Target-specific extracellular vesicles

Abstract

A method for producing a protein comprising an extracellular vesicle (EV) surface protein TED is provided, comprising the steps of: modifying, by mutagenesis, a polynucleotide comprising a nucleotide sequence encoding the ED of the EV surface protein, within at least one modified region within the ED amino acid sequence having a length of 3 to 20 consecutive amino acids, flanked at the N- and C-termini by regions of the wild-type ED sequence, to incorporate a target binding site within the ED, thereby producing a repertoire of polynucleotides encoding diverse target-specific vesicle ectodomains (TEDs), each comprising a different target binding site; selecting a TED that specifically recognizes a predetermined target; and producing a protein comprising the selected TED.

Description

Technical Field

[0001] The present invention relates to the field of manipulating target-specific extracellular vesicles. The present invention further relates to methods for the production of said target-specific extracellular vesicles.

Background Art

[0002] In the past decade, research on exosomes has increased because they are recognized as important mediators of cell-cell communication [1]. Exosomes are released from multivesicular bodies when their fusion with the plasma membrane occurs, and the released vesicles function as delivery vehicles that transfer transmembrane proteins including functional RNA, exosomal DNA, and receptors to cells in the surrounding environment [2]. Regarding their advantages as potential therapeutic moieties, including desirable properties such as low immunogenicity and low cytotoxicity, interest in their application has spurred further development such as encapsulation, improvement of cytoplasmic release of exosomal contents, enhancement of cell uptake, and new methods for more specific cell targeting [3].

[0003] Exosome vesicle uptake is cell type-specific, may involve membrane fusion or endocytosis, and can even be induced by stimulation of oncogenic cancer receptors [4]. To achieve tissue-specific delivery, the source cells can be engineered to overexpress exosomal membrane proteins such as tetraspanins, and exosome targeting can be optimized by having receptor-specific ligand peptides as recognition units [5]. Tetraspanins are known as molecular facilitators that associate in large cell signaling complexes known as the tetraspanin web, which also includes members of other protein families such as integrins and co-receptor molecules. Furthermore, such large membrane protein assemblies can associate with lipid rafts [6-7]. The functions of tetraspanin CD9, CD63, and CD81 ligands for exosome endocytosis have been reported [8-10], however, the mechanism of uptake has not yet been clearly defined.

[0004] Despite its almost ubiquitous distribution, the tetraspanin protein CD81 is a major protein enriched in the exosome fraction of multivesicular bodies

[11] . The large extracellular loop (LEL) of CD81, which is topologically located between transmembrane domains 3 and 4, is characterized by five helix elements that form a mushroom-like structure stabilized by two pairs of cysteines [12-13]. This motif is conserved among protein members of the tetraspanin family

[14] , and the oxidation of cysteine bonds is an essential condition for the high-affinity binding of the E2 envelope protein of hepatitis C virus (HCV), the natural ligand of CD81

[15] . The correct pairing of cysteines has been described with respect to recognition by antibody M38

[16] , which does not bind to denatured or reduced proteins but can react with membrane-bound hCD81 as well as the native form of purified soluble hCD81.

[0005] The crystal structure of hCD81 LEL solved at 1.6 Å revealed a new type of protein fold

[12] , and subsequent sequence analysis of 160 tetraspanin family members showed that their folds and important structural features are conserved

[17] . Apart from cysteine crosslinks, hCD81 LEL is stabilized by invariant residues Gly157 and Pro176 that are positioned to accommodate cysteine linkages, and Tyr127 that is completely buried and contributes to a hydrogen bond network together with His151 and Cys190. Soluble hCD81 LEL assembles into dimers that surround a two-fold axis, and the contacts between protomers are in low-polarity regions between helices of each interaction partner and the C-terminal residues of the protomer on the opposite side of helix B. The N- and C-termini of the protomers fit into the central region of the opposite face of the assembled dimer, similar to dimer assembly at the cell surface, where transmembrane segments are also present. A second low-polarity region includes the solvent-exposed surfaces of helices C and D, which is energetically unfavorable. According to solution studies, helix D is mostly unstructured and adopts a helical conformation only when bound to a specific antigen

[18] . Sequence alignments of tetraspanin family members actually show an increased variability in this region, including insertions and deletions

[19] . This surface area may be involved in species- or tetraspanin-specific recognition processes

[20] , which may hint at the possibility of heterodimeric tetraspanin species assembly

[21] . In particular, segment D of CD81 should be able to lead to specific homomeric clustering

[22] .

[0006] WO2014 / 168548 discloses a therapeutic delivery vesicle, which can be, for example, an exosome or a microvesicle, having a polypeptide construct attached to its membrane, the polypeptide construct comprising a carrier polypeptide fused to a decoy receptor having no signaling ability.

[0007] WO2016 / 073864 discloses a B cell targeting agent comprising a CD19 or CD21 targeting antibody conjugated to a nanoparticle, a lipid-based carrier molecule, or an extracellular vesicle. WO2018 / 075825 discloses a biotechnologically produced exosome comprising a fusion protein comprising a segment of an exosomal protein fused to a cancer stem cell targeting peptide.

[0008] WO2018 / 015535A1 discloses EVs coated with a protein containing an Fc-binding domain. Exemplary EVs carry a fusion construct comprising an exosomal protein fused to an Fc-binding factor such as protein A / G, the Z domain or ZZ domain of protein A.

[0009] US2018 / 0015182A1 discloses exosomes that deliver bioactive cargo by engineering tetraspanins to include a fusion with a protein or by attaching a protein to the exosome, including at a terminal or loop peptide attachment site.

[0010] El Andaloussi et al. (Advanced Drug Delivery Reviews 2013, 65:391-397) describe exosomes for targeted siRNA delivery. Exemplary targeted exosomes include a Lamp2b brain-specific peptide (RVG, 29mer peptide) fusion protein.

[0011] Drummer et al. (Journal of Virology 2002, 76(21):11143-11147) describe the binding site on the LEL for CD81 binding to hepatitis C virus E2 glycoprotein.

[0012] To enhance their potential as next-generation therapeutic carriers, exosome-mediated delivery systems need to be further developed, especially to improve their inherently low efficiency of cellular uptake. There is a specific need for exosome-mediated delivery systems that have increased stability and improved target specificity and that include exosomal membrane proteins. SUMMARY OF THE INVENTION PROBLEMS TO BE SOLVED BY THE INVENTION

[0013] It is an object of the present invention to provide target - specific extracellular vesicles with improved target - binding characteristics. It is a further object to provide target - specific EV surface proteins and their binding domains with improved target - binding characteristics.

Means for Solving the Problems

[0014] The problem is solved by the present invention. According to the present invention, within at least one modified region having a length of 3 to 20 consecutive amino acids, where regions of the wild - type extracellular domain (ED) sequence are adjacent to the N - terminus and C - terminus of the modified region, a polynucleotide containing a nucleotide sequence encoding the ED of an extracellular vesicle (EV) surface protein is modified by a mutagenesis method to incorporate a target - binding site into the ED, thereby producing a repertoire of polynucleotides encoding diverse target - specific extracellular domains (TEDs), each containing a different target - binding site, and a step of selecting a TED that specifically recognizes a predetermined target, and a step of producing a protein containing the selected TED. A method for producing a protein containing a TED of an EV surface protein is provided.

[0015] It is specifically understood that all properties of the protein, particularly the target - binding molecules ( "binding factors", "target - specific molecules") described herein, are properties that characterize the method of the present invention, and vice versa.

[0016] Specifically, the protein containing the TED can consist of the TED, or a protein containing the TED and one or more additional regions, such as another ED and / or a transmembrane domain, or in particular a recombinant fusion protein containing a heterologous sequence.

[0017] Specifically, the repertoire of polynucleotides presents diverse TEDs on the outer surface and preferably is included in a gene package that employs a display system selected from the group consisting of yeast, phage, bacteria, ribosome, mRNA, or mammalian cell display.

[0018] Specifically, the modified region has a lower target binding affinity when isolated from the TED. Specifically, the target binding site includes at least one additional binding region that is within an additional modified region at least 2 amino acids apart or within the wild-type ED sequence. The at least one additional binding region is preferably of the same ED and / or is located within the same TED at a specific distance further described herein.

[0019] Specifically, the TED includes at least 70% sequence identity with the wild-type ED. Specifically, the wild-type ED is derived from (or is) an EV surface protein selected from the group consisting of a tetraspanin-like protein, an integrin family protein, a proteoglycan, a 5-transmembrane domain protein, a type I transmembrane protein, a Notch family protein, an enzyme membrane protein, an immune regulatory surface protein, a surface marker of mesenchymal stem cells, a glycoprotein, or a channeling protein.

[0020] Specifically, the tetraspanin-like protein is preferably a) CD81 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 87; b) CD9 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 89; c) CD53 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 90; d) TSPAN32 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 91; e) CD82 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 92; f) CD63 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 93; g) CD151 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 94; and h) CD37 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 95 which is a tetraspanin selected from the group consisting of or the tetraspan-like protein is a lysosome-associated membrane protein, preferably LAMP2 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 96.

[0021] Specifically, wild-type ED is a) CD81 in which ED comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 130 or SEQ ID NO: 131; b) CD9 in which ED comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 132, SEQ ID NO: 182, or SEQ ID NO: 133; c) CD53 in which ED comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 134 or SEQ ID NO: 135; d) TSPAN32 in which ED comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 136 or SEQ ID NO: 137; e) CD82 in which ED comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 138 or SEQ ID NO: 139; f) CD63 in which ED comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 140 or SEQ ID NO: 141; g) CD151 in which ED comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 142 or SEQ ID NO: 143; h) CD37 in which ED comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 144 or SEQ ID NO: 145; or i) ED is LAMP2 that contains or consists of the amino acid sequence specified as SEQ ID NO: 146. It is any one of the following.

[0022] Specifically, the protein contains a loop structure in the ED amino acid sequence, and the loop structure is stabilized by one or more cysteine(s) at one or more position(s) that allow the formation of one or more disulfide bonds.

[0023] Specifically, the modified region is located within the loop region of ED. Specifically, ED is a) of CD81, the amino acid sequence is preferably modified between positions 134 and 144 and / or 130 and 146 and / or 135 and 168 to introduce cysteine(s) that allow the formation of one or more disulfide bonds not naturally present in the wild-type ED sequence, and the numbering is that of human CD81 specified as SEQ ID NO: 87, or b) of CD9, the amino acid sequence is preferably modified between positions 20 and 28 to introduce cysteine(s) that allow the formation of one or more disulfide bonds not naturally present in the wild-type ED sequence, and the numbering of the position is that of the large extracellular loop (LEL, SEQ ID NO: 118) of CD9.

[0024] Specifically, ED is of CD81, the modified region is located within the range of positions 160 and 172, and the numbering is that of human CD81 specified as SEQ ID NO: 87. Specifically, TED contains at least one additional binding region located between positions 132 and 141 or between positions 180 and 189, and the numbering is that of human CD81 specified as SEQ ID NO: 87.

[0025] Specifically, the ED is that of CD9, and the modified region is located within any one of the ranges of positions 155-166, 128-142, 130-140, or 169-180, and the numbering is that of human CD9 specified as SEQ ID NO: 89.

[0026] Specifically, the target is selected from the group consisting of a cell target, preferably a mitogen receptor, cytokine receptor, asialoglycoprotein receptor, membrane transporter, lipoprotein, lipopolysaccharide, glycoprotein, proteoglycan, or a cell-free target, preferably a cytokine, artificial protein, or artificial surface structure.

[0027] Specifically, the protein containing the TED is a target-specific EV surface protein (TSP) containing the said TED and at least one transmembrane domain. Specifically, the transmembrane domain contains at least 70% sequence identity with a transmembrane domain originating from a mammalian EV surface protein.

[0028] Specifically, the transmembrane domain is a) CD81, wherein the transmembrane domain contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 147, SEQ ID NO: 148, SEQ ID NO: 149, or SEQ ID NO: 150; b) CD9, wherein the transmembrane domain contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 151, SEQ ID NO: 152, SEQ ID NO: 153, or SEQ ID NO: 154; c) CD53, wherein the transmembrane domain contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 155, SEQ ID NO: 156, SEQ ID NO: 157, or SEQ ID NO: 158; d) TSPAN32, wherein the transmembrane domain contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 159, SEQ ID NO: 160, SEQ ID NO: 161, or SEQ ID NO: 162; e) CD82 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 163, SEQ ID NO: 164, SEQ ID NO: 165, or SEQ ID NO: 166; f) CD63 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 167, SEQ ID NO: 168, SEQ ID NO: 169, or SEQ ID NO: 170; g) CD151 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 171, SEQ ID NO: 172, SEQ ID NO: 173, or SEQ ID NO: 174; h) CD37 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 175, SEQ ID NO: 176, SEQ ID NO: 177, or SEQ ID NO: 178; or i) LAMP2 in which the transmembrane domain comprises or consists of the amino acid sequence specified as SEQ ID NO: 179 is any one of the above.

[0029] Specifically, both the ED and the transmembrane domain originate from (or are) the same EV surface protein, and preferably, the EV surface protein is a) CD81 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 87; b) CD9 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 89; c) CD53 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 90; d) TSPAN32 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 91; e) CD82 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 92; f) CD63 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 93; g) CD151 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 94; h) CD37 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 95; and i) LAMP2 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 96 selected from the group consisting of.

[0030] The present invention provides a) a polynucleotide encoding a protein comprising TED obtainable by the method described herein; b) introducing the polynucleotide into a source cell or source cell mixture; b) culturing the cell(s) under conditions for producing extracellular vesicles; c) isolating a fraction containing target-specific extracellular vesicles (TEV) containing the target binding site of TED; and d) producing a preparation of TEV contained in the fraction to further provide a method for producing a TEV preparation.

[0031] Specifically, the protein containing TED is a target-specific EV surface protein (TSP) containing the TED and at least one transmembrane domain, and the TEV presents a target binding site on the outer surface of its membrane.

[0032] Specifically, the method described herein further includes the step of loading an intracellular cargo into the TEV, and the cargo includes a peptide, polypeptide, protein domain, protein, lipid, gene, mRNA, miRNA, nucleic acid such as an RNAi mediator molecule, particularly a locked nucleic acid, or a plasmid such as phosphorothioate, DNA, DNA fragment, minicircle DNA, a drug such as a small molecule, particularly any one or more of a chemotherapeutic drug or a senolytic drug.

[0033] Specifically, the source cell or source cell mixture is of eukaryotic or prokaryotic origin (or is itself), preferably of body tissue, body fluid, or cell culture.

[0034] Specifically, the source cell or source cell mixture is obtained from a subject, and the TEV preparation is formulated for self-use. According to a specific embodiment, the present invention provides a TEV preparation obtainable or obtained by the method described herein.

[0035] According to a specific embodiment, the present invention provides an autologous TEV preparation obtainable or obtained by the method described herein, wherein the source cell or source cell mixture is obtained from a subject, and the TEV preparation is administered to the same subject.

[0036] The present invention further provides a medical use of such an autologous TEV preparation for use in treating a subject in need thereof, wherein the source cell or source cell mixture is obtained from the subject.

[0037] Specifically, the subject is a patient, particularly a human patient suffering from a disorder or disease. According to a specific embodiment, the present invention provides a protein comprising a target-specific extracellular domain (TED) of an extracellular vesicle (EV) surface protein obtainable or obtained by the method described herein.

[0038] According to a specific embodiment, the present invention provides a target-specific extracellular domain (TED) of an EV surface protein, comprising at least 70% sequence identity with the wild-type extracellular domain (ED) of a mammalian extracellular vesicle (EV) surface protein sequence, and at least one modified region having a length of 3 to 20 consecutive amino acids, wherein regions of the wild-type ED sequence are adjacent to the N-terminus and C-terminus of the modified region, the modified region is at least part of a target binding site not naturally present in the wild-type ED, the modified region has a lower target binding affinity when isolated from the TED, and / or the target binding site comprises at least one additional binding region within at least two amino acids away in a further modified region or within the wild-type ED sequence.

[0039] According to a specific embodiment, the present invention provides a target-specific extracellular vesicle (EV) surface protein (TSP) comprising at least one transmembrane domain, at least 70% sequence identity with the wild-type extracellular domain (ED) of a mammalian EV surface protein sequence, and at least one modified region having a length of 3 to 20 contiguous amino acids, wherein regions of the wild-type extracellular domain (ED) sequence are adjacent to the N-terminus and C-terminus of the modified region. The modified region is at least part of a target binding site that does not naturally exist in the wild-type ED, the modified region has a lower target binding affinity when isolated from the TSP, and / or the target binding site comprises at least one additional binding region within at least two amino acids away in a further modified region or within the wild-type ED sequence.

[0040] According to a specific embodiment, the present invention provides a polynucleotide encoding any one of the target-specific molecules described herein, particularly a protein comprising a TED described herein, a TED described herein, or any one of the TSPs described herein. Specifically, the polynucleotide is a cDNA molecule.

[0041] According to a specific embodiment, the present invention provides a target-specific extracellular vesicle (TEV) comprising a membrane and presenting a target-specific molecule on the outer surface of the membrane, wherein the target-specific molecule is any one of the target-specific molecules described herein, particularly a protein comprising a TED described herein, a TED described herein, or any one of the TSPs described herein.

[0042] According to a specific embodiment, the present invention provides a pharmaceutical preparation comprising any one of the target-specific molecules described herein, particularly a protein comprising a TED described herein, a TED described herein, or any one of the TSPs described herein, or a TEV described herein, and a pharmaceutically acceptable carrier, preferably for intradermal, subcutaneous, intravenous, topical, or oral use.

[0043] According to a specific embodiment, the present invention provides a library of target-specific extracellular vesicles (TEVs) comprising a variety of at least 10 2 individual target-specific extracellular vesicles (TEVs), wherein the diversity comprises or consists of TEVs having different modified regions with the same region of the same wild-type extracellular domain (ED) adjacent at the N-terminus and C-terminus of the modified region.

[0044] Specifically, the TEV library comprises a repertoire of TEVs comprising the target-specific molecules described herein, the repertoire covering at least 10 2 different modified regions or target binding sites.

[0045] According to a specific embodiment, the present invention a) providing a repertoire of polynucleotides encoding a variety of target-specific EV surface proteins (TSPs), each comprising a different target binding site; b) introducing said repertoire into said source cell(s); and b) isolating a fraction comprising a repertoire of target-specific extracellular vesicles (TEVs) having different target binding specificities to produce a library of TEVs A method for producing a library of target-specific extracellular vesicles (TEVs) is provided, which comprises mutating a polynucleotide encoding an EV surface protein by a mutagenesis method to obtain a mutation of the ED of the EV surface protein within at least one modified region having a length of 3 to 20 contiguous amino acids with regions of the wild-type extracellular domain (ED) sequence adjacent at the N-terminus and C-terminus of the modified region, and incorporating a target binding site within the ED, whereby a repertoire of polynucleotides is produced.

[0046] According to a specific embodiment, the present invention provides a TEV library obtainable or obtained by the method described herein, preferably comprising at least 10 2 individual TEVs having different target specificities.

[0047] According to a specific embodiment, the present invention provides a target-specific extracellular domain (TED) of an EV surface protein, comprising at least 70% sequence identity with the wild-type extracellular domain (ED) of a mammalian extracellular vesicle (EV) surface protein sequence, and at least one modified region having a length of 3 to 20 contiguous amino acids, wherein regions of the wild-type ED sequence are adjacent to the N-terminus and C-terminus of the modified region. The modified region is at least part of a target binding site not naturally present in the wild-type ED, and the TED is characterized as further described herein.

[0048] Specifically, the modified region has a lower target binding affinity when isolated from the TED. Specifically, the TED comprises at least any one of 70%, 80%, 85%, 90%, or 95% sequence identity with each wild-type ED sequence.

[0049] Specifically, the wild-type ED is derived from an EV surface protein further described herein. Specific EV surface proteins include a) tetraspanin-like proteins, b) proteins of the integrin family, c) proteoglycans, d) five-transmembrane domain proteins, e) type I transmembrane proteins, f) Notch family proteins, g) membrane proteins, particularly those having enzymatic activity (enzymatic TM proteins), h) immunomodulatory surface proteins, i) surface markers of mesenchymal stem cells, j) glycoproteins, k) channeling proteins, or l) a variety of exosome (vesicle) surface proteins any one (or a combination thereof) of the above.

[0050] According to a specific embodiment, the EV surface protein is a tetraspanin-like protein, particularly a tetraspanin or a lysosome-associated membrane protein. Specifically, the tetraspanin-like protein is preferably a) CD81 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 87; b) CD9 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 89; c) CD53 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 90; d) TSPAN32 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 91; e) CD82 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 92; f) CD63 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 93; g) CD151 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 94; and h) CD37 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 9 selected from the group consisting of those of the tetraspanin junctional complex superfamily, such as class I or II tetraspanins.

[0051] Specifically, the lysosome-associated membrane protein is LAMP2 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 96. Specifically, the originating protein is a human wild-type EV surface protein, or an artificial protein (or a wild-type protein of a non-human animal) comprising at least 90% sequence identity thereto.

[0052] Specifically, the originating ED is a wild-type ED originating from a human wild-type EV surface protein, or an artificial protein (or a wild-type protein of a non-human animal) comprising at least 90% sequence identity thereto.

[0053] Specifically, wild-type ED is that of a mammalian EV surface protein, such as one containing or consisting of a human or non-human animal amino acid sequence, etc. Specifically, wild-type ED includes at least any one of 90%, 95%, 98%, 99% sequence identity with any one of the ED amino acid sequences included in or composed of the wild-type EC sequence, particularly that of a human EV surface protein, or includes 100% sequence identity, which is a) that of CD81, wherein ED contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 130 or SEQ ID NO: 131; b) that of CD9, wherein ED contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 132, SEQ ID NO: 182, or SEQ ID NO: 133; c) that of CD53, wherein ED contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 134 or SEQ ID NO: 135; d) that of TSPAN32, wherein ED contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 136 or SEQ ID NO: 137; e) that of CD82, wherein ED contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 138 or SEQ ID NO: 139; f) that of CD63, wherein ED contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 140 or SEQ ID NO: 141; g) that of CD151, wherein ED contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 142 or SEQ ID NO: 143; h) that of CD37, wherein ED contains or consists of any one of the amino acid sequences specified as SEQ ID NO: 144 or SEQ ID NO: 145; or i) that of LAMP2, wherein ED contains or consists of the amino acid sequence specified as SEQ ID NO: 146 and is any one of them.

[0054] Specifically, depending on the method for determining the region of EV or ED that presents the EV surface protein, the sequence of ED can vary at one or both ends such that ED becomes longer or shorter by several amino acids, such as 1, 2, 3, 4, or 5 amino acids.

[0055] Specifically, ED can be of any non-human mammalian origin, such as those containing or consisting of the extracellular loop sequences of each EC or each non-human homolog.

[0056] Specifically, the mammalian wild-type tetraspanin-like protein contains at least one of 90, 95, 98, or 99% sequence identity with the mammalian wild-type tetraspanin-like protein, or 100% sequence identity.

[0057] According to a specific example, the EV surface protein is CD81, the modified region is located within the range of positions 160 to 172, and the numbering is that of human CD81 specified as SEQ ID NO: 87. Preferably, such a TED containing the modified region contains at least one additional binding region located between positions 132 and 141 or between positions 180 and 189, and the numbering is that of human CD81 specified as SEQ ID NO: 87.

[0058] According to another specific example, the EV surface protein is CD9, and the modified region is located within any one of the ranges of positions 155 to 166, 128 to 142, 130 to 140, or 169 to 180, and the numbering is that of human CD9 specified as SEQ ID NO: 89.

[0059] According to specific embodiments, the at least one modified region within the TED (in particular, within at least two separate regions or within the ED contained in the target binding site) contains solvent-exposed residues, and preferably, the at least one modified region is located in a loop and / or helix region of the ED. Solvent exposure at a position typically indicates the desirable accessibility for target binding. The assays for determining solvent exposure are solvent-accessible surface area and relative accessible surface area. Specifically, solvent-exposed residues are those having a relative accessible surface area greater than 20%.

[0060] According to further specific embodiments, the at least one modified region is located in an alpha-helix region of the ED that contains solvent-exposed residues. Specifically, the alpha-helix region contains a series of coils, such as a coil amino acid sequence, in particular a coil repeat sequence containing a repeating pattern of hydrophobic and charged amino acid residues, thereby forming a peptidic alpha-helix. Specifically, the helix region of the ED has a length ranging from 5 to 30 amino acids, preferably 7 to 18 amino acids. Specifically, at least one of the helix regions is part of the target binding site. Since helix regions typically tend to dimerize or multimerize, when the target binding site is contained in an ED monomer, the target binding sites appropriately prevent dimerization and multimerization respectively.

[0061] Specifically, the ED is of a tetraspanin such as CD81 or CD9, and the target binding site contains or is associated with at least one helix and / or loop structure of the tetraspanin, particularly within the LEL of CD81 and CD9 respectively. Specifically, the EV surface protein is CD81 or CD9, particularly human CD81 or CD9. Specifically, the surface protein is monomeric CD81 or monomeric CD9.

[0062] However, in some cases, the ED or EV surface protein incorporates the target binding site only when present as a dimer or multimer. In such cases, it is preferred to engineer the target binding site within the non-helical region to ensure dimerization and multimerization, as well as effective target binding, respectively.

[0063] Specifically, the placement of the target binding site varies among different types of ED or EV surface proteins. Certain motifs are conserved, while others can vary within a protein family or among similar sequences of different species. For example, helix D in the tetraspanin protein family is mostly unstructured and adopts a helical conformation only when bound to a specific antigen. Sequence alignments of tetraspanin family members show increased natural variability in this region, including insertions and deletions. Specifically, such natural variability indicates good tolerance for site-directed mutagenesis.

[0064] Specifically, the modified region of TED described herein is located within the loop region of the wild-type ED sequence, particularly within the large extracellular loop region. The EV surface protein can have a tertiary structure when attached to an EV containing a loop structure, exposing the loop to the vicinity of the vesicle. Such loop regions are particularly suitable for engineering the target binding site within ED.

[0065] For example, a tetraspanin-like molecule can have one or more small extracellular loops and / or one or more large extracellular loops (LELs). Exemplary LEL sequences of CD81 and CD9 are further disclosed herein. In the case of human CD81, the LEL sequence is specified as SEQ ID NO: 7, and in the case of human CD9, the LEL sequence is specified as SEQ ID NO: 118.

[0066] Specifically, the TED or EV surface protein described herein contains a loop structure in its amino acid sequence, and the loop structure is stabilized by one or more cysteines (plural possible) at one or more positions (plural possible) that allow the formation of one or more disulfide bonds. Specifically, the TED or EV surface protein containing the TED described herein contains at least one loop region, which is stabilized by at least one intramolecular bond, such as a disulfide bond, that connects at least two amino acid side chains.

[0067] The length of the loop of the ED or EV surface protein can vary. Specifically, the ED or EV surface protein contains at least one large extracellular loop and smaller-sized loops. The large extracellular loop typically has a length ranging from 75 to 140 amino acids, preferably 78 to 132 amino acids. Specifically, the small loop has a length ranging from 25 to 35 amino acids, preferably 26 to 32 amino acids.

[0068] Specifically, the ED or EV surface protein contains at least one helix region or domain, such as 1, 2, 3, or 4 helix regions. Specifically, the helix region is within the loop region or within the terminal region of the surface protein.

[0069] Specifically, the TED or EV surface protein containing the TED described herein exists as a monomer on the surface of the EV. Although extracellular or extracellular surface proteins tend to dimerize or oligomerize for biological functions, the surface protein is engineered to be target-binding as a monomer.

[0070] Specifically, at least one of the loop and / or helix regions is mutagenized to generate a modified region within the ED so as to form at least part of the target-binding site.

[0071] Specifically, the TED described herein is included in a target-specific EV (TEV) surface protein that contains at least one loop and / or helix structure typically fixed to the transmembrane domain of an EV, typically an EV.

[0072] Specifically, the modified region within the TED described herein has a length of 3 to 20 consecutive amino acids, preferably at least 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12, up to for example 20, 19, 18, 17, 16, 15, 14, 13, or 12. The modification typically results in at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or up to 20 point mutations within any such modified region.

[0073] Specifically, the modification includes substitution, insertion, or deletion of one amino acid at one position, preferably several point mutations including amino acid substitution, for example at least (or at most) any one of 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20.

[0074] Specifically, several point mutations at consecutive positions within the first modified region, for example a number ranging from 3 to 20, particularly 3, 4, 5, 6, 7, 8, 9, or 10; and optionally, several point mutations within the second modified region at consecutive positions distant from the first modified region, for example a number ranging from 3 to 20, particularly 3, 4, 5, 6, 7, 8, 9, or 10, are preferred, and both the first and second modified regions are included in the target binding site. The distance between the first and second modified regions is typically composed of adjacent sequences originating from the wild-type ED.

[0075] Specifically, in the modified region, regions of the wild-type ED sequence flank the N-terminus and C-terminus of the modified region, such that the flanking regions are proximal to each respective end of the modified region. Typically, the flanking regions are characterized by wild-type amino acid sequences of wild-type ED length, and the wild-type flanking sequences have a length of at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 contiguous amino acids that are identical to the respective (unmodified) sequences of wild-type ED.

[0076] However, according to specific embodiments, one (but not both) of the flanking regions may be absent. For example, the modified region may be a terminal region that includes the C-terminus of ED. In such a case, the modified region is flanked by a region of the wild-type ED sequence only at the N-terminus of the modified region, and the modified region is the C-terminal region of ED.

[0077] Specifically, when the modified region is located within TED, it is bound to the target and thus involves the secondary or tertiary structure surrounding TED for specifically recognizing the target. According to specific embodiments, the target binding site includes at least the modified region, as well as the wild-type region of TED, and / or another modified region that is within TED, within TSP, or on the surface of an EV that includes TED and / or TSP.

[0078] When isolated from TED or produced as a separate peptide consisting of the same amino acid sequence as the modified region, typically such an isolated modified region has a lower target binding affinity or even lacks specificity or selectivity for binding to the target. Compared to the target binding of the modified region within TED, the isolated modified region specifically has binding constants or binding kinetics that are at least 10-fold, or at least 100-fold, at least 1000-fold different, as determined using the same assay or in a comparable setting, and has a lower affinity or lower selective binding.

[0079] Suitable assays for comparing the target binding properties of the TEDs described herein and the modified regions isolated from said TEDs are any of the following assays: ELISA, such as Biacore, biolayer interferometry, fluorescence measurement of cells presenting the TED incubated with the cognate antigen, isothermal titration microcalorimetry, affinity determination using fluorescence correlation spectroscopy.

[0080] Specifically, the target binding site comprises at least two regions (at least one or two of which are modified regions further described herein), such as EDs separated by at least one transmembrane domain of an EV surface protein, within at least two different EDs of the same EV surface protein.

[0081] Such target binding sites incorporated within the TED and containing or alternatively accompanied by said at least one modified region within the same ED or within at least two EDs have the particular advantage of improved binding properties. Such improvement in binding properties is typically obtained by introducing mutations into a predetermined region within the ED such that the modified region has binding properties embedded within the ED and selecting the appropriate binding factor according to their binding specificity and / or affinity.

[0082] The binding properties are specifically improved compared to comparable fusions of specific (peptide) binding factors to the ED, thereby producing a fusion protein. This is because fusion typically changes the binding properties, such that comparable fusion proteins have lower affinity or lower selective binding upon fusion of the binding factor to the ED compared to the isolated binding factor.

[0083] Specifically, the modified region is at least part of or consists of the target binding site. According to a specific embodiment, the modified region comprises all the contact amino acid residues of the binding site.

[0084] According to specific embodiments, the target binding site includes a number of modified binding regions greater than one, and at least the first modified region is located at a specific distance from the second modified region. Specifically, the target binding site includes at least one additional binding region within at least one additional modified region separated from any one of 2, 3, 4, 5, 6, 7, 8, 9, or 10 consecutive amino acids or within the wild-type ED sequence. Specifically, the region between the at least two separated modified regions includes one of the regions adjacent to the modified regions described herein at the N-terminus or C-terminus of the modified region, which is a region of the wild-type ED sequence.

[0085] Specifically, the target binding site includes binding residues within at least two separated regions of the TED. Specifically, the TED is modified or mutagenized within the at least two separated regions to incorporate the target binding site.

