Multispecific antibodies for neurological disorders

Abstract

The invention relates to multispecific polypeptides comprising a first protein for the prevention, treatment or detection of a neurological disorder in the central nervous system, a second protein binding specifically to a protein expressed on the blood brain barrier, wherein the protein expressed on the brain blood barrier mediates enhanced uptake of compounds over the BBB, and a third protein binding specifically to a protein in the CNS.

Description

[0001] MULTISPECIFIC ANTIBODIES FOR NEUROLOGICAL DISORDERS FIELD OF THE INVENTION The invention relates to antibodies binding to MOG and carrying a therapeutic protein for the treatment of neurological disorders. BACKGROUND OF THE INVENTION Therapeutic proteins for the treatment of neurological disease are promising although limited by difficulties for passing the blood-brain barrier (BBB). Therefore, no therapeutic proteins have thus far been approved that cross the BBB to exert their effect. To facilitate the effective delivery of protein therapeutics into the brain, researchers have fused therapeutic mAb (fragments) to brain targeting mAb (fragments) thereby enabling brain accumulation through different mechanisms. The most well-studied approach is the use of mAb (fragments) targeting nutrient receptors that shuttle the BBB (transferrin receptor (TfR), insulin receptor, CD98hc, ^), enabling transport of these bispecific antibodies to the brain. More recently, researchers have explored the alternative possibility to decrease the brain efflux of antibodies after their limited passive diffusion into the brain. Several studies have shown the success of this non-invasive TfR BBB shuttling technology to increase brain concentrations of therapeutics [Wouters et al. (2020) Fluids Barriers CNS 17, 62; Kariolis et al. (2020) Sci. Transl. Med. 12, eaay1359; Wouters et al. (2022) Fluids Barriers CNS 19, 79; Atwal et al. (2011) Science trans. Med.3, 84ra43]. However, the main downside of using TfR for RMT, is the widespread expression of TfR in multiple peripheral organs, leading to fast peripheral clearance and hence lower central exposure. Therefore, the observed therapeutic effect is rather short and in a clinical setting, more frequent intravenous dosing might be required to keep levels high enough to elicit a therapeutic effect. More recently, an alternative possibility to decrease the brain efflux of antibodies after their limited passive diffusion into the brain was explored [Nakano et al. (2019) PLOS ONE 14, e0214404]. To decrease its efflux, a therapeutic protein was fused to a mAb fragment targeting an abundant brain protein, i.e. myelin oligodendrocyte glycoprotein (MOG), which resulted in significantly increased brain accumulation of the antibody-drug conjugate. However, the achieved brain concentrations remained rather low, and no therapeutic effect was observed [Nakano et al. cited above; WO2018123979]. SUMMARY OF THE INVENTION The invention relates to multispecific antibodies, comprising a therapeutic protein for a neurological disease and an antibody or antigen binding fragment thereof binding to a structural protein in the brain. The retention to the structural protein results in an accumulation of the protein and / or a delayed clearance from the brain. Therapeutic antibodies for the treatment of neurological disease show great potential, but their applications are still rather limited due to limited brain exposure. Finding ways to increase brain exposure of antibodies by increasing their influx or decreasing their efflux is a critical step to boost the development of innovative antibody drugs for neurological disorders. The most well-studied approach to enhance brain influx of protein therapeutics into the brain, is receptor-mediated transcytosis (RMT) by targeting nutrient receptors to shuttle protein therapeutics over the blood-brain barrier (BBB), often along with their endogenous cargos. While, higher brain exposure is achieved with this approach, the timeframe is short due to rather fast brain clearance. Here, we report the discovery of VHHs binding to mouse, human, and cynomolgus myelin oligodendrocyte glycoprotein (MOG), a CNS specific protein. When fused to antibodies shuttling the BBB through RMT, those anti-MOG VHHs decrease brain efflux of the fused antibodies leading to a dramatically increased brain half-life and a prolonged therapeutic effect of the fused antibodies. The invention is further summarised in the following statements: 1. A multispecific polypeptide comprising: - a first protein for the prevention, treatment or detection of a neurological disorder in the central nervous system, - a second protein binding specifically to a protein expressed on the blood brain barrier, wherein the protein expressed on the blood brain barrier mediates enhanced uptake of compounds over the BBB, characterised in further comprising a third protein binding specifically to a protein in the CNS, more typically to a protein in the brain. Herein ^first^, ^second^ and ^third^ protein are different proteins. The multispecific polypeptide can be a single polypeptide, but typically consists of more that one polypeptide chain bound by protein-protein interaction and / or disulfide bridges, as shown for example in the multispecific antibodies shown in figure 1, and discussed in detail in the description. Furthermore, in the multispecific protein, a ^first^, ^second^ and ^third^ protein can be present in multiple copies to enhance the pharmaceutic activity, or to increase binding to the BBB or a protein in the brain. 2. The multispecific polypeptide according to statement 1, wherein the first protein is a protein for prevention or treatment of a neurological disorder in the central nervous system. 3. The multispecific polypeptide according to statement 1 or 2, wherein the first protein is a protein for the prevention, treatment or detection of a neurological disorder in the brain. 4. The multispecific polypeptide according to any one of statements 1 to 3, wherein the first protein is a protein for the treatment of a neurological disorder in the brain. 5. The multispecific polypeptide according to any one of statements 1 to 4, wherein the first protein is an antibody or antigen binding fragment thereof binding specifically to a therapeutic target in the brain. 6. The multispecific polypeptide according to claim 5, wherein the therapeutic target is BACE1. 7. The multispecific polypeptide according to any one of statements 1 to 6, wherein the second protein binds to a membrane protein involved in receptor- mediated transcytosis . 8. The multispecific polypeptide according to any one of statements 1 to 7, wherein the second protein binds to a receptor involved in receptor-mediated transcytosis. 9. The multispecific polypeptide according to statement 8, wherein the receptor involved in receptor -mediated transcytosis over the blood brain barrier is transferrin receptor (TfR). 10. The multispecific polypeptide according to any one of statements 1 to 9, wherein the third protein is a protein binding specifically to a structural protein in the CNS, more specifically to a structural protein in the brain. 11. The multispecific polypeptide according to statement 10, wherein the structural protein is MOG (myelin oligodendrocyte glycoprotein). 12. The multispecific polypeptide according to any one of statements 1 to 11, wherein the second protein and / or the third protein is an antibody or antigen binding fragment thereof. 13. The multispecific polypeptide according to statement 12, wherein one or more of the antibodies are single domain antibodies (sdAbs). 14. The multispecific protein according to any one of statement 1 to 13 comprising: - an antibody or antigen binding fragment thereof binding specifically to a therapeutic target of a neurological disorder in the brain, - an antibody or antigen binding fragment thereof binding specifically to the transferrin receptor, - one or more antibodies or antigen binding fragments thereof binding specifically to MOG. 15. The multispecific protein according to any one of statements 1 to 14, wherein the antibody binding to MOG is a single domain antibody defined by: CDR1 sequence GFSLDYYAIG [SEQ ID NO:2], CDR2 sequence CIGSRDGRTF [SEQ ID NO:3] and CDR3 sequence AYNTAGQRCPSNGYDYDYWG [SEQ ID NO:4]. 16. The multispecific antibodies according to any one of statements 1 to 14, wherein the antibody binding to MOG is a single domain antibody defined by: CDR1 sequence GFSLDYYAIG [SEQ ID NO:2], CDR2 sequence CISSRDGRTF [SEQ ID NO:6], and CDR3 sequence AYNTIGQRCPSNGYDYDYWG [SEQ ID NO:7]. 17. A polynucleotide or polynucleotides encoding the multifunctional polypeptide as defined in any one of statement 1 to 16. 18. A multispecific polypeptide according to any one of statements 1 to 16, for use in the prevention, treatment or detection of a neurological disorder in the brain. 19. A polynucleotide or polynucleotides encoding the polypeptides as defined in any one of statement 1 to 16, for use in the prevention, treatment or detection of a neurological disorder in the brain. 20. A single domain antibody defined by: CDR1 sequence GFSLDYYAIG [SEQ ID NO:2], CDR2 sequence CIGSRDGRTF [SEQ ID NO:3], and CDR3 sequence AYNTAGQRCPSNGYDYDYWG [SEQ ID NO:4]. 21. A single domain antibody defined by: CDR1 sequence GFSLDYYAIG [SEQ ID NO:2], CDR2 sequence CISSRDGRTF [SEQ ID NO:6], and CDR3 sequence AYNTIGQRCPSNGYDYDYWG [SEQ ID NO:7]. DETAILED DESCRIPTION Figure legends Figure 1. Multispecific antibody constructs. Schematic overview of produced antibodies binding to BACE1 or SarsCov, TfR or GFP, and MOG. Figure 2. mMOG / hMOG / cMOG affinities of VHHs and antibodies. Flow cytometry binding curves of VHHs and antibodies to CHO cells overexpressing mouse MOG (A / D), human MOG (B / E), and cynomolgus MOG (C / F). Data represents mean ± SEM (BBB00498-BBB00500: n = 2, BBB00497- BBB00499- BBB00500 -BBB00501-BBB00502: n = 1, biMOGHigh: n = 2; monoMOGHigh, biMOGLow, monoMOGLow: n = 3). Figure 3. MOG binding valency and affinity affects in vivo pharmacokinetics / pharmacodynamics . Mice received a single intravenous injection (80nmol / kg) with PBS (control), biMOGHigh:monoTfR:BACE1, monoMOGHigh: monoTfR:BACE1, biMOGLow: monoTfR: BACE1, monoMOGLow:monoTfR:BACE1 or monoTfR:BACE1 (= noMOG). Pharmacokinetics / pharmacodynamics were evaluated one and three days post treatment. (A) Plasma concentration of treatment antibodies, fast plasma clearance was observed for all antibodies. (B) Effect of treatment on A 1-40 levels in plasma, a similar decrease in plasma A 1-40levels was observed for all antibodies. Three days post treatment the decrease was less pronounced then one day post treatment. (C) Brain concentration of treatment antibodies, no major differences were observed one day post treatment while MOG binding significantly increased brain antibody levels three days post treatment. Bivalent binding showed the highest antibody levels in brain. (D) Effect of treatment on A 1-40 levels in brain, similar decrease in A 1-40 levels was observed one and three days post treatment for all MOG binding antibodies. Bar graphs represent mean ± SEM (n = 4 per group). Statistical test: two-way ANOVA with Dunnett^s multiple comparisons test compared to monoTfR:BACE1 injected mice for antibody concentrations and to PBS injected control mice for A 1-40 levels. (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001) Figure 4. Enhanced and prolonged brain uptake in mice of biMOGLow:monoTfR:BACE1. Mice received a single intravenous injection of monoTfR:BACE1 (160 nmol / kg) or biMOGLow:monoTfR:BACE1 (80 nmol / kg, 160 nmol / kg) or PBS (control). Pharmacokinetics / pharmacodynamics were evaluated over a period of 7-14 days. (A) Schematic representation of biMOGLow:monoTfR:BACE1 and monoTfR:BACE1 . (B) Plasma concentration of treatment antibodies over time showing fast peripheral clearance. (C) Effect of treatment on A 1-40 levels in plasma, a decrease in therapeutic effect is observed over time. (D) Brain concentration of treatment antibodies indicating a higher brain antibody exposure with biMOGLow:monoTfR:BACE1. (E) Effect of treatment on A 1-40 levels in brain, a prolonged therapeutic effect is observed with biMOGLow:monoTfR:BACE1 compared to