Single-chain trimer PMHC-i antigen-presenting vesicles as preventative and therapeutic vaccines for cancer and infection

Abstract

Provided herein are antigen-presenting vesicles and compositions of antigen-presenting vesicles presenting antigenic peptides as preventative and therapeutic vaccines for cancer and infection through potent and selective activation of antigen-specific T cells.

Description

WSGR Docket No. 48295-718.601SINGLE-CHAIN TRIMER PMHC-I ANTIGEN-PRESENTING VESICLES AS PREVENTATIVE AND THERAPEUTIC VACCINES FOR CANCER AND INFECTIONCROSS REFERENCE

[0001] This application claims the benefit of the U.S. Provisional Application No. 63 / 688,753 filed August 29, 2024, which is incorporated herein by reference in its entirety.SEQUENCE LISTING

[0002] The instant application contains a Sequence Listing in electronic format. The Sequence Listing is provided as a file entitled 48295-718.601_SL, created on August 8, 2025, which is 528,858 bytes in size. The information in the electronic format of the Sequence Listing is incorporated by reference in its entirety.INCORPORATION BY REFERENCE

[0003] All publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.BRIEF SUMMARY

[0004] Disclosed herein is an antigen-presenting vesicle (APV) comprising a first fusion protein that comprises a single-chain trimer peptide major histocompatibility complex class I (SCT- pMHC-I) molecule and a second fusion protein that comprises a costimulatory ligand molecule, wherein the first or second fusion protein is expressed at a valency of at least about 100 copies on a surface of the antigen-presenting vesicle (APV).

[0005] Disclosed herein is a method of selectively activating antigen-specific T cells in culture or in vivo with antigen-presenting vesicles (APVs).

[0006] Disclosed herein is a method of inducing protective immunity against cancer or viral or bacterial infection in a subject in need thereof comprising administering an antigen-presenting vesicle (APV) disclosed herein or a composition disclosed herein.

[0007] Disclosed herein is a method of preventing or treating cancer or viral infection in a subject in need thereof comprising: (a) obtaining T cells from the subject; (b) contacting the T cells ex vivo with an antigen-presenting vesicle (APV) disclosed herein or a composition disclosed herein to activate the T cells; and (c) administering the activated and expanded T cells from step (b) to the subject.WSGR Docket No. 48295-718.601BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The novel features of the invention are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings of which:

[0009] Figure 1A and IB illustrates (A) a schematic of the vesicle display vector with oligomerization domains and (B) a list of oligomerization domains utilized in the vesicle display vector.

[0010] Figure 2 illustrates schematics of VSVG-based monomeric vector for vesicle display. VSV-G post-fusion D4 trimerization domain is not included in the vector.

[0011] Figure 3A-3C illustrates production of vesicles displaying monomeric peptide on (A) VLPs with viral genomes, (B) VLPs without viral genomes, and (C) extracellular vesicles (EV).

[0012] Figure 4A-4C illustrates the characterization of monomeric ACE2-VM-MVPs, including (A) quantitative Western-blot analyses of ACE2-MVPs. (B) Western-blot analyses of ACE2-VM-MVPs in the reducing and non-reducing conditions. (C) Viral neutralization activity of the ACE2-VM-MVPs determined in a pseudovirus neutralization assay.

[0013] Figure 5 illustrates schematics of leucine-zipper-based vector for dimeric vesicle display. The displayed peptide is fused leucine-zipper dimeric domain, VSV-G transmembrane domain, and VSV-G cytosolic tail.

[0014] Figure 6A-6C illustrates production of vesicles displaying dimeric peptide on (A) VLPs with viral genomes, (B) VLPs without viral genome, and (C) EVs.

[0015] Figure 7A-7C illustrates characterization of dimeric ACE2-LZ-MVPs. (A) Quantitative Western-blot analyses of ACE2-LZ-MVPs. (B) Western-blot analyses of ACE2-LZ-MVPs in the reducing and non-reducing conditions. (C) Viral neutralization activity of the ACE2-LZ-MVPs determined in a pseudovirus neutralization assay.

[0016] Figure 8 illustrates schematics of D4-based trimeric vector for vesicle display. VSV-G post-fusion D4 trimerization domain linked to the VSV-G transmembrane and cytosolic domains in the vector.

[0017] Figure 9 illustrates schematics of Dengue E protein (DE)-based trimeric vector for vesicle display. Dengue E protein-based trimerization domain linked to the VSV-G transmembrane and cytosolic domains in the vector.

[0018] Figure 10 illustrates schematics of Foldon-based trimeric vector for vesicle display. Foldon protein-based trimerization domain linked to the VSV-G transmembrane and cytosolic domains in the vector.WSGR Docket No. 48295-718.601

[0019] Figure 11A-11C illustrates production of vesicles displaying dimeric peptide on (A) VLPs with viral genomes, (B) VLPs without viral genome, and (C) EVs.

[0020] Figure 12A-12C illustrates the characterization of trimeric ACE2-MVPs, including (A) Quantitative Western-blot analyses of trimeric ACE2-MVPs. (B) Western-blot analyses of trimeric ACE2-MVPs in the reducing and non-reducing conditions. (C) Viral neutralization activity of the trimeric ACE2-MVPs determined in a pseudovirus neutralization assay.

[0021] Figure 13A-13C illustrates characterization of ACE2-MVPs with various oligomerization domains including (A) Quantitative Western-blot analyses of ACE2-MVPs. (B) Western-blot analyses of ACE2-MVPs under reducing and non-reducing conditions. (C) Viral neutralization activity of ACE2-MVPs determined in a pseudovirus neutralization assay.

[0022] Figure 14 illustrates schematics of neuraminidase stem-based tetrameric vector for type II transmembrane protein display on vesicles. Neuraminidase stem-based tetrameric domain linked to an optional additional oligomerization domain and an extracellular domain of a type II transmembrane protein.

[0023] Figure 15A-15C illustrate production of vesicles displaying tetrameric type II peptide on (A) VLPs with viral genomes, (B) VLPs without viral genome, and (C) EVs.

[0024] Figure 16A-16C illustrates characterization of DPP4-MVPs with variations of neuraminidase-stem tetrameric domains. (A) Quantitative Western-blot analyses of DPP4- MVPs. (B) Western-blot analyses of DPP4-MVPs under reducing and non-reducing conditions. (C) Viral neutralization activity of DPP4-MVPs determined in a pseudovirus neutralization assay.

[0025] Figure 17A-17C illustrates production of vesicles displaying mixed oligomer, combination peptide on (A) VLPs with viral genomes, (B) VLPs without viral genome, and (C) EVs.

[0026] Figure 18A-18C illustrates characterization of mixed oligomer, bi-specific Ab-MVPs displaying monomeric and trimeric scFv targeting SARS CoV-2 spike proteins. (A) Quantitative Western-blot analyses of Ab-MVPs. (B) Western-blot analyses of Ab-MVPs under reducing and non-reducing conditions. (C) Viral neutralization activity of Ab-MVPs determined in a pseudovirus neutralization assay.

[0027] Figure 19A-19C illustrates characterization of VLPs displaying various peptides. (A) Western-blot analyses of VLPs displaying innate cell receptors. (B) Quantitative Western-blot analyses of VLPs displaying B cell receptors and cytokines. (C) Quantitative Western-blot analyses of VLPs displaying scFv antibodies.

[0028] Figure 20A-20D illustrates production and characterization of EVs displaying trimeric ACE2 targeting SARS CoV-2 spike proteins. (A) Production of ACE2-D4-EVs with D4-trimericWSGR Docket No. 48295-718.601 display vector.(B) Quantitative Western-blot analyses of ACE2-D4-EVs. (C) Particle size analyses of ACE2-D4-EVs using qNano. (D) Viral neutralization activity of H2A / ACE2-D4-EVs determined in a pseudovirus neutralization assay.

[0029] Figure 21A-21D illustrates schematics of Linel coiled-coil (Llcc)-based trimeric vector for vesicle display. (A). Schematics of LINE1 open reading frame protein 1. (B) The monomeric and trimeric structure of LINE 1 orfl coiled-coil peptide. (C) Design of Llcc proteinbased trimeric display vector with Llcc linked to the LI transmembrane and cytosolic domains in the vector.

[0030] Figure 22A-22C illustrates production of vesicles displaying Llcc-trimeric peptide on (A) VLPs with viral genomes, (B) VLPs without viral genome, and (C) EVs.

[0031] Figure 23A-23C illustrates production and characterization of VLPs displaying trimeric Llcc-ACE2 targeting SARS CoV-2 spike proteins. (A) Quantitative Western-blot analyses of ACE2-Llcc-MVPs. (B) Particle size analyses of ACE2-Llcc-MVPs using qNano. (C) Viral neutralization activity of ACE2-Llcc-MVPs determined in a pseudovirus neutralization assay.

[0032] Figure 24A-24C illustrates schematics depicting the molecular construction of singlechain trimer pMHC-I antigen-presenting vesicles (SCT-APVs). (A) T cells employ multiple costimulatory molecules to modulate T cell activation, proliferation, differentiation, and function. (B) Molecular components of the displayed molecules for TCR engagement. (C) Schematics of the displayed single-chain trimer (SCT), consisting of linked peptide, MHC-I, and P2m, and displayed type I and type II costimulatory (costim) molecules with corresponding oligomerization domains and membrane anchoring domain. “C” stands for the C-terminus of type I protein and “N” stands for the N-terminus of type II protein.

[0033] Figure 25 illustrates production of single-chain-trimer pMHC-I antigen-presenting vesicles (SCT-APVs) by displaying high-density oligomerized SCT-pMHC-I and combination of costimulatory molecules on VLPs or EVs.

[0034] Figure 26A-26C illustrate effects of costim subtraction on naive OT-1 T cell activation by EV-based ova-APVs displaying ova-SCT and combination of costim molecules. (A) A matrix of costim molecules displayed on ova-APVs. (B) Effects of ova-APVs displaying varied costim molecules on T cell activation. Naive OT1 T cells were stimulated with ova-APVs and analyzed by FACS to determine the up-regulation of CD69 and CD25 expression at day 1 post stimulation. (C) Representative FACS plots are shown.

[0035] Figure 27A-27C illustrates effects of costim addition on T cell activation by EV-based ova-APVs displaying ova-SCT and combination of costim molecules. (A) A matrix of costim molecules displayed on ova-APVs. All ova-APVs displayed ova-SCT, GITRL, and ICAM-1 butWSGR Docket No. 48295-718.601 with varied CD80, CD86, XO40L, and CD48. (B) Effects of ova-APVs displaying varied costim molecules on T cell activation. Pan OT1 and Pmel T cells, including both naive and antigen- experienced T cells were stimulated with ova-APVs and analyzed by FACS to determine the upregulation of CD69 and CD25 expression at day 1 post stimulation. (C) Representative FACS plots are shown.

[0036] Figure 28A-28B illustrates the production of ova-APVs displaying SCT-pMHC-I with oligomerization domains with varied degree of oligomerization. (A) Schematics of SCT-pMHC-I display vector with varied oligomerization domain. (B) The list of display vectors used to produce ova-APVs displaying ova-SCT, CD86, GITRL, and ICAM-1 and their corresponding oligomerization domains. Coiled-coil helix domains with varied degree of oligomerization were used to optimize SCT-pMHC-I display on EVs.

[0037] Figure 29A-29D illustrates the effects of oligomerization domain for SCT-pMHC-I display on the function of ova-APVs on T cell activation and proliferation. (A) FACS gating to distinguish the effects of ova-APVs on antigen-specific OT-1 T cells (Ly5.2) and non-specific T cells (Ly5.1). (B) Effects of ova-APVs displaying SCT-MHC-I with varied oligomerization domain on T cell activation. Naive OT1 and non-specific Ly5.1 cells were stimulated with various ova- APVs and analyzed by FACS to determine the up-regulation of CD69 and CD25 expression at day 2 post stimulation. Representative FACS plots are shown. (C) Effects of ova-APVs with varied oligomerization domain on T cell activation as indicated by the percent of CD69 positive cells or CD69 and CD25 double-positive cells among the stimulated T cell mixture. (D) Effects of ova- APVs with varied oligomerization domain on T cell proliferation as determined by the fold of T cell expansion or total cell increases.

[0038] Figure 30A-30C illustrates optimized EV- and VLP -based APVs with displayed high- density oligomerized SCT, CD80, ICAM1, and GITRL. (A) Schematics of the displayed SCT / D4, consisting of linked peptide, MHC-I, P2m and D4, and displayed CD80 / D4, ICAM1 / D4, and GITRL / NA costimulatory molecules. “C” stands for the C-terminus of type I protein and “N” stands for the N-terminus of type II protein. (B) Production of APVs by displaying high-density oligomerized SCT, CD80, ICAM1, and GITRL on VLPs or EVs.

[0039] Figure 31A-31E Illustrates production and characterization of APV(86IG) displaying high-density oligomerized SCT-pMHC-I and combination of chimeric costimulatory molecules, including CD86 / D4, GITRL / NA, and ICAM1 / D4. (A, B) Particle size analyses of (A) exosome- based ova-APV(86IG) and (B) VLP -based ova-APV(86IG) using qNano. (C) Quantitative westernblot analyses to determine the copies of ova-H2Kb SCT class I MHC molecules on EV- or VLP- based ova-APV(86IG). (D) Quantitative western-blot analyses to determine the copies ofWSGR Docket No. 48295-718.601CD86 / D4, GITRL / NA, and ICAM1 / D4 molecules on EV- or VLP-based ova-APV(86IG). (E) The copies of ova-H2Kb SCT, CD86 / D4, GITRL / NA, and ICAM1 / D4 on EV- or VLP-based ova- APV(86IG).

[0040] Figure 32A-32E illustrates potent and specific activation of ova-specific OT-1 T cells with EV- or VLP-based ova-APVs displaying high-density oligomerized ova-H2Kb SCT-pMHC-I and combination of chimeric costimulatory molecules, including CD86 / D4, GITRL / NA, and ICAM1 / D4. (A, B) EV- or VLP-based ova-APVs are as potent as anti-CD3 / CD28 dynabeads at activating OT1 T cells. (A) Representative FACS plots of CD69 and CD25 expression or (B) total percent of CD69 and CD25 positive cells at day 1 after ova-specific OT-1 T cells were activated with anti-CD3 / CD28 dynabeads, or EV-ova-APVs or VLP-ova-APVs. (C, D). EV-ova-APVs selectively activate ova-specific OT-1 T cells but not non-specific Ly5.1 T cells. (E) The ratio of fully-activated (CD69 and CD25 double-positive) and partially-activated T cells (CD69 positive only) among the EV-ova-APVs activated OT-1 and Ly5.1 T cells at different time point post stimulation.

[0041] Figure 33A-33C illustrates does-dependent, selective activation of ova-specific OT-1 T cells with EV-based ova-APVs displaying high density of ova-H2Kb SCT-pMHC-I, CD86, GITRL, and ICAMl.(A) Representative FACS plots of CD69 and CD25 expression at day 1, 2, and 4 after mixed ova-specific OT-1 T cells and LY5.1 T cells were activated with EV-ova-APVs at increasing particle to T cell ratios. (B, C). EV-ova-APVs preferentially activate ova-specific OT-1 T cells and not non-specific Ly5.1 T cells at various APV to T cell ratios. (B) Percent of CD69 positive cells or (C) proportion of CD69 single positive T cells vs the percent of CD69 / CD25 double-positive T cells among stimulated OT-1 T cells as determined by FACS analysis.

[0042] Figure 34A-34C illustrates does-dependent, selective activation of ova-specific OT-1 T cells with VLP-based ova-APVs displaying high density of ova-H2Kb SCT-pMHC-I, CD86, GITRL, and ICAM1. (A) Representative FACS plots of CD69 and CD25 expression at day 1, 2, and 4 after mixed ova-specific OT-1 T cells and LY5.1 T cells were activated with VLP-ova-APVs at increasing particle to T cell ratios. (B). VLP-ova-APVs preferentially activate ova-specific OT-1 T cells and not non-specific Ly5.1 T cells at various APV to T cell ratios. (B) Percent of CD69 positive cells or (C) Percent of CD69 single positive T cells vs percent of CD69 and CD25 doublepositive T cells among the stimulated OT-1 T cells as determined by FACS analysis.

[0043] Figure 35A-35D illustrates ova-APVs presenting variant ova-peptides have differential effects on antigen-specific T cell activation. (A) Variants of ova peptide presented on the EV-based ova-APVs, which display ova-H2Kb SCT-pMHC-I, CD86, GITRL, and ICAM-1. (B) Percent of CD69 single positive T cells and CD69 and CD25 double-positive cells among naive OT-1 T cellsWSGR Docket No. 48295-718.601 stimulated with variant ova-APVs as determined by FACS analysis. (C) Representative FACS plots of CD69 and CD25 expression on naive OT-1 T cells at day 2 after stimulation with ova-APVs presenting variant ova peptides. (D). Effects of ova-APVs presenting variant ova peptides on the proliferation of activated naive OT-1 T cells.

[0044] Figure 36A-36D illustrates mice immunized with APVs presenting antigenic peptides from melanoma-specific antigens are resistant to Bl 6F 10 challenge. (A) List of antigenic peptides from melanoma-associated antigen gplOO or neoantigen Hsf2_K72N presented on VLP -based APVs. (B) Schematic illustrating melanoma antigen- APV immunization and re-challenge study design. 5E10 particles for each melanoma antigen- APV was mixed and injected into mice intravenously following the immunization schedule depicted. Concurrently, lOpg of high molecular weight poly(I:C) (HMW-polylC) is also delivered through intramuscular injection. Immunized mice were challenged with wild-type tumor cells at day 35 post-primary immunization. VLP -based APVs displaying peptide-H2Kb SCT-pMHC-I, CD86, GITRL, and ICAM-1 were used in the study. (C, D) B16F0 melanoma tumor growth in (C) naive, unimmunized mice or (D) mice immunized with mixed melanoma-specific APVs. Number of tumor free mice are indicated..

[0045] Figure 37A-37C illustrates therapeutic treatment with antigen-specific APVs inhibits the growth of established tumors in the B16F0 mouse melanoma model. (A) Mice bearing ova- B16F0 tumors were treated with intravenously delivered ova-APVs at day 7 and day 8 or at day 14 post-tumor implantation. (B, C) Effects of ova-APV treatment on the growth of ova-B16F0 tumors are plotted in (B) the medium tumor sizes of each treatment group and (C) tumor sizes of individual mouse within each treatment group (n=5 / treatment group).

[0046] Figure 38A-38C illustrate the differential effects of variant ova-APVs in controlling the growth of established tumors in the ova-B16F10 mouse melanoma model. (A) Mice bearing ova- B16F10 tumors were treated with intravenously delivered variant ova-APVs at day 7 and day 8 and at day 21 post-tumor implantation. (B, C) Effects of variant ova-APV treatment on the growth of ova-B16F10 tumors are plotted as (B) the average tumor sizes of each treatment group and (C) individual mouse tumor growth curves within each treatment group (n=5). .

[0047] Figure 39A-39C illustrates the therapeutic treatment with antigen-specific APVs inhibits the growth of established tumors in the MC38 mouse tumor model. (A) Mice bearing ova- MC38 tumors were treated with intravenously delivered ova-APVs at day 7 and day 8 post-tumor implantation. (B, C) Effects of ova-APV treatment on the growth of ova-MC38 tumors are plotted as (B) the average tumor sizes of each treatment group and (C) tumor sizes of individual mice within each treatment group (n=5).WSGR Docket No. 48295-718.601

[0048] Figure 40A-40C illustrates production of off-the-shelf APVs with human HLA alleles. (A) Human HLA alleles may be used to create APVs matched to an individual patient’s HLA type. EURCAU, European Caucasian; AAFA, African American; NCHI, Chinese. (B) Production of APVs by displaying high-density oligomerized SCT, CD80, ICAM1, and GITRL on VLPs or EVs. (C) Schematics depicting APVs with displayed SCT-pMHCI, CD86 (or CD80), and GITRL. VSVG membrane anchoring domain and oligomerization domains, such as D4 trimeric domain or NA (Neuraminidase stem), are used to increase the local and global valency of display molecules on EVs and VLP.

[0049] Figure 41A-41B illustrates off-the-shelf SCT-APVs including (A) variants with single peptide antigens or (B) mixed antigenic peptides.

[0050] Figure 42A-42B illustrates in vivo and ex vivo T cell therapy with SCT-APVs. (A) Autologous T cell therapy with ex vivo expanded T cells. (B) Using antigen-specific APVs as preventative and therapeutic vaccines..

[0051] Figure 43A-43D illustrates combination SCT-APVs and anti-PDl / Ll therapy for cancer. (A) Anti-PDl / Ll immune checkpoint blockade is effective in eliminating tumors with activated anti-tumor T cells. (B) Anti-PDl / Ll immune checkpoint blockade is ineffective in eliminating tumors with tolerized anti-tumor T cells or few activated T cells. (C) In vivo activation of anti-tumor T cells with APVs may synergize with anti-PDl / Ll immune checkpoint blockade therapy. (D) In vivo selective activation of anti -tumor T cells with APVs may be essential to improve the long-term survival of cancer patients received anti-PDl / Ll or anti-CTLA-4 therapies.DETAILED DESCRIPTION

[0052] The present disclosure employs, unless otherwise indicated, conventional molecular biology techniques, which are within the skill of the art. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as is commonly understood by one of ordinary skill in the art.Definitions

[0053] Throughout this disclosure, various embodiments are presented in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of any embodiments. Accordingly, the description of a range should be considered to have specifically disclosed all the possible subranges as well as individual numerical values within that range to the tenth of the unit of the lower limit unless the context clearly dictates otherwise. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed subranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual valuesWSGR Docket No. 48295-718.601 within that range, for example, 1.1, 2, 2.3, 5, and 5.9. This applies regardless of the breadth of the range. The upper and lower limits of these intervening ranges may independently be included in the smaller ranges, and are also encompassed within the disclosure, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the disclosure, unless the context clearly dictates otherwise.

[0054] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of any embodiment. As used herein, the singular forms “a,” “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and / or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. As used herein, the term “and / or” includes any and all combinations of one or more of the associated listed items.

[0055] Unless specifically stated or obvious from context, as used herein, the term “about” in reference to a number or range of numbers is understood to mean the stated number and numbers + / - 10% thereof, or 10% below the lower listed limit and 10% above the higher listed limit for the values listed for a range.

[0056] The term “sequence identity” means that two polynucleotide sequences are identical (i.e., on a nucleotide-by-nucleotide basis) over the window of comparison. The term “percentage of sequence identity” is calculated by comparing two optimally aligned sequences over the window of comparison, determining the number of positions at which the identical nucleic acid base (e.g., A, T, C, G, U, or I) occurs in both sequences to yield the number of matched positions, dividing the number of matched positions by the total number of positions in the window of comparison (i.e., the window size), and multiplying the result by 100 to yield the percentage of sequence identity. Typically, techniques for determining sequence identity include comparing two nucleotide or amino acid sequences and the determining their percent identity. Sequence comparisons, such as for the purpose of assessing identities, may be performed by any suitable alignment algorithm, including but not limited to the Needleman-Wunsch algorithm (e.g., the EMBOSS Needle aligner), the BLAST algorithm (see, e.g., NCBI BLAST alignment tool), and the Smith-Waterman algorithm (see, e.g., the EMBOSS Water aligner). Optimal alignment may be assessed using any suitable parameters of a chosen algorithm, including default parameters. The “percent identity,” also referred to as “percent homology,” between two sequences may be calculated as the number of exact matches between two optimally aligned sequences divided by the length of the referenceWSGR Docket No. 48295-718.601 sequence and multiplied by 100. Percent identity may also be determined, for example, by comparing sequence information using the advanced BLAST computer program, including version 2.2.9, available from the National Institutes of Health. The BLAST program is based on the alignment method of Karlin and Altschul, Proc. Natl. Acad. Sci. USA 87:2264-2268 (1990) and as discussed in Altschul, et al., J. Mol. Biol. 215:403-410 (1990); Karlin and Altschul, Proc. Natl. Acad. Sci. USA 90:5873-5877 (1993); and Altschul et al., Nucleic Acids Res. 25:3389-3402 (1997). Briefly, the BLAST program defines identity as the number of identical aligned symbols (i.e., nucleotides or amino acids), divided by the total number of symbols in the shorter of the two sequences. The program may be used to determine percent identity over the entire length of the sequences being compared. Default parameters are provided to optimize searches with short query sequences, for example, with the blastp program. The program also allows use of an SEG filter to mask-off segments of the query sequences as determined by the SEG program of Wootton and Federhen, Computers and Chemistry 17: 149-163 (1993). High sequence identity generally includes ranges of sequence identity of approximately 80% to 100% and integer values there between.Artificial antigen-presenting cells

[0057] Selective activation of antigen-specific T cells underlies the immune defense against cancer and infection. Activated cytotoxic T lymphocytes (CTLs) can rapidly expand many orders of magnitude within days to eliminate mutated or infected somatic cells and while generating long- lasting immune memory to safeguard against disease recurrence. Antigen recognition by T cells involves the engagement of complex protein machinery on both antigen-presenting cells (APCs) and T cells. APCs, which include dendritic cells, macrophages, and B cells, play a critical role in T cell activation. APCs capture and process antigens into short peptides and then present some peptides on their surface as peptide-MHC (pMHC) complexes. Each APC typically presents a few foreign peptides among abundant tolerogenic self-peptides, all in the form of pMHC complexes. T cells have T cell receptors (TCRs) on their surface that recognize and bind to pMHC complexes on APCs. Notably, humans have billions of T cells, each with a unique TCR. TCR-pMHC engagement must be highly selective and sensitive to enable only a few antigen-specific T cells, among the billions of off-target T cells, to recognize and discriminate rare viral or cancer antigens from an overwhelming number of self-antigens. The difference between activating and non-activating peptides in pMHCs may be a single amino acid variation and a minor change in affinity. Nevertheless, T cells can be activated by as few as ten agonist peptides presented on APCs, illustrating the incredible sensitivity and specificity of TCR and pMHC recognition. Thus, one has to mimic the mechanisms APCs use to activate antigen-specific T cells therapeutically.WSGR Docket No. 48295-718.601Multivalent and combinatorial engagement between T cells and APCs

[0058] The multivalent engagement between TCR and pMHC complexes is vital to the sensitivity and specificity of T cells in antigen recognition. TCR-pMHC engagement leads to the formation of TCR-pMHC supramolecular activation complexes (SMAC) between the T cell and APC, termed immunological synapses. Immunological synapses are comprised of hundreds or thousands of copies of dozens of different proteins, including TCR subunits and signaling proteins, adhesion molecules, T cell co-stimulatory molecules, and inhibitor receptors on the T cell membrane. Formation of immunological synapses depends on the fluidity of the lipid bilayer membrane, which permits lateral movement of membrane proteins and dynamic clustering of dozens of proteins in multivalent format. In particular, T cell activation also requires a second signal provided by the interaction of co-stimulatory receptors on the T cell with their respective ligands on APCs. This co-stimulation follows the triggering event of TCR engagement with pMHC and a third inflammatory priming signal from cytokines. These signals act in concert to further control T cell activation, proliferation, and differentiation. Upon proper activation, T cells undergo clonal expansion, producing copies of themselves, and differentiate into effector T cells. These effector T cells can directly kill infected or cancerous cells (in the case of cytotoxic T cells) or help other immune cells respond to the infection (in the case of helper T cells). Errors in antigen recognition by T cells may lead to autoimmunity, where the immune system attacks the body's own cells, immunodeficiency, where the immune system cannot effectively respond to infections or cancer-causing mutations, or hypersensitivity, where the immune system over-reacts to benign foreign antigens. While the importance of therapeutically manipulating antigen-specific T cells is well-recognized, there are tremendous challenges given the complexity of TCR and pMHC engagement in terms of the number of proteins involved, the multivalent and aggregative features, and the cascades of signals following the recognition event.Artificial antigen-presenting systems

[0059] Artificial antigen-presenting systems, which include cell-based technology and acellular technologies, have been developed to induce antigen-specific T-cell responses in vitro and in vivo. Specifically, various vehicles, such as insect cells, mouse fibroblasts, human leukemia cell lines, magnetic beads, lipid nanoparticles, or cell-derived vesicles (i.e., exosomes and viral-like particles), have been engineered to recapitulate some of the essential aspects of the antigen-presenting machinery of APCs, including dendritic cells. Typically, the vectors were engineered to present pMHC complexes and co-stimulatory molecules as signals 1 and 2, respectively, for TCR engagement and T-cell activation. As an example of cell-based artificial antigen-presenting cells (aAPCs), K562-based aAPC cells, a human erythroleukemia cell line engineered to express HLA,WSGR Docket No. 48295-718.601CD80, and CD83, can effectively activate and expand antigen-specific T cells in culture. K562 cells also express ICAM-1 (CD54) and LFA-3 (CD58) adhesion molecules required to form effective immunologic synapses. K562-based aAPCs showed some efficacy in the ex vivo generation of antigen-specific T-cells for adoptive T-cell therapies to treat cancer and chronic virus infection in clinical settings. Cell-based aAPCs have largely been limited to ex vivo uses and have yet to be used for in vivo induction of T cell immunity.Deficiencies of current Artificial antigen-presenting systems

[0060] To overcome the deficiencies of the cell-based aAPCs, many attempts have been made to develop acellular aAPC systems based on synthetic beads of 5-6-micron diameter or cell-derived vesicles, such as exosomes and viral-like particles (VLPs). Specifically, pMHC and co-stimulatory signal-engaging molecules, including antibodies against CD28 or the CD80 / CD86 (the ligand of CD28), are conjugated to the beads / synthetic liposomes or genetically displayed on VLPs. However, these acellular aAPCs often have weaker ex vivo T-cell expansion activities than cellbased aAPCs. Notably, none of these aAPC strategies have aimed to recapitulate the multivalent characteristics exhibited by T cell-DC interactions. They often display less than 100 copies of the combinatorial signals required to induce proper T cell activation, including pMHC and costimulatory ligands. They are yet to be proven to have strong effects in vivo as therapeutics or preventative vaccines for cancer. Intriguingly, DC-derived exosomes, which maintain the essential TCR-engaging apparatus from corresponding DCs, can weakly stimulate antigen-specific T cells in vitro but more potently prime antigen-specific immunity in vivo. However, DC-derived exosomes seem to stimulate T cell response in vivo primarily through indirect antigen transfer to bystander APCs. They cannot prime naive T cells without the help of endogenous APCs. Moreover, early clinical testing of DC-derived exosomes as cancer vaccines has shown limited efficacy in advanced cancer patients. In summary, concurrent acellular aAPC systems still need much improvement to fully recapitulate the immunological synapse engagement process.Engineered antigen-presenting vesicles

[0061] Described herein, in some embodiments, are engineered antigen-presenting vesicles particles comprising a first fusion protein that comprises a single-chain trimer peptide major histocompatibility complex class I (SCT-pMHC-I) molecule and a second fusion protein that comprises a costimulatory ligand molecule. In some embodiments, are engineered antigen- presenting vesicle particles comprising a first fusion protein that comprises a single-chain trimer peptide major histocompatibility complex class I (SCT_pMHC-I) molecule and a second fusion protein that comprises a costimulatory ligand molecule, wherein the first or second fusion protein is expressed at a valency of at least about 100 copies on a surface of the antigen-presenting vesicle. InWSGR Docket No. 48295-718.601 some embodiments, the SCT-pMHC-I molecule binds specifically to an antigen-specific T cell receptor. In some embodiments, the SCT-pMHC-I molecule comprises an antigenic peptide, P-2- microglobulin (P2m), and an MHC-I alpha chain. In some embodiments, the antigenic peptide binds to the MHC-I alpha chain and is fused to P2m. In some embodiments, the pMHC-I molecule comprises a single-chain trimer (SCT) peptide MHC-I molecule (SCT-pMHC-I). In some embodiments, the antigenic peptide is derived from a disease-associated antigen, a model antigen, a viral antigen, a mutated tumor antigen, a tumor-specific antigen, a differentiated antigen, an overexpressed antigen, or a neoantigen. In some embodiments, the costimulatory ligand molecule binds specifically to a T-cell surface costimulatory receptor. In some embodiments, the antigen- presenting vesicle is capable of activating or stimulating a cognate antigen-specific T cell.

[0062] Antigen-presenting vesicles are described herein. In some embodiments, the multivalent particle is an antigen-presenting vesicle, Various antigen-presenting vesicles are contemplated herein. In some embodiments, the antigen-presenting vesicle is recombinant. In some embodiments, the antigen-presenting vesicle does not comprise viral genetic material. In some embodiments, the antigen-presenting vesicle is a viral-like particle or virus-like particle. As used herein, viral-like particle and virus-like particle interchangeably. In some embodiments, the viral- like particle does not comprise viral genetic material. In some embodiments, the antigen-presenting vesicle is an extracellular vesicle. In some embodiments, the antigen-presenting vesicle is an exosome. In some embodiments, the antigen-presenting vesicle is an ectosome.

[0063] Antigen-presenting vesicles as described herein, in some embodiments, comprise a first fusion protein, wherein the fusion protein is expressed at multiple copies on a surface of the antigen-presenting vesicle. In some embodiments, antigen-presenting vesicles as described herein comprise a second fusion protein, wherein the fusion protein is expressed at multiple copies on a surface of the antigen-presenting vesicle. In some embodiments, the first or second fusion protein of the antigen-presenting vesicle is expressed at a valency of at least or about 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 75, 100, 125, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, 400, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950, 1000, 1250, 1500, 1750, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, or more than 1000 copies on a surface of the multivalent particle. In some embodiments, the fusion protein is expressed at a valency of at least or about 100 to about 250, about 100 to about 500, about 100 to about 1000, about 100 to about 2000, about 100 to about 4000, about 100 to about 6000, about 100 to about 8000, about 250 to about 500, about 250 to about 1000, about 250 to about 2000, about 250 to about 4000, about 250 to about 6000, about 250 to about 8000, about 500 to about 1000, about 500 to about 2000, about 500 to about 4000, about 500 to about 6000, about 500 to about 8000, about 1000 to about 2000, about 1000 to about 4000,WSGR Docket No. 48295-718.601 about 1000 to about 6000, about 1000 to about 8000, about 2000 to about 4000, about 2000 to about 6000, about 2000 to about 8000, about 4000 to about 6000, about 4000 to about 8000, or about 6000 to about 8000 copies on a surface of the antigen-presenting vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 100 copies on a surface of the antigen-presenting vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 250 copies on a surface of the antigen-presenting vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 500 copies on a surface of the antigen-presenting vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 1000 copies on a surface of the antigen-presenting vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 2000 copies on a surface of the antigen-presenting vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 4000 copies on a surface of the antigen-presenting vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 6000 copies on a surface of the antigen-presenting vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 8000 copies on a surface of the antigen-presenting vesicle.

[0064] In some embodiments, the antigen-presenting vesicle is a viral-like particle. Viral-like particles as described herein, in some embodiments, comprise a first fusion protein, wherein the fusion protein is expressed at multiple copies on a surface of the viral-like particle. In some embodiments, viral-like particles as described herein comprise a second fusion protein, wherein the fusion protein is expressed at multiple copies on a surface of the antigen-presenting vesicle In some embodiments, the first or second fusion protein of the viral-like particle is expressed at a valency of at least or about 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 75, 100, 125, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, 400, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950, 1000, 1250, 1500, 1750, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, or more than 1000 copies on a surface of the viral-like particle. In some embodiments, the fusion protein is expressed at a valency of at least or about 100 to about 250, about 100 to about 500, about 100 to about 1000, about 100 to about 2000, about 100 to about 4000, about 100 to about 6000, about 100 to about 8000, about 250 to about 500, about 250 to about 1000, about 250 to about 2000, about 250 to about 4000, about 250 to about 6000, about 250 to about 8000, about 500 to about 1000, about 500 to about 2000, about 500 to about 4000, about 500 to about 6000, about 500 to about 8000, about 1000 to about 2000, about 1000 to about 4000, about 1000 to about 6000, about 1000 to about 8000, about 2000 to about 4000, about 2000 to about 6000, about 2000 to about 8000, about 4000 to about 6000, about 4000 to about 8000, or about 6000 to about 8000 copies on a surface of the viral-like particle. In some embodiments, the fusion protein is expressed at a valency of at least or about 100 copies on aWSGR Docket No. 48295-718.601 surface of the viral-like particle. In some embodiments, the fusion protein is expressed at a valency of at least or about 250 copies on a surface of the viral-like particle. In some embodiments, the fusion protein is expressed at a valency of at least or about 500 copies on a surface of the viral-like particle. In some embodiments, the fusion protein is expressed at a valency of at least or about 1000 copies on a surface of the viral-like particle. In some embodiments, the fusion protein is expressed at a valency of at least or about 2000 copies on a surface of the viral-like particle. In some embodiments, the fusion protein is expressed at a valency of at least or about 4000 copies on a surface of the viral-like particle. In some embodiments, the fusion protein is expressed at a valency of at least or about 6000 copies on a surface of the viral-like particle. In some embodiments, the fusion protein is expressed at a valency of at least or about 8000 copies on a surface of the viral- like particle.

[0065] In some embodiments, the antigen-presenting vesicle is a extracellular vesicle. Extracellular vesicles as described herein, in some embodiments, comprise a first fusion protein, wherein the fusion protein is expressed at multiple copies on a surface of the extracellular vesicle. In some embodiments, extracellular vesicles as described herein comprise a second fusion protein, wherein the fusion protein is expressed at multiple copies on a surface of the antigen-presenting vesicle In some embodiments, the first or second fusion protein of the extracellular vesicle is expressed at a valency of at least or about 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 75, 100, 125, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, 400, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950, 1000, 1250, 1500, 1750, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, or more than 1000 copies on a surface of the extracellular vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 100 to about 250, about 100 to about 500, about 100 to about 1000, about 100 to about 2000, about 100 to about 4000, about 100 to about 6000, about 100 to about 8000, about 250 to about 500, about 250 to about 1000, about 250 to about 2000, about 250 to about 4000, about 250 to about 6000, about 250 to about 8000, about 500 to about 1000, about 500 to about 2000, about 500 to about 4000, about 500 to about 6000, about 500 to about 8000, about 1000 to about 2000, about 1000 to about 4000, about 1000 to about 6000, about 1000 to about 8000, about 2000 to about 4000, about 2000 to about 6000, about 2000 to about 8000, about 4000 to about 6000, about 4000 to about 8000, or about 6000 to about 8000 copies on a surface of the extracellular vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 100 copies on a surface of the extracellular vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 250 copies on a surface of the extracellular vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 500 copies on a surface of the extracellular vesicle. In someWSGR Docket No. 48295-718.601 embodiments, the fusion protein is expressed at a valency of at least or about 1000 copies on a surface of the extracellular vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 2000 copies on a surface of the extracellular vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 4000 copies on a surface of the extracellular vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 6000 copies on a surface of the extracellular vesicle. In some embodiments, the fusion protein is expressed at a valency of at least or about 8000 copies on a surface of the extracellular vesicle.

[0066] In some embodiments, the antigen-presenting vesicle is an exosome. Exosomes as described herein, in some embodiments, comprise a first fusion protein, wherein the fusion protein is expressed at multiple copies on a surface of the exosome. In some embodiments, exosomes as described herein comprise a second fusion protein, wherein the fusion protein is expressed at multiple copies on a surface of the antigen-presenting vesicle In some embodiments, the first or second fusion protein of the exosome is expressed at a valency of at least or about 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 75, 100, 125, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, 400, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950, 1000, 1250, 1500, 1750, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, or more than 1000 copies on a surface of the exosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 100 to about 250, about 100 to about 500, about 100 to about 1000, about 100 to about 2000, about 100 to about 4000, about 100 to about 6000, about 100 to about 8000, about 250 to about 500, about 250 to about 1000, about 250 to about 2000, about 250 to about 4000, about 250 to about 6000, about 250 to about 8000, about 500 to about 1000, about 500 to about 2000, about 500 to about 4000, about 500 to about 6000, about 500 to about 8000, about 1000 to about 2000, about 1000 to about 4000, about 1000 to about 6000, about 1000 to about 8000, about 2000 to about 4000, about 2000 to about 6000, about 2000 to about 8000, about 4000 to about 6000, about 4000 to about 8000, or about 6000 to about 8000 copies on a surface of the exosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 100 copies on a surface of the exosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 250 copies on a surface of the exosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 500 copies on a surface of the exosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 1000 copies on a surface of the exosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 2000 copies on a surface of the exosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 4000 copies on a surface of the exosome. In some embodiments, the fusion protein isWSGR Docket No. 48295-718.601 expressed at a valency of at least or about 6000 copies on a surface of the exosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 8000 copies on a surface of the exosome.

[0067] In some embodiments, the antigen-presenting vesicle is an ectosome. Ectosomes as described herein, in some embodiments, comprise a first fusion protein, wherein the fusion protein is expressed at multiple copies on a surface of the ectosome. In some embodiments, ectosomes as described herein comprise a second fusion protein, wherein the fusion protein is expressed at multiple copies on a surface of the antigen-presenting vesicle In some embodiments, the first or second fusion protein of the ectosome is expressed at a valency of at least or about 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 75, 100, 125, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, 400, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950, 1000, 1250, 1500, 1750, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10000, or more than 1000 copies on a surface of the ectosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 100 to about 250, about 100 to about 500, about 100 to about 1000, about 100 to about 2000, about 100 to about 4000, about 100 to about 6000, about 100 to about 8000, about 250 to about 500, about 250 to about 1000, about 250 to about 2000, about 250 to about 4000, about 250 to about 6000, about 250 to about 8000, about 500 to about 1000, about 500 to about 2000, about 500 to about 4000, about 500 to about 6000, about 500 to about 8000, about 1000 to about 2000, about 1000 to about 4000, about 1000 to about 6000, about 1000 to about 8000, about 2000 to about 4000, about 2000 to about 6000, about 2000 to about 8000, about 4000 to about 6000, about 4000 to about 8000, or about 6000 to about 8000 copies on a surface of the ectosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 100 copies on a surface of the ectosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 250 copies on a surface of the ectosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 500 copies on a surface of the ectosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 1000 copies on a surface of the ectosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 2000 copies on a surface of the ectosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 4000 copies on a surface of the ectosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 6000 copies on a surface of the ectosome. In some embodiments, the fusion protein is expressed at a valency of at least or about 8000 copies on a surface of the ectosome.

[0068] Described herein, in some embodiments, are antigen-presenting vesicles that comprise an oligomerization domain within the displayed fusion proteins. In some embodiments, the first orWSGR Docket No. 48295-718.601 second fusion protein of the antigen-presenting vesicle comprises an oligomerization domain. In some embodiments, the oligomerization domain is a dimerization domain. In some embodiments, the dimerization domain comprises a leucine zipper dimerization domain. In some embodiments, the dimerization domain comprises a coiled coil dimerization domain. In some embodiments, the coiled coil dimerization domain comprises an Apple4 domain. In some embodiments, the coiled coil dimerization domain comprises Factor XI. In some embodiments, the coiled coil dimerization domain comprises Fibrinogen. In some embodiments, the oligomerization domain is a trimerization domain. In some embodiments, the trimerization domain comprises a post-fusion oligomerization domain of viral surface protein. In some embodiments, the trimerization domain comprises a D4 post-fusion trimerization domain of VSV-G protein. In some embodiments, the trimerization domain comprises a Dengue E protein post-fusion trimerization domain. In some embodiments, the trimerization domain comprises a foldon trimerization domain. In some embodiments, the trimerization domain comprises a coiled coil trimerization domain. Non-limiting examples of coiled coil trimerization domains include LINE1 retrotranspson orflp coiled coil, collagenXV coiled coil, lung surfactant protein D, tetranectin coiled coil, and mannose binding coiled coil domains. In some embodiments, the trimerization domain comprises a LINE1 retrotransposon coiled coil domain (Llcc). In some embodiments, the trimerization domain comprises a collagen XV coiled coil domain (CXVcc). In some embodiments, the trimerization domain comprises a tetranectin coiled coil domain (TNTNcc). In some embodiments, the oligomerization domain is a tetramerization domain. In some embodiments, the tetramerization domain comprises an influenza neuraminidase stem domain. In some embodiments, the oligomerization domain is a hexamerization domain. In some embodiments, the hexamerization domain comprises a cc-hex domain. In some embodiments, the oligomerization domain is a heterohexamerization domain. In some embodiments, the heterohexamerization domain comprises a L24D domain. In some embodiments, the heterohexamerization domain comprises a L24H domain. In some embodiments, the oligomerization domain is a heptamerization domain. In some embodiments, the heptamerization domain comprises a cc-hept domain. In some embodiments, the oligomerization domain is an octamerization domain. In some embodiments, the octamerization domain comprises a wazOct domain.

[0069] Described herein, in some embodiments, are antigen-presenting vesicles comprising a first or second fusion protein with a transmembrane anchoring polypeptide for vesicle targeting. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of a Vesicular Stomatitis virus glycoprotein (VSV-G). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of a Vesicular StomatitisWSGR Docket No. 48295-718.601 virus glycoprotein (VSV-G). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of a Dengue E protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of a Dengue E protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of influenza Hemagglutinin (HA). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of influenza Hemagglutinin (HA). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of HIV surface glycoprotein GP120 or GP41. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of HIV surface glycoprotein GP120 or GP41. In some embodiments, the transmembrane domain comprises the transmembrane polypeptide of measles virus surface glycoprotein hemagglutinin (H) protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of measles virus surface glycoprotein hemagglutinin (H) protein. In some embodiments, the transmembrane domain comprises the transmembrane polypeptide of measles virus hemagglutinin envelope protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of measles virus envelope glycoprotein fusion protein. In some embodiments, the transmembrane domain comprises the transmembrane polypeptide of measles virus envelope glycoprotein fusion protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of measles virus envelope glycoprotein fusion protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of influenza Neuraminidase (NA). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of influenza Neuraminidase (NA). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of spike protein SI. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of spike protein SI. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of spike protein S2. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of spike protein S2. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of Sindbis virus envelope (SINDBIS) protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of Sindbis virus envelope protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of lentiviral RD114 envelope protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of lentiviral RD114 envelope protein. In some embodiments, the transmembraneWSGR Docket No. 48295-718.601 polypeptide comprises the transmembrane domain of lentiviral BaEV envelope protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of lentiviral BaEV envelope protein. In some embodiments, the transmembrane polypeptide comprises the wild-type transmembrane domain of costimulatory ligand molecