VHH antibodies and uses thereof

Abstract

The present invention relates to heavy chain-only variable domain (VHH) antibodies that specifically bind to transferrin receptor 1 (TfR1) and their use to transport therapeutic or imaging agents intracellularly and across the blood-brain barrier.

Description

[Technical Field]

[0001] The present invention relates to heavy chain-only variable domain (VHH) antibodies capable of binding to the transferrin receptor, and the use of such VHH antibodies to transport molecules across the blood-brain barrier to relevant targets in the brain. [Background technology]

[0002] Brain exposure of drugs targeting central nervous system (CNS) diseases is inherently challenging because the blood-brain barrier (BBB) ​​protects the brain from unwanted substances present in the peripheral circulation, including antibodies and other proteins that may have therapeutic effects in the brain. Some small molecules are excluded from the brain compartment due to properties such as lipophilicity or lack of substrates for efflux pumps. Typically, compounds smaller than 600 daltons (Da) can cross the BBB unless excluded by other forces. Larger molecules, such as proteins, do not readily enter the brain in significant quantities unless assisted by specific active transport. In the case of therapeutic antibodies and protein-based drugs, it is estimated that less than 0.1% of systemically injected therapeutic antibodies reaches the brain compartment. Several strategies to overcome this narrow barrier are being tested and evaluated.

[0003] The BBB is composed of brain endothelial cells (BECs) as the first barrier to entry into the brain. Other cells of the so-called neurovascular unit (NVU) are also important for transport and interaction with target cells within the brain parenchyma. In the human brain, the cerebral blood vessels total 20 m 2 , representing a large surface area, which may present an opportunity for brain exposure for circulatory therapies.

[0004] Therapeutic antibodies or other protein-based drugs have great potential for treating CNS pathologies. However, the low availability of such therapeutic molecules in brain compartments remains a major problem. Recently, therapeutic monoclonal antibodies targeting the brain, such as amyloid beta protofibrils, have reported clinical efficacy. However, the exposure of these therapeutic molecules in human brain compartments after each administration is not considered optimal.

[0005] Therefore, to improve the safety, dosing, and overall cost of CNS therapeutics, there is a need to increase the brain exposure of therapeutic molecules.

[0006] Receptor-mediated transcytosis, a natural mechanism using endogenous receptors expressed on the luminal surface of the BBB, has been reported to be successful in increasing the brain exposure of therapeutic molecules and to be clinically effective and safe.

[0007] WO2020 / 144233 discloses camelid heavy chain-only variable domain (VHH) molecules that bind to the transferrin receptor (TfR) and their uses for the delivery of molecules of pharmaceutical or diagnostic interest within cells and organs in pathological conditions, including cancer.

[0008] WO2016 / 077840, WO2019 / 089395, WO2020 / 056327, and WO2022 / 103769 disclose TfR-specific binding moieties that can be used to transport biomolecules across membranes, including the BBB and the gastrointestinal tract. These TfR-specific binding moieties comprise single-domain nurse shark variable domains of novel antigen receptor (VNAR) antibodies that bind to TfR.

[0009] WO2016 / 081643 relates to anti-transferrin receptor antibodies and methods of using same.

[0010] There remains a need for effective transporters of therapeutic molecules into the brain. These transporters should be capable of being conjugated or fused to therapeutic or diagnostic molecules in a manner that does not affect the target binding of the transporter or the therapeutic effect of the therapeutic molecule. Summary of the Invention

[0011] It is a general objective to provide VHH molecules that are specific for transferrin receptor 1 and are able to efficiently transport cargo to a desired compartment by receptor-mediated transcytosis.

[0012] These and other objects are achieved by embodiments of the present invention.

[0013] The invention is defined in the independent claims. Further embodiments of the invention are defined in the dependent claims.

[0014] An embodiment of the present invention relates to a heavy chain-only variable domain (VHH) antibody that specifically binds to transferrin receptor 1 (TfR1). The VHH antibody comprises a complementarity determining region 1 (CDR1) having an amino acid sequence selected from the group consisting of GSIFGSKR as defined in SEQ ID NO:1 and GSIFGFNA as defined in SEQ ID NO:2. The VHH antibody also comprises a CDR2 having an amino acid sequence selected from the group consisting of ITYRGTT as defined in SEQ ID NO:3 and IAVAGST as defined in SEQ ID NO:4. The VHH antibody further comprises a CDR3 having an amino acid sequence selected from the group consisting of WMFTTDNY as defined in SEQ ID NO:5 and WMYATANY as defined in SEQ ID NO:6. When CDR1 has the amino acid sequence defined in SEQ ID NO:2, CDR3 has the amino acid sequence defined in SEQ ID NO:6. The VHH antibody does not comprise a CDR1 having the amino acid sequence defined in SEQ ID NO:2, a CDR2 having the amino acid sequence defined in SEQ ID NO:4, and a CDR3 having the amino acid sequence defined in SEQ ID NO:6.

[0015] Another aspect of the present invention relates to a VHH antibody that specifically binds to TfR1. The VHH antibody comprises a CDR1 consisting of the amino acid sequence X1X2IX3GSKR as defined in SEQ ID NO:7, where X1 is G or E, X2 is S, D, or I, and X3 is F or N. The VHH antibody also comprises a CDR2 consisting of the amino acid sequence ITX4X5GTT as defined in SEQ ID NO:8, where X4 is Y or V, and X5 is R, H, or G. The VHH further comprises a CDR3 consisting of the amino acid sequence WMFTTX6NY as defined in SEQ ID NO:9, where X6 is D, T, or N.

[0016] A further aspect of the present invention relates to a fusion molecule comprising a VHH antibody as described above linked to at least one molecule.

[0017] A related aspect of the present invention defines the above-described fusion molecules for use as a medicament, wherein at least one molecule is a therapeutic agent or for use in treating a disease or disorder of the central nervous system (CNS), and at least one molecule is a therapeutic agent capable of treating a disease or disorder of the CNS.

[0018] Yet another aspect of the present invention relates to a pharmaceutical composition comprising the above-described fusion molecule and a pharmaceutically acceptable vehicle, wherein at least one molecule is a therapeutic agent.

[0019] Other aspects of the present invention relate to nucleic acid molecules encoding the above-described VHH antibodies or fusion molecules, expression vectors comprising the above-described nucleic acid molecules operably linked to a promoter, and host cells comprising the above-described nucleic acid molecules or the above-described expression vectors.

[0020] The VHH antibodies of the present invention specifically bind to TfR1 without interfering with the binding of transferrin to TfR1. The binding properties of the VHH antibodies are tailored to optimize transcytosis on endothelial cells at the BBB. This allows the VHH antibodies to be used as transporters of various molecules, including therapeutic agents or diagnostic imaging agents, to the brain compartment when administered systemically.

[0021] The embodiments, together with further objects and advantages thereof, may best be understood by reference to the following description taken in conjunction with the accompanying drawings, in which: [Brief explanation of the drawings]

[0022] [Figure 1A]Graphical representation of llama immunization with immunogens. Immunization of llamas (Llama glama) (N=2) was performed at Modiquest Research BV (The Netherlands) according to the ModiPhage™ method. The primary immunization was performed on day 1 with the human transferrin receptor 1 ectodomain (hTfR1) (500 μg protein + complete Freund's adjuvant (CFA), intramuscularly (im)). A boosting dose (500 μg protein + incomplete Freund's adjuvant (IFA), im) was administered on day 21 for mouse TfR1 (mTfR1) and day 42 for hTfR1. Peripheral blood was withdrawn on day 54, and the immune response was tested by enzyme-linked immunosorbent assay (ELISA) to test for the presence of both immunoglobulin G1 (IgG1) and IgG2 / 3 antibodies and their reactivity to mTfR1 and hTfR1. Immunizations were repeated on days 86 (hTfR1 + IFA) and 107 (hTfR1 + IFA). New peripheral blood samples were analyzed on day 117. An additional pre-harvest boost dose with a 1:1 mixture of hTfR1 and mTfR1 (250 μg + 250 μg + IFA) was given on day 120, followed by peripheral blood lymphocyte (PBL) harvest on day 124. [Figure 1B]Strategy for generating TfR1-binding VHHs. Graphical overview of phage library establishment. PBLs were isolated by density gradient centrifugation using approximately 1.5 x 109 Ficoll-Paque™ PLUS. Ribonucleic acid (RNA) was extracted, followed by reverse transcription into complementary deoxyribonucleic acid (cDNA). The cDNA was then used as a template in polymerase chain reaction (PCR) to amplify the IgG repertoire (IgG2 / 3, variable heavy chain (VH) domain to constant heavy chain 2 (CH2), heavy chain-only antibodies). A first PCR reaction was performed to amplify all antibodies (IgG1 and IgG2 / 3) and a second nested PCR reaction was performed to amplify and isolate the VHH repertoire (described in Pardon et al., A general protocol for the generation of Nanobodies for structural biology. Nat Protoc 9:674-693 (2014)). After vector digestion, DNA amplification was performed as described above, and a total of 1600 ng of DNA was used to electroporate TG1 E. coli, yielding a library with an estimated size of 3.2 x 10 8 . [Figure 2] Graphical representation of VHH and various VHH-containing fusion proteins. Figure 2 illustrates a schematic of a VHH monomer and various VHH-containing fusion proteins. Free VHH binds as a monomer to one binding site on TfR1. Bivalent VHH fusion proteins were formed between two VHH and a human fragment crystallizable (Fc) region. Functional monovalent fusion proteins were produced between one VHH and a single-chain variable fragment (scFv) or between one VHH and a non-antibody-derived molecule (X). Figure 2 also shows a bivalent VHH fusion protein with two non-antibody-derived molecules (X). [Figure 3]Affinity measurement of monomeric VHH binding to hTfR1 using SPR. The figure shows representative sensorgrams of KB_A01 (A) at increasing concentrations ranging from 6.25 to 100 nM binding to human transferrin (hTf) loaded onto a dextran-coated gold (CM5) chip (Cytiva) by amine coupling. As a control, the full-length monoclonal antibody BA1 (B) was run at the same concentration. [Figure 4] Affinity measurement of dimeric VHH-Fc proteins binding to hTfR1 using SPR. The figure shows representative sensorgrams of hTfR1 at increasing concentrations ranging from 0.16 to 100 nM binding to the Fc-fusion KB_A01 (A) loaded onto a Protein A-coated chip (Cytiva). As control standards, the full-length monoclonal antibody BA2 (B) and the reference VHH-Fc BV (C) were also tested. KB_A01, BA2, and BV were also tested against mTfR1, but only BV bound with detectable affinity (D). [Figure 5] Affinity measurements of dimeric KB_A01-Fc for hTfR1 and cTfR1 in the absence and presence of hTf. The figure shows representative sensorgrams of KB_A01-Fc fusion protein binding to hTfR1 (A-B) or cynomolgus monkey TfR1 (cTfR1) (C-D) in the absence (A and C) and presence (B and D) of an excess concentration of 250 nM hTf. [Figure 6] Affinity measurements of reference antibodies for hTfR1 in the absence and presence of hTf. The figure shows representative sensorgrams of reference antibody BA2 (A-B) or reference antibody BA1 (C-D) binding to hTfR1 in the absence (A and C) and presence (B and D) of an excess concentration of 250 nM hTf. [Figure 7] Sequence alignment of sequences closely related to KB_A01. The figure shows the amino acid sequences of KB_A01 (SEQ ID NO: 23) and three closely related clones, KB_A09 to KB_A11, in single-letter code. Dots indicate amino acids identical to the reference sequence (KB_A01), and boxes indicate the three complementarity-determining regions (CDRs). [Figure 8] Binding to the apical and extracellular domains of hTfR1 measured by ELISA. The apical domain of hTfR1 was expressed in M13-phage as a stabilized unit and used to determine the apical binding of KB_A01, the reference antibody BA2, and the reference VHH-Fc BV, as well as its lack of binding to BV-Fc (A). Variants of KB_A01 were also assayed in this system, and representative clones confirmed binding to the apical domain (B). Validation ELISAs were performed against representative clones or the indicated reference antibodies using soluble ectodomain (without M13 phage) at 2.5 μg / mL for full-length IgG and 1.25 μg / mL for VHH-Fcs (C). [Figure 9] Affinity measurement of VHH and scFv fusion proteins binding to hTfR1 using SPR. Human TfR1 was amine-coupled to the surface of a CM5 chip and exposed to KB_A01 genetically fused to scFv at the C-terminus (VHH-scFv, A) or N-terminus (scFv-VHH, B) at increasing concentrations ranging from 0.25 to 64 nM in 1:4 step increments. [Figure 10A]TfR1-mediated cellular uptake in HEK293T cells. HEK293T cells were cultured in 96-well plates until confluence was reached (4-5 days). Test compounds (KB_A01-Fc, BA1, BA2, and a negative non-binding VHH-Fc, negative control) were added at 3.3-20 nM, diluted in DMEM, and incubated for 15 or 45 min. Cells were then rinsed with PBS and fixed with 4% PFA (Figures 10A and 10B). For longer incubation times (Figure 10C), the medium was changed twice: at t = 30 min incubation and again at t = 120 min with fresh DMEM. After incubation times of 15, 40-45, or 240 min, cells were immunostained and analyzed by confocal microscopy. Representative images of cellular uptake after 40 min at 20 nM for KB_A01-Fc, BA1, and the negative control are shown here (10A), or comparing 15 and 45 min of incubation for KB_01-Fc and BA1 at 3.3 nM (10B). The incubations of KB_A01-Fc, BA1, and BA2, as well as the negative control, are shown as inverted grayscale images at 240 min (10C). Results at t = 240 min show that BA1 has significantly less signal (intracellular presence) at this time point compared to BA2 and KB_A01-Fc, indicating a higher degree of intracellular degradation. Figures 10A and 10C also show the absence of signal using the non-hTfR1-binding VHH-Fc negative control. Images are shown as 8-bit grayscale images (10A, 10B) or as inverted grayscale images (10C). Scale bar: 50 μm. [Figure 10B]TfR1-mediated cellular uptake in HEK293T cells. HEK293T cells were cultured in 96-well plates until confluence was reached (4-5 days). Test compounds (KB_A01-Fc, BA1, BA2, and a negative non-binding VHH-Fc, negative control) were added at 3.3-20 nM, diluted in DMEM, and incubated for 15 or 45 min. Cells were then rinsed with PBS and fixed with 4% PFA (Figures 10A and 10B). For longer incubation times (Figure 10C), the medium was changed twice: at t = 30 min incubation and again at t = 120 min with fresh DMEM. After incubation times of 15, 40-45, or 240 min, cells were immunostained and analyzed by confocal microscopy. Representative images of cellular uptake after 40 min at 20 nM for KB_A01-Fc, BA1, and the negative control are shown here (10A), or comparing 15 and 45 min of incubation for KB_01-Fc and BA1 at 3.3 nM (10B). The incubations of KB_A01-Fc, BA1, and BA2, as well as the negative control, are shown as inverted grayscale images at 240 min (10C). Results at t = 240 min show that BA1 has significantly less signal (intracellular presence) at this time point compared to BA2 and KB_A01-Fc, indicating a higher degree of intracellular degradation. Figures 10A and 10C also show the absence of signal using the non-hTfR1-binding VHH-Fc negative control. Images are shown as 8-bit grayscale images (10A, 10B) or as inverted grayscale images (10C). Scale bar: 50 μm. [Figure 10C]TfR1-mediated cellular uptake in HEK293T cells. HEK293T cells were cultured in 96-well plates until confluence was reached (4-5 days). Test compounds (KB_A01-Fc, BA1, BA2, and a negative non-binding VHH-Fc, negative control) were added at 3.3-20 nM, diluted in DMEM, and incubated for 15 or 45 min. Cells were then rinsed with PBS and fixed with 4% PFA (Figures 10A and 10B). For longer incubation times (Figure 10C), the medium was changed twice: at t = 30 min incubation and again at t = 120 min with fresh DMEM. After incubation times of 15, 40-45, or 240 min, cells were immunostained and analyzed by confocal microscopy. Representative images of cellular uptake after 40 min at 20 nM for KB_A01-Fc, BA1, and the negative control are shown here (10A), or comparing 15 and 45 min of incubation for KB_01-Fc and BA1 at 3.3 nM (10B). The incubations of KB_A01-Fc, BA1, and BA2, as well as the negative control, are shown as inverted grayscale images at 240 min (10C). Results at t = 240 min show that BA1 has significantly less signal (intracellular presence) at this time point compared to BA2 and KB_A01-Fc, indicating a higher degree of intracellular degradation. Figures 10A and 10C also show the absence of signal using the non-hTfR1-binding VHH-Fc negative control. Images are shown as 8-bit grayscale images (10A, 10B) or as inverted grayscale images (10C). Scale bar: 50 μm. [Figure 11]Transcytosis experiments in brain-like endothelial cells in an in vitro BBB model. An in vitro BBB transwell assay using brain-like endothelial cells was performed as described in Sjoestroem et al., Transport study of interleukin-1 inhibitors using a human in vitro model of the blood-brain barrier, Brain Behavior, Immunity Health 16:100307 (2021). Test compounds (500 nM in physiological buffer) were added to donor wells, and compartments were harvested at 180 minutes (n = 3 independent wells per sample). Cells were lysed, and the contents of all compartments were analyzed by ELISA reactive to human Fc. The control IgG used was a known anti-IL1 beta antibody. All test items were reactive to human Fc and analyzed based on their own standard curves and IgG standards on the same sample plate. KB_A01-Fc demonstrated higher transcytosis capacity than the control IgG antibody and the reference antibody BA2. [Figure 12A] PrismA™ enabled mutations of KB_A01. The figure shows KB_A01, KB_A12, and positive control 1, all with His-tags, produced in E. coli, purified using IMAC (Ni-NTA) and run on SDS-PAGE. The IMAC purified material was then purified using a Protein A-based resin (PrismA™ resin). [Figure 12B] PrismA™-Effective Mutations of KB_A01 Acid-eluted fractions were analyzed by SDS-PAGE. [Figure 12C] PrismA™-Effective Mutations of KB_A01 Flow-through fractions were analyzed by SDS-PAGE. [Figure 12D] PrismA™-Effective Mutations of KB_A01 KB_A01 was also tested for binding to PrismA™ on a pre-immobilized chip (Cytiva) by SPR over a concentration range of 1.95-500 nM. [Figure 12E]The PrismA™-enabling mutation of KB_A01, KB_A12, was also tested for binding to PrismA™ on a pre-immobilized chip (Cytiva) by SPR over a concentration range of 1.95-500 nM. [Figure 13] Brain and Blood Distribution of VHH-Fc Fusion KB_A01 after Systemic Administration in TfR1 Extracellular Domain-Humanized Mice Radiolabeled [I]VHH-Fc fusion protein was injected into human extracellular domain chimeric TfR1 (hECD-TfR1) mice with homozygous (HOM), heterozygous (HET), and wild-type (WT) genotypes, as described in Example XI. (A) Blood concentrations from blood samples taken at t = 5 min, t = 30 min, t = 1 h, and terminal (2.0 h) time points are shown. (B) and (C) Two hours after injection, animals were euthanized, transcardially perfused with NaCl, and brains were excised. Radioactivity, blood compartments, and brain concentrations at 2.0 h were analyzed. Blood concentrations are expressed as (% of injected dose, radioactivity), and brain uptake is expressed as standardized uptake value, SUV (% of injected dose corrected for animal weight). DETAILED DESCRIPTION OF THE INVENTION

[0023] The present invention relates to heavy chain-only variable domain (VHH) antibodies, also called single domain antibodies, in particular to such VHH antibodies capable of binding to transferrin (Tf) receptor 1 (TfR1), and to the use of such VHH antibodies for transporting molecules across the blood-brain barrier (BBB).