[0086] According to specific embodiments, the target binding site includes contact amino acid residues within the ED but outside the modified region. Such additional contact amino acids may be located in one or more additional regions of the ED, and the one or more additional regions may include a modified (mutated) amino acid sequence containing, for example, one or more point mutations, or may include the wild-type sequence of the ED.

[0087] According to specific embodiments, the ED or the target binding site includes the at least one modified region and one or more additional point mutations, or specifically a series of at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 consecutive point mutations within an amino acid sequence at a specific distance, for example, at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids apart.

[0088] According to specific embodiments, the ED or the target binding site includes at least two or three of the modified regions. In a specifically preferred embodiment, the target binding site is a three-dimensional structural binding site that extends over at least two regions that are each at a certain distance from each other, for example, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acids apart, typically up to 100 or 80 amino acids apart, and involves a binding surface that extends over at least two non - contiguous regions within the same ED or at least two different EDs. Specifically, the separated binding regions are not adjacent to each other. Specifically, the separated binding regions are not continuous in the sequence of each EV surface protein.

[0089] Specifically, the binding site includes each of the separated binding regions, and optionally one or more additional contact points, or regions of TED. Specifically, the target binding site includes contact points within at least two modified (e.g., synthetic or mutagenized) regions, and optionally at least one additional region that is not synthetically or mutagenized and is a natural (wild - type) region of the ED.

[0090] Specifically, the target binds to the three - dimensional structural paratope of the binding site incorporated within the extracellular portion of the EV surface protein (especially the portion including TED described herein), and the paratope includes contact areas, such as loop and / or helix regions (s), in separated regions of TED. Appropriate TEDs and their respective EV surface proteins are conveniently produced by introducing mutations into selected areas within individual regions. For example, the TEDs described herein are mutagenized in one, two, or three separated regions.

[0091] Specifically, the target binding site is a novel binding site that recognizes a specific predetermined target that does not naturally exist in the wild - type ED or wild - type EV surface protein, also referred to herein as an "artificial" binding site.

[0092] According to specific embodiments, the target binding site comprises a new or additional binding site within one or more EDs, whereby the ED(s) has a new or additional specificity for binding to the target. Such a new target binding site may have the same specificity for binding to the same target as any naturally occurring binding site of the ED(s), or may have a different specificity, for example, recognizing a different target.

[0093] According to another specific embodiment, the target binding site can be a modified naturally occurring binding site of the ED(s), whereby the ED(s) has the same specificity for binding to the same target as the naturally occurring binding site (although the fine specificity or affinity can vary).

[0094] According to a specific aspect, a TED provided as a separate molecule, or a protein comprising such a TED (such as an EV surface protein or TSP), can comprise at least one or two different target binding sites that specifically recognize the same or different targets.

[0095] The TEDs described herein are specifically used for medical purposes, for example, to deliver a therapeutically effective amount of a protein comprising the TEDs described herein, or in particular a target-specific EV surface protein (TSP) comprising such TEDs when attached to an EV, thereby providing target-specific extracellular vesicles (TEVs) further described herein.

[0096] Specifically, the target is selected from the group consisting of cell targets, preferably mitogen receptors, cytokine receptors, asialoglycoprotein receptors, membrane transporters, lipoproteins, lipopolysaccharides, glycoproteins, proteoglycans, or acellular targets, preferably cytokines, artificial proteins, or artificial surface structures.

[0097] According to a specific aspect, the target is a human cell, for example, a cell originating from a healthy or diseased subject, or a cell lysate thereof. Human cells of a diseased phenotype are preferably used as the target.

[0098] According to certain embodiments, the target binding site specifically recognizes a novel target, i.e., a target that would not bind if the target binding site were a naturally occurring ED or EV surface protein that lacks the target binding site.

[0099] According to another embodiment, the target binding site specifically recognizes a target that is a natural ligand of an ED or EV surface protein, but with modified binding properties such as selectivity, fine specificity, affinity, and / or avidity, for example, for improved target binding. For example, natural ligands of tetraspanins are, for example, antigens or pathogens of pathogens such as cytopathic agents or viruses. By modifying the tertiary structure of the tetraspanin presented as a surface protein by TED, TSP, or TEV described herein, the binding properties can be improved to achieve, for example, an increase in the selectivity and / or affinity of binding that targets each pathogen.

[0100] Specifically, the target consists of an antigen or an antigenic structure of an antigen, particularly an epitope that is alternatively recognized by a target-specific antibody. Specifically, the target is a cell receptor. According to a specific example, an EV comprising a TED or TSP described herein that targets a cell receptor can fuse directly with the recipient cell membrane and thus incorporate its membrane proteins into the plasma membrane and deliver their cargo to the cytoplasm of the recipient cell.

[0101] Specifically, the target is an antigen, such as a naturally occurring antigen or a synthetic antigen. In certain embodiments, the target antigen is present in the blood of an affected patient, and the antigen is bound by the surface protein of the EV and is thus removed from the cardiovascular and / or lymphatic systems of the patient. Specific examples are undesired natural agents such as pathogens, toxins, diseased or cancerous cells, cytokines, or metabolites. In certain further embodiments, the target antigen is a synthetic antigen on the solid surface such as a graft or implant, or a soluble compound such as a chemical or synthetic compound, which can be removed when effective binding by the specific binding factors (e.g., TED, TSP, or TEV) described herein occurs.

[0102] The specific target can be a natural target typically recognized by wild-type EVs. However, the target binding sites included in the binding factors described herein can specifically recognize novel targets that are not recognized by wild-type EVs without them. Among the natural targets are pathogens such as viral antigens. The specific binding factors (e.g., TED, TSP, or TEV) described herein can bind to such natural targets through novel binding sites that can have improved binding characteristics such as binding affinity, avidity, or specificity.

[0103] According to a specific embodiment, the present invention provides a target-specific EV surface protein (TSP) comprising at least one ED comprising at least one transmembrane domain and at least 70% sequence identity with the wild-type extracellular domain (ED) of the mammalian EV surface protein sequence, and at least one modified region having a length of 3 to 20 contiguous amino acids, wherein regions of the wild-type ED sequence are adjacent at the N-terminus and C-terminus of the modified region, and the modified region is at least part of a target binding site not naturally present in the wild-type ED, and the TSP is characterized as further described herein.

[0104] Specifically, the modified region has a lower target binding affinity when isolated from the TSP. Specifically, the TSP described herein is characterized by one or more extracellular domains, at least one of which is target-binding ("target-specific"). Specifically, the ED of the described TSP is the TED described herein.

[0105] Specifically, the at least one transmembrane domain is that of a vesicular membrane protein or an artificial transmembrane domain produced, for example, by mutagenesis of a wild-type transmembrane domain or by de novo synthesis of an appropriate amino acid sequence.

[0106] Specifically, the transmembrane domain (TM) comprises at least any one of 70%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% sequence identity with the respective wild-type transmembrane domain sequence of a mammalian EV surface protein, such as any of the EV surface proteins further described herein, or 100% sequence identity.

[0107] Specifically, the wild-type TM is that of a mammalian EV surface protein, such as one comprising or consisting of a human or non-human animal amino acid sequence. Specifically, the TM is that of a vesicular membrane protein that is of exosome, microvesicle, or apoptotic body origin, particularly a wild-type exosome, microvesicle, apoptotic body protein, or one of such wild-type proteins that includes a modification in the extracellular region, for example.

[0108] Specifically, the wild-type TM comprises at least any one of 90%, 95%, 98%, 99% sequence identity with any one of the TM amino acid sequences included in or composed of the wild-type sequence of an EV surface protein that can be incorporated into the membrane of a cell or vesicle when the EV surface protein, particularly a human EV surface protein, is bound to the respective cell or vesicle, or 100% sequence identity.

[0109] Specifically, the transmembrane domain is a) CD81 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 147, SEQ ID NO: 148, SEQ ID NO: 149, or SEQ ID NO: 150; b) CD9 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 151, SEQ ID NO: 152, SEQ ID NO: 153, or SEQ ID NO: 154; c) CD53 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 155, SEQ ID NO: 156, SEQ ID NO: 157, or SEQ ID NO: 158; d) TSPAN32 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 159, SEQ ID NO: 160, SEQ ID NO: 161, or SEQ ID NO: 162; e) CD82 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 163, SEQ ID NO: 164, SEQ ID NO: 165, or SEQ ID NO: 166; f) CD63 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 167, SEQ ID NO: 168, SEQ ID NO: 169, or SEQ ID NO: 170; g) CD151 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 171, SEQ ID NO: 172, SEQ ID NO: 173, or SEQ ID NO: 174; h) CD37 in which the transmembrane domain comprises or consists of any one of the amino acid sequences specified as SEQ ID NO: 175, SEQ ID NO: 176, SEQ ID NO: 177, or SEQ ID NO: 178; or i) LAMP2 in which the transmembrane domain comprises or consists of the amino acid sequence specified as SEQ ID NO: 179 and is any one of the above.

[0110] Specifically, depending on the method for determining the region of the transmembrane domain of the EV presenting the EV surface protein, the sequence of the transmembrane domain can vary at one or both ends such that the transmembrane domain becomes longer or shorter by several amino acids, for example 1, 2, 3, 4, or 5 amino acids.

[0111] According to a specific embodiment, both the wild-type ED and the at least one transmembrane domain are preferably a) CD81 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 87; b) CD9 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 89; c) CD53 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 90; d) TSPAN32 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 91; e) CD82 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 92; f) CD63 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 93; g) CD151 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 94; h) CD37 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 95; and i) LAMP2 comprising or consisting of the amino acid sequence specified as SEQ ID NO: 96 and are derived from the same mammalian EV surface protein selected from the group consisting of.

[0112] Specifically, the EV surface protein is derived from exosomes, microvesicles, or apoptotic bodies, particularly wild-type exosomes, microvesicles, apoptotic body proteins, or vesicle membrane proteins such as those of such wild-type proteins containing modifications in the extracellular region.

[0113] Specifically, the EV surface protein is a protein derived from cell exosomes, i.e., an exosome protein. Specifically, the EV surface protein is a protein derived from cell microvesicles, i.e., a microvesicle protein.

[0114] Specifically, the EV surface protein is a protein derived from apoptotic bodies, i.e., an apoptotic body protein. It is well understood that the exosome protein, microvesicle protein, or apoptotic body protein used herein for modification purposes is of cellular origin, i.e., it can be produced by the respective cells, or it is artificial and can be produced by de novo synthesis.

[0115] Specifically, the EV surface protein is a polypeptide, protein domain, or protein having a structure incorporating an artificial target binding site. Such a polypeptide, protein domain, or protein can be naturally occurring or can be partially or fully synthetic.

[0116] Specifically, one or two transmembrane domains are fused to the ED of the surface protein, and the ED can be attached to the membrane of the EV. According to a specific embodiment, the present invention provides a target-specific extracellular vesicle (TEV) comprising a lipid bilayer membrane and the TED described herein or the TSP described herein, presenting a target binding site on the outer surface of the lipid bilayer membrane.

[0117] Specifically, the TED described herein can be bound to the EV through the TSP described herein or any other suitable means, such as by conjugation, fusion, or affinity binding.

[0118] Specifically, the TSP described herein can be bound to the EV through the at least one transmembrane domain contained in the TSP, particularly the transmembrane domain(s) is / are within the vesicle membrane, whereby the TSP presents the TED and optionally at least one additional ED (which may or may not be target-binding) on the outer surface of the EV.

[0119] Specifically, said at least one transmembrane domain can act as an anchor that can be fused to any type of target-binding molecule incorporating a TED described herein, such as a TSP described herein.

[0120] Specifically, the TSP comprises at least two, three, or four transmembrane domains of the same wild-type EV surface protein, preferably all transmembrane membranes. Specifically, a number of transmembrane domains greater than one of a series of one or more (e.g., the same type or different) proteins, or artificial transmembrane domains, can be used, thereby creating an extracellular loop region, i.e., a loop structure outside the EV. Specifically, the number of transmembrane domains used to anchor at least one loop structure is two, three, or four. Specifically, the surface protein comprises one or more loops, e.g., at least 1, 2, 3, or 4 loops. However, in certain cases, the surface protein comprises a tertiary structure without loops.

[0121] Specifically, two transmembrane domains are used to present one extracellular loop structure on the surface of EVs containing the TSP described herein. According to specific embodiments, three or four transmembrane domains are used to present two extracellular loop structures, and four or five transmembrane domains are used to present three extracellular loop structures. Specifically, within one TSP described herein, following one transmembrane domain, a loop structure containing one or more intramolecular bonds, such as those spanning from the transmembrane domain to another transmembrane domain or stabilizing a loop or stem structure, etc., is connected to the vesicle surface by an amino acid sequence, etc., and the loop structure continues. It is understood that the loop structure may be adjacent to at least one transmembrane domain so as to bind to the vesicle membrane, for example, by either an N-terminal or C-terminal fusion of a peptide loop sequence. When only one transmembrane domain is adjacent, the opposite end (the N or C terminus of the loop sequence on the side opposite to the fusion with the transmembrane domain) is typically not fixed to the membrane, for example, is a released end. When two transmembrane domains are adjacent, for example, when both an N-terminal and a C-terminal fusion of a peptide loop sequence to the transmembrane domain are adjacent, the loop structure is fixed to the EV membrane on both sides.

[0122] Specifically, the TSP is a tetraspanin-like protein comprising several transmembrane domains spanning one or more loop sequences, particularly four transmembrane domains that can present the wild-type loop structure of the tetraspanin protein on the surface of EVs, particularly two extracellular loops fixed by four transmembrane domains.

[0123] Specifically, TEV is of eukaryotic or prokaryotic source cells from body tissues, body fluids, or cell cultures. According to specific embodiments, the TEV transports intracellular cargos. For example, the cargos include any one or more of peptides, polypeptides, protein domains, proteins, lipids, genes, mRNA, miRNA, nucleic acids such as RNAi mediators, particularly locked nucleic acids, or plasmids such as phosphorothioate, DNA, DNA fragments, minicircle DNA, drugs such as small molecules, particularly any one or more of chemotherapeutic drugs or senolytics drugs.

[0124] The EV can be a membrane vesicle, an ultramicroscopic vesicle containing a lipid bilayer membrane such as the membrane of exosomes, microvesicles, or apoptotic bodies. Specifically, the EV is produced by cells or by synthesis. When produced by cells, the EV can be released by the cells into the extracellular space by a biological process, and the surface proteins can be manipulated before or after cell release. The EV can be surface decorated, for example, using carbohydrate structures and / or by fusion with amino acids or amino acid sequences and / or by conjugation with chemical compounds such as drugs, labels, or tags.

[0125] Specifically, the lipid bilayer membrane is fixed to the TED or EV surface protein (TSP) through fusion of the at least one transmembrane domain of the membrane surface protein, each coding sequence, or by chemical and / or affinity binding. In certain cases, the surface protein is fixed to the lipid bilayer membrane of the extracellular vesicle by means other than transmembrane domains. Such anchors can involve, for example, affinity-based interactions such as streptavidin-biotin or protein A-Fc interactions, or covalent bond-based interactions such as maleimide-free cysteine. The fixation can be achieved, for example, by click chemistry conjugation.

[0126] Exemplary techniques for binding surface proteins to the membrane include S-S linkage (disulfide bridging), binding by bioconjugation, e.g., by click chemistry, or otherwise covalent bonding, including linking atoms of amino acid side chains. Alternative techniques employ affinity binding factors such as biotin and avidin to link surface proteins to the membrane.

[0127] In certain cases, one or more (artificial) target binding sites are included in one or two surface proteins of the TEV described herein (the same or different proteins). Specifically, two different surface proteins can be used to generate at least two different binding sites, e.g., recognizing different epitopes of the same or different antigens. Specifically, the TEV described herein can be mono- or bispecific, or even oligospecific.

[0128] In certain cases, the additional target binding sites of the TEV described herein can originate from any binding structure, such as proteins, polypeptides, or peptides including antibodies and antibody fragments, or complex molecules having a binding moiety. Specifically, the binding site can be that of the antigen-binding portion of an antibody, or of any one of the ligand-binding portions of an enzyme, adhesion protein, ligand, or receptor, and the binding site can bind to the cognate structure of the binding partner. The EV surface protein can particularly include the protein domain of an antibody or antibody fragment, or one, two, or more than two variable antibody domains, e.g., Fab, Fv, VH / VL dimer, scFc, dAb, F(ab)2-containing ones, etc., each antibody domain or fragment, or one or more binding sites of other biological binding factors such as soluble T cell receptors, Darpins.

[0129] A modified tetraspanin protein is specifically described herein that contains a novel target binding site characterized by at least any one of 75%, 80%, 85%, or 90% sequence identity, preferably reaching 80-90% sequence identity, with the amino acid sequence of the wild-type human tetraspanin protein LEL, such as the LEL specified as SEQ ID NO: 7 in the case of CD81, or the LEL specified as SEQ ID NO: 118 in the case of CD9.

[0130] Specific examples of the modified tetraspanin protein containing a novel target binding site are characterized by at least any one of 75%, 80%, 85%, 90%, 95%, 96%, 97%, or 98% sequence identity, preferably reaching 90-98% sequence identity, with the amino acid sequence of the full-length wild-type human tetraspanin protein, such as the amino acid sequence specified as SEQ ID NO: 87 in the case of CD81, or the amino acid sequence specified as SEQ ID NO: 89 in the case of CD9.

[0131] According to specific examples, the novel target binding site contains a modified region (binding region) within the LEL of CD81 specified as SEQ ID NO: 7, particularly within the region of aa113-201 of the full-length protein specified as SEQ ID NO: 87. A specific embodiment refers to the novel target binding site within aa130-201.

[0132] Specific examples refer to the modification in the amino acid sequence of CD81 (SEQ ID NO: 87) that substitutes at least one amino acid at positions 160-162 and 181-189. According to a further specific embodiment, the EV surface protein is CD9, and the binding residues are arranged within at least one or two different regions selected from the following regions: 155-166; 128-142; 130-140; 169-180. Specifically, CD9 is human CD9 specified by SEQ ID NO: 89. The preferred Cys residues for disulfide bridges are 132 and 140, and the numbering follows SEQ ID NO: 89.

[0133] According to a further specific embodiment, the EV surface protein is CD53, and the binding residues are arranged in at least two different regions selected from the following regions: 147-160; 121-134; 123-129; 164-169. Specifically, CD53 is human CD53 specified by SEQ ID NO: 90.

[0134] According to a further specific embodiment, the EV surface protein is TSPAN32, and the binding residues are arranged in at least two different regions selected from the following regions: 158-171; 130-144; 132-139; 174-183. Specifically, TSPAN32 is human TSPAN32 specified by SEQ ID NO: 91.

[0135] According to a further specific embodiment, the EV surface protein is CD82, and the binding residues are arranged in at least two different regions selected from the following regions: 153-172; 178-188; 128-137; 194-215. Specifically, CD82 is human CD82 specified by SEQ ID NO: 92.

[0136] According to a further specific embodiment, the EV surface protein is CD63, and the binding residues are arranged in at least two different regions selected from the following regions: 149-165; 171-176; 127-135; 179-189. Specifically, CD63 is human CD63 specified by SEQ ID NO: 93.

[0137] According to a further specific embodiment, the EV surface protein is CD151, and the binding residues are arranged in at least two different regions selected from the following regions: 159-177; 186-191; 135-145; 194-206. Specifically, CD151 is human CD151 specified by SEQ ID NO: 94.

[0138] According to a further specific embodiment, the EV surface protein is CD37, and the binding residues are arranged in at least two different regions selected from the following regions: 156-177; 183-206; 131-141; 218-229. Specifically, CD37 is human CD37 identified by SEQ ID NO: 95.

[0139] The EV surface protein or each ED (or any domain other than the transmembrane domain contained in the EV surface protein) can be used to introduce mutations in a predetermined region to introduce one or more novel target binding sites described herein.

[0140] Further EV surface proteins containing novel target binding sites can be produced by introducing mutations into each region. To identify suitable regions, the sequences of the large extracellular loops of tetraspanin class I (e.g., CD9, CD53, and TSPAN32) have been analyzed using the Swissmodel modeling server with a template such as CD81. The sequence of the large extracellular loop of CD82, a representative of tetraspanin class II, has been modeled using the program for protein structure and function prediction (i-Tasser modeling server). The sequences of tetraspanin class II (CD63, CD151, and CD37) have been modeled using the CD82 model coordinates.

[0141] According to specific embodiments, an EV surface protein (e.g., CD81 or any other tetraspanin) is stabilized or thermostabilized by introducing one or more cysteine(s) at position(s) that allow the formation of one or more additional disulfide bonds that stabilize the extracellular loop structure of the protein. Specifically, at least one additional cysteine is introduced into at least one region of the extracellular domain of the surface protein. Specifically, cysteines are introduced into at least two different regions of the surface protein. Introducing the novel cysteine(s) allows connection of regions via novel (artificial) disulfide bonds, which are normally formed by oxidation of the sulfhydryl (-SH) groups of cysteines.

[0142] In particular, the tertiary structure of the surface protein can be modified by one or more additional small loop(s) resulting from one or more additional intramolecular bonds. Such additional small loop(s) can be used to create additional contact points for an artificial binding site.

[0143] One or more additional intramolecular disulfide bond(s) have successfully increased the stability of the surface protein, as measured by a method for determining the thermal stability of proteins by standard methods. The stabilized EV surface protein can be conveniently used as a scaffold to generate a repertoire of EV surface proteins, each having different target binding properties or different target binding characteristics, by introducing mutations into a predetermined region within the EV surface protein. Such a repertoire can be appropriately used as a library to select binding factors to a target of interest.

[0144] Specifically, since the number of residues to be endosed is theoretically decisive for the free energy change with respect to the non-mutated protein, the candidate positions for introducing mutations into the cysteine pairs and the subsequent formation of cysteine bonds can be preselected in a first step by visual inspection of the crystal structure. Specifically, at least one new cysteine bond is introduced into the preselected candidates, preferably a new cysteine bond is created by the introduction of at least one or one pair of cysteines by mutagenesis.

[0145] A specifically preferred example refers to a modified tetraspanin containing a new disulfide bond that connects Cys residues when reduced, and the Cys residues are introduced by mutating the tetraspanin sequence (e.g., one of the ECs, especially the LEL of tetraspanin) at two distant sites, such as the N and C termini of the loop, thereby stabilizing the loop structure. Any such loop structure stabilized by at least one disulfide bond is preferably used to introduce mutations into the tetraspanin to engineer a new target binding site within such a loop structure.

[0146] Specifically, at least two cysteines are introduced either by insertion or substitution. According to a specific example, additional cysteines are introduced into CD81 to stabilize and / or modify the tertiary structure of the LEL.

[0147] According to a specific embodiment, the EV surface protein is CD81, and the amino acid sequence is modified to introduce cysteines that enable the formation of one or more disulfide bonds that do not naturally exist in the wild-type ED sequence, preferably between positions 134 and 144 and / or positions 130 and 146 and / or positions 135 and 168.

[0148] Specifically, CD81 is human CD81, the first cysteine is introduced at a position within the range of amino acids 120 and 200, at least the second cysteine is introduced at a position within the range of amino acids 143 and 201, and the numbering is that of human CD81 specified as SEQ ID NO: 87.

[0149] Specifically, the first cysteine is introduced at a position within the range of amino acids 130 and 140, and the second cysteine is introduced at a position within the range of amino acids 144 and 170. Specifically, cysteine is introduced at positions 134 and 144 in the human CD81 sequence (in particular, the ED of CD81 such as LEL), substituting with A134C and L144C respectively, thereby connecting helix A and helix B of the LEL of CD81 by an additional intramolecular disulfide bond. Alternatively, additional cysteines are introduced at positions 135 and 168 respectively, substituting with V134C and S144C respectively, thereby connecting helix A and helix C of the large extracellular loop of CD81 by an additional intramolecular disulfide bond.

[0150] According to a specific embodiment, Cys residues are introduced into the CD81 sequence specified as SEQ ID NO: 87 (in particular, the ED of CD81 such as LEL) by, for example, Ala134Cys and Lys144Cys mutations, thereby introducing a new disulfide bond spanning the cysteines at positions 134 and 144.

[0151] According to a further embodiment, (additional or alternative) Cys residues are introduced into the CD81 sequence specified as SEQ ID NO: 87 by Val135Cys and Ser168Cys mutations, thereby introducing a new disulfide bond spanning the cysteines at positions 135 and 168. According to a further embodiment, (additional or alternative) Cys residues are introduced into the CD81 sequence specified as SEQ ID NO: 87 by Ala130Cys and Ala146Cys mutations, thereby introducing a new disulfide bond spanning the cysteines at positions 130 and 146.

[0152] According to a further embodiment, an (additional or alternative) Cys residue is introduced into the CD81 sequence specified as SEQ ID NO: 87 by the Val135Cys and Ser168Cys mutations, thereby introducing a novel disulfide bond spanning the cysteines at positions 135 and 168.

[0153] Specifically, at least one of the EDs of human CD81, particularly the LEL, is modified to introduce additional cysteines at positions 134 and 144 and at positions 135 and 168, thereby connecting helix A to helix B and helix A to helix C. Such CD81 variants having the combination of the novel disulfide bonds Ala134Cys / Lys144Cys and Val135Cys / Ser168Cys that strongly stabilize in the LEL show an increase in the positive shift of the melting temperature of at least 20 °C.

[0154] Specifically, at least one of the EDs of human CD9, particularly the LEL, is modified to introduce an additional cysteine, thereby obtaining one or more new (additional) disulfide bridges. Preferably, the disulfide bridge connects positions 20 and 28, which can be obtained by mutating to Lys20Cys and Arg28Cys, and the numbering of the positions is that of the CD9 LEL (SEQ ID NO: 118). The sequence resulting from the stabilized variant specifically contains or consists of SEQ ID NO: 125.

[0155] Specifically, the increased thermal stability of the mutated preselected candidate is shown by an increase in the temperature at which thermal unfolding occurs by at least 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, or 55 °C compared to the wild-type protein.

[0156] Specifically, any one of TED, TSP, or TEV described herein is 10 -5 M or less, preferably 10 -6 M, 10-7 M or 10 -8 less than M or even 10 -9 K less than M D and has an affinity to bind to said target.

[0157] Generally, the binding factor is K D is regarded as a high-affinity binding factor having <10 nM, and in some cases, for example, for therapeutic purposes, a higher affinity, such as K D <1 nM or K D <0.1 nM or K D <0.01 nM or K D <pM (picomole = 10 -12 M) are provided.

[0158] Once it is proven that the selected TED, TSP, or TEV binds to the target of interest, the selected binding ED or EV surface protein can be affinity matured by standard methods of affinity maturation, such as those typically used to produce affinity matured antibodies. For this purpose, just a few point mutations, such as 1, 2, 3, 4, or 5, up to 10 point mutations, can be introduced within one region of the molecule or throughout the molecule to generate a new repertoire of target binding factors that can be selected to isolate binding factors with increased binding affinity. Such affinity matured binding factors can exhibit increased binding affinity with at least a 1 or 2 logarithm K D difference.

[0159] Specific binding can be determined in a suitable binding assay such as a conventional immunoassay. There are numerous methods known in the art for detecting binding in immunoassays. Various immunoassays known in the art can be used, including competitive and non-competitive assay systems using techniques such as radioimmunoassay, ELISA (enzyme-linked immunosorbent assay), immunoradiometric assay, gel diffusion precipitation reaction, immunodiffusion assay, Western blot, BIAcore, etc.

[0160] According to a preferred embodiment, the TEVs described herein are of eukaryotic or prokaryotic source cells. The source cells are understood herein as donor cells that can produce the EVs described herein. However, the source cells may simply use, for example, one or more components (such as transmembrane domains and / or surface proteins, etc.) that would otherwise be produced by the source cells, and serve as templates for constructing TEVs with the characteristics described herein without relying on cell transport mechanisms to produce the respective synthetic EVs in vitro.

[0161] Exemplary eukaryotes are mammals, plants, insects, fungi, or yeasts. Specific examples are cells of mammalian, plant, especially mammalian origin including human or non-human animals, particularly Chinese hamster-derived cells such as CHO cells, plants, especially Arabidopsis thaliana or Zea mays, or fungi, particularly Saccharomyces cerevisiae or Pichia pastoris.

[0162] Exemplary prokaryotes are bacteria. EVs derived from Gram-negative bacteria are known as outer membrane vesicles (OMVs). Specific examples are EVs of Lactobacillus or Mycobacteria pathogens, or Salmonella enterica, M. tuberculosis, Moraxella catarrhalis, or Haemophilus influenzae.

[0163] Specifically, the source cells are from body tissues, body fluids, or cell cultures, preferably from animal or plant cells; or from mammalian body fluids or tissues, preferably from blood, urine, amniotic fluid, ascites, cerebrospinal fluid, saliva, synovial fluid, or bone marrow. Specifically, the EVs described herein are produced by cell culture of the source cells.

[0164] Specifically, the tissue is from organs such as the kidney, brain, or placenta. Specifically, the tissue is tumor or metastatic tissue, or benign tissue. Specifically, the source cells are stem cells such as mesenchymal stem cells (MSCs), amniotic stem cells, or induced pluripotent stem (iPS) cells, dendritic cells, hematopoietic cells, epithelial cells, endothelial cells, nerve cells, blood cells, or immune cells. Specifically, the source cells of EVs can be amniotic-derived pluripotent progenitor cells, chorionic-derived mesenchymal stem cells, induced pluripotent stem cells, keratinocytes, fibroblasts, embryonic stem cells, ectodermal stromal cells, endodermal stromal cells, olfactory nerve sheath cells, dental pulp stem cells, or immortalized mesenchymal stem cells.

[0165] Specifically, the source cells are selected from the group consisting of normal or immortalized human cells such as induced pluripotent stem cells or adult stem cells, epithelial cells, and cancer cells. Specifically, the source cells are cell lines of recombinant host cells such as mammalian host cells, for example, human primary cells, telomerase-immortalized cell lines, or cell lines immortalized by a combination of viral oncogenes including adenovirus E1A, HPV-derived E6, EBV-derived oncogenes, SV40, or transcription factors, etc., which are cell lines used as cell factories. Specifically, cell lines including telomerase-immortalized endothelial cells or mesenchymal stem cells, HEK293, CHO, Vero, HEK, or CAP.

[0166] Cells appropriately employed in large-scale EV production include mesenchymal stem cells, dendritic cells, and HEK cells or 293T cells. Specifically, the source cells are mammalian stem cells or dendritic cells, preferably of human origin.

[0167] According to a specific embodiment, the EV surface protein is endogenous to the source cells. However, the endogenous EV surface protein is typically presented as a modified surface protein by the TEV described herein.

[0168] According to further specific embodiments, the EV surface protein is heterologous to the source cell. The heterologous EV surface protein can originate from different types of source cells or can be a synthetic surface protein that does not occur naturally. When using a synthetic surface protein, a novel target binding site can be synthesized intramolecularly without any further modification. Unlike a modified native surface protein, a synthetic surface protein typically does not have sequence identity with the native (wild-type) surface protein (e.g., less than 50% sequence identity).

[0169] Specifically, the TEV described herein has a size ranging from 10 to 1000 nm, preferably from 30 to 150 nm. Specifically, the TEV described herein has a buoyant density ranging from 1.0 to 1.4, preferably from 1.1 to 1.2 (g / cm 3 ), as measured by, for example, density gradient ultracentrifugation.

[0170] According to specific embodiments, the TEV described herein carries an intracytoplasmic cargo. Specifically, the cargo is within a volume range of 10 -14 ~10 -10 μl per EV. Specifically, the cargo contains an active substance or a mixture of active substances, for example, with a loading efficiency of 5 to 90%.

[0171] According to specific embodiments, the binding factors described herein (in particular, the TED, TSP, or TEV described herein) are provided for medical use in treating a subject in need thereof. The medical use includes treatment for a disease state by administering the TEV described herein or by ex vivo use as a reagent or affinity matrix, for example, for removing undesired substances from a body fluid. Further medical uses are for, for example, inducing an immune response to present an antigen to the subject's immune system for active immunotherapy and the like.

[0172] EVs can be used as a therapeutic agent in itself or as a delivery system for delivering specific payloads. According to a specific example, the intracellular payload is an active substance or drug encapsulated within the vesicle membrane. Specifically, active substances that act in conjunction with elements naturally present in EVs are used. According to another specific example, EVs may only act as a vehicle for reaching specific targets, which may be highly protected from conventional administration routes.

