WO2024107831A1 - Use of extracellular vesicles as a molecular screening platform - Google Patents

Abstract

The present invention provides a library composition comprising a plurality of extracellular vesicles (EVs). The EVs comprise a plurality of surface-expressed peptides of interest and a plurality of nucleic acids encoding the surface expressed peptides of interest. Methods of making the library and methods of using the library are also provided.

Description

USE OF EXTRACELLULAR VESICLES AS A MOLECULAR SCREENING PLATFORM

CROSS-REFERENCE TO RELATED APPLICATIONS

The present application claims priority to U.S. Provisional Patent Application No. 63/425,592, filed November 15, 2022, the entire contents of which are hereby incorporated by reference.

SEQUENCE LISTING

A Sequence Listing accompanies this application and is submitted as an ASCII text file of the sequence listing named “203128_00099_Sequence_Listing.xml” which is 11 l,457bytes in size and was created on November 14, 2023. The sequence listing is electronically submitted via Patent Center with the application and is incorporated herein by reference in its entirety.

BACKGROUND

The natural capacity of extracellular vesicles (EVs) to mediate intercellular communication via molecular transfer provides tremendous potential as a drug delivery vehicle. Despite a few cell sources generating EVs with specificity, targeting a specific tissue or cells remains a challenge for discovery. An EV engineering platform that can be readily modified for various cellular and tissue targets is essential for rapid development of EV-based therapeutics with clinical applications via screening.

Accordingly, there is a remaining need in the art for methods of molecular screening using EVs and development of their use as a therapeutic delivery tool.

SUMMARY

The present disclosure relates to a method of employing extracellular vesicles (EVs) as a molecular screening platform for EV-mediated targeted gene delivery.

In a first aspect, the disclosure provides a library composition comprising a plurality of extracellular vesicles (EVs) comprising a plurality of surface-expressed peptides of interest and a plurality of nucleic acids encoding the surface expressed peptides of interest.

In some aspects, the plurality of EVs comprises greater than 10 distinct peptides of interest, and in some aspects the plurality of EVs comprises 10, 100, 1000, 10⁴, 10⁵, or 10⁶ distinct peptides of interest. In some aspects, the plurality of surface-expressed peptides of interest comprises an affinity reagent selected from the group consisting of a monobody, an scFv, an antibody, an antibody mimetic, a nanobody, and a ligand; in some aspects, at least one of the surface- expressed peptides of interest is a monobody, and in some aspects, the monobody is E626. In some aspects, at least one of the surface-expressed peptides of interest is a ligand, and in some aspects the ligand is p88.

In some aspects, the surface-expressed peptide of interest comprises a peptide linked to lactadherin C1C2, and in some aspects, the linker is at least 10 amino acids long.

In some aspects, the plurality of EVs further comprise an imaging agent.

In some aspects, the nucleic acid encoding the surface-expressed peptide of interest is a plasmid or mini circle, and in some aspects, the plasmid is greater than 3 kB.

In another aspect, the disclosure provides a method of gene delivery to a cell or tissue comprising contacting the cell or tissue with any one of the compositions described herein.

In another aspect, the disclosure provides a method of screening for affinity reagents, the method comprising: (a) contacting a plurality of cells with the library composition comprising any of the EVs described herein; harvesting the portion of the library composition bound to the cells in step (a); (c) extracting DNA from the portion of the library composition harvested in step (b); and (d) amplifying the nucleic acid encoding the surface-expressed peptide from the prepared DNA of step (c) to produce a first screened nucleic acid pool.

In some aspects, the method further comprises: (e) generating a first screen library composition from the first screened nucleic acid pool of step (d) by (i) using the first screened nucleic acid pool to transfect cells and (ii) harvesting EVs from the cells; and (f) repeating steps (a)-(d) using the first screen library composition to produce a second screened nucleic acid pool. In some aspects, the method further comprises repeating steps (a) through (d) after step (f), and in some aspects steps (a) through (f) are repeated at least five times.

In some aspects, the method further comprises sequencing the screened nucleic acid pool. In some aspects, the method further comprises generating EVs using the screened nucleic acid pool. In some aspects, the EVs are used for delivering an agent to a cell, and in some aspects, the agent is selected from the group consisting of small molecule, chemotherapeutic, peptide, polypeptide, and enzyme. In some aspects, the cell is selected from the group consisting of pancreatic cell, kidney cell, spleen cell, liver cell, brain cell, and tumor cell. In some aspects, the cell is contacted in vitro, and in other aspects, the cell is contacted in vivo. In some aspects in which the cell is contacted in vivo, the plasmid DNA is recovered from a tissue, and in some aspects, the tissue is selected from the group comprising liver, heart, lung, brain, kidney, pancreas, and spleen.

In another aspect, the disclosure provides a method of making the library composition comprising the EVs described herein, the method comprising: (a) amplifying the library backbone from a DNA template comprising a polynucleotide encoding lactadherin C1C2; (b) amplifying a polynucleotide encoding the peptide of interest; (c) joining the products of step (a) and (b) to create the library construct; (d) introducing the library construct of step (c) into bacterial cells; (e) harvesting the library construct from the bacterial cells; (f) introducing the library construct into eukaryotic cells; and (g) harvesting the EVs from the eukaryotic cells, wherein the harvested EVs are the library composition. In some aspects, the bacterial cells are E. coll cells. In some aspects, step (d) of the method further comprises selecting for bacterial cells comprising a selectable marker, and in some aspects, the selectable marker is ampicillin. In some aspects, step (c) of the method is performed via Seamless Cloning Ligation Cell Extract (SLiCE). In some aspects, the eukaryotic cells are HEK293T cells. In another aspect, the disclosure provides a monobody expressed on the surface of the EVs harvested in step (g) of any one of the methods of making the library composition comprising the EVs described herein.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows a schematic of the monobody-library EV generation by transient transfection of monobody-EV display constructs into HEK293T cells. Each EV will contain unique monobody-coding DNA and display its protein on its surface.

FIG. 2 shows the schematic of EV-display technology. Fusing Fn3HP -based monobody- coding DNA with the lactadherin signal peptide and C1C2 domain (lactadherin C1C2) allows the monobody to be displayed on the EV surface.

FIG. 3 shows in vitro uptake demonstrating the genotype-phenotype link of EV- molecular display. EGFR positive (A431) cells were treated with EVs prepared from the mixed ratio of EGFR-targeting (E626) and non-targeting (RDG) monobodies. qPCR analysis was performed on recovered plasmid DNA following 10-, 30-, and 60-min treatment. RIO indicates EVs prepared from 100% of RDG DNA, R9E10 from RDG90%EGFR10%, R5E5 from RDG50%EGFR50%, R1E9 from RDG10%EGFR90% and E10 from EGFR100%. Data is one of two independent experiments with similar pattern.

FIG. 4 shows library injection and successful recovery of plasmid DNA from various organs. The table shows the plasmid DNA copy numbers determined by qPCR of isolated plasmid DNA retrieved from each organ following 5 xlO⁸ monobody library EV injection and 1-hr circulation, using primer/probe against ampicillin gene. The bottom photo is an agarose gel showing the monobody fragments recovered from each organ using the re-cloning primer set. (control-plasmid DNA as a template).

FIG. 5 shows a schematic illustration of the Monobody -EV library screening strategy. Overview of EV-based monobody screening processes, involving monobody library pDNA generation, pDNA transfection to EV donor cells (HEK293T), EV isolation from the cell culture media, EV treatment or administration, DNA extraction from cell or organ, monobody amplification and re-cloning to generate enriched monobody library pool. This process can be repeated to enrich for targeting monobodies (or other affinity reagent specific for the target recipient cell or organ).

FIG. 6 shows EV characterization showing the successful EV isolation, pDNA loading and EV-surface display with monobody protein. (A) Nanoparticle tracking analysis (NTA) showing the library EVs' representative size distribution. (B) Size distribution plot showing the peak sizes of library EVs from 26 library EV preparations. (C) a distribution plot showing plasmid copy number per EV collected from 26 library EV preparations. The numbers were generated based on NTA and qPCR analysis. (D) Representative images of library EVs stained with EV markers (CD63+CD81-cyan, CD9-yellow) and with monobodies (HA-Pink), with molecular counts taken by Super-resolution microscopy.

FIG. 7 shows enrichment of the target sequence on in vitro EV-Monobody library Screening. (A) Reduction of no-binder Monobody (RDG) sequence and (B) enrichment of targeting Monobody (E626) following five rounds of panning using A431 cells. (C) Validation of the fold change of RDG and E626 in A431 cells by qPCR. (D) Enrichment growth rate of each variant calculated using: Final count = initial count * Exp (enrichment growth rate * time) (E) Phylogenetic tree showing the relatedness between positive control (E626), negative control (RDG), and several novel lead monobody variants (or “clones”).

FIG. 8 shows evaluation of selected clones using Bioluminescence imaging (BLI). A431 and MCF7 cells were treated with each monobody displayed EV co-labeled with NanoLuc. The total photon flu (p/s) from EVs bound to the cells was quantified using IVS. The value represents the graph's means ± SD (n=3). Two-way ANOVA was used to assess the effect of the time course in the group. In all figures, significance against 0 min is expressed as follows: * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0001, if not otherwise specified. NP (nonpeptide, non-binder negative control). FIG. 9 shows pancreas-enriched monobody library EVs show higher accumulation in the pancreas. Time course biodistribution of EV library showing the accumulation of pancreas- enriched EV library in the pancreas in vivo. (A) Time course in vivo imaging of EV library following iv injection of the original EV library (PO, left) and the pancreas-enriched EV library (pan-P5, right) (EVs labeled with NanoLuc) in 30 min sequential imaging (interval: 2min) (B) Ex vivo Bioluminescence imaging (BLI) of organs 30 min after the injection.