[0024] Brain exposure of drugs targeting central nervous system (CNS) diseases is inherently challenging because the BBB protects the brain from unwanted substances present in the peripheral circulation, including antibodies, proteins, and other molecules that may have therapeutic effects in the brain. Receptor-mediated transcytosis, a natural mechanism that uses endogenous receptors expressed on the luminal surface of the BBB, has been suggested to increase BBB exposure of drugs. One such endogenous receptor that can be used to achieve receptor-mediated transcytosis is the transferrin receptor.

[0025] The transferrin receptor (TR) is a glycoprotein that binds iron (Fe) and mediates its transport through plasma. It internalizes the transferrin-iron complex via receptor-mediated endocytosis. Humans and other mammals have two transferrin receptors: transferrin receptor 1 (TfR1) and transferrin receptor 2 (TfR2). While TfR1 is a ubiquitously expressed receptor with high affinity, TfR2 expression is restricted to specific cell types and is not affected by intracellular iron concentration. TfR2 binds transferrin with 25-30 times lower affinity than TfR1. As used herein, the term "transferrin receptor" refers to the TfR1 homolog, also known as cluster of differentiation 71 (CD71), encoded by the TFRC gene in humans.

[0026] The extracellular domains of human TfR1 (hTfR1) and mouse TfR1 (mTfR1) were used to immunize llamas (Llama glama) to generate a library of VHH antibody-expressing clones. These VHH antibodies were screened for binding to hTfR1, and a lead VHH antibody, KB_A01, with desirable binding properties was selected. Additional VHH antibodies were generated through selected CDR modifications and the identification of VHH antibodies with sequence similarity to KB_A01. The VHH antibodies of the present invention specifically bind to hTfR1 and also to cynomolgus monkey TfR1 (cTfR1), possessing binding properties favorable for transcytosis, including binding affinity, non-interference with the transferrin-binding site, and binding to the apical domain of hTfR1. As shown in the experimental section, these VHH antibodies retain TfR1-binding properties when present as VHH-containing fusion proteins, including bivalent VHH fusion proteins formed between two VHH antibodies and a human constant antibody fragment crystallizable region (Fc), and functional fusion proteins fused to drug molecules, such as single-chain variable fragments (scFv). Experimental data further demonstrate that VHH antibodies, including VHH-containing fusion proteins, were taken up by human cells expressing hTfR1 and transcytosed through human brain-like endothelial cell monolayers used as an in vitro model of the BBB.

[0027] The VHH antibodies of the present invention further have advantages over the VHH antibodies disclosed in WO2020 / 144233. First, the VHH antibodies of the present invention bind to different epitopes on hTfR1 compared to these prior art VHH antibodies disclosed in Example VIII. The epitopes bound by the prior art VHH antibodies are not within the so-called apical domain of hTfR1, to which the VHH antibodies of the present invention bind. Binding of the ectodomain of TfR1 to the apical domain is advantageous over binding to the helical or protease-like domain for the purposes of BBB delivery and receptor-mediated transcytosis because these do not interfere with transferrin binding (Daniels et al., The transferrin receptor part I: Biology and targeting with cytotoxic antibodies for the treatment of cancer. Clin Immunol. 2006;121:144-158 (2006); WO 2016 / 081643). Furthermore, binding of the VHH antibodies of the present invention to the apical domain of TfR1 does not interfere with the binding of holo-transferrin to the TfR1. See Figure 5 and Table 5. Another significant advantage of the VHH antibodies of the present invention compared to the VHH antibodies disclosed in WO2020 / 144233 is that the VHH antibodies have binding properties, both in terms of their affinity range for hTfR1 and their release from hTfR1, that allow for efficient crossing of the BBB and high relative uptake in the brain. The VHH antibodies disclosed in WO2020 / 144233 have significantly higher affinity for hTfR1 and are vulnerable to lysosomal degradation if endocytosed into endothelial cells of the BBB rather than crossing the BBB. This can result in the VHH antibodies of WO2020 / 144233 becoming trapped in endothelial cells and ultimately being degraded there. Another significant advantage of the VHH antibodies of the present invention compared to the VHH antibodies disclosed in WO2020 / 144233 is that there is no significant change in the affinity of the VHH antibodies of the present invention for hTfR1 in the presence or absence of human transferrin (hTf).However, the affinity of the VHH antibodies disclosed in WO2020 / 144233 for hTfR1 was significantly reduced in the presence of hTf compared to the absence of hTf, as shown in Table 5. This dependence of the affinity for hTfR1 on the hTf concentration makes it difficult to select the appropriate amount of VHH antibody required to achieve the desired receptor-mediated transcytosis across the BBB.

[0028] Thus, an embodiment of the present invention relates to a heavy chain-only variable domain (VHH) antibody that specifically binds to transferrin receptor 1 (TfR1). The VHH antibody comprises a complementarity determining region 1 (CDR1) having an amino acid sequence selected from the group consisting of GSIFGSKR as defined in SEQ ID NO:1 and GSIFGFNA as defined in SEQ ID NO:2. The VHH antibody also comprises a CDR2 having an amino acid sequence selected from the group consisting of ITYRGTT as defined in SEQ ID NO:3 and IAVAGST as defined in SEQ ID NO:4. The VHH antibody further comprises a CDR3 having an amino acid sequence selected from the group consisting of WMFTTDNY as defined in SEQ ID NO:5 and WMYATANY as defined in SEQ ID NO:6. According to the present invention, if CDR1 has the amino acid sequence defined in SEQ ID NO:2, then CDR3 has the amino acid sequence defined in SEQ ID NO:6. With the proviso that the VHH antibody does not comprise a CDR1 having the amino acid sequence defined in SEQ ID NO:2, a CDR2 having the amino acid sequence defined in SEQ ID NO:4, and a CDR3 having the amino acid sequence defined in SEQ ID NO:6.

[0029] The VHH antibodies of this embodiment were generated based on CDR shuffling between KB_A01 and a VHH antibody containing several alanine mutations in the CDRs compared to KB_A01. The strategy was to treat the VHH antibody containing alanine mutations in the CDRs (referred to herein as KB_ref) as a naturally occurring alanine scan, but to replace the entire CDR rather than a single amino acid residue in KB_A01. KB_ref was initially identified as an hTfR1 binder when analyzed by ELISA. However, unexpectedly, this VHH antibody, KB_ref, did not bind to hTfR1 when immobilized with Protein A and analyzed using surface plasmon resonance (SPR). However, several VHH antibodies obtained by selected CDR shuffling, KB_A03, KB_A04, KB_A06, and KB_A07, were indeed able to bind to VHHs with the desired binding properties. This was highly unexpected, considering that the VHH antibody KB_ref, from which part of the CDRs of KB_A03, KB_A04, KB_A06, and KB_A07 were derived, was unable to bind to hTfR1.

[0030] More specifically, the scaffold VHH antibody lacking alanine residues in the CDRs, i.e., KB_A01, has CDR1 defined as SEQ ID NO: 1, CDR2 defined as SEQ ID NO: 3, and CDR3 defined as SEQ ID NO: 5, while the VHH antibody containing alanine mutations in the CDRs, i.e., KB_ref, has CDR1 defined as SEQ ID NO: 2, CDR2 defined as SEQ ID NO: 4, and CDR3 defined as SEQ ID NO: 6.

[0031] Replacing all CDRs of KB_A01 with those of KB_ref results in a loss of binding to hTfR1, KB_A08, as shown in Table 7. Thus, the VHH antibody of this embodiment does not comprise a CDR1 having the amino acid sequence defined in SEQ ID NO:2, a CDR2 having the amino acid sequence defined in SEQ ID NO:4, and a CDR3 having the amino acid sequence defined in SEQ ID NO:6. Furthermore, replacing the CDR1 of KB_A01 with that of KB_ref resulted in a loss of hTfR1 binding unless the CDR3 of KB_A01 was also replaced with that of KB_ref. See KB_A02 and KB_A05 in Table 5. Thus, if the CDR1 of a VHH antibody of this embodiment has the amino acid sequence defined in SEQ ID NO:2, then the CDR3 has the amino acid sequence defined in SEQ ID NO:6.

[0032] Table 6 summarizes the CDRs of VHH antibodies KB_A01 to KB_A08, and Table 7 shows the binding kinetics of these VHH antibodies to hTfR1 in the form of dimeric VHH-Fc fusion proteins.

[0033] In one embodiment, the VHH antibody does not comprise a CDR1 having the amino acid sequence defined in SEQ ID NO: 2, a CDR2 having the amino acid sequence defined in SEQ ID NO: 3 or 4, and a CDR3 having the amino acid sequence defined in SEQ ID NO: 6. In this embodiment, the VHH antibody does not comprise a CDR1 having the amino acid sequence defined in SEQ ID NO: 2, a CDR2 having the amino acid sequence defined in SEQ ID NO: 3, and a CDR3 having the amino acid sequence defined in SEQ ID NO: 6. Furthermore, in this particular embodiment, the VHH antibody does not comprise a CDR1 having the amino acid sequence defined in SEQ ID NO: 2, a CDR2 having the amino acid sequence defined in SEQ ID NO: 4, and a CDR3 having the amino acid sequence defined in SEQ ID NO: 6.

[0034] In one embodiment, the CDR1 has the amino acid sequence defined in SEQ ID NO: 1, the CDR2 has an amino acid sequence selected from the group consisting of SEQ ID NOs: 3 and 4, and the CDR3 has the amino acid sequence defined in SEQ ID NO: 5.

[0035] In a specific embodiment, the CDR1 has the amino acid sequence defined in SEQ ID NO:1, the CDR2 has the amino acid sequence defined in SEQ ID NO:4, and the CDR3 has the amino acid sequence defined in SEQ ID NO:5.

[0036] In another embodiment, the CDR1 has the amino acid sequence defined in SEQ ID NO: 1, the CDR2 has the amino acid sequence defined in SEQ ID NO: 3, and the CDR3 has the amino acid sequence defined in SEQ ID NO: 5.

[0037] In another embodiment, the CDR1 has an amino acid sequence selected from the group consisting of SEQ ID NO: 1 and 2, the CDR2 has an amino acid sequence defined in SEQ ID NO: 3, and the CDR3 has an amino acid sequence defined in SEQ ID NO: 6.

[0038] In a specific embodiment, the CDR1 has the amino acid sequence defined in SEQ ID NO:1, the CDR2 has the amino acid sequence defined in SEQ ID NO:3, and the CDR3 has the amino acid sequence defined in SEQ ID NO:6.

[0039] In another embodiment, the CDR1 has the amino acid sequence defined in SEQ ID NO:2, the CDR2 has the amino acid sequence defined in SEQ ID NO:3, and the CDR3 has the amino acid sequence defined in SEQ ID NO:6.

[0040] In a further embodiment, the CDR1 has the amino acid sequence defined in SEQ ID NO:1, the CDR2 has the amino acid sequence defined in SEQ ID NO:4, and the CDR3 has the amino acid sequence defined in SEQ ID NO:6.

[0041] In one embodiment, the VHH antibody is of the formula: framework region 1 (FR1)-CDR1-FR2-CDR2-FR3-CDR3-FR4.

[0042] In one embodiment, the FR1 has the amino acid sequence QVQLQESGGGSVQAGGSLSLSCAAS as defined in SEQ ID NO:57.

[0043] In one embodiment, the FR2 has the amino acid sequence MGWFRQAPGEQRDVVAT as defined in SEQ ID NO:58.

[0044] In one embodiment, the FR3 has the amino acid sequence EYADSVKGRFTISRDNAKNTVYLQMNNLKPEDTAVYYC as defined in SEQ ID NO:59.

[0045] In one embodiment, the FR4 has the amino acid sequence WGQGTQVTVSS as defined in SEQ ID NO:22.

[0046] In certain embodiments, the VHH antibody has the amino acid sequences of FR1, FR2, FR3 and FR4 defined in SEQ ID NOs: 57, 58, 59 and 22.

[0047] In one embodiment, the VHH antibody has the amino acid sequence defined in SEQ ID NO: 23. Such a VHH antibody is referred to herein as KB_A01.

[0048] In another embodiment, the VHH antibody has the amino acid sequence defined in SEQ ID NO: 24. Such a VHH antibody is referred to herein as KB_A03.

[0049] In a further embodiment, the VHH antibody has the amino acid sequence defined in SEQ ID NO: 25. Such a VHH antibody is referred to herein as KB_A04.

[0050] In a further embodiment, the VHH has the amino acid sequence set forth in SEQ ID NO: 44. Such a VHH antibody is referred to herein as KB_A06.

[0051] In yet another embodiment, the VHH antibody has the amino acid sequence defined in SEQ ID NO: 26. Such a VHH antibody is referred to herein as KB_A07.

[0052] In one embodiment, the VHH antibody has an amino acid sequence selected from the group consisting of SEQ ID NOs: 23-26 and 44, preferably selected from the group consisting of SEQ ID NOs: 23-26.

[0053] Another aspect of the present invention relates to a VHH antibody that specifically binds to TfR1. The VHH antibody comprises a CDR1 consisting of the amino acid sequence X1X2IX3GSKR as defined in SEQ ID NO:7, where X1 is G or E, X2 is S, D, or I, and X3 is F or N. The VHH antibody also comprises a CDR2 consisting of the amino acid sequence ITX4X5GTT as defined in SEQ ID NO:8, where X4 is Y or V, and X5 is R, H, or G. The VHH further comprises a CDR3 consisting of the amino acid sequence WMFTTX6NY as defined in SEQ ID NO:9, where X6 is D, T, or N.

[0054] This aspect of the invention relates to a family of VHH antibodies that have closely related CDR regions and all bind with high affinity to hTfR1.

[0055] In one embodiment, CDR1 consists of the amino acid sequence X1X2IX3GSKR as defined in SEQ ID NO: 7, where X1 is G or E, X2 is D or I, and X3 is F or N. In this embodiment, CDR2 consists of the amino acid sequence ITX4X5GTT as defined in SEQ ID NO: 8, where X4 is Y or V, and X5 is R, H, or G. Additionally, CDR3 consists of the amino acid sequence WMFTTX6NY as defined in SEQ ID NO: 9, where X6 is D, T, or N.

[0056] In one embodiment, CDR1 consists of the amino acid sequence GDIX3GSKR as defined in SEQ ID NO: 11, wherein X3 is F or N. In this particular embodiment, the CDR2 consists of the amino acid sequence ITVX5GTT as defined in SEQ ID NO: 12, wherein X5 is R or G. Furthermore, the CDR3 consists of the amino acid sequence WMFTTX6NY as defined in SEQ ID NO: 9, wherein X6 is T or N.

[0057] In a preferred embodiment, the CDR1 consists of the amino acid sequence GDINGSKR defined in SEQ ID NO: 13, the CDR2 consists of the amino acid sequence ITVRGTT defined in SEQ ID NO: 14, and the CDR3 consists of the amino acid sequence WMFTTTNY defined in SEQ ID NO: 10.

[0058] In another embodiment, the CDR1 consists of the amino acid sequence GDIFGSKR as defined in SEQ ID NO: 15, the CDR2 consists of the amino acid sequence ITVGGTT as defined in SEQ ID NO: 16, and the CDR3 consists of the amino acid sequence WMFTTNNY as defined in SEQ ID NO: 55.

[0059] In a further preferred embodiment, the CDR1 consists of the amino acid sequence EIINFGSKR as defined in SEQ ID NO: 17, the CDR2 consists of the amino acid sequence ITYHGTT as defined in SEQ ID NO: 18, and the CDR3 consists of the amino acid sequence WMFTTDNY as defined in SEQ ID NO: 5.

[0060] In one embodiment, the VHH antibody is of the formula: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4.

[0061] In one embodiment, the FR1 has the amino acid sequence QVQLQESGGGX7VQAGGSLX8LSCAAS as defined in SEQ ID NO: 19, wherein X7 is S or L and X8 is S or R.

[0062] In one embodiment, the FR2 has the amino acid sequence MGWFRQAPGX9X as defined in SEQ ID NO: 20. 10 RDX 11 VAT, where X9 is E, K, or Q, and X 10 is Q or A, and X 11 is V or L.

[0063] In one embodiment, the FR3 comprises the amino acid sequence X defined in SEQ ID NO:21. 12 YX 13 DSVKGRFTISRDNAX 14NTVYLQMNX 15 LKPEDTAX 16 YYC, wherein X 12 is E or K, and X 13 is A or E, and X 14 is K or N, and X 15 is N or S, and X 16 is V or F.

[0064] In one embodiment, the FR4 has the amino acid sequence WGQGTQVTVSS as defined in SEQ ID NO:22.

[0065] In certain embodiments, the VHH antibody has the amino acid sequences of FR1, FR2, FR3 and FR4 defined in SEQ ID NOs: 19, 20, 21 and 22.

[0066] The present invention also encompasses VHH antibodies having an FR1 amino acid sequence comprising or consisting of SEQ ID NO: 19 or 57, or a variant thereof having at least 88% sequence identity to SEQ ID NO: 19 or 57, preferably at least 92% sequence identity, more preferably at least 96% sequence identity.

[0067] The present invention also encompasses VHH antibodies having an FR2 amino acid sequence comprising or consisting of SEQ ID NO: 20 or 58, or a variant thereof having at least 82% sequence identity to SEQ ID NO: 20 or 58, preferably at least 88% sequence identity, more preferably at least 94% sequence identity.

[0068] The present invention also encompasses VHH antibodies having an FR3 amino acid sequence comprising or consisting of SEQ ID NO: 21 or 59, or a variant thereof having at least 92% sequence identity to SEQ ID NO: 21 or 59, preferably at least 94% sequence identity, more preferably at least 97% sequence identity.

[0069] The present invention also encompasses VHH antibodies having an FR4 amino acid sequence comprising or consisting of SEQ ID NO: 22, or a variant thereof having at least 72% sequence identity to SEQ ID NO: 22, preferably at least 81% sequence identity, more preferably at least 90% sequence identity.

[0070] As used herein, the term "sequence identity %" can be determined using methods well known in the art. For example, sequence identity % is calculated as follows: The query sequence is aligned to the target sequence using the CLUSTAL W algorithm. Comparison is performed over a window corresponding to the shortest of the aligned sequences. The shortest of the aligned sequences may in some cases be the target sequence. In other examples, the query sequence may constitute the shortest of the aligned sequences. The amino acid residues at each position are compared, and the percentage of positions in the query sequence that have the same correspondence in the target sequence is reported as sequence identity %.

[0071] An amino acid sequence having a specified % sequence identity with a reference amino acid sequence is preferably obtained by amino acid substitutions, such as conservative amino acid substitutions, also called conservative amino acid substitutions or mutations, which are amino acid substitutions in an amino acid sequence that change a given amino acid into a different amino acid with similar biochemical, structural and / or chemical properties.

[0072] For example, amino acids can be divided into six major classes based on their structure and the general chemical properties of their side chains (R groups): · Aliphatic: Isoleucine (I), Leucine (L), Glycine (G), Alanine (A), Valine (V); · Hydroxyl or sulfur / selenium containing: serine (S), cysteine ​​(C), threonine (T), methionine (M); Cyclic: Proline (P) · Aromatic: phenylalanine (F), tyrosine (Y), tryptophan (W); Basic: histidine (H), lysine (K), arginine (R); and · Acids and their amides: aspartic acid (D), glutamic acid (E), asparagine (N), glutamine (Q).

[0073] This means that an amino acid sequence having a specified % sequence identity with a reference amino acid sequence is obtained by one or more conservative amino acid substitutions of one or more amino acid residues in the reference amino acid sequence, preferably with a respective amino acid from the same R group as the given amino acid residue, as represented above.

[0074] In one embodiment, the VHH antibody has the amino acid sequence defined in SEQ ID NO: 27. Such a VHH antibody is referred to herein as KB_A09.

[0075] In another embodiment, the VHH antibody has the amino acid sequence defined in SEQ ID NO: 28. Such a VHH antibody is referred to herein as KB_A10.

[0076] In a further embodiment, the VHH antibody has the amino acid sequence defined in SEQ ID NO: 29. Such a VHH antibody is referred to herein as KB_A11.

[0077] In yet another embodiment, the VHH antibody has the amino acid sequence defined in SEQ ID NO: 23. Such a VHH antibody is referred to herein as KB_A01.

[0078] Figure 7 shows a sequence alignment of the closely related VHH antibodies KB_A01, KB_A09, KB_A10, and KB_A11. Table 7 shows the binding kinetics of VHH antibodies KB_A09, KB_A10, and KB_A11 to hTfR1 in the form of dimeric VHH-Fc fusion proteins.

[0079] In one embodiment, the VHH antibody has an amino acid sequence selected from the group consisting of SEQ ID NOs: 23, 27-29, preferably selected from the group consisting of SEQ ID NOs: 27-29.

[0080] In some embodiments, the VHH has an amino acid sequence selected from the group consisting of SEQ ID NOs: 23-29, 44. In one preferred embodiment, the VHH has an amino acid sequence selected from the group consisting of SEQ ID NOs: 23-29.

[0081] The VHH antibodies of the present invention specifically bind to TfR1.

[0082] "Specifically binds to" and similar expressions mean that a molecule of interest, such as a VHH antibody, specifically binds to a target antigen without significantly binding to other molecules. The specificity of an antibody can be determined based on affinity and / or avidity. The equilibrium dissociation constant (K D Affinity, expressed as K, is a measure of the binding strength between an antigenic determinant, i.e., an epitope, and the antigen-binding site of the antibody. D The smaller the value of K, the stronger the binding strength between the antigenic determinant and the antibody. Alternatively, the affinity can be expressed as 1 / K D The equilibrium binding constant (K A As will be apparent to those skilled in the art, affinity can be determined in a manner known per se depending on the particular antigen of interest.

[0083] Typically, antibodies have an equilibrium dissociation constant (K D )10 -5 ~10 -12 moles / liter (M) or less, and preferably 10 -7 ~10 -12 M or less, and more preferably 10 -8 ~10 -12 M, i.e., the affinity constant (K A )10 5 ~10 12 M -1 or more, and preferably 10 7 ~10 12 M -1 More preferably, 10 8 ~10 12 M -1 Generally, 10 -4Any K greater than M D value (or 10 4 M -1 Any K less than A values) are considered to represent nonspecific binding.

[0084] Specific binding of an antibody to an antigen or antigenic determinant can be determined by any suitable method known per se, including, for example, Scatchard analysis and / or competitive binding assays, such as radioimmunoassays (RIA), enzyme immunoassays (EIA) and sandwich competition assays, surface plasmon resonance (SPR), biolayer interferometry (BLI), and various variants thereof known per se in the art.

[0085] The affinity of VHH antibodies for TfR1 needs to be adjusted to be within a certain range to achieve efficient BBB penetration and high relative uptake in the brain. Generally, if the affinity of the antibody for TfR1 is too low, e.g., a K above about 1 μM, D In the case of , the affinity of the antibody is too low to efficiently bind to TfR1 and be taken up by the endothelial cells of the BBB, meaning that most of the antibody remains in the peripheral blood system. Correspondingly, if the affinity of the antibody is too high, e.g., K D If the affinity (K) is significantly lower than 1 nM (<<1 nM), the antibody will be endocytosed but will instead undergo lysosomal degradation in the endothelial cells of the BBB. This suggests that VHH antibodies should have an affinity range (K) of low nM to nM for optimal receptor-mediated endocytosis, i.e., receptor-mediated cellular uptake, and transcytosis, i.e., transport beyond the interior of the cell. D ) (Bien-Ly N, et al., Transferrin receptor (TfR) trafficking determines brain uptake of TfR antibody affinity variants. J Exp Med. 211(2):233-244(2014), WO2016 / 081643).

[0086] As shown in Table 2, the VHH antibody KB_A01 of the invention has an affinity (K D ) Furthermore, K D VHH-containing fusion proteins (see Figure 2), such as dimer-VHH Fc fusion proteins (see Table 4) with a K D When included in a fusion with a 3.2-3.4 nM scFv (see Table 9), the affinity remains in the desired low nM to nM range. Other VHH antibodies of the present invention also have affinities for TfR1 in the low nM to nM range, making them suitable for crossing the BBB by receptor-mediated endocytosis and transcytosis, and can be expressed as dimeric VHH Fc fusion proteins with K D ranges from 0.15 to 5.1 nM, see Table 7.

[0087] In embodiments, the VHH antibodies of the invention specifically bind to human TfR1 (hTfR1).

[0088] In one embodiment, the VHH antibody of the present invention has an affinity (K D In a specific embodiment, the VHH antibodies of the invention in monovalent form specifically bind to hTfR1 at a K selected within the range of 1 to 150 nM, preferably within the range of 1 to 100 nM, more preferably within the range of 1 to 50 nM. D It specifically binds to hTfR1.

[0089] In embodiments, the VHH antibodies of the present invention specifically bind to cTfR1. In certain embodiments, the VHH antibodies of the present invention specifically bind to not only hTfR1 but also cTfR1. See Table 5.

[0090] In embodiments, the VHH antibodies of the present invention do not specifically bind to mTfR1.

[0091] Structurally, hTfR1 is a dimeric transmembrane glycoprotein with a large ectodomain (residues 90-760), an intramembrane region (residues 62-89), and the remaining 61 residues in the cytoplasm. The ectodomain contains three domains: a helical domain (residues 606-760), a protease-like domain (residues 121-183, 384-605), and an apical domain (residues 184-383). The helical domain is involved in receptor dimerization. Transferrin binds to the helical and protease-like domains.

[0092] In embodiments, the VHH antibodies of the invention specifically bind to the apical domain of TfR1, such as hTfR1. Binding of the ectodomain of TfR1 to the apical domain is advantageous compared to binding to the helical or protease-like domain for the purposes of BBB delivery and receptor-mediated transcytosis because these do not interfere with transferrin binding (Daniels et al., The transferrin receptor part I: Biology and targeting with cytotoxic antibodies for the treatment of cancer. Clin Immunol. 2006;121:144-158 (2006); WO 2016 / 081643). Furthermore, binding of the VHH antibodies of the invention to the apical domain of TfR1 does not interfere with holo-transferrin binding to the TfR1 (see Figure 5 and Table 5). Thus, the VHH antibodies of the invention do not interfere with transferrin binding for iron uptake by cells using TfR1. Therefore, it is preferred that the VHH antibody binds to the apical domain of TfR1 in terms of transcytosis ability.

[0093] In an embodiment, the VHH antibody of the invention is a Camelid VHH antibody.

[0094] In another embodiment, the VHH antibody of the present invention is a humanized VHH antibody. For example, the CDR regions of a VHH antibody can be grafted onto a human skeleton. Soler et al., Effect of Humanizing Mutations on the Stability of the Llama Single-Domain Variable Region, Biomolecules 11(2):163(2021), identified several amino acid positions and an N-terminal Gln as hot spots for converting camelid VHHs to human consensus germline sequences.

[0095] Hereinafter, a sequence alignment between the framework regions of KB_A01, the reference human VH (sVH), the universal VHH (uVHH), and the humanized VHH sequence (hVHH) as disclosed in the above Biomolecules article is presented.

[0096] QVQLQESGGGSVQAGGSLLSCAAS KB_A01 FR1 ----VQ----L---P----R------ sVH FR1 ----V---------P----R---T-- uVHH FR1 ----V----L――P----R------ hVHH FR1 MGWFRQAPGEQRDVVAT KB_A01 FR2 -SV-----KGLEW-SP sVH FR2 L--------QE-EA―-A uVHH FR2 L--------QGLEA―-A hVHH FR2 EYADSVKGRFTISRDNAKNTVYLQMNNLKPEDTAVYYC KB_A01 FR3 Y--------------S---L----T-RA--------- sVH FR3 Y--------------------T------------I--- uVHH FR3 Y--------------S---L----S-RA--------- hVHH FR2 WGQGTQVTVSS KB_A01 FR4 -----M----- sVH FR4 ----------- UVHH FR4 -----L----- sVH FR4

[0097] The VHH antibodies of the present invention can be humanized by modification, for example, amino acid substitution, in FR1, FR2, FR3, and / or FR4. The amino acid positions in FR1 to FR4 referred to herein are those in KB_A01 (SEQ ID NO: 23) shown in Figure 7.

[0098] As an example, the humanized positions in FR1 can be selected from E1 or Q1; V5; Q6 or E6, S11 or L11, P14; and / or R19. In one embodiment, any one of these humanized positions in FR1 is used for humanized FR1, or any combination of two to all of these positions is humanized. As an example, humanized FR1 can be based on SEQ ID NO: 57, but with Q1 replaced with E, Q5 replaced with V, and S19 replaced with R, or with two or all of these amino acids replaced.

[0099] The humanized positions in FR2 may be selected from L34 or M34 (the first amino acid position of FR2), S35 or G35 (the second amino acid position of FR2), V37 or F37 (the fourth amino acid position of FR2), Q43 or K43 (the tenth amino acid position of FR2), G44 (the eleventh amino acid position of FR2), L45 (the twelfth amino acid position of FR2), E46 (the thirteenth amino acid position of FR2), W47 or A47 (the fourteenth amino acid position of FR2), S49 (the sixteenth amino acid position of FR2), and / or P50, A50, V50, or G50 (the seventeenth amino acid position of FR2). In one embodiment, any one of these humanized positions in FR2 is used for the humanized FR2, or any combination of two up to all of these positions are humanized. As an illustrative example, the humanized positions in FR2 can be V37, G44, L45 and W47, or F37, G44, L45 and A47.

[0100] The humanized positions in FR3 can be selected from Y58 (the 1st amino acid position of FR3), S74 (the 17th amino acid position of FR3), K75 (the 18th amino acid position of FR3), N76 (the 19th amino acid position of FR3), L78 or I78 (the 21st amino acid position of FR3), N84, S84 or T84 (the 27th amino acid position of FR3), R86 (the 29th amino acid position of FR3), A87 (the 30th amino acid position of FR3), A96 (the 39th amino acid position of FR3), and / or R97 or A97 (the 40th amino acid position of FR3). In one embodiment, any one of these humanized positions in FR3 is used for the humanized FR3, or any combination of two up to all of these positions are humanized.

[0101] The humanized position in FR4 can be L109 or M109 (the sixth amino acid position in FR4).

[0102] I78 also has a stabilizing effect on VHH. Thus, Ile at position 78 stabilizes VHH.

[0103] Example XIV herein produced and tested various humanized VHH antibodies based on the FR region of VHH antibody KB_A01. The positions indicated below as X could be successfully mutated to humanize VHH antibody KB_A01 while maintaining binding to hTfR1.

[0104] FR1 (SEQ ID NO: 78): X1VQLX2ESGGGX3VQX4GGSLX5LSCAAS FR2 (SEQ ID NO: 79): MGWFRQAPGX6X7X8X9VVAT FR3 (SEQ ID NO: 80): EYADSVKGRFTISRDNX 10 KNTX 11 YLQMNX 12 LX 13 PEDTAVYYC FR4 (SEQ ID NO: 81): WGQGTX 14 VTVSS The wild-type FR region of KB_A01 is X1=Q, X2=Q, X3=S, X4=A, X5=S, X6=E, X7=Q, X8=R, X9=D, X 10 =A, X 11 =V, X 12 =N, X 13 = K and X 14 =Q.

[0105] Examples of the humanized VHH antibodies of the present invention include VHH antibodies having the above-described FR1 to FR4 regions (SEQ ID NOs: 78 to 81), and VHH antibodies having the following X1 to X2 regions (SEQ ID NOs: 78 to 81). 14 and a VHH antibody having at least one of X1 to X2 by the wild-type FR region of KB_A01 as described above. 14 Any remaining non-mutated one(s): X1=E, X2=V, X3=L, X4=P, X5=R, X6=K, X7=G, X8=L, X9=E, X 10 =S, X 11 =I,X 12 =S, X 13 = R and X 14=L.

[0106] In one embodiment, a VHH antibody of the invention binds to Protein A. In certain embodiments, a VHH antibody of the invention binds to a Protein A-based resin, such as a Protein A chromatography resin, such as a MabSelect PrismA™ resin.

[0107] Protein A is a 42 kDa surface protein originally found in the cell wall of the bacterium Staphylococcus aureus. It is encoded by the spa gene. It has found use in biochemical research due to its ability to bind immunoglobulins. It is composed of five homologous Ig-binding domains, each folded into a three-helical bundle. Each domain can bind proteins from many mammalian species, most notably IgG. It binds to the heavy chains within the Fc region of most immunoglobulins and, in the case of the human VH3 family, also within the Fab region. Protein A generally does not bind, or only weakly binds, camelid VHH antibodies.

[0108] Henry KA, et al., A Rational Engineering Strategy for Designing Protein A-Binding Camelid Single-Domain Antibodies, PLoS One 11(9):e0163113(2016) discloses methods for generating non-Protein A binding VHH antibodies that can bind to Protein A. See Table 5.

[0109] More particularly, amino acid residue 15 should be G or D, amino acid residue 17 should be S or A, and amino acid residue 19 should be R. The consensus FR1 sequence of SEQ ID NO: 19 is a protein A-binding FR1 region when X8 is R. In a preferred embodiment, the FR1 region of KB_A01 is modified by replacing S at amino acid residue 19 with R to obtain the protein A-binding FR1 sequence shown below and in SEQ ID NO: 82.

[0110] QVQLQESGGGSVQA G G S L S LSCAAS KB_A01 FR1 QVQLQESGGGXVQA G G S L X LSCAAS consensus FR1 ------------------R------ Protein A binding FR1

[0111] Furthermore, amino acid residue 59 should be Y, amino acid residue 64 should be K or E, amino acid residue 65 should be G, amino acid residue 66 should be R, amino acid residue 68 should be T or A, amino acid residue 70 should be S, amino acid residue 75 should be A, E, K, Q, or R, amino acid residue 81 should be Q, amino acid residue 83 should be N, and amino acid residue 84 should be S, N, or G. The FR3 region of KB_A01 is protein A-binding without amino acid substitutions. However, in an optional embodiment, the N at amino acid residue 84 can be replaced with S. The consensus FR3 sequence of SEQ ID NO:21 is X 14 is A, E, K, Q, or R, and X 15 is S, N, or G, it is a protein A-binding FR3 region.

[0112] E Y ADSV KGR F T I S RDNA K NTVYL Q M NN LKPEDTAVYYC KB_A01 FR3 X Y XDSV KGR F T I S RDNA X NTVYL Q M NX LKPEDTAXYYC consensus FR3 -----------------K--------S----------- Protein A binding FR3

[0113] In one embodiment, the FR1 region of KB_A01 is modified by replacing S with R at amino acid residue 19, and the FR3 region of KB_A01 is modified by replacing N with S at amino acid residue 84.

[0114] In one embodiment, the VHH antibody has a FR1 region according to QVQLQESGGGSVQAGGSLRLSCAAS as defined in SEQ ID NO: 82 and a FR3 region according to EYADSVKGRFTISRDNAKNTVYLQMNSLKPEDTAVYYC as defined in SEQ ID NO: 83.

[0115] In one embodiment, the VHH antibody of the invention is an isolated VHH antibody.

[0116] The term "isolated," when used in reference to a VHH antibody, such as in the expression "isolated VHH antibody," means that the VHH antibody has been purified and removed from its original environment. As used herein, an isolated VHH antibody is intended to refer to a VHH antibody that is substantially free of other antibodies with different antigen specificities; for example, an isolated VHH antibody that specifically binds to TfR1, particularly human TfR1 (hTfR1), is substantially free of antibodies that specifically bind to antigens other than TfR1. However, an isolated VHH antibody that specifically binds to hTfR1 may have cross-reactivity to other antigens, such as TfR1 molecules from other species, such as cTfR1. Furthermore, an isolated VHH antibody may be substantially free of other cellular material and / or chemicals. For example, isolated VHH antibodies may be purified to greater than 95% or 99% purity, as determined, for example, by electrophoretic methods (e.g., sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), isoelectric focusing (IEF), capillary electrophoresis) or chromatographic methods (e.g., ion exchange or reverse-phase high performance liquid chromatography (HPLC)). One of skill in the art will understand that isolated VHH antibodies are referred to herein even when the word "isolated" is not explicitly mentioned every time the term "VHH antibody" and similar expressions are used.

[0117] Further aspects of the embodiments include nucleic acid molecules encoding the VHH antibodies of the embodiments or the fusion molecules of the embodiments, see further below. As used herein, a nucleic acid molecule includes polynucleotides, oligonucleotides, and nucleic acid sequences, and generally refers to a polymer of DNA or RNA, which may be single-stranded or double-stranded, and which may contain natural, non-natural, or modified nucleotides, and which may contain natural, non-natural, or modified internucleotide linkages, such as phosphoramidate or phosphorothioate linkages instead of the phosphodiesters found between nucleotides in unmodified oligonucleotides. The term "nucleic acid molecule" also includes complementary DNA (cDNA) and messenger RNA (mRNA).

[0118] In one embodiment, the nucleic acid molecule is an isolated nucleic acid molecule encoding a VHH antibody according to the embodiments, or a fusion molecule according to the embodiments.

[0119] The nucleic acid molecule may encode a single VHH antibody according to the embodiments, multiple copies of a single VHH antibody according to the embodiments, or one or more copies of different VHH antibodies according to the embodiments.

[0120] The nucleic acid molecule may also encode other molecules than the VHH antibodies of the embodiments, for example VHH antibodies genetically fused to another molecule, i.e. in the form of a VHH-containing fusion protein as further described herein.

[0121] Another aspect of the embodiments relates to a vector comprising a nucleic acid molecule according to the embodiments.

[0122] The vector is preferably an expression vector, i.e., a vector comprising at least one nucleic acid molecule comprising a coding sequence capable of being expressed, e.g., transcribed and translated, in a host cell comprising the expression vector. Thus, an expression vector comprises a nucleic acid molecule according to an embodiment operably linked to a promoter.

[0123] As used herein, operably linked means that the nucleic acid molecule is in the correct functional location and / or orientation with respect to the promoter to allow expression of the nucleic acid molecule in a host cell, i.e., the promoter controls transcription of the nucleic acid molecule (constitutively or inducibly).

[0124] In one embodiment, the expression vector is selected from a DNA molecule, an RNA molecule, a plasmid, an episomal plasmid, and a viral vector. In one such embodiment, the expression vector contains, in addition to a nucleic acid sequence encoding the VHH antibody, regulatory sequences necessary for producing the VHH antibody in the host cell. For example, the nucleic acid sequence encoding the VHH antibody is under the transcriptional control of a promoter sequence contained in the expression vector. A promoter is a DNA sequence to which a protein binds and initiates transcription of an RNA molecule from the DNA (gene) downstream. The promoter is preferably selected based on the specific host cell in which the VHH antibody will be expressed, such as T5, T7, lac, or BL21 for expression of VHH antibodies in bacterial cells, GAL1, MET25, CUP1, LAC4, ADH2, SUC2, or GAPDH for expression of VHH antibodies in yeast, and EF1α, CMV, or CAG promoter for expression in mammalian cells such as human cells. The promoter may be a constitutive or inducible promoter. Constitutive promoters, also called constitutively active promoters, are active under all conditions in host cells. Inducible promoters, also called inducible promoters, are regulated and activated in host cells only in response to specific stimuli, such as chemically inducible promoters, temperature-inducible promoters, or light-inducible promoters. Expression vectors can optionally contain other regulatory sequences, such as enhancers. Enhancers are short regions of DNA that can be bound by activators to increase the likelihood of transcription of a particular gene. An optional signal peptide can be provided at the N- or C-terminus of the polypeptide encoded by the expression vector.

[0125] A further aspect of the embodiment relates to a host cell comprising a nucleic acid molecule or an expression vector according to the embodiment.

[0126] The nucleic acid molecule or expression vector can then be transcribed in a host cell to produce the VHH antibody intracellularly.

[0127] In one embodiment, the cell is selected from the group consisting of a bacterial cell, a yeast cell, and a mammalian cell.

[0128] Yet another aspect of this embodiment relates to a VHH antibody according to the present invention linked to at least one molecule. Accordingly, this aspect of the present invention relates to a fusion molecule or protein between a VHH antibody and at least one other molecule. The fusion molecule or protein is preferably in the form of a genetic fusion between the VHH antibody and at least one other molecule. In such cases, the nucleic acid molecule encodes not only the VHH antibody but also at least one other molecule. For example, the nucleic acid molecule may comprise, from its 5'-end to its 3'-end, a nucleic acid sequence encoding a VHH molecule and at least one other molecule, or a nucleic acid sequence encoding at least one other molecule and a nucleic acid sequence encoding a VHH molecule. Thus, a VHH molecule may be covalently linked to at least one other molecule at its N-terminus or C-terminus. See Figure 2.

[0129] Furthermore, it is possible to generate fusion molecules that comprise multiple molecules in addition to a VHH antibody, for example, multiple different molecules or multiple copies of a single molecule may be attached or linked to the N-terminus of a VHH antibody, the C-terminus of a VHH antibody, or at least one molecule attached or linked to the N-terminus of a VHH antibody and at least one molecule attached or linked to the C-terminus of the VHH molecule.

[0130] In one embodiment, the fusion molecule comprises at least one other molecule attached or bound to the C-terminus of the VHH antibody.

[0131] As mentioned above, the VHH antibody is preferably linked to at least one other molecule by genetic fusion to the other molecule. In such embodiments, the nucleic acid molecule encodes the VHH antibody and the at least one other molecule as a VHH-containing fusion protein. This means that the nucleic acid molecule comprises a nucleic acid sequence encoding a VHH antibody connected or linked to a nucleic acid sequence encoding at least one other molecule, and these nucleic acid sequences are under the transcriptional control of the same promoter.

[0132] However, a VHH antibody may be linked, connected, bound, or conjugated to at least one molecule by means other than genetic fusion, for example, by chemically connecting the VHH antibody to at least one molecule. For example, a VHH antibody may be covalently attached to at least one molecule by reaction between a maleimide and an amine or thiol, often referred to as thiol-maleimide or amine-maleimide click chemistry. In one such embodiment, additional amino acid residues such as cysteine, arginine, or lysine may be added to the N- and / or C-terminus of the VHH antibody to enable such reaction with a maleimide.

[0133] The VHH antibody and at least one molecule can be enzymatically linked, for example, by a transglutaminase (TGase) enzyme, which catalyzes the formation of an isopeptide bond between the γ-carboxamide group (-(C=O)NH2) of a glutamine residue side chain and the ε-amino group (-NH2) of a lysine residue side chain, followed by the release of ammonia (NH3). In one embodiment, additional amino acid residues, such as glutamine or lysine, can be added to the N-terminus and / or C-terminus of the VHH antibody to enable such an enzymatic reaction. Site-specific linkage sites can also be prepared via a sortase-mediated reaction (Guimaraes et al., Site-specific C-terminal and internal loop labeling of proteins using sortase-mediated reactions, Nate Protocols 8: 1787-1799 (2013)).

[0134] In one embodiment, the VHH antibody according to the invention is covalently bound to at least one molecule.

[0135] The VHH antibody can be directly linked, e.g., fused, to at least one molecule. Alternatively, the VHH antibody is covalently linked to at least one molecule via a linker. Various such linkers, particularly peptide linkers, can be used according to embodiments, including, but not limited to, G n a linker (wherein n is an integer of 1 or more, and typically an integer of 10 or less), S m Liker (wherein m is an integer of 1 or more, and usually 10 or less), A q linkers (wherein q is an integer equal to or greater than 1, and is usually an integer equal to or less than 10), various GS-linkers or GA-linkers, i.e., combinations of one or more G with one or more S or one or more A, such as (G n S m ) p or (S m G n ) p or (G n A q )p or (A q G n ) p (wherein p is an integer equal to or greater than 1, and is usually an integer equal to or less than 10). For example, linkers such as G4A, G4S, G3S, and combinations thereof, such as G4A-G4A-G4S or G4S-G3S, can be used. Other commonly used peptide linkers are disclosed in Table 1 of Vishnu Priyanka Reddy Chichil, et al., Linkers in the structural biology of protein-protein interactions, Protein Science 22("):153-167 (2013), the linkers listed in Table 1 on pages 156-157 of which are incorporated herein by reference as illustrative, but non-limiting examples of peptide linkers that can be used to covalently link a VHH antibody to at least one molecule.

[0136] The fusion molecules or VHH antibodies of the present invention may comprise one or more tags, including but not limited to affinity purification tags such as His-tag, C-tag, Q-tag, and / or myc-tag. The tags may then be present at the N-terminus and / or C-terminus of the fusion molecule or VHH antibody. The addition of the tags to enable purification may also be accompanied by a protease site, such as but not limited to a TEV site, for efficient subsequent removal of the purification tag from the VHH antibody or fusion molecule.

[0137] The fusion molecule may also comprise multiple VHH antibodies in addition to at least one molecule shown in FIG.

[0138] In one embodiment, the fusion molecule is monovalent with respect to the VHH antibody, i.e., preferably comprises only a single VHH antibody in addition to the last molecule. There is in vitro and in vivo evidence that monovalent binding modes promote transcytosis, whereas bivalent binding modes result in lysosomal sorting (Niewoehner et al., Increased brain penetration and potency of a therapeutic antibody using a monovalent molecular shuttle, Neuron. 81(1):49-60(2014)).

[0139] The at least one molecule can be any molecule, such as a drug or pharmaceutical agent, a diagnostic agent, an imaging agent, a tracer, a half-life extender, etc. Examples of such molecules include, but are not limited to, antibiotics, antivirals, immunomodulators, antitumor agents, anti-inflammatory agents, adjuvants, peptides, polypeptides and proteins, such as enzymes, hormones, neurotrophic factors, neuropeptides, cytokines, apolipoproteins, growth factors, antigens, antibodies or antibody portions, adjuvants, etc., nucleic acids, such as RNA or DNA, including encoding genes, ribozymes, antisense, interfering nucleic acids, complete genomes or portions thereof, inhibitory nucleic acids, such as plasmids.

[0140] In one embodiment, at least one molecule of the fusion molecule is selected from the group consisting of a therapeutic agent and an imaging agent.

[0141] As used herein, an imaging agent refers to an agent or molecule that is administered to a patient and used to visualize the imaging agent during the imaging process, such as when taken up by cells. Examples of such imaging agents include radioactive atoms or isotopes, e.g., radioactive tracers such as position emission tomography (PET) tracers, e.g., 18 F-containing PET tracer, 11 C-containing PET tracer, 64 Radioactive atoms or isotopes such as Cu-containing PET tracers, or e.g. 99m Tc or 111Included are agents or molecules containing single-photon emission computed tomography (SPECT) tracers, such as In-containing SPECT tracers. Other examples of imaging agents include fluorescent probes, luminescent probes, metal complex-containing probes, near-infrared (NIR) fluorescent probes, etc.

[0142] In general, there is an urgent need for diagnostic tools that allow early detection of diseases or medical conditions, particularly in brain compartments, which would facilitate efficient disease-modifying therapies, particularly for assessing therapeutic efficacy in preclinical and clinical trials of new drug candidates.

[0143] Specific targeting of disease-causing proteins, including but not limited to amyloid beta, tau, and alpha-synuclein, or false markers of disease, such as inflammation, is difficult with classical PET ligands or tracers based on small molecules that can cross the BBB. Generally, it is preferable to use PET ligands or tracers based on specific binding molecules, such as antibodies or antibody domains. However, monoclonal antibody (mAb)-based PET tracers generally have a long circulatory half-life, making them unsuitable for any use as PET ligands or tracers. This means that radioactivity generated from radiolabeled mAbs in the cerebral blood circulation interferes with specific signals at target sites in the brain. Therefore, to be useful in PET applications, antibody-based PET ligands or tracers must have a high blood-to-brain ratio and rapid blood clearance. Smaller antibody domains or fragments, such as VHH antibodies (average molecular weight ≤60 kDa), are excreted through the kidneys and are therefore excreted more rapidly than mAbs. Therefore, VHH antibodies are suitable as PET ligands or tracers due to their smaller size compared to mAbs (Syvaenen, et al., A bispecific tribody PET radioligand for visualization of amyloid-beta protofibrils—a new concept for neuroimaging. Neuroimage 148:55-63. (2017); Vandesquille, et al., Chemically-defined camelid antibody bioconjugate for the magnetic resonance imaging of Alzheimer's disease. mAbs 9:1016-1027 (2017)).

[0144] In these embodiments, the fusion molecules can be used as diagnostic agents for the diagnosis of various diseases or disorders.

[0145] For example, the VHH antibody of the embodiments may be linked to another antibody, antigen-binding fragment or domain thereof, including another VHH antibody that specifically binds to a disease-specific marker. In such cases, the other antibody is preferably labeled to detect its binding to the disease-specific marker upon delivery, such as to a brain compartment, by receptor-mediated transcytosis using the VHH antibody as a BBB transporter.

[0146] As used herein, an antigen-binding fragment or domain of an antibody may be selected from the group consisting of a single chain antibody, an Fv fragment, an scFv fragment, an Fab fragment, an F(ab')2 fragment, an Fab' fragment, an Fd fragment, a single domain antibody (sdAb), an scFv-Fc fragment, a di-scFv fragment, and two or more CDRs.

[0147] For therapeutic purposes, the fusion molecule may contain, in addition to the therapeutic agent, a half-life extender or group, also referred to as a stabilizer or stabilizing group. Such a half-life extender is then included to increase the half-life of the fusion molecule when administered to a patient. Any such half-life extender that can be linked to a VHH antibody without adverse biological effects can be used, including, but not limited to, serum proteins such as the Fc fragment of IgG, human serum albumin (HSA), albumin-binding protein scaffolds, IgG, various polyethylene glycol (PEG) molecules, or unstructured polypeptides such as XTEN, or PAS as an alternative to PEGylation.

[0148] The present invention also relates to a pharmaceutical composition comprising the fusion molecules defined above, wherein at least one molecule is a therapeutic agent, and further comprising a pharmaceutically acceptable vehicle or excipient.

[0149] The pharmaceutically acceptable vehicle can be any pharmaceutically acceptable vehicle or carrier that is compatible with the other component(s) of the pharmaceutical composition.The pharmaceutically acceptable vehicle can be selected from conventionally used vehicles according to each mode of administration.The pharmaceutical composition can be a solid pharmaceutical composition such as a tablet, pill, powder, or granule, a semi-solid pharmaceutical composition such as a suppository, or a liquid pharmaceutical composition such as a soft capsule or an injection solution.

[0150] Pharmaceutically acceptable vehicles or excipients include, but are not limited to, diluents such as lactose, dextrose, sucrose, mannitol, sorbitol, cellulose, and glycine; lubricants such as silica, talc, stearic acid (including its salts), and polyethylene glycol; binders such as magnesium and aluminum silicate, starch paste, gelatin, tragacanth, methylcellulose, sodium carboxymethylcellulose, and polyvinylpyrrolidone; disintegrating agents such as starch, agar, alginic acid, and sodium alginate; absorbents, colorants, flavorings, sweeteners; and liquid vehicles such as polypropylene glycol, water, saline, aqueous dextrose, glycerol, ethanol, and oils.

[0151] Embodiments also relate to fusion molecules according to the above for use as a medicament, wherein at least one molecule is a therapeutic molecule.

[0152] In one embodiment, the therapeutic agent is capable of treating a disease or disorder of the CNS.

[0153] In certain embodiments, the fusion molecules described above are for use in treating a CNS disease or disorder. In such embodiments, at least one molecule is a therapeutic agent capable of treating a CNS disease or disorder.

[0154] A further embodiment is directed to a method for treating a CNS disease or disorder in a patient. The method comprises administering to the patient an effective amount of a fusion molecule according to the above or a pharmaceutical composition according to the above. In such an embodiment, at least one molecule is a therapeutic agent capable of treating a CNS disease or disorder.

[0155] Illustrative, but non-limiting examples of diseases or disorders of the CNS include Alzheimer's disease (AD), Bell's palsy, cerebral palsy, epilepsy, motor neuron diseases (MND) such as amyotrophic lateral sclerosis (ALS), progressive bulbar palsy (PBP), pseudobulbar palsy, progressive muscular atrophy (PMA), primary lateral sclerosis (PLS), spinal muscular atrophy (SMA) and monomeric muscular atrophy (MMA), multiple sclerosis (MS), neurofibromatosis, Parkinson's disease (PD), lysosomal storage diseases, neuropathic lysosomal storage diseases, ischemic stroke, intracerebral hemorrhage, traumatic brain injury (TBI), vascular dementia, frontotemporal dementia, amyloidosis, tauopathy, Creutzfeldt-Jakob disease, neuroinflammation and neuropathic pain.

[0156] Malignant cells often overexpress TfR1, and increased expression can be associated with poor prognosis in different types of cancer (Candelaria et al., Antibodies Targeting the Transferrin Receptor 1 (TfR1) as Direct Anti-cancer Agents. Front. Immunol. 12: 607692(2021)). TfR1 is overexpressed in many different types of cancer cells, often at levels several times higher than in normal cells. Indeed, TfR1 has been identified as a universal cancer marker. Increased expression of TfR1 correlates with advanced stage and / or poor prognosis in several cancers, including solid tumors such as esophageal squamous cell carcinoma, breast cancer, ovarian cancer, lung cancer, cervical cancer, bladder cancer, osteosarcoma, pancreatic cancer, cholangiocarcinoma, renal cell carcinoma, hepatocellular carcinoma, adrenocortical carcinoma, glioblastoma multiforme (GBM), and cancers of the nervous system, including brain metastases of peripheral cancers, as well as hematopoietic malignancies, such as acute lymphoblastic leukemia (ALL), chronic lymphocytic leukemia (CLL), and non-Hodgkin's lymphoma (NHL) (Candelaria et al., Antibodies Targeting the Transferrin Receptor 1 (TfR1) as Direct Anti-cancer Agents. Front. Immunol. 12:607692 (2021); Ramalho et al., Transferrin Receptor-Targeted Nanocarriers: Overcoming Barriers to Treat Glioblastoma. Pharmaceutics 14(2): 279(2022)).

[0157] The elevated expression levels of TfR1 on malignant cells, along with its extracellular accessibility, ability to be internalized, and central role in cancer cell pathology, make this receptor an attractive target for antibody-mediated therapy. TfR1 can be targeted by the VHH antibodies of the present invention for cancer treatment by using VHH antibodies conjugated to anticancer, immunotherapeutic, and / or adjuvant therapeutic agents that are internalized by receptor-mediated endocytosis.

[0158] In another embodiment, the therapeutic agent is capable of treating cancer.

[0159] In certain embodiments, the fusion molecules according to the above are for use in the treatment of cancer, in which at least one molecule is a therapeutic agent capable of treating cancer.

[0160] A further embodiment is directed to a method of treating cancer in a patient, the method comprising administering to the patient an effective amount of a fusion molecule according to the above, or a pharmaceutical composition according to the above, in which at least one molecule is a therapeutic agent capable of treating cancer.

[0161] hTfR1 is also a target for delivering therapeutic agents, including oligonucleotides, to muscle (Desjardins et al., Enhanced exon skipping and prolonged dystrophin restoration achieved by TfR1-targeted delivery of antisense oligonucleotides using FORCE TM (Sugo et al., Development of antibody-siRNA conjugate targeted to cardiac and skeletal muscles, J Control Release 237: 1-13(2016)). Therefore, the fusion molecules of the present invention can be used to treat various muscle diseases, particularly muscular dystrophies.

[0162] The fusion molecules could also be used to treat muscular dystrophies such as Duchenne muscular dystrophy (DMD), Becker muscular dystrophy (BMD), Mitton muscular dystrophy (DM), limb-girdle muscular dystrophy (LGMD), facioscapulohumeral muscular dystrophy (FSHD), congenital muscular dystrophy (CMD), distal muscular dystrophy (DD), oculopharyngeal muscular dystrophy (OPMD), or Emery-Dreifuss muscular dystrophy (EDMD), particularly DMD, FSHD, or DM.

[0163] In a further embodiment, the therapeutic agent can affect muscular dystrophy.

[0164] In certain embodiments, the fusion molecules according to the above are for use in treating muscular dystrophy, in which at least one molecule is a therapeutic agent capable of treating muscular dystrophy.

[0165] A further embodiment is directed to a method of treating muscular dystrophy in a patient, comprising administering to the patient an effective amount of a fusion molecule according to the above, or a pharmaceutical composition according to the above, wherein at least one molecule is a therapeutic agent capable of treating muscular dystrophy.

[0166] As used herein, an effective amount refers to an amount effective at a dosage and for a period of time necessary to achieve a desired therapeutic result. An effective amount may vary depending on factors such as the state of the disease, age, sex, and weight of the patient. As used herein and as is well understood in the art, treating or treatment refers to an approach for obtaining beneficial or desired results, such as clinical results. Beneficial or desired clinical results may include, for example, alleviation or amelioration of one or more symptoms or conditions, reduction in the extent of disease, stable disease, i.e., prevention of disease worsening, prevention of disease spread, delay or slowing of disease progression, improvement or palliation of the condition, reduction in disease recurrence, and remission. Treating or treatment may also prolong survival compared to the expected survival without treatment.