[0173] Specifically, the binding factors described herein are provided for any cosmetic, food, or industrial purpose. Specific embodiments refer to such TEVs provided in the form of a lipid or oily composition or encapsulated, for example, for cosmetic or food purposes. Industrial purposes include analytical or preparative purposes, such as analyzing or preparing (on an industrial scale) specific binding factors respectively.

[0174] Specifically, TEVs carry a payload containing autologous or heterologous active substances, particularly heterologous compounds, for medical use as a targeting vector in treating subjects in need of targeted therapy with a compound.

[0175] According to a specific aspect, the present invention provides a method of treating a subject in need thereof by administering to the subject an effective amount of the binding factors described herein, particularly the TEVs or TEV preparations described herein, for therapeutic or diagnostic purposes, to ameliorate or detect a particular condition, particularly a disease condition.

[0176] Specifically, an effective amount of TEV is used, and the dosage is measured according to the intracellular payload administered. Specifically, the carrier contains at least one autologous or heterologous compound. Specifically, the carrier contains any one or more of peptides, polypeptides, protein domains, proteins, lipids, genes, nucleic acids such as mRNA, miRNA, RNAi-mediated molecules, especially locked nucleic acids, or plasmids such as phosphorothioates, DNA, DNA fragments, minicircle DNA, drugs such as small molecules, especially any one or more of chemotherapeutic drugs or senolytic drugs.

[0177] Small molecule drugs are understood herein as low molecular weight (<900D) organic compounds that can regulate biological processes. Small molecules can have diverse biological functions or applications and act as cell signaling molecules, drugs in medicine, insecticides in agriculture, and in many other roles. These compounds can be natural (such as secondary metabolites) or artificial (such as antiviral or chemotherapeutic drugs); they can have a beneficial effect on diseases (such as drugs) or be harmful (such as teratogens and carcinogens).

[0178] According to a specific embodiment, the TEV originates from a source cell that is autologous to the subject. Specifically, depending on the needs of the subject, autologous TEV that is modified and / or carried in vitro (outside the subject's body) for in vivo administration is used.

[0179] According to a specific embodiment, the present invention provides a TEV preparation containing isolated TEV. The TEV preparation is specifically characterized by a homogeneous population of EVs consisting of at least 50%, preferably 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 95% of at least any one of the EVs having the same target specificity.

[0180] According to a specific embodiment, the TEV preparation is a homogeneous preparation having a median diameter of 100 - 150 nm or 120 - 140 nm. The specific yield is preferably at least 1000 EVs per source cell, more preferably at least 1500 or at least 1600 EVs per source cell.

[0181] Specifically, the TEV preparation is provided as a stable aqueous solution for storage or as a lyophilized preparation. The present invention further provides a pharmaceutical preparation comprising any one of the TED or TSP or TEV described herein and a pharmaceutically acceptable carrier, preferably in a formulation for intradermal, subcutaneous, intravenous, topical, or oral use.

[0182] According to a specific embodiment, the present invention modifies a polynucleotide comprising a coding sequence encoding the extracellular domain (ED) of an extracellular vesicle (EV) surface protein by a mutagenesis method, such that the regions of the wild-type ED sequence are adjacent to the N-terminus and C-terminus of the modified region, obtaining a mutation in the ED amino acid sequence within at least one modified region having a length of 3 to 20 consecutive amino acids, incorporating a target binding site into the ED, thereby producing a repertoire of polynucleotides encoding diverse target-specific extracellular domains (TEDs) each containing a different target binding site, as well as selecting a protein comprising a TED that specifically recognizes a predetermined target, and producing the selected protein, thereby providing a method for producing a protein comprising a TED of an EV surface protein.

[0183] Specifically, the repertoire of polynucleotides presents diverse TEDs on the outer surface and is preferably included in a genetic package employing a display system selected from the group consisting of yeast, phage, bacteria, ribosome, mRNA, or mammalian cell display.

[0184] Specifically, the protein comprising the selected TED is the TSP described herein. Specifically, the selected TED is characterized as further described herein.

[0185] The present invention a) introducing a polynucleotide encoding the TSP described herein into a source cell or a source cell mixture; b) culturing the cell(s) under conditions that produce extracellular vesicles; c) isolating a fraction containing TEV that includes the target binding site of TSP; and d) producing a preparation of TEV contained in the fraction Further provided is a method for producing a TEV preparation described herein, which includes the above steps.

[0186] Specifically, the source cell or source cell mixture is obtained from a biological sample of interest. Specifically, the biological sample of interest is selected from the group consisting of blood, urine, amniotic fluid, ascites, cerebrospinal fluid, saliva, synovial fluid, or bone marrow.

[0187] Specifically, the source cell is isolated before culturing in cell culture or cultured within the biological sample. Specifically, the subject is an animal such as a mammal including a human or a non-human animal.

[0188] According to a further specific embodiment, a) introducing a polynucleotide or gene encoding a TEV surface protein into a stem cell; b) culturing the stem cell under conditions that produce extracellular vesicles; c) isolating a fraction containing TEV, for example according to target binding specificity; and d) producing a TEV preparation A method for producing a TEV preparation described herein that is derived from stem cells of a subject is provided, which includes the above steps.

[0189] Specifically, the stem cell is isolated before culturing in cell culture or cultured within the biological sample. According to a specific embodiment, EVs are obtained from mesenchymal stem cells (MSCs). MSCs can be prepared by in vitro proliferation of cell cultures, for example, by dispersing embryonic stem cell colonies. Isolation of EVs, particularly exosomes, from MSCs can be performed in mesenchymal stem cell-conditioned medium. The medium can be obtained by culturing MSCs, their progeny, or cell lines derived therefrom in a cell culture medium and isolating the cell culture medium.

[0190] Specifically, the source cell or source cell mixture is obtained from a subject, and the TEV preparation is formulated for autologous use. Specifically, provided herein is an autologous TEV preparation produced by the methods described herein, wherein the source cell or source cell mixture is obtained from a subject and the TEV preparation is administered to the same subject.

[0191] According to a specific embodiment, TEVs can be targeted to a tumor, and TEVs carrying an antigen directly obtained from the tumor can be produced. Specifically, the polynucleotide or gene introduced into the source cell or source cell mixture encodes a TSP, and the ED of the TSP is bound to the surface of the EV via at least one of the transmembrane domains of the TSP.

[0192] According to a specific embodiment, introduction of the polynucleotide or gene encoding TSP into the source cell can be achieved by transfection. Specifically, the gene is transfected into the cell(s) before culturing. Specifically, the coding gene is introduced into the cell by any of the commonly used transfection methods, such as electroporation, or transfection using an apoptosis inducer such as siRNA, particularly liposome-based transfection.

[0193] According to a specific embodiment, the source cell or source cell mixture is cultured in cell culture under conditions that produce membrane vesicles and release TEVs, thereby obtaining TEVs in the culture supernatant.

[0194] Specifically, the cell culture conditions are adapted to various source cells or various biological samples containing source cells. According to a specific embodiment, the biological sample consists of a biological fluid (bone marrow, peripheral blood, etc.) derived from a subject, a culture supernatant, a cell lysate, a pre-purified solution, or any other composition containing membrane vesicles. The specific cell culture method for the production of TEV may further involve a step of inducing oxidative stress. The oxidative stress can be induced by externally added cytokines or by oxidants such as hydrogen peroxide.

[0195] Exosomes can also be synthesized or artificially manufactured, that is, they are not isolated from human or non-human cells. Instead of being isolated, exosomes can be synthesized by various lipid formation techniques.

[0196] Specifically, the source cells or source cell mixture are preferably cultured in a cell culture containing an active substance, such as a heterologous compound, under conditions that produce extracellular vesicles that carry the compound by intracellular loading into the vesicles.

[0197] Specifically, the intracellular loading is by incubation, optionally by disrupting the membrane, or by binding the cargo to a membrane component. In particular, TEV is carried by any suitable transfection technique, including reagent-based methods (calcium phosphate, polyethyleneimine, cationic polymers, DEAE-dextran, activated dendrimers, magnetic beads) or device-based methods (electroporation, sonication, biolistic technology, microinjection, laserfection, optoinjection). Alternatively, TEV can be carried by binding or fusing a compound to the membrane, for example, by binding to or fusing with a membrane lipoprotein.

[0198] The step of carrying TEV can be performed in vitro, in vivo, or ex vivo. TEV can be carried either before or after the production of extracellular vesicles.

[0199] Specifically, the source cells can be further surface decorated by any suitable method that employs biological, enzymatic, and / or chemical reactions for producing TEV including modifications, for example, the step of modifying surface protein glycosylation (e.g., by sialylation, fucosylation, or deglycosylation), post-translational modification, and / or the step of conjugating chemical compounds, drugs, labels, tags, or enzymatic (e.g., enzyme substrates) or chemical reactive groups.

[0200] Regarding the isolation of TEV, particularly microvesicles or exosomes, the medium of cell culture of the source cells is recovered, cells and debris are pre-purified, and subjected to a series of (ultra)centrifugation steps. Subsequently, the resulting TEV pellet is usually subjected to sucrose density gradient centrifugation to separate a homogeneous EV population. Specifically, the culture supernatant is treated to be enriched with membrane vesicles. In particular, a pre-purified solution obtained from the culture supernatant of a population of membrane vesicle-producing cells or from a biological sample is subjected to treatments such as centrifugation, purification, ultrafiltration, nanofiltration, and / or affinity chromatography.

[0201] The cell culture medium or supernatant can be filtered, particularly by filtration using tangential force or ultrafiltration, through a membrane having, for example, a specific pore size or a specific molecular weight cut-off.

[0202] Specifically, the TEV-containing fraction is isolated and concentrated, if necessary, by any one or more of binding to an affinity ligand, centrifugation, chromatography, purification, ultrafiltration, or nanofiltration.

[0203] According to a specific embodiment, a method of preparing TEV, particularly purifying from a biological sample, includes at least one anion exchange chromatography step. Anion exchange chromatography can be carried out using various types of anion exchange materials, including cellulose, poly(styrene - divinylbenzene), agarose, dextran, acrylamide, silica, ethylene glycol - methacrylate copolymer, or a mixture thereof, such as an agarose - dextran mixture. The EV retained on the column can be eluted in various ways, particularly using the passage of an increasing concentration saline gradient. Typically, the various fractions purified in this way are detected by measuring their optical density at the column outlet using continuous spectrophotometric readings.

[0204] As an alternative or in addition to the anion exchange chromatography step, gel permeation chromatography can be used. Typically, the gel permeation chromatography step is carried out using a material selected from silica, acrylamide, agarose, dextran, ethylene glycol - methacrylate copolymer, or a mixture thereof, such as an agarose - dextran mixture.

[0205] The present invention further provides a TED library comprising at least any one of a variety of 10 2 、10 3 、10 4 、10 5 、or 10 6 TEDs, each having a different modified region with the same region of the same wild - type extracellular domain (ED) adjacent at the N - terminus and C - terminus of the modified region.

[0206] Preferably, the library of TEDs comprises 10 2 、10 3 、10 4 、10 5 、or 10 6includes at least any one of the individual TEDs. Specifically, the repertoire of the TED library is at least 2×10 6 10 7 2×10 7 10 8 or 2×10 8 individual TEDs.

[0207] The present invention further provides a TSP library including at least any one of a variety of 10 2 10 3 10 4 10 5 or 10 6 individual TSDs, each having a different modified region where the same region of the same wild-type extracellular domain (ED) is adjacent to the N-terminus and C-terminus of the modified region.

[0208] Preferably, the library of TSDs includes at least any one of 10 2 10 3 10 4 10 5 or 10 6 individual TSDs, each having a different modified region and / or target specificity. Specifically, the repertoire of the TSD library is at least 2×10 6 10 7 2×10 7 10 8 or 2×10 8 individual TSDs.

[0209] The present invention further provides a TEV library including at least any one of a variety of 10 2 10 3 10 4 10 5 or 10 6 individual TEVs, each having a different modified region where the same region of the same wild-type extracellular domain (ED) is adjacent to the N-terminus and C-terminus of the modified region.

[0210] Preferably, the library of TEVs comprises at least any one of 10 2 , 10 3 , 10 4 , 10 5 , or 10 6 TEVs, each having a different modification region and / or target specificity. Specifically, the repertoire of the library of TEVs comprises at least 2×10 6 , 10 7 , 2×10 7 , 10 8 , or 2×10 8 TEVs having different target specificities.

[0211] According to a specific embodiment, the present invention further provides a library of binding factors such as TED, TSP, or TEV described herein, produced by the method further described herein. Specifically, the method for producing any such library comprises introducing mutations into a nucleic acid sequence comprising a polynucleotide encoding the ED of an EV surface protein by a mutagenesis method, to obtain a mutation of the ED within at least one predetermined modification region having a length of 3 to 20 contiguous amino acids, wherein regions of the wild-type ED sequence are adjacent at the N-terminus and C-terminus of the modification region, and incorporating a novel target binding site into the ED or the EV surface protein.

[0212] The present invention a) providing a repertoire of polynucleotides encoding diverse target-specific extracellular vesicle (TSP) surface proteins, each comprising a different target binding site; b) introducing said repertoire into said source cell(s); and b) isolating a fraction comprising a repertoire of target-specific extracellular vesicles (TEVs) having different target binding specificities to produce a library of TEVs A method for producing a library of target-specific extracellular vesicles (TEVs), comprising introducing mutations into a polynucleotide encoding an EV surface protein by a mutagenesis method, such that regions of the wild-type extracellular domain (ED) sequence flank the N-terminus and C-terminus of the modified region, obtaining a mutation in the ED of the EV surface protein within at least one modified region having a length of 3 to 20 contiguous amino acids, and incorporating a target binding site into the ED, thereby producing a repertoire of polynucleotides, and further providing a method for producing a library of TEVs.

[0213] Specifically, the repertoire can be produced by modifying the polynucleotide or gene within at least a predetermined region of the ED by a mutagenesis method that produces the modified regions described herein. Specifically, the mutagenesis method is applied to obtain a randomized amino acid sequence within the modified region. Specifically, the mutagenesis method employs mutagenesis of at least one or two discrete regions of the ED or EV surface protein described herein, thereby producing a repertoire of polynucleotides encoding diverse EDs or EV surface proteins, each having a different binding specificity, and selecting a polynucleotide encoding a TED or TEV that specifically recognizes a target. Specifically, the repertoire of polynucleotides is included in a genetic package that presents diverse surface proteins on the outer surface of extracellular vesicles.

[0214] Specifically, a display package further described herein, such as a replicable genetic package encoding a surface protein described herein, is used, and the display package is presented on extracellular vesicles of natural or artificial origin, preferably selected from the group consisting of exosomes, microvesicles, nanoparticles, and liposomes.

[0215] Optionally, a polynucleotide encoding a TED or TSP is selected for producing each TED or TSP, respectively, in cell culture and / or for producing each TEV comprising the TED or TSP.

[0216] Introducing mutations into the surface protein or its respective coding sequences can be carried out with or without using EVs, such that one or more point mutations (plural possible) can be introduced, etc., within a predetermined region of the EV surface protein, particularly within such regions that can be presented on the outer surface of the EV or the cell (i.e., the extracellular or cell exterior portion of the surface protein).

[0217] Specifically, a repertoire of polynucleotides is generated that is included in a gene package presenting various surface proteins on the outer surface. Specifically, the display package is a replicable gene package encoding a surface protein described herein, such as those selected from the group consisting of bacteriophage, phagemid, and cell display packages, preferably bacteria, mammalian, or insect cells, or yeast, or in vitro display systems such as ribosome display systems. Such in vitro display systems specifically translate nucleic acid information into corresponding protein sequences presenting related binding specificities and affinities. Specifically preferred methods employ a display system selected from the group consisting of phage, yeast, bacteria, ribosome, mRNA, or mammalian cell display.

[0218] According to a specific example, the ED or EV surface protein described herein can be presented by a bacteriophage or phagemid via an anchor protein selected from the group consisting of the protein products of gene III, gene VI, gene VII, gene VIII, or gene IX. Gene III is often desirable because it is located at the N-terminus of the phagemid or phage.

[0219] According to another specific example, the ED or EV surface proteins described herein can be presented by yeast that routinely presents proteins, for example, by fusion to the Aga2 yeast surface protein. This is routinely used in yeast display libraries. Immobilization to yeast cells can also be achieved, for example, by co-expression of an anchor protein such as Protein A. Alternatively, the soluble surface protein can be secreted from yeast cells and captured after derivatization of the yeast cell surface by the external addition of Protein A or an antibody that specifically recognizes the desired target sequence.

[0220] For example, a series of display packages presenting various surface proteins with various binding properties can be provided. In particular, a series of display packages presenting various modifications of the same surface protein, for example, various modifications within the artificial binding site of the surface protein, can be provided to present such proteins with various binding properties.

[0221] In the methods described herein, the display library is specifically contacted with the target such that a series of members of the display library are selected according to their target binding properties. Among the members of the display library that recognize the target, one or more library members can be identified as having high target specificity and / or affinity. Such identification can be specifically understood as the step of selecting any kind of binding factor, that is, selection according to binding specificity.

[0222] The specific method enables co-selection according to both conjugation and functionality by using any of the following methods and means: Fluorescence-activated cell sorting (FACS) is routinely utilized for these investigations using any suitable target cells in combination with any detection antibody, binding protein, or dye such as annexin V that records binding or phenotypic changes such as apoptosis. Any similar high-sensitivity system can track and characterize binding events to target cells by means of an evolved microscopic sensitivity imaging or light detection system. FACS is also the main technique for fractionating binding factors and characterizing their corresponding phenotypic changes in cells.

[0223] In addition, there is a wide range of techniques used in high-throughput proteomics and cell biology, and its highly automated microscopic analysis is one of the most useful for this purpose. Improvement can be achieved using automated high-information-content digital microscopy that robotically analyzes individual cells in a sample.

[0224] Once an appropriate binding factor is selected, its coding sequence can be used to produce a surface protein, or a fusion protein comprising the TSP, such as a fusion of a transmembrane domain and the TED, can be engineered, and recombinant cells expressing such a protein on the surface of the vesicle membrane and / or cell membrane can be produced, or alternatively, a TEV as described herein comprising the TSP and / or presenting the TED on its outer surface can be produced.

[0225] Also provided herein is a method of selecting and isolating candidate binding factors that specifically recognize a given target from the libraries described herein by contacting the repertoire with the target under conditions that permit specific target binding, and selecting candidate binding factors having demonstrated target binding specificity.

[0226] In the method described herein, the library is targeted, for example, specifically contacted with a target antigen or target cell so that a member of the library binds to the target. Among the members of the library that recognize the target, one or more library members can be identified as having high target specificity and / or affinity. Such identification can be specifically understood as a step of selecting any kind of binding factor, that is, selection according to binding specificity.

Brief Description of the Drawings

[0227]

Figure 1-1

Figure 1-2

Figure 1-3

Figure 1-4

Figure 1-5

Figure 1-6

Figure 1-7

Figure 1-8

Figure 1-9

Figure 1-10

Figure 1-11

Figure 1-12

Figure 1-13

Figure 1-14

Figure 1-15

Figure 1-16

Figure 1-17

Figure 1-18

Figure 1-19

Figure 1-20

Figure 1-21

Figure 1-22

Figure 1-23

Figure 1-24

Figure 1-25

Figure 1-26

Figure 1-27

Figure 2

Figure 3

Figure 4

Figure 5-1

Figure 5-2

Figure 6

Figure 7-1

Figure 7-2

Figure 8-1

Figure 8-2

Figure 9-1

Figure 9-2

Mode for Carrying Out the Invention

[0228] Detailed Description Unless otherwise specified or defined, all terms used herein have their ordinary meanings in the art that would be apparent to one of ordinary skill in the art. References are made, for example, to standard manuals such as Sambrook et al., "Molecular Cloning: A Laboratory Manual" (2nd Edition), Volumes 1-3, Cold Spring Harbor Laboratory Press (1989); Lewin, "Genes IV", Oxford University Press, New York, (1990), and Janeway et al., "Immunobiology" (5th Edition, or more recent editions), Garland Science, New York, 2001, etc.

[0229] The subject matter of the claims specifically refers to artificial products or methods of employing or producing such artificial products, which may be variants of natural (wild-type) products. The materials, methods, and uses of the present invention, which specifically refer to, for example, isolated nucleic acid sequences, amino acid sequences, expression constructs, transformed host cells, and recombinant proteins, although there may be a certain degree of sequence identity with natural structures, are "artificial" or synthetic and are therefore not considered as a result of "laws of nature".

[0230] The term "domain" with respect to a protein domain such as an ED or transmembrane domain is understood herein as a polypeptide or protein of a continuous amino acid sequence that is at least a particular portion (or the full length) of a polypeptide or protein. A domain may be contained within a larger protein. However, a protein domain is also called a domain and is isolated from a protein larger than the domain.

[0231] The term "expression" is understood as follows. For example, a nucleic acid molecule containing a desired coding sequence of an expression product such as an antibody described herein, and a control sequence such as a promoter in operable linkage, can be used for the purpose of expression. A host transformed or transfected with these sequences can produce the encoded protein. To cause transformation, an expression system can be included in a vector; however, the relevant DNA can also be integrated into the host chromosome. Specifically, the term refers to a host cell and a compatible vector under appropriate conditions for the expression of a protein encoded by foreign DNA, for example, carried by a vector and introduced into a host cell.

[0232] Coding DNA is a DNA sequence that encodes a specific amino acid sequence for a specific polypeptide or protein such as an antibody. Promoter DNA is a DNA sequence that initiates, regulates, or alternatively mediates or controls the expression of coding DNA. Promoter DNA and coding DNA can be from the same gene or different genes and can be from the same or different organisms. A recombinant cloning vector often contains one or more replication systems for cloning or expression, one or more markers for selection in a host, such as antibiotic resistance, and one or more expression cassettes.

[0233] As used herein, an "expression vector" or "vector" is defined as a cloned recombinant nucleotide sequence, i.e., a DNA sequence required for the transcription of a recombinant gene and the translation of its mRNA in a suitable host organism.

[0234] An "expression cassette" refers to a DNA coding sequence or segment of DNA encoding an expression product that can be inserted into a vector at defined restriction sites. Cassette restriction sites are designed to ensure the insertion of the cassette in the proper reading frame. Generally, foreign DNA is inserted into one or more restriction sites of the vector DNA and then carried by the vector into the host cell together with the transmissible vector DNA. A segment or sequence of DNA having inserted or added DNA, such as an expression vector, may also be referred to as a "DNA construct".

[0235] An expression vector contains an expression cassette and usually additionally includes an origin for autonomous replication in a host cell or a genomic integration site, one or more selectable markers (e.g., an amino acid synthesis gene, or a gene conferring resistance to an antibiotic such as zeocin, kanamycin, G418, or hygromycin), several restriction enzyme cleavage sites, a suitable promoter sequence, and a transcription terminator, and the components are operably linked together. As used herein, the term "vector" includes nucleotide sequences that replicate autonomously as well as genomic integration-type nucleotide sequences. A commonly seen type of vector is a "plasmid", which is generally a self-contained molecule of double-stranded DNA that can readily accept additional (foreign) DNA and be readily introduced into a suitable host cell. Plasmid vectors often contain coding DNA and promoter DNA and have one or more restriction sites suitable for inserting foreign DNA. Specifically, the term "vector" or "plasmid" refers to a vehicle that can introduce a DNA or RNA sequence (e.g., a foreign gene) into a host cell, transform the host, and promote the expression (e.g., transcription and translation) of the introduced sequence.

[0236] The term "extracellular vesicle" abbreviated as "EV", including such EVs with target binding specificity such as TEV described in this specification, is understood herein as a cell vesicle or such a vesicle of cell origin provided outside the cell. EVs can be produced in vivo or ex vivo by their respective source or donor cells, and can also be artificially produced by in vitro methods that manipulate liposomes or nanoparticles to produce synthetic EVs containing, for example, EV properties described herein without using cells.

[0237] EVs are membrane-packed vesicles typically secreted by a variety of cell types including T cells, B cells, dendritic cells, platelets, mast cells, epithelial cells, endothelial cells, neurons, cancerous cells, oligodendrocytes, Schwann cells, embryonic cells, and MSCs. EVs also naturally exist in physiological fluids such as normal urine, blood, bronchoalveolar lavage fluid, breast milk, saliva, cerebrospinal fluid, amniotic fluid, synovial fluid, and malignant ascites. EVs have been demonstrated to play an important role in cell-cell communication. They mediate cell-cell communication and enable the transfer of functional nucleic acids from the originating cell to recipient cells. They are involved in processes such as immune response, homeostasis maintenance, coagulation, inflammation, cancer progression, angiogenesis, and antigen presentation. Therefore, EVs participate in many physiological and pathological states.

[0238] EVs can contain biomolecules or synthetic cargo referred to herein as "payloads". Therefore, they create an attractive delivery vehicle for targeted therapies or diagnostic methods due to their stability, biocompatibility, low immunogenicity, and toxicity profile in circulation. EVs can specifically transport compounds between cells, including neurons. Advantageously, they can cross the blood-brain barrier. This natural trafficking ability gives extracellular vesicles the potential to be used as delivery vehicles for various self or heterologous compounds.

[0239] Exemplary EVs described herein are exosomes or microvesicles. Exosomes are a type of vesicles enclosed by an extracellular membrane and contain molecular components of the cells that secreted them.

[0240] Exosomes form one of the major subclasses of EVs and have an endosomal origin. Exosomal EVs are nanometer-sized vesicles of endocytic origin formed by budding of the limiting membrane of multivesicular endosomes (MVEs) inward. Thus, their size is comparable to that of intraluminal vesicles within MVEs, which generally ranges from 30 nm to 120 nm, preferably 50 - 100 nm.

[0241] The biogenesis of exosomal EVs occurs via the endocytosis-exocytosis pathway when cells absorb a small amount of intracellular fluid in specific membrane regions to form early endosomes. The early endosomes begin to mature and expand into late endosomes; then, intraluminal vesicles or multivesicular bodies (MVBs) are formed by internal budding of the endosomal membrane. The MVBs then fuse with the cell membrane and are released into the extracellular environment. At this point, the vesicles are named exosomes and are released via exocytosis, which is regulated by p53 and under the control of the cytoskeletal activation pathway but not affected by calcium. Exosomal EVs can have a diameter of 30 - 100 nm and a density of 1.13 - 1.19 g / mL in a sucrose gradient; they can be recovered by centrifugation, for example, at 100,000 g. After isolation, they can be stored as a lyophilized product or in an aqueous solution, for example, at room temperature or refrigerator temperature (2° - 8°C), or frozen (at -80°C or up to -18°C at the highest) for a period exceeding six months without any toxic cryoprotectant while maintaining their function.

[0242] Exosomal EVs may contain a large number of surface proteins such as annexins, tetraspanins such as CD63, CD81, and CD9, and heat shock proteins including Hsp60, Hsp70, and Hsp90. They also express Alix, tumor susceptibility gene 101 (Tsg101), and clathrin. Exosomal EVs specifically contain a lipid bilayer membrane that protects their contents and enables them to move long distances in tissues. The membrane typically possesses a small amount of phosphatidylserine as well as large amounts of cholesterol, ceramide, and sphingolipids.

[0243] Microvesicles are a type of membrane-enclosed vesicles derived from fragments of the plasma membrane. Microvesicular EVs typically bud from the cell surface, and their size can vary from 50 nm to 1000 nm. Artificial microvesicular vesicles such as semi-synthetic EVs and fully synthetic EVs can have sizes ranging from 10 to 1000 nm, preferably 10 to 500 nm, and even more preferably 10 to 100 nm.

[0244] Microvesicular EVs typically have a density of 1.04 - 1.07 g / mL in a sucrose gradient and are isolated by ultracentrifugation. Microvesicular EVs typically contain a large amount of phosphatidylserine-containing proteins associated with lipid rafts, are enriched in the surface marker CD40 as well as cholesterol, sphingomyelin, and ceramide. Specifically, they are also encapsulated in a lipid bilayer membrane and contain transmembrane proteins such as tetraspanins.

[0245] Typically, apoptotic body EVs are released by outward blebs and fragmentation of the cell membrane of apoptotic cells and have a broad size range with diameters of 50 - 2,000 nm. EVs typically interact with targets, such as target cells, via surface ligands and adhesion molecules. In some cases, they can enter cells via endocytic uptake or by direct fusion of vesicles with the cell membrane. They can also transmit their contents through adhesion to the cell surface mediated by lipid-ligand receptor interactions. These interactions indicate that EVs can play a central role in cell-cell communication and immune modulation in various physiological and pathological states.

[0246] Nanoscale EVs are an excellent alternative means for drug delivery. Since the composition of the EV membrane is typically derived from the source or donor cells (e.g., stem cells), these particles are inherently non-immunogenic, enabling them to resist rapid clearance from the circulation, thereby increasing the drug delivery efficiency to the target tissue. They are known to naturally possess specific cell tropism or homing ability by cell-type specific proteins, which is one of the important requirements for targeted drug delivery (using their surface ligands and adhesion molecules). However, natural targets are limited, and problems regarding affinity and specificity are commonly seen. By engineering EVs containing the surface proteins described herein, target-specific EVs can be provided that can be targeted with high specificity and affinity to any desired target or cell type.

[0247] The EVs described herein are particularly useful for medical purposes, for example, to diagnose or treat diseases or disease states, especially those diseases for which targeted therapies have been proven to improve the disease state.

[0248] EVs derived from mesenchymal stem cells, especially exosomal EVs, can be particularly useful regarding their use as a regenerative therapy. EVs originating from mesenchymal stem cells (MSCs) can carry bioactive molecules that can migrate to target cells and exert their therapeutic effects such as suppressing the inflammatory response that modulates the immune system and regenerating tissue injury, as well as many other beneficial effects. Therefore, EVs can be an effective, safe, and inexpensive therapeutic approach in cell-free regenerative medicine.

[0249] EVs can be a suitable drug delivery system, especially for crossing biological barriers (e.g., the blood-brain barrier) and delivering their payloads to sites that would otherwise be inaccessible. EVs can be formulated to exhibit the intended drug delivery activity by various methods, including biological, chemical, and physical means. Encapsulation of active substances or drugs (e.g., chemicals, RNAs, DNAs, proteins, or lipids) within EVs can significantly increase their bioavailability by maintaining their integrity and bioactivity in vivo. The lipid membrane derived from donor cells is suitable for avoiding phagocytosis, degradation, and modification in the host circulation. In particular, not only autologous but also heterologous EVs typically avoid capture in the reticuloendothelial system (also known as the mononuclear phagocyte system) and are non-immunogenic in most, if not all, parameters.

[0250] Various methods can be utilized to load active substance agents into EVs. These include (1) the step of loading purified EVs ex vivo, or (2) the step of pre-loading donor (source) cells prior to EV production, followed by isolation and purification as necessary in each case.

[0251] Ex vivo delivery strategies mostly utilize passive packaging of therapeutic molecules, ranging from simple incubation to more sophisticated chemical and / or physical methods. Hydrophobic (i.e., lipophilic) molecules such as antioxidants, anticancer drugs, and lipophilic dyes can be spontaneously packaged into EVs under ambient conditions. In fact, successful loading of curcumin, doxorubicin, and paclitaxel into EVs has been demonstrated. Compared to standard liposomes composed of phosphatidylcholine and cholesterol, EVs exhibit higher loading efficiency and capacity for hydrophobic chemical drugs.

[0252] Many active substances cannot freely penetrate the EV membrane, and thus EV loading typically occurs by means such as electroporation, sonication, permeabilization, fusogenic liposomes, polymeric carriers, and / or other means of physical damage. Sonication and extrusion, or permeabilization using saponin, have been shown to result in stable EV reformation with high loading efficiency.

[0253] Electroporation specifically applies an electric field to transiently create pores in the EV membrane, thereby enabling the movement of active substances into the EV lumen. Electroporation is also known to induce vesicle aggregation, thereby affecting vesicle integrity. One skilled in the art can select several parameters, including the EV source and concentration, the cargo molecule (payload), and the applied voltage using the time for optimal payload loading.

[0254] EVs are conveniently loaded by electroporation. Studies have demonstrated enhanced efficacy with a reduction in the harmful effects typically associated with chemotherapeutic drugs when compared to either drug-free EVs or drug-loaded liposomes.