DETAILED DESCRIPTION

The present invention provides library compositions and methods of making the library compositions. Methods of screening for affinity reagents using the library compositions are also provided. The present inventors have previously described in vitro and in vivo successes in target-specific delivery of extracellular vesicles (EVs). Here, the present inventors disclose compositions and methods of using the target- specificity of previous work using EVs to develop target specific molecular screening tools and methods. Extracellular vesicles (EV) refer to naturally released, cell-derived vesicles that mediate intracellular communication, in part, by transferring genetic information. Cells secrete heterogeneous populations of lipidbilayer membranous nano-sized particles such as exosomes, microvesicles (MVs) and apoptotic bodies whose composition may vary depending on the cell of origin, physiological and pathological condition of the cells or surrounding tissues. While the sizes of these particles largely overlap with each other, the differences are consistent with the distinctive biogenesis of exosomes and MVs. Exosomes (40-150 nm in diameter) derive from the inward budding of endosomal multivesicular bodies (MVBs) and are released from the cell upon MVB fusion with the cell membrane. MVs (50-1,000 nm in diameter) are generally larger vesicles and are the product of direct budding from the plasma membrane. EVs are released from many different cell types into various body fluids, including milk, saliva, sweat and plasma, to mediate molecular transfer to other cell types in both physiological and pathological conditions. The inventors demonstrate that EV can be engineered to express peptides on the surface of the EVs and that the plasmids which encode the surface-expressed peptides can be recovered from cells with which the EVs interact. This finding allows for the development of library compositions comprising EVs comprising surface expressed peptides and plasmids encoding those peptides. These libraries can then be used in methods of screening to identify EVs engineered to express proteins on their surface that target the EVs to various cell types, both in vivo and in vitro. The targeting peptides can then be identified by sequencing the DNA in the EVs. These targeting EVs can be used for delivery of therapeutic compositions. Library Compositions:

In a first aspect, the present invention provides library compositions comprising a plurality of extracellular vesicles (EVs). The EVs comprise a plurality of surface-expressed peptides (which may also be polypeptides as used herein) of interest and a plurality of nucleic acids encoding the surface expressed peptides of interest.

The “peptides of interest” includes peptides that may bind to or target the EVs to a cell, tissue, organ or site of disease in a subject (or multiple cells, tissues, organs, or sites of disease). The peptides of interest may be peptides known to bind to cells or to other proteins, carbohydrates, or lipids or may be a portion of a protein which may bind to a protein, carbohydrate or lipid and may be included in the library to determine if the peptide is capable of mediating binding to a particular cell, tissue, organ or site. The peptides of interest in the library are engineered to be surface expressed on the EVs and suitably are detectable on the surface of the EVs in the library. The plurality of EVs comprises greater than 10 distinct peptides of interest. “Distinct” in this specific context refers to peptides of interest possessing unique sequences, functions, conformations, or other characteristics which may distinguish them from other peptides. The plurality of EVs may comprise 10, 100, 1000, 10⁴, 10⁵, or 10⁶ or more distinct peptides of interest.

The peptides of interest may include but are not limited to, affinity reagents such as a monobody, an scFv, an antibody, an antibody mimetic, a nanobody, and a ligand. As used herein, the term “affinity reagent” refers to a molecular tool that has the ability to specifically recognize and bind other cell constituents such as proteins, carbohydrates and lipids. Suitably, the peptides of interest have affinity for a cell surface such that the presence of the peptide of interest on the surface of the EV allows the EV to bind to, interact with or even fuse with (target) the cell. As used herein, the term “monobody” refers to a class of small non-antibody scaffolds derived from the thermodynamically stable human fibronectin type III (FN3) domain. Additional hydrophilic mutations can be incorporated into this scaffold (Fn3HP) to improve processing and in vivo biodistribution. Site-wise modifications of three solvent exposed loops (akin to antibody CDRs) on the hydrophilic monobody scaffold has enabled strong, specific binding interactions against a diverse panel of clinically relevant targets including to EGFR (for example, E626). As used herein, the term “antibody” refers to a protein that comprises at least one antigen-binding domain from an immunoglobulin molecule. Suitable antibody molecules include, without limitation, whole antibodies (e.g., IgG, IgA, IgE, IgM, IgD), monoclonal antibodies, polyclonal antibodies, chimeric antibodies, humanized antibodies, and antibody fragments, including single chain variable fragments (scFv), single domain antibodies, and antigen-binding fragments. Any format of antibody or antibody fragment may be used with the present invention. As used herein, the term “antibody mimetic” refers to a molecule that can bind to antigens similar to an antibody; however, antibody mimetics are not generated by the immune system and have no structural relation to the antibodies. As used herein, the term “nanobody” refers to an antibody fragment consisting of a single monomeric variable antibody domain A nanobody can also be referred to as single-domain antibody (sdAb). As used herein, the term “ligand” refers to a peptide or protein that binds to a receiving molecule or protein (otherwise known as a receptor). An exemplary peptide that can serve as a targeting ligand in this invention is the p88 peptide which is known to bind (FXYD2)ya on pancreatic P-cells. In some embodiments, at least one of the surface-expressed peptides of interest is a monobody, and in some embodiments, the monobody is E626. Using an EV-based monobody display screening strategy, we were able to display HEK293T-derived monobody proteins on the surfaces of EV library EVs that resulted in the enrichment of cell-type specific monobody sequences that possessed high binding affinities. In some embodiments, at least one of the surface-expressed peptides of interest is a ligand, and in some embodiments, the ligand is p88.

The peptide of interest may be linked to lactadherin C1C2 (SEQ ID NO: 41) to allow for surface expression. As used herein, the term “lactadherin” refers to the phosphatidylserine (PS) binding protein milk fat globule-EFG factor 8 (MFG-E8) and belongs to the secreted extracellular matrix protein family. Fusing targeting moieties to the C1C2 domain of lactadherin comprising the Cl and C2 domains (SEQ ID NOs: 42 and 43, respectively) enables versatile EV surface display since it can be used to decorate the EV surface when expressed from producer cells, or when added to purified EVs. The peptide of interest may be linked to the lactadherin via a peptide linker. The linker (that is, the sequence or composition linking the peptide of interest to lactadherin C1C2) may include any length of amino acids including but not limited to 1, 2, 3, 4, 5, 10, 15, 20, 25, 30, or 40 amino acids. The linker may include any amino acid sequence or amino acid motifs. For example, the linker may include a glycineserine linker including a number of glycine amino acids followed by a serine e.g. G4S repeats, or Pro, Ala and Ser (PAS) repeats. For example, the linker may include but not be limited to 1, 2, 3 (SEQ ID NO: 26), 4, 5, 10, or 15 G₄S or PAS repeats.

In some embodiments, the plurality of EVs further comprise an imaging agent. As used herein, the term “imaging agent” can refer to the agent used in any method of biological imaging which includes, but is not limited to, enzymatic agents, luciferase, radio labels, and magnetic labels. Imaging agents can be expressed on a separate plasmid or as part of a fusion protein with the peptide of interest. For example, GFP, mCherry or luciferase may be linked to the affinity peptide and lactadherin C1C2. Several of these luminescent or fluorescent proteins are now available as tags that may be expressed as small partial proteins to provide a signal only when paired with the remainder of the protein may also be used herein. The imaging agent may also be an epitope tag, such as a FLAG-tag, His-tag or other tag that may interact with a second protein to produce a detectable signal.

The nucleic acid encoding the surface-expressed peptide of interest may be a plasmid or minicircle. In some embodiments, the nucleic acid is greater than 3 kB, but may be smaller than lOkb. As used herein, the term “plasmid” refers to a small circular DNA molecule that can replicate independently from chromosomal DNA. In nature, plasmids are commonly found in bacteria, and artificial plasmids are widely used as vectors in molecular cloning. As used herein, the term “minicircle” refers to small (less than or approximately 4 kB) replicons or plasmids. Replicons are regions of DNA that are independently replicated from a single origin of replication. The nucleic acids encoding the peptides of interest may also include a promoter operably linked to the nucleic acid encoding the peptide of interest. The promoter should allow for expression of the peptide of interest in the cells in which the user plans to prepare the EVs for the library composition. In one embodiment the promoter is a CMV promoter, but those of skill in the art are aware of the vast number of promoters that may be used. The nucleic acid may also further encode the lactadherin C1C2 as well as epitope tags, imaging labels, or a selectable marker.

Methods of Using the Library Compositions:

In another aspect, the present invention provides methods of gene delivery to a cell or tissue. The method includes contacting the cell or tissue with any one of the library compositions described above. Gene delivery may refer to the transportation of an encoding molecule into the cell or tissue. Non-limiting examples of encoding molecules include DNA, RNA, guide RNA (gRNA), shRNA, miRNA, ssDNA, pDNA, and sgRNA. Delivery of gRNA using the methods herein may be utilized in CRISPR/Cas gene editing, for example. However, the methods of gene delivery may be sensitive to molecule stability. In such cases, DNA and its derivatives may be the only encoding molecule capable of delivery in the present methods due to its stability. Genes or other payloads of interest — including, but not limited to therapeutics, peptides, and imaging agents — may be delivered in a targeted fashion. That is, cells, tissues, organs, or sites of disease of interest may specifically be targeted by the using the methods described herein. For example, pancreas-specific targeting may be performed using the P88 ligand. In another aspect, the present invention provides methods of screening the peptides of interest to find affinity reagents for targeting cells. The methods include contacting cells with the library composition described above. The contacted cells and EVs associated with the cells are then harvested and DNA is extracted. The nucleic acid encoding the surface-expressed peptide is amplified from the extracted DNA to produce a first screened nucleic acid pool. This can be accomplished using PCR based amplification using primers specific for the library composition. This first screened pool can be analyzed to determine the sequence of the nucleic acid directly or after amplification of the library composition based on library specific primers and its encoded peptide can be determined to identify peptides that mediate binding to and/or targeting of the cells by the surface-expressed peptide of interest.

As used herein, the term “contacting” includes contacting cells directly or indirectly in vivo, in vitro, or ex vivo. Contacting encompasses administration to a cell, tissue, mammal, patient, or human. Further, contacting a cell may include adding an agent, such as EVs of the library composition, to a cell culture. Other suitable methods may include introducing or administering an agent to a cell, tissue, mammal, or subject using appropriate procedures and routes of administration. The cells here may be contacted directly by adding the library composition to the media in which the cells are growing. Cells or tissues may also be contacted in vivo by administration of the library composition to a tissue directly for example oral administration to cells of the mouth or alimentary canal or topical administration to the skin, inhalation into the nasal passages. The library composition may also be administered via injection. Methods of administration are well known in the art and include, but are not limited to, oral administration, transdermal administration, administration by inhalation, nasal administration, topical administration, intravaginal administration, ophthalmic administration, intraaural administration, intracerebral administration, rectal administration, sublingual administration, buccal administration, and parenteral administration, including injectable such as intravenous administration, intra-arterial administration, intramuscular administration, intradermal administration, intrathecal administration, and subcutaneous administration. Injection or administration may be systemic or local.

After the cells or tissues are contacted with the library composition or the library composition is administered, the cells or tissues should be washed if in vitro or allowed enough time to remove unbound EVs if in vivo. Once the free and unbound EVs are removed or at least diminished, the cells or tissues are harvested. For cells this may be completed by centrifugation of non-adherent cells or by trypsinization or scraping to remove adherent cells followed by centrifugation. For tissues this will require harvesting the tissue and homogenization or protease treatment to prepare a cell suspension. The DNA can then be extracted using procedures well-known to those of skill in the art and using materials and kits available to those of skill in the art such as those described in the Examples. Suitably, the DNA is selected either before or after the extraction to prepare plasmid DNA or DNA similar in size to the library composition and to select against the cellular genomic DNA. Extraction of DNA (plasmid or otherwise) may be referred to as “DNA recovery” (“plasmid DNA recovery”, for example) and grammatical variants thereof.

The extracted DNA can then be amplified to produce a first screened nucleic acid pool using methods known to those of skill in the art. As used herein, the term “amplifying” or “amplification” refers to a template-dependent process that results in an increase in the concentration of a nucleic acid molecule relative to its initial concentration. A “templatedependent process” is a process in which the sequence of the newly synthesized nucleic acid is dictated by the rules of complementary base pairing. The amplification step of the present methods can be performed using any amplification method known in the art. Exemplary amplification methods include polymerase chain reaction (PCR), loop-mediated isothermal amplification (LAMP), strand displacement amplification (SDA), ligase chain reaction (LCR), and transcription-mediated-amplification (TMA). However, in preferred embodiments, the amplification step is performed using PCR.

In some embodiments, the method further comprises reiterative screening to produce a selected pool from the library composition with increased affinity for or ability to target a particular cell or tissue type. These methods take the first screened nucleic acid pool from the first round of screening and use the first screened nucleic acid pool to generate a first screen library composition. Briefly, the first screened nucleic acid pool is used to transfect cells and EVs are harvested from the cells. The generated EVs of the first screen library composition are then used as the starting EVs in the screening method to produce a second screened nucleic acid pool. This cycle can be repeated multiple times. In one embodiment the cycle is repeated at least five times.

As used herein, the term “transfect” or “transfection” refers to the introduction of DNA, RNA, other genetic material, protein or organelle into a target cell. Methods of transfection are known in the art, and exemplary methods can include, but are not limited to, transient transfection, stable transfection, co-transfection, electroporation, cationic lipid transfection, cationic polymer transfection, and some are described in the Examples.

In some embodiments, the method further comprises sequencing the screened nucleic acid pool(s). As used herein, the term “sequencing” refers to determining the order of nucleic acids in a nucleic acid molecule or amino acids in a polypeptide. Methods of sequencing are commonly known in the art.

Other methods of harvesting cells and/or DNA, transfection, sequencing, washing, extracting DNA, and amplifying nucleic acids are commonly known in the art and may be used with the methods herein. The present disclosure provides non-limiting examples of such methods which may be found in the present Examples

In some embodiments, the method further comprises generating EVs using the screened nucleic acid pool. Methods of generating the EVs using the screened nucleic acid pool can be extracted from the Examples. In some embodiments, the method comprises using the EVs generated by using the screened nucleic acid pool(s) for delivering an agent to a cell or tissue. The agent may be a small molecule, chemotherapeutic, peptide, polypeptide, or enzyme. The cells may be derived from or tissues may be is pancreatic, kidney, spleen, liver, brain or tumor.

As used herein, the term “agent” refers to a substance that brings about a chemical, biological, or physical effect or causes a chemical, biological, or physical reaction. Suitable agents for use with the present invention include a, without limitation, small molecule, chemotherapeutic, peptide, polypeptide, and enzyme. As used herein, the term “small molecule” refers to a low molecular weight (< 1000 Daltons) organic compound that may regulate a biological process, with a size on the order of 1 nm. Many drugs are small molecules; the terms are equivalent in the literature. Larger structures such as nucleic acids and proteins, and many polysaccharides are not small molecules, although their constituent monomers (ribo- or deoxyribonucleotides, amino acids, and monosaccharides, respectively) are often considered small molecules. Small molecules may be used as research tools to probe biological function as well as leads in the development of new therapeutic agents. Some can inhibit a specific function of a protein or disrupt protein-protein interactions. As used herein, the term “chemotherapeutic” refers to anti-cancer drugs that may or may not rely on non-specific intracellular interactions to inhibit mitosis. Chemotherapeutics can also be referred to as “cytotoxic agents”. As used herein, the term “enzyme” refers to a substance, typically a protein, which acts as a catalyst to bring about a specific biochemical reaction.

Circulation duration (that is, the length of time for which the libraries are allowed circulate within the host) may be about 15 minutes, about 30 minutes, about 1 hour, about 2 hours, about 3 hours, about 5 hours, about 10 hours, and so forth. Circulation duration may optimally be 1 hour. Methods of Making the Library Compositions:

In a third aspect, the present invention provides a method of making any one of the embodiments of the library composition described above. The methods of making the library compositions begin by generating the nucleic acid encoding the surface expressed peptides of interest. While multiple methods of generating plasmid or minicircle DNAs with peptides exist and may be used to generate the library compositions provided herein the inventors used the Seamless Cloning Ligation Cell Extract (SLiCE) method. SLiCE works similarly to other commercially available seamless assemble cloning methods, such as Gibson assembly, NEBuilder HiFi DNA Assembly, In-Fusion® and GeneArt®, and these are suitable alternatives for SLiCE. The library backbone was amplified from a DNA template comprising a polynucleotide encoding lactadherin C1C2 and a polynucleotide encoding the peptide of interest was also amplified from a template The products of these two amplification reactions were assembled together to create the library construct. The library construct was introduced into bacterial cells for propagation and storage. The library construct can be introduced into the bacterial cells using any means to transfect or transform bacterial cells and E. coll may be used. Those of skill in the art are capable of working with bacterial cells for propagation, storage and growth of these cells and the library contained therein. The library construct is then harvested from the bacterial cells and introduced (generally via a transfection method) into eukaryotic cells capable of producing EVs. The EVs are harvested from the supernatant in which the eukaryotic cells are growing and the harvested EVs are the library composition. The EVs may be collected as a pellet, and that EV-containing pellet may be resuspended in PBS and/or an EV storage buffer as described in Kawai-Harada, et al (2023). The EVs may similarly be administered, contacted, and/or delivered to the cells in the same PBS and/or EV storage buffer (which may be referred to as an EV delivery buffer). In some embodiments, the eukaryotic cells are HEK293T cells. HEK293T cells are efficiently transfected and efficiently produce EVs; these characteristics may make them optimal candidates for use in the methods described herein. CHO cells may prove to be a suitable alternative to HEK293T cells. The Examples provide methods of preparing the library constructs described herein and methods of producing and harvesting the library compositions.

As used herein, the term “backbone” refers to the portion of the library composition which is consistent between varying library compositions. As used herein, the term “introducing” refers to a process by which exogenous polypeptides or polynucleotides are introduced into a recipient cell. Suitable introduction methods will depend on the cell-type being used and include, without limitation, bacteriophage or viral infection, electroporation, heat shock, lipofection, microinjection, transformation, and particle bombardment. In some embodiments, the exogenous polynucleotides, constructs, or vectors described herein are transfected into a cell using a suitable carrier Suitable carriers are known and used in the art, including, but not limited to, lipid carriers (e.g., Lipofectamine), and polymeric nanocarriers.

In some embodiments, step (d) further comprises selecting for bacterial cells comprising a selectable marker, and in some embodiments, the selectable marker is ampicillin. As used herein, the term “selectable marker” refers to a protein that protects an organism from a selective agent that would normally kill it or prevent its growth. A selectable marker may be a compound that confers resistance to an otherwise toxic compound. For example, in some embodiments, the selectable marker confers resistance to an antibiotic (e.g., puromycin, penicillin, streptomycin, neomycin, or hygromycin). Cells that have been transfected with a construct encoding a selectable marker can be exposed to the selection agent to select for cells comprising the construct.

Monobodies:

In another aspect, the present invention provides monobodies derived from the methods described herein. The library composition provided herein may comprise a monobody backbone (the fibronection FN3 domain) comprising random sequences in each of the three loop regions that may mediate binding of the monobody to its target protein. In the Examples the inventors used a monobody library to screen for new monobodies capable of binding EGFR + cells. Using the methods of affinity reagent screening described herein, the inventors found at least 5 candidate monobodies. Using an EV-based monobody display screening strategy, we were able to display novel HEK293T-derived monobody proteins on the surfaces of EV library EVs that resulted in the enrichment of cell-type specific monobody sequences that possessed high binding affinities Those of skill in the art will recognize that the monobodies have three loops that are required for binding the target protein and that the monobodies can tolerate some additional amino acid substitutions outside of these regions without affecting binding to the target. These monobodies may be utilized in the methods described herein as an affinity reagent for further iterations of library construction or as a gene-delivery protein targeting the pancreas, for two examples.