monoTfR / BACE1. Curves represent mean ± SEM (n = 3-8 per group). Statistical test: Two-way ANOVA, all conditions compared to monoTfR:BACE1. (D) Significant treatment*time interaction effect for high dose (p < 0.0001) and low dose (p = 0.0001). (E) Significant treatment*time interaction effect for high dose (p = 0.0208) and low dose (p = 0.0413), only points considered till day 7. (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001) Figure 5. Brain and plasma pharmacokinetics / pharmacodynamics of MOG binding without active shuttling. Mice received a single intravenous injection of 160 nmol / kg of monoGFP:BACE1 or biMOGLow:monoGFP:BACE1 or PBS (control). Pharmacokinetics / pharmacodynamics were evaluated over a period of 14 days. (A) Schematic representation of biMOGLow:monoGFP:BACE1 and monoTfR:BACE1 . (B) Plasma concentration of treatment antibodies over time showing a slower peripheral clearance compared to TfR binding equivalents. (C) Effect of treatment on A 1-40 levels in plasma, the decrease in plasma A 1-40 levels remained constant over time. (D) Brain concentration of treatment antibodies indicating a longer half-life for biMOGLow:monoGFP:BACE1 but overall very low brain concentrations. (E) Effect of treatment on A 1-40 levels in brain, no therapeutic effect is observed. Curves represent mean ± SEM (n = 3-6 per group). Statistical test: Two-way ANOVA. (D) Significant treatment*time interaction effect (p = 0.0001). (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001) Figure 6. Plasma and brain pharmacokinetics of antibodies with non-targeted Fabs. Mice received a single intravenous injection of 160 nmol / kg of monoTfR:SarsCov, biMOGLow:monoTfR:SarsCov or biMOGLow:monoGFP:SarsCov (dashed lines). Pharmacokinetics / pharmacodynamics were evaluated over a period of 7-14 days. (A) Schematic representation of the antibodies. (B) Plasma concentration of treatment antibodies over time showing similar profiles for SarsCov and BACE1 binding antibodies. (C) Brain concentration of actively shuttled treatment antibodies indicating a higher brain antibody exposure with biMOGLow:monoTfR:SarsCov, which is still detectable in brain 49 days post treatment. (D) Brain concentration of antibodies without active shuttling, showing an increased brain exposure with non-targeting Fabs. Curves represent mean ± SEM (n = 3-6 per group). In the figures description of multispecific antibodies the abbreviations mono or bi are used to indicate respectively the presence of one or two of the same nanobodies in the compound. Furthermore the indication low and high describes the affinity of the nanobody for mouse MOG (High = BBB00498, Low = BBB00500). Thus biMOGLow:monoTfR:BACE1 is a multispecific antibody with two anti-MOG nanobodies with low affinity, one anti-TfR nanobody and one anti-BACE1 Fab- fragment. An aspect of the invention relates to a multispecific protein comprising: - a therapeutic protein for a neurological disorder in the central nervous system and especially in the brain. Apart from therapeutic proteins, proteins used for prevention and also for diagnosis in the CNS and in the brain, are envisaged. - an antibody or antigen binding fragment thereof binding specifically to a protein in the brain, typically a structural protein in the brain. -an antibody or antigen binding fragment thereof binding specifically to a protein expressed on the BBB and mediating enhanced uptake, thus enabling shuttling of compounds over the blood brain. Such proteins may be receptors, more particularly involved in receptor-mediated transcytosis. Herein examples of receptors involved in enabling shuttling over the blood brain barrier and / or involved in are transferrin receptor, the insulin receptor, low-density lipoprotein receptor related proteins 1 and 2, diphtheria toxin receptor, CD98hc, SLC7A5, SLC7A1, IGF1R and TMEM30A. Neurological disorders in the context of the present invention include Frontotemporal Dementia, Amyotrophic Lateral Sclerosis, Alzheimer^s Disease, Parkinson's Disease, Multiple Sclerosis, Traumatic Brain Injury, Huntington^s Disease, lysosomal storage disorders (Hunter syndrome, Hurler Syndrome, Sanfilippo A & B syndrome ^) and cancer (e.g. glioblastoma or metastatic cancer in the brain). - Targets for such neurological disorders include:Cell surface target in the brain: TREM2, PILRA, CD33, CR1, ABCA1, ABCA7, MS4A4A, MS4A6A, MS4A4E, HLA- DR5, HLA-DR1, IL1RAP, TREML2, IL-34, SORL1, ADAM17, BACE1, gamma secretase, and Siglec11. - Biological targets on tumor cells: ALK, AXL, CD25, CD44v6, CD46, CD56 (NCAM), CDH6 (cadherin 6), CEACAM 5 (CD66E), EGFR, EGFR viii, ETBR, FGFR (1-4), Folate Receptor alpha, GAL-3BP (galectin binding protein), GD2, GD3, GloboH (globohexasylceramide), gp100, gpNMB, HER2, HER3, HER4, IGFR1, KIT, LIV1A, LRRC15 (leucine rich repeat containing 15), MET , NaPi2B, PDL1, PMEL17, PRAME, PSMA, PTK7 (CCK4; colon carcinoma kinase), RON, ROR1, TF (tissue factor), and TROP2. - Cell surface targets on a hematological cancer cell: B7H3, BCMA, CD125, CD166, CD19, CD20, CD205, CD22, CD25, CD30, CD37, CD39, CD73, and CD79b. - Other biological targets: alpha-synuclein or derivatives or fragments thereof, amyloid-beta peptide or derivatives of fragments thereof, Tau or derivatives or fragments thereof, pTau, huntingtin, transthyretin, or TAR DNA-binding protein 43 (TDP-43) or derivatives or fragments thereof. Apart from antibody based therapies, examples of non-antibody therapeutic proteins include: - Proteins & peptides (e.g. brain-derived neurotorophic factor (BDNF) and glial- derived neurotorophic factor (GDNF), ^) - Enzyme replacement therapies such as iduronate-2-sulfatase, -L- iduronidase, glucocerebrosidase, heparan-N-sulfatase, -N- acetylglucosaminidase, and N-sulfoglucosamine sulfohydrolase. Myelin oligodendrocyte glycoprotein (MOG) is the preferred embodiment of structural protein in the brain. Alternatives include microtubules, actin filaments, neurofilaments, myelin basic protein (MBP), and glial fibrillary acidic protein (GFAP): In preferred embodiments the apparent affinity for the bispecific or trispecific antibodies for the structural protein is below 1 µM, below 750 nm, below 500 nm, below 250 nm, below 100 nm, for example between 1 to 50 nm, 10-40 nm, 15-30 nm, 5 to 25 nm, 5-15 nM or 8 to 12 nm. Specific antibodies and antigen fragments thereof binding to a structural protein in the brain are MOG binding antibodies. An antibody can be a mouse antibody, a rat antibody, a rabbit antibody, an alpaca antibody, a camel antibody, a llama antibody, a chimeric antibody, a humanized antibody, a synthetic antibody, an in silico designed antibody, and a human antibody. Fragment of antibodies are for example a Fab, Fab , F(ab )2, a single chain antibody (scFv), a dimerized V region (diabody), a disulfide-stabilized V region (dsFv), VHH, VH, VL, and, a peptide containing the 3 CDRs. Other binding moieties apart from antibodies and providing alternative scaffolds and antibody mimetics are equally envisaged and include for example Fynomers, dual affinity re-targeting (DART), anticalins, FN3 monobodies, DARPins, Affibodies, Affilins, Affimers, Affitins, Alphabodies, Avimers, Im7, VLR, VNAR, Trimab, CrossMab and TRIDENT. Multispecific antibodies of the present invention can be administered as protein as well as DNA constructs (e.g. via intranasal delivery). In particular embodiments, the antibodies are single domain antibodies. Particular embodiments thereof are the single domain antibodies against MOG depicted below: BBB00498 MOG binding nanobody high affinity [SEQ ID NO:1] QVQLQESGGG LVQSGGSLRL SCAASGFSLD YYAIGWFRQA PGKERERVSC IGSRDGRTFY ADSVKGRFTI SKDHAKNIVY LQMTSLKPED TAVYSCAAYN TAGQRCPSNG YDYDYWGKGT QVTVSS CDR 1 GFSLDYYAIG [SEQ ID NO:2] CDR 2 CIGSRDGRTF [SEQ ID NO:3] CDR 3 AYNTAGQRCPSNGYDYDYWG [SEQ ID NO:4] BBB00500 MOG binding nanobody lower affinity [SEQ ID NO:5] QVQLQESGGG LVQSGGSLRL SCAASGFSLD YYAIGWFRQA PGKERERVSC ISSRDGRTFY ADSVKGRFTI SKDNAKNTVY LQMNSLKPED TAVYSCAAYN TIGQRCPSNG YDYDYWGQGT QVTVSS CDR 1 GFSLDYYAIG [SEQ ID NO:2] CDR 2 CISSRDGRTF [SEQ ID NO:6] CDR 3 AYNTIGQRCPSNGYDYDYWG [SEQ ID NO:7] The present invention further provides an additional set of antibodies (depicted at the end of the application) which can be tested for affinity against MOG, either as single domain antibodies, either as part of bispecific or trispecific antibodies. Different examples of bispecific and multi-specific antibodies are disclosed in WO2024050524 and are the following: Antibodies may be heterodimeric bi- and tri-(or more) specific Ig antibodies and Fc fusion proteins. Exemplary structures include, but are not limited to, IgG, IgM, mono- , di-, tri-, or more scFv-Fcs. For example, bispecific, trispecific, and multispecific formats include, but are not limited to, bispecific and trispecific IgG, IgG-scFv, IgG- dAb, scFv-Fc-scFv, knob-in-hole (KIH)-IgG, -body, KIH0Fc-Fab / scFv, tandem scFv, KIH trispecific, bispecific Fc fusion (N- or C-terminal, with or without KIH). Multispecific antibody molecules can include more than one antigen-binding site, where different sites are specific for different antigens. A multispecific antibody molecule can bind more than one (e.g., two or more) epitopes on the same antigen. Bispecific antibody molecules can be classified into five different structural groups: (i) bispecific immunoglobulin G (BsIgG); (ii) IgG appended with an additional antigen-binding moiety; (iii) bispecific antibody fragments; (iv) bispecific fusion proteins; and (v) bispecific antibody conjugates. BsIgG is a format that is monovalent for each antigen. Exemplary BsIgG formats include but are not limited to crossMab, DAF (two-in-one), DAF (four-in-one), DutaMab, DT-IgG, knobs-in-holes common LC, knobs-in-holes assembly, charge pair, Fab-arm exchange, SEEDbody, triomab, LUZ-Y, Fcab, kappa-lamda-body, orthogonal Fab. BsIgG can include heavy chains that are engineered for heterodimerization. For example, heavy chains can be engineered for heterodimerization using a "knobs-into- holes" strategy, a SEED platform, a common heavy chain (e.g., in Kk-bodies), and use of heterodimeric Fc regions. Strategies that have been used to avoid heavy chain pairing of homodimers in BsIgG include knobs-in-holes, duobody, azymetric, charge pair, HA-TF, SEEDbody, and differential protein A affinity. See Id. BsIgG can be produced by separate expression of the component antibodies in different host cells and subsequent purification / assembly into a BsIgG. BsIgG can also be produced by expression of the component antibodies in a single host cell. BsIgG can be purified using affinity chromatography, e.g., using protein A and sequential pH elution. IgG appended with an additional antigen-binding moiety is another format of bispecific antibody molecules. For example, monospecific IgG can be engineered to have bispecificity by appending an additional antigen-binding unit onto the monospecific IgG, e.g., at the N- or C-terminus of either the heavy or light chain. Exemplary additional antigen-binding units include single domain antibodies (e.g., variable heavy chain or variable light chain), engineered protein scaffolds, and paired antibody variable regions (e.g., single chain variable fragments or variable fragments). Examples of appended IgG formats include dual variable domain (DVD) IgG (DVD- Ig), IgG(H)-scFv, scFv-(H)IgG, IgG(L)-scFv, scFv-(L)IgG, IgG(L,H)-Fv, IgG(H)-V, V(H)-IgG, IgG(L)-V, V(L)-IgG, KIH IgG-scFab, 2scFv-IgG, IgG-2scFv, scFv4-Ig, zybody, and DVI-IgG (four-in-one). Bispecific antibody fragments (BsAb) are a format of bispecific antibody molecules that lack some or all of the antibody constant domains. For example, some BsAb lack an Fc region. Bispecific antibody fragments can include heavy and light chain regions that are connected by a peptide linker that permits efficient expression of the BsAb in a single host cell. Exemplary bispecific antibody fragments include but are not limited to nanobody, nanobody-HAS, BiTE, Diabody, DART, TandAb, scDiabody, scDiabody-CH3, Diabody-CH3, triple body, miniantibody, minibody, TriBi minibody, scFv-CH3 KIH, Fab-scFv, scFv-CH-CL-scFv, F(ab')2, F(ab')2-scFv2, scFv-KIH, Fab- scFv-Fc, tetravalent HCAb, scDiabody-Fc, Diabody-Fc, tandem scFv-Fc, and intrabody. See Id. For example, the BiTE format includes