[0255] EVs are natural carriers of various nucleic acid molecules, such as mRNA, miRNA, and various non-coding RNAs, or DNA molecules, and thus are suitable vehicles for nucleic acid translocation. Nucleic acid molecules are effective means for regulating genes of interest, but their low stability and transfection ability in circulation determine the need for a vehicle that can protect these therapeutic molecules and deliver them to target cells and tissues. Also, electroporation can be performed to load materials into EVs.

[0256] Sonication can be a suitable alternative for the active loading of molecules with minimal aggregation and degradation. There are various methods in the art for preloading drugs into donor cells prior to EV release. For example, the active substance can be incorporated into EVs derived from recombinant host cells that overexpress the protein or cell metabolite of interest. As described herein, EVs can be isolated from donor cells transfected with a heterologous gene in addition to a gene encoding a surface protein incorporating a target binding site. Since the cargo can include proteins, the loading of recombinant proteins expressed by host cells can be an attractive mode of protein delivery by the loaded EVs. Several model proteins, including ovalbumin, catalase, and glial cell line-derived neurotrophic factor (GDNF), have already been successfully loaded into EVs derived from genetically modified host cells.

[0257] The use of target-specific EVs described herein is a next-generation drug delivery system with the ability to cross complex biological barriers such as the blood-brain barrier, while avoiding or overcoming some safety concerns regarding drugs or vehicles, such as cytotoxicity, short in vivo distribution, and low efficiency of target delivery. Chemical drugs and biological molecules with low stability in circulation and / or low transfection ability into target cells can efficiently translocate into the cytoplasm of target cells without undergoing endosomal and lysosomal degradation.

[0258] As described herein, EVs can be produced to target a specific tissue, cell, artificial surface, or soluble compound by an artificial target binding site. EVs can be engineered for the purpose of expressing an appropriate surface protein and structure incorporating the binding site. For example, the surface protein can be overexpressed in the source cell so as to be expressed on the surface of the EV by employing an appropriate recombinant expression system in the source cell.

[0259] The artificial target binding site is engineered either before or after vesicle formation. For example, the binding site can be incorporated into the surface protein described herein in a predetermined region, such as to include loop, helix, and / or linear (peptide) structures. The specific target contact surfaces or binding residues in at least two separate regions of the surface protein can be generated in situ, i.e., by methods of recombining nucleic acid molecules such as by employing a method of introducing mutations into the source cell that produces point mutations in each respective surface protein, and / or by further modification of the EVs involving biological, enzymatic, and / or chemical reactions when producing the EVs.

[0260] The EVs described herein are suitably provided in the state of an EV preparation containing isolated EVs. The EVs can be characterized by certain specific traits that can be determined by appropriate quality control measurements, such as those that determine the size, density, amount, and composition of the cargo, target binding affinity and / or selectivity, purity, etc.

[0261] Exemplary methods of quality control are further described in the Examples section. The term "extracellular domain", abbreviated as ED, as described herein, is understood to encompass protein domains located on the outer surface (extracellular surface) of the EV when attached or bound to the EV. However, the protein domain is also referred to as ED, while being isolated from the EV or from the EV surface protein.

[0262] As used herein with respect to elements of an amino acid sequence or a nucleotide sequence, the terms "adjacent" or "flanked" are understood herein as follows: A first sequence element is said to be "flanked" by a second sequence element when the first sequence element is disposed in immediate proximity to the second sequence element, thereby providing a contiguous sequence of said first and second sequences. A linear first sequence element can be flanked by another element at only one of its ends or at both ends, i.e., on one or both sides.

[0263] In the TED described herein, the modified region is specifically flanked by two adjacent sequences, one at the N-terminus of the modified region and one at its C-terminus. Thus, such a modified region is located between the flanking sequences and is also referred to as "embedded".

[0264] As used herein, the term "host cell" refers to a primary target cell that has been transformed to produce a specific recombinant protein such as a surface protein described herein, or to produce an EV described herein and any progeny thereof. Although not all progeny may be exactly identical to the parental cell (due to intentional or inadvertent mutations or differences in the environment), such altered progeny are included in these terms as long as the progeny retain the same functionality as that of the originally transformed cell. The term "host cell line" refers to a cell line of host cells used to express a recombinant gene to produce a recombinant polypeptide or protein. As used herein, the term "cell line" refers to an established clone of a specific cell type that has acquired the ability to grow over an extended period of time. Such host cells or host cell lines can be maintained under cell culture and / or cultured to produce a recombinant polypeptide.

[0265] As used herein with respect to the modified region of ED, the term "isolated" or "isolation" refers to a peptide consisting of the amino acid sequence of the modified region that is sufficiently separated from the adjacent sequences of ED as long as it exists in a "substantially pure" form. "Isolated" does not necessarily mean the exclusion of an artificial or synthetic peptide, or a mixture of such a peptide with other compounds or materials, or the presence of impurities that may be present due to, for example, incomplete purification, which do not interfere with the basic binding activity. The term "isolated" is also intended to include those chemically synthesized.

[0266] In particular, the modified region can be isolated from the TED described herein, or can be isolated from the TSP described herein, or can be provided as each isolated peptide, for the purpose of comparing its binding properties such as target binding affinity and / or specificity, compared to the same modified region (not isolated therefrom) contained in TED and TSP, respectively.

[0267] As used herein with respect to the EV described herein, the term "isolated" or "isolation" refers to such vesicles that are sufficiently separated from the environment in which they would be naturally associated and exist in a "substantially pure" form. "Isolated" does not necessarily mean the exclusion of an artificial or synthetic mixture with other compounds or materials, or the presence of impurities that may be present due to, for example, incomplete purification, which do not interfere with the basic activity. In particular, the isolated EVs described herein are also intended to include those chemically synthesized.

[0268] With respect to the polypeptides or proteins such as EV surface proteins described herein, the term "isolated" specifically refers to compounds that do not contain or substantially contain the materials to which they are naturally associated, such as other compounds that are found together with them in their natural environment or in the environment (e.g., cell culture) in which such preparations are made if such preparations are by recombinant DNA techniques practiced in vitro or in vivo.

[0269] As used herein, a "library" with respect to the binding factors described herein, particularly TED, TSP, or TEV described herein, is understood to include a repertoire of binding factors that includes a number of different target-binding species (library members) that contain binding factors that vary to a certain degree. Libraries typically contain library members that can be distinguished by their functional binding and can thus be selected according to the desired binding properties.

[0270] The TED libraries described herein specifically include a set or collection of TEDs, particularly TEV as described herein, each having a different modified region embedded in the same wild-type sequence of the same wild-type ED.

[0271] The TSP libraries described herein specifically include a set or collection of TSPs, particularly TSP as described herein, each having a different TED embedded in the same wild-type sequence of the same wild-type EV surface protein. TSP libraries can include libraries of EDs, such as TED libraries. TED libraries are suitably produced by introducing mutations into ED or the EV surface protein, thereby producing a repertoire of ED and EV surface protein variants as soluble proteins or on the surface of EVs, respectively.

[0272] The TEV libraries described herein specifically include TEV, particularly the sets or collections of TEV described herein, each having different TEDs embedded in the same wild-type sequence of the same wild-type EV surface protein bound to the membrane of the vesicle. The TEV library may include a library of surface proteins, such as a TSP library.

[0273] The library can be constructed by well-known techniques involving appropriate methods of mutagenesis, such as site-directed mutagenesis of the extracellular domain of the surface protein. The libraries described herein preferably contain at least 10 2 library members, more preferably at least 10 3 members, more preferably at least 10 4 members, more preferably at least 10 5 members, more preferably at least 10 6 members, more preferably at least 10 7 members, more preferably at least 10 8 members, more preferably at least 10 9 members, more preferably at least 10 10 members, more preferably at least 10 11 members, and at most 10 12 members of the library.

[0274] Specifically, the library contains at least 10 2 , 10 3 , 10 4 , 10 5 , or 10 6 library members, and each library member has at least one nucleotide different in the sequence of the modified surface protein. Specifically, the library contains at least 10 6 library members, and each library member has a different target binding site or specificity.

[0275] For example, any protein or gene diversity library, including a number of individual library members, can be used for the purposes described herein to create diverse sequences or, for example, to employ a pre-selected library rich in stabilized or functionally active library members.

[0276] For example, a display system can conjugate a given protein, here a surface protein described herein, with its coding nucleic acid, such as its coding mRNA, cDNA, or gene. Thus, each member of the library contains a nucleic acid encoding a modified surface protein that is presented thereon. Display systems include, without limitation, cells, viruses such as phage, ribosomes, eukaryotic cells such as yeast, DNA including plasmids, and mRNA display.

[0277] As is well known in the art, there are various display and selection techniques that can be used to identify and isolate proteins having certain binding characteristics and affinities, including display techniques such as cell and cell-free methods, particularly mobilization display systems. Among cell systems, phage display, virus display, yeast, or other eukaryotic cell displays such as mammalian or insect cell display can be used. Mobilization systems are related to display systems in soluble form, such as in vitro display systems, among which ribosome display, mRNA display, or nucleic acid display.

[0278] A specific library is provided herein that presents diverse surface proteins and / or diverse surface proteins immobilized on EVs or cells. Preferably, the library is a phage display or yeast library. Specifically, yeast host cells present the surface proteins described herein on the surface of the yeast cells. Phage and phagemid display systems are well known for their versatility and potential to streamline the selection process. Yeast display confers several attractive features: the eukaryotic transcription and translation machinery is well-suited for protein expression, and the use of flow cytometry allows for high-throughput quantitative analysis of individual clones in real time using a scaffold ligand.

[0279] Yeast host cells are preferably selected from the genera Saccharomyces, Pichia, Hansenula, Schizisaccharomyces, Kluyveromyces, Yarrowia, and Candida. The most preferred host cell is Saccharomyces cerevisiae.

[0280] In certain cases, as in an RNA or DNA display library, the repertoire of surface proteins described herein is presented such that an entity containing the DNA, RNA, or cDNA encoding the surface proteins described herein can be directly linked to the surface protein it encodes. In such cases, surface protein variants are created by modification of methods of cell-free protein synthesis.

[0281] Screening for binding activity (or any other desired activity) in the library is performed according to methods well known in the art, for example, from phage display technology. For example, a target immobilized on a solid phase can be used to identify and isolate the binding members of the repertoire. Screening allows for the selection of members of the repertoire according to the desired characteristics.

[0282] In a method of selecting a suitable binding agent for a target, it is advantageous to provide a hyper-multiplicity of each binding agent, e.g., at least 10 copies, in the repertoire of library members to increase the chance of selecting one or more candidate binding sequences, which can be further characterized for suitability in engineering target-specific extracellular vesicle constructs.

[0283] The step of screening the library for library members containing a target binding structure can be performed by any suitable selection method. The screening step can include one or several rounds of selection (also referred to as panning).

[0284] One or several rounds of selection can include, for example, 1, 2, or preferably 3 rounds, and can include 4, 5, 6, 7, 8, 9, or 10 rounds of selection. In particular, a round of selection can include the step of incubating the library in the presence of the target to select proteins that bind to the target or an epitope thereof.

[0285] As used herein, the term "mutagenesis" refers to any technique recognized in the art for altering a polynucleotide or polypeptide sequence. Preferred types of mutagenesis include error-prone PCR mutagenesis, saturation mutagenesis, or other site-directed mutagenesis. Any known mutagenesis method can be employed to introduce point mutations at desired positions, e.g., by randomization techniques. In some cases, positions are randomly selected, e.g., using any possible or preferred amino acid, to randomize the antibody sequence.

[0286] As used herein, the term "recombinant" shall mean "prepared by or as a result of genetic manipulation." Alternatively, the term "engineered" is used. For example, surface proteins can be mutated to produce variants by manipulating each parental sequence to produce variants of the parental sequence. A recombinant host specifically contains a recombinant expression vector or cloning vector, or it has been genetically engineered to contain a recombinant nucleic acid sequence by adopting a nucleotide sequence that is particularly foreign to the host. Recombinant proteins are produced by expressing the respective recombinant nucleic acids in a host.

[0287] As used herein, "point mutation" is understood to be, in particular, an operation of a polynucleotide that results in the expression of an amino acid sequence different from the unengineered amino acid sequence in terms of substitution or exchange, deletion, or insertion of one or more amino acids for various amino acids at a particular position (one position) in the amino acid sequence. Specifically, one or more single (discontinuous) or double amino acid residues can be subjected to point mutations. Specifically preferred methods of introducing mutations provide point mutations at selected positions, preferably substitution of one amino acid by another at one (predetermined) amino acid position, or substitution of a plurality of amino acids only at a predetermined amino acid position. One or more point mutations can be within the modified region, particularly point mutations at discontinuous or contiguous positions within the region.

[0288] As used herein, the term "repertoire" shall refer to a collection of variants, such as variants of modified surface proteins, having diverse specificities that bind to a target with high affinity. Typically, the structure of a surface protein containing an extracellular domain having helices and loop regions is the same in such a repertoire. The diversity will specifically reflect the diversity of binding sites containing binding residues within one or more predetermined positions or regions to be modified, for example, within at least two distant regions that are part of the same target binding site that specifically recognizes and binds the target.

[0289] The repertoires described herein are specifically provided within a library that is a heterogeneous mixture of surface proteins, target-specific extracellular vesicle constructs, or targets. The library can take the form of a simple mixture of proteins or EVs, or in the form of isolated polypeptides or proteins, or nucleic acids encoding such polypeptides or proteins, each in the form of a binding region of such a modified surface protein, or further, organisms or cells expressing the nucleic acid, such as, for example, bacteria, viruses, animals, or plant cells transformed with a library of nucleic acids, etc., reflecting the diversity of target-specific binding factors of the repertoire.

[0290] As used herein, the term "subject" shall refer to warm-blooded mammals, particularly humans or non-human animals. Thus, the term "subject" can also refer to animals, including, in particular, dogs, cats, rabbits, horses, cows, pigs, and birds. In particular, the antibodies described herein are provided for medical use in treating subjects or patients in need of prevention or treatment of a disease state. The term "patient" includes human and other mammalian subjects undergoing either prophylactic or therapeutic treatment. The term "treatment" is thus intended to include both prophylactic and therapeutic treatment.

[0291] As used herein, the term "surface protein", including "EV surface protein", refers to a protein disposed on or at the surface of an EV that is immobilized within the lipid bilayer membrane of the EV. For this purpose, "immobilized" is understood herein to mean binding to the cell surface by fusion to a protein domain disposed within the membrane. Surface proteins are understood herein to include at least one ED and at least one transmembrane domain. Such surface proteins can be bound to the EV through incorporation of the at least one transmembrane protein domain into the membrane of the EV. The transmembrane domain(s) can be part of the surface protein or can be fused to the surface protein, for example, for the purpose of immobilizing the surface protein to the EV.

[0292] The term "EV surface protein" specifically includes "exosome proteins" that can be used to transport polypeptides or protein constructs to appropriate vesicular structures or EVs, and such proteins of EVs. Specifically, the term includes any protein that enables the transport, trafficking, or shuttling of polypeptides or protein constructs to vesicular structures such as EVs. Examples of such EV surface proteins include those identified by the sequences provided herein (in particular, in FIG. 1) (excluding signal sequences even if any), or one or more types of such EV surface proteins as described herein, including cloned isoforms, in whole or in part (as fragments), for example, isolated, synthetic, and / or recombinant amino acid sequences.

[0293] Exemplary EV surface proteins include a) i) Any one of the group consisting of tetraspanins such as CD81 (human CD81, SEQ ID NO: 87, etc.), CD9 (human CD9, SEQ ID NO: 89, etc.), CD53 (human CD53, SEQ ID NO: 90, etc.), TSPAN32 (human TSPAN32, SEQ ID NO: 91, etc.), CD82 (human CD82, SEQ ID NO: 92, etc.), CD63 (human CD63, SEQ ID NO: 93, etc.), CD151 (human CD151, SEQ ID NO: 94, etc.), CD37 (human CD37, SEQ ID NO: 95, etc.), TSPAN8 (human TSPAN8, SEQ ID NO: 184, etc.), TSPAN14 (human TSPAN14, SEQ ID NO: 185, etc.), or CD231 (TSPAN7) (human CD231 (TSPAN7), SEQ ID NO: 186, etc.); or ii) Lysosome-associated membrane proteins such as LAMP2 (human LAMP2, SEQ ID NO: 96, etc.) including tetraspanin-like proteins. b) Proteins of the integrin family, such as CD49d (human CD49d, SEQ ID NO: 187, etc.), ITGB5 (human ITGB5, SEQ ID NO: 188, etc.), ITGB6 (human ITGB6, SEQ ID NO: 189, etc.), ITGB7 (human ITGB7, SEQ ID NO: 190, etc.), CD71 (human CD71, SEQ ID NO: 191, etc.), CD29 (human CD29, SEQ ID NO: 249, etc.); c) Proteoglycans, such as CD138 (syndecan-1) (human CD138 (syndecan-1), SEQ ID NO: 192, etc.), syndecan-2 (human syndecan-2, SEQ ID NO: 193, etc.), syndecan-3 (human syndecan-3, SEQ ID NO: 194, etc.), syndecan-4 (human syndecan-4, SEQ ID NO: 195, etc.), HSPG2 (human HSPG2, SEQ ID NO: 196, etc.); d) Family of five-transmembrane domain proteins, such as CD133 (human CD133, SEQ ID NO: 195, etc.); e) Type I transmembrane proteins, such as (human CD50, SEQ ID NO: 196, etc.), CD102 (human CD102, SEQ ID NO: 197, etc.); f) Notch family, such as NOTCH1 (human NOTCH1, SEQ ID NO: 198, etc.), NOTCH2 (human NOTCH2, SEQ ID NO: 199, etc.), NOTCH3 (human NOTCH3, SEQ ID NO: 200, etc.), NOTCH4 (human NOTCH4, SEQ ID NO: 201, etc.), DLL1 (human DLL1, SEQ ID NO: 202, etc.), DLL4 (human DLL4, SEQ ID NO: 203, etc.), JAG1 (human JAG1, SEQ ID NO: 204, etc.), JAG2 (human JAG2, SEQ ID NO: 205, etc.), CD11a (human CD11a, SEQ ID NO: 206, etc.), CD11b (human CD11b, SEQ ID NO: 207, etc.), CD11c (human CD11c, SEQ ID NO: 208, etc.), CD18 / ITGB2 (human CD18 / ITGB2, SEQ ID NO: 209, etc.), CD41 (human CD41, SEQ ID NO: 210, etc.), CD51 (human CD51, SEQ ID NO: 211, etc.), CD61 (human CD61, SEQ ID NO: 212, etc.), CD104 (human CD104, SEQ ID NO: 213, etc.); g) Membrane proteins with enzymatic activity (enzymatic TM proteins), such as CD13 (human CD13, SEQ ID NO: 214, etc.), CD73 (human CD73, SEQ ID NO: 247, etc.); h) Immunomodulatory surface proteins, including, for example, Fc receptors, T cell receptors, complement receptors, interleukin receptors, immunoglobulins, MHC I, or MHC-II components; exemplary proteins include CD2 (human CD2, SEQ ID NO: 215, etc.), CD3 epsilon (human CD3 epsilon, SEQ ID NO: 216, etc.), CD3 zeta (human CD3 zeta, SEQ ID NO: 217, etc.), CD18 (human CD18, SEQ ID NO: 218, etc.), CD19 (human CD19, SEQ ID NO: 219, etc.), CD30 (human CD30, SEQ ID NO: 220, etc.), CD34 (human CD34, SEQ ID NO: 221, etc.), CD36 (human CD36, SEQ ID NO: 222, etc.), CD40 (human CD40, SEQ ID NO: 223, etc.), CD40L (human CD40L, SEQ ID NO: 224, etc.), CD44 (human CD44, SEQ ID NO: 225, etc.), CD45 (human CD45, SEQ ID NO: 226, etc.), CD47 (human CD47, SEQ ID NO: 227, etc.), CD86 (human CD86, SEQ ID NO: 228, etc.), CD110 (human CD110, SEQ ID NO: 229, etc.), CD111 (human CD111, SEQ ID NO: 230, etc.), CD115 (human CD115, SEQ ID NO: 231, etc.), CD117 (human CD117, SEQ ID NO: 232, etc.), CD125 (human CD125, SEQ ID NO: 233, etc.), CD135 (human CD135, SEQ ID NO: 234, etc.), CD184 (human CD184, SEQ ID NO: 235, etc.), CD200 (human CD200, SEQ ID NO: 236, etc.), CD279 (human CD279, SEQ ID NO: 237, etc.), CD273 (human CD273, SEQ ID NO: 238, etc.), CD274 (human CD274, SEQ ID NO: 239, etc.), CD362 = syndecan-2 (human, SEQ ID NO: 193, etc.), EGFR (human EGFR, SEQ ID NO: 240, etc.), L1CAM (human L1CAM, SEQ ID NO: 241, etc.), LFA-1 (human LFA-1, SEQ ID NO: 242, etc.), LGALS3BP (human LGALS3BP, SEQ ID NO: 243, etc.), MFGE8 (human MFGE8, SEQ ID NO: 244, etc.), SLlT2 (human SLlT2, SEQ ID NO: 245, etc.), STX3 (human STX3, SEQ ID NO: 246, etc.); i) Surface markers of mesenchymal stem cells, such as CD44 (human CD44, SEQ ID NO: 225, etc.), CD45 (human CD45, SEQ ID NO: 226, etc.), CD71 (human CD71, SEQ ID NO: 191, etc.), CD73 (human CD73, SEQ ID NO: 247, etc.), CD90 (human CD90, SEQ ID NO: 248, etc.), CD29 (human CD29, SEQ ID NO: 249, etc.), CD105 (human CD105, SEQ ID NO: 250, etc.), CD106 (human CD106, SEQ ID NO: 251, etc.), CD146 (human CD146, SEQ ID NO: 252, etc.), CD164 (human CD164, SEQ ID NO: 253, etc.), CD166 (human CD166, SEQ ID NO: 254, etc.), STRO-1 (human STRO-1, SEQ ID NO: 255, etc.); j) Glycoproteins, such as CD54 (human CD54, SEQ ID NO: 256, etc.), CD235a (human CD235a, SEQ ID NO: 257, etc.), CD106 (human CD106, SEQ ID NO: 251, etc.); k) Channeling proteins including Ca-channel proteins, such as GLUR2 (human GLUR2, SEQ ID NO: 258, etc.), GLUR3 (human GLUR3, SEQ ID NO: 259, etc.), HLA-DM (human HLA-DM, SEQ ID NO: 260, etc.); or l) A wide variety of exosome (vesicle) surface proteins, such as FLOT2 (human FLOT2, SEQ ID NO: 261, etc.) are as follows.

[0294] Additional diverse EV surface proteins are selected from TCRA, TCRB, TCRD, TCRG, and T cell receptors (T cell receptor loci) having variable amino acid sequences. Those skilled in the art can readily identify appropriate EV surface proteins based on the information provided herein or from respective databases, such as databases providing genomic loci and amino acid sequences of human EV surface proteins (fragments containing at least one ED or TM, or isoforms of such EV surface proteins), or homologs or analogs from non-human animals.

[0295] Specifically, the EV surface protein includes a tertiary structure having regions with loop, helix, and / or linear (peptide) structures, in particular, a tetraspan-like tertiary structure such as that described for CD81, including at least one large loop and one or more helix regions. Such a tertiary structure can be appropriately manipulated to incorporate an artificial binding site containing contact points in distant regions of the tertiary structure.

[0296] In certain cases, the surface protein is immobilized on the lipid bilayer membrane of the extracellular vesicle via a linker. Such a linker can be, for example, an amino acid linker, a hydrophilic and uncharged polymer-based linker such as polyethylene glycol (PEG) and polysaccharide, or can be composed of a zwitterionic polymer containing both cationic and anionic groups.

[0297] Specific surface proteins are those of mammalian origin, in particular, those that occur naturally in species including warm-blooded animals, especially dogs, cats, rabbits, horses, cows, pigs, and birds, humans or non-human mammals.

[0298] As used herein, the term "transmembrane domain" refers to a protein domain that spans the lipid membrane, which is typically hydrophobic. Specifically, the surface protein includes at least two transmembrane domains that anchor the extracellular loop to the EV. Tetraspanin proteins typically include four domains that span the membrane. The transmembrane domains are typically located within the membrane of the EV when attached or bound to the EV. However, the protein domain is also referred to as a transmembrane domain, while being isolated from the EV or from the EV surface protein.

[0299] As used herein, "tetraspanin," also referred to as "tetraspan" or "tetraspanin protein," is a protein superfamily of transmembrane 4 superfamily proteins that organize membrane microdomains called tetraspanin-enriched microdomains by forming clusters and interacting with a diverse array of transmembrane and cytoplasmic signaling proteins (also referred to as the "tetraspanin superfamily"). Tetraspanins are typically cell surface proteins characterized by the presence of four hydrophobic transmembrane domains. Naturally occurring tetraspanin proteins mediate signaling events that play roles in the regulation of cell development, activation, growth, and motility.

[0300] Tetraspanins, as understood herein, typically consist of an extracellular domain (also referred to as the extracellular domain ED), a transmembrane domain, and an intracellular domain. For example, the N and C termini of tetraspanins are typically located within the EV, while the transmembrane domain is located within the lipid bilayer membrane and the extracellular domain is placed on the outer surface of the EV. Specific examples of tetraspanins are glycosylated.

[0301] The extracellular (extravascular) domain, also known as the small extracellular domain ED, is the most variable region in tetraspanins and may be involved in binding to targets. EC1 (the first extracellular loop) is also known as the small extracellular loop (SEL). The EC2 of tetraspanins (the "large extracellular loop", LEL) has been investigated using CD81LEL as a model protein for all tetraspanins. The LEL domain is divided into a constant region with conserved A, B, and E helices that are suggested to mediate homodimerization through a hydrophobic surface, and a variable region with helices C and D adjacent to such sequences involved in protein-protein interactions. Specifically, EC2 contains one cysteine residue proximal to the transmembrane that is present four times in all tetraspanins, and at least two conserved cysteine residues that form disulfide bonds for EC2 folding (the CCG motif), the Pro-Xaa-Xaa-Cys (PXXC, SEQ ID NO: 183, X can be any amino acid) motif in the majority of tetraspanins.

[0302] Among tetraspanins, CD9, CD63, CD81, CD82, and CD151 have a wide tissue distribution, while others such as Tssc6, CD37, and CD53 in hematopoietic cells are restricted to specific tissues. Immunoelectron microscopy studies have shown that tetraspanins are abundant in various types of endocytic membranes and are widely used as exosome markers.

[0303] The tetraspanin protein CD81 is a major protein enriched in the exosome fraction of multivesicular bodies. Human CD81 contains or consists of the amino acid sequence specified as SEQ ID NO: 87 (the coding sequence specified as SEQ ID NO: 88).

[0304] The large extracellular loop of CD81, which is topologically located between transmembrane domains 3 and 4, is characterized by five helix elements that form a mushroom-like structure stabilized by two pairs of cysteines. This motif is conserved among protein members of the tetraspanin family, and oxidation of the cysteine bond is involved in high-affinity binding of, for example, the E2 envelope protein of hepatitis C virus (HCV), the natural ligand of CD81.

[0305] Tetraspan proteins can be expressed as soluble proteins or presented on the surface of tetraspanin-expressing cells or their respective EVs. The crystal structure of hCD81 LEL solved at 1.6 Å revealed a new type of protein fold, and subsequent sequence analysis of 160 tetraspanin family members showed that their folds and important structural features are conserved. Apart from cysteine cross-links, hCD81 LEL can be stabilized by invariant residues Gly157 and Pro176 that are positioned to accommodate cysteine linkages, and Tyr127 that is fully buried and contributes to the hydrogen bond network together with His151 and Cys190. Soluble hCD81 LEL can assemble into dimers that surround a two-fold axis, and the contacts between protomers are low-polarity regions between helices of each interacting partner and the C-terminal residues of the protomer on the opposite side of helix B. The N- and C-termini of the protomer fit into the central region of the opposite face of the assembled dimer, as well as in the cell surface dimer assembly, where transmembrane segments are also present. A second low-polarity region includes the solvent-exposed surfaces of helices C and D. According to solution studies, helix D is mostly unstructured and adopts a helical conformation only when bound to a specific antigen. Sequence alignments of tetraspanin family members actually show an increased variability in this region, including insertions and deletions. This surface area may be involved in species- or tetraspanin-specific recognition processes, which may hint at the possibility of heterodimeric tetraspanin species assembly. In particular, segment D of CD81 can lead to specific homomeric clustering.

[0306] According to specific embodiments, the biophysical properties of tetraspanins such as CD81 are improved by the introduction of de novo pairs of cysteine residues that form novel disulfide bridges that stabilize the protein. Specifically, the formation of the novel disulfide bridges increases the stability of tetraspanins such as CD81, which enables the production of mutants with higher stability, for example mutants incorporating one or more modified regions, or novel target-specific binding sites containing binding residues within said one or more modified regions. Specifically, the amino acid sequence can be modified by targeted or random mutagenesis to incorporate binding residues (in particular by their substitution) at a given position or region(s) that create a binding site, or a library can be created that contains diverse binding sites containing binding residues.

[0307] Similar to CD81, the tetraspanin CD9 is a cell surface protein containing four hydrophobic transmembrane domains and two extracellular domains ED (such as comprising or consisting of EC1 and EC2, etc.).

[0308] Naturally occurring CD9 consists of 228 amino acids and has a weight of 24 - 27 kDa. Each of the four small and highly conserved hydrophobic transmembrane domains contains 24 - 27 amino acids. It has a small N-terminal (11 amino acids) and C-terminal cytoplasmic (7 amino acids) tail, as well as a very small intracellular domain (4 amino acids). The remaining portion of the protein consists of two extracellular domains (a small one of 20 amino acids, EC1, and a large one of 83 amino acids, EC2). Two disulfide bonds made by four well-conserved cysteine residues (C) stabilize the large extracellular domain (EC2). CD9 also contains a tetraspanin signature (amino acids 65 - 89) and a CCG motif (amino acids 152 - 154). CD9 is one of the most ubiquitously expressed proteins on the surface of exosomes and is therefore regarded as an exosome marker. Although there are variations in the amino acid sequence in the extracellular loop, the CD9 protein sequence is very well conserved among species (90% among human, mouse, and rat). CD9 also shares some homology with other tetraspanins, particularly in the transmembrane domain.

[0309] Wild-type CD9 can interact with or form complexes with many other proteins, including other tetraspanins, integrins, EWI molecules, TGF-α, diphtheria toxin receptor, or tyrosine kinases, pregnancy-specific glycoproteins, as well as proteins of the immune system such as MHC class II molecules and members of the Ig superfamily. Furthermore, CD9 is involved in platelet activation and aggregation, as well as cell adhesion, spreading, cell motility, and tumor metastasis. CD9 also regulates paranodal junction formation and is required for gamete fusion. Additionally, CD9 promotes myocyte fusion and supports myotube maintenance.

[0310] As described herein, tetraspanin CD63 is a highly glycosylated cell surface protein containing four transmembrane domains and three putative N-glycosylation sites.

[0311] Wild-type CD63 is typically present in late endosomes, lysosomes, secretory vesicles, and the plasma membrane and moves between these compartments. CD63 is extensively and variably glycosylated, and its EC2 region contains three potential N-linked glycosylation sites (N130, N150, and N172). It is often used as a marker for multivesicular bodies enriched in CD63. It has numerous natural interaction partners, including CD82; HLA-DR, HLA-DM, and HLA-DO of MHC class II molecules; several integrins; and other tetraspanins such as phosphatidylinositol 4 kinase. CD63 also contains a tyrosine-based motif at its most distal C-terminus. Tyrosine-based motifs in the cytoplasmic domain of membrane proteins are recognized by clathrin adaptor complexes and play important roles in endocytosis, lysosomal targeting, and basolateral targeting. The tyrosine-based motif in CD63 mediates its interaction with the μ subunits of adaptor protein complexes 2 and 3 (AP-2 and AP-3).

[0312] As described herein, the tetraspanin CD151 has the characteristic structure of tetraspanins. It is a 253-amino acid protein with a single N-glycosylation site in its LEL and is palmitoylated at several cysteine residues. Immunoblots reveal apparent doublet bands of 28 and 32 kDa representing non-glycosylated and glycosylated forms of CD151. Human CD151 comprises or consists of the amino acid sequence specified as SEQ ID NO: 94.