Miscellaneous:

The terms “protein,” “polypeptide,” and “peptide” are used interchangeably herein to refer to a series of amino acid residues connected by peptide bonds between the alpha-amino and carboxy groups of adjacent residues, forming a polymer of amino acids. Proteins may include modified amino acids. The term “nucleic acid” as used herein includes “polynucleotide,” “oligonucleotide,” and “nucleic acid molecule,” and generally refers to a polymer of DNA or RNA, which may be single-stranded or double-stranded, synthesized or obtained (e g., isolated and/or purified) from natural sources, which may contain natural, non-natural or altered nucleotides. It is widely known in the art that nucleic acid length can be referred to by its number of base pairs and often recorded in terms of “bases”, “kilobases”, or “kB”.

In the Examples, the inventors disclose a screening platform designed to overcome technical limitations — including a lack of controlled generation of EVs, inefficient library loading, and inadequate tissue specificity — that can be utilized for molecular screening.

The present disclosure is not limited to the specific details of construction, arrangement of components, or method steps set forth herein. The compositions and methods disclosed herein are capable of being made, practiced, used, carried out and/or formed in various ways that will be apparent to one of skill in the art in light of the disclosure that follows. The phraseology and terminology used herein is for the purpose of description only and should not be regarded as limiting to the scope of the claims. Ordinal indicators, such as first, second, and third, as used in the description and the claims to refer to various structures or method steps, are not meant to be construed to indicate any specific structures or steps, or any particular order or configuration to such structures or steps. All methods described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The use of any and all examples, or exemplary language (e.g., "such as") provided herein, is intended merely to facilitate the disclosure and does not imply any limitation on the scope of the disclosure unless otherwise claimed. No language in the specification, and no structures shown in the drawings, should be construed as indicating that any non-claimed element is essential to the practice of the disclosed subject matter. The use herein of the terms “including,” “comprising,” or “having,” and variations thereof, is meant to encompass the elements listed thereafter and equivalents thereof, as well as additional elements. Embodiments recited as “including,” “comprising,” or “having” certain elements are also contemplated as “consisting essentially of’ and “consisting of’ those certain elements.

The phrase “and/or,” as used herein in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and/or” should be construed in the same fashion, i.e., “one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the “and/or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and/or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc. As used herein in the specification and in the claims, “or” should be understood to have the same meaning as “and/or” as defined above. For example, when separating items in a list, “or” or “and/or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one, of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as “only one of’ or “exactly one of,” or, when used in the claims, “consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term “or” as used herein shall only be interpreted as indicating exclusive alternatives (i.e. “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.”

Recitation of ranges of values herein are merely intended to serve as a shorthand method of referring individually to each separate value falling within the range, unless otherwise indicated herein, and each separate value is incorporated into the specification as if it were individually recited herein. For example, if a concentration range is stated as 1% to 50%, it is intended that values such as 2% to 40%, 10% to 30%, or l% to 3%, etc., are expressly enumerated in this specification. These are only examples of what is specifically intended, and all possible combinations of numerical values between and including the lowest value and the highest value enumerated are to be considered to be expressly stated in this disclosure. Use of the word “about” to describe a particular recited amount or range of amounts is meant to indicate that values very near to the recited amount are included in that amount, such as values that could or naturally would be accounted for due to manufacturing tolerances, instrument and human error in forming measurements, and the like.

While the claims provided herein are directed to methods of treating a subject, both human and non-human subjects are envisioned. In addition, use of the compositions provided herein as medicaments for uses in therapy or for treating disease are also provided herein. Use of the compositions provided herein in the manufacture of a medicament for the treatment of a disease or condition are also encompassed.

No admission is made that any reference, including any non-patent or patent document cited in this specification, constitutes prior art. In particular, it will be understood that, unless otherwise stated, reference to any document herein does not constitute an admission that any of these documents forms part of the common general knowledge in the art in the United States or in any other country. Any discussion of the references states what their authors assert, and the applicant reserves the right to challenge the accuracy and pertinence of any of the documents cited herein. All references cited herein are fully incorporated by reference, unless explicitly indicated otherwise. The present disclosure shall control in the event there are any disparities between any definitions and/or description found in the cited references.

The following examples are meant only to be illustrative and are not meant as limitations on the scope of the invention or of the appended claims.

EXAMPLES

Example 1: Use of extracellular vesicles as a molecular screening platform

In the following example, the inventors describe a method of employing extracellular vesicles (EVs) as a molecular screening platform for EV-mediated targeted delivery. Previously published methods of engineering targeting EVs are available in Nanotheranostics (Komuro 2021) and Bioengineering (Komuro 2022), which are both publicly available and are incorporated in their entirety herein. The following describes adapting the methods of engineering targeting EVs for use as amolecular screening platform.EVs are naturally released, cell-derived vesicles that mediate intracellular communication, in part, by transferring genetic information and, thus, have the potential to be modified for use as a therapeutic gene or drug delivery vehicle. We previously demonstrated that EV surface modification with tissuespecific molecules accomplished targeted EV-mediated DNA delivery (Komuro 2021; Komuro 2022). Here, we describe reliable methods for (i) generating EGFR tumor-targeting EVs via the display of high-affinity monobodies, (ii) in vitro measurement of EV binding using fluorescence and bioluminescence labeling, and (iii) enriched delivery of engineered EVs displaying an organ targeting peptide specific to the pancreas.

Monobodies are a well-suited class of small (10 kDa) non-antibody scaffolds derived from the human fibronectin type III (FN3) domain. A recombinant protein consisting of the EGFR-targeting monobody fused to the EV-binding domain of lactadherin (C1C2) enable the monobody to be displayed on the surface of the EVs. In addition, the use of bioluminescence or fluorescence molecules on the EV surface allows for the assessment of EV binding to the target cells in vitro. Additionally, intravenous administration of a P-cell-specific recombinant protein consisting of the peptide p88 fused to the EV-binding domain of lactadherin (C1C2) showed an altered pattern of EV localization and improved DNA delivery to the pancreas relative to control EVs. Here, we describe methods of EV engineering to generate targeted delivery vehicles using monobodies that will have diverse applications to furnish future EV therapeutic development, including qualitative and quantitative evaluation for their binding capacity.

It has been demonstrated that producer cells, transfected with an expression plasmid, package plasmid DNA into EVs, and if that plasmid encodes a surface protein that is expressed on EVs then the EVs will contain the DNA representing the surface expressed protein (Kanada 2015). EVs from cells containing such plasmid DNAs would then have a feature of bacteriophages that make phage display screens possible. Towards developing in vitro and in vivo EV display screening protocols, we evaluated cell and tissue targeting and EV packaging. Previously, we used C1C2 to display p88, a peptide with known affinity for the human and mouse membrane small ion transport regulator (FXYD2)ya, to generate pancreas targeting EVs (Komuro 2021). Similarly, we used the EV surface display approach with the use of an engineered monobody (E626), a clinically approved binding molecule with high affinity towards EGFR tumors (Komuro 2022). We use this method to identify novel monobody affinity reagents to demonstrate this method of affinity reagent display and selection works. Materials and Methods (in vitro) p88 EV Construction

Seamless Ligation Cloning Extract (SLiCE) mediates in vitro DNA assembly through a RecA- independent recombination mechanism between DNA fragments with short homologous ends. The SLiCE reagent, a bacterial lysate, was prepared from E.coli DH5a strain and used for all the cloning in this work following a previously described (Motohashi 2015; Zhang 2014). Briefly, EV-display constructs were created by SLiCE assembly of PCR fragments into pcDNA6.0 V5/His (Invitrogen) digested with Nhel and Agel. The signal peptide and lactadherin C1C2 domain were amplified from psd44-Lactadherin46 (a gift from Agnese Mariotti Addgene, plasmid # 46830) using the primer sets (SEQ ID NOs: 1 and 2) which included overhangs. The PCR fragment was amplified from the pcDNA-C!C2 plasmid and assembled with the synthetic double-stranded oligonucleotide consisting of pepl coding sequence and the (GGGGS)3 linker sequence (SEQ ID NO: 26) (pepl-3xG4S- C1C2; SEQ ID NO: 23) to generate pepl-EV display construct. Similarly, a fragment was amplified from the pepl-EV display construct and assembled from synthetic oligonucleotides consisting of p88 coding sequence with homologous ends (P885-3-1 and P885-3-2; SEQ ID NOs: 24 and 25, respectively) to create p88-EV display construct. pcDNA backbone for both pepl (SEQ ID NOs: 7 and 8) and p88 (SEQ ID NOs: 9 and 10) was further down-sized by removing unnecessary sequences including the mammalian selectable marker (Blasticidin) and the phage origin of replication through the single-piece SLiCE reaction of the PCR fragments. This resulted in pcS-p88-ClC2 and pcS-pepl-ClC2. Another fragment was amplified from the pcS- p88-ClC2 using a primer set (HA-3xG4S-F and HA-R; SEQ ID NOs: 15 and 16, respectively) and assembled to generate non-Peptide display construct, pcS-NP- C1C2 (pcS primer sets displayed in SEQ ID NOs: 11 and 12). Each construct was purified using the QIAprep Miniprep kit (QIAGEN) and sequenced for validation. The Plasmid Plus Midi Kit (QIAGEN) was used for large scale plasmid preparation. Similar methods may be used to generate a library comprising multiple peptides of interest instead of the p88 peptide used in the above-method. EV Monobody Display Plasmid Construction

The EV monobody display constructs were created using Seamless Ligation Cloning Extract (SliCE) assembly, as previously described. PCR fragment of EV display backbone from pcS-p88-ClC2 was fused with the synthetic double stranded DNA fragments coding for monobody (E626 or RDG) and G4S- PAS linker (Stern 2016) including 15 bp overhangs. The three double-stranded DNA fragments were joined together by homologous recombination using SliCE cloning. Fragment 1: PCR amplified product of EV display backbone amplified from pcS-p88-ClC2 with 15 bp overlaps at the HA and the G4S-PAS linker; Fragment 2: synthetic DNA of the monobody (E626 or RDG) with 15 bp overlaps with the end of HA tag and the start of the linker; Fragment 3: synthetic DNA of the linker that overlaps with both fragments 1 and 2 at each end. All the synthetic DNAs were purchased from Integrated DNA Technologies (IDT; Coralville, IA, USA).