tandem scFvs, where the component scFvs bind to CD3 on T cells and a surface antigen on cancer cells. Bispecific fusion proteins include antibody fragments linked to other proteins, e.g., to add additional specificity and / or functionality. An example of a bispecific fusion protein is an immTAC, which includes an anti-CD3 scFv linked to an affinity-matured T-cell receptor that recognizes HLA-presented peptides. The dock-and-lock (DNL) method can be used to generate bispecific antibody molecules with higher valency. Also, fusions to albumin binding proteins or human serum albumin can be extend the serum half-life of antibody fragments. Chemical conjugation, e.g., chemical conjugation of antibodies and / or antibody fragments, can be used to create BsAb molecules. An exemplary bispecific antibody conjugate includes the CovX-body format, in which a low molecular weight drug is conjugated site-specifically to a single reactive lysine in each Fab arm or an antibody or fragment thereof. The conjugation can improves the serum half-life of the low molecular weight drug. A multispecific molecule can further includes a heavy chain constant region (e.g., an Fc region) chosen from the heavy chain constant regions of IgG1, IgG2, and IgG4, more particularly, the heavy chain constant region of human IgG1, IgG2 or IgG4. The heavy chain constant region (e.g., an Fc region) can optionally be linked to, e.g., covalently linked to, one or both of the ApoL1-containing complex-binding antibody molecule and the second antibody molecule. The heavy chain constant region (e.g., an Fc region) can be altered, e.g., mutated, to increase or decrease one or more of: Fc receptor binding, antibody glycosylation, the number of cysteine residues, effector cell function, or complement function. Also, the interface of a first and second heavy chain constant regions (e.g., Fc region) can be altered, e.g., mutated, to increase or decrease dimerization, e.g., relative to a non-engineered interface. The dimerization of the heavy chain constant region (e.g., Fc region) can be enhanced by providing an Fc interface of a first and a second Fc region with one or more of: a paired cavity-protuberance ("knob-in-a hole"), an electrostatic interaction, or a strand-exchange, such that a greater ratio of heteromultimer:homomultimer forms, e.g., relative to a non-engineered interface. The heavy chain constant region (e.g., Fc region) can include an amino acid substitution at a position chosen from one or more of 347, 349, 350, 351, 366, 368, 370, 392, 394, 395, 397, 398, 399, 405, 407, or 409, e.g., of the Fc region of human IgG1, numbered based on the Eu numbering system. Unless otherwise specified herein, numbering of amino acid residues in the Fc region or constant region is according to the EU numbering system, also called the EU index, as described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md., 1991. The heavy chain constant region (e.g., Fc region) may include an amino acid substitution chosen from: T366S, L368A, or Y407V (e.g., corresponding to a cavity or hole), or T366W (e.g., corresponding to a protuberance or knob), or a combination thereof, numbered based on the Eu numbering system. The heavy chain constant region (e.g., an Fc region) may include one or more mutations that increase or decrease one or more of: Fc receptor binding, antibody glycosylation, the number of cysteine residues, effector cell function, or complement function, relative to a naturally-existing heavy chain constant region. The first and / or second heavy chain constant region (e.g., a first and / or second Fc region, e.g., a first and / or second IgG1 Fc region) can include an amino acid substitution at a position chosen from one or more of 347, 349, 350, 351, 366, 368, 370, 392, 394, 395, 397, 398, 399, 405, 407, or 409, numbered based on the Eu numbering system. The first and / or second heavy chain constant region (e.g., a first and / or second Fc region, e.g., a first and / or second IgG1 Fc region) can include an amino acid substitution chosen from: T366S, L368A, Y407V, or Y349C (e.g., corresponding to a cavity or hole), or T366W or S354C (e.g., corresponding to a protuberance or knob), or a combination thereof, numbered based on the Eu numbering system. The present invention provides a delivery strategy with binding to a brain specific protein, MOG, leading to a drastically improved central half-life of the antibodies. VHHs targeting the brain specific protein MOG were identified. VHHs originate from heavy-chain only antibodies found in camelids and can easily be fused to a wide variety of therapeutics, including antibodies, allowing the design of multispecific antibodies. While species cross-reactivity remains a challenge for certain antibody / VHH targets such as TfR, the identified MOG binding VHHs are cross- reactive for mouse, human and cynomolgus MOG. The optimal MOG affinity for brain retention is not known. Strong binding might lead to more accumulation, but potentially less free antibody concentration in the brain and hence lower efficacy; vice versa, lower affinity could lead to lower accumulation, and hence lower efficacy. Indeed, constructs with an apparent affinity between 19.6 to 36.5 nM showed under the experimental conditions disclosed herein an improved brain accumulation, while an apparent affinity of 573.8 nM seems to lead to lower brain accumulation (Fig. 3C). A full 14-day pharmacokinetics / pharmacodynamics evaluation was performed for biMOGLow:monoTfR:BACE1 and compared to monoTfR:BACE1. While plasma clearance is similar for both antibodies, brain pharmacokinetics significantly improved by binding to MOG. While antibody is undetectable in plasma after 7 days, we could still detect MOG binding antibody in the brain 14 days post treatment. Also the therapeutic effect was prolonged in time, up to 7 days. By eliminating TfR binding, the peripheral half-life of the antibodies is improved but the concentration reaching the brain is too low to elicit a therapeutic effect. Up to now, the short peripheral half- life of TfR mediated BBB crossing antibodies was mostly considered a disadvantage as the brain pharmacokinetics of such antibodies are following the peripheral pharmacokinetics , leading to fast clearance from the brain. In the present invention the antibody brain pharmacokinetics seem to be more disconnected from the peripheral pharmacokinetics as the antibody remains detectable in the brain when it is already completely cleared in plasma. This scenario of short peripheral half-life combined with long central half-life can be an advantage for drugs exerting peripheral side effects. These would be short-term due to the fast peripheral clearance, while central, desired therapeutic effects would last longer. Even more, when exchanging the BACE1 binding Fab for a non-targeting Fab (anti-SarsCov), kinetics of brain exposure improve further with an approximately two-fold higher Cmax and brain antibody levels still around 30 nM 49 days post treatment. This is a huge advancement for the field as, to the best of or knowledge, this is the first time that antibodies are detected in brain 7 weeks after a single antibody injection, even at therapeutically relevant concentrations. In conclusion, the present invention relates to mouse / human / cyno cross-reactive anti-MOG VHHs and their ability to drastically increase brain exposure of antibodies, resulting in both a higher Cmax and a longer brain half-life. Combining MOG and TfR binding leads to distinct pharmacokinetics, biodistribution, and brain exposure, differentiating it from the highly investigated TfR-shuttling. Long brain antibody exposure is demonstrated after one single dose. The antibodies of the present invention enhance brain accumulation of biological agents, overcoming a significant hurdle in the development of new treatments for neurological disorders. To find mouse / human cross-reactive anti-MOG VHHs, a VHH-displaying phage library originating from camelids immunized with human MOG (hMOG) and mouse MOG (mMOG), was subjected to one or two in vitro selection rounds on CHO-mMOG overexpressing cell lines and optional one selection round on recombinant hMOG. Single clones from libraries with promising enrichment factors were sequenced and clustered according to homology. 93 unique clones were selected, expressed and periplasmic extracts were prepared. Screening results revealed 59 potential mMOG binders and potential 54 hMOG binders of which 24 VHHs were mouse / human cross- reactive. One mouse specific binder (BBB00497) and 5 cross-reactive VHHs (BBB00498-BBB00502) were recloned in an expression vector harboring a c-terminal 3xFLAG and a hexahistidine tag. VHHs were expressed in TG1 and purified from the periplasmic extract with Ni-charged Magbeads. In vitro binding characteristics of these six VHHs were determined with flow cytometry on mouse, human and cynomolgus MOG overexpressing cell lines using mouse-anti-FLAG-iFluor647 detection antibody. Full binding curves were generated for the three different species. Except for BBB00497, all VHHs were cross-reactive for mouse, human and cynomolgus MOG (Fig. 2A-C). Most cross-reactive VHHs showed a higher affinity towards hMOG compared to mMOG or cynomolgus MOG (cMOG) (Table 1). Table 1. mMOG / hMOG / cMOG affinities of VHHs and antibodies. EC50 values (nM) of flow cytometry on CHO cells overexpressing mouse MOG, human MOG, and cynomolgus MOG. VHH / Ab Mouse EC50 Human EC50 Cynomolgus EC50 (nM) (nM) (nM) BBB00497 3.1 > 1000 844.5 BBB00498 3.9 3.1 6.8 BBB00499 800.6 2.6 285.5 BBB00500 67.4 2.8 71.9 BBB00501 304.1 3.4 234.2 BBB00502 434.7 3.3 225.3 biMOGHigh 29.6 25.0 46.33 monoMOGHigh 36.5 43.3 184.0 biMOGLow 19.6 20.3 24 monoMOGLow 573.8 50.28 519.6 Two cross-reactive VHHs with a more than 10-fold difference (BBB00498: EC503.9 nM, BBB00500: EC50 67.4 nM) in affinity for mMOG were selected for in vivo evaluation. Nb62, a VHH binding to mouse transferrin receptor (mTfR), delivers a therapeutic anti-BACE1 antibody (1A11) to the CNS and elicit a short-term therapeutic effect by decreasing A 1-40 levels in the brain [Wouters et al. 2022, cited above]. BACE1 is an enzyme involved in the cleavage of amyloid precursor protein (APP) and the formation of A peptides, presenting a therapeutic readout for brain penetration of antibody constructs. To prolong the therapeutic effect, this 1A11-Nb62 antibody- VHH fusion is supplemented with one or two anti-MOG VHHs with different affinities (High: BBB00498, Low: BBB00500) (Fig. 1). Binding of these antibody-VHH fusions to mMOG was evaluated using flow cytometry. Three out of four constructs show a similar affinity ranging from 19.6 to 36.5 nM (monoMOGHigh:monoTfR:BACE1, biMOGLow:monoTfR:BACE1 and biMOGHigh:monoTfR:BACE1), while one of the constructs had a 15-30 fold lower affinity (monoMOGLow:monoTfR:BACE1; 573.8 nM). The affinity of similar multispecific antibody constructs harbouring Nb62 for mTfR was previously determined to be 37 nM. To investigate the effect of anti-MOG VHH affinity and avidity on CNS half-life, 5 different antibodies (monoTfR:BACE1, monoMOGLow:monoTfR:BACE1, monoMOGHigh:monoTfR:BACE1, biMOGLow:monoTfR:BACE1, biMOGHigh:monoTfR:BACE1), were tested in vivo in mice at a dose of 80 nmol / kg. Plasma and brain samples were collected from treated mice one and three days post treatment. Antibody concentrations in plasma were similar for all antibodies and decreased rapidly, as expected given the high TfR expression in multiple peripheral organs, resulting in a peripheral sink for these antibodies (Fig. 3A). A 1-40 levels in plasma decreased significantly for all antibodies compared to PBS treated control mice at both evaluated timepoints (Fig. 3B). Brain pharmacokinetics was similar for all antibodies one day post treatment. However, three days post treatment, antibody concentrations in brain were higher for all MOG binding antibodies compared to the monoTfR:BACE1 control (biMOGHigh: p = < 0.0001, monoMOGHigh: p = 0.0023, biMOGLow: p < 0.0001, monoMOGLow: p = 0.1689). Bivalent MOG binding constructs (biMOGLow:monoTfR:BACE1, biMOGHigh:monoTfR:BACE1) showed the greatest improvement in brain pharmacokinetics (Fig. 3C). A 1-40 levels in brain decreased around 40 % compared to PBS treated control mice for all antibodies one day post treatment and for all MOG binding antibodies three days post treatment. The therapeutic effect diminished already three days post treatment for the monoTfR:BACE1 control (Fig. 3D). 