[0313] Wild-type CD151 has a broad cellular and tissue distribution, including epithelial, endothelial, muscle, renal glomeruli and proximal and distal renal tubules, Schwann cells, and dendritic cells, and a single RNA species is observed in most human adult tissues. CD151 is highly expressed in platelets and megakaryocytes. Similar to other tetraspanins, CD151 associates with several integrins in the cell membrane.

[0314] Although all immune cells express tetraspanins, most of these are present in a variety of other cell types. There are several found in the hematopoietic system, such as CD37, CD53, TSPAN32 (Tssc6), and TSPAN33.

[0315] As described herein, the tetraspanin CD37 is a cell surface glycoprotein that has the typical structure of a tetraspanin known to complex with integrins and other transmembrane 4 superfamily proteins. Alternative splicing results in multiple transcript variants encoding various isoforms. Human CD37 comprises or consists of the amino acid sequence specified as SEQ ID NO: 95.

[0316] CD37 is known to be expressed by cells of the immune system, with the highest abundance in mature B cells and lower expression found in T cells and myeloid cells. Wild-type CD37 controls both humoral and cellular immune responses. CD37 deficiency in mice leads to the spontaneous development of B cell lymphomas, and patients with CD37-negative lymphomas have a poor clinical outcome.

[0317] As described herein, the tetraspanin CD53 is a pan-leukocyte surface glycoprotein that traverses the plasma membrane four times and is a member of the transmembrane 4 superfamily. The protein sequence and gene structure of mouse CD53 (Cd53) were determined by the isolation of both genomic and cDNA clones. CD53 is highly conserved in evolution, as mouse Cd53 was 91% identical to rat CD53 and 82% identical to human CD53. The tetraspanin CD53 has four transmembrane domains and is glycosylated twice in its second extracellular loop. It has a length of 219 amino acids and is located in the cell membrane, endosomes, and the lipid bilayer membrane of exosomes. Human CD53 comprises or consists of the amino acid sequence specified as SEQ ID NO: 90.

[0318] Wild-type CD53 is expressed by virtually all immune cells, a subset of hematopoietic stem cells, and in diverse hematologic malignancies. There are several tetraspanins present in platelets, including CD9, CD151, Tssc6, and CD63. Recent studies in knockout mouse models have revealed that CD151 and Tssc6 are physically and functionally involved in regulating the "outside-in" signaling properties of the majority of platelet integrins, integrin alpha(IIb)beta(3), and thrombus stability in vivo.

[0319] As used herein, the tetraspanin Tssc6, also referred to as TSPAN32, is a member of the tetraspanin superfamily. The protein has a size of 320 amino acids and is expressed ubiquitously at low levels. High levels of TSPAN32 expression are typically restricted to hematopoietic tissues, including peripheral blood leukocytes, thymus, and spleen.

[0320] Human TSPAN32 comprises or consists of the amino acid sequence specified as SEQ ID NO: 91. As used herein, lysosome-associated membrane protein 2 (LAMP2) is a membrane glycoprotein associated with lysosomes. LAMP2 is an essential membrane protein having two conserved luminal domains (comprising 90% of the whole protein), a single transmembrane (TM) domain (about 20 amino acids), and a short (10 - 12 amino acids) C-terminal cytoplasmic tail. Glycosylation is found in its luminal domains. Human LAMP2 comprises or consists of the amino acid sequence specified as SEQ ID NO: 96. LAMP2 as used herein preferably includes modifications in the extracellular loop region so as to incorporate an artificial binding site.

[0321] Wild-type LAMP2 plays an important role in chaperone-mediated autophagy, a process that mediates the lysosomal degradation of proteins in response to various stresses and as part of the normal metabolic turnover of proteins having a long biological half-life.

[0322] As used herein, the term "tetraspanin-like protein" (also sometimes referred to as tetraspaninoid) refers to an EV surface protein comprising at least two transmembrane domains and at least one ED located between said at least two transmembrane domains, preferably the region between said at least two transmembrane domains comprises or consists of one ED.

[0323] Specific examples of tetraspanin-like proteins include naturally occurring or modified tetraspanin proteins such as lysosome-associated membrane glycoproteins (LAMP) and others, or recombinant or synthetic proteins, such as chimeric proteins that contain one or more transmembrane domains of one protein and one or more EDs of another protein, thereby obtaining a recombinant tetraspanin-like protein.

[0324] "Sequence identity" or "percent amino acid sequence identity (%)" with respect to a protein sequence and its variants is defined as the percentage of amino acid residues in a candidate sequence that are identical to the amino acid residues in a specific polypeptide sequence being compared (the "parent sequence") after aligning the sequences and introducing gaps if necessary to achieve the maximum percentage of sequence identity, and without considering any conservative substitutions as part of the sequence identity. One of ordinary skill in the art can determine appropriate parameters for measuring the alignment, including any algorithm required to achieve the maximum alignment over the entire length of the sequences being compared.

[0325] As used herein, the terms "specificity", "target-specific", or "specific binding" refer to a binding reaction that determines the cognate ligand of interest in a heterogeneous population of molecules. Thus, under specified conditions (e.g., immunoassay conditions), the modified surface protein binds to its specific target and does not bind in significant amounts to other molecules present in the sample. Specific binding means that the binding is selective in terms of the selected target identity, high, intermediate, or low binding affinity or avidity. Selective binding is usually achieved when the binding constant or binding kinetics differ by at least 10-fold, preferably by at least 100-fold, more preferably by at least 1000-fold.

[0326] Specific binding does not exclude some degree of cross-reactivity with similar antigens, or the same antigen (analog) from different species. For example, the binding entity may preferably cross-react with rodent targets similar to human targets, facilitating preclinical animal studies.

[0327] As used herein, the term "target" shall specifically include all antigens and target molecules that can be recognized by the binding site of an antibody. The surface proteins described herein are engineered to include an artificial target binding site that specifically recognizes an antigenic structure or epitope, such as an antibody.

[0328] Specific targets are cell targets or soluble targets. Often, the target is a receptor located on the surface of tumor cells, or an autoantigen such as a cytokine or growth factor that may be present in the circulation of a subject or patient. Further targets can be of pathogen origin, such as those of microbial or viral pathogens.

[0329] The target antigen can be the entire target molecule or a fragment of such a molecule that is immunologically relevant, i.e., recognizable by natural or monoclonal antibodies, in particular lower structures, such as polypeptide or carbohydrate structures of the target, commonly referred to as "epitopes" (e.g., B-cell epitopes, T-cell epitopes).

[0330] Specifically, the target antigen is selected from the group consisting of cellular target antigens including receptors, particularly erbB receptor tyrosine kinases (such as EGFR, HER2 including Her2neu, HER3, and HER4, etc., particularly such epitopes of the extracellular domain of such receptors, such as the 4D5 epitope). In addition, further antigens can be targeted, for example, molecules of the TNF receptor superfamily, which are, for example, Apo-1 receptor, TNFR1, TNFR2, nerve growth factor receptor NGFR, CD40, CD40 ligand, OX40, TACI, BCMA, BAFF receptor, T cell surface molecules, T cell receptor, T cell antigens, Apo-3, DR4, DR5, DR6, decoy receptors, such as DcR1, DcR2, CAR1, HVEM, GITR, ZTNFR-5, NTR-1, TNFL1, IGFR-1, c-Met, but are not limited to these molecules, B cell surface antigens, such as CD10, CD19, CD20, CD21, CD22, DC-SIGN, antigens or markers of solid tumor or blood cancer cells, cells of lymphoma or leukemia, other blood cells including blood platelets, but are not limited to these molecules.

[0331] The terms "effective amount" or "sufficient amount" of a compound, such as a binding factor described herein, particularly TEV described herein, which are used interchangeably herein with the term "therapeutically effective amount", are an amount or activity sufficient to produce a beneficial or desired result, including clinical results, when administered to a subject, and thus the effective amount or its synonyms depend on the context in which it is applied.

[0332] The effective amount is intended to mean the amount of the compound sufficient to treat, prevent, or inhibit a disease or disorder. In the context of a disease, the therapeutically effective amount of a binding factor or TEV described herein is specifically used to treat, modulate, alleviate, ameliorate, or affect a disease or condition that benefits from the interaction of the binding factor or TEV with its target antigen, such as a tumor cell.

[0333] The amount of the compound that would correspond to such an effective amount will vary depending on various factors such as the given drug or compound, pharmaceutical formulation, route of administration, type of disease or disorder, the nature of the subject or host being treated, etc., but can nevertheless be routinely determined by one of ordinary skill in the art.

[0334] The binding factors described herein can be specifically used in pharmaceutical compositions. Accordingly, pharmaceutical compositions are provided that include the binding factors described herein and a pharmaceutically acceptable carrier or excipient. These pharmaceutical compositions can be suitably administered as a bolus injection or infusion, or by continuous infusion. Except for parenteral administration, topical or oral administration can be preferred. Pharmaceutically acceptable carriers suitable to facilitate such means of administration are well known in the art.

[0335] Pharmaceutically acceptable carriers generally include any suitable solvent, adjuvant, dispersion medium, coating, isotonic agent, absorption delaying agent, and the like that are physiologically compatible with the binding factors provided herein. Further examples of pharmaceutically acceptable carriers include sterile water, physiological saline, phosphate buffered saline, dextrose, glycerol, ethanol, etc., as well as any combination thereof.

[0336] Suitable pharmaceutically acceptable carriers or excipients specifically include one or more of any conventional solvent, dispersion medium, filler, solid carrier, aqueous solution, coating, vehicle suitable for topical administration, other antimicrobial agents, isotonic and absorption enhancing or delaying agents, or activity enhancing or delaying agents for pharmaceutically active substances. Commonly seen pharmaceutically acceptable additives are disclosed, for example, in Remington: the Science & Practice of Pharmacy, 20th Edition, by Alfonso Gennaro, Lippencott Williams & Wilkins, (2000).

[0337] In one embodiment, suitable pharmaceutically acceptable carriers include, but are not limited to, inert solid fillers or diluents, and sterile aqueous or organic solutions (e.g., polyethylene glycol, propylene glycol, polyvinylpyrrolidone, ethanol, benzyl alcohol, etc.). In certain such embodiments, suitable pharmaceutically acceptable excipients include, but are not limited to, water, saline solutions, alcohol, polyethylene glycol, gelatin, lactose, amylase, magnesium stearate, talc, silicic acid, viscous paraffin, hydroxymethylcellulose, polyvinylpyrrolidone, fillers such as saccharides (e.g., lactose, sucrose, mannitol, or sorbitol), and cellulose preparations (e.g., corn starch, wheat starch, rice starch, potato starch, gelatin, tragacanth gum, methylcellulose, hydroxypropylmethylcellulose, sodium carboxymethylcellulose, and / or polyvinylpyrrolidone PVP).

[0338] In one such aspect, the binding agent can be combined with one or more carriers suitable for the desired route of administration, for example, lactose, sucrose, starch, cellulose esters of alkanoic acids, stearic acid, talc, magnesium stearate, magnesium oxide, sodium and calcium salts of phosphoric and sulfuric acids, acacia, gelatin, sodium alginate, polyvinylpyrrolidone, polyvinyl alcohol, and can be admixed with any of them, and optionally, can be further tableted or encapsulated for conventional administration. Other carriers, adjuvants, and modes of administration are well known in the pharmaceutical art. The carrier can include, alone or with waxes or other materials well known in the art, controlled release or time delay materials such as glyceryl monostearate or glyceryl distearate.

[0339] Additional pharmaceutically acceptable carriers are known in the art and are described, for example, in REMINGTON’S PHARMACEUTICAL SCIENCES. Liquid formulations can be solutions, emulsions, or suspensions and can contain excipients such as suspending agents, solubilizing agents, surfactants, preservatives, and chelating agents.

[0340] Pharmaceutical compositions are contemplated in which the binding factors and one or more therapeutic active agents described herein are formulated. Stable formulations of the binding factors described herein are prepared for storage in the form of lyophilized formulations or aqueous solutions by mixing the construct having the desired degree of purity with a pharmaceutically acceptable carrier, excipient, or stabilizer as required. Formulations for in vivo administration are preferably sterile and are, for example, in the form of a sterile aqueous solution. This is readily accomplished by filtration through a sterile filtration membrane or other suitable sterilization methods.

[0341] Administration of the pharmaceutical compositions containing the binding factors described herein can be effected in a variety of ways, including orally, subcutaneously, intravenously, intranasally, intraotically, transdermally, mucosally, topically, e.g., in the form of tablets, gels, ointments, plasters, suppositories, patches, lotions, creams, etc., intraperitoneally, intramuscularly, intrapulmonary, vaginally, parenterally, rectally, or intravitreally.

[0342] In one embodiment, the pharmaceutical composition is administered orally, intravenously, or by inhalation. In a specific embodiment, the binding factor is administered in a dosage form selected from the group consisting of solid dosage forms, creams, aqueous mixtures, lyophilized aqueous mixtures, and aerosols.

[0343] Exemplary formulations for parenteral administration include those suitable for subcutaneous, intramuscular, or intravenous injection, such as, for example, sterile solutions, emulsions, or suspensions. The binding factors described herein can be specifically used in diagnostic compositions, for example, for in vitro or in vivo use. Therefore, diagnostic compositions are provided that include the binding factors described herein and, optionally, diagnostic reagents, in the form of a composition or a kit of parts.

[0344] The diagnostic kit preferably includes all essential components for qualitatively or quantitatively determining a target in a biological sample, without commonly used or non-specific substances or components such as water, buffer, or excipient, as required. A storage-stable kit can be provided that preferably has a shelf life of at least six months, more preferably at least one or two years. It can be composed of dry (e.g., lyophilized) components and / or can contain preservatives.

[0345] Preferred diagnostic kits are provided as packaged or prepackaged units that facilitate routine experiments, for example, where the components are contained in a single package. Such a package can contain the reagents required for one or more tests, for example, suitable for performing a series of biological sample tests. The kit can further appropriately contain a standard substance or a reference control.

[0346] The diagnostic composition can be a reagent that can be used immediately in a reaction mixture with a biological sample, or a storage form of such a reagent, for example, lyophilization; rapid freezing (e.g., in liquid nitrogen), cryopreservation (-70 °C and -80 °C), refrigerated storage (-20 °C and 5 °C), and controlled room temperature (15 °C to 27 °C) and other storage-stable forms; standard sample storage such as glycerol stocks, tissue paraffin blocks, (oral) swabs, and other standard biological sample storage methods. The storage form of the reagent can be reconstituted or prepared to obtain a reagent that can be used immediately. Such a reagent that can be used immediately is typically in the form of an aqueous solution, specifically in a (physiological) buffer state (e.g., EDTA-buffered, phosphate buffer, HBSS, citrate buffer, etc.).

[0347] Specifically, further diagnostic reagents are reagents that specifically react with a binding factor and / or the reaction product of the binding of the binding factor to its target. Suitable diagnostic reagents can be solvents, buffers, dyes, anticoagulants, and ligands that specifically bind to the binding factors and / or binding factor-target complexes described herein.

[0348] Specifically, the present invention provides a diagnostic preparation of a binding factor described herein, which optionally contains a binding factor or a reagent that specifically recognizes a complex of a binding factor and its respective target, such as a binding factor having a label and / or a further diagnostic reagent having a label, and / or a solid phase for immobilizing at least one of the binding factor and the diagnostic reagent.

[0349] Specifically, further diagnostic reagents are diagnostic labels or reagents that specifically react with a binding factor and / or the reaction product of the binding of the binding factor to its target. EV or a diagnostic reagent can be directly labeled or indirectly labeled. Indirect labeling can include a labeled binder that forms a complex with a binding factor or a diagnostic reagent for a target.

[0350] Labels are typically molecules or parts of molecules that can be detected in an assay. Exemplary labels are chromophores, fluorescent dyes, or radioactive molecules. In some embodiments, EV or a diagnostic reagent is conjugated to a detectable label, which is a molecule that is itself detectable (e.g., a fluorescent moiety, an electrochemical label, a metal chelate, etc.), as well as a molecule that can be indirectly detected by the production of a detectable reaction product (e.g., an enzyme such as horseradish peroxidase, alkaline phosphatase, etc.) or a specific binding molecule that is itself detectable (e.g., biotin, digoxigenin, maltose, oligohistidine, 2,4-dinitrobenzene, phenylarsine oxide, ssDNA, dsDNA, etc.).

[0351] Preferred diagnostic preparations or assays include the EVs described herein immobilized on a solid phase, such as latex beads, gold particles, etc. The following items are regarded as specific embodiments of the present invention.

[0352] 1. A target-specific extracellular vesicle (EV) comprising a lipid bilayer membrane immobilizing a surface protein comprising an artificial binding site that specifically recognizes a target, wherein the binding site comprises binding residues within at least two discrete regions of the surface protein, the EV.

[0353] 2. The EV according to item 1, wherein the surface protein comprises a transmembrane domain of a vesicle membrane protein, and the transmembrane domain anchors the surface protein to the EV. 3. The EV according to item 2, wherein the membrane protein is a tetraspanin protein, preferably selected from the group consisting of CD81, CD9, CD37, CD53, CD63, and CD82, or LAMP2.

[0354] 4. The EV according to any one of items 1 to 3, wherein the surface protein is CD81, and the binding residues are located within a first region between positions 160 and 172 and within at least one additional region located between positions 132 and 139 or between positions 180 and 189, and the numbering is that of CD81 specified as SEQ ID NO: 87.

[0355] 5. The EV according to item 4, wherein the surface protein is CD81, and it is stabilized by introducing one or more cysteine(s) at position(s) that allow the formation of one or more additional disulfide bonds that stabilize the extracellular loop structure of the protein.

[0356] 6. The EV according to any one of items 1 to 5, which is of eukaryotic or prokaryotic source cells of body tissues, body fluids, or cell cultures. 7. It transports intracellular contents, and the contents include any one or more of peptides, polypeptides, protein domains, proteins, lipids, genes, mRNAs, miRNAs, nucleic acids such as RNAi mediator molecules, especially locked nucleic acids, or plasmids such as phosphorothioates, DNA, DNA fragments, minicircle DNAs, drugs such as small molecules, especially chemotherapeutic drugs or senolytic drugs, and the EV according to any one of items 1 to 6.

[0357] 8. The target is selected from the group consisting of cell targets such as mitogen receptors, cytokine receptors, asialoglycoprotein receptors, membrane transporters, lipoproteins, lipopolysaccharides, glycoproteins, proteoglycans, or cell-free targets such as cytokines, artificial proteins, or artificial surface structures, and the EV according to any one of items 1 to 7.

[0358] 9. A pharmaceutical preparation comprising the EV according to any one of items 1 to 8 and a pharmaceutically acceptable carrier, preferably in a formulation for intradermal, subcutaneous, intravenous, topical, or oral use.

[0359] 10. a) Introducing a gene encoding a surface protein into a source cell or a source cell mixture; b) Culturing the cell(s) under conditions for producing extracellular vesicles; c) Isolating a fraction containing EVs having target-binding specificity; and d) Producing an EV preparation A method for producing a preparation of the EV according to any one of items 1 to 8, comprising the above steps.

[0360] 11. Modifying the gene by a mutagenesis method to incorporate an artificial binding site that specifically recognizes a target, and the mutagenesis method employs mutagenesis of at least two distant regions of the surface protein, thereby producing a repertoire of polynucleotides encoding diverse surface proteins each having different target specificities, and selecting a polynucleotide encoding a surface protein that specifically recognizes a target, according to the method of item 10.

[0361] 12. The repertoire of polynucleotides presents various surface proteins on the outer surface and preferably is included in a gene package that employs a display system selected from the group consisting of phage, yeast, bacteria, ribosome, mRNA, or mammalian cell display, the method according to item 11.

[0362] 13. An EV preparation produced according to any one of items 10 to 12 for self - use, wherein the source cell or source cell mixture is obtained from a subject and the EV preparation is administered to the same subject, the EV preparation.

[0363] 14. a) Introduce a repertoire of genes encoding various surface proteins into the said source cell(s); and b) Isolate a repertoire of EVs each having a different target - binding specificity to produce a library of EVs comprising a repertoire of target - binding EVs having various target - binding specificities, wherein the repertoire is produced by introducing mutations into the gene within at least two predetermined separated regions of the gene, The method according to any one of items 10 to 12 for producing a library of EVs.

[0364] 15. Preferably, the repertoire is a library of EVs produced by the method according to item 14, comprising at least 10 2 EVs having different target specificities. The examples described herein are illustrative of the invention and are not intended to be limiting thereof. Various embodiments of the invention are described in accordance with the invention. Many modifications and variations can be made to the techniques described and illustrated herein without departing from the spirit and scope of the invention. Therefore, it should be understood that the examples are merely illustrative and do not limit the scope of the invention.

Examples

[0365] Example 1: Yeast display of CD81 LEL The wild-type human CD81 LEL sequence was cloned as a C-terminal fusion protein with Aga2, with an Xpress tag added first and his tag and V5 tag added to the C-terminus.

[0366] The amino acid sequence of CD81 LEL was

[0367]

Chemical formula

[0368] as follows. The coding nucleotide sequence was

[0369]

Chemical formula

[0370] as follows. The primers for amplification of human CD81 LEL were

[0371]

Chemical formula

[0372] as follows. The PCR product was digested with BamHI and NotI and ligated with the corresponding digested vector pYD1 (Thermo Fisher Scientific). The ligation mixture was transformed into electrocompetent E. coli TOP10, and the transformants were selected on ampicillin plates. The plasmid was isolated by minipreparation and transformed into S. cerevisiae EBY100 using chemical transformation. A starter culture of EBY100 (Thermo Fisher Scientific) in 20 ml of YPD medium (2% peptone, 1% yeast extract, 2% glucose) (Merck) was incubated overnight at 30 °C and 180 rpm. The culture was then diluted to an OD 600 of 0.4 and incubated at 30 °C and 180 rpm for approximately 5 hours. Next, an aliquot of 50 ml of the cell culture was pelleted at 1000 g for 5 minutes at room temperature, then washed with 25 ml of AD and pelleted again. The cells were resuspended in 3 ml of 100 mM Li-acetate and incubated at 30 °C for 15 minutes in a shaking incubator. Then, 0.3 ml of the cell suspension was pelleted and the supernatant removed. The components of the transformation mix were added as follows: 240 μl of 50% PEG 3350, 36 μl of 1.0 M Li-acetate, 50 μl of 2 mg / ml ssDNA (salmon sperm carrier DNA, pre-heated to a maximum of 95 °C for 5 minutes and then placed on ice) (Sigma Aldrich), and 1 μg of the pYD1-CD81 LEL plasmid. The cell pellet was resuspended in the transformation mix and incubated at 30 °C for 30 minutes in a shaking incubator, then given a heat shock at 42 °C for 45 minutes. The yeast cells were pelleted at 1000 g for 5 minutes at room temperature, resuspended in AD, and the transformants were selected on MDL medium at 30 °C for 3 days.

[0373] The transformants were seeded into SD-CAA (1% casamino acids (Becton Dickinson), 100 mM K-phosphate buffer, pH 6.0 (Merck), 1×YNB (Becton Dickinson), 2% glucose (Merck)) and cultured overnight at 30°C. Induction was carried out at 37°C overnight or at 20°C for 2 days with SG / R-CAA medium (1% casamino acids (Becton Dickinson), 100 mM K-phosphate buffer, pH 6.0 (Merck), 1×YNB (Becton Dickinson), 2% galactose (Merck), 1% raffinose (Merck)). Subsequently, yeast cultures were examined for the expression of the recombinant protein. FACS analysis revealed presentation levels similar to those of yeast transformed with an unmodified expression vector encoding only the reporter tag. Furthermore, when yeast cultures presenting wild-type CD81 LEL were stained with the structural report antibodies M38 and 1.3.3.22 (Thermo Fisher Scientific), the results showed the same expression levels for the tags examined. FACS analysis revealed similar presentation levels for cultures induced at 20°C or under stress conditions at 37°C. Tag expression was at levels similar to those found for yeast transformed with a vector encoding only that tag. Furthermore, the yeast cells expressing wild-type CD81 were stained with anti-CD81 antibody in the same manner as the cells induced at 20°C.

[0374] Example 2: Phage display of CD81 LEL The sequence encoding wild-type CD81 LEL was cloned into the multiple cloning site of the expression vector fdmyc, which enables the expression of phage particles by a recombinant protein located at the N-terminus from the c-myc tag and the g3p protein.

[0375] The primers for amplification were

[0376]

Chemical formula

[0377] It was. The PCR product was digested with the restriction enzymes ApaLI and NotI and ligated into the corresponding digested vector fdmyc. The ligation mixture was transformed into E. coli TG1 cells (Thermo Fisher Scientific) and selected on TYE plates containing tetracycline (1.5% agar, 1.6% peptone, 1% yeast extract (Merck)). After overnight culture in the tetracycline-containing medium at 30°C, 10 11 ~10 12 A high titer of phage particles could be obtained, indicating that the expressed protein is not harmful to phage growth. The expression level of the fusion protein was tested using SDS-PAGE and analysis of phage particles by Western blotting. Detection of the presented protein was performed with an anti-g3p antibody (New England Biolabs), and it was found to be 50% fused with wild-type CD81 LEL. CD81 LEL of the phage could be detected with both CD81-specific antibodies M38 and 1.3.3.22, indicating correct folding of the phage display molecule.

[0378] Example 3: Expression of soluble CD81 LEL The CD81 LEL sequence was amplified by primers.

[0379] [Chemical formula]

[0380] The PCR fragment was digested with the restriction enzymes NheI and BamHI and ligated into the vector pTT22SSP4 (CNRC) digested with the same enzymes. The ligation mixture was transformed into electrocompetent Escherichia coli TOP10 (Thermo Fisher Scientific), and the transformants were selected on ampicillin plates. The plasmid was isolated by minipreparation and transfected into HEK-293-6E cells (CNRC) exactly according to the manufacturer's instructions. TN-20 was added to a final concentration of 0.5% and protein expression was continued for 5 days. Then, hCD81 LEL was purified by Ni-NTA chromatography using standard protocols. The supernatant was buffered with PBS and 20 mM imidazole and a pH adjusted to 7.5. An Excel Ni-NTA column (GE Healthcare) was equilibrated with PBS and 20 mM imidazole at pH 7.5, and the buffered supernatant was loaded. Elution was performed at pH 7.5 with a linear gradient of imidazole from 20 mM to 500 mM in PBS over 5 column volumes. The protein-containing fractions were pooled and dialyzed overnight at 4 °C against 100 volumes of PBS.

[0381] SEC analysis under native conditions revealed a monodisperse elution profile corresponding to a dimeric form of the protein, similar to soluble wild-type CD81 LEL. Alternatively, the protein was expressed in the ExpiCHO expression system (Thermo Fisher Scientific) using the MaxTiter protocol according to the manufacturer's instructions. Subsequently, hCD81 LEL was purified by Ni-NTA chromatography using a standard protocol. The supernatant was diluted with an equal volume of AD and buffered with PBS and 20 mM imidazole and a pH adjusted to 7.5. An Excel Ni-NTA column (GE Healthcare) was equilibrated with PBS and 20 mM imidazole at pH 7.5, and the buffered supernatant was loaded. Elution was performed at pH 7.5 with a linear gradient of 20 mM to 500 mM imidazole in PBS over 5 column volumes. The protein-containing fractions were pooled and dialyzed against 100 volumes of PBS at 4 °C overnight.

[0382] Example 4: Construction of CD81 library 1 A yeast display library of CD81 LEL mutants randomly mutated at a total of 12 amino acid residues in the C and D segments of CD81 LEL was constructed. The amino acid residues randomized were: 160 - 162 and 181 - 189 (numbered as in 1G8Q). The yeast display library was prepared at a size of 7×10 7 of non-dependent members. The PCR fragment for recombination was amplified using Q5 HiFi polymerase (New England Biolans) with oligonucleotides

[0383] [wherein N is any one of A, C, G, or T]

[0384] [wherein N is any one of A, C, G, or T]

[0385] [wherein N is any one of A, C, G, or T]

[0386] [wherein, N is any one of A, C, G, or T, and M is any one of A or C] It was prepared using and purified after gel electrophoresis. To facilitate recombination of the PCR fragment, the recipient vector was modified. All mutagenesis steps were performed using the QuikChange Lightning Mutagenesis kit (Agilent) exactly according to the manufacturer's instructions.

[0387] The BamHI site was introduced into pYD1_CD81 LEL by

[0388]

Chem.

[0389] and then the naturally occurring BamHI site was removed using the oligonucleotide

[0390]

Chem.

[0391] . The vector was linearized with BamHI and ClaI, and the vector backbone was purified from the agarose gel. Chemical transformation was used to introduce the PCR fragment and the vector backbone into budding yeast EBY100. A starter culture of EBY100 in 20 ml of YPD medium was incubated overnight at 30 °C and 180 rpm. Then, the culture was adjusted to OD 600Diluted to =0.4 and incubated at 30 °C and 180 rpm for about 5 h. Subsequently, an aliquot of 50 ml of the cell culture was pelleted at 1000 g for 5 min at room temperature, then washed with 25 ml of AD and pelleted again. The cells were resuspended in 3 ml of 100 mM Li-acetate and incubated at 30 °C for 15 min in an orbital incubator. The cells were pelleted and the supernatant removed. The components of the transformation mix were added as follows; 2400 μl of 50% PEG 3350, 360 μl of 1.0 M Li-acetate, 500 μl of 2 mg / ml ssDNA (salmon sperm carrier DNA, pre-heated to a maximum of 95 °C for 5 min and then placed on ice), 10 μg of linearized recipient vector, and 7 μg of DNA fragment. The cell pellet was resuspended in the transformation mix and incubated at 30 °C for 30 min in an orbital incubator and given a heat shock at 42 °C for 45 min.

[0392] The cells were collected by centrifugation at 1000 g for 5 min at room temperature and the supernatant removed, after which the pellet was resuspended in 10 ml of SD-CAA medium (1% casamino acids (Becton Dickinson), 100 mM K phosphate buffer, pH 6.0 (Merck), 1× YNB (Becton Dickinson), 2% glucose (Merck)). To determine library size, an aliquot for dilution plating was removed, 10 μl of the cell suspension was diluted in 990 μl of SD-CAA medium and 100 μl of this was plated onto MDL plates (1.5% agar (Merck), 1× YNB (Becton Dickinson), 2% glucose (Merck), and 0.01% leucine (Sigma-Aldrich)) and incubated at 30 °C for 3 days. The yeast cells were diluted in 50 ml of SD-CAA medium and incubated at 30 °C at 180 rpm for 24 h, subcultured into fresh SD-CAA medium at a 1:20 dilution and cultured for a further 24 h under the same conditions. The cells were harvested by centrifugation at 1000 g for 5 min at 4 °C and the pellet resuspended in an equal volume of 30% glycerol and then frozen at -80 °C.

[0393] Example 5: Selection of CD81 LEL Library 1 with Mouse EGFR-Fc Mouse EGFR-Fc was purchased from Sino Biological. For biotinylation, the EZ-Link™ Sulfo-NHS-LC-LC-Biotin reagent was used at a molar ratio of 3:1. The antigen was reconstituted to a concentration of 0.25 μg / μl according exactly to the manufacturer's instructions. Incubation with the biotinylation reagent was continued for 1 hour at room temperature with shaking. Unbound biotin was removed by dialysis against 100-fold volume of PBS at 4 °C overnight with stirring using a Snakeskin dialysis tube (Thermo Fisher Scientific) with a MWCO of 10,000 Da.

[0394] For selection, the library was seeded into SD-CAA and cultured overnight at 30 °C. Induction was carried out with SG / R-CAA medium overnight at 37 °C. The induced cell suspension was adjusted to 10 8Dilute to the cells and centrifuge at 1000 g for 5 minutes at 20°C. To block the cells, resuspend the pellet in 1 ml of 10% BSA-PBS and incubate on a rotator at 20°C for 30 minutes. Centrifuge the cells at 3000 rpm for 5 minutes at 20°C and resuspend in 250 μl of 10% BSA-PBS containing 1 μM biotinylated EGFR-Fc. Incubate on a rotator at room temperature for 30 minutes, then centrifuge the cells at 3000 rpm for 5 minutes at 4°C. To wash the cells, resuspend the pellet in 1 ml of ice-cold PBS and centrifuge at 1000 g for 5 minutes at 4°C. Then resuspend the cells in 250 μl of 10% BSA-PBS with streptavidin-Alexafluor 647 (1:800) and anti-V5-FITC antibody (1:100) (Thermo Fisher Scientific) and incubate on ice for 30 minutes. Centrifuge the cells at 3000 rpm for 5 minutes at 4°C, then resuspend them in 1 ml of ice-cold PBS and centrifuge again. Finally, resuspend the cells in 250 μl of ice-cold PBS and maintain on ice until sorted by FACS Aria™. The first sort covered 2.5-fold library size and collected 1% false-positive yeast cells. In the second sort, 20-fold output of the first sort was processed and 0.1% false-positive yeast cells grew. In the following two sorting rounds, at least 100-fold output of the previous sort was processed and again 0.1% yeast cells were collected. After the fourth sort, single yeast display clones were plated out to characterize. Twenty-three selected clones were screened using staining with mouse EGFR-Fc and all except clones 9, 10, and 23 were significantly stained by the antigen but not by the secondary reagent streptavidin-Alexafluor 647.