Library DNA preparation

The method and the backbone constructs were adopted from our previous work. (Komuro 2022; Komuro 2021)

Briefly, the library backbone was amplified from a template pcS-RDG-ClC2 using, devoid of the variable loop regions. Monobody library fragments coding for variable loop region containing 15 bp overlap at both ends were amplified (Woldring 2015). These fragments were joined together via Seamless cloning Ligation Cell Extract (SliCE). Following the cleanup by QIAquick PCR purification kit and the concentration measurement by the Qubit system (Invitrogen), the assembled DNA was electroporated into electrocompetent E. coli cells (NEB) and pre-cultured at 37°C for 1 hour without antibiotics, and further cultured in the LB- ampicillin in a flask for 8 hours in the bacterial shaking incubator. The library DNA was extracted by the Midiprep kit (QIAGEN). The concentration was determined by NanoDrop. Cell Culture and Treatment Before the experiments, the following cell lines from American Type Culture Collection (ATCC), were tested for mycoplasma: HEK293T (Human Embryonic Kidney cell line), A431 (Human carcinoma cell line), and MCF-7 (human breast cancer cell line). The cells were cultured in high-glucose DMEM (Gibco) supplemented with lOOU/mL penicillin, 100 pg/mL streptomycin, and 10% (v/v) fetal bovine serum (FBS, Gibco). All cells were maintained in a humidified incubator with 5% CO2 at 37°C. For EV production, HEK293T cells were seeded at 2xl0⁶ in a 10 cm tissue culture dish 24 h prior to transfection. The next day, 10 pg DNA was mixed with PEI in a ratio of 1:2.5 (DNA/PEI) in non-supplemented DMEM, pulse- vortexed for 30 s, incubated at room temperature for 10 min and added to the cells.1 Following 24 h incubation, cells were washed twice with PBS, and the culture media was replaced with 20 mL of DMEM supplemented with Insulin-Transferrin-Selenium (ITS) (Corning), lOOU/mL penicillin and 100 pg/mL streptomycin (conditioned media) and incubated for another 24 h for library EV generation or single-monobody EV generation. For gLuc binding assay and co-culture experiments, EVs were co-labeled with imaging molecules by co-transfecting 5pg of monobody-display plasmid and 5ug of pcS-gLuc-ClC2 (forward and reverse primers represented by SEQ ID NOs: 5 and 6) or pcS-mCherry-ClC2 (forward and reverse primers represented by SEQ ID NOs: 3 and 4).

EV Isolation

EVs were purified from 20 mL of conditioned media by differential centrifugation. Briefly, the media was centrifuged at 400g for 10 min and then 600g for 30 min to remove the cell and cell debris, and the supernatant was further centrifuged at 2000g for 30 min to remove apoptotic bodies. The supernatant was then ultracentrifuged in PET Thin-Walled ultracentrifuge tubes (Thermo Scientific 75000471) at 100,000g with a Sorvall WX+ Ultracentrifuge equipped with an AH-629 rotor (k factor = 242.0) for 90 min at 4°C to pellet the EVs. The pellet containing EVs was resuspended in 100 pL EV storage buffer (Kawai- Harada 2023).

In vitro EV library screening

For the EV library screening, A431 cells were seeded at 0.3xl0⁶ cells/well in 6-well plates 24 h prior to EV treatment. The cells were treated with 2.0xl0⁷ library EVs in 2 mL media for 30 min at 37°C. Following the PBS wash to remove residue of EVs, cells were harvested using trypsin. The plasmid DNA was isolated from the cells following a modified protocol for plasmid isolation from the cells following a modified protocol for plasmid isolation from organ homogenates using QIAprep Spin Miniprep Kit, which was used as a template for the next round of library DNA preparation described above. A series of EV library screenings was repeated 5 times to enrich targeting monobody sequences. At each round, the variable loop region was subjected to next generation sequencing and analyzed for enrichment.

In vitro Bioluminescent assay

In this assay, 10⁷ naive EVs, 10⁷ gLuc EVs and 10⁷ #l-#20-monobody-gLuc EVs were placed in wells of a 96 well plate (UV-Star® Microplate, 96 well, COC, F-Bottom (Chimney Well), uClear®, Clear; Greiner Bio-one) in triplicate. 95 pl of DPBS was added to each well and then treated with 50 pL 1.5 pM Coelenterazine-H (CTZ; Regis Technologies). The luminescence was recorded using an in vivo imaging system (IVIS; Spectrum Perkin Elmer) and the particle numbers emitting equal amounts of luminescence/radiance (photons/sec/cm2/sr) was calculated. For control, 5pL of DPBS and DMEM media were used and treated in the same manner.

Materials and Methods (In vivo)

In vivo p88

Library DNA preparation

The method and the backbone constructs were adopted from our previous work. (Komuro 2022; Komuro 2021).

Briefly, the library backbone was amplified from a template pcS-RDG-ClC2, devoid of the variable loop regions. Monobody library fragments coding for variable loop region containing 15 bp overlap at both ends were amplified (Woldring 2015). These fragments were joined together via Seamless cloning Ligation Cell Extract (SliCE). Following the clean-up by QIAquick PCR purification kit and the concentration measurement by the Qubit system (Invitrogen), the assembled DNA was electroporated into electrocompetent E. coli cells (NEB) and pre-cultured at 37°C for 1 hour without antibiotics, and further cultured in the LB- ampicillin in a flask for 8 hours in the bacterial shaking incubator. The library DNA was extracted by the Midiprep kit (QIAGEN). The concentration was determined by NanoDrop. Cell Culture and Treatment

Before the experiments, the HEK293T (Human Embryonic Kidney cell line) from American Type Culture Collection (ATCC), were tested for mycoplasma. The cells were cultured in high-glucose DMEM (Gibco) supplemented with lOOU/mL penicillin, 100 pg/mL streptomycin, and 10% (v/v) fetal bovine serum (FBS, Gibco). All cells were maintained in a humidified incubator with 5% CO2 at 37°C. For EV production, HEK293T cells were seeded at 2xl0⁶ in a 10 cm tissue culture dish 24 h prior to transfection. The next day, 10 pg DNA was mixed with PEI in a ratio of 1:2.5 (DNA/PEI) in non-supplemented DMEM, pulse- vortexed for 30 s, incubated at room temperature for 10 min and added to the cells.1 Following 24 h incubation, cells were washed twice with PBS, and the culture media was replaced with 20 mL of DMEM supplemented with Insulin-Transferrin-Selenium (ITS) (Corning), 1 OOU/mL penicillin and 100 pg/mL streptomycin (conditioned media) and incubated for another 24 h for library EV generation or single- monobody EV generation. For gLuc binding assay and coculture experiments, EVs were co-labeled with imaging molecules by co-transfecting 5 pg of monobody-display plasmid and 5 pg of pcS-gLuc-ClC2 or pcS-mCherry-ClC2.

In vivo EV library screening

In this study, 6-week old female Balb/c mice were used for animal experiments. Animals were purchased from Jackson Laboratories and housed in the University Laboratory Animal Resources Facility at Michigan State University. All the experimental procedures for the animal study were performed with the approval of the Institutional Animal Care and Use Committee of Michigan State University.

Approximately x 10⁹ monobody-EVs in 100 pl EV storage buffer were injected into mice intravenously (iv). Control mice were injected with naive EVs. After 1-h of EVs administration, the mice were sacrificed and the visceral organs (heart, lung, liver, kidney, pancreas and spleen) were dissected and homogenized using Triple-Pure High Impact 2.8mm Steel Beads (Benchmark Scientific) and BeadBug 6 Microtube Homogenizer (Benchmark Scientific). The plasmid DNA was isolated from the organ homogenates using QIAprep Spin Miniprep Kit following a modified protocol for plasmid isolation from mammalian cells which was used as a template for the next round of library DNA preparation described above. The copy number of the plasmids was assessed by qPCR-based TaqMan assay. A series of EV library screenings was repeated 5-7 times to enrich targeting monobody sequences of nearly 100 clones. The 20-100 variable loop fragments were re-cloned into the EV display construct for individual monobody assays.

Quantitative Real-time Polymerase Chain Reaction (qPCR) qPCR was performed using Taq DNA polymerase (Fisher BioReagents). Each reaction contains 200 pM dNTP, 500 nM each of forward/reverse primer, 400 nM probe, 0.5 U Taq DNA polymerase, lx Assay buffer A and 1 pL sample DNA or isolated EV in a total reaction volume of 10 pL using CFX96 Touch Real-Time PCR Detection System (BIO-RAD). The PCR amplification cycle was as follows: 95°C for 2 min; 40 cycles of 95°C for 20 seconds, 65°C for 30 seconds. The plasmid DNA copy number was determined by absolute quantification using the standard curve method, and the copy number of EV encapsulated plasmid DNA per vesicles was calculated based on NTA and qPCR results. The plasmid DNA copy numbers recovered from each organ were calculated by qPCR and compared after normalization to the organ weights.

Next-Generation Sequencing

The samples for next-generation sequencing (Illumina MiSeq) were prepared by PCR amplification using a primer set with a sequencing index for the sequencing reaction. A fluorometric method (Qubit) was used to quantify the PCR products before the submission. All the samples were normalized to the same concentration and agarose gel electrophoresis was used to confirm the product size. Sequencing was performed at the MSU Genomics Core facility using MiSeq Reagent Kit v3 for 250 bp paired-end (PE) reads. The generated FASTQ format file was extracted, processed and clustered by sequence similarity using our custom software, ScaffoldSeq (Woldring 2016).