40 % decrease in brain A 1-40 levels is assumed to be the maximum effect possible with antibody 1A11, as the same decrease is observed after direct intracranial injection. Despite the pharmacokinetics differences between the different MOG binding constructs, no significant difference is observed in the therapeutic effect, possibly by a saturation of the pharmacodynamics effect with all brain concentrations achieved except for the noMOG control. As biMOGLow:monoTfR:BACE1 was the best performing antibody construct in terms of brain pharmacokinetics, this antibody was selected for further in vivo evaluation. biMOGLow:monoTfR:BACE1 was compared head-to-head with monoTfR:BACE1 at a dose of 80 nmol / kg and 160 nmol / kg, and a follow-up period of maximum 14 days. Plasma pharmacokinetics and pharmacodynamics were similar for both antibodies and characterized by a fast clearance.7 days post treatment no antibody or decrease in A 1-40 levels could be detected in plasma anymore (Fig. 4B and 4C). However, binding to MOG leads to a significantly increased concentration over the whole two- week time period (p < 0.0001). Even with a lower dose of biMOGLow:monoTfR:BACE1, significantly higher brain concentrations were achieved then with a high dose of monoTfR:BACE1 (p = 0.0001) (Fig.4D). This is translated in an increased therapeutic effect, i.e. significantly lower A 1-40 levels, up to one week post treatment (high dose: p = 0.0208, low dose: p = 0.0413) (Fig. 4E). The increased antibody concentration when binding to MOG was also observed in spinal cord. To avoid the peripheral sink of antibody mediated by TfR binding, MOG binding antibody-based constructs without active shuttling mechanism were generated to evaluate if brain accumulation of antibodies after passive diffusion is sufficient to elicit a therapeutic effect. Peripheral clearance of biMOGLow:monoGFP:BACE1 and monoGFP:BACE1 was reduced compared to the TfR binding counterparts (with monoGFP being a VHH targeting green fluorescent protein (GFP)). Plasma concentrations one day post treatment were four fold higher and antibodies were still detectable in plasma after 14 days (Fig. 5B). Consistently, A 1-40 levels in plasma were decreased at all evaluated timepoints (Fig. 5C). Even without active shuttling there is a significant difference between MOG binding and no MOG binding antibodies. 14 days post treatment the levels of biMOGLow:monoGFP:BACE1 were 5-fold higher then monoGFP:BACE1 (Fig. 5D). However, brain concentrations were too low to elicit any therapeutic effect (Fig. 5E). A similar antibody pharmacokinetics profile for these constructs was observed in spinal cord. Above described results were all based on therapeutic antibodies binding BACE1. However, the target of the therapeutic Fab-arm could potentially influence antibody kinetics, for instance by target mediated clearance. To assess the contribution of BACE1 mediated clearance, the BACE1 binding Fab arm was exchanged for a Fab fragment binding Sars-Covid 2, a target not present in the animal model used. Plasma pharmacokinetics remained unchanged compared to their BACE1 binding equivalents (Fig. 6B). While there was almost no difference in brain pharmacokinetics of monoTfR:BACE1 vs monoTfR:SarsCov, a major improvement was observed for biMOGLow:monoTfR:SarsCov compared to biMOGLow:monoTfR:BACE1 (Fig. 6C). Brain concentrations for both constructs peaked 3 days post treatment but were almost 2- fold higher when eliminating the BACE1 binding. Afterwards, brain antibody concentration decreased to approximately 40 nM at day 21 post treatment, after which a minimal decline was observed up till 49 days post treatment, the final timepoint investigated in this study (Fig. 6C). Also biMOGLow:monoGFP:SarsCov showed increased brain exposure reaching concentrations of 10 nM in the brain (Fig. 6D). Similar results were obtained for spinal cord pharmacokinetics. When comparing brain antibody exposure (area under the curve (AUC)) over all evaluated treatments and timepoints, biMOGLow:monoTfR:SarsCov showed a 4-fold higher exposure then biMOGLow:monoTfR:BACE1, a 15-fold higher exposure then monoTfR:SarsCov, the current state-of-the-art for brain antibody delivery, and a 63-fold higher exposure then monoGFP:BACE1, a negative control antibody. EXAMPLES Example 1. VHH Library Generation Three alpacas were subjected to three rounds of four bi-weekly DNA immunizations (VIB Nanobody Core, Brussels, Belgium), at a one-month interval, using a pool of recombinant pVAX1 plasmids encoding various proteins, including human myelin oligodendrocyte glycoprotein (hMOG). Seven months later, the same animals were immunized with another pool of recombinant pVAX1 plasmid DNA encoding various proteins, including mouse MOG (mMOG). The immunization with the mouse equivalent consisted of two rounds, round one consisting of one and round two of 3 bi-weekly DNA immunizations. DNA solutions were injected intradermally at multiple sites at front and back limbs near the draining lymph nodes followed by electroporation. A total amount of 2 mg of each pool of the plasmid DNA was used per injection per animal. Four and 8 days after the first and the last mMOG immunization, 100 mL anticoagulated blood was collected from each animal for the preparation of peripheral blood lymphocytes (PBLs). The VHH encoding genes were recovered and the phagemid library was cloned as described in Pardon et al. (2014) Nat. Prot. 9, 674-693. In short, total RNA from peripheral blood lymphocytes was used as template for first strand cDNA synthesis with oligodT primers. This cDNA was used to amplify the VHH-encoding open reading frames by PCR, digested with PstI and NotI, and cloned into a phagemid vector. The library was transformed into electro-competent E.coli TG1 cells, which resulted in 108independent transformants, of which at least 85 % contained the vector with a right insert size. Example 2. Cell line generation Stable Chinese hamster ovary (CHO) cell lines, overexpressing mouse, human or cynomolgus MOG (cMOG), were generated with the Flp-InTMCHO system (Invitrogen, Waltham, MA, USA). DNA encoding mouse, human or cynomolgus MOG, tagged with c-terminal hemagglutinin (HA) followed by green fluorescent protein (GFP) under the control of an internal ribosome entry site (IRES), was synthesized and subcloned into the pcDNA5 / FRT mammalian expression vector (Invitrogen, Waltham, MA, USA). This target expressing vector and the Flp-InTMrecombinase vector pOG44, were co- transfected in the Flp-InTMCHO cell line using Xtreme gene HP DNA transfection reagent (Sigma, St. Louis, MO, USA). Cells were cultured in Ham^s F-12 Nutrient Mix medium supplemented with GlutaMAX (Gibco, Thermo Fisher Scientific, Waltham, MA, USA), 10 % fetal bovine serum (FBS) (Gibco, Thermo Fisher Scientific, Waltham, MA, USA) and 100 µg / mL zeocin selection antibiotic (Invivogen, San Diego, CA, USA) before transfection or 700 µg / mL Hygromycin B gold (Invivogen, San Diego, CA, USA) after transfection to select for stably transfected cells. Stable cell lines were amplified and aliquots frozen with 10 % dimethylsulfoxide (DMSO) (Sigma, St. Louis, MO, USA) for further use. Example 3. Isolation of anti-MOG VHHs To select species cross-reactive anti-MOG VHHs, two rounds of cell selections on 5x106mouse MOG (mMOG) expressing cells (in-house cells derived from Flp-InTMCHO Cell Line, Invitrogen, Waltham, MA, USA) were performed followed by one round of in solution selection with 50 nM in-house biotinylated human MOG (hMOG) (8535- MO, R&D Systems, Minneapolis, MN, USA). Thereafter, the library was subcloned into an expression vector (pBDS100, a modified pHEN6 vector with an OmpA signal peptide and a C-terminal 3xFlag / 6xHis tag). The expression library was used to transform TG1 E.coli after which VHHs were expressed from single colonies. These VHHs were screened for direct binding to the biotinylated hMOG using the AlphaScreen FLAG (M2) Detection Kit (6760613, PerkinElmer, Waltham, MA, USA) and to the mMOG overexpressing CHO cells using flow cytometry with mouse-anti- FLAG-iFluor647 Ab (A01811, Genscript, Piscataway, NJ, USA). The cross-reactive hits were sequenced and clustered according to sequence homology. One representative of each cluster was expressed following the protocol by Pardon et al. cited above and purified using Ni-Charged MagBeads and an AmMag SA Plus System (Genscript, Piscataway, NJ, USA). Example 4. Monoclonal antibodies The control antibody consisted of either the anti^BACE1 Fab (1A11) or an anti- SarsCov2 Fab, and Nb62 binding to TfR [Wouters et al.2020, cited above] or an anti- GFP VHH, fused to a human IgG1 Fc domain with LALA-PG effector and complement knock-out mutations (L234A, L235A, P329G) and knob-into-hole mutations (T350V, T366L, K392L, T394W / T350V, L351Y, F405A, Y407V) to generate heterodimeric antibodies (monoTfR:BACE1 and monoTfR:SarsCov). MOG binding antibodies were created by genetically fusing these controls with either one (monovalent) or two (bivalent) MOG binding VHHs with different affinities (BBB00498 and BBB00500) at the c-terminal side of the Fc (Fig. 1). Encoding genes were ordered at Twist Bioscience (San Francisco, CA, USA) in pTwist CMV BG WPRE Neo vector (Twist Bioscience, San Francisco, CA, USA). Antibodies were expressed in expiCHO-STMcells (Thermo Fisher Scientific, Waltham, MA, USA) using CHOgro high yield expression system (MIR6270, Mirus Bio, Madison, WI, USA) according to the manufacturer^s protocol and purified following the protocol by Nesspor et al. (2020) Sci Rep 10, 7557. The purification protocol consisted of protein A purification using AmMagTMProtein A Magnetic Beads (Genscript, Piscataway, NJ, USA), followed by purification over a CaptureSelect^ CH1-XL Pre-packed Column (Thermo Fischer Scientific, Waltham, MA, USA) and size exclusion chromatography on a Superdex 200 Increase 10 / 300 GL (Cytiva, Marblorough, MA, USA). All different antibody constructs were given dedicated numbers: monoTfR:BACE1 [BBB00611], biMOGhigh:monoTfR:BACE1 [BBB00609], monoMOGhigh:monoTfR :BACE1 [BBB00610], biMOGlow:monoTfR:BACE1 [BBB00623], and monoMOGlow:monoTfR:BACE1 [BBB00624], monoGFP:BACE1, biMOGLow: monoGFP: BACE1, monoTfR:SarsCov, biMOGLow:monoTfR:SarsCov, monoGFP:SarsCov, biMOGLow:monoGFP:SarsCov. (Figure 1) The sequences of antibodies of figure 1 are detailed below: Protein building blocks: BBB00498 MOG binding nanobody high affinity [SEQ ID NO:1] QVQLQESGGG LVQSGGSLRL SCAASGFSLD YYAIGWFRQA PGKERERVSC IGSRDGRTFY ADSVKGRFTI SKDHAKNIVY LQMTSLKPED TAVYSCAAYN TAGQRCPSNG YDYDYWGKGT QVTVSS CDR 1 SEQ ID NO:2 GFSLDYYAIG CDR 2 SEQ ID NO:3 CIGSRDGRTF CDR 3 SEQ ID NO:4 AYNTAGQRCPSNGYDYDYWG BBB00500 MOG binding nanobody lower affinity [SEQ ID NO:5] QVQLQESGGG LVQSGGSLRL SCAASGFSLD YYAIGWFRQA PGKERERVSC ISSRDGRTFY ADSVKGRFTI SKDNAKNTVY LQMNSLKPED TAVYSCAAYN TIGQRCPSNG YDYDYWGQGT QVTVSS CDR 1 GFSLDYYAIG [SEQ ID NO:2] CDR 2 CISSRDGRTF [SEQ ID NO:6] CDR 3 AYNTIGQRCPSNGYDYDYWG [SEQ ID NO:7] BBB00062 Transferrin Receptor binding nanobody [SEQ ID NO:8] QVQLQESGGG LVQAGDSLRL SCATSGRTFS SYNMGWFRQA PGKEREFVSV INWSGGITSY AESVEGRFTI SRDNAKNTVY LQMNSLKPED TAVYYCAAGR RSLVRGDEYD YWGQGTQVTV SS Constant Domain Kappa Light Chain [SEQ ID NO:9] RTVAAPSVFI FPPSDEQLKS GTASVVCLLN NFYPREAKVQ WKVDNALQSG NSQESVTEQD SKDSTYSLSS TLTLSKADYE KHKVYACEVT HQGLSSPVTK SFNRGEC** Human IgG1 backbone CH1-CH2a-CH3a [SEQ ID NO:10] ASTKGPSVFP LAPSSKSTSG GTAALGCLVK DYFPEPVTVS WNSGALTSGV HTFPAVLQSS GLYSLSSVVT VPSSSLGTQT YICNVNHKPS NTKVDKKVEP KSCDKTHTCP PCPAPEAAGG PSVFLFPPKP KDTLMISRTP EVTCVVVDVS HEDPEVKFNW YVDGVEVHNA KTKPREEQYN STYRVVSVLT VLHQDWLNGK EYKCKVSNKA LGAPIEKTIS KAKGQPREPQ VYVYPPSRDE LTKNQVSLTC LVKGFYPSDI AVEWESNGQP ENNYKTTPPV LDSDGSFALV SKLTVDKSRW QQGNVFSCSV MHEALHNHYT QKSLSLSPGK Human IgG1 backbone CH2b-CH3b [SEQ ID NO:11] EPKSCDKTHT CPPCPAPEAA GGPSVFLFPP KPKDTLMISR TPEVTCVVVD VSHEDPEVKF NWYVDGVEVH NAKTKPREEQ YNSTYRVVSV LTVLHQDWLN GKEYKCKVSN KALGAPIEKT ISKAKGQPRE PQVYVLPPSR DELTKNQVSL LCLVKGFYPS DIAVEWESNG QPENNYLTWP PVLDSDGSFF LYSKLTVDKS RWQQGNVFSC SVMHEALHNH YTQKSLSLSP GK Linkers GGGGSGGGGS [SEQ ID NO:12] GGGGSGGGGSGGGGSGGGGS[SEQ ID NO:13]