[0395]

Table 1

[0396] Six different arrays were identified in array determination (Table 2).

[0397]

Table 2

[0398] Example 6: Selection of CD81 Library 1 with Human EGFR-Fc Human EGFR-Fc was purchased from Sino Biological. For biotinylation, the EZ-Link™ Sulfo-NHS-LC-LC-Biotin reagent was used at a molar ratio of 3:1. The antigen was reconstituted to a concentration of 0.25 μg / μl exactly according to the manufacturer's instructions. Incubation with the biotinylation reagent was continued for 1 hour at room temperature with shaking. Unbound biotin was removed by dialysis against 100-fold volume of PBS at 4°C overnight with stirring using a Snakeskin dialysis tube (Thermo Fisher Scientific) with a MWCO of 10,000 Da.

[0399] For selection, first, the library was cultured in SD-CAA medium supplemented with penicillin-streptomycin at 30°C overnight with shaking, and then the expression of the recombinant protein was induced by resuspending the yeast cells in SG / R-CAA medium with penicillin-streptomycin, and it was incubated at 37°C overnight with shaking.

[0400] For MACS, 10 9The induced yeast cells were pelleted at 2500 g for 5 minutes, and the pellet was washed by resuspending it in 50 ml of washing buffer (PBS, pH 7.2, 0.25% BSA, 2 mM EDTA), and then pelleted again at 2500 g for 5 minutes. The washed cells were resuspended in 25 ml of washing buffer containing 0.25 μM biotinylated antigen, incubated at room temperature for 30 minutes with gentle agitation, and further incubated in an ice bath for 10 minutes. Subsequently, the cells and antigen solution were pelleted at 4 °C at 2500 g for 5 minutes, washed twice with 50 ml of washing buffer, and pelleted after each washing step. The cells were resuspended in 5 ml of washing buffer + 25 μl of streptavidin microbeads (μMACS streptavidin kit, MACS Miltenyi), and incubated on ice for 10 minutes. Finally, 15 ml of washing buffer was added.

[0401] The LS column was positioned in the separator and 3 ml of washing buffer was applied to precondition the column. Then the cell solution was loaded in 7 ml batches. The dripping of the column was stopped once, and the column was briefly removed from the magnet and repositioned on the magnet to direct the captured cells through the column in order to change the orientation of the beads within the column. 1 ml of washing buffer was applied to the column before loading the next 7 ml cell suspension. These steps were repeated until all the cells were loaded. Then the column was washed with 3 ml of washing buffer, briefly removed from the magnet, and washed again with 3 ml of washing buffer. To elute the bound cells, the column was removed from the magnet and 5 ml of washing buffer was added. The cells were pushed out of the column into a new tube using the supplied plunger. The fraction of bound cells was pelleted at 2500 g for 5 minutes. The pellet was resuspended in 10 ml of SD-CAA medium and incubated overnight at 30 °C at 180 rpm. 10 μl of the eluted cells was diluted in 990 μl of SD-CAA, 100 μl was seeded onto an MDL plate, and incubated at 30 °C for 3 days to evaluate the output of the MACS procedure.

[0402] In subsequent FACS selection rounds, induced cell suspensions were diluted to 10 per ml in 10% BSA-PBS. 8 The cells were diluted to 1 ml of PBS and centrifuged at 3000 rpm for 5 min at 20° C. To block the cells, the pellet was resuspended in 1 ml of 10% BSA-PBS and incubated on a rotating platform at 20° C. for 30 min. The cells were centrifuged at 3000 rpm for 5 min at 20° C. and resuspended in 250 μl of 10% BSA-PBS containing 1 μM biotinylated EGFR-Fc. After incubation on a rotating platform at 20° C. for 30 min, the cells were centrifuged at 3000 rpm for 5 min at 4° C. To wash the cells, the pellet was resuspended in 1 ml of ice-cold PBS and centrifuged at 3000 rpm for 5 min at 4° C. The cells were then resuspended in 250 μl of 10% BSA-PBS with streptavidin-Alexafluor 647 (1:800) and anti-V5-FITC antibody (1:100) (Thermo Fisher Scientific) and incubated on ice for 30 minutes. The cells were centrifuged at 3000 rpm for 5 minutes at 4°C, after which they were resuspended in 1 ml of ice-cold PBS and centrifuged again. Finally, the cells were resuspended in 250 μl of ice-cold PBS and kept on ice until sorting by FACS Aria™. At least 20 times the output of the previous round of sorting was processed, and 0.1% of the top anti-V5 antibody positive yeast cells were collected. The fifth round of sorting was plated out to characterize single yeast display clones.

[0403] The following binder sequences were identified (Table 3).

[0404] [Table 3]

[0405] Example 7: Construction of CD81 LEL library 2 To expand the potential repertoire in antigen-specific CD81 LEL binders, a library was designed in which amino acid residues different from those randomized in CD81 LEL library 1 could form potential antigen recognition sites. This design involved the randomization of 11 amino acid residues: namely, amino acid residues 132 - 133 in helix A, amino acid residues 136 - 139 in the AB loop, and amino acid residues 162 - 165, 167, and 171 - 172 in helix C. Here, the scaffold is a CD81 LEL mutant having novel disulfide bonds: one connecting helix A and B (C4) and one connecting helix A and C (C9). Subsequently, this CD81 LEL mutant with the combination of the novel disulfide bonds Ala134Cys / Lys144Cys and Val135Cys / Ser168Cys was produced in HEK293 - 6E cells (CNRC) and its thermal stability was tested. Thermal unfolding was recorded up to a maximum of 130 °C, which continued in a single event with a Tm of 109.40 ± 0.25 °C, higher than 43 °C for wild-type hCD81LEL. This protein, called C4C9, migrated as a single sharp peak in the time profile of the wild-type protein in SEC under native conditions. Importantly, the stabilized mutant was able to bind to the structure-reporter antibody M38 (Thermo Fisher Scientific) to the same extent as wild-type hCD81 LEL. Examination of the Far-UV CD spectrum for this mutant found it to be identical to that obtained for wild-type hCD81 LEL, which was typical of a highly α-helix-containing protein with two characteristic minima at 208 nm and 222 nm, similar to previously published results.

[0406] First, in a series of mutagenesis steps, the recipient vector for yeast display pyd1_CD81LEL was modified by deletion of the native BamHI and HindIII restriction sites. All mutagenesis steps were performed using the Quickchange Lightning Mutagenesis Kit (Agilent) with oligonucleotides, deleting the BamHI site by using oligonucleotide

[0407]

Chem.

[0408] and deleting the HindIII site by using oligonucleotide

[0409]

Chem.

[0410] respectively. To facilitate recombination of library fragments after linearization, a new BamHI restriction site was introduced by using oligonucleotide

[0411]

Chem.

[0412] and a new HindIII restriction site was introduced by using oligonucleotide

[0413]

Chem.

[0414] respectively. The recipient vector DNA was isolated by midipreparation using Nucleobond (Macherey-Nagel), and the vector was linearized using restriction enzyme treatment with BamHI and HindIII. The vector backbone was isolated using purification after preparative gel electrophoresis.

[0415] Furthermore, to facilitate the correct recombination of the library encoding the fragments, the template insertion of CD81 was modified by the mutations Ala134Cys and Lys144Cys that form a new disulfide bond and Val135Cys and Ser168Cys that form another new disulfide bond, introducing a sequence with a restriction site. The template for the modification was the pTT22SSP4_CD81 LEL C4C9 mutant. Oligonucleotide

[0416]

Chemical formula

[0417] was used to introduce a HindIII site, and oligonucleotide

[0418]

Chemical formula

[0419] was used to introduce a BamHI site. PCR fragments with a randomized nucleotide sequence were amplified using oligonucleotides

[0420]

Chemical formula

[0421] [wherein N is any one of A, C, G, or T], and

[0422] [Chemical formula]

[0423] [In the formula, N is any one of A, C, G, or T; and M is any one of A or C] was prepared by (According to the IUPAC nucleotide code) In the PCR reaction, Q5 HiFi polymerase (New England Biolabs) was used to purify the library fragments after gel electrophoresis.

[0424] For transformation, a starter culture of EBY100 (Thermo Fisher Scientific) in 20 ml of YPD medium (2% peptone, 1% yeast extract, 2% glucose) (Merck) was incubated overnight at 30 °C and 180 rpm. Then, the culture was diluted to an OD 600 of 0.4 and incubated at 30 °C and 180 rpm for about 5 hours. Then, an aliquot of 50 ml of the cell culture was pelleted at room temperature at 1000 g for 5 minutes, then washed with 25 ml of AD and pelleted again. The cells were resuspended in 3 ml of 100 mM Li-acetate and incubated at 30 °C for 15 minutes in an orbital incubator. The cells were pelleted and the supernatant was removed. The components of the transformation mix were added as follows: 2400 μl of 50% PEG 3350, 360 μl of 1.0 M Li-acetate, 500 μl of 2 mg / ml ssDNA (salmon sperm carrier DNA, pre-heated to a maximum of 95 °C for 5 minutes and then placed on ice) (Sigma-Aldrich), 10 μg of linearized recipient vector, and 7 μg of DNA fragment. The cell pellet was resuspended in the transformation mix and incubated at 30 °C for 30 minutes in an orbital incubator and given a heat shock at 42 °C for 45 minutes.

[0425] Cells were collected by centrifugation at 1000 g for 5 minutes at room temperature, the supernatant was removed, and the pellet was resuspended in 10 ml of SD-CAA medium. An aliquot was removed for dilution plating to determine the library size, 10 μl of the cell suspension was diluted in 990 μl of SD-CAA medium, 100 μl of this was seeded onto MDL plates and incubated at 30 °C for 3 days. Yeast cells were diluted in 50 ml of SD-CAA medium, incubated at 30 °C at 180 rpm for 24 hours, subcultured into fresh SD-CAA medium at a 1:20 dilution, and cultured for a further 24 hours under the same conditions. Cells were harvested by centrifugation at 1000 g for 5 minutes at 4 °C, the pellet was resuspended in an equal volume of 30% glycerol and then frozen at -80 °C.

[0426] Yeast display transformation of the CD81_LEL library 2 resulted in: Library 2_4: 9.5×10 6 non-dependent members Library 2_5: 2.1×10 7 non-dependent members Library 2_6: 1.7×10 7 non-dependent members which were generated.

[0427] Example 7: Selection of CD81 LEL library 2 with human EGFR-Fc For selection, first the library was cultured in SD-CAA medium supplemented with penicillin-streptomycin with shaking overnight at 30 °C, then expression of the recombinant protein was induced by resuspending the yeast cells in SG / R-CAA medium with penicillin-streptomycin and incubating this with shaking overnight at 37 °C.

[0428] Next, 10 9The cells were pelleted at 2500 g for 5 minutes and then washed by resuspending the pellet in 50 ml of wash buffer (1× D-PBS (Thermo Fisher Scientific), 0.25% BSA (Sigma Aldrich), 2 mM EDTA (Sigma Aldrich)), and pelleted again at 2500 g for 5 minutes. The washed cells were resuspended in 25 ml of wash buffer containing 0.25 μM biotinylated antigen, incubated at room temperature for 30 minutes with gentle agitation, and further incubated in an ice bath for 10 minutes. Then, the cells and antigen solution were pelleted at 2,500 g for 5 minutes at 4°C, washed twice with 50 ml of wash buffer, and pelleted after each wash step. The cells were resuspended in 5 ml of wash buffer + 25 μl of streptavidin microbeads and incubated on ice for 10 minutes. Finally, 15 ml of wash buffer was added.

[0429] The LS column (Milteny Biotec) was positioned in the separator and 3 ml of wash buffer was applied to precondition the column. Then, the cell solution was loaded in 7 ml batches. The drip of the column was stopped once, and the column was briefly removed from the magnet and repositioned on the magnet to allow the captured cells to perfuse through in order to change the orientation of the beads inside the column. 1 ml of wash buffer was applied to the column before loading the next 7 ml cell suspension. These steps were repeated until all the cells were loaded. Then, the column was washed with 3 ml of wash buffer, briefly removed from the magnet, and washed again with 3 ml of wash buffer. To elute the bound cells, the column was removed from the magnet and 5 ml of wash buffer was added. Using the supplied plunger, the cells were pushed out of the column into a new tube. The fraction of bound cells was pelleted at 2500 g for 5 minutes. The pellet was resuspended in 10 ml of SD-CAA medium and incubated overnight at 30°C at 180 rpm. Further, 100 μl of 10 -2 dilution was seeded onto MDL plates and incubated at 30°C.

[0430] The induced cell suspension was diluted to 10 per ml in 10% BSA-PBS. 8 The cells were diluted to 1 ml of PBS and centrifuged at 3000 rpm for 5 min at 20°C. To block the cells, the pellet was resuspended in 1 ml of 10% BSA-PBS and incubated on a rotating platform at 20°C for 30 min. The cells were centrifuged at 3000 rpm for 5 min at 20°C and resuspended in 250 μl of 10% BSA-PBS containing 1 μM biotinylated EGFR-Fc (Sino Biological). After incubation on a rotating platform at 20°C for 30 min, the cells were centrifuged at 3000 rpm for 5 min at 4°C. To wash the cells, the pellet was resuspended in 1 ml of ice-cold PBS and centrifuged at 3000 rpm for 5 min at 4°C. The cells were then resuspended in 250 μl of 10% BSA-PBS with streptavidin-Alexafluor 647 (1:800) and anti-V5-FITC antibody (1:100) (Thermo Fisher Scientific) and incubated on ice for 30 minutes. The cells were centrifuged at 3000 rpm for 5 minutes at 4° C., after which they were resuspended in 1 ml of ice-cold PBS and centrifuged again. Finally, the cells were resuspended in 250 μl of ice-cold PBS and kept on ice until sorting by FACS Aria™.

[0431] After sorting, cells were resuspended in an appropriate volume of SD-CAA and incubated at 30° C. and 180 rpm before induction as described above.

[0432] [Table 4]

[0433] Example 8: Expression of modified CD81 LEL in a soluble form The sequence encoding CD81 was amplified by oligonucleotide

[0434] [ka]

[0435] were amplified from yeast display clones using The PCR fragments were digested with the restriction enzymes NheI and BstEII and ligated into the vector pTT28(CNRC) digested with the same enzymes. The ligation mixture was transformed into electrocompetent Escherichia coli TOP10 (Thermo Fisher Scientific), and the transformants were selected on ampicillin plates. The plasmid was isolated by minipreparation and transfected into ExpiCHO cells (Thermo Fisher Scientific). Expression in the ExpiCHO expression system (Thermo Fisher Scientific) is based on the MaxTiter protocol following exactly the manufacturer's instructions. The hCD81 LEL variant was then purified by Ni-NTA chromatography using standard protocols. The supernatant was diluted with an equal volume of AD and buffered with PBS and 20 mM imidazole and a pH adjusted to 7.5. An Excel Ni-NTA column (GE Healthcare) was equilibrated at pH 7.5 with PBS and 20 mM imidazole, and the buffered supernatant was loaded. Elution was performed at pH 7.5 in five column volumes with a linear gradient of 20 mM to 500 mM imidazole in PBS. The protein-containing fractions were pooled and dialyzed overnight at 4°C against 100 volumes of PBS.

[0436] SEC analysis under native conditions revealed a monodisperse elution profile corresponding to a dimeric form of the protein, similar to soluble wild-type CD81 LEL. Alternatively, the protein was expressed in the ExpiCHO expression system (Thermo Fisher Scientific) according to the MaxTiter protocol exactly following the manufacturer's instructions. Then, hCD81 LEL was purified by Ni-NTA chromatography using a standard protocol. The supernatant was diluted with an equal volume of AD and buffered with PBS, 20 mM imidazole, and a pH adjusted to 7.5. An Excel Ni-NTA column (GE Healthcare) was equilibrated with PBS and 20 mM imidazole at pH 7.5, and the buffered supernatant was loaded. Elution was performed at pH 7.5 with a linear gradient of 20 mM to 500 mM imidazole in PBS over 5 column volumes. The protein-containing fractions were pooled and dialyzed overnight at 4°C against 100 volumes of PBS.

[0437] Example 9: Thermal Stabilization of CD81 LEL A fragment encoding hCD81 LEL (fragment Phe113-Lys201) (numbered based on SEQ ID NO: 87) was amplified from a synthetic construct by the full-length CD81 sequence (Geneart). DSDBASE (Vinayagam et al., 2004, Nucleic Acids Research, Volume 32, Issue suppl_1, pp. D200-D202, https: / / doi.org / 10.1093 / nar / gkh026) was used as a prediction tool to identify positions that might have cysteine residues suitable for the formation of intradomain disulfide bonds. The algorithm was used to analyze the hCD81 LEL crystal structure 1G8Q for the distance between the C α atom and the C β atom of adjacent amino acid residues, as well as the torsion angle and the resulting S-S bond length. Of the 36 predicted possible disulfide bonds, 11 of the most likely to succeed as judged by visual inspection of the crystal structure were selected. Five of these were predicted by the DSDBASE program in both promoter A and promoter B of hCD81 LEL.

[0438] The introduction of mutations of a single selected amino acid residue to cysteine was carried out using the QuikChange Lightning Mutagenesis kit (Agilent) exactly according to the manufacturer's instructions using the oligonucleotides listed in Table 5.

[0439] [Table 5]

[0440] The hCD81 LEL variants were cloned into the pTT22SSP4 mammalian expression vector (CNRC) and expressed in two different expression systems. For preliminary screening, the constructs were grown in HEK293-6E cells (CNRC) at 2 mL scale in F17 medium supplemented with 4 mM glutamine and 50 μg / mL G-418 (Thermo Fisher Scientific) on an orbital shaker at 180 rpm at 37 °C for 4 days under 5% CO2, with TN-20 supplied to a final concentration of 0.8% on day 2 post-transfection. Mutant C5 did not express, C6 expressed poorly, and C10 formed distinct dimers and were thus omitted from further analysis. The mutants selected for further characterization (C1, C2, C3, C4, C7, C8, C9, and C11) were transfected into ExpiCHO cells (Thermo Fisher Scientific) according to the manufacturer's instructions. Cell culture was continued according to the MaxTiter protocol (Thermo Fisher Scientific). After 14 days, the supernatant was harvested and purified using Ni-NTA affinity chromatography. After clarification, the sample was buffered with phosphate-buffered saline (PBS) with 20 mM imidazole at pH 7.5 and passed through an Excel Ni-NTA column (GE Healthcare) equilibrated with the same buffer. His-tagged hCD81 LEL was eluted with a gradient of 20–500 mM imidazole in 5 column volumes. Fractions containing the target protein were pooled and dialyzed twice against 100 volumes of PBS overnight at 4 °C. The protein was stored at -80 °C until use.

[0441] SEC-HPLC was performed using a Shimadzu LC-20A Prominence system equipped with a diode array detector and a refractive index detector on a Superdex 200 Increase 10 / 300 GL column (GE Healthcare). The mobile phase buffer used was PBS with 200 mM NaCl. Chromatography was performed at a constant flow rate of 0.75 mL / min. A total of 200 μg of protein at approximately 2 mg / mL was loaded onto the column for analysis. Column calibration was performed using a set of molecular weight standards (Bio-Rad) in the range of 10 - 500 kDa. Mutants C1 and C8 did not show well-resolved peaks, but all other mutants were similar to the wild-type protein.

[0442] DSC experiments were performed using an automated MicroCal PEAQ-DSC Automated system (Malvern) with an 80 μM protein solution diluted in PBS at pH 7.4. Heating was performed from 20 °C to 110 °C at a heating rate of 1 °C / min. The protein solution was then cooled in situ and the same thermal scan was performed to obtain a baseline for subtraction from the first scan. All measurements were performed in duplicate. Approximation was carried out by the MicroCal PEAQ-DSC software using a non-2-state transition mechanism.

[0443] Measurement of the thermal stability of wild-type CD81 LEL revealed a single melting point at 66.15 °C, but revealed a low enthalpy of 2.4×10 4 kcal / mol.

[0444]

Table 6

[0445] A differential thermal analysis scan was performed, and furthermore, a rescan of the denatured protein solution to be subtracted as background was carried out. Here, in contrast to the wild-type CD81 LEL and other stabilized variants, the mutant C2 was surprisingly found to exhibit reversible unfolding up to a maximum of 110 °C.

[0446] Next, an hCD81 LEL mutant with a combination of the newly discovered strongly stabilizing disulfide bonds Ala134Cys / Lys144Cys and Val135Cys / Ser168Cys was produced and its thermal stability was tested. Thermal unfolding was recorded up to a maximum of 130 °C, which continued in a single event with a Tm of 109.40 ± 0.25 °C, higher than 43 °C of the wild-type CD81 LEL. This protein, called C4C9, migrated as a single sharp peak in the time profile of the wild-type protein in SEC under native conditions.

[0447] Importantly, the stabilized mutant was able to bind to the structure-reporter antibody M38 to the same extent as the wild-type CD81 LEL. ELISA plates (Maxisorp, NUNC) were coated with 5 μg / mL of anti-hCD81 M38 antibody (Thermo Fisher Scientific) in PBS for 1 h at room temperature. After blocking with 5% bovine serum albumin (BSA)-PBS for 1 h at room temperature, the supernatant of HEK293-6E cells transfected with the hCD81 LEL variant or the purified variant of hCD81 LEL diluted with 2.5% BSA-PBS was allowed to bind for 1 h at room temperature. After thorough washing, the binding of the mutant protein was detected with an anti-his-horseradish peroxidase (HRP)-labeled antibody (QIAgen) diluted 1:2000 in 2.5% BSA-PBS. Antibody binding was confirmed with 3,3’,5,5’-tetramethylbenzidine (TMB) (Sigma Aldrich), and the reaction was stopped by adding an equivalent amount of 30% H2SO4, and the absorbance was read at 450 / 620 nm.

[0448]

Table 7

[0449] The Far-UV CD spectra of the mutant C4C9 and wild-type CD81 were examined. The CD spectra were measured in a Chirascan spectropolarimeter (Applied Photophysics). A 1-mm quartz cuvette was used. The protein preparation was diluted to 1 mg / mL in PBS. The spectra were found to be identical in wild-type hCD81 LEL and C4C9. The observed spectra were typical of highly α-helix-containing proteins, having two characteristic minima at 208 nm and 222 nm, as in previously published results.

[0450]

Table 8-1

[0451]

Table 8-2

[0452] Example 10: Production of Target-Specific EVs in Cell Culture 10.1 Transient Transfection of a Target-Specific EV Donor Cell Line Using HeLa Cells HeLa cells expressing recombinant CD81 fused to GFP at the C-terminus were transiently transfected using electroporation to secrete target-specific CD81-GFP-containing exosomes (CD81-GFP-exosomes). Two to three days prior to transfection, HeLa cells (5×10 6Cells were seeded into T175 cell culture flasks (Sigma, Kremsmunster, Germany). For regular culture of donor HeLa cells, complete RPMI-1640 medium (Merck, Darmstadt, Germany) containing 10% FCS and 4 mM L-glutamine was used. On the day of transfection (day 0), HeLa cells (optimally at a culture density of 70 - 85%) were harvested using 0.1% Trypsin solution (5-minute treatment, neutralized with complete RPMI-1640). HeLa cells detached in RPMI-1640 were quantified using a ViCell Cell Counter. 3×10 6 cells were transfected with the CD81-GFP plasmid (pTT5, NRC Canada, Ottawa, Canada, 4 μg) to fill one T175 flask. Electroporation was performed using an Amaxa Nucleofector I device (Lonza, Basel, Switzerland) according to the manufacturer's brochure (program I-013 for HeLa cells) using in-house prepared electroporation buffer (5 mM KCl, 15 mM MgCl2, 120 mM Na2HPO4 / NaH2PO4 pH 7.2, 50 mM mannitol, 0.005% Pluronic F-68). The electroporated cells were recovered in complete RPMI-1640 and cultured for 24 hours to attach the cells to the culture flask surface. Indeed, various forms of the human transferrin receptor targeting recombinant CD81 were localized to the plasma membrane, suggesting functional integration of the recombinant tetraspanin (Figure 4). 10.2 Isolation and purification of target-specific EVs On the first day (24 hours after transfection), the culture medium was removed and replaced with an equal volume (45 mL for 1×T175) of EV-depleted secretion medium (RPMI-1640). The supernatant fraction was collected from complete RPMI-1640 (containing 20% FCS) that had been ultracentrifuged (14 - 18 hours, 4°C, 100,000 g), and the medium was diluted to 10% FCS with RPMI-1640 and supplemented with L-glutamine to a final concentration of 4 mM to prepare EV-depleted RPMI-1640. To secrete target-specific EVs, HeLa cells were incubated with the EV-depleted medium for 72 hours (37°C, 5% CO2 atmosphere).

[0453] At 72 hours after the collection time, EV isolation was performed by using ultracentrifugation. The conditioned cell culture medium was collected and centrifuged at 4°C for 30 minutes (3,100×g) to remove cells and cell debris. The supernatant was passed through a 450 nm syringe filter unit (Merck, Darmstadt, Germany) and transferred to a 100 mL sealable ultracentrifuge tube. Then, the supernatant was centrifuged at 4°C for 90 minutes (100,000×g) using an ultracentrifuge (Optima L-60, Beckman Coulter, Brea, USA). The supernatant was removed, and the pellet was collected in the remaining 1 mL of medium. By performing a second step of ultracentrifugation (100,000×g, 90 minutes, 4°C), a smaller amount of EVs was collected. The supernatant was removed again, and the remaining EV pellet was resuspended in 200 - 400 μl of HEPES-based 1×Live Cell Imaging Solution (ThermoFisher Scientific, Waltham, USA). Using this method, approximately 10 11 EV / ml was purified from the HeLa supernatant.

[0454] Example 11: Generation of EVs Retaining a Trackable and Purifiable Tag 11.1 Sequence and Cloning of CD81-Snorkel Tag in a Retroviral System The wild-type human CD81 sequence (amino acid sequence number 87; nucleic acid sequence number 88) was cloned into a pBMN vector with a SNORKEL tag fused to the C-terminus (Brown et al. 2013, PLoS ONE 8(9): e73255. doi:10.1371 / journal.pone.0073255). The SNORKEL tag enables the tag to be presented on the surface of the vesicle membrane. The PCR product of CD81 was digested with NcoI and AgeI, and the PCR product of the SNORKEL tag was digested with AgeI and NotI. The pBMN plasmid was digested with NcoI and NotI. The digested PCR products ligated with NcoI and NotI digested the pBMN plasmid. The ligation mixture was transformed into competent E. coli cells, and the transformants were selected on ampicillin plates. The plasmid was isolated by an endotoxin-free plasmid preparation (Qiagen Maxiprep). 11.2 Stable expression of cD81-Snorkel tag in EV donor cells using retroviral transfection Phoenix cells (5×10 6The cells were seeded onto a T75 plate (Sigma, Kremsmunster, Germany) one day before transfection (day 0). The cells were cultured in a complete growth medium using DMEM with 4.5 g glucose, 10% FCS, and 4 mM L-glutamine. When the cells reached 60 - 70% confluence, the culture medium was removed and replaced with DMEM with 4.5 g glucose, 4 mM L-glutamine without FCS. The cells were transfected with 10 μg of the pBMN-CD81-SNORKEL tag (SNORKEL tag fused to the C-terminus of CD81) plasmid using the jetPRIME transfection reagent (Polyplus, France). Twenty-four hours after transfection (day 1), the medium was changed to 7 ml of complete growth medium. The target cells (HeLa) were seeded in a 6-well plate (Sigma, Kremsmunster, Germany) in RPMI-1640 medium containing 10% FCS and 4 mM L-glutamine at 1×10 6Cells / wells were seeded. Twenty-four hours after seeding (day 2), HeLa cells (50-80% confluence) can be infected. 7 mL of supernatant from Phoenix cells (containing virus particles) was filtered using a 0.45 μm filter (Merck, Darmstadt, Germany), and polybrene (final concentration 8 μg / mL) was mixed into this filtrate. Fresh growth medium was added onto the Phoenix cells. The supernatant from the target cells was removed, and up to 2 mL of virus-containing supernatant was added. The entire plate was covered with paraffin and centrifuged at 800×g for 60 minutes at room temperature. Then, the virus supernatant was discarded, and fresh growth medium was added to the HeLa cells. Viral infection was carried out for the next 3 days until the cells expressed the CD81-SNORKEL tag at 95-100%. The tagged CD81 was localized to the cell membrane, which indicated correct folding and localization (Figure 5A). The method was also successfully carried out with the CD63 fusion protein in HeLa cells together with the SNORKEL tag fused to the C-terminus of CD81 (Figure 5B). Similar results were observed in human adipose-derived stem cells (ASCs). EVs were positive for both the HA tag and CD81 when electron microscopy was performed after immunogold labeling with an anti-HA antibody conjugated to 10 nm gold particles and an anti-CD81 antibody conjugated to 18 nm gold particles. 11.3 Characterization of EVs for Size, Number, and Incorporation of the CD81-SNORKEL Tag HeLa cells stably expressing the CD81-SNORKEL tag were seeded in 3 T75 plates (5×10 6(Cells / Flask). 24 hours after seeding (Day 1), the culture medium was removed and replaced with the respective amount (12 mL for 1×T75) of EV-depleted secretion medium (RPMI-1640). The supernatant fraction was collected from ultracentrifuged (14 - 18 hours, 4°C, 100,000 g) complete RPMI-1640 (containing 20% FCS), and the medium was diluted with RPMI-1640 to 10% FCS, and L-glutamine was added to a final concentration of 4 mM to prepare EV-depleted RPMI-1640. To secrete EVs with SNORKEL tags, HeLa cells were incubated with the EV-depleted medium for 48 hours (37°C, 5% CO2 atmosphere).

[0455] After culturing the cells for 48 hours in the EV-depleted medium, isolation of EVs retaining the SNORKEL tag was performed. The conditioned medium was collected and centrifuged at 700 g for 5 minutes at 4°C to remove cell debris. A second round of centrifugation of the supernatant at 2000 g for 10 minutes at 4°C was performed to remove apoptotic bodies and larger particles. Next, the supernatant was passed through a 0.22 μm filter (Merck, Darmstadt, Germany) to remove larger particles such as microvesicles. The processed supernatant was concentrated to 1 / 20 of the initial volume using a tangential flow filtration column equipped with a 300 kDa pore size hollow fiber (SpectrumLabs, Netherlands). The concentrated conditioned medium was used to further characterize the EVs. 10 μl of the concentrated conditioned medium was diluted up to 1000-fold in filtered DPBS (Merck, Darmstadt, Germany) and used for nanoparticle tracking analysis (NTA) in the scattering mode by nanosight NS500. All acquisition parameters were the same for size and concentration measurements. All experiments were measured in experimental triplicates. The size of EVs from various isolates was approximately 110 - 140 nm.