In vivo imaging of mice and organs

Anesthetized mice were intravenously injected with gLuc or Monobody-gLuc EVs emitting an average radiance of X photons/sec/cm2/sr. Naive EVs were used as control. Immediately before imaging, 150 pL CTZ (lOOpg) was injected intravenously into each mouse. IVIS was used for the bioluminescence imaging (BLI) of the live mouse immediately after substrate injection. After imaging, the mice were sacrificed and the following visceral organs were dissected and placed on a transparent sheet: heart, lungs, liver, kidneys, pancreas and spleen. Ex vivo images of BLI were taken following re-application of CTZ to the resected organs by IVIS.

Results:

We have previously described the design, generation and characterization of engineered EVs displaying pancreatic p-cell targeting peptides (Komuro 2021). There we showed EVs engineered to display p88 (p88-EVs) exhibited higher binding capacity to P-cells in co-cultured cells in vitro. However, the in vitro assay does not represent the biological response in vivo and is not conclusive. Further in vivo imaging and analysis of harvested organs demonstrated that small peptide-based ligands can impart affinity to EVs upon being displayed on the surface. EV-mediated targeted delivery was achieved without any observed toxicity or visible side effects on the mice. We have additionally described a method of EV surface engineering using lactadherin C1C2 gene fusion in HEK293T cells, comprehensive characterization of engineered EVs, and verification of EV binding in vitro using an EGFR-targeting monobody (Komuro 2022).

Here, we utilize the EV surface engineering described previously to describe a molecular screening platform. A schematic of the use of cell-derived vesicles as an in vivo screening platform in mammals is shown in FIG. 1. Monobody-library EV display constructs are transiently transfected into HEK293T cells to generate the EV library. The EVs will contain monobody-coding DNA and display the protein on its surface. The EV display technology is shown in FIG. 2. Fusing Fn3HP -based monobody- coding DNA with the lactadherin C1C2 domain allows the monobody to be displayed on the EV surface.

In vitro uptake demonstrates the genotype-phenotype link of EV-molecular display (FIG. 3). EGFR positive (A431) cells were treated with EVs prepared from the mixed ratio of EGFR-targeting (E626) and non-targeting (RDG) monobodies. qPCR analysis was performed on recovered plasmid DNA following 10-, 30-, and 60-min treatment. RIO indicates EVs prepared from 100% of RDG DNA, R9E10 from RDG90%EGFR10%, R5E5 from RDG50%EGFR50%, R1E9 from RDG10%EGFR90% and E10 from EGFR100%. The ratios of input plasmid and the plasmids recovered after binding demonstrate that the plasmid DNA recovered is that of the protein allowing targeting of the cell. This assay thus validates the screening methods provided herein.

Plasmid DNA from injected libraries can be recovered from various organs (FIG. 4). Copy numbers determined by qPCR of isolated plasmid DNA retrieved from organs such as pancreas, spleen, kidney, and liver and agarose gel showing the monobody fragments recovered from each organ using the re-cloning primer set shows that plasmid DNA can be efficiently recovered from targeted organs. Recovered DNA can be successfully enriched following recovery from target organs as demonstrated by in vitro uptake of targeting EV-pDNA from the library pool (FIG. 8C). EGFR positive (A431) cells were treated with EVs prepared from the library monobody spiked in with 1% each of EGFR-targeting (E626) and non-targeting (RDG) monobodies.

Discussion:

Unlike synthetic nanocarriers, EVs feature more robust stability in vivo since they do not provoke strong immunogenic responses or toxic side effects (Yanez-Mo 2015). Furthermore, targeting and non-targeting EVs can be engineered with relative ease by fusing targeting moi eties to known EV surface proteins like Lamp2b, tetraspanins (CD63, CD81, CD53, CD37, and CD82), and Lactadherin (Alvarez-Erviti 2011; Yang 2017; Liang 2018; Tian 2014; Salunkhe 2020). Targeting is key to effective delivery of therapeutics allowing precise localization to diseased tissues and thus eliminating side effects derived from off-target effects of large drug dosage. Similarly, this approach could deliver genes to create producer cells in target tissues to generate bystander effects which can influence groups of surrounding cells as was reported by Kanada et al (2019). EV generator cells with engineered surface ligands cloned into the genome may allow efficient EV production at scale.

Our engineered EV generation technique is simple, robust, and efficient. Previous studies demonstrates that small peptide-based ligands and monobodies can impart affinity to EVs upon being displayed on the surface using lactadherin C1C2 gene fusion. EV-mediated targeted delivery was achieved without any observed toxicity in the cell lines or visible side effects on the mice.

The methods including EV engineering, synthesis, isolation, mass production, and analytical tools are evolving rapidly, yet they have considerable room for improvement and verification. Here, we disclose a screening platform designed to overcome technical limitations — including a lack of controlled generation of EVs, inefficient library loading, and inadequate tissue specificity — that can be utilized for molecular screening

This screening platform, outlined in FIG. 1, can utilize a library composed of EVs with surface-expressed peptides of interest (FIG. 2) and nucleic acids encoding the surface peptides of interest. The study described here can serve as an example of such a platform that uses EGFR-targeting monobodies (E626) as the peptide of interest (FIG. 3). The current study shows the library compositions can be successfully recovered from target and non-target tissues (FIG. 4) and subsequently enriched (FIG. 7C). DNA recovered and amplified from the library compositions recovered from tissues can be used thereafter to produce a pool of screened nucleic acids. This method can be further iterated to transfect cells with the screened nucleic acids, harvesting the EVs produced, and repeating the process for a novel method of generating reiterative screened nucleic acid pools. Reiterative screened nucleic acid pools can then produce reiterative, screened, target-specific EVs.

The EV-mediated targeted delivery can be leveraged for treating human diseases. Moreover, conjugating therapeutic molecules/drugs/imaging probes with engineered EVs can be applied for investigating targeted delivery in other clinically significant organs.

References

Alvarez-Erviti L, Seow Y, Yin H, Betts C, Lakhal S, Wood MJ. Delivery of siRNA to the mouse brain by systemic injection of targeted exosomes. Nature biotechnology. 2011; 29: 341-5.

Kanada M, Bachmann MH, Hardy JW, Frimannson DO, Bronsart L, Wang A, et al. Differential fates of biomolecules delivered to target cells via extracellular vesicles. Proc Natl Acad Sci U S A. 2015; 112: E1433-42. Kanada M, Kim BD, Hardy JW, Ronald JA, Bachmann MH, Bernard MP, et al. Microvesicle- Mediated Delivery of Mini circle DNA Results in Effective Gene-Directed Enzyme Prodrug Cancer Therapy. Mol Cancer Then 2019; 18: 2331-42.

Kawai-Harada, Y., El Itawi, H., Komuro, H. & Harada, M. Evaluation of EV Storage Buffer for Efficient Preservation of Engineered Extracellular Vesicles. Int J Mol Sci 24, doi:10.3390/ijms241612841 (2023).

Komuro, H. et al. 2021. Engineering Extracellular Vesicles to Target Pancreatic Tissue. Nanotheranostics 5, 378-390, doi:10.7150/ntno.54879.

Komuro, H., Aminova, S., Lauro, K , Woldring, D. & Harada, M. 2022. Design and Evaluation of Engineered Extracellular Vesicle (EV)-Based Targeting for EGFR-Overexpressing Tumor Cells Using Monobody Display. Bioengineering (Basel) 9, doi : 10.3390/bioengineering9020056.

Liang G, Kan S, Zhu Y, Feng S, Feng W, Gao S. Engineered exosome-mediated delivery of functionally active miR-26a and its enhanced suppression effect in HepG2 cells. Int J Nanomedicine. 2018; 13: 585-99.

Motohashi K. A simple and efficient seamless DNA cloning method using SliCE from Escherichia coli laboratory strains and its application to SliP site-directed mutagenesis. BMC Biotechnology. 2015; 15: 47.

Salunkhe S, Dheeraj, Basak M, Chitkara D, Mittal A. Surface functionalization of exosomes for target-specific delivery and in vivo imaging & tracking: Strategies and significance. J Control Release. 2020; 326: 599-614.

Stern, L ; Schrack, I. A.; Johnson, S.M.; Deshpande, A.; Bennett, N.R.; Harasymiw, L.A.; Gardner, M.K.; Hackel, B.J. Geometry and expression enhance enrichment of functional yeast-displayed ligands via cell panning. Biotechnol. Bioeng. 2016, 113, 2328-2341.

Tian Y, Li S, Song J, Ji T, Zhu M, Anderson GJ, et al. A doxorubicin delivery platform using engineered natural membrane vesicle exosomes for targeted tumor therapy. Biomaterials. 2014; 35: 2383-90.

Woldring, D. R., Holec, P. V., Zhou, H. & Hackel, B. J. 2015. High-Throughput Ligand Discovery Reveals a Sitewise Gradient of Diversity in Broadly Evolved Hydrophilic Fibronectin Domains. PloS One 10, e0138956, doi: 10.1371/joumal. pone.0138956.

Woldring, D. R., Holec, P. V. & Hackel, B. J. 2016. ScaffoldSeq: Software for characterization of directed evolution populations. Proteins 84, 869-874, doi:10.1002/prot.25040. Yanez-Mo M, Siljander PR, Andreu Z, Zavec AB, Borras FE, Buzas El, et al. Biological properties of extracellular vesicles and their physiological functions. J Extracell Vesicles. 2015; 4: 27066.

Yang J, Zhang X, Chen X, Wang L, Yang G. Exosome Mediated Delivery of miR-124 Promotes Neurogenesis after Ischemia. Mol Ther Nucleic Acids. 2017; 7: 278-87.

Zhang Y, Werling U, Edelmann W. Seamless ligation cloning extract (SliCE) cloning method. DNA Cloning and Assembly Methods: Springer; 2014. P. 235-44.

Example 2: EV monobody display directs pancreas-specific accumulation

Using an EV-based monobody display screening strategy, we were able to display HEK293T-derived monobody proteins on the surfaces of EV library EVs. This resulted in the enrichment of cell-type specific monobody sequences along with a high-binder control. Binding assays confirmed that the monobody, E626, and five novel individual monobody clones possessed a high binding affinity. Lastly, pancreas-enriched monobody library EVs accumulated in the pancreas of mice in vivo. This at least shows that this library display strategy may be utilized for organ-specific protein library synthesis and may be used for organ-specific delivery of an EV payload. This development will allow for screening and development of EV targeting strategies so that EVs can be used as specific therapeutic delivery vehicles.