[0167] As used herein, and as well understood in the art, prevention or prophylaxis refers to an approach in which the risk of developing a disease or condition is reduced or prevented, including prolonging or delaying the onset of the disease. For example, a patient who is predisposed to developing a disease, such as due to a genetic or inherited predisposition, may benefit from administration of a fusion molecule or pharmaceutical composition to prevent, reduce the risk of, delay and / or delay the onset of the disease.

[0168] The patient is preferably a human patient, however, embodiments may also be applied to veterinary uses, i.e., non-human patients, such as non-human mammals, including primates, monkeys, apes, cows, sheep, pigs, goats, horses, cats, dogs, mice, rats, and guinea pigs.

[0169] Fusion molecules or pharmaceutical compositions according to embodiments can be administered to patients by a variety of routes, including, for example, parenteral routes such as subcutaneous, intravenous, intraperitoneal, or intramuscular injection, oral, rectal, topical, intranasal, perilingual, or intraocular routes. In specific embodiments, the route of administration is subcutaneous or intravenous. [Example]

[0170] Example I - Protein production for llama immunization and assay The extracellular domains of hTfR1 (amino acids R121 to F760, SEQ ID NO: 30), mTfR1 (S122 to F763, SEQ ID NO: 31), and cTfR1 (R121 to F760, SEQ ID NO: 32) were cloned into the pCDNA3.4-TOPO plasmid optimized for expression in Critulus griseus using 5' cloning sites: HindIII, KpnI, and NheI, and 3' cloning sites: XhoI and PacI, with stop codons TGA, TGA, and a Kozak sequence (generated with the Geneart tool (Thermo Fischer Scientific)). An N-terminal signal peptide (MDWLRNLLFLMAAAQSINA, SEQ ID NO: 33) followed by a single amino acid linker "A", a histidine tag (6xHis, SEQ ID NO: 38), and an SG linker (hTfR1, cTfR1) or G linker (mTfR1) were used for expression. Plasmids were ordered from Geneart (Thermo Fischer Scientific, Germany). Transient transfection of Chinese hamster ovary (CHO) cells was performed according to standard procedures. Cultures were harvested on day 6. The supernatant was filtered (0.45 μm syringe filter) and then purified by immobilized metal ion chelate (IMAC) using a HisTrapExcel 5 mL column (Cytiva). The following buffers were used: equilibration / wash (50 mM phosphate, 500 mM NaCl, pH 8), primary wash buffer (50 mM phosphate, 500 mM NaCl, 20 mM imidazole, pH 8), and elution buffer (50 mM phosphate, 500 mM, 500 mM NaCl, imidazole, pH 8). The eluted fraction was concentrated using an Amicon Ultra 10000 MWCO spin concentrator. The protein was dialyzed overnight with 10% glycerol (Medicago tablets, glycerol diluted from 80% stock solution) in PBS with a 10 kDa cutoff (Pierce) at 4°C (3 ml cassette slide alyzer). PBS was set to pH 8. Protein concentration was measured using absorbance at 280 nm.

[0171] Example II - Generation and selection of a VHH library that binds to human TfR1 Immunizations for antibody generation in llama (Lama glama) animals (N=2) were performed at Modiquest Research BV (The Netherlands) according to the ModiPhage™ method. Based on monitoring the specific immune response to the immunogens hTfR1 and mTfR1 (produced as described in Example I), primary immunizations were performed in two llama animals using an adaptive immunization protocol. Immunizations were administered intramuscularly (i.m.) with a dose of 500 μg of protein on days 1 (hTfR1), 21 (mTfR1), and 42 (hTfR1), as shown in Figure 1 . Blood was collected on day 54, and the immune response was tested by ELISA. Serum samples were serially diluted in 3-fold steps (1:100 to 1:72900) and screened against plate-immobilized hTfR1, mTfR1, and the control protein bovine serum albumin (BSA) (all at 1 μg / ml). Specific binding of anti-llama IgG1, IgG2 / IgG3 antibodies was detected with goat anti-rabbit horseradish peroxidase (HRP) and goat anti-mouse HRP, respectively, and detected with 3,3',5,5'-tetramethylbenzidine (TMB) substrate at 480 nm absorbance in a spectrophotometric plate reader. Two additional doses with 500 μg of hTfR1 were delivered on days 86 and 107. Blood was collected on day 117, and ELISA was performed as described above. One animal was selected for the final collection and also received an additional boost dose with a 1:1 mixture of hTfR1 and mTfR1 (250 μg + 250 μg) on ​​day 120. Four days later (day 124), approximately 1.5 × 10 9 Peripheral blood was collected for isolation of PBLs by density gradient centrifugation using TRIzol (Thermo Fischer Scientific). Approximately 0.7 × 10 9 RNA extraction was performed from PBLs, and reverse transcription to cDNA was performed by oligo(dT)-primed reverse transcription of 400 μg of freshly isolated RNA.

[0172] All subsequent steps for constructing a phage display library of VHHs were carried out as described in Pardon et al., A general protocol for the generation of Nanobodies for structural biology. Nat Protoc 9:674-693 (2014). See Figure 1B. More specifically, the above cDNA was used as a template in PCR reactions to amplify the IgG2 / 3 repertoire. A first PCR reaction was performed to amplify all antibodies (IgG1 and IgG2 / 3), followed by a second nested PCR reaction using primers CALL001 (SEQ ID NO: 34) and CALL002 (SEQ ID NO: 35) to amplify and isolate the VHH repertoire, including overhanging regions for subcloning into the phagemid pADL10b. A 3 μg quantity of cDNA template was used in a High Fidelity® kit from ThermoFischer Scientific: 10 μl of 5× High Fidelity buffer, 1 μl of dNTPs (dinucleotide triphosphates; each 10 mM, 0.2 mM final concentration), 25 pmol per primer (0.5 μM final primer concentration), 0.5 μl of Phusion® polymerase (1 U / reaction), 50 μl of ddHO (double-distilled water), 1-2 μl of template, and a cycling program of 98°C for 5 min, 98°C for 30 s (30×), 72°C for 30 s (30×), 72°C for 10 min, and storage at 4°C. A total of 36 wells were pooled and then purified to remove PCR dimers using Genejet Enzyme Cleanup (ThermoFischer, catalog no. K0832). The mixture was purified using four columns and eluted with 4 x 20 μl of elution buffer. The eluate was loaded onto a 1% agarose gel with SYBRSafe DNA GelStain (Sigma-Aldrich) for gel extraction, and the extracted product was used as a template in the subsequent PCR reaction. Next, a nested PCR reaction containing the pADL10 cloning site was used for amplification using the following oligonucleotides: VHH-R-SfiI (SEQ ID NO: 36) and VHH-F-SfiI (SEQ ID NO: 37).Two PCR mixtures were prepared with 0.5 μM each primer (VHH-F-SfiI and VHH-R-SfiI) and 0.1 μM each primer, and thermocycling was performed with 1 μl of template as follows: 98°C for 5 min, 98°C for 30 s (20x), 72°C for 1 min (20x), 72°C for 10 min, and storage at 4°C. Approximately 50 ng of amplified and gel-extracted heavy chain antibody was used as template in each well. After amplification, all wells were pooled and purified using Genejet PCR cleanup according to the manufacturer's protocol, followed by elution with 10 mM Tris-HCl pH 8.5. The PCR product concentration was measured using a Nanodrop (result: 160 ng / μl). The product was digested with SfiI restriction enzyme (Thermo Fischer Scientific). Using the Genejet gel extract, a 400 bp DNA fragment was extracted on a 2% agarose gel.

[0173] The pADL10b vector was digested with SfiI restriction enzyme (Thermo Fischer Scientific) overnight at 37°C according to the manufacturer's instructions. The digested vector was purified on a 1% agarose gel and visualized using Sybr Safe (Thermo Fischer Scientific). This procedure was repeated to a final concentration of 154.6 ng / μl.

[0174] After PCR amplification, the library was digested (using Sfil) at 50°C for 2 hours. The enzyme was removed using a Genejet PCR cleanup column as described above. The resulting material was eluted with 20 μl of elution buffer (10 mM Tris-HCl pH 8.5). The DNA concentration was measured to be 73 ng / μl. The mixture was then religated. First, the material was heated at 50°C, slowly cooled to room temperature (RT, 20-25°C), and then incubated overnight at 16°C with shaking. Cleanup was performed as described above and eluted with 10 mM Tris-HCl pH 8.5. A total of 1600 ng of DNA was used. 1 or 2 μl of the ligation mixture was electroporated into 10 μl of TG1 E. coli in a total reaction volume of 25 μl (5.48 ms / 1.92 kV and 5.48 ms / 1.92 kV, respectively). Cells were removed from the -70°C freezer and thawed on ice immediately before use. Transformants were distributed onto large agar plates (20 cm) and grown under standard conditions until colonies were present and ready to be picked.

[0175] The cells were harvested by adding Luria Broth (LB) medium. Colonies were counted, and the library size was 3.2 × 10 8 After overnight growth on agar plates, cultures were prepared from the final library pool. The cultures were shaken at 200 revolutions per minute (rpm) for 2 hours at 37°C. Hyperphage was added at a multiplicity of infection (MOI) of 2 for 30 minutes at 37°C without shaking. The medium was removed by centrifugation at 4500g for 5 minutes in a Falcon tube. The cells were then resuspended in an equal volume of Terrific Broth (TB) medium supplemented with 100 μg / ml ampicillin, 50 μg / ml kanamycin, and 1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG). The cultures were shaken at 200 rpm overnight at 30°C.

[0176] A 5 mL culture was prepared in LB from an overnight culture of the selected clone in a 15 mL Falcon tube. It was supplemented with 100 μg / mL ampicillin and glucose was omitted. The overnight culture was diluted 1:100 into 5 mL of LB and grown for 2 hours at 150 rpm until the culture became turbid. 1 μL of Hyperphage helper phage (Hyperphage, titer: 1.4 × 10) was added. 12 cfu / ml (Progen) was added to each culture. Helper phage was allowed to infect for 30 minutes at 37°C. The phage-containing LB was discarded by centrifugation at 4500g. The bacterial pellet was resuspended in 5 ml of TB medium supplemented with 100 μg / ml ampicillin, 50 μg / ml kanamycin, and 1 mM IPTG. The culture was shaken overnight at 200 rpm at 30°C. The culture was centrifuged at 4500g for 4 minutes to remove bacteria, and 0.25×V volume of 20% PEG8000 / 2.5M NaCl was added and the centrifuge tube was incubated on ice for 2 hours. The phage / bacteria were pelleted at 15000g for 10 minutes and resuspended in 0.5 ml of PBS. The phage-containing supernatant was transferred to a separate microcentrifuge tube preblocked with PBS / casein.

[0177] Panning with Dynabeads® using biotinylated TfR1 Biotinylation of hTfR1, cTfR1, and mTfR1 was performed according to standard procedures using the Biotinylation Kit EZ-link Sulfo NHS PEG4 (Thermo Fisher Scientific) with a mixture of 1 mg of target protein and 2 mM biotin at a 1:1 molar ratio at pH 8. Free biotin was removed by dialysis against 1 L of PBS buffer, pH 8, for 3 x 1 h at 4 °C in a Pierce 3 mL 10 kDa MWCO (molecular weight cut-off) dialysis unit. Protein was concentrated to 850 μl using an Amicon spin concentrate with a 10 kDa MWCO. Phage supernatants were preblocked with PBS / casein and mixed with 100 nM biotinylated hTfR1, cTfR1, or mTfR1 and incubated for 1 h. Bead blocking was performed with 50 μl of Streptavidin Dynabeads® (M280, Invitrogen, Thermo Fischer Scientific) using hTfR1 (1.1 mg / mL) and mTfR1 (1 mg / mL) and washed with PBS containing 0.15% Tween® 20 (PBST). The beads were resuspended in 1 ml of ELISA blocker (PBS / casein ELISA blocker (Thermo Fischer Scientific)) and incubated at room temperature for 30 minutes on a rotating mixer. The blocking was replaced with the phage + target mixture and incubated at room temperature for 30 minutes with gentle rotation to capture biotinylated TfR1 on the beads. The beads were washed five times with PBS. Both mTfR1 and hTfR1 captures were eluted with 400 μl of PBS using 0.25 mg / mL trypsin.

[0178] Phages were trypsinized for 30 minutes at room temperature before titration and rescue. TG1 E. coli cultures were initiated by inoculating a 10 ml culture in 10 ml LB medium with a single colony of TG1 grown on an M9 minimal agar plate. The culture was incubated at OD 600 until the OD reaches the log phase (approximately 4 hours). 600The phage input was trypsinized by mixing 75 μl of phage stock with 25 μl of TG1 (1 mg / ml trypsin) and incubated at room temperature for 30 minutes before dilution and titration. The remaining phage output was mixed with 2 ml of TG1 E. coli and incubated at room temperature for 5 minutes without shaking. The mixtures were plated onto two large LB-Lennox agar plates (Merck-Sigma-Aldrich) and allowed to dry completely before being incubated at 37°C.

[0179] The procedure for selection (and subsequent rescue) with capture Dynabeads® was repeated in three subsequent rounds using either hTfR1, cTfR1, or mTfR1 at 100 nM, 25 nM, 10 nM, and 5 nM, respectively, according to Table 1 below (the last two rounds in the presence of 1 μM human transferrin (recombinant human transferrin made in rice, Sigma-Aldrich)). After the fourth selection round, 25 colonies representing each track and elution method (trypsin or citrate / phosphate buffer, pH 5.5, containing 0.15% Tween® 20) were randomly picked from the agar plates and sent for sequencing with primer pADL 10fwd at the Karolinska Institutet (Sweden) Genetic Analysis Core Facility.

[0180] [Table 1]

[0181] Example III - Binding with ELISA to select VHHs that bind to hTfR1 and cTfR1 Prior to ELISA, selected phage samples were diluted 1:10 in blocking buffer (casein ELISA blocker) containing 20% ​​v / v PEG8000 in 2.5 M NaCl. Targets (hTfR1, cTfR1, mTfR1) were coated onto 96-well untreated Greiner plates (100 μl) at 1 μg / ml in PBS, pH 7.4, overnight at 4°C. The coating buffer was discarded and replaced with 100 μl of blocking solution. The plates were covered and incubated at room temperature for 2 hours at 900 rpm. Anti-human TfR1 antibody BA3 (MEM189, Sigma, 1 mg / ml) and anti-mTfR1 (8D3, ​​Novus biologicals, 1 mg / ml) were diluted in blocking buffer to a starting concentration of 100 nM and serially diluted 1:2 by transferring 100 μl of blocking buffer. The plates were incubated at 900 rpm at room temperature for 1 hour, and the wells were washed 3 times with 300 μl of PBST (PBS containing 0.1% Tween® 20) using a 50TS microplate washer (Biotek). 100 μl of anti-M13 antibody (anti-PVIII GE Healthcare) diluted 1:4000 in blocking buffer was added to the target-coated wells, followed by incubation at 900 rpm at room temperature for 1 hour. Washing with PBST followed. Secondary anti-mouse antibody (Sigma) 1:4000 in blocking buffer was incubated at 900 rpm at room temperature for 1 hour. After washing in PBST, the color was developed with TMB substrate. The reaction was stopped by adding 2 M H2SO4 to each well, and absorbance readings were taken at 450 nm using a SpectraMax 3000 (Molecular Devices). One clone (KB_A01) used in the following examples was selected based on cross-reactive binding in ELISA to both cTfR1 and hTfR1, but not mTfR1. The selection criteria were >10-fold reactivity to both human and cynomolgus receptors compared to the blank (negative control). Selection criteria in this workflow were also based on non-competition with Tf for the Tf binding site (see Example II), which is an unfavorable binding site for brain transport of therapeutic agents.

[0182] Example IV - Production of free VHH or fusion variants The selected VHH clone (KB_A01, SEQ ID NO: 23) was produced as a single VHH unit (12-14 kDa) (see Figure 2). Following a C-terminal 6xHis tag (SEQ ID NO: 38), a "C-tag" (C-terminal amino acids EPEA, SEQ ID NO: 39) was ordered and codon-optimized from Genscript Inc. using the expression vector pET-22b(+). The protein was obtained from the periplasmic space of Escherichia coli and purified using a one-step purification on a Ni-column.

[0183] A human Fc region was genetically fused to a VHH using a GG spacer to generate a bivalent, TfR1-binding functional entity (see Figure 2). More specifically, a VHH-Fc fusion protein (SEQ ID NO: 41) between human IgG1 Fc (SEQ ID NO: 40) and KB_A01 (SEQ ID NO: 23), containing a signal peptide from the mouse IgG kappa light chain (SEQ ID NO: 61), was cloned into the vector pcDNA3.4 backbone containing codon-optimized sequences (GeneArt, Thermo Fischer Scientific) and transiently expressed in 200 mL cultures of Expi293™ cells (Thermo Fischer Scientific), grown under standard conditions, and harvested 6 days post-transfection. Purification of the culture medium was performed using MabSelect Sure™ and polished on HiLoad Superdex 200, yielding a fraction of dimeric VHH-Fc fusion with a purity of >99% as verified by analytical size-exclusion chromatography (SEC). The same process was followed to generate a fusion protein for the reference VHH (SEQ ID NO: 60) and human IgG1 Fc as disclosed in WO2020 / 144233. This reference VHH-containing fusion protein is referred to herein as BV.

[0184] Following bioinformatics evaluation, additional VHH-Fc fusions were made between human IgG1 Fc and KB_A02 (SEQ ID NO: 42), KB_A03 (SEQ ID NO: 24), KB_A04 (SEQ ID NO: 25), KB_A05 (SEQ ID NO: 43), KB_A06 (SEQ ID NO: 44), KB_A07 (SEQ ID NO: 26), KB_A08 (SEQ ID NO: 45), KB_A09 (SEQ ID NO: 27), KB_A10 (SEQ ID NO: 28), and KB_A11 (SEQ ID NO: 29). These additional VHH-Fc fusions were generated in the Turbo-CHO™ high performance platform system after sequence optimization and cloned into proprietary plasmids from GenScript Biotech Corporation in a volume of 4 mL, based on the same human IgG1 Fc and GG linker sequence and signal peptide (SEQ ID NO: 56) as above. Here, the final sample was also taken 6 days post-transfection and purified with a protein A-based system in Genscript, yielding 95–99% purity as verified by either SDS-PAGE or analytical size-exclusion chromatography (SEC).

[0185] Functional fusion proteins were produced in the form of fusions between VHH and single-chain variable domains (scFv). Such fusion proteins have monomeric TfR1 binding but have the potential for bispecific binding due to the presence of scFv domains that can bind to targets different from TfR1 or different epitopes on the same target. The amyloid beta-binding antibody 3D6 was selected as a functional representative scFv unit. The scFv of the 3D6 antibody was composed of a VH (SEQ ID NO: 46) and a VL (SEQ ID NO: 47) fused to either the N- or C-terminus of KB_A01 with a 3×(G4S) linker (SEQ ID NO: 48), resulting in KB_A01-scFv (SEQ ID NO: 49) or scFv-KB_A01 (SEQ ID NO: 50). See Figure 2. The gene constructs were optimized for expression in CHO cells by GenScript Biotech Corporation and transiently transfected and expressed in TurboCHO™ high-performance platform cells.