[0002] monoTfR:BACE1 [BBB00611] (1A11-mNb62) Polypeptide 1:Variable Domain Light Chain binding to BACE 1 (therapeutic target) fused to Constant Domain Kappa Light Chain [Variable Domain Light Chain binding to BACE 1]- RTVAAPSVFI FPPSDEQLKS GTASVVCLLN NFYPREAKVQ WKVDNALQSG NSQESVTEQD SKDSTYSLSS TLTLSKADYE KHKVYACEVT HQGLSSPVTK SFNRGEC [SEQ ID NO:14] Polypeptide 2: variable Domain HC binding to BACE1 (therapeutic target) + Human IgG1 backbone CH1-CH2a-CH3a [Variable Domain Heavy Chain binding to BACE 1]- ASTKGPSVFP LAPSSKSTSG GTAALGCLVK DYFPEPVTVS WNSGALTSGV HTFPAVLQSS GLYSLSSVVT VPSSSLGTQT YICNVNHKPS NTKVDKKVEP KSCDKTHTCP PCPAPEAAGG PSVFLFPPKP KDTLMISRTP EVTCVVVDVS HEDPEVKFNW YVDGVEVHNA KTKPREEQYN STYRVVSVLT VLHQDWLNGK EYKCKVSNKA LGAPIEKTIS KAKGQPREPQ VYVYPPSRDE LTKNQVSLTC LVKGFYPSDI AVEWESNGQP ENNYKTTPPV LDSDGSFALV SKLTVDKSRW QQGNVFSCSV MHEALHNHYT QKSLSLSPGK [SEQ ID NO:15] Polypeptide 3: Transferrin Receptor binding nanobody + linker + Human IgG1 backbone CH2b-CH3b QVQLQESGGG LVQAGDSLRL SCATSGRTFS SYNMGWFRQA PGKEREFVSV INWSGGITSY AESVEGRFTI SRDNAKNTVY LQMNSLKPED TAVYYCAAGR RSLVRGDEYD YWGQGTQVTV SSGGGGSGGG GSEPKSCDKT HTCPPCPAPE AAGGPSVFLF PPKPKDTLMI SRTPEVTCVV VDVSHEDPEV KFNWYVDGVE VHNAKTKPRE EQYNSTYRVV SVLTVLHQDW LNGKEYKCKV SNKALGAPIE KTISKAKGQP REPQVYVLPP SRDELTKNQV SLLCLVKGFY PSDIAVEWES NGQPENNYLT WPPVLDSDGS FFLYSKLTVD KSRWQQGNVF SCSVMHEALH NHYTQKSLSL SPGK [SEQ ID NO:16] biMOGhigh:monoTfR:BACE1 [BBB00609] (1A11-mNb62-bMOG (High Affinity)) Polypeptide 1 : as in MonoTfR:BACE1 [BBB00611] Polypeptide 2 : Variable Domain HC binding to BACE1 + Human IgG1 backbone CH1-CH2a-CH3a + linker + MOG binding nanobody high affinity [Variable Domain heavy Chain binding to BACE 1] - ASTKGPSVFP LAPSSKSTSG GTAALGCLVK DYFPEPVTVS WNSGALTSGV HTFPAVLQSS GLYSLSSVVT VPSSSLGTQT YICNVNHKPS NTKVDKKVEP KSCDKTHTCP PCPAPEAAGG PSVFLFPPKP KDTLMISRTP EVTCVVVDVS HEDPEVKFNW YVDGVEVHNA KTKPREEQYN STYRVVSVLT VLHQDWLNGK EYKCKVSNKA LGAPIEKTIS KAKGQPREPQ VYVYPPSRDE LTKNQVSLTC LVKGFYPSDI AVEWESNGQP ENNYKTTPPV LDSDGSFALV SKLTVDKSRW QQGNVFSCSV MHEALHNHYT QKSLSLSPGK GGGGSGGGGS GGGGSGGGGS QVQLQESGGG LVQSGGSLRL SCAASGFSLD YYAIGWFRQA PGKERERVSC IGSRDGRTFY ADSVKGRFTI SKDHAKNIVY LQMTSLKPED TAVYSCAAYN TAGQRCPSNG YDYDYWGKGT QVTVSS** [SEQ ID NO:17] Polypeptide 3: Transferrin Receptor binding nanobody + linker + Human IgG1 backbone CH2b-CH3b + linker + MOG binding nanobody high affinity QVQLQESGGG LVQAGDSLRL SCATSGRTFS SYNMGWFRQA PGKEREFVSV INWSGGITSY AESVEGRFTI SRDNAKNTVY LQMNSLKPED TAVYYCAAGR RSLVRGDEYD YWGQGTQVTV SSGGGGSGGG GSEPKSCDKT HTCPPCPAPE AAGGPSVFLF PPKPKDTLMI SRTPEVTCVV VDVSHEDPEV KFNWYVDGVE VHNAKTKPRE EQYNSTYRVV SVLTVLHQDW LNGKEYKCKV SNKALGAPIE KTISKAKGQP REPQVYVLPP SRDELTKNQV SLLCLVKGFY PSDIAVEWES NGQPENNYLT WPPVLDSDGS FFLYSKLTVD KSRWQQGNVF SCSVMHEALH NHYTQKSLSL SPGKGGGGSG GGGSGGGGSG GGGSQVQLQE SGGGLVQSGG SLRLSCAASG FSLDYYAIGW FRQAPGKERE RVSCIGSRDG RTFYADSVKG RFTISKDHAK NIVYLQMTSL KPEDTAVYSC AAYNTAGQRC PSNGYDYDYW GKGTQVTVSS** [SEQ ID NO:18] monoMOGhigh:monoTfR:BACE1 [BBB00610] (1A11-mNb62-mMOG (High Affinity)) Polypeptide 1 : as in monoTfR:BACE1 [BBB00611] Polypeptide 2 : as in biMOGhigh:monoTfR:BACE1 [BBB00609] Polypeptide 3 : as in monoTfR:BACE1 [BBB00611] biMOGlow:monoTfR:BACE1 [BBB00623] (1A11-mNb62-bMOG (Low Affinity)) Polypeptide 1 as in monoTfR:BACE1 [BBB00611] Polypeptide 2 : Variable Domain HC binding to BACE1 + Human IgG1 backbone CH1-CH2a-CH3a + linker + MOG binding nanobody lower affinity [Variable Domain heavy Chain binding to BACE 1] - ASTKGPSVFP LAPSSKSTSG GTAALGCLVK DYFPEPVTVS WNSGALTSGV HTFPAVLQSS GLYSLSSVVT VPSSSLGTQT YICNVNHKPS NTKVDKKVEP KSCDKTHTCP PCPAPEAAGG PSVFLFPPKP KDTLMISRTP EVTCVVVDVS HEDPEVKFNW YVDGVEVHNA KTKPREEQYN STYRVVSVLT VLHQDWLNGK EYKCKVSNKA LGAPIEKTIS KAKGQPREPQ VYVYPPSRDE LTKNQVSLTC LVKGFYPSDI AVEWESNGQP ENNYKTTPPV LDSDGSFALV SKLTVDKSRW QQGNVFSCSV MHEALHNHYT QKSLSLSPGK GGGGSGGGGS GGGGSGGGGS QVQLQESGGG LVQSGGSLRL SCAASGFSLD YYAIGWFRQA PGKERERVSC ISSRDGRTFY ADSVKGRFTI SKDNAKNTVY LQMNSLKPED TAVYSCAAYN TIGQRCPSNG YDYDYWGQGT QVTVSS** [SEQ ID NO:19 Polypeptide 3 : Transferrin Receptor binding nanobody + linker + Human IgG1 backbone CH2b-CH3b + linker + MOG binding nanobody lower affinity QVQLQESGGG LVQAGDSLRL SCATSGRTFS SYNMGWFRQA PGKEREFVSV INWSGGITSY AESVEGRFTI SRDNAKNTVY LQMNSLKPED TAVYYCAAGR RSLVRGDEYD YWGQGTQVTV SSGGGGSGGG GSEPKSCDKT HTCPPCPAPE AAGGPSVFLF PPKPKDTLMI SRTPEVTCVV VDVSHEDPEV KFNWYVDGVE VHNAKTKPRE EQYNSTYRVV SVLTVLHQDW LNGKEYKCKV SNKALGAPIE KTISKAKGQP REPQVYVLPP SRDELTKNQV SLLCLVKGFY PSDIAVEWES NGQPENNYLT WPPVLDSDGS FFLYSKLTVD KSRWQQGNVF SCSVMHEALH NHYTQKSLSL SPGKGGGGSG GGGSGGGGSG GGGSQVQLQE SGGGLVQSGG SLRLSCAASG FSLDYYAIGW FRQAPGKERE RVSCISSRDG RTFYADSVKG RFTISKDNAK NTVYLQMNSL KPEDTAVYSC AAYNTIGQRC PSNGYDYDYW GQGTQVTVSS** [SEQ ID NO:18] monoMOGlow:monoTfR:BACE1 [BBB00624] (1A11-mNb62-mMOG (Low Affinity)) Polypeptide 1 : as in monoTfR:BACE1 [BBB00611] Polypeptide 2 : as in biMOGlow:monoTfR:BACE1 [BBB00623] Polypeptide 3 : as in monoTfR:BACE1 [BBB00611] Example 5. In vitro binding and affinity determination Binding of purified VHHs or antibodies to m / h / cMOG expressing cells was evaluated by flow cytometry. 105MOG / GFP expressing CHO cells were incubated, 30 min on ice, with the purified proteins in a concentration range between 0.05 and 5000 nM. Cells were washed twice with PBS containing 2 % FBS (Gibco, Thermo Fisher Scientific, Waltham, MA, USA). After washing, cells were stained, 30 min on ice in the dark, with mouse-anti-FLAG-iFluor647 (1:500) (Genscript, Piscataway, NJ, USA) or anti-human IgG Fc Alexa Fluor 647 (1:200) (Thermo Fisher Scientific, Waltham, MA, USA), for VHHs or antibodies respectively, and eBioscience Fixable Viability Dye eFluor 780 (1:2000) (Invitrogen, Waltham, MA, USA) in staining buffer (PBS + 2 % FBS), for protein detection and live / dead staining respectively. After staining, cells were washed with staining buffer, followed by fixation with 4 % paraformaldehyde (PFA) (Alfa Aesar, Haverhill, MA, USA) (15 min, RT, protected from light), and resuspension in staining buffer. Cells were kept at 4 °C before analysis. Attune NxT Flow Cytometer (Thermo Fisher Scientific, Waltham, MA, USA) was used for sample analysis. CHO-GFP expressing cells were used for background correction (aspecific binding of VHHs to cells or fluorescence spillover in other detectors). Data analysis was performed using FCS Express 7 software (De Novo Software, Pasadena, CA, USA). Example 6. Animals Male and female Balb / c mice (Balb / cAnNRj), 7-8 weeks old, with an approximate weight of 18^24 g were purchased at Janvier (Janvier, Le Genest-Saint-Isle, France). All animal experiments were approved by the KU Leuven Animal Ethics Committee (project P091 / 2022) following governmental and EU guidelines. Mice were housed under a 12-h light / dark cycle with water and standard rodent diet ad libitum. Example 7. Intravenous injection Intravenous injections of antibodies were performed by putting the mice in a heating chamber at 40 °C for 10 minutes. Afterwards, mice were put in a restrainer and mAbs were injected in the tail vein at volumes between 45 and 120 µL. Example 8. Plasma / brain sampling Blood collection was performed by heart puncture at time of brain collection in K3 EDTA coated tubes (Sarstedt, Nümbrecht, Germany). Blood was processed to plasma by centrifugation (10 min, 2000g, 4 °C and 10 min, 16 000g, 4 °C), and stored at 20 °C. Mice were euthanized with a Dolethal® (pentobarbital) (Vetoquinol, Niel, Belgium) overdose (150^200 mg / kg) injected peritoneally. Brains were harvested after transcardial perfusion with heparinized (Leo Pharma, Ballerup, Denmark) PBS, snap frozen by submerging collection tubes in liquid nitrogen and stored at -80 °C until further processing. Example 9. A 1-40 detection using MSD ELISA A brain hemisphere was homogenized in buffer containing 0.4 % diethylamine (Sigma, St. Louis, MO, USA) and 50 mM NaCl (Fisher Scientific, Waltham, MA, USA), supplemented with cOmplete^ protease inhibitor cocktail EDTA-free (Roche Diagnostics, Mannheim, Germany) using Ceramic Bead Lysing Matrix D 1.4 mm (MP Biomedicals, Irvine, CA, USA) and a FastPrep-24 classic homogenizer (MP Biomedicals, Irvine, CA, USA)(6 m / s, 3x 20 seconds, 5 min cooling on ice between different cycles). Homogenate was subjected to high speed centrifugation (100000g, 1 h, 4 °C) and supernatant was neutralized with 0.5M Tris-HCl (pH 6.8) (Sigma, St.Louis, MO, USA). A 1-40 levels in brain and plasma samples were quantified by ELISA using Meso Scale Discovery (MSD, Meso Scale Diagnostics, Rockville, MD, USA) 96-well plates and in-house produced antibodies. MAb LTDA_A 40, which recognizes the C-terminus of A 1-40, was used as a capture antibody and LTDA_hA N, a rodent sulfoTAG-fused antibody recognizing the N-terminus as the detection antibody. MSD plates were coated overnight (ON) at 4 °C with mAb LTDA_A 40 (1.5 µg / mL in PBS, 50 µL / well). Next day, plates were washed 3x with PBS-T (0.05 % Tween 20, Sigma, St. Louis, MO, USA) and blocked with PBS 0.1 % casein (Sigma, St. Louis, MO, USA) (150 µL / well) for 4 hours at RT, shaking 700 rpm. Plates were washed 3x with PBS- T and incubated with samples (50 µL / well) ON at 4 °C. Sample mix consisted of plasma or 1:2 diluted brain homogenate in PBS 0.1 % casein, diluted 1:2 with LTDA_hA N sulfoTAG (1:2000 in PBS / casein), pre-incubated 15 min at RT, shaking 700 rpm. Serial two-fold dilutions of rodent A 40, with concentrations ranging between 2.4 and 2500 pg / mL, were used as calibration curve. Next day, plates were washed 3x with PBS-T and signal measured with MSD plate reader after adding 150 µL of READ buffer (R92TC-1 diluted 1:2 in water, Meso Scale Diagnostics, Rockville, MD, USA). Sample concentrations were calculated based on the A 40 calibration curve using a nonlinear fit, Log(agonist) vs response ^ variable slope (4 parameters) (GraphPad Prism 10, GraphPad Software, San Diego, CA, USA). Example 10. Antibody detection in brain and plasma The antibody levels in plasma and brain homogenate were quantified using an in- house developed ELISA. A brain hemisphere was homogenized with 9 volumes of PBS containing 1 % Igepal (Alfa Aesar, Haverhill, Massachusetts, USA), supplemented with cOmplete^ protease inhibitor cocktail EDTA-free (Roche Diagnostics, Mannheim, Germany) using Ceramic Bead Lysing Matrix D 1.4 mm (MP Biomedicals, Irvine, CA, USA) and a FastPrep-24 classic homogenizer (MP Biomedicals, Irvine, CA, USA) (6.5 m / s, 2x45 seconds, 1 min cooling on ice between different cycles). After a 45 min incubation on ice, homogenate was centrifuged (16000g, 10 min, 4°C) and supernatant collected. This extraction procedure was repeated two additional times with the obtained pellet. Plates were coated overnight at 4 °C with 2 µg / mL mouse anti-human IgG Fc (Genscript, Piscataway, NJ, USA) in PBS. Plates were blocked with PBS 0.1 % casein (Sigma, St. Louis, Missouri, USA) for two hours at room temperature (RT). Then, samples were diluted in PBS 0.1 % casein, added to the plate and incubated for two hours at RT under shaking conditions (300 rpm). Serial two-fold dilutions of the purified antibody constructs in corresponding matrix (diluted plasma or brain homogenate), with concentrations ranging between 16 and 0,125 ng / mL, were used as calibration curves. Detection of the captured human IgGs was performed with mouse anti-human IgG Fab-HRP (Genscript, Piscataway, NJ, USA) (1:5000 dilution in PBS 0.1 % casein). Each incubation step, except blocking, was preceded by a washing step with PBS 0.002 % Tween 80. Plates were developed for 30 minutes using 1-step Ultra TMB-ELISA substrate solution (Thermo Fisher Scientific, Waltham, MA, USA). The reaction was stopped with 2 M H2SO4. Absorption was measured at 450 nm using an ELx808 Absorbance Microplate Reader (BioTek Instruments, Bad Friedrichshall, Germany). IgG concentrations in the samples were calculated based on the human IgG calibration curve using a linear regression fit (GraphPad Prism 10, GraphPad Software, San Diego, California, USA). Example 11. Statistical Analysis Statistical analysis was performed using GraphPad Prism 10 (GraphPad Software, San Diego, CA, USA). Parametric methods were used for analysis as data distribution was sufficiently normal for all experiments. Two-way ANOVAs (variables time and treatment) were performed. The reported F statistic and associated p-value were derived on the interaction term for comparison of PK / PD curves or on Dunnett^s multiple comparisons test for comparing single timepoint data to a control group.

[0003] EXAMPLE 15. Additional single domains against MOG mBBBMP062D07 [SEQ ID NO:20] QVQLQESGGGLVQPGGSLRLSCAASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTV SRDNAKNTLYLEMNSLKVEDTAVYYCAKGLRPFRYCSGYGAGCYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGAGCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:88 mBBBMP052B07 [SEQ ID NO:21] QVQLQESGGGLVQPGGSLRLSCTASGFAFRDYWMYWVRQAPGKGLEWVSTLTTGGEDPYYSDSVKGRFTV SRDNAKNTLYLEMNSLKVEDTAVYYCAKGLRPFRYCSGYGAGCYDTHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TLTTGGEDPY CDR3: GLRPFRYCSGYGAGCYDTHP SEQ ID NO:86 SEQ ID NO:89 SEQ ID NO:90 mBBBMP063A12[SEQ ID NO:22] QVQLQESGGGSVQPGGSLRLSCAASGFAFEDYWMYWVRQAPGKGLEWVSTLTTGGEDPYYSDSVKGRFTV SRDNAKNTLYLEMNSLKVEDTAVYYCAKGLRPFRYCSGYGAGCYDTHPPGPGTQVTVSS CDR1: GFAFEDYWMY CDR2: TLTTGGEDPY CDR3: GLRPFRYCSGYGAGCYDTHP SEQ ID NO:91 SEQ ID NO:89 SEQ ID NO:90 mBBBMP052B11[SEQ ID NO:2] QVQLQESGGGLVQPGGSLRLSCATSGFAFKDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKNTLYLEMNSLRVEDTAVYYCAKGLRPFRYCSGYGAGCYDAHPPGPGTQVTVSS CDR1: GFAFKDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGAGCYDAHP SEQ ID NO:92 SEQ ID NO:87 SEQ ID NO:88 mBBBMP061D07[SEQ ID NO:2824] QVQLQESGGGSVQPGGSLRLSCAASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKNTLYLEMNSLRVEDTAVYYCAKGLRPFRYCSGYGAGCYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGAGCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:88 mBBBMP061E05[SEQ ID NO:25] QVQLQESGGGLVQPGGSLRLSCVASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKNTLYLEMDSLKIDDTAVYYCAKGLRPFRYCSGYGSGSQCYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSQCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:93 mBBBMP062E01[SEQ ID NO:26] QVQLQESGGGLVQAGGSLRLSCAASGFAFTDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTV SRDNAKNTLYLEMNSLSAEDTAVYYCVKGLRPFRYCSGYGSGSGCYDAHPPGPGTQVTVSS CDR1: GFAFTDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:94 SEQ ID NO:87 SEQ ID NO:95 mBBBMP052A04[SEQ ID NO:17] QVQLQESGGGLVEAGGSLRLSCAAS GFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTISRDNAKNTLYLEMNSLKAEDTAVYY CAKGLRPFRYCSGYGSGSGCYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:95 mBBBMP052E11[SEQ ID NO:28] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDFWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKNTLYLEMNSLKAEDTAVYYCAKGLRPFRYCSGYGSGSGCYDAHPPGPGTQVTVSS CDR1:GFAFSDFWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:96 SEQ ID NO:87 SEQ ID NO:95 mBBBMP052C07[SEQ ID NO:29] QVQLQESGGGLVQPGGSLRLSCAASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKNTLYLEMNSLKDEDTAVYYCAKGLRPFRYCSGYGSGSACYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSACYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:97 mBBBMP052D08[SEQ ID NO:30] QVQLQESGGGLVQPGGSLRLSCAAS GFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTISRDNAKNTLYLEMNSLKAEDTAVYY CAKGLRPFRYCSGYGSGSGCYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:95 mBBBMP061G06[SEQ ID NO:31] QVQLQESGGGLVQPGGSLRLSCAASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKSTLYLEMNSLKAEDTAVYYCAKGLRPFRYCSGYGSGSGCYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:95 mBBBMP052A07[SEQ ID NO:32] QVQLQESGGGLVQPGGSLRLSCVASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKNTLYLEMNSLKVEDTAVYYCAKGLRPFRYCSGYGSGSGCYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:95 mBBBMP052A05[SEQ ID NO:33] QVQLQESGGGLVQPGGSLRLSCAASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKNTLYLEMNSLKVEDTAVYYCAKGLRPFRYCSGYGSGSGCYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:95 mBBBMP052G06[SEQ ID NO:34] QVQLQESGGGLVQPGGSLRLSCAASGFAFTDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKNTLYLEMNSLKAEDTAVYYCAKGLRPFRYCSGYGSGCYDAHPPGPGTQVTVSS CDR1: GFAFTDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGCYDAHP SEQ ID NO:94 SEQ ID NO:87 SEQ ID NO:98 mBBBMP062E11[SEQ ID NO:35] QVQLQESGGGLVQPGGSLRLSCAASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKKTLYLEMNSLKAEDSAVYYCAKGLRPFRYCSGYGTGSGCYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGTGSGCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:99 mBBBMP052E06[SEQ ID NO:36] QVQLQESGGGLVQPGGSLRLSCVASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGDSPYYADSVKGRFTI SRDNAKNTLYLEMNSLKAEDTAVYYCAKGLRPFRYCSGYGSGSGCYDAHPPGQGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGDSPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:86 SEQ ID NO:100 SEQ ID NO:95 mBBBMP052H07[SEQ ID NO:37] QVQLQESGGGLVQPGGSLRLSCAASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFTI SRDNAKNTLYLEMNSLKVEDTAVYYCAKGLRPFRYCSGYGSGSGCYDAHPPGQGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGDDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:86 SEQ ID NO:101 SEQ ID NO:95 mBBBMP052C10[SEQ ID NO:38] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDSPYYADSVKGRFTI SRDNAKNTLYLEMNSLKAEDTAVYYCAKGLRPFRYCSGYGSGSGCYDAHPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDSPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:103 SEQ ID NO:100 SEQ ID NO:95 mBBBMP052G03[SEQ ID NO:39] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFTI SRDNAKNTVYLEMNSLKAEDTAVYYCTKGLRPFRYCSGYGSGSGCYDAHPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:95 mBBBMP061C12[SEQ ID NO:40] QVQLQESGGGSVQPGGSLRLSCAASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDAVKGRFTI SRDNAKNTLHLEMNSLKPEDTAVYYCAKGLRPFRYCSGYGSGSGCYDAHPVGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:95 mBBBMP061F02[SEQ ID NO:41] QVQLQESGGGLVQPGGSLRLSCAASGFAFRDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKKTLYLQMDNLKLEDTAVYFCAKGLRPFRYCSGYGSGSGCYDAHPPGPGTQVTVSS CDR1: GFAFRDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:86 SEQ ID NO:87 SEQ ID NO:95 mBBBMP061A01[SEQ ID NO:42] QVQLQESGGGLAQAGDSLRLSCAASGFAFSDYWMYWVRQIPGKGLEWVSTITTGGEDPYYADSVQGRFTI SRDNAKNTLYLEMNSLKVEDTAVYYCAKGLRPFRYCSGYGSGSGCYDAHPPGPGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSGSGCYDAHP SEQ ID NO:103 SEQ ID NO:87 SEQ ID NO:95 mBBBMP052A11[SEQ ID NO:43] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDDPYYSDSVKGRFTI SRDNAKNTLYLDMNSLKTEDSAVYYCAKGSRPFSYCRGSGCYGARPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GSRPFSYCRGSGCYGARP SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:104 mBBBMP052A03[SEQ ID NO:44] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDSPYYADSVKGRFTI SRDNAKNTLYLEMNSLKAEDTAVYYCAKGLRPFSYCRGYGCYGARPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDSPY CDR3: GLRPFSYCRGYGCYGARP SEQ ID NO:103 SEQ ID NO:100 SEQ ID NO:105 mBBBMP052E12[SEQ ID NO:45] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFTI SRDNAKNTLYLEMNSLKAEDTAVYYCAKGLRPFSYCRGSGCYGARLPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GLRPFSYCRGSGCYGARL SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:106 mBBBMP052D04[SEQ ID NO:46] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFTI SRDNAKNTLYLEMNSLKAEDTAVYYCAKGLRPFSYCRGSGCYGARPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GLRPFSYCRGSGCYGARP SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:107 mBBBMP052F09[SEQ ID NO:47] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFTI SRDNDKNTLYLEMNSLKAEDTAVYYCAKGLRPFSYCRGSGCYGARPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GLRPFSYCRGSGCYGARP SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:107 mBBBMP052C08[SEQ ID NO:48] QVQLQESGGGLVQAGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFTI SRDNDKNTLYLEMNSLKAEDTAVYYCAKGLRPFSYCRGSGCYGARPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GLRPFSYCRGSGCYGARP SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:107 mBBBMP052D12[SEQ ID NO:49] QVQLQESGGGSVQPGGSLRLSCTASGFVFSDYWMYWIRQAPGKGLEWVSTITTGGEDPYYADAVKGRFTI SRDNAKNTMYLEMNSLKSEDTAVYYCAKGLRPFSYCRGVGCYGARPPGQGTQVTVSS CDR1: GFVFSDYWMY CDR2: TITTGGEDPY CDR3: GLRPFSYCRGVGCYGARP SEQ ID NO:108 SEQ ID NO:87 SEQ ID NO:109 mBBBMP052B06[SEQ ID NO:50] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFII SRDSAKNTVYLEMNNLKAEDTAVYYCAKGLRPFSYCLGTGCYGARPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GLRPFSYCLGTGCYGARP SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:110 mBBBMP052B12[SEQ ID NO:51] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFII SRDNAKNTVYLDLNNLKTEDTAVYYCAKGLRPFSYCLGTGCYGARPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GLRPFSYCLGTGCYGARP SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:111 mBBBMP052G07[SEQ ID NO:52] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRRAPGKGLEWVSTITTGGDDPYYADSVKERFTI SRDNAKNTLYLEMNNLKTEDTAVYYCAKGSRPFSYCRGTGCYGAPPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GSRPFSYCRGTGCYGAPP SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:112 mBBBMP052A02[SEQ ID NO:53] QVQLQESGGGSVQPGGSLRLSCAASGFAFEDYWMYWVRQIPGKGLEWVSTITTGGEDPYYADSVQGRFTI