Materials and Methods

DNA construct

The cloning method and the backbone constructs were adopted from our previous work and synthetic double-stranded DNA coding for NanoLuc¹ were purchased from (Twist Bioscience) to generate pcS-NanoLuc-ClC2.^2,3 Library DNA preparation

The method and the backbone constructs were adopted from our previous work.^2,3 Briefly, the library backbone was amplified from a template pcS-RDG-ClC2 using primers Lib-BB-F and Lib-BB-R, followed by Dpn I treatment to eliminate template DNA. Monobody library fragments coding for variable loop region containing 44 bp at 3 ’-end and 45 bp overlap at 5 ’-end were amplified using primers Lib-IN-F and Lib-IN-R⁴ and fused to the backbone using Seamless cloning Ligation Cell Extract (SliCE) method.^2,3 Following the clean-up using QIAquick PCR purification kit (QIAGEN) and the concentration measurement by the Qubit dsDNA BR Kit (Invitrogen), the assembled DNA was electroporated into electrocompetent E. coli cells (NEB) and pre-cultured at 37°C for 1 hour without antibiotics, and further cultured in the LB-ampicillin in a flask for 8 hours in the bacterial shaking incubator. The library DNA was extracted by the Midiprep kit (QIAGEN). The DNA concentration was determined by NanoDrop (Thermo Fisher Scientific).

Cell Culture and Treatment

Before the initial experiments, the following cell lines obtained from American Type Culture Collection (ATCC), were tested for mycoplasma: HEK293T (Human Embryonic Kidney cell line), A431 (Human carcinoma cell line), and MCF-7 (human breast cancer cell line). The cells were cultured in high-glucose DMEM (Gibco) supplemented with lOOU/mL penicillin, 100 pg/mL streptomycin, and 10% (v/v) fetal bovine serum (FBS, Gibco). All cells were maintained in a humidified incubator with 5% CO2 at 37 °C. For EV production, HEK293T cells were seeded at IxlO⁶ in a 10 cm tissue culture dish 24 h prior to transfection. The next day, 10 pg DNA was mixed with PEI in a ratio of 1:2.5 (DNA/PEI) in nonsupplemented DMEM, pulse-vortexed for 30 s, incubated at room temperature for 10 min and added to the cells.²² Following 24 h incubation, cells were rinsed once with PBS, and the culture media was replaced with 20 mL of DMEM supplemented with Insulin-Transferrin-Selenium (ITS) (Corning), lOOU/mL penicillin and 100 pg/mL streptomycin (conditioned media) and incubated for another 24 h for library EV generation or single-monobody EV generation. For NanoLuc binding assay, EVs were co-labeled with imaging molecules by co-transfecting 8 pg of monobody-display plasmid and 2 pg of pcS-NanoLuc-ClC2.

EV isolation

EVs were purified from 20 mL of conditioned media by differential centrifugation. Briefly, the media was centrifuged at 600g for 30 min to remove the cell and cell debris, and the supernatant was further centrifuged at 2000g for 30 min to remove apoptotic bodies. The supernatant was then ultracentrifuged in PET Thin-Walled ultracentrifuge tubes (Thermo Scientific 75000471) at 12,000g with a Sorvall WX+ Ultracentrifuge equipped with an AH- 629 rotor (k factor = 242.0) for 90 min at 4 °C to pellet the EVs. The pellet containing EVs was resuspended in 100 pL EV storage buffer.⁵ Dnase I Treatment of EVs

The 5 pL of Monobody Library EVs were incubated at room temperature for 15 min with 1 U of Dnase I (Zymo Research) and DNA Digestion Buffer. The plasmid DNA was isolated from the EVs using Qiamp Miniprep kits and quantified by qPCR.

Quantitative Real-time Polymerase Chain Reaction (qPCR) qPCR was performed using Dream Taq DNA polymerase (ThermoFisher). Each reaction contains 200 pM dNTP, 500 nM each of forward/reverse primer, 400 nM probe (Table 1), 0.5 U DreamTaq DNA polymerase, lx Dream Taq buffer A and 1 pL sample DNA in a total reaction volume of 10 uL using CFX96 Touch Real-Time PCR Detection System (BIORAD). The PCR amplification cycle was as follows: 95°C for 2 min; 40 cycles of 95°C for 20 seconds, 65°C for 30 seconds. The pDNA copy number were determined by absolute quantification using the standard curve method, and the copy number of EV encapsulated pDNA per vesicles was calculated based on NTA and qPCR results. qPCR primer and probes are represented by SEQ ID NOs: 13 and 14; primer and probe for peptide region are represented by SEQ ID NOs: 17 and 18; primer and probes for ampicillin region are represented by SEQ ID NOs: 19 and 20; primer and probe for C1C2 region are represented by SEQ ID NOs: 21 and 22.

In vitro EV library screening

For the EV library screening, A431 cells were seeded at 0.3xl0⁶ cells/well in 6-well plates 24 h prior to EV treatment. The cells were treated with 2.0xl0⁷ library EVs in 2 mL media for 30 min at 37 °C. Following the PBS wash to remove residue of EVs, cells were harvested using trypsin. The plasmid DNA was isolated from the cells following a modified protocol for plasmid isolation from the organ homogenates using QIAprep Spin Miniprep Kit,⁶ which was used as a template for the next round of library DNA preparation described above. A series of EV library screenings was repeated 5 times to enrich targeting monobody sequences.

In vitro Biohimine scent assay

A431 cells were seeded at 0.02xl0⁶ cells/well in 96-well plates (UV-Star® Microplate, 96 well, COC, F-Bottom (Chimney Well), uClear®, Clear; Greiner Bio-one) 24 h prior to EV treatment. In this assay, 5xl0⁶ EVs were placed in wells in triplicate. After incubation at 37°C, cells were washed twice with PBS to remove residual un-bound EVs. 50 pL lug/mL Coelenterazine-H (CTZ; Regis Technologies) was added to each well right before imaging. The luminescence was recorded using an in vivo imaging system (IVIS; Spectrum Perkin Elmer) and the particle numbers emitting equal amounts of luminescence/radiance (photons/sec/cm2/sr) was calculated.

Super Resolution Microscopy

Isolated monobody Library EVs was analyzed with EV Profiler V2 Kit for Nanoimager (ONI) by following manufacture’s protocol. The imaging data was analyzed by CODI software (ONI).

Next-Generation Sequencing The sample for next-generation sequencing (Illumina MiSeq) were prepared by PCR amplification using a primer pair (CS 1 -LibHA-F, CS2-G4S-R) with a sequencing index for the sequencing library amplification. A fluorometric method (Qubit) were used to quantify the PCR products before the submission. All the samples were normalized to the same concentration and agarose gel electrophoresis was used to confirm the product size. Sequencing were performed at the MSU Genomics Core facility using MiSeq Reagent Kit v3 for 250 bp paired-end (PE) reads. The generated FASTQ format file was extracted, processed and clustered by sequence similarity using our custom software, ScaffoldSeq⁷.

In vivo EV library screening

In this study, 8- to 12-week-old female Balb/c mice were used for animal experiments. Animals were purchased from Jackson Laboratories and housed in the University Laboratory Animal Resources Facility at Michigan State University. All the experimental procedures for the animal study were performed with the approval of the Institutional Animal Care and Use Committee of Michigan State University. Approximately 5* 10⁹ Monobody Library EVs in EV storage buffer⁵ were injected into mice intravenously (iv). After 1-h of EVs administration, the mice were sacrificed and the visceral organs (heart, lung, liver, kidney, pancreas and spleen) were dissected and homogenized using Bulk Ceramic Beads 2.8mm (Fisher Scientific) and BeadBug 6 Microtube Homogenizer (Benchmark Scientific). The plasmid DNA was isolated from the organ homogenates using QIAprep Spin Miniprep Kit following a modified protocol for plasmid isolation from mammalian cells which was used as a template for the next round of library DNA preparation described above.^3,6 A series of EV library screenings was repeated 5 times to enrich targeting monobody sequences. The enriched variable loop fragments were re-cloned into the EV display construct for individual monobody assays.

In vivo and ex vivo imaging

Animal was administered IP (intraperitoneal) with 10 ug/g CTZ and after 5 min, injected approximately 5* 10⁹ Monobody Library EVs co-labeled with NanoLuc via tail vein. The isoflurane sedated animal was imaged in IVIS for luminescence every 2 minutes by 30 minutes. After in vivo imaging, the mice were sacrificed and the following visceral organs were dissected and placed on a transparent sheet: heart, lungs, liver, kidneys, pancreas and spleen. Ex vivo images of BLI were taken by IVIS. Table 1. qPCR probes.

Results

EV-based monobody display screening strategy and design

We adopted our previous EV-surface display design and DNA constructs.^2,8 We incorporated a monobody library to generate the EV-display monobody library DNA to transfect HEK293 cells to generate an EV-Monobody library. EV-Monobody library is isolated from the conditioned media and quantified for the numbers and pDNA loading prior to the downstream applications. Briefly EV-monobody library is used either to treat cells or to inject into animals and cells/organs are collected to extract DNA after the incubation or circulation. Monobody-coding region is PCR amplified and re-cloned into the EV-display backbone. The entire process was repeated 5 times to evaluate the enrichment of the targeting monobodies as illustrated in the FIG. 5.