[0186] Reference antibody 128.1 (BA1) was expressed as full-length human IgG1 (SEQ ID NO: 51) in transiently transfected FreeStyle™-293 cells using the vector pcDNA3.4, recovered from the culture medium, and purified in a single step on Protein A resin. This reference antibody, BA1, binds to a region of the extracellular portion of hTfR1 and has the ability to cross the BBB (WO 93 / 10819; Friden PM, et al., Characterization, receptor mapping, and blood-brain barrier transcytosis of antibodies to the human transferrin receptor. J Pharmacol Exp Ther 278:1491-1498 (1996)).

[0187] Reference antibody JCR-IgG (BA2), a clinically validated CNS-targeting TfR1-binding antibody, was designed based on information from US 9,994,641 as an IgG1 antibody that reacts with hTfR1 through its Fab'2 (SEQ ID NO: 51). The heavy chain (HC) and light chain (LC) sequences were optimized, cloned into separate plasmids, and expressed in transiently transfected ExpiCHO™ cells using standard reagents and harvested on day 8. Proteins were purified using MabSelect Sure™ (Cytiva) followed by Superdex 16 / 60 (Cytiva) and analyzed according to standard procedures known to those skilled in the art.

[0188] Binding properties of Example V-KB_A01 The binding of VHH KB_A01 to hTfR1 and cTfR1 was identified during phage display panning toward hTfR1 and cTfR1, and the reactivity of the clone was confirmed by ELISA. The binding kinetics was further analyzed by surface plasmon resonance (SPR) and biolayer interferometry (BLI). VHH KB_A01 was tested for binding affinity to TfR1 and for the lack of interference with affinity by transferrin (Tf). SPR analysis was performed on a Biacore T200 or Biacore 8K instrument (Cytiva, Sweden). SPR Series S Chip SA and Series S CM5 chips, or Series S Protein A chips (Cytiva, Uppsala, Sweden) were used. Biotinylated transferrin was bound to the SA chip.

[0189] The binding characteristics of monomeric VHH KB_A01 and reference antibody BA1 were evaluated by SPR on a Biacore T200 instrument using single-cycle kinetics. Human Tf was amine-coupled (to approximately 3000 RU) on a CM5 chip, and then hTfR1 was captured on immobilized human transferrin. Human TfR1 was injected at 100 nM, followed by injections of increasing concentrations of analyte (VHH KB_A01 or BA1) ranging from 6.25 to 100 nM in two-fold increments. The contact time was 150 s for each concentration at 30 μL / s, with a final dissociation step of 500 s. At the end of each single-cycle run, the hTf surface was fully regenerated by injecting 10 mM glycine pH 2 for 30 s. Similarly, mTfR1 was amine-coupled to approximately 2200 RU on a CM5 chip. The same concentration range of analyte (VHH KB_A01 or BA1) was applied to human TfR1 (6.25-100 nM, 2-fold increments), and the above SPR cycles without the loading step were applied to these surfaces. At the end of each single-cycle run, the mTfR1 surface was fully regenerated by injecting 10 mM NaOH for 30 seconds. A blank run was performed after each single-cycle kinetic run and used for baseline correction. The association rate constant (ka), dissociation rate constant (kd), and equilibrium dissociation constant (K) were measured.D ) were evaluated using the Biacore T200 software evaluation tool, applying the Langmuir 1:1 model.

[0190] Monomeric KB_A01 and the reference antibody BA1 bound to hTfR1 (Table 2 and Figure 3), but neither of the two antibodies bound to mTfR1. Because hTfR1 was immobilized by loading onto hTf, analyte binding meant that neither KB_A01 nor BA1 competed with hTf for binding to hTfR1. Furthermore, a comparison of KB_A01 and BA1 in terms of affinity and binding kinetics showed that monomeric (free) KB_A01 bound to the bivalent BA1 (K D = 0.0199 nM) compared to a 206-fold lower affinity (K D ) bound to the precomplex hTfR1-Tf at a significantly higher affinity than BA1, which was due to a dramatic difference in dissociation rate (kd). It has been shown that too high an affinity can result in retention of VHHs and any proteins fused to them within the vasculature and / or lysosomal degradation of TfR1-anti-TfR1 complexes (Bien-Ly N, et al., Transferrin receptor (TfR) trafficking determines brain uptake of TfR antibody affinity variants. J Exp Med. 211(2):233-244(2014)). Therefore, VHH KB_A01 has lower affinity for hTfR1 compared to BA1, making it more suitable as a drug transporter.

[0191] [Table 2]

[0192] Example VI - Mapping the epitope of monomer KB_A01 via binning experiments against known hTfR1 binding antibodies BA1, BA2, and BA3 by BLI Recombinant KB_A01 was biotinylated using the EZ-Link NHS-PEG4-Biotin Biotinylation Kit (ThermoFisher) according to the manufacturer's instructions. A vial of biotin stock solution was dissolved at a concentration of 20 nM in DMSO and stored at -70 °C. It was thawed, freshly diluted to 2 mM in ddH2O, and mixed with VHH protein (12 nmol to 57 nmol protein) at an equimolar ratio with biotin in PBS pH 7.4. PBS was added to a total volume of 0.5 or 1 mL. The mixture was incubated in a rotor mixer at room temperature for 30 min, followed by incubation on a shaker at 4 °C overnight. Samples were concentrated using a 500 μL spin concentrator (13,000 rpm for 5 min, 3,000 MWCO) and repeated until the concentration was sufficient for the particular sample. Samples were then desalted using a NAP-5 column equilibrated with PBS pH 7.4 (Medicago) according to the manufacturer's instructions. Briefly, samples were loaded onto the resin, and 400 μL of PBS was added. Protein was eluted by adding 500 μL of PBS. Protein concentration was estimated using absorbance at 280 (lambda). Yields were calculated for the samples and ranged from 24% to 46%.

[0193] Analysis of hTfR1 binding and competition between KB_A01 and reference antibodies BA1, BA2, and BA3 was performed using BLI on an OctetRED96 instrument (ForteBio / Pall Life Sciences). Samples were loaded into a 96-well black polypropylene microplate (Greiner Bio catalog no. 655209). First, biotinylated KB_A01 (as free VHH protein, as described above) was immobilized at 2 μM on a Streptavidin Dip and Read™ biosensor (Pall Life Sciences, ForteBio) for kinetic analysis. hTfR1 protein was added to each sensor at 50 nM, followed by regeneration with glycine pH 2 (5 s, followed by four 5 s regenerations in run buffer [Kinetics Buffer 10X (ForteBio) diluted with PBS pH 7.4]), repeated for four cycles. In the final cycle, competitor protein probes KB_A01 (control), BA1, and BA3 (purchased from Sigma-Aldrich / Merck) were added at 10 nM, and binding was recorded. The parameters for kinetic measurements were: baseline 100 s, loading 300 s, quenching 1200 s, association 300 s, and dissociation 200 s between steps. Data were recorded and analyzed using Octet System Data Acquisition, Release 10.0 (ForteBio, Pall Life Sciences) and Octet System Data Analysis, Release 10.0 - Kinetics Module (ForteBio, Pall Life Sciences). The resulting binding of biotinylated monomer KB_A01 is shown in the first row of Table 3.

[0194] Next, full-length reference antibodies BA1, BA2, and BA3 were immobilized at 1 μM on kinetic protein A (ProA) dip and read biosensors (Pall Life Sciences, Fortebio). hTfR1 (50 nM) was added to each biosensor according to the same principle as above, followed by the addition of a competitive probe, VHH KB_A01 (not biotinylated), or antibodies BA1 or BA3 (all at 10 nM). Due to the loading configuration on protein A, it was not possible to compare the monoclonal antibodies with each other (indicated as nd = not performed in Table 3). As shown in Table 3 (rows 2–4), KB_A01 blocked BA1 from binding to hTfR1. In contrast, KB_A01 and reference antibody BA2, as well as KB_A01 and BA3, were able to simultaneously bind to hTfR1. Thus, KB_A01 binds to a specific epitope that overlaps but is not identical to BA1. Antibody BA1 and antibody BA3 are known to bind to overlapping epitopes in the apical domain and thus compete for binding to hTfR1 (Sade et al. A Human Blood-Brain Barrier Transcytosis Assay Reveals Antibody Transcytosis Influenced by pH-Dependent Receptor Binding, PLOS One 4(1):e936340(2014)). Therefore, we can further conclude that KB_A01 also binds to the apical domain.

[0195] [Table 3]

[0196] Example VII - Characterization of Fc-fused VHHs Affinity measurements were performed by SPR on a Biacore T200 (Cytiva) by loading KB_A01-Fc fusion, BV (Fc fusion), and BA2 (Example IV) onto a Protein A chip (Cytiva) at high temperature (100 nM for 60 seconds). They were then exposed to increasing concentrations of hTfR1 ranging from 0.16 to 100 nM in 5-fold increments. The data are summarized in Table 4, and the sensorgrams are presented in Figure 4. The dimer, KB_A01-Fc, bound to immobilized hTfR1 (K D = 4.1 nM, see Table 2) compared to the free monomeric KB_A01. D As controls, the reference BV (VHH-Fc) and BA2 (IgG1) were run under the same conditions and bound to hTfR1 with affinities (K) of 0.00228 and 0.0249 nM, respectively. D Mouse TfR1 was also tested under the same conditions, but neither KB_A01-Fc nor the reference antibody BA2 bound to mTfR1 with a detectable signal. In contrast, the reference VHH-Fc (BV) showed a high affinity (K D ) bound to mTfR1 (Figure 4D). These experiments demonstrated that KB_A01 exhibited desirable binding properties in the nM affinity range and a low K for receptor release (off). D In contrast to BV, KB_A01 binds only to hTfR1, but not to mTfR1.

[0197] [Table 4]

[0198] To further investigate the effect of soluble human transferrin on binding to human and cynomolgus monkey TfR1, the SPR assay described above was performed in the presence and absence of human Tf. Mouse TfR1 was also included as a control. KB_A01-Fc, BV-Fc, BA1, and BA2 were immobilized on a Protein A sensor (Cytiva) and then exposed to increasing concentrations of receptor ranging from 0.156 to 40 nM (4-fold increments) in the absence or presence of a saturating concentration of 250 nM hTf. There was little difference when comparing affinities in the absence or presence of 250 nM hTf, suggesting that KB_A01 and the two reference antibodies, BA1 and BA2, do not interfere with Tf binding to human or cynomolgus monkey TfR1s (Table 5, Figures 5A–6D and 6A–6D). However, the affinity of BV-Fc for human TfR1 (K ) in the presence or absence of hTf was D ) were significantly different. Furthermore, the data showed high affinity for cynomolgus monkey TfR1, while no binding to mouse TfR1 was recorded (Table 5). This experiment demonstrates that KB_A01 has desirable and unique binding properties for using TfR1 molecules in non-human primates and humans to increase delivery to the CNS.

[0199] [Table 5]

[0200] Example VIII - Bioinformatics and analysis of additional clones KB_A01 was aligned to all sequenced clones from the initial screening (panning) of clones using the Jalview Multiple Sequence Alignment Editor and Analysis Workbench version 2 (Burton Group, University of Dundee) software. Based on sequence identity to KB_A01 and positive binding confirmed by ELISA, three additional clones, KB_A09, KB_A10, and KB_A11 (see Figure 7 for alignment), were selected and produced as fusions to human IgG1 Fc (Example IV). Next, CDR combinations were designed, and CDRs from one sequenced clone (KB_ref, SEQ ID NO: 53) were grafted onto the KB_A01 framework to generate KB_A03, KB_A04, KB_A06, and KB_A07. Here, CDR2 (KB_A03), CDR3 (KB_A04), CDR1 and CDR3 (KB_A06), and CDR2 and CDR3 (KB_A07) were replaced in KB_A01. Replacing CDR1 alone (KB_A02), or in combination with CDR2 (KB_A05), or in combination with CDR2 and CDR3 (KB_A08) in KB_A01 resulted in VHHs unable to bind to hTfR1, as shown in Table 7.

[0201] [Table 6]

[0202] Based on the above, KB_A02-KB_A11 were generated as fusions to human IgG1 Fc (see Example IV, SEQ ID NOS: 62-71) and tested for affinity to hTfR1 and mTfR1 using SPR. The dimeric Fc-fused VHHs were immobilized on a Protein A chip as described above. Briefly, using a Biacore 8K and single-cycle kinetics, the VHH-Fc constructs were diluted to 5 nM and captured on a Protein A chip (Cytiva) in eight increments at 5 μL / s for 120 s, followed by injections of receptor concentrations ranging from 0.8 to 500 nM in 5-fold increments (contact time: 120 s, dissociation time: 600 s). After each complete single-cycle kinetics cycle, the surface was regenerated with 10 mM glycine, pH 1.5, at 30 μL / s for 30 s. The kinetic data for KB_A02 to KB_A11 as Fc fusions are presented in Table 7. KB_A03, KB_A04, KB_A06, KB_A07, KB_A09, KB_A10, and KB_A11 all exhibited the same K D The KB_ref fused to human IgG1 Fc (SEQ ID NO: 85) bound to hTfR1 with binding affinity ranging from 0.150 to 5.12 nM (Table 7). In contrast, KB_0A2, KB_A05, and KB_A08 did not bind to hTfR1 when immobilized on Protein A. No binding to hTfR1 was observed for KB_ref fused to human IgG1 Fc (SEQ ID NO: 85).

[0203] [Table 7]

[0204] Epitope binning was then performed by combinatorially testing hTfR1 binders pairwise using SPR on a Biacore 8K. KB_A01 and KB_A09, along with reference antibodies BA1, BA2, and reference VHH-Fc (BV), were immobilized on a CM5 chip series S (Cytiva) by amine coupling. The immobilized proteins were diluted to 25 μg / mL with sodium acetate at pH 5.5, resulting in an SPR signal of approximately 6,000 resonance units (RU). 50 nM of hTfR1 ectodomain was used in the first injection (contact time 120 s, 10 μL / s), and 100 nM of each analyte was used in the second injection. KB_A01, KB_A03, KB_A04, KB_A09, KB_A10, KB_A11, BA1, BA2, BV, mouse transferrin (mTf, purified, Jackson Immunoresearch Europe Ltd), and human transferrin (hTf, purified, Jackson Immunoresearch European Ltd) were applied at 10 μL / s for 150 seconds. After each cycle, the surface was regenerated with 10 mM glycine, pH 2.1, at 30 μL / s for 30 seconds. Data were analyzed using Biacore Insight evaluation package version 4.0.8.20368. Results showed that KB_A03, KB_A09, KB_A10, and KB_A11 all competed with KB_A01, KB_A09, and antibody BA1 (Table 8), but not with BA2 or BV. Antibody BA1, known to bind to the apical domain of hTfR1 (Helguera G. et al., An antibody recognizing the apical domain of human transferrin receptor 1 efficiently inhibits the entry of all new world hemorrhagic fever arenaviruses. J Virol. 86(7):4024-4028(2012)), competed with BA2 and the VHH-Fc fusion constructs described above, but not with BV. Quite surprisingly, KB_A04 did not compete with BA1 (Table 8).Thus, KB_A01, KB_A03, KB_A09, KB_A10, and KB_A11 form distinct bins that partially, but not completely, overlap with the KB_A04 and BA1 bins. The control VHH-Fc molecule BV did not block the binding of other anti-TfR1 molecules, nor mTf or hTf, to hTfR1. Thus, the BV VHH antibody binds to an epitope that is completely different from those of KB_A01, KB_A03, KB_A09, KB_A10, and KB_A11.

[0205] [Table 8]

[0206] To demonstrate functionality when fused to non-Fc proteins, KB_A01 was also produced as a fusion with a single-chain variable domain located at either the N-terminus (scFv-KB_A01) or C-terminus (KB_A01-scFv). For details, see Example IV and Figure 2. To verify that these fusion proteins still bind to hTfR1, binding kinetics were examined by SPR using an 8K Biacore. Briefly, the receptor hTfR1 was immobilized on a CM5 chip (Cytiva) by amine coupling and exposed to increasing concentrations of either scFv-KB_A01 or KB_A01-scFv in four-fold increments ranging from 0.25 to 64 nM using single-cycle kinetics. Kinetic binding data and sensorgrams are presented in Table 9 and Figure 8.

[0207] [Table 9]

[0208] Validation of apical domain binding of KB_A01 family binders by ELISA To further evaluate whether KB_A01 and related anti-hTfR1 VHHs bind within the apical domain of hTfR1, we expressed the apical domain by display on M13 phage. This approach was undertaken because apical domains expressed as a single soluble domain are known to be unstable. A synthetic gene encoding a recombinant variant of the apical domain of hTfR1 (covering regions Q191-F297 and S326-V380 of the full-length hTfR, corresponding to Q80-F186 and S215-V269 in the hTfR ectodomain, SEQ ID NO: 30), with mutations at Y222S and L329S for increased solubility, was provided by Genscript Biotech Corp. (SEQ ID NO: 54; see SEQ ID NO: 72 for the amino acid sequence). The gene was flanked by restriction enzyme sites AgeI and NotI. The AP1 gene was cloned into the phagemid pTG3 using restriction digestion and T4 ligation (New England Biolabs). The resulting phagemid pTG3-AP1 was transformed into E. coli XL1-Blue via heat shock and plated on LA-ampicillin (100 μg / ml) plates. Transformants were grown in 2xYT medium supplemented with ampicillin (100 μg / ml) and phagemid minipreps (Monarch Plasmid Mini Prep Kit, New England Biolabs). The sequence of the pTG3-AP1 phagemid prep was confirmed using Sanger sequencing (Eurofins Genomics GmbH).

[0209] E. coli XL1-Blue transformed with pTG3-AP1 was grown at OD in 30 ml of 2xYT medium supplemented with 5 μg / ml tetracycline and 100 μg / ml ampicillin. 600 The cells were cultured at 37°C until the pH reached 0.7. 0.5 ml (3.1 × 10 11pfu / ml) was added to the main culture, followed by incubation at 37°C for 1 hour. Kanamycin 50 μg / ml and IPTG 0.25 mM were added to the culture, followed by incubation overnight at 30°C. The culture was centrifuged, and the supernatant was transferred to a new tube. 7.5 ml of NaCl-PEG (20% PEG-8000 / 2.5M NaCl) was added to the supernatant, followed by incubation on ice for 30 minutes. The supernatant was centrifuged for 30 minutes at 4°C. The supernatant was discarded, and the phage pellet was dissolved in 2 ml of 1% BSA PBS. The resuspended phage preparation was centrifuged briefly to remove larger aggregates and then 0.2 μm filtered.