SRDNAKNTLYLEMRGLKPEDTAVYYCTKGLQPFRYCSGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFEDYWMY CDR2: TITTGGEDPY CDR3: GLQPFRYCSGYGSRPGCYDARL SEQ ID NO:91 SEQ ID NO:87 SEQ ID NO:113 mBBBMP064F03[SEQ ID NO:54] QVQLQESGGGSVQPGGSLRLSCAASGFAFEDYWMYWVRQIPGKGLEWVSTITTGGEDPYYADSVQGRFTI SRDNAKNTLYLEMRGLKSEDTAVYYCTKGLQPFRYCSGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFEDYWMY CDR2: TITTGGEDPY CDR3: GLQPFRYCSGYGSRPGCYDARL SEQ ID NO:91 SEQ ID NO:87 SEQ ID NO:113 mBBBMP061A07[SEQ ID NO:55] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGEDPYYADSVQGRFTI SRDNAKNTLYLEMRGLKPEDTAVYYCTKGLQPFRYCSGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGEDPY CDR3: GLQPFRYCSGYGSRPGCYDARL SEQ ID NO:103 SEQ ID NO:87 SEQ ID NO:113 mBBBMP052E08[SEQ ID NO:56] QVQLQESGGGSVQPGGSLRLSCVASGFAFEDYWMYWVRQIPGKGLEWVSTITTGGEDPYYADSVQGRFTI SRDNAKNTLYLEMNSLKPEDTAVYYCTKGLRPFRYCSGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFEDYWMY CDR2: TITTGGEDPY CDR3: GLRPFRYCSGYGSRPGCYDARL SEQ ID NO:91 SEQ ID NO:87 SEQ ID NO:114 mBBBMP061E01[SEQ ID NO:57] QVQLQESGGGSVQPGGSLRLSCVASGFAFEDYWMYWVRQIPGKGLEWVSTITTGGEDPYYADSVQGRFTI SRDNAKNTLYLEMNSLKPEDTAVYYCTKGLQPFRYCSGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFEDYWMY CDR2: TITTGGEDPY CDR3: GLQPFRYCSGYGSRPGCYDARL SEQ ID NO:91 SEQ ID NO:87 SEQ ID NO:113 mBBBMP052B05[SEQ ID NO:58] QVQLQESGGGLVQPGGSLRLSCAASGFAFEDYWMYWVRQIPGKGLEWVSTITTGGEDPYYADSVQGRFTI SRDNAKNTLYLEMNSLKPEDTAVYYCTKGLQPFRYCSGYGSRPGCYDARLPGQGTQVTVSSA CDR1: GFAFEDYWMY CDR2: TITTGGEDPY CDR3: GLQPFRYCSGYGSRPGCYDARL SEQ ID NO:91 SEQ ID NO:87 SEQ ID NO:113 mBBBMP052B04[SEQ ID NO:59] QVQLQESGGGSVRPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDSPYYADSVKGRFTI SRDNAKNTLYLEMNSLKPEDTAVYYCTKGLQPFRYCSGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDSPY CDR3: GLQPFRYCSGYGSRPGCYDARL SEQ ID NO:103 SEQ ID NO:100 SEQ ID NO:113 mBBBMP052H08[SEQ ID NO:60] QVQLQESGGGSVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDSPYYADSVKGRFTI SRDNAKNTLYLEMNSLKPEDTAVYYCTKGLQPFRYCSGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDSPY CDR3: GLQPFRYCSGYGSRPGCYDARL SEQ ID NO:103 SEQ ID NO:100 SEQ ID NO:113 mBBBMP052F05[SEQ ID NO:61] QVQLQESGGGLGQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDSPYYADSVKGRFTI SRDNAKNTLYLEMNSLKPEDTAVYYCTKGLQPFRYCSGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDSPY CDR3: GLQPFRYCSGYGSRPGCYDARL SEQ ID NO:103 SEQ ID NO:100 SEQ ID NO:113 mBBBMP062B09[SEQ ID NO:62] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDSPYYADSVKGRFTI SRDNAKNTLYLEMNSLKAEDTAVYYCTKGLRPFRYCPGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDSPY CDR3: GLRPFRYCPGYGSRPGCYDARL SEQ ID NO:103 SEQ ID NO:100 SEQ ID NO:102 mBBBMP052A10[SEQ ID NO:63] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFII SRDNAKNTVYLEMNNLKAEDTAVYYCTKGLRPFRYCSGYGSRPACYDARLPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GLRPFRYCSGYGSRPACYDARL SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:115 mBBBMP052A08[SEQ ID NO:64] QVQLQESGGGLVQPGGSLRLSCATSGFAFKDYWMYWVRQAPGKGLEWVSTITTGGEDPYYSDSVKGRFTI SRDNAKNTVYLQMNSLQPEDTAVYYCTKGLQPFRYCSGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFKDYWMY CDR2: TITTGGEDPY CDR3: GLQPFRYCSGYGSRPGCYDARL SEQ ID NO:92 SEQ ID NO:87 SEQ ID NO:113 mBBBMP052A06[SEQ ID NO:65] QVQLQESGGGSVQPGGSLRLSCAAAGFAFSDFWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFTV SRDNAKNTVYLEMNSLKPEDTAVYYCTKGLRPFRYCPGYGSRPGCYDARLPGQGTQVTVSS CDR1:GFAFSDFWMY CDR2: TITTGGDDPY CDR3: GLRPFRYCPGYGSRPGCYDARL SEQ ID NO:96 SEQ ID NO:101 SEQ ID NO:102 mBBBMP061A12[SEQ ID NO:66] QVQLQESGGGSVQPGGSLRLSCAAAGFAFSDFWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFTV SRDNAKNTVYLEMNSLKPEDTAVYYCTKGLRPFRYCPGYGSRPGCYDGRLPGQGTQVTVSS CDR1:GFAFSDFWMY CDR2: TITTGGDDPY CDR3: GLRPFRYCPGYGSRPGCYDGRL SEQ ID NO:96 SEQ ID NO:101 SEQ ID NO:116 mBBBMP063C06[SEQ ID NO:67] QVQLQESGGGSVQPGGSLRLSCTTTGFAFSDFWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFTV SRDNAKNTVYLEMNSLKPEDTAVYYCTKGLRPFRYCPGYGSRPGCYDARLPGQGTQVTVSS CDR1:GFAFSDFWMY CDR2: TITTGGDDPY CDR3: GLRPFRYCPGYGSRPGCYDARL SEQ ID NO:96 SEQ ID NO:101 SEQ ID NO:102 mBBBMP061F06[SEQ ID NO:68] QVQLQESGGGLVESGGSLKLSCAAAGFAFSDFWIYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRFTV SRDNAKNTVYLEMNSLKPEDTAVYYCTKGLRPFRYCPGYGSRPGCYDARLPGQGTQVTVSS CDR1: GFAFSDFWIY CDR2: TITTGGDDPY CDR3: GLRPFRYCPGYGSRPGCYDARL SEQ ID NO:117 SEQ ID NO:101 SEQ ID NO:102 mBBBMP052B02[SEQ ID NO:69] QVQLQESGGGSVEPGGSLRLSCAASGFAFGDYWMYWVRQAPGKGLEWVSTITTGGDSPYYTDSVKGRFTI SRDNAKNMLYLQMNSLKPEDTAVYYCAKGLRPFRYCSGYGSRPGCYDARPPGQGTQVTVSS CDR1: GFAFGDYWMY CDR2: TITTGGDSPY CDR3: GLRPFRYCSGYGSRPGCYDARP SEQ ID NO:118 SEQ ID NO:100 SEQ ID NO:119 mBBBMP052E04[SEQ ID NO:70] QVQLQESGGGLVQPGGSLRLSCAASGFPFGDYWMYWVRQAPGKGLEWISTITTGGDSPYYADSVKGRFTI SRDNAKNRLYLEMNSLKAGDTAVYYCAKGLRPYSYCLGYGCYGARPLGQGTQVTVSS CDR1: GFPFGDYWMY CDR2: TITTGGDSPY CDR3: GLRPYSYCLGYGCYGARP SEQ ID NO:120 SEQ ID NO:100 SEQ ID NO:121 mBBBMP052F11[SEQ ID NO:71] QVQLQESGGGLVQPGGSLRLSCAASGFPFGDYWMYWVRQAPGKGLEWISTITTGGDSPYYADSVKGRFTI SRDNAKNRLYLEMNSLKAGDTAVYYCAKGLRPFSYCLGYGCYGARPLGQGTQVTVSS CDR1: GFPFGDYWMY CDR2: TITTGGDSPY CDR3: GLRPFSYCLGYGCYGARP SEQ ID NO:120 SEQ ID NO:100 SEQ ID NO:122 mBBBMP052B10[SEQ ID NO:72] QVQLQESGGGLVQPGGSLRLSCAASGFPFGDYWMYWVRQAPGKGLEWISTITTGGDSPYYADSVKGRFTI SRDNAKNRLYLEMNSLKPGDTAVYYCAKGLRPYSYCLGPGCYGARPLGQGTQVTVSS CDR1: GFPFGDYWMY CDR2: TITTGGDSPY CDR3: GLRPYSYCLGPGCYGARP SEQ ID NO:120 SEQ ID NO:100 SEQ ID NO:123 mBBBMP052C11[SEQ ID NO:73] QVQLQESGGGLVQPGGSLRLSCAASGFAFSDYWMYWVRQAPGKGLEWVSTITTGGDDPYYADSVKGRSTT SRDNDKNTVYLELNSLKPADTAVYYCAKGLRPFSYCVKNGCYGARPLGEGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGDDPY CDR3: GLRPFSYCVKNGCYGARP SEQ ID NO:103 SEQ ID NO:101 SEQ ID NO:124 mBBBMP052H11[SEQ ID NO:74] QVQLQESGGGLVQPGGTLRLSCTASGFAFSDYWMYWVRQAPGKGPEWVSTITTGGIDPYYADSVKGRFII SRDNARNTVDLEMNNLKVEDTAVYYCAKGLRPFSYCLGPGCYGSRPPGQGTQVTVSS CDR1: GFAFSDYWMY CDR2: TITTGGIDPY CDR3: GLRPFSYCLGPGCYGSRP SEQ ID NO:103 SEQ ID NO:125 SEQ ID NO:126 mBBBMP052B03[SEQ ID NO:75] QVQLQEAGGGLVQSGGSLRLSCAASGFSLDYYAIGWFRQAPGKERERVSCISSRDGRTFYADSVKGRFTI SKDNAKNTVYLQMNSLKPEDTAVYSCAAYNTIGQRCPSNGYDYDYWGQGTQVTVSS CDR1: GFSLDYYAIG CDR2: CISSRDGRTF CDR3: YNTIGQRCPSNGYDYDY SEQ ID NO:2 SEQ ID NO:6 SEQ ID NO:127 mBBBMP052D11[SEQ ID NO:76] QVQLQESGGGLVQSGGSLRLSCAASGFSLDYYAIGWFRQAPGKERERVSCISSRDGRTFYADSVKGRFTI SKDNAKNTVYLQMNSLKPEDTAVYSCAAYNTIGQRCPSNGYDYDYWGQGTQVTVSS CDR1: GFSLDYYAIG CDR2: CISSRDGRTF CDR3: YNTIGQRCPSNGYDYDY SEQ ID NO:2 SEQ ID NO:6 SEQ ID NO:127 mBBBMP064E06[SEQ ID NO:77] QVQLQESGGGLVQSGXSLRLSCAASGFXLDYYAIGWFRQAPGKERERVSCISSRDGRTFYADSVKGRFTI SKDNAKNTVYLQMNSLKPEDTAVYSCAAYNTIGQRCPSNGYDYDYWGQGTQVTVSS CDR1: GFXLDYYAIG CDR2: CISSRDGRTF CDR3: YNTIGQRCPSNGYDYDY SEQ ID NO:128 SEQ ID NO:6 SEQ ID NO:127 mBBBMP063D10[SEQ ID NO:78] QVQLQESGGGLVQSGGSLRLSCAASGFSLDYYAIGWFRQAPGKERERVSCISSRDGRTFYADSVKGRFTI SKDNAKNTVYLQMNSLKPEDTAVYSCAAYNTIGQRCPSNGYDYDYWGRGTQVTVSS CDR1: GFSLDYYAIG CDR2: CISSRDGRTF CDR3: YNTIGQRCPSNGYDYDY SEQ ID NO:2 SEQ ID NO:6 SEQ ID NO:127 mBBBMP062G02[SEQ ID NO:79] QVQLQESXGGXVQPGGSLRLSCAASGFXLXYYAIGWXRQAPGKERERVSCIXSRDGRTFYADSVKGRFTI SKDNAKNTVYLQMNSLKPEDTAVYSCAAYNTIGQRCPSNGYDYDYWGQGTQVTVSS CDR1: GFXLXYYAIG CDR2: CIXSRDGRTF CDR3: YNTIGQRCPSNGYDYDY SEQ ID NO:129 SEQ ID NO:130 SEQ ID NO:127 mBBBMP064G11[SEQ ID NO:80] QVQLQESGGGLVQSGGSLRLSCAASGFSLDYYAIGWFRQAPGKERERVSCIGSRDGRTFYADSVKGRFTI SKDHAKNIVYLQMTSLKPEDTAVYSCAAYNTAGQRCPSNGYDYDYWGKGTQVTVSS CDR1: GFSLDYYAIG CDR2: CIGSRDGRTF CDR3: YNTAGQRCPSNGYDYDY SEQ ID NO:2 SEQ ID NO:3 SEQ ID NO:131 mBBBMP052D03[SEQ ID NO:81] QVQLQESGGGLVRSGGSLRLSCVVSGFSLDYYAIGWFRQAPGKERERVSCISSRDGYTFYADSVKGRFTT SKDNAKNTVYLQMISLKPEDTAVYSCAAYNTMGRRCPSNGYEYDYWGQGTQVTVSS CDR1: GFSLDYYAIG CDR2: CISSRDGYTF CDR3: YNTMGRRCPSNGYEYDY SEQ ID NO:2 SEQ ID NO:132 SEQ ID NO:133 mBBBMP062G04[SEQ ID NO:82] QVQLQESGGGLVRSGGSLRLSCVVSGFSLDYYAIGWFRQAPGKERERVSCISSRDGPTEYADSVKGRFTT AKDNMSNTVFLQMNSLKPEDTAVYYCAAYNTIGRRCPSNPYEYDYWGQGTQVTVSS CDR1: GFSLDYYAIG CDR2: CISSRDGPTE CDR3: YNTIGRRCPSNPYEYDY SEQ ID NO:2 SEQ ID NO:134 SEQ ID NO:135 mBBBMP063F08[SEQ ID NO:83] QVQLQESGGGLVQPGGSLRLSCVASGFSLNYYAIGWFRQAPGKERERVSCISSRDGPTYYADSVKGRFTI AKDNMKNTVFLQMNSLKPEDTAVYYCAAYNTIGRRCPSNPYEYDYWGQGTQVTVSS CDR1: GFSLNYYAIG CDR2: CISSRDGPTY CDR3: YNTIGRRCPSNPYEYDY SEQ ID NO:136 SEQ ID NO:137 SEQ ID NO:135 mBBBMP052G11[SEQ ID NO:84] QVQLQESGGGLVQPGGSLSVSCVASGFSLDYYAIGWFRQAPGKERERVSCISSRDGTTYYADSVKGRFTV AKDNTKNTVFLQMNSLKPEDTAVYYCAAYNTIGPRCPSNHYDYDYWGQGTQVTVSS CDR1: GFSLDYYAIG CDR2: CISSRDGTTY CDR3: YNTIGPRCPSNHYDYDY SEQ ID NO:2 SEQ ID NO:138 SEQ ID NO:139 mBBBMP062G12[SEQ ID NO:85] QVQLQESGGGLVQPGESLRVSCVVSGSMFSSVVMGWYRQAPGKERELVADITPGGSDMYADSVKGRFTLS RDNARHTVYLQMTSLKPEDTAVYYCNARTWSNRNYWGQGTQVTVSS CDR1: GSMFSSVVMG CDR2: DITPGGSDM CDR3: RTWSNRNY SEQ ID NO:140 SEQ ID NO:141 SEQ ID NO:142

Claims

CLAIMS 1. A multispecific polypeptide comprising: - a first protein for the prevention, treatment or detection of a neurological disorder in the central nervous system, - a second protein binding specifically to a protein expressed on the blood brain barrier, wherein the protein expressed on the blood brain barrier mediates enhanced uptake of compounds over the BBB, characterised in further comprising a third protein binding specifically to a protein in the CNS.

2. The multispecific polypeptide according to claim 1, wherein the first protein is a protein for prevention or treatment of a neurological disorder in the central nervous system.

3. The multispecific polypeptide according to claim 1 or 2, wherein the first protein is a protein for the prevention, treatment or detection of a neurological disorder in the brain.

4. The multispecific polypeptide according to any one of claims 1 to 3, wherein the first protein is a protein for the treatment of a neurological disorder in the brain.

5. The multispecific polypeptide according to any one of claims 1 to 4, wherein the first protein is an antibody or antigen binding fragment thereof binding specifically to a therapeutic target in the brain.

6. The multispecific polypeptide according to claim 5, wherein the therapeutic target is BACE1.

7. The multispecific polypeptide according to any one of claims 1 to 6, wherein the second protein binds to a membrane protein involved in receptor- mediated transcytosis.

8. The multispecific polypeptide according to any one of claims 1 to 7, wherein the second protein binds a receptor involved in receptor-mediated transcytosis.

9. The multispecific polypeptide according to claim 8, wherein the receptor involved in receptor-mediated transcytosis over the blood brain barrier is transferrin receptor (TfR).

10. The multispecific polypeptide according to any one of claims 1 to 9, wherein the third protein is protein binding specifically to a structural protein in the CNS.

11. The multispecific polypeptide according to claim 10, wherein the structural protein is MOG (myelin oligodendrocyte glycoprotein).

12. The multispecific polypeptide according to any one of claims 1 to 11, wherein the second protein and / or the third protein is an antibody or antigen binding fragment thereof.

13. The multispecific polypeptide according to claim 12, wherein one or more of the antibodies are single domain antibodies (sdAbs).

14. The multispecific protein according to any one of claim 1 to 13 comprising: - an antibody or antigen binding fragment thereof binding specifically to a therapeutic target of a neurological disorder in the brain, - an antibody or antigen binding fragment thereof binding specifically to the transferrin receptor, - one or more antibodies or antigen binding fragments thereof binding specifically to MOG.

15. The multispecific protein according to any one of claims 1 to 14, wherein the antibody binding to MOG is a single domain antibody defined by: CDR1 sequence GFSLDYYAIG [SEQ ID NO:2], CDR2 sequence CIGSRDGRTF [SEQ ID NO:3] and CDR3 sequence AYNTAGQRCPSNGYDYDYWG [SEQ ID NO:4].

16. The multispecific antibodies according to any one of claims 1 to 14, wherein the antibody binding to MOG is a single domain antibody defined by:CDR1 sequence GFSLDYYAIG [SEQ ID NO:2], CDR2 sequence CISSRDGRTF [SEQ ID NO:6], and CDR3 sequence AYNTIGQRCPSNGYDYDYWG [SEQ ID NO:7].

17. A polynucleotide or polynucleotides encoding the multifunctional polypeptide as defined in any one of claim 1 to 16.

18. A multispecific polypeptide according to any one of claims 1 to 16, for use in the prevention, treatment or detection of a neurological disorder in the brain.

19. A polynucleotide or polynucleotides encoding the polypeptides as defined in any one of claim 1 to 16, for use in the prevention, treatment or detection of a neurological disorder in the brain.

20. A single domain antibody defined by: CDR1 sequence GFSLDYYAIG [SEQ ID NO:2], CDR2 sequence CIGSRDGRTF [SEQ ID NO:3], and CDR3 sequence AYNTAGQRCPSNGYDYDYWG [SEQ ID NO:4].

21. A single domain antibody defined by: CDR1 sequence GFSLDYYAIG [SEQ ID NO:2], CDR2 sequence CISSRDGRTF [SEQ ID NO:6], and CDR3 sequence AYNTIGQRCPSNGYDYDYWG [SEQ ID NO:7].

Images