HEK293T-derived monobody library EVs display Monobody protein on their surfaces and package and protect pDNA

For quality control purposes, each batch of EV library was characterized by NTA and qPCR to measure the quantity of EVs and encapsulated pDNA. The peak sizes of EVs were consistently around 120-140 nm in size (FIG. 6A,B) measured by NTA. To ensure free-floating and non-bound pDNA elimination, pDNA quantification was performed on the DNA isolated from EVs post-Dnasel treatment. Thus, FIG. 6C shows the distribution of EV-packaged monobody-coding pDNA per EV, revealing the average packaging efficacy was around 1.5 copies per particle in our library preparation. Monobody protein display was confirmed by the super-resolution microscopy analysis (ONI nanoimager), by co-localization of EV markers (CD63, CD9 and CD81) along with HA-tag fused to the monobody fragment (FIG. 6D). EV-based monobody library screening enriched cell type-specific monobody sequences along with the high binder control in vitro

First, we ran the time-course experiments using A431 EGFR+ cells treated with EGFR- specific monobody and non-binder control to identify the optimal EV-treatment duration of 30 min by comparing the pDNA uptake to the recipient cells. To screen the EV-monobody library, 1% each of EGFR high-binder monobody positive control (E626 (SEQ ID NO: 39)) and nonbinder monobody negative control (RDG) were added to the library to the transfection mixture. The cells were treated for 30 min with the monobody library EVs following washes and DNA extraction for sequencing. We monitored six lines of 5 individual screenings by collecting samples and sequencing each line. Non-binder RDG monobody was consistently depleted throughout the screening from the sequencing data (FIG. 7A), which was validated by qPCR (FIG. 7C), while EGFR targeting monobody (E626) was enriched with slight variability in the sequence analysis (FIG. 7B) but found to be stably enriched 10-fold compared to the initial library (P0) by the qPCR analysis (FIG. 7C). We further identified the enrichment of E626 variants and novel clones, suggesting the enrichment of competitively high-affinity monobody selection (FIG. 7D,E).

Novel variants were enriched from a naive monobody library. In FIG. 7C, each dot represents a unique monobody variant. The size of the dot is proportional to how many times the sequence appeared across all replicates. RDG variants include the negative control sequence and any sequences that were within two mutations of RDG (SEQ ID NO: 40). Similarly, the E626 variants include any sequence that was within two mutations of the positive control E626 sequence. Positive growth rate indicates that the variant was enriched during iterative rounds of panning while negative growth rates indicate depletion. Enrichment growth rate is calculated using: Final count = initial count * Exp(enrichment growth rate * time). Binding assay confirmed the monobody with high binding affinity to the target cells in vitro

Next, we have re-cloned the five individual high-binder candidates (Clones 1-5) into EV-display clone and test the binding using bioluminescence binding assay using EVs display monobody and NanoLuc on their surface. The binding of two (Clones 3 and 4) monobodies and the positive control monobody (E626) showed higher binding affinity in the statistically significant level (FIG. 8).

To determine the optimal EV circulation time, we administrated the library EVs to mice via the tail vein, euthanatized the animals, and isolated pDNA from the liver, kidney, pancreas, and spleen after 1, 4, and 24 hours to identify the optimal time for the circulation. The pDNA recovery was consistent or decreased after 4 or 24 hours, depending on the organ. Thus, we have decided to use 1-hour circulation for the library screening. In addition, we confirmed the pDNA recovery and monobody fragment amplification from the heart, lung, kidney, liver, pancreas, and spleen.

Pancreas-enriched monobody library EVs accumulate in the pancreas.

We compared the biodistribution of the initial monobody library EVs (PO) to the pool of pancreas-enriched monobody library EVs (pan-P5: pooled monobody-library pDNA after five rounds of 3 independent lines in vivo screening) using BLI in vivo and ex vivo. Biodistribution of EV-library in 30 min sequential imaging shows the distinct pattern of EV accumulation (FIG. 9A). Further, ex vivo imaging shows the accumulation of pancreas- enriched monobody library EVs in the pancreas compared to the original library, providing solid evidence of enrichment of the pancreas monobody selection by the screening system (FIG. 9B).

References

1 Gaspar, N. et al. Evaluation of NanoLuc substrates for bioluminescence imaging of transferred cells in mice. J Photochem Photobiol B 216, 112128, doi:10.1016/j.jphotobiol.2021.112128 (2021).

2 Komuro, H., Aminova, S., Lauro, K , Woldring, D. & Harada, M. Design and Evaluation of Engineered Extracellular Vesicle (EV)-Based Targeting for EGFR- Overexpressing Tumor Cells Using Monobody Display. Bioengineering (Basel) 9, doi:10.3390/bioengineering9020056 (2022).

3 Komuro, H. et al. Engineering Extracellular Vesicles to Target Pancreatic Tissue In Vivo. Nanotheranostics 5, 378-390, doi:10.7150/ntno.54879 (2021).

4 Woldring, D. R., Holec, P. V., Zhou, H. & Hackel, B. J. High-Throughput Ligand Discovery Reveals a Sitewise Gradient of Diversity in Broadly Evolved Hydrophilic Fibronectin Domains. PLoS One 10, e0138956, doi: 10.1371/journal.pone.0138956 (2015).

5 Kawai-Harada, Y., El Itawi, H., Komuro, H. & Harada, M. Evaluation of EV Storage Buffer for Efficient Preservation of Engineered Extracellular Vesicles. Ini J Mol Sci 24, doi:10.3390/ijms241612841 (2023).

6 Isolation of plasmid DNA from mammalian cells using QIAprep kit. QIAGEN News 2, 11 (1995).

7 Woldring, D. R , Holec, P. V. & Hackel, B. I. ScaffoldSeq: Software for characterization of directed evolution populations. Proteins 84, 869-874, doi:10.1002/prot.25040 (2016). 8 Komuro, H. el al. Engineering Extracellular Vesicles to Target Pancreatic Tissue <i>In Vivo</i>. Nanotheranostics 5, 378-390, doi:10.7150/ntno.54879 (2021).

Claims

We claim:

1. A library composition comprising a plurality of extracellular vesicles (EVs) comprising a plurality of surface-expressed peptides of interest and a plurality of nucleic acids encoding the surface expressed peptides of interest.

2. The composition of claim 1, wherein the plurality of EVs comprises greater than 10 distinct peptides of interest.

3. The composition of claim 2, wherein the plurality of EVs comprises 10, 100, 1000, 10⁴, 10⁵, or 10⁶ distinct peptides of interest.

4. The composition of any one of the preceding claims, wherein the plurality of surface- expressed peptides of interest comprises an affinity reagent selected from the group consisting of a monobody, an antibody, an antibody mimetic, a nanobody, and a ligand.

5. The composition of claim 4, wherein at least one of the surface-expressed peptides of interest is a monobody or a ligand.

6. The composition of claim 5, where in the monobody is E626.

7. The composition of claim 5, wherein the ligand is p88.

8. The composition of any one of the preceding claims, wherein the surface-expressed peptide of interest comprises a peptide linked to lactadherin C1C2.

9. The composition of claim 8, wherein the linker is at least 10 amino acids long.

10. The composition of any one of the preceding claims, wherein the plurality of EVs further comprise an imaging agent.

11. The composition of any one of the preceding claims, wherein the nucleic acid encoding the surface-expressed peptide of interest is a plasmid or minicircle.

12. The composition of claim 11, wherein the plasmid is greater than 3 kB.

13. A method of gene delivery to a cell or tissue comprising contacting the cell or tissue with any one of the compositions of claims 1-12.

14. A method of screening for affinity reagents, the method comprising:

(a) contacting a plurality of cells with the library composition of any one of claims 1- 13;

(b) harvesting the portion of the library composition bound or delivered to the cells in step (a);

(c) extracting DNA from the portion of the library composition harvested in step (b); and

(d) amplifying the nucleic acid encoding the surface-expressed peptide from the prepared DNA of step (c) to produce a first screened nucleic acid pool.

15. The method of claim 14, further comprising:

(e) generating a first screen library composition from the first screened nucleic acid pool of step (d) by (i) using the first screened nucleic acid pool to transfect cells and (ii) harvesting EVs from the cells; and

(f) repeating steps (a)-(d) using the first screen library composition to produce a second screened nucleic acid pool.

16. The method of claim 15, further comprising repeating steps (a) through (d) after step (f).

17. The method of claim 16, wherein steps (a) through (f) are repeated at least five times.

18. The method of any one of claims 14-17, further comprising sequencing the screened nucleic acid pool.

19. The method of any one of claims 14-18, further comprising generating EVs using the screened nucleic acid pool. 0. The method of claim 19, wherein the EVs are used for delivering an agent to a cell.

21. The method of claim 20, wherein the agent is selected from the group consisting of small molecule, chemotherapeutic, peptide, polypeptide, and enzyme.

22. The method of claim 20 or 21, wherein the cell is selected from the group consisting of pancreatic cell, kidney cell, spleen cell, liver cell, brain cell, and tumor cell.

23. The method of any one of claims 14-22, wherein the cell is contacted in vitro.

24. The method of any one of claims 14-22, wherein the cell is contacted in vivo.

25. The method of claim 24, wherein the plasmid DNA is recovered from a tissue.

26. The method of claim 25, wherein the tissue is selected from the group comprising liver, heart, lung, brain, kidney, pancreas, and spleen.

27. A method of making the library composition of any one of claims 1-13, the method comprising:

(a) amplifying the library backbone from a DNA template comprising a polynucleotide encoding lactadherin C1C2;

(b) amplifying a polynucleotide encoding the peptide of interest;

(c) joining the products of step (a) and (b) to create the library construct;

(d) introducing the library construct of step (c) into bacterial cells;

(e) harvesting the library construct from the bacterial cells;

(f) introducing the library construct into eukaryotic cells; and

(g) harvesting the EVs from the eukaryotic cells, wherein the harvested EVs are the library composition.

28. The method of claim 27, wherein the bacterial cells are E. coli cells.

29. The method of any one of claims 27-28, wherein step (d) further comprises selecting for bacterial cells comprising a selectable marker.

30. The method of claim 29, wherein the selectable marker is ampicillin.

31. The method of any one of claims 27-30, wherein step (c) is performed via Seamless Cloning Ligation Cell Extract (SLiCE).

32. The method of any one of claims 27-31, wherein the eukaryotic cells are HEK293T cells.

33. A monobody expressed on the surface of the EVs harvested in step (g) of the method of any one of claims 27-32.