[0210] Immunoplates (Nalge Nunc International) were coated with the reference antibody BA2 and the BV-Fc-containing constructs KB_A01-Fc and KB_A02-KB_A11, tested at a concentration of 10 μg / ml (5 μg / ml for Fc fusions) in 0.05 M sodium carbonate buffer, pH 9.4. The plates were incubated overnight at 4°C. The plates were washed twice with PBS-Tween. Blocking buffer, 1% BSA in PBS, was added and incubated for 1 hour at room temperature. The plates were then washed twice with PBST. An M13 phage displaying the hTfR1-apical region (SEQ ID NO: 54) was added in a dilution series starting at a 1:1 dilution in PBST. For the hTfR1 ectodomain (SEQ ID NO: 30), the plates were coated with 10 μg / ml hTfR1, and the antibodies were added in a dilution series as described above. The plates were incubated for 2 hours at room temperature and then washed four times with PBST. Anti-M13-HRP antibody (product number 11973-MM05T-H, Sino Biological) or anti-Fc HRP (Pierce product number 31423, 1 mg / mL), both diluted 1:5000 in PBS-Tween, was added, followed by incubation at room temperature for 1 hour. The plate was washed four times with PBST, and 1-Step™ Ultra TMB-ELISA substrate solution (ThermoFisher Scientific) was added and developed for 30 minutes at room temperature. The reaction was stopped with 2 M H2SO4, and absorbance was measured at 450 nm. This series of experiments demonstrated that KB_A01, KB_A04, KB_A09, and KB_A11 and BA2 bound to the apical domain but not BV (Figures 8A and 8B). Control experiments using the above ELISA method confirmed that both KB_A01, BA2, and BV bound to the ectodomain of hTfR1 using the same methodology (Figure 8C). This was confirmed in all the SPR experiments mentioned above.

[0211] Combined with the above binning experiments, we can conclude that KB_A01, KB_A03, KB_A04, KB_A07, KB_A09, and KB_A11 bind to the same restricted region of the apical domain of hTfR1. Although not tested here, KB_A07 shares sequence homology with other apical-binding VHH clones and blocks them in the binning experiments, making it highly unlikely that it does not bind to the apical domain. Furthermore, the reference VHH antibodies in the BV were found not to bind to the apical domain of hTfR1.

[0212] Example IX - TfR1-mediated uptake in human cells Next, we tested whether the KB_A01 VHH-Fc fusion protein could be internalized into TfR1-expressing cells. Adherent HEK293T cells (ATCC), known to express TfR1, were grown in collagen-coated 96-well plates (ThermoFischer Scientific) and seeded at a density of 30,000 cells / well. Cells were grown in DMEM (Dulbecco's Modified Eagle's Medium) for 3-5 days until approximately 80-90% confluence. KB_A01-Fc, as well as known reference binders BA1 and BA2, and a negative control (a similar VHH-Fc known not to bind hTfR1) were used in this assay.

[0213] The cell assay was established based on titration of test items KB_A01, BA1, BA2, and a negative control at concentrations ranging from 0.85 nM to 82.5 nM for 5, 15, 30, and 45 minutes at 37°C in 5% CO2. To compare the cellular fate of added proteins after prolonged exposure, after a 30-minute incubation, the medium was replaced with fresh DMEM for 90 minutes (incubated at 37°C in 5% CO2), followed by two subsequent medium changes. After 120 minutes, cells were transferred to the bench, immediately washed with PBS (1x), and fixed with 4% paraformaldehyde for 10–15 minutes. After fixation, cells were washed with PBS, permeabilized with 0.1% Triton-X100™ in PBS, and then blocked with 1% bovine serum albumin (BSA) in PBS. Alexa Fluor-488-anti-human IgG (Fc fragment-specific, stock 0.75 mg / ml) (Jackson Immunoresearch) was then added. 300 U of Texas Red-phalloidin (1:200 dilution of 10 μg / mL stock) was added to the wells during the final 20 minutes of secondary antibody incubation. After this, cells were washed at least three times with PBS and stored at 4°C until imaging. Cells were imaged on an inverted Zeiss 710 laser scanning confocal microscope using lasers 488, 555, and 647 and a 20x air objective.

[0214] Results from these experiments demonstrated that KB_A01, as well as BA1 and BA2, were readily internalized into HEK293T cells at various concentrations (0.7–85 nM) and incubation times from 5 to 240 minutes. As shown in Figure 10A, KB_A01-Fc was internalized as efficiently as BA1. At shorter incubation times (15–45 minutes), the amount was comparable to that of KB_A01-Fc and BA1. At longer incubation times, BA1 showed a clear decrease in intracellular abundance compared to both KB_A01 and BA2 (Figure 10C), i.e., its abundance in cells was low after 240 minutes, indicating lysosomal degradation of BA1. The negative control (VHH-Fc, not bound to hTfR1) did not accumulate intracellularly and was absent at any time point or concentration tested, indicating that the assay and internalization were TfR1-dependent.

[0215] Example X - Evaluation of transcytosis potential Experiments were performed to evaluate the transcytosis potential of KB_A01 and compare it to selected reference IgGs and a standard TfR1-binding BA2. In vitro BBB transwell assays were performed using brain-like endothelial cells in a tight monolayer on a transwell system as described in Sjoestroem et al., Transport study of interleukin-1 inhibitors using a human in vitro model of the blood-brain barrier, Brain Behavior, Immunity Health 16:100307 (2021). Assays were performed at Laboratoire de la barriere Hematoencephalique (LBHE), Universite d'Artois, France. To evaluate the transcytosis potential of VHH-Fc fusions, 500 nM of test compound (VHH-Fcs or full-length antibody) was added in triplicate to donor wells and incubated at 37°C for 3 hours. At the end of the experiment, all compartments were collected. Cells were lysed, and the contents from all compartments (donor, cell, and receiver chambers) were analyzed by ELISA reactive to human Ig (Fc-specific) from Mabtech, Sweden. Plates were coated with anti-human IgG mAb MT145 (Fc-specific) and blocked with PBST containing 0.1% BSA. Sample and standard dilutions were added. Detection reagents were then added in two steps: first, biotinylated anti-human IgG mAb MT78 (Fc-specific), then HRP-conjugated streptavidin. Washes with PBST were performed between each step. An equal volume of TMB was added, and finally, the reaction was stopped with 2 M HCl, and the plate was read. The substrate incubation time was set to 10 min. A control monoclonal antibody lacking active transcytosis activity (reactive with interleukin-1-beta) was used as a reference. This control monoclonal antibody does not bind to TfR1 and exhibits poor BBB penetration. Additionally, the clinically validated transporter BA2 was used (Example IV).The method was validated by an internal control in every well through co-incubation and measurement of human serum albumin (HSA) conjugated to fluorescent Alexa-647 and sodium fluorescein (NaF) to control for monolayer tightness. All wells included in the comparison were tight and showed similar proportional distribution of HSA across the polarized cell monolayer.

[0216] As shown in Figure 11, KB_A01 VHH-Fc exhibits a clearly superior ability to transcytose and remain in the luminal compartment of polarized cell monolayers. This experiment demonstrates that KB_A01 as a VHH-Fc fusion was capable of directional transcytosis through human polarized brain-like endothelial cell monolayers.

[0217] Example XI - In vivo and ex vivo evaluation of VHH-Fc of KB_A01 in TfR1 extracellular domain humanized mice (hECD-TfR1-mice) To evaluate the in vivo brain targeting properties of KB_A01, the distribution characteristics of KB_A01 were examined in homozygous, heterozygous, and wild-type hECD-TfR1 mice. Homozygous mice express only partially humanized TfR1, i.e., human-mouse chimeric TfR1. Humanized mice expressing a chimeric receptor containing the extracellular domain of hTfR1 and the intramembrane and transmembrane domains of mTfR1 were generated at Taconic Biosciences GmbH, Leverkusen, Germany.

[0218] A genetically engineered mouse model was generated by Taconic Biosciences GmbH, Leverkusen, Germany, in which the endogenous mouse Tfrc gene was partially humanized so that the engineered mice produced a chimeric TFRC protein containing the human TFRC extracellular domain. The targeting vector was constructed using methods known to those skilled in the art using BAC-derived mouse and human genomic DNA fragments and including standard selection cassettes.

[0219] Embryonic stem cells (ES) derived from C57BL / 6NTac mice were grown on a mitotically inactivated feeder cell layer composed of mouse embryonic fibroblasts in ES cell medium containing leukemia inhibitory factor and fetal bovine serum. Cells were electroporated with a linearized DNA targeting vector according to methods known to those skilled in the art. Homologous recombinant clones were isolated using methods known to those skilled in the art. ES cell colonies (ES clones) with distinct morphologies were isolated and analyzed by Southern blotting and / or PCR. Homologous recombination at the 3' and 5' sites and single integration were confirmed by PCR and Southern blot using conventional methods known to those skilled in the art. Homologous recombinant ES cell clones were expanded and frozen in liquid nitrogen.

[0220] Generation of chimeras and heterozygotes: After hormone administration, superovulated BALB / c females were mated with BALB / c males. Blastocysts were removed from the uterus at 3.5 days post-transplant (dpc) and placed in a drop of DMEM containing 15% fetal calf serum (FCS) under mineral oil. Using a microinjection pipette, 10–15 targeted ES cells were injected into each blastocyst. After retrieval, approximately eight injected blastocysts were transferred into each uterine horn of pseudopregnant NMRI females. The degree of chimera formation was assessed in chimeras (F0) by observing the contribution of ES cell coat color to the BALB / c host (black / white). In vitro fertilization was performed using oocytes from superovulated C57BL / 6NTac females and thawed semen from previously cryopreserved sperm from male chimeras. Fertilized embryos were then transferred into pseudopregnant Swiss-Webster recipient females. Germline transmission was initially identified by the presence of black-coated pups (C57BL / 6NTac strain) and confirmed by PCR-mediated genotyping of the black pups. F1 mice heterozygous for the modified Tfrc allele were subsequently bred to generate homozygous (HOM), heterozygous (HET), and wild-type (WT) genotypes, which were used in the following studies.

[0221] KB_A01-Fc was prepared as described in Greenwood et al., The preparation of131 Iodine-125 (I-labeled human growth hormone of high specific radioactivity) was measured using the chloramine T method as described in Biochem J 89:114-123 (1963). 125 Protein amounts of 15–80 μg were radiolabeled with 260 ± 26 MBq / μg or 61 ± 2.4 kBq / μg stocks. 125I (PerkinElmer Inc, Waltham, MA, USA) and 5 μg of chloramine T (Sigma-Aldrich) were mixed in PBS. The reaction mixture (110 μL) was incubated for 90 s and then quenched with 10 μg of Na-metabisulfite (Sigma-Aldrich). The radiolabeled sample was immediately purified on a Zeba-column (ThermoFischer) with a 7 kDa cutoff. In vivo experiments were performed in 6-10 week-old C57BL / 6 mice. Immediately after radiolabeling, mice were intravenously administered KB_A01-Fc (5 nmol / kg, 3.4 MBq / nmol) via the tail vein. Blood samples (8 μL) were obtained from the tail vein at 5 min, 0.5 h, 1 h, and finally 2 h after injection. Two hours after injection, the mice were euthanized by cardiac blood collection followed by transcardial perfusion with 40 mL of NaCl for 2.5 minutes. The brain and major organs were dissected. Blood was separated into plasma and a blood cell pellet by centrifugation at 10,000 × g for 5 minutes. The brain was separated into left and right hemispheres. The cerebellum was removed from the left hemisphere. The remaining tissues of the left hemisphere are hereafter referred to as the "brain." Radioactivity was then measured in the brain, blood fractions, and major organs using a γ-counter (2480 Wizard™, Wallac Oy PerkinElmer, Turku, Finland). As shown in Figures 13A and 13B, the pharmacokinetic blood profile showed the highest blood clearance in HOM mice, followed by HET mice, and the lowest in WT mice. Furthermore, the distribution of KB_A01 among whole blood, pellet, and plasma showed the highest abundance in the blood pellet (after spinning whole blood as described above) in HOM mice, followed by HET mice, while the lowest abundance in WT mice, indicating binding to hTfR1 on erythrocytes in vivo. Figure 13C shows the normalized brain uptake (SUV), which was significantly higher in HOM mice compared to WT mice, while in HET mice it was nearly 50% lower than in HOM mice. This clearly demonstrates the specific brain targeting ability of KB_A01 in functional hECD-TfR1 in vivo.

[0222] Example XII - Epitope Characterization To determine the precise contribution of amino acids in the apical domain of human TfR1, the binding sites of KB_A01, KB_A03, and KB_A09 in the form of Fc fusions described in Example IV were investigated together with two reference antibodies, BA1 and BA2. Epitope mapping was performed based on massively parallel mutagenesis scanning using Seqitope™ (AAX Biotech AB, Sweden). Seqitope™ outputs a binding ratio (BR) value for each amino acid residue in the target antigen, which correlates with the importance of that amino acid residue in antibody binding. A high BR value (>2.5) indicates a high contribution of that amino acid residue to antibody binding, but this value will vary depending on the affinity of the antibody being investigated. A BR value of 1.0 indicates no contribution of that amino acid residue to antibody binding.

[0223] The target antigen was the apical domain of hTfR1 expressed in M13 phage as described in Example VIII.

[0224] KB_A01, KB_A03, and KB_A09 as well as the reference antibodies BA1 and BA2 tested bound to the apical domain expressed on M13 phage, but the reference BV did not.

[0225] The epitopes of reference antibodies BA2 (D204, K205, N206, R208, V210, E369) and BA1 (D242, L246, P249, E369, V366, G351, R364) on hTfR1 and antibodies KB_A01 (P354, L246, D245, G351, D242, R364, S355, D356), KB_A03 (D356, P354, D242, D245, G351, L246, S355) and KB_A09 (D356, D245, P354, G351, D242, L246, S355) were isolated in this case (see Table 10). Antibodies within the KB_A01 group (KB_A01, KB_A03, and KB_A09) bound to similar epitopes in the apical domain.

[0226] Detailed separation of the epitopes obtained by Seqitope™ revealed that the epitope bound by BA1 partially overlaps with the epitope bound by BA2 (E369), but also partially overlaps with the epitopes bound by antibodies bound by the KB_A01 group (D242, L246, and G351). The conclusion is that the epitope bound by the KB_A01 group is distinct from, but partially overlaps with, the BA1 epitope and does not overlap at all with the BA2 epitope. This is consistent with the binning data in Example VIII, in which KB_A01, KB_A03, KB_A09, and BA1 all competed with each other, BA1 and BA2 competed with each other, and BA2 did not compete with members of the KB_A01 group. However, the Seqitope™ data provided further detail, demonstrating the differentiation of the current antibodies compared to the reference antibody. The reference VHH BV was not included because it had previously been shown not to bind to the apical domain (see Figure 8A). The epitope of BA1 was mapped to a stretch of the apical domain between Ser324 and Ser368, where three of the identified amino acids are located (Helguera G. et al., An antibody recognizing the apical domain of human transferrin receptor 1 efficiently inhibits the entry of all new world hemorrhagic fever arenaviruses. J Virol. 86(7):4024-4028(2012)).

[0227] Although the epitopes of the three VHH-Fc variants tested largely overlapped, the individual contribution of each amino acid in hTfR1 varied to some extent. Most strikingly, D356 contributed most strongly to the binding of KB_A03 and KB_A09, whereas this amino acid was the most weakly associated with the binding of KB_A01.

[0228] [Table 10]

[0229] Example XIII - Generation of Protein A-binding antibodies Although Llama VHHs do not normally bind to the Protein A-based resin PrismA™, amino acids within the framework sequences can be mutated to generate VHHs that will bind PrismA™ and thereby bind Protein A. To investigate the potential of KB_A01, KB_A09, and KB_A11, and fusion proteins comprising KB_A01, KB_A09, and KB_A11, to enable Protein A binding, these three VHHs were mutated according to the consensus Protein A binding sequence by Henry et al., A Rational Engineering Strategy for Designing Protein A-Binding Camelid Single-Domain Antibodies, PLoS One 11(9):e0163113 (2016). KB_A01 was mutated from serine to arginine at position 19 and asparagine to serine at position 84 (KB_A01:S19R, N84S, KB_A12 designated as SEQ ID NO:73), KB_A09 was mutated from serine to arginine at position 19 (KB_A09:S19R, KB_A13 designated as SEQ ID NO:74), and KB_A11 was mutated from serine to arginine at position 19 (KB_A11:S19R, KB_A14 designated as SEQ ID NO:75). A positive control (Positive Control 1) (SEQ ID NO:76) known to bind to Protein A was also included. As an expression control, green fluorescent protein (GFP) was expressed and can be seen on the blot in Figure 12 but was not included in any other analyses. All variants were fused to a His6 tag via a short Gly2 linker (GGHHHHHH, SEQ ID NO: 77) to enable Ni-NTA-based purification. A plasmid (pNIC28_Bsa4) containing the codon-optimized sequence was ordered from GeneArt Thermo Fisher and used to transfect E. coli BL21(DE3)pRARE2 cells according to standardized procedures at the Protein Science Facility, Karolinska Institute, Sweden. Briefly, a preculture was grown overnight and used to inoculate the expression culture.Expression cultures were grown at 37°C in 2 x 1 ml Terrific Broth supplemented with 8 g / L glycerol and the appropriate antibiotic. On day 2, the temperature was reduced to 18°C, and cells were induced 1 hour later with 0.5 mM isopropyl β-D-1-thiogalactopyranoside (IPTG). Protein expression continued overnight at 18°C, after which cells were harvested by centrifugation (4000 x g for 20 minutes). Cells were stored at -80°C.

[0230] Cells were resuspended in lysis buffer (100 mM HEPES, 500 mM NaCl, 10% glycerol, 10 mM imidazole, 0.5 mM TCEP, pH 8.0, 1 mg / ml lysozyme, 0.1% DDM, 1 mM MgSO, protease inhibitor (Roche cOmplete™ EDTA-free 0.5 tab / ml), benzonase (0.125 U / mL)) and lysed by three freeze-thaw cycles. Whole cell lysates were clarified by centrifugation (4000 × g, 30 min). The supernatant was mixed with 25 μl of nickel resin pre-equilibrated with equilibration buffer (20 mM HEPES, 500 mM NaCl, 10% glycerol, 10 mM imidazole, 0.5 mM TCEP, pH 7.5) in a 96-well plate equipped with filter units and incubated at room temperature for 30 min with gentle shaking. Unbound proteins were then collected by centrifugation (200 × g, 1 min) at 4 °C. The resin was washed three times with wash buffer (20 mM HEPES, 500 mM NaCl, 10% glycerol, 30 mM imidazole, 0.5 mM TCEP, pH 7.5) and resuspended in 40 μl of elution buffer (20 mM HEPES, 500 mM NaCl, 10% glycerol, 500 mM imidazole, 0.5 mM TCEP, pH 7.5) and incubated at room temperature. Elution fractions were collected by centrifugation. Aliquots of the whole cell lysate (Figure 12A), clarified cell lysate (Figure 12B), and elution fractions (Figure 12C) were analyzed by SDS-PAGE according to standard procedures.

[0231] Elution fractions of wild-type (WT) KB_A01, KB_A09, and KB_A11, as well as protein A mutant variants of KB_A01, KB_A09, and KB_A11 (referred to herein as KB_A12, KB_A13, and KB_14), and positive control 1 were added to 25 μl of MabSelect PrismA™ resin pre-equilibrated with binding buffer (PBS; 10 mM phosphate, 140 mM NaCl, 2.7 mM KCl, pH 7.4) in a 96-well plate containing filter units and incubated for 45 minutes at room temperature with gentle shaking. Unbound protein was collected by centrifugation (200 × g, 1 minute) at 4 °C. The resin was washed with 2 × 250 μl of wash buffer (PBS pH 7.4). To elute the bound protein, the resin was resuspended in 40 μl of elution buffer (100 mM sodium citrate pH 3.0), neutralized with 1 M Tris-HCl pH 9.0, and incubated at room temperature. The elution fraction was collected by centrifugation. Aliquots of the flow-through and elution fractions (Figures 12B and 12C) were analyzed by SDS-PAGE.

[0232] The binding ability of WT KB_A01, KB_A09, and KB_A11 and the corresponding Protein A-enabled mutation variants KB_A12, KB_A13, and KB_A14, along with the positive control 1, to the target hTfR1 was assessed at a single concentration by SPR analysis using a Biacore T200 instrument. Briefly, hTfR1 was covalently immobilized on the surface of a CM5 chip (Cytiva) by amine coupling using standard reagents. Human TfR1 was diluted to 25 μg / mL in 10 mM sodium acetate, pH 5.5, and immobilized for 600 seconds, resulting in a very high signal level of approximately 2300 response units (RU). All six VHHs were then diluted to 100 nM in running buffer (PBS supplemented with 0.05% Tween® 20 (PBS-T)). Signals equal to or greater than the corresponding WT were considered "WT-like binding" (see Table 11).

[0233] The ability of WT KB_A01, KB_A09, and KB_A11 and the corresponding Protein A-enabled mutation variants KB_A12, KB_A13, and KB_A14, along with positive control 1, to bind to PrismA™ resin was examined in detail by SPR analysis using a Biacore T200 instrument. Each construct was diluted to 500 nM in running buffer (PBS-T, pH 7.4) and then serially diluted 1:4 four times (500, 125, 31.2, 7.81, and 1.95 nM) and run on a PrismA™ chip (Cytiva). The contact time for each concentration was 30 seconds at 30 μL / min, and the dissociation time was 30 seconds at 30 μL / min. The surface was regenerated by washing with running buffer at 30 μL / min for 300 seconds.

[0234] Ni-NTA-based purification of wild-type KB_A01, KB_A09, and KB_A11, and MabSelect PrismA™ resin-based purification of the Protein A validated mutation variants and Positive Control 1, yielded highly pure products in the elution fractions (see Figure 12B for representative SDS-PAGE of PrismA™ eluates of KB_A01, KB_A12, and Positive Control 1). The majority of WT KB_A01, KB_A09, and KB_A11 was found in the flow-through of the PrismA™-based purification (see Figure 12C for SDS-PAGE of the flow-through fractions of KB_A01, KB_A12, and Positive Control 1), indicating that mutations at positions 19 and 84 of KB_A01 and at position 19 of KB_A09 and KB_A11 yielded VHHs capable of being purified using PrismA™ resin.

[0235] Binding to hTfR1 was tested by SPR at a single concentration of 100 nM. All three Protein A-active mutation variants bound to hTfR with similar affinity to the corresponding WT VHH. In contrast, positive control 1 did not bind to hTfR1 (see Table 11).

[0236] Binding to PrismA™ resin was also verified by SPR using a precoated chip (Cytiva Series S Sensor Chip PrismA™). WT KB_A01 (Figure 12D), KB_A09, and KB_A11 bound very little to PrismA™ resin even at the highest concentration tested (500 nM), whereas KB_A12 (Figure 12E), KB_A13, and KB_A14, as well as positive control 1, bound at the highest concentration (500 nM), reaching a signal of >1400 RU, confirming the functionality of VHH-mediated binding to PrismA™ resin (Table 11).

[0237] [Table 11]

[0238] Example XIV - Humanization of VHHs Seventeen VHH variants of KB_A01 (without His-tag) (see Table 12, KB_A12, KB_A16 to KB_A31, each mutated at one or several amino acids compared to KB_A01) were produced in CHO cells by ProteoGenix SAS (France). A positive control (Positive Control 2) known to bind Protein A (SEQ ID NO: 84) was also included. cDNA encoding the VHH was chemically synthesized and optimized for expression in CHO cells by ProteoGenix and subcloned into Proteogenix's proprietary mammalian cell expression vector. The vector was transfected into XtenCHO™ cells using the XtenCHO™ transfection protocol. A total of 3.5 mL of culture medium was harvested 8 days post-transfection and purified using Mabselect PrismA™ resin (Cytiva). The medium was clarified by filtration (0.22 μm) and loaded onto pre-equilibrated (PBS pH 7.5) PrismA™ resin. Bound VHHs were eluted by pH shift and the eluate was immediately neutralized by addition of 1 M Tris-HCl pH 9.0. Protein concentration was determined by A280 spectrophotometry.

[0239] The ability of KB_A01 and 18 mutant variants of KB_A01 produced in CHO cells, all with two protein A-effective mutations (S19R, N84S, see Example XII), to bind to the target hTfR1 was assessed at a single concentration by SPR using a Biacore T200 instrument. E. coli -produced variants of KB_A01, KB_A12, KB_A15, KB_A16, KB_A17, KB_A18, and KB_A19, all fused to a His6 tag (Example XII), were also included in the same experiment and run under the same conditions. Briefly, hTfR1 was covalently immobilized on the surface of a CM5 chip (Cytiva) by amine coupling using standard reagents. Human TfR1 was diluted to 15 μg / mL in 10 mM sodium acetate, pH 5.5, with an immobilization level of 1,000 response units (RU) targeted. All VHHs were then diluted to 50 nM in running buffer (PBS supplemented with 0.05% Tween® 20). A signal similar to or greater than the corresponding WT KB_A01 (for E. coli-produced material) or KB_A12 (CHO-produced) was considered "bound" (see Table 12). The reason for using KB_A12 as a reference for CHO-produced material is that WT KB_A01 does not have PrismA™ binding capacity and was therefore not produced as a free VHH without a His6 tag. Similarly, KB_A15 was also tested only fused to a His6 tag.

[0240] Eighteen mutant VHH variants of KB_A01 were analyzed for binding in screening mode at 50 nM using an intermediate immobilization of 1000 RU of hTfR. KB_A01 and KB_A15 were tested only as His-tagged proteins produced in E. coli, whereas KB_A12, KB_A15, KB_A17, KB_A18, and KB_A19 were tested both as His-tagged proteins produced in E. coli and as free VHHs produced in CHO cells. KB_A20 to KB_A31 were produced only in CHO cells and purified by PrismA™ resin. As can be seen in Table 12, mutations at positions 1, 5, 11, 14, 19, 43, 44, 45, 46, 74, 78, 84, 86, and 109, by themselves and / or in some combination of the above, do not interfere with binding to hTfR1 when analyzed as a single concentration assay at 50 nM. In contrast, mutations at positions 37, 49, 50, 96, and 97, in addition to the 10 positions already mutated in KB_A18, completely prevented binding to hTfR1. Mutation at position 47 dramatically reduced binding, while mutations at positions 58 and 87 more modestly reduced affinity.

[0241] [Table 12] TIFF2025539346000014.tif44159

[0242] [Table 13] TIFF2025539346000016.tif239159TIFF2025539346000017.tif242159TIFF2025539346000018.t if240159TIFF2025539346000019.tif246159TIFF2025539346000020.tif242159TIFF20255393460 00021.tif235159TIFF2025539346000022.tif240159TIFF2025539346000023.tif237159TIFF202 5539346000024.tif244159TIFF2025539346000025.tif235159TIFF2025539346000026.tif246159

[0243] The above-described embodiments should be understood as a few illustrative examples of the present invention. Those skilled in the art will understand that various modifications, combinations, and changes can be made to the embodiments without departing from the scope of the present invention. In particular, different part solutions in different embodiments can be combined in other configurations, if technically possible. However, the scope of the present invention is defined by the appended claims.

Claims

1. A heavy chain-only variable domain (VHH) antibody that specifically binds to transferrin receptor 1 (TfR1), Complementarity determining region 1 (CDR1) having an amino acid sequence selected from the group consisting of GSIFGSKR as defined in SEQ ID NO: 1 and GSIFGFNA as defined in SEQ ID NO: 2; a CDR2 having an amino acid sequence selected from the group consisting of ITYRGTT as defined in SEQ ID NO:3 and IAVAGST as defined in SEQ ID NO:4; and a CDR3 having an amino acid sequence selected from the group consisting of WMFTTDNY as defined in SEQ ID NO:5 and WMYATANY as defined in SEQ ID NO:6; when the CDR1 has the amino acid sequence defined in SEQ ID NO:2, the CDR3 has the amino acid sequence defined in SEQ ID NO:6; However, the VHH antibody does not comprise a CDR1 having the amino acid sequence defined in SEQ ID NO: 2, a CDR2 having the amino acid sequence defined in SEQ ID NO: 4, and a CDR3 having the amino acid sequence defined in SEQ ID NO:

6.

2. However, the VHH antibody does not include a CDR1 having the amino acid sequence defined in SEQ ID NO: 2, a CDR2 having the amino acid sequence defined in SEQ ID NO: 3, and a CDR3 having the amino acid sequence defined in SEQ ID NO:

6.

3. the CDR1 has the amino acid sequence defined in SEQ ID NO: 1; the CDR2 has an amino acid sequence selected from the group consisting of SEQ ID NOs: 3 and 4; A VHH antibody according to claim 1 or 2, wherein the CDR3 has the amino acid sequence defined in SEQ ID NO:

5.

4. the CDR1 has the amino acid sequence defined in SEQ ID NO: 1; the CDR2 has the amino acid sequence defined in SEQ ID NO:4; The VHH antibody of claim 3, wherein the CDR3 has the amino acid sequence defined in SEQ ID NO:

5.

5. the CDR1 has the amino acid sequence defined in SEQ ID NO: 1; the CDR2 has the amino acid sequence defined in SEQ ID NO:3; The VHH antibody of claim 3, wherein the CDR3 has the amino acid sequence defined in SEQ ID NO:

5.

6. the CDR1 has an amino acid sequence selected from the group consisting of SEQ ID NOs: 1 and 2; the CDR2 has the amino acid sequence defined in SEQ ID NO:3; The VHH antibody of claim 1, wherein the CDR3 has the amino acid sequence defined in SEQ ID NO:

6.

7. the CDR1 has the amino acid sequence defined in SEQ ID NO: 1; the CDR2 has the amino acid sequence defined in SEQ ID NO:3; The VHH antibody of claim 6, wherein the CDR3 has the amino acid sequence defined in SEQ ID NO:

6.

8. the CDR1 has the amino acid sequence defined in SEQ ID NO: 1; the CDR2 has the amino acid sequence defined in SEQ ID NO:4; A VHH antibody according to claim 1 or 2, wherein the CDR3 has the amino acid sequence defined in SEQ ID NO:

6.

9. said VHH antibody having the formula: framework region 1 (FR1), -CDR1-FR2-CDR2-FR3-CDR3-FR4, FR1 has the amino acid sequence QVQLQESGGGSVQAGGSLSLSCAAS as defined in SEQ ID NO:57; FR2 has the amino acid sequence MGWFRQAPGEQRDVVAT as defined in SEQ ID NO:58; FR3 has the amino acid sequence EYADSVKGRFTISRDNAKNTVYLQMNNLKPEDTAVYYC as defined in SEQ ID NO:59; The VHH antibody according to any one of claims 1 to 8, wherein FR4 has the amino acid sequence WGQGTQVTVSS defined in SEQ ID NO:

22.

10. The VHH antibody of claim 1, wherein the VHH antibody has an amino acid sequence selected from the group consisting of SEQ ID NOs: 23 to 26.

11. A heavy chain-only variable domain (VHH) antibody that specifically binds to transferrin receptor 1 (TfR1), Amino acid sequence X defined in SEQ ID NO:7 １ X ２ IX ３ A complementarity determining region 1 (CDR1) consisting of GSKR, wherein X １ is G or E, and X ２ is S, D, or I, and X ３ is F or N, The amino acid sequence ITX defined in SEQ ID NO:8 ４ X ５ GTT, wherein X ４ is Y or V, and X ５ is R, H or G, The amino acid sequence WMFTTX defined in SEQ ID NO:9 ６ NY, wherein X ６ is D, T, or N, The VHH antibody comprising:

12. The CDR1 is the amino acid sequence X defined in SEQ ID NO:7 １ X ２ IX ３ GSKR, wherein X １ is G or E, and X ２ is D or I, and X ３ is F or N, The CDR2 has the amino acid sequence ITX defined in SEQ ID NO:8 ４ X ５ GTT, wherein X ４ is Y or V, and X ５ is R, H or G, The CDR3 has the amino acid sequence WMFTTX defined in SEQ ID NO:9 ６ NY, wherein X ６ The VHH antibody of claim 11, wherein is D, T, or N.

13. the CDR1 has the amino acid sequence GDIX defined in SEQ ID NO: 11 ３ GSKR, wherein X ３ is F or N, the CDR2 has the amino acid sequence ITVX defined in SEQ ID NO: 12 ５ GTT, wherein X ５ is R or G, The CDR3 has the amino acid sequence WMFTTX defined in SEQ ID NO:9 ６ NY, wherein X ６ The VHH antibody of claim 12, wherein is T or N.

14. the CDR1 consists of the amino acid sequence GDINGSKR as defined in SEQ ID NO: 13; the CDR2 consists of the amino acid sequence ITVRGTT as defined in SEQ ID NO: 14; 14. The VHH antibody of claim 13, wherein the CDR3 consists of the amino acid sequence WMFTTTNY defined in SEQ ID NO:

10.

15. the CDR1 consists of the amino acid sequence GDIFGSKR defined in SEQ ID NO: 15; the CDR2 consists of the amino acid sequence ITVGGTT as defined in SEQ ID NO: 16; 14. The VHH antibody of claim 13, wherein the CDR3 consists of the amino acid sequence WMFTTNNY defined in SEQ ID NO:

54.

16. the CDR1 consists of the amino acid sequence EIINFGSKR defined in SEQ ID NO: 17; the CDR2 consists of the amino acid sequence ITYHGTT as defined in SEQ ID NO: 18; 13. The VHH antibody of claim 12, wherein the CDR3 consists of the amino acid sequence WMFTTDNY defined in SEQ ID NO:

5.

17. said VHH antibody having the formula: framework region 1 (FR1), -CDR1-FR2-CDR2-FR3-CDR3-FR4, FR1 is the amino acid sequence QVQLQESGGGX defined in SEQ ID NO: 19 ７ VQAGGSLX ８ LSCAAS, wherein X ７ is S or L, and X ８ is S or R, FR2 has the amino acid sequence MGWFRQAPGX defined in SEQ ID NO: 20 ９ X １０ RDX １１ VAT, wherein X ９ is E, K or Q, preferably K or Q, X １０ is Q or A, and X １１ is V or L, FR3 is the amino acid sequence X defined in SEQ ID NO:21 １２ YX １３ DSVKGRFTISRDDNAX １４ NTVYLQMNX １５ LKPEDTAX １６ YYC, wherein X １２ is E or K, and X １３ is A or E, and X １４ is K or N, and X １５ is N or S, and X １６ is V or F, The VHH antibody according to any one of claims 11 to 16, wherein FR4 has the amino acid sequence WGQGTQVTVSS defined in SEQ ID NO:

22.

18. The VHH antibody of claim 11, wherein the VHH antibody has an amino acid sequence selected from the group consisting of SEQ ID NOs: 23, 27 to 29, preferably selected from the group consisting of SEQ ID NOs: 27 to 29.

19. The VHH antibody according to any one of claims 1 to 18, which specifically binds to human TfR1.

20. The VHH antibody of claim 19, wherein the VHH antibody specifically binds to the apical domain of human TfR1.

21. The VHH antibody has an affinity K selected in the range of 0.1 nM to 150 nM, preferably in the range of 0.1 nM to 100 nM. Ｄ 21. A VHH antibody according to claim 19 or 20, which binds to human TfR1 at

22. The VHH antibody according to any one of claims 1 to 21, which specifically binds to cynomolgus monkey TfR1.

23. The VHH antibody according to any one of claims 1 to 22, wherein the VHH antibody is a Camelidae VHH antibody.

24. The VHH antibody according to any one of claims 1 to 22, wherein the VHH antibody is a humanized VHH antibody.

25. The VHH antibody is X defined in SEQ ID NO: 78 １ VQLX ２ ESGGGX ３ VQX ４ GGSLX ５ framework region 1 (FR1) by LSCAAS; MGWFRQAPGX defined in SEQ ID NO: 79 ６ X ７ X ８ X ９ FR2 by VVAT FR2, EYADSVKGRFTISRDNX defined in SEQ ID NO: 80 １０ KNTX １１ YLQMNX １２ LX １３ FR3 by PEDTAVYYC, and WGQGTX defined in SEQ ID NO: 81 １４ Includes FR4 by VTVSS, In the formula, X １ ~X １４ At least one of １ = E, X ２ = V, X ３ = L, X ４ = P, X ５ = R, X ６ = K, X ７ = G, X ８ = L, X ９ = E, X １０ = S, X １１ = I, X １２ = S, X １３ = R and X １４ = L, X １ ~X １４ Any of the remaining (multiple) is X １ = Q, X ２ = Q, X ３ = S, X ４ = A, X ５ = S, X ６ = E, X ７ = Q, X ８ = R, X ９ = D, X １０ = A, X １１ = V, X １２ = N, X １３ = K and X １４ 25. A VHH antibody according to claim 24, wherein Q is the amino acid sequence of the VHH antibody.

26. The VHH antibody according to any one of claims 1 to 25, wherein the VHH antibody binds to Protein A.

27. The VHH antibody of claim 26, wherein the VHH antibody has framework region 1 (FR1) of QVQLQESGGGSVQAGGSLRLSCAAS according to SEQ ID NO: 82 and FR3 of EYADSVKGRFTISRDNAKNTVYLQMNSLKPEDTAVYYC according to SEQ ID NO:

83.

28. A fusion molecule comprising a VHH antibody according to any one of claims 1 to 27 linked to at least one molecule.

29. The fusion molecule of claim 28, wherein the fusion molecule comprises a VHH antibody of any one of claims 1 to 27 linked to the at least one molecule via a linker.

30. 30. The fusion molecule of claim 28 or 29, wherein the at least one molecule is selected from the group consisting of a therapeutic agent and an imaging agent.

31. 31. The fusion molecule of claim 30, wherein the therapeutic agent is capable of treating a disease or disorder of the central nervous system.

32. 31. The fusion molecule of claim 30, wherein the imaging agent is selected from the group consisting of a position emission tomography tracer, a single photon emission computed tomography tracer, a fluorescent probe, a luminescent probe, a metal complex-containing probe, and a near-infrared fluorescent probe.

33. The fusion molecule of any one of claims 28 to 31, wherein the at least one molecule is a therapeutic agent; and A pharmaceutical composition comprising a pharmaceutically acceptable vehicle.

34. A nucleic acid molecule encoding a VHH antibody according to any one of claims 1 to 27 or a fusion molecule according to any one of claims 28 to 31.

35. 35. An expression vector comprising the nucleic acid molecule of claim 34 operably linked to a promoter.

36. 36. A host cell comprising the nucleic acid molecule of claim 34 or the expression vector of claim 35.

37. The fusion molecule of any one of claims 28 to 31 for use as a medicament, wherein said at least one molecule is a therapeutic agent.

38. 32. The fusion molecule of any one of claims 28 to 31 for use in treating a disease or disorder of the central nervous system (CNS), wherein said at least one molecule is a therapeutic agent capable of treating a disease or disorder of the CNS.

Images