providing a molecule of any one of claims 1 to 26, wherein the sialylated sugar of the molecule comprises a chemical group that can be conjugated to a compatible chemical group of the desired moiety, wherein the molecule is either the antibody or is a component part of the antibody; and

(b) conjugating the desired moiety to the molecule via said chemical group, wherein if the molecule is a component part of an antibody, rather than the antibody itself, the method further comprises assembling the whole antibody.

50. A method of labelling a molecule, wherein the method comprises:

(a) providing a molecule of any one of claims 1 to 26 wherein the sialylated sugar of the molecule comprises a chemical group that can be conjugated to a desired label which comprises a compatible chemical group to allow conjugation;

(b) providing the label with the compatible group that allows conjugation to said chemical group of the molecule of (a); and

(c) contacting the molecule of (a) and the label of (b) under suitable conditions to bring about conjugating of the two.

51. Use of a molecule of any one of claims 1 to 26 as a capture agent for a desired moiety wherein the sialylated sugar of the molecule comprises a chemical group that can be conjugated to a desired moiety and the desired moiety comprises a compatible conjugation group for the conjugation.

52. A cell encoding a polypeptide which comprises one or more copies of the following glycosylation amino acid sequence:

Xi Thr Pro X₂ X₃ wherein

Xi, X₂, and X₃ are any amino acid; and the Threonine (T) amino acid residue at the second position is O- glycosylated with a sialylated sugar.

53. The cell of claim 52, wherein the molecule comprises a chemical group, wherein:

(a) the chemical group can be conjugated to a moiety; or

(b) the chemical group is conjugated to a moiety.

54. A cell encoding a molecule according to any one of claims 1 to 26.