[0456] The incorporation of the CD81-SNORKEL tag into EVs was confirmed by Western blot analysis. 1E10 EVs quantified by nanosight were loaded per well together with 50 μg of cell lysate quantified by the BCA kit (Thermo fisher scientific) in a NuPAGE 4–12% Bis-Tris protein gel, and SDS-PAGE was performed. Proteins on the gel were transferred to a PVDF membrane using a Trans-Blot turbo transfer system (Bio-Rad). Specific antibodies against EV-specific markers with the SNORKEL tag for quantification. Indeed, EVs showed markers of typical EV proteins in Western blots including CD63, CD81, and TSG101, as well as the absence of the intracellular protein calnexin. 11.4 Isolation of EVs exclusively retaining the SNORKEL tag using anti-HA matrix and PreScission protease The conditioned medium from HeLa cells expressing the CD81-SNORKEL tag was treated by fractionation centrifugation (700 g and 2000 g), filtered using a 0.22 μm filter (Merck, Darmstadt, Germany), and concentrated to 1 ml using a tangential flow filtration column equipped with 300 kDa pore size hollow fibers (SpectrumLabs, Netherlands). 350 μl of the concentrated conditioned medium was mixed with 200 μl of anti-HA antibody conjugated to agarose beads (Sigma) and incubated overnight at 4 °C in a tube rotator. After 16 - 18 hours of incubation, the agarose beads were rotated at 500 g for 5 minutes at 4 °C, the flow-through (unbound EVs) was collected, and the beads were washed with 1 ml of filtered DPBS by rotating the agarose beads at 500 g for 5 minutes at 4 °C. Next, 4 μl of 8 units of PreScission protease treatment (sigma) in 50 mM Tris (Sigma), 150 mM NaCl (Sigma) at pH 7 - 7.4 was applied to the agarose beads conjugated with anti-HA antibody that were bound to the SNORKEL tag on the EVs in a tube rotator at 4 °C for 16 - 18 hours. After 16 - 18 hours of PreScission protease treatment, the agarose beads were rotated at 500 g and the supernatant was collected. To quantify the number of EVs retaining the purified SNORKEL tag, all samples (input, flow-through, wash, and eluate) were used for NTA analysis at dilutions from 1-fold to 1000-fold. After PreScission digestion, the band was not detectable in the eluate fraction using the cleaved HA tag, but the remaining CLIP tag was sufficiently visible in the eluate fraction, so purification using PreScission protease resulted in ~80 - 90% of the vesicles gently eluted from the affinity matrix. By comparing the eluate fraction with the fraction remaining on the beads without elution, a yield of ~80 - 90% was observed (Figure 6).

[0457] Insertion of an antigen-specific ligand, such as Gas6, etc., upstream of the PreScission protease site in the SNORKEL tag enables purification of specific EVs retaining the SNORKEL tag and targeting of the EVs after purification.

[0458] Example 12: Characterization of Target-Specific Extracellular EVs 12.1 Characterization of EV Size and Incorporation Ratio of Recombinant Proteins 10 μl of the EV isolate was diluted with 6 mL of filtered DPBS (Merck, Darmstadt, Germany) and used for nanoparticle tracking analysis (NTA) in scattering and fluorescence modes by ZetaView (Particle Metrix, Inning, Germany), followed by evaluation using NTA software (Particle Metrix, Inning, Germany). All acquisition parameters were identical for size and concentration measurements. All experiments were measured in experimental triplicates.

[0459] The ratio of the number of particles measured in scattering and fluorescence modes is equal to the ratio between recombinant EVs with the target-specific CD81-LEL moiety and fluorescent GFP and non-recombinant and non-fluorescent native EVs. For various batches of recombinant EV production, the ratio obtained from NTA measurements indicated that approximately 30 - 50% of all isolated EVs were recombinants. The general size median of EVs isolated by the isolation method mentioned above was approximately 120 - 140 nm, and the typical yield of target-specific fluorescent EVs reached 1600 - 5000 recombinant EVs per HeLa cell in the size range of 60 - 400 nm, with a peak at 130 nm in the diameter of the particle size (Figure 7). 12.2 Qualification and Quantification of Specific EV Uptake in vitro (a) Qualification of EV Uptake in Target Antigen-Positive Cells by Fluorescence Microscopy Recipient cells (e.g., Caco-2 cells) were seeded in an 8-well ibidi glass bottom plate (2.5×10 per well) 4Cells were allowed to adhere for 24 hours in complete culture medium (37 °C, 5% CO2 atmosphere). After the incubation time, 10 9 of recombinant EVs were added to each well. Cells were cultured for an additional 24 hours to allow recipient cells to take up the recombinant particles. Finally, cells were washed with PBS and surface-bound EVs were removed by treating on ice with acidic glycine buffer (100 mM NaCl, 100 mM glycine, pH 3.5). Cells were then imaged at 40× magnification using a fluorescence microscope (DMI3000B; Leica Microsystems, Wetzlar, Germany). Microscopy imaging parameters (exposure time, contrast, and gain) were the same in all experiments. Indeed, recombinant EVs were readily taken up by recipient cells (Figure 8A).

[0460] (b) Quantification of EV uptake in target antigen-positive cells by flow cytometry Recipient cells (e.g., Caco-2 cells) were seeded in 24-well plastic-bottom plates (0.25 × 10 6 cells per well) and allowed to adhere for 24 hours in complete culture medium (37 °C, 5% CO2 atmosphere). After the incubation time, 5 × 10 9 of a defined number of recombinant EVs were added to each well. Cells were cultured for an additional 24 hours to allow recipient cells to take up the recombinant particles. Finally, cells were washed with DPBS, treated with 0.1% trypsin (5 minutes, 37 °C), and neutralized with complete medium containing 10% FBS. Cells were then centrifuged at 300 g and the pellet was resuspended in ice-cold DPBS.

[0461] Samples were maintained on ice and measured by Cytoflex S flow cytometry (Beckman Coulter, Brea, USA). Data were analyzed with Cytoflex software. Mean fluorescence intensity was normalized to control / untreated cell samples (ΔMFI). Here, a ~30 - 60% increase in the uptake of recombinant EVs targeting the transferrin receptor was observed depending on the recombinant CD81 constructs (Tfr1EVsP1, Tfr2EVs P1, Tfr2CP4EVs P1) when compared to EVs (CP4EVs) or wild - type EVs (wtEVs) overexpressing non - Tfr1 targeting CD81 (Figure 8B).

[0462] Example 13: Preparation of Therapeutic Agent - Loaded Target - Specific EVs Using Apoptosis - Inducing siRNA 13.1 Transfection of Recombinant EVs with siRNA Purified EVs retaining either non - recombinant or recombinant CD81 targeting the human transferrin receptor were transfected with either apoptosis - inducing siRNA (TOX transfection control, Dharmacon) or non - targeting control siRNA (ON - TARGETplus Non - targeting siRNA, Dharmacon) using a liposome - based transfection reagent, Dharmafect (Dharmacon, Lafayette, USA). EVs were transfected with each targeting or non - targeting siRNA according to the manufacturer's brochure (up to a final concentration of 25 nM siRNA). siRNA loading was controlled for qPCR use. 13.2 Cell - Based Assay for Determination of Target - Specific Cytotoxicity Target antigen - expressing cells were seeded in complete medium at a density of 7.5×10 3 cells / well in a 96 - well plate and then 5×10 8It was incubated for 72 hours with the recombinant / siRNA transfected EV. The experiment was performed in six replicates per EV administration or EV control. After removal of the medium and washing with PBS, the cells were incubated for 1 hour, 2 hours, and 4 hours, or overnight with 1×WST-1 (Sigma, St. Louis, USA) or 1×AlamarBlue (ThermoFisher Scientific, Waltham, USA) in complete medium. The assay readout was performed in a microplate reader Infinite 200 Pro (Tecan, Mannedorf, Switzerland) according to the respective manufacturer's brochure. In fact, by using, among others, the Tfr1CP4 recombinant CD81 construct, about 30% higher cytotoxicity was observed in EV targeting the transferrin receptor compared to the control (Figure 10).

[0463] Example 14: Design of Library CD81 LEL_L3 The CD81 LEL_L3 library design is based on a reversibly refolding CD81 LEL stabilized mutant (SEQ ID NO: 180).

[0464]

Chemical formula

[0465] In this library, the stabilized mutants Ala130Cys / Ala146Cys and Val135Cys / Ser168Cys function additively to increase the midpoint of the thermal transition of CD81 LEL to 93.4 °C, and this combinatorial mutant can reversibly refold when heated up to 110 °C. In this library, by randomizing the amino acid residues at positions 132 - 133, 136 - 141, 162 - 165, 167, and 171 - 172, a composite surface that can be utilized for antigen binding was formed (X in SEQ ID NO: 1) (SEQ ID NO: 97).

[0466]

Chemical formula

[0467] In this case, X is any amino acid. Example 15: Construction of Yeast Display Libraries CD81 LEL_L2 and L3 For the yeast display library CD81 LEL_L2, a PCR recombinant product encoding a randomized insert was generated in a 100 μl aliquot using Q5 HiFi polymerase MasterMix (New England Biolabs), 10 ng / μl of template (pTT22SSP4_C4C9), and 50 pmol of oligonucleotides LIB2FWD (SEQ ID NO: 98) and EFrev (SEQ ID NO: 99). After an initial denaturation at 98°C for 30 seconds, 35 cycles were performed with denaturation at 98°C for 20 seconds each, annealing at 55°C for 20 seconds, and extension at 72°C for 20 seconds, followed by an incubation step at 72°C for 5 minutes to complete. The PCR product was purified using the Illustra GFX purification kit and eluted in AD.

[0468] [Chemical formula]

[0469] [wherein, N is any one of A, C, G, or T].

[0470] [Chemical formula]

[0471] [wherein, N is any one of A, C, G, or T; and M is any one of A or C]. For the yeast display library CD81 LEL_L3, a PCR recombinant product encoding a randomized insert was generated in 100 μl aliquots using Q5 HiFi Polymerase MasterMix (New England Biolabs), 10 ng / μl of template (pTT22SSP4_C2C9), and 50 pmol of oligonucleotides LIB3FWD (SEQ ID NO: 100) and LIB2REV2 (SEQ ID NO: 99). After an initial denaturation at 98 °C for 30 seconds, 35 cycles were performed with denaturation at 98 °C for 20 seconds each, annealing at 55 °C for 20 seconds, and extension at 72 °C for 20 seconds, followed by an incubation step at 72 °C for 5 minutes to complete. The PCR product was purified using the Illustra GFX purification kit (GE Healthcare) and eluted in AD.

[0472] [Chemical formula]

[0473] [wherein N is any one of A, C, G, or T]. The recipient vector pYD1 delbamdelhind_bamhind was linearized using BamHI and HindIII, purified using preparative agarose gel electrophoresis and the Illustra GFX purification kit (GE Healthcare), and eluted in AD.

[0474] Yeast Saccharomyces cerevisiae EBY100 was transformed with the linearized recipient vector and the PCR recombinant product using the chemical PEG3350 / Li-acetate / ssDNA transformation method. Two libraries, L2A and L2B and L3A and L3B, were generated with each of the recombinant fragments. For each library, 250 ml of YPD medium was inoculated and an OD of 0.4 600They were cultured overnight. After 5 hours of incubation in a shaking incubator at 30 °C, yeast cells were harvested in 50 ml aliquots. The supernatant was removed, and the yeast cells were washed with 25 ml of AD per aliquot at 3000 rpm for 5 minutes at room temperature. The cell pellet was resuspended in 3 ml of 100 mM Li-acetate per aliquot and shaken at 200 rpm for 15 minutes at 30 °C. The cells were collected by centrifugation at 3000 rpm for 5 minutes at room temperature. To each yeast cell aliquot, a mix of 2750 μl of 50% PEG3350 solution, 360 μl of 1 M Li-acetate, 500 μl of heat-shocked ssDNA, 10 μg of linearized recipient vector, and 10 μg of recombinant PCR fragment was added. Incubation was carried out at 200 rpm for 30 minutes at 30 °C, followed by a heat shock at 42 °C for 45 minutes. Yeast cells were collected by centrifugation at 3000 rpm for 5 minutes at room temperature and diluted into 250 ml of SD-CAA medium with penicillin and streptomycin. Incubation was continued at 200 rpm for 24 hours at 30 °C. 12.5 ml of the culture was transferred to 250 ml of fresh SD-CAA medium with penicillin and streptomycin at 30 °C for 24 hours, pelleted by centrifugation at 3000 rpm for 10 minutes at 4 °C, mixed with the same volume of 30% glycerol, and then frozen. The size of the library was determined using dilution plating, and it was determined that there were 1.1×10 8 non-dependent members for L2 and 1.2×10 8 non-dependent members for L3.

[0475] To determine the level of library validity, plasmid DNA was isolated from 10 μl of pelleted yeast cells and transformed into E. coli TOP10 using electroporation. The expression cassette of the CD81 LEL mutant was amplified using primers pydfwd (SEQ ID NO: 101) and pydrev (SEQ ID NO: 102), and sequencing was performed using one of these primers. The validity of library L2A was found to be 62.5%, L2B was 87.5%, L3A was 62.5%, and L3B was 100%.

[0476]

Chemical formula

[0477] For quality control, the yeast cells were induced in SG / RCAA medium with penicillin and streptomycin either at 20 °C for 48 hours or at 37 °C for 24 hours. For staining, the yeast cells were adjusted to an OD of 1 600At room temperature, they were blocked in 2% BSA-PBS solution for 30 minutes. Then, they were resuspended in 100 μl aliquots and stained with anti-Xpress antibody (Thermo Sientific) (1:1000) that reacts with the Xpress tag located at the N-terminus, and M38 antibody (Thermo Scientific) (1 μg / ml), which detect properly folded CD81 LEL in 2% BSA-PBS at room temperature for 1 hour. The cells were pelleted at 3000 rpm for 5 minutes at 4°C and resuspended in 2% BSA-PBS with goat anti-mouse (Fab’)2-FITC conjugate (Sigma Aldrich) diluted 1:200 in 2% BSA-PBS. Other stainings were anti-his-tag-Alexa Fluor 488 (QIAgen) diluted 1:200 in 2% BSA-PBS, and anti-V5-tag-FITC (Thermo Scientific) diluted 1:100 in 2% BSA-PBS, which were used to detect the C-terminal his tag and the C-terminal V tag to determine the appropriate read-through of the clone. Incubation with the fluorescent antibodies was performed on ice for 30 minutes. Finally, the cells were collected at 3000 rpm for 5 minutes at 4°C and resuspended in 200 μl of ice-cold PBS. The fluorescence of the stained samples and unstained controls was determined using Guava EasyCyte flow cytometry. The percentage of positive cells was determined and presented in Table 9.

[0478]

Table 9

[0479] Example 16: EGFR-Specific Binding Based on CD81 LEL 16.1 Selection of CD81 LEL Libraries L2 and L3 with Human EGFR-Fc Human EGFR-Fc was purchased from Sino Biological. For biotinylation, the EZ-Link™ Sulfo-NHS-LC-LC-biotin reagent (Thermo Scientific) was used at a biotin to protein molar ratio of 3:1. The antigen was reconstituted to a concentration of 0.25 μg / μl according to the manufacturer's instructions exactly. Incubation with the biotinylation reagent was continued for 1 hour at room temperature with shaking. Unbound biotin was removed by dialysis against 100-fold volume of PBS at 4°C overnight with stirring using a Snakeskin dialysis tube (Thermo Scientific) with a 10,000 Da MWCO.

[0480] For selection, first, libraries 2A, 2B, 3A, and 3B were cultured in SD-CAA medium supplemented with penicillin-streptomycin with shaking overnight at 30°C, and then the expression of the recombinant protein was induced by resuspending the yeast cells in SG / R-CAA medium supplemented with penicillin-streptomycin, and it was incubated with shaking overnight at 37°C.

[0481] For MACS, 10 9 induced yeast cells were pelleted by centrifugation at 1000 g for 5 minutes, and the pellet was washed by resuspending it in 50 ml of wash buffer (PBS, pH 7.2, 0.25% BSA, 2 mM EDTA), and then pelleted again by centrifugation at 2500 g for 5 minutes. The washed cells were resuspended in 5 ml of wash buffer containing 0.5 μM biotinylated antigen and incubated at room temperature for 30 minutes with gentle agitation. Antigen binding was quenched by adding 10 ml of ice-cold wash buffer, and the cells and antigen solution were pelleted by centrifugation at 1000 g for 5 minutes at 4°C. The cells were resuspended in 5 ml of wash buffer + 25 μl of streptavidin microbeads (μMACS streptavidin kit, MACS Miltenyi) and incubated on ice for 10 minutes. Finally, 15 ml of wash buffer was added and the cells were passed through a 40 μl cell strainer.

[0482] An LS column (MACS Miltenyi) was placed in the separator and 3 ml of wash buffer was applied to precondition the column. Then, the cell solution was loaded in 7-ml batches. The column drip was stopped once and the column was briefly removed from the magnet and repositioned on the magnet to allow the captured cells to perfuse in order to change the orientation of the beads within the column. 1 ml of wash buffer was applied to the column before loading the next 7-ml cell suspension. These steps were repeated until all the cells were loaded. The column was then washed with 3 ml of wash buffer, briefly removed from the magnet and washed again with 3 ml of wash buffer. To elute the bound cells, the column was removed from the magnet and 5 ml of wash buffer was added. The cells were extruded from the column into a new tube by using the supplied plunger. The fraction of bound cells was pelleted at 2500 g for 10 minutes. The pellet was resuspended in 10 ml of SD-CAA medium, 10 μl of the eluted cells was diluted in 990 μl of SD-CAA, 100 μl was seeded onto an MDL plate and incubated at 30 °C for 3 days to evaluate the output of the MACS procedure, and the remaining cells were incubated overnight at 30 °C and 180 rpm.

[0483] In subsequent FACS selection rounds, the induced cell suspension was adjusted to 10 cells per 1 ml of 10% BSA-PBS 8Dilute to the cells and centrifuge at 3000 rpm for 5 minutes at 20°C. To block the cells, resuspend the pellet in 1 ml of 10% BSA-PBS and incubate on a rotary shaker at 20°C for 30 minutes. Centrifuge the cells at 3000 rpm for 5 minutes at 20°C and resuspend in 250 μl of 10% BSA-PBS containing 0.5 μM biotinylated EGFR-Fc. Incubate on a rotary shaker at 20°C for 1 hour, then centrifuge the cells at 3000 rpm for 5 minutes at 4°C. To wash the cells, resuspend the pellet in 1 ml of ice-cold PBS and centrifuge at 3000 rpm for 5 minutes at 4°C. Then resuspend the cells in 250 μl of 10% BSA-PBS with streptavidin-Alexa Fluor 647 (1:800) and anti-V5-FITC antibody (1:100) and incubate on ice for 30 minutes. Centrifuge the cells at 3000 rpm for 5 minutes at 4°C, resuspend in 250 μl of ice-cold PBS, and maintain on ice until sorted by a Sony cell sorter SH8000. Process at least 20-fold the output of the previous sorting round and collect the top 0.1% anti-V5 antibody-positive yeast cells. After visual enrichment, plate out the sorting to characterize individual yeast display clones. For some enriched sorts, perform additional selection rounds using an antigen concentration of 100 nM.

[0484] For screening, induced yeast cells were diluted to 1 OD over 30 minutes at room temperature. 600They were blocked using 2% BSA-PBS. Subsequently, they were incubated with biotinylated human EGFR-Fc at a two-fold serial dilution starting at 100 nM in 2% BSA-PBS for 30 minutes at room temperature. Binding was detected with streptavidin-Alexa Fluor 647 diluted 1:1000 in 2% BSA-PBS over 30 minutes on ice. Finally, the cells were resuspended in 200 μl of ice-cold PBS and the fluorescence of the samples was recorded on a Guava EasyCyte flow cytometer. The percentage of antigen-binding cells was evaluated (Table 10), and the EGFR-specific binder L2B_EU1_1 was identified by the amino acid sequence of SEQ ID NO: 103. Residues different from the parental clone are shown in boldface.

[0485]

Table 10

[0486]

Chem.

[0487] 16.2 Mammalian display system for confirmation of antigen binding of EGFR-specific clones The sequences of wild-type CD81 LEL and the EGFR-specific clone L2BEU1_1 were cloned between the SfiI and SalI restriction sites of the pDisplay vector (Thermo Scientific). This display system enables the expression of the protein of interest anchored to the C-terminus between an N-terminal HA tag and a C-terminal c-myc tag. The construct was transfected into HEK293-6E cells (CNRC). Cells were harvested after 48 or 72 hours, blocked for 30 minutes on ice in 4% BSA-PBS, and stained for 30 minutes on ice with an antibody against CD81 (M38, Thermo Scientific) at 10 μg / ml in 4% BSA-PBS and an anti-c-myc antibody (A-14, sc789, Santa Cruz) at 10 μg / ml in 4% BSA-PBS. After incubation, their binding was detected on ice for 30 minutes with secondary reagents, an anti-mouse F(ab)2-Alexa Fluor 555 (Thermo Scientific) diluted 1:1000 in 4% BSA-PBS and an anti-rabbit (H+L) antibody conjugated to Alexa Fluor 488 (Thermo Scientific) diluted 1:100 in 4% BSA-PBS. The cells were then resuspended in PBS and maintained on ice until analyzed by Guava EasyCyte flow cytometry (Merck Millipore). Antigen reactivity was determined after incubation with biotinylated human EGFR-Fc at 300 nM and detection with streptavidin-Alexa Fluor 647 at 1:1000 in 4% BSA-PBS. Both wild-type CD81 LEL and the EGFR-binding mutant showed good levels of presentation on the mammalian cell surface as judged by reactivity with the anti-c-myc antibody (Table 11). Wild-type CD81 LEL reacted well with the anti-CD81 antibody, while the mutant showed no reactivity, presumably due to modification of the relevant epitope by library mutagenesis (Table 11).Antigen binding in the mammalian cell display format could be confirmed in the antigen-binding CD81 LEL mutants (Table 11).

[0488]

Table 11

[0489] 16.3 Characterization of Specificity, Species Cross-Reactivity, and Epitope Mapping of EGFR Binders HEK293-6E cells were transfected with pDisplay constructs encoding wild-type CD81 LEL and the anti-EGFR mutant CD81 LEL L2B_EU1_1. 1×10 5 cells were blocked in 2% BSA-PBS for 30 minutes on ice and then stained with biotinylated human EGFR-Fc, biotinylated human Her2 / neu-Fc, and biotinylated mouse EGFR-Fc, each at 500 nM, for 30 minutes on ice in 2% BSA-PBS. After centrifugation at 300 g for 5 minutes at 4°C, antigen binding was eluted with streptavidin-Alexa Fluor 647 (Thermo Scientific) diluted 1:1000 for 30 minutes on ice. The cells were then centrifuged at 300 g for 5 minutes at 4°C and resuspended in 200 μl of ice-cold PBS. The induced cultures were stained with an anti-c-myc antibody (A-14, sc789, Santa Cruz) at 10 μg / ml in 2% BSA-PBS and an anti-rabbit (H+L) antibody conjugated to Alexa Fluor 488 (Thermo Scientific) diluted 1:1000 in 2% BSA-PBS for 30 minutes on ice to measure the display. Fluorescence was determined by Guava EasyCyte flow cytometry. The anti-EGFR clones showed binding only to their cognate antigen and not to human Fc and Her2 / neu Fc proteins (Table 12). The EGFR-reactive clones were shown to be cross-reactive with mouse EGFR (Table 12).

[0490]

Table 12

[0491] In the epitope mapping experiment, a 300 nM antigen solution was incubated with cetuximab (Li et al., Cancer Cell 7 (4), 301 - 311 (2005). DOI: 10.1016 / j.ccr.2005.03.003), an anti - EGFR antibody with a 3 - fold molar excess of efficacy confirmed, and matuzumab (Schmiedel et al., Cancer Cell 13 (4), 365 - 373 (2008). doi: 10.1016 / j.ccr.2008.02.019.2005), or a commercially available human IgG kappa isotype antibody (Sigma - Aldrich), and then used for staining mutant CD81 LEL presenting cells. The binding of the antibody with confirmed efficacy to the antigen was allowed to proceed for 30 minutes at room temperature in 2% BSA - PBS, and then the solution was used to stain CD81 LEL mutant presenting cells. Detection was continued with streptavidin - Alexa Fluor 647 diluted 1:1000 in 2% BSA - PBS for 30 minutes on ice. The MFI values of the presenting cells were recorded. The data show that the binding site of the L2B_EU1_1 clone overlaps with the binding site in the cetuximab antibody (Table 13). Still, L2B_EU1_1 was able to bind to the antigen in the presence of an excess of matuzumab antibody.

[0492]

Table 13

[0493] 16.4 Re - randomization of the modified loop of the EGFR - binder based on CD81 LEL The purpose of this experiment is to re-randomize the mutated extensions at the residues of the EGFR-binding mutant and show that the binding of the clones depends on the amino acid residues in both extensions of the polypeptide chain that were randomized in the parental clone to introduce the antigen-binding site.

[0494] A mutant primer LIB2FWD (SEQ ID NO: 98) that randomizes the mutated residues in helix A and the AB loop in library L2BE1_A was used once, and a mutant primer LIB2REV2 (SEQ ID NO: 99) that randomizes the mutated residues in helix C in library L2BE1_B was used once to generate PCR fragments for recombination. The PCR fragments were generated using Q5 High-Fidelity Polymerase (New England Biolabs). For the recombinant fragments, primer LIB2FWD was used with primer AP2 (SEQ ID NO: 104) for library L2BE1_A, and primer LIB2REV2 (SEQ ID NO: 99) was used with primer AP1 (SEQ ID NO: 105) for library L2BE1_B.

[0495]

Chemical formula

[0496] Each recombinant fragment was transformed together with the BamHI / HindIII linearized recipient vector pYD1 with the cloned CD81 LEL_dellbamdelhind_bamhind sequence using chemical transformation into S. cerevisiae EBY100. For library construction, 35 μg of the linearized vector and 25 μg of the PCR fragment were used. The size of the library was 8.4×10 6 independent members for library L2BE1_A and 1.18×10 7It was a non-dependent member. Quality control of the library included induction of library members as stress temperature, subsequent measurement of proteins presented on the yeast surface by the N-terminal tag detected by the anti-Xpress tag antibody, and subsequent incubation with goat anti-mouse (Fab’)2-FITC (Sigma Aldrich) and determination of the correct reading frame of the presented protein by detection of the C-terminal V5 tag with the anti-V5-FITC antibody (Thermo Scientific). The percentage of antigen-binding cells in each library was determined after staining with human EGFR-Fc at 500 nM and detection with goat anti-human gamma chain-PE conjugate (Sigma Aldrich). The percentage of positive cells in the library is presented in Table 14 along with the values characteristic of the parental clone. Re-randomization of mutated residues in both target regions decreased the number of antigen-positive clones, suggesting that amino acid residues in both randomized extensions contribute to antigen binding.

[0497]

Table 14

[0498] Example 17: Binders Based on CD81 LEL to Human Placental Laminin 17.1 Selection of CD81 LEL Libraries L2 and L3 with Human Laminin Human laminin was purchased from Sigma-Aldrich. After dialysis of the antigen against 100-fold volume of PBS overnight at 4°C, the EZ-Link™ Sulfo-NHS-LC-LC-biotin reagent (Thermo Scientific) was used at a molar ratio of 3:1 for biotinylation. Incubation with the biotinylation reagent was continued for 1 hour at room temperature with shaking. Unbound biotin was removed by dialysis against 100-fold volume of PBS overnight at 4°C with stirring using a Snakeskin dialysis tube (Thermo Scientific) with a MWCO of 10,000 Da.

[0499] For selection, first, libraries 2A, 2B, 3A, and 3B were cultured in SD-CAA medium supplemented with penicillin-streptomycin with shaking overnight at 30 °C, and then the expression of the recombinant protein was induced by resuspending the yeast cells in SG / R-CAA medium with penicillin-streptomycin and incubating it with shaking overnight at 37 °C.

[0500] For MACS, 10 9 induced yeast cells were pelleted at 1000 g for 5 minutes and washed by resuspending the pellet in 50 ml of wash buffer (PBS, pH 7.2, 0.25% BSA, 2 mM EDTA), and then pelleted again at 1000 g for 5 minutes. The washed cells were resuspended in 5 ml of wash buffer containing 1 μM biotinylated antigen and incubated at room temperature for 30 minutes with gentle agitation. Antigen binding was quenched by adding 10 ml of ice-cold wash buffer, and the cells and antigen solution were pelleted at 4 °C at 1000 g for 5 minutes. The cells were resuspended in 5 ml of wash buffer + 25 μl of streptavidin microbeads (μMACS streptavidin kit, MACS Miltenyi) and incubated on ice for 10 minutes. Finally, 15 ml of wash buffer was added and the cell suspension was filtered through a 40 μm strainer.

[0501] The LS column (MACS Miltenyi) was placed in the separator and 3 ml of wash buffer was applied to precondition the column. Then, the cell solution was loaded in 7 ml batches. The dripping of the column was stopped once and the column was briefly removed from the magnet and repositioned on the magnet to allow the captured cells to perfuse in order to change the orientation of the beads within the column. Before loading the next 7 ml cell suspension, 1 ml of wash buffer was applied to the column. These steps were repeated until all the cells were loaded. Then, the column was washed with 3 ml of wash buffer, briefly removed from the magnet, and washed again with 3 ml of wash buffer. To elute the bound cells, the column was removed from the magnet and 5 ml of wash buffer was added. The cells were extruded from the column into a new tube by using the supplied plunger. The fraction of bound cells was pelleted at 2500 g for 10 minutes. The pellet was resuspended in 10 ml of SD-CAA medium, 10 μl of the eluted cells was diluted in 990 μl of SD-CAA, 100 μl was seeded onto an MDL plate and incubated at 30 °C for 3 days to evaluate the output of the MACS procedure, and the remaining cells were incubated overnight at 30 °C and 180 rpm.

[0502] In subsequent FACS selection rounds, the induced cell suspension was at 10 per 1 ml of 10% BSA-PBS 8Diluted to the cells of , and centrifuged at 3000 rpm for 5 minutes at 20 °C. To block the cells, the pellet was resuspended in 1 ml of 10% BSA-PBS and incubated at 20 °C for 30 minutes on a rotary shaker. The cells were centrifuged at 3000 rpm for 5 minutes at 20 °C, and resuspended in 250 μl of 10% BSA-PBS containing 1 μM biotinylated laminin. After incubation at 20 °C for 1 hour on a rotary shaker, the cells were centrifuged at 3000 rpm for 5 minutes at 4 °C. To wash the cells, the pellet was resuspended in 1 ml of ice-cold PBS and centrifuged at 3000 rpm for 5 minutes at 4 °C. Then, the cells were resuspended in 250 μl of 10% BSA-PBS with streptavidin-Alexa Fluor 647 (1:800) and anti-V5-FITC antibody (1:100), and incubated for 30 minutes on ice. The cells were centrifuged at 3000 rpm for 5 minutes at 4 °C, resuspended in 250 μl of ice-cold PBS, and maintained on ice until sorted by a Sony cell sorter. Processed at least 20-fold the output of the previous sorting round, and collected 0.1% of the top anti-V5 antibody-positive yeast cells. After visual enrichment, the sorting was plated out to characterize single yeast display clones. For several enriched sorts, an additional selection round was performed using 100 nM antigen. 17.2 Expression in mammalian cells The unique binder sequences amplified using the oligonucleotide sequences CD81hnhe1 (SEQ ID NO: 106) and LELp28_bste2 (SEQ ID NO: 107) and cloned between the NheI and BstEII sites of the mammalian expression vector pTT28 and the sequence of the construct were verified using the Sanger sequencing method.

[0503]

Chemical formula

[0504] Recombinant proteins were expressed in HEK293-6E following exactly the manufacturer's instructions. The supernatant of 25 ml of culture was recovered by centrifugation at 3500 rpm for 15 minutes at 4°C, buffered with PBS and 20 mM imidazole, filtered through a 0.45 μm filter, and then loaded onto a 1 ml His Excel column (GE Healthcare) equilibrated with PBS / 20 mM imidazole at pH 7.5. The column was then washed with the same buffer, and the his-tagged protein was eluted at pH 7.5 with a 20 - 500 mM gradient of imidazole in 5 column volumes of PBS. Fractions 4 - 6 were analyzed by SDS-PAGE for the presence of the eluted protein and then stained with Coomassie. Expression was confirmed for the following mutants: L2A_LU1 (SEQ ID NO: 108), L2A_LU1_1 (SEQ ID NO: 109), L2B_LU1 (SEQ ID NO: 110), L3B_LU1_2 (SEQ ID NO: 111). Residues different from the parental clone are represented in boldface type.

[0505]

Chem.

[0506] 17.3 Antigen binding in soluble form The eluted proteins were tested for their binding to their cognate antigens and BSA as a control antigen. Streptavidin activation plates (Immobilizer, NUNC) were coated with 5 μg / ml biotinylated laminin in PBS for 30 min at room temperature and blocked with 4% BSA-PBS for 1 h at room temperature. The eluted proteins were added to 2% BSA-PBS and allowed to bind for 1 h at room temperature. After three washing steps with PBS, binding was detected by a 1:3000 dilution of anti-pentaHis antibody-HRP conjugate (QIAgen) in 2% BSA-PBS for 30 min, and revealed by the addition of TMB (Sigma-Aldrich). The reaction was stopped by the addition of H2SO4 and absorbance was read at 450 / 620 nm (Table 15). Specific reactivity with the target protein could be confirmed in three laminin-specific clones: L2A_LU1, L2B_LU1 and L3B_LU1_2.

[0507]

Table 15

[0508] 17.4 Expression of laminin binders in mammalian cell display systems The sequences of the laminin-binding clones were cloned between the SfiI and SalI restriction sites of the pDisplay vector (Thermo Scientific). This display system enables the expression of the protein of interest anchored to the C-terminus between the N-terminal HA tag and the C-terminal c-myc tag. The constructs were transfected into HEK293-6E cells (CNRC). Cells were harvested after 48 or 72 hours, blocked on ice for 30 minutes in 4% BSA-PBS, and stained on ice for 30 minutes with anti-c-myc antibody (A-14, sc789, Santa Cruz) at 10 μg / ml in 4% BSA-PBS. After incubation, their binding was detected on ice for 30 minutes with a secondary anti-rabbit (H+L) antibody conjugated to Alexa Fluor 488 (Thermo Scientific A11034) diluted 1:1000 in 4% BSA-PBS. Antigen reactivity was determined after incubation with biotinylated human laminin at 500 nM and detection with streptavidin-Alexa Fluor 647 at 1:1000 in 4% BSA-PBS. The cells were then resuspended in PBS and maintained on ice until analyzed by Guava EasyCyte flow cytometry (Merck Millipore). Presentation on the mammalian cell surface could be detected for all tested clones (Table 16). Reactivity with the recombinant antigen could be established for nine clones (Table 16).

[0509]

Table 16

[0510] For L3A_LU1 (SEQ ID NO: 112), L2B_LU1_1 (SEQ ID NO: 113), 81L1 (SEQ ID NO: 110, the same as L2B_LU1), 81L_13 (SEQ ID NO: 114), 81L_17 (SEQ ID NO: 115), 81L_21 (SEQ ID NO: 116), the sequences of the discovered clones were determined. Residues different from the parental clones are represented in boldface type.

[0511]

Chem.

[0512] Example 18: Anti-laminin CD81 mutants expressed on the surface of EVs 18.1 Preparation of EVs HeLa cells were transduced with wild-type CD81 or CD81 with a modified LEL corresponding to the sequences of L2A_LU1 (SEQ ID NO: 12) or L3B_LU1_2 (SEQ ID NO: 15) cloned into the pBMN expression vector in a frame with eGFP. Extracellular vesicles (EVs) were prepared from the transduced HeLa cells cultured to 80% confluence in RPMI (10% FCS, 4 mM L-glutamine). Subsequently, the medium was changed and EV collection was performed over 72 - 96 hours in OptiPRO SFM. To remove cells and cell debris, the cell culture supernatant was centrifuged at 3000 g for 30 minutes. Larger particles were excluded by filtering the supernatant through a 0.45 μm cellulose acetate filter. Finally, the EVs were pelleted by ultracentrifugation at 120000 g for 90 minutes and resuspended in Live Cell Imaging Solution (Thermo Scientific). Recombinant EVs with CD81-eGFP fusions were verified using human CD9 capture beads for flow cytometry-based detection (ImmunoStep). 18.2 Competitive assay showing specific antigen binding of anti-laminin CD81 mutants expressed on the surface of EVs 5×10 9 of recombinant EVs with wild-type CD81-eGFP and the respective laminins targeting variants L2A_LU1 and L3B_LU1_2 were at 6×10 3It was incubated overnight with CD9+ capture beads. 5 μg / mL of biotinylated human placental laminin was added to the beads, and in some parallel samples, 15 μg / mL of unlabeled laminin was added for competition. To detect EV-bound laminin, NeutrAvidin-PE (1:800) was used. Background (BG) fluorescence was determined by staining with NeutrAvidin-PE alone. The fluorescence of eGFP was measured at 488 nm, and the PE-fluorescence was measured at 561 nm. Specific binding to the target antigen was shown for both laminin-binding CD81 mutants, while this was not observed for wild-type CD81. Furthermore, outcompeting by unlabeled laminin resulted in a decrease in the signal indicating specific binding (Table 17).

[0513]

Table 17

[0514] 18.3 Uptake of Laminin-Targeted EVs by the Model Hepatocellular Carcinoma Cell Line Huh7 2.4×10 10 、1.6×10 10 、and 0.8×10 10 of recombinant EVs with CD81-eGFP and its targeted variants were incubated with 0.4×10 6 Huh7 cells for 3 hours. The cells were trypsinized, neutralized, and resuspended in 100 μl of PBS. Flow cytometry was used to measure the uptake level of cells incorporating eGFP-positive EVs, and the MFI value of the major population (gated by FSC / SSC) was determined. Table 18 shows the biological triplicates obtained as a result of a single batch of EVs, normalized to the MFI max of CD81 wild-type eGFP, which shows a 2- to 3-fold increase in the uptake of laminin-targeted EVs.

[0515]

Table 18

[0516] In another experiment, the uptake of various batches of laminin-targeted EVs into the hepatocellular carcinoma cell line Huh7 was repeated. Three independent EV batches were tested in duplicates and normalized against MFI max Recombinant EVs with 2.4×10 10 CD81-eGFP or their respective targeted variants were incubated with 0.4×10 6 Huh7 cells for 3 hours. Cells were trypsinized, neutralized and resuspended in 120 μl of PBS. Flow cytometry was used to measure the uptake level of cells incorporating eGFP-positive EVs and the MFI values of the major population (gated by FSC / SSC) were determined. Values were normalized against MFI max. The mean and standard deviation of duplicates are presented in Table 19. EVs with overexpressed laminin-binding CD81 showed higher uptake into Huh7 cells.

[0517]

Table 19

[0518] Example 19: CD81-LEL-based binder against human Her2 / neu 19.1 Selection of CD81-LEL libraries L2A and L2B by human Her2 / neu Human Her2 / neu-Fc was purchased from SinoBiological. For biotinylation, the EZ-Link™ Sulfo-NHS-LC-LC-biotin reagent (Thermo Scientific) was used at a 3:1 molar ratio of biotin to protein. Incubation with the biotinylation reagent was continued for 1 hour at room temperature with shaking. Unbound biotin was removed by dialysis against 100-fold volume of PBS at 4°C overnight with stirring using a Snakeskin dialysis tube (Thermo Scientific) with a 10,000 Da MWCO, and aliquots of the labeled antigen were stored at -80°C until further use.

[0519] For selection, first, libraries 2A and 2B were cultured overnight with shaking at 30 °C in SD-CAA medium supplemented with penicillin-streptomycin, and then the expression of the recombinant protein was induced by resuspending the yeast cells in SG / R-CAA medium with penicillin-streptomycin, and it was incubated overnight with shaking at 37 °C.

[0520] For MACS, 4×10 9 induced yeast cells were pelleted by centrifugation at 1000 g for 5 minutes and blocked in 10% BSA-PBS for 30 minutes at room temperature in a rotating wheel. The cells were pelleted and resuspended in 10% BSA-PBS with 0.5 μM biotinylated antigen and incubated for 30 minutes at room temperature in a rotating wheel. Antigen binding was quenched by adding 10 volumes of ice-cold PBS, and the cells were pelleted at 4 °C by centrifugation at 1000 g for 5 minutes. The cells were resuspended in 5 ml of wash buffer + 200 μl of streptavidin microbeads (MACS Miltenyi) and incubated on ice for 15 minutes. Finally, 15 ml of MACS wash buffer (0.5% BSA, 2 mM EDTA in PBS; pH 7.4) was added. The cell suspension was filtered through a 40 μm cell strainer.

[0521] The LS column (MACS Miltenyi) was placed in the separator, and 3 ml of wash buffer was applied to precondition the column. Then, the cell solution was loaded in 7 ml batches. The dripping of the column was stopped once, and the column was briefly removed from the magnet and repositioned on the magnet to change the orientation of the beads in the column and allow the captured cells to perfuse. 1 ml of wash buffer was applied to the column before loading the next 7 ml cell suspension. These steps were repeated until all the cells were loaded. Then, the column was washed three times with 5 ml of wash buffer. To elute the bound cells, the column was removed from the magnet and 5 ml of wash buffer was added. The supplied plunger was used to extrude the cells from the column into a new tube. The fraction of bound cells was pelleted at 2500 g for 10 minutes. The pellet was resuspended in 10 ml of SD-CAA medium, 10 μl of the eluted cells was diluted in 990 μl of SD-CAA, 100 μl was seeded onto an MDL plate, and incubated at 30 °C for 3 days to evaluate the output of the MACS procedure, and the remaining cells were incubated overnight at 30 °C and 180 rpm.

[0522] In subsequent FACS selection rounds, the induced cell suspension was adjusted to 10 cells per 1 ml of 10% BSA-PBS 8Diluted to the cells, and centrifuged at 3000 rpm for 5 minutes at 20°C. To block the cells, the pellet was resuspended in 1 ml of 10% BSA-PBS and incubated on a rotator at 20°C for 30 minutes. The cells were centrifuged at 3000 rpm for 5 minutes at 20°C and resuspended in 150 μl of 10% BSA-PBS containing 0.5 μM biotinylated Her2 / neu-Fc. After a 1-hour incubation at room temperature on a rotator, antigen binding was quenched by adding 1 ml of ice-cold PBS. The cells were centrifuged at 3000 rpm for 5 minutes at 4°C and resuspended in 800 μl of 10% BSA-PBS with streptavidin-Alexa Fluor 647 (Thermo Fisher Scientific) (1:800) and anti-V5-FITC antibody (Thermo Scientific) (1:100), and incubated on ice for 30 minutes. The cells were centrifuged at 3000 rpm for 5 minutes at 4°C, resuspended in 250 μl of ice-cold PBS, and maintained on ice until sorted by a Sony SH8000 sorting instrument. Processed at least 20-fold the output of the previous sorting round, and collected 0.1% of the top anti-V5 antibody-positive yeast cells. Five sorting rounds were performed on the enriched pool, and single yeast clones were plated out for screening. Five identified sequences were cloned into a pDisplay expression vector, expressed in HEK293-6E cells, and tested for binding to human Her2 / neu. After 48 hours or 72 hours, the cells were harvested, blocked on ice in 4% BSA-PBS for 30 minutes, and stained with an anti-c-myc antibody (A-14, sc789, Santa Cruz) at 10 μg / ml in 4% BSA-PBS on ice for 30 minutes. After incubation, their binding was detected by a secondary anti-rabbit (H+L) antibody conjugated to Alexa Fluor 488 (Thermo Scientific A11034) diluted 1:1000 in 4% BSA-PBS on ice for 30 minutes.Antigen reactivity was determined after incubation with biotinylated human Her2 / neu / Fc in a two-fold dilution series starting from 300 nM and detection with streptavidin-Alexa Fluor 647 at 1:1000 in 4% BSA-PBS. Cells were then resuspended in PBS and maintained on ice until analyzed by Guava EasyCyte flow cytometry (Merck Millipore).

[0523] Clone 81_H2_11 (SEQ ID NO: 117) was able to specifically bind to the human Her2 / neu protein (Table 20). Residues that differ from the parental clone are shown in boldface.

[0524]

Chemical formula

[0525]

Table 20

[0526] 19.2 Characterization of the specificity of Her2 / neu binders based on CD81 LEL HEK293-6E cells were transfected with a pDisplay construct encoding wild-type CD81 LEL and an anti-Her2 / neu targeting CD81 LEL 81H2-11. 1×10 5The cells were blocked in 2% BSA-PBS on ice for 30 minutes and then stained with biotinylated human EGFR-Fc, biotinylated human Her2 / neu-Fc, and biotinylated mouse EGFR-Fc, each at 500 nM, on ice for 30 minutes in 2% BSA-PBS. After centrifugation at 300 g for 5 minutes at 4°C, antigen binding was eluted with streptavidin-Alexa Fluor 647 diluted 1:1000 on ice for 30 minutes. The cells were then centrifuged at 300 g for 5 minutes at 4°C and resuspended in 200 μl of ice-cold PBS. The induced cultures were stained on ice for 30 minutes with an anti-c-myc antibody (A-14, sc789, Santa Cruz) at 10 μg / ml in 2% BSA-PBS and an anti-rabbit (H+L) antibody conjugated to Alexa Fluor 488 (Thermo Scientific A11034) diluted 1:1000 in 2% BSA-PBS to measure presentation. Fluorescence was determined by Guava EasyCyte flow cytometry. The anti-Her2 / neu clone showed binding only to its cognate antigen (Table 21).

[0527]

Table 21

[0528] Example 20: Stabilization of Human CD9 LEL 20.1 Design and Construction of Stabilized CD9 LEL Mutants The Genbank entry with the full-length human CD9 sequence was identified, and BLAST comparisons were performed to define the boundaries of the LEL region. Using Swissmodel (Waterhouse et al., 2018) with CD81 LEL PDB:1iv5 as the proposed closest model, homology modeling of the CD9 LEL region was performed as defined in Seigneuret, 2006. The resulting structure can be aligned to the CD81 LEL crystal structure 1g8q with an RMSD of 0.413 Å. The sequence of CD9 LEL (SEQ ID NO: 118) was cloned between the NheI and BstEII cloning sites of the pTT28 vector (CNRC) using the oligonucleotides CD9hnhe1 (SEQ ID NO: 119) and CD9p28_bste2 (SEQ ID NO: 120).

[0529]

Chem.

[0530] To generate the stabilized variant CD9 LEL_20_28 having the amino acid sequence as in (SEQ ID NO: 125), the oligonucleotides CD9_L20C (SEQ ID NO: 121) and CD9_20Ca (SEQ ID NO: 122) and CD9_28C (SEQ ID NO: 123) and CD9_28Ca (SEQ ID NO: 124) were used to replace the amino acids at positions 20 and 28 with cysteine residues according to the manufacturer's instructions using the QuickChange Lightning Mutagenesis Kit (Agilent), and thus the mutagenesis reaction was carried out.

[0531]

Chem.

[0532] According to the manufacturer's instructions, CD9 LEL and CD9 LEL_20_28 were expressed in the ExpiCHO system according to the MaxTiter protocol and purified by one-step Ni-NTA affinity chromatography in the supernatants of 35 mg / L and 70 mg / L, respectively. The purified proteins were analyzed by SDS-PAGE and both were monomers. The thermal stability was determined using a differential scanning calorimeter (DSC). The melting point of the wild-type protein was determined to be 52.7 °C, and that of the stabilized variant 20 - 28 was 81.9 °C, indicating that stabilization was successfully achieved. 20.2 Yeast display library construction based on CD9 LEL for binder selection 20.2.1 Yeast display of wild-type CD9 LEL Between the BamHI and NotI cloning sites of the pYD1 vector, the sequence of CD9 LEL was cloned using oligonucleotides CD9YDbam1 (SEQ ID NO: 126) and CD9YDnot2 (SEQ ID NO: 127), and the construct was transformed into S. cerevisiae using chemical transformation.

[0533]

[0534] ​Transformants were selected on MDL plates, cultured at 30 °C in SD-CAA, and induced at 20 °C for 48 h and at 37 °C for 24 h. Post-measurement of the protein presented on the yeast surface by the N-terminal tag was performed by detection with a 1:2000 dilution of the anti-Xpress tag antibody (Thermo Scientific), followed by incubation with goat anti-mouse (Fab’)2-Alexa Fluor 555 (Thermo Scientific) at 1:1000. The correct reading frame of the presented protein was determined by detection of the C-terminal his tag with an anti-his-Alexa Fluor 488 antibody (QIAgen) at 1:200 and detection of the C-terminal V5 tag with an anti-V5-FITC antibody (Thermo Scientific) at 1:100, both in 2% BSA-PBS. Furthermore, yeast cells were first incubated with 10 μg / ml antibody in 2% BSA-PBS, and after detecting binding with 1:1000 goat anti-mouse (Fab’)2-Alexa Fluor 555 (Thermo Scientific) in 2% BSA-PBS, reactivity with the anti-CD9 specific antibody MEM-61 (Thermo Scientific) was determined. After resuspension in 200 μl of ice-cold PBS, the percentage of positive yeast cells was determined by a Guava EasyFlow Cytometer (Table 22).

[0535]

Table 22

[0536] 20.2.2 Site-directed mutagenesis of CD9 LEL for the design of binding clones Oligonucleotide CD9PFOREC1 (SEQ ID NO: 128) combined with oligonucleotide CD9NOTREC2 (SEQ ID NO: 129) was used to introduce mutations at residues 18-19, 21-25, and 27 of the CD9 LEL_20_28 sequence to generate the PCR recombinant fragment.

[0537]

Chemical formula

[0538] [wherein, N is any one of A, C, G, or T].

[0539]

Chemical formula

[0540] This fragment was linearized with the enzymes PfoI and NotI for use with the vector pYD1CD9 and transformation of S. cerevisiae EBY100. The resulting library of sizes of the non-dependent members was selected for binders with human EGFR-Fc by one round of MACS-based selection rounds and several rounds of FACS-based selection rounds until enrichment of the binding clones was achieved. The sequences of the binders were recloned into the mammalian pDisplay vector, and the resulting plasmid was used to transfect HEK293-6E cells. After 48 - 72 hours, the cells were harvested and stained with antigen and secondary reagents to detect specifically binding clones. 8 References References 1. Kalra, H.; Drummen, G.P.; Mathivanan, S. Focus on Extracellular Vesicles: Introducing the Next Small Big Thing. Int J Mol Sci 2016, 17 (2), 170, 10.3390 / ijms17020170. 2. Iraci, N.; Leonardi, T.; Gessler, F.; Vega, B.; Pluchino, S. Focus on Extracellular Vesicles: Physiological Role and Signalling Properties of Extracellular Membrane Vesicles. Int J Mol Sci 2016, 17 (2), 171, 10.3390 / ijms17020171. 3. Luan, X.; Sansanaphongpricha, K.; Myers, I.; Chen, H.; Yuan, H.; Sun, D. Engineering exosomes as refined biological nanoplatforms for drug delivery. Acta Pharmacol Sin 2017, 38 (6), 754 - 763, 10.1038 / aps.2017.12. 4. Tian, T.; Zhu, Y.L.; Zhou, Y.Y.; Liang, G.F.; Wang, Y.Y.; Hu, F.H.; Xiao, Z.D. Exosome uptake through clathrin - mediated endocytosis and macropinocytosis and mediating miR - 21 delivery. J Biol Chem 2014, 289 (32), 22258 - 67, 10.1074 / jbc.M114.588046. 5. El Andaloussi, S.; Lakhal, S.; Mager, I.; Wood, M.J. Exosomes for targeted siRNA delivery across biological barriers. Adv Drug Deliv Rev 2013, 65 (3), 391 - 7, 10.1016 / j.addr.2012.08.008. 6. Rubinstein, E.; Le Naour, F.; Lagaudriere-Gesbert, C.; Billard, M.; Conjeaud, H.; Boucheix, C. CD9, CD63, CD81, and CD82 are components of a surface tetraspan network connected to HLA-DR and VLA integrins. Eur J Immunol 1996, 26 (11), 2657-65, 10.1002 / eji.1830261117. 7. Levy, S.; Shoham, T. Protein-protein interactions in the tetraspanin web. Physiology (Bethesda) 2005, 20, 218-24, 10.1152 / physiol.00015.2005. 8. Morelli, A.E.; Larregina, A.T.; Shufesky, W.J.; Sullivan, M.L.; Stolz, D.B.; Papworth, G.D.; Zahorchak, A.F.; Logar, A.J.; Wang, Z.; Watkins, S.C.; Falo, L.D., Jr.; Thomson, A.W. Endocytosis, intracellular sorting, and processing of exosomes by dendritic cells. Blood 2004, 104 (10), 3257-66, 10.1182 / blood-2004-03-0824. 9. Svensson, K.J.; Christianson, H.C.; Wittrup, A.; Bourseau-Guilmain, E.; Lindqvist, E.; Svensson, L.M.; Morgelin, M.; Belting, M. Exosome uptake depends on ERK1 / 2-heat shock protein 27 signaling and lipid Raft-mediated endocytosis negatively regulated by caveolin-1. J Biol Chem 2013, 288 (24), 17713-24, 10.1074 / jbc.M112.445403. 10. Berditchevski, F.; Odintsova, E. Tetraspanins as regulators of protein trafficking. Traffic 2007, 8 (2), 89-96, 10.1111 / j.1600-0854.2006.00515.x. 11. Escola, J.M.; Kleijmeer, M.J.; Stoorvogel, W.; Griffith, J.M.; Yoshie, O.; Geuze, H.J. Selective enrichment of tetraspan proteins on the internal vesicles of multivesicular endosomes and on exosomes secreted by human B-lymphocytes. J Biol Chem 1998, 273 (32), 20121-7, 12. Kitadokoro, K.; Galli, G.; Petracca, R.; Falugi, F.; Grandi, G.; Bolognesi, M. Crystallization and preliminary crystallographic studies on the large extracellular domain of human CD81, a tetraspanin receptor for hepatitis C virus. Acta Crystallogr D Biol Crystallogr 2001, 57 (Pt 1), 156-8, 13. Kitadokoro, K.; Ponassi, M.; Galli, G.; Petracca, R.; Falugi, F.; Grandi, G.; Bolognesi, M. Subunit association and conformational flexibility in the head subdomain of human CD81 large extracellular loop. Biol Chem 2002, 383 (9), 1447-52, 10.1515 / BC.2002.164. 14. Kitadokoro, K.; Bordo, D.; Galli, G.; Petracca, R.; Falugi, F.; Abrignani, S.; Grandi, G.; Bolognesi, M. CD81 extracellular domain 3D structure: insight into the tetraspanin superfamily structural motifs. EMBO J 2001, 20 (1-2), 12-8, 10.1093 / emboj / 20.1.12. 15. Higginbottom, A.; Quinn, E.R.; Kuo, C.C.; Flint, M.; Wilson, L.H.; Bianchi, E.; Nicosia, A.; Monk, P.N.; McKeating, J.A.; Levy, S. Identification of amino acid residues in CD81 critical for interaction with hepatitis C virus envelope glycoprotein E2. J Virol 2000, 74 (8), 3642-9, 16. Imai, T.; Yoshie, O. C33 antigen and M38 antigen recognized by monoclonal antibodies inhibitory to syncytium formation by human T cell leukemia virus type 1 are both members of the transmembrane 4 superfamily and associate with each other and with CD4 or CD8 in T cells. J Immunol 1993, 151 (11), 6470-81, 17. Seigneuret, M. Complete predicted three-dimensional structure of the facilitator transmembrane protein and hepatitis C virus receptor CD81: conserved and variable structural domains in the tetraspanin superfamily. Biophys J 2006, 90 (1), 212-27, 10.1529 / biophysj.105.069666. 18. Rajesh, S.; Sridhar, P.; Tews, B.A.; Feneant, L.; Cocquerel, L.; Ward, D.G.; Berditchevski, F.; Overduin, M. Structural basis of ligand interactions of the large extracellular domain of tetraspanin CD81. J Virol 2012, 86 (18), 9606-16, 10.1128 / JVI.00559-12. 19. Seigneuret, M.; Delaguillaumie, A.; Lagaudriere-Gesbert, C.; Conjeaud, H. Structure of the tetraspanin main extracellular domain. A partially conserved fold with a structurally variable domain insertion. J Biol Chem 2001, 276 (43), 40055-64, 10.1074 / jbc.M105557200. 20. Homsi, Y.; Schloetel, J.G.; Scheffer, K.D.; Schmidt, T.H.; Destainville, N.; Florin, L.; Lang, T. The extracellular delta-domain is essential for the formation of CD81 tetraspanin webs. Biophys J 2014, 107 (1), 100-13, 10.1016 / j.bpj.2014.05.028. 21. Schmidt, T.H.; Homsi, Y.; Lang, T. Oligomerization of the Tetraspanin CD81 via the Flexibility of Its delta-Loop. Biophys J 2016, 110 (11), 2463-74, 10.1016 / j.bpj.2016.05.003. 22. Homsi, Y.; Lang, T. The specificity of homomeric clustering of CD81 is mediated by its delta-loop. FEBS Open Bio 2017, 7 (2), 274-283, 10.1002 / 2211-5463.12187. 23. Michel Seigneuret: Complete Predicted Three-Dimensional Structure of the Facilitator Transmembrane Protein and Hepatitis C Virus Receptor CD81: Conserved and Variable Structural Domains in the Tetraspanin Superfamily. Biophys J. 2006 Jan 1; 90(1): 212-227. doi: 10.1529 / biophysj.105.069666 24. Andrew Waterhouse, Martino Bertoni, Stefan Bienert, Gabriel Studer, Gerardo Tauriello, Rafal Gumienny, Florian T Heer, Tjaart A P de Beer, Christine Rempfer, Lorenza Bordoli, Rosalba Lepore, Torsten Schwede: SWISS-MODEL: homology modelling of protein structures and complexes. Nucleic Acids Res. 2018 Jul 2; 46(Web Server issue): W296-W303. Published online 2018 May 21. doi: 10.1093 / nar / gky427 In one aspect, the present invention may be as follows. [Aspect 1] A method for producing a protein comprising a target-specific extracellular domain (TED) of an extracellular vesicle (EV) surface protein, comprising A polynucleotide comprising a nucleotide sequence encoding the extracellular domain (ED) of an EV surface protein is modified by a mutagenesis method within at least one modified region in the ED amino acid sequence having a length of 3 to 20 consecutive amino acids adjacent to the N-terminus and C-terminus in the region of the wild-type ED sequence to incorporate a target binding site into the ED, thereby producing a repertoire of polynucleotides encoding diverse TEDs, each containing a different target binding site, and selecting a TED that specifically recognizes a predetermined target, and producing a protein containing the selected TED, The above method comprising. [Aspect 2] The repertoire of polynucleotides is presenting diverse TEDs on the outer surface and preferably contained in a gene package employing a display system selected from the group consisting of yeast, phage, bacteria, ribosome, mRNA, or mammalian cell display, the method according to aspect 1. [Aspect 3] The modified region has a lower target binding affinity when isolated from the TED, the method according to aspect 1 or 2. [Aspect 4] The target binding site is within at least one additional modified region separated by at least 2 amino acids or contains at least one additional binding region within the wild-type ED sequence, the method according to any one of aspects 1 to 3. [Aspect 5] T...

Claims

1. A method for producing a protein comprising a target-specific extracellular domain (TED) of an extracellular vesicle (EV) surface protein, comprising: Modifying a polynucleotide comprising a nucleotide sequence encoding an extracellular domain (ED) of an EV surface protein by a mutagenesis method to manipulate a target binding site containing at least two modified regions each having a length of 3 to 20 consecutive amino acids adjacent to the N-terminus and C-terminus in the region of the wild-type ED sequence, thereby producing a repertoire of polynucleotides encoding diverse TEDs each containing a different target binding site; and Selecting a TED that specifically recognizes a predetermined target; and Producing a protein containing the selected TED, wherein the EV surface protein is a tetraspanin, the ED is that of CD81 and the amino acid sequence is at least one selected from the group consisting of the following (i) to (iii): (i) positions 134 and 144; (ii) positions 130 and 146; and (iii) positions 135 and 168; and is modified to introduce cysteines that enable the formation of one or more disulfide bonds not naturally present in the wild-type ED sequence, and the numbering is that of human CD81 specified as SEQ ID NO: 87, the above method.

2. The method according to claim 1, wherein the repertoire of polynucleotides presents diverse TEDs on the outer surface.

3. The method according to claim 2, wherein the gene package employs a display system selected from the group consisting of yeast, phage, bacteria, ribosome, mRNA, or mammalian cell display.

4. The method according to any one of claims 1 to 3, wherein the target binding site comprises at least one additional binding region within at least two additional modified regions that are at least 2 amino acids apart, or at least one additional binding region within the wild-type ED sequence.

5. The method according to any one of claims 1 to 4, wherein the TED contains at least 70% sequence identity with the wild-type ED.

6. The tetraspanin is CD81 containing the amino acid sequence specified as SEQ ID NO: 87, according to any one of claims 1 to 5.

7. The wild-type ED is that of CD81, and the method according to any one of claims 1 to 6, wherein the ED comprises any one of the amino acid sequences specified as SEQ ID NO: 130 or SEQ ID NO:

131.

8. The method according to any one of claims 1 to 7, wherein the protein comprises a loop structure in the ED amino acid sequence, and the loop structure is stabilized by one or more cysteine(s) at one or more position(s) enabling the formation of one or more disulfide bonds.

9. The method according to any one of claims 1 to 8, wherein the modified region is located within the loop region of ED.

10. The method according to any one of claims 1 to 9, wherein the ED is that of CD81, the modified region is located within the range of positions 160 and 172, and the numbering is that of human CD81 specified as SEQ ID NO:

87.

11. The method according to claim 10, wherein TED comprises at least one additional binding region located between positions 132 and 141 or between positions 180 and 189, and the numbering is that of human CD81 specified as SEQ ID NO:

87.

12. The method according to any one of claims 1 to 11, wherein the target is selected from the group consisting of a cell target and a cell-free target.

13. The method according to claim 12, wherein the cell target is one of a mitogen receptor, a cytokine receptor, an asialoglycoprotein receptor, a membrane transporter, a lipoprotein, a lipopolysaccharide, a glycoprotein, or a proteoglycan.

14. The method according to claim 12, wherein the cell-free target is one of a cytokine, an artificial protein, or an artificial surface structure.

15. The method according to any one of claims 1 to 14, wherein the protein comprising TED is a target-specific EV surface protein (TSP) comprising the said TED and at least one transmembrane domain.

16. The method according to claim 15, wherein the transmembrane domain comprises at least 70% sequence identity with a transmembrane domain originating from a mammalian EV surface protein.

17. The method according to claim 15 or 16, wherein the transmembrane domain is that of CD81, and the transmembrane domain comprises any one of the amino acid sequences specified as SEQ ID NO: 147, SEQ ID NO: 148, SEQ ID NO: 149, or SEQ ID NO:

150.

18. The method according to any one of claims 15 to 17, wherein both the ED and the transmembrane domain originate from the same EV surface protein.

19. The method according to claim 18, wherein the EV surface protein is CD81 comprising the amino acid sequence specified as SEQ ID NO:

87.

20. a) providing a polynucleotide encoding a protein comprising a TED obtainable by the method according to any one of claims 1 to 19; b) introducing said polynucleotide into a source cell or a source cell mixture; c) culturing said cell(s) under conditions for producing extracellular vesicles; d) isolating a fraction comprising target-specific extracellular vesicles (TEV) comprising a target binding site for the TED; and e) producing a preparation of TEV contained in said fraction A method for producing a TEV preparation by the above.

21. The method according to claim 20, wherein in method step a), the polynucleotide encodes a protein which is a target-specific EV surface protein (TSP) comprising said TED and at least one transmembrane domain, and in method step d), the fraction comprises TEV presenting a target binding site on the outer surface of the vesicle membrane.

22. Further comprising the step of loading an intravesicular cargo into the TEV, the cargo comprising any one or more of a peptide, polypeptide, protein domain, protein, lipid, gene, mRNA, miRNA, nucleic acid such as an RNAi mediator molecule, particularly a locked nucleic acid, or a plasmid such as phosphorothioate, DNA, DNA fragment, minicircle DNA, a drug such as a small molecule, particularly a chemotherapeutic drug or a senolytic drug.

23. The method according to any one of claims 20 to 22, wherein the source cell or the source cell mixture originates from a eukaryotic or prokaryotic source.

24. The method according to claim 23, wherein the source cell or the source cell mixture is from a body tissue, body fluid, or cell culture.

25. The method according to any one of claims 20 to 24, wherein the source cell or the source cell mixture is obtained from a subject and the TEV preparation is formulated for autologous use.

26. A method for producing an autologous TEV preparation by the method according to claim 25, wherein the source cell or the source cell mixture is obtained from a subject and the TEV preparation is formulated for autologous use in the same subject.

27. A TEV preparation obtainable by the method according to any one of claims 20 to 26.

28. A protein comprising a target-specific extracellular domain (TED) of a cell surface protein of extracellular vesicles (EVs), obtainable by the method according to any one of claims 1 to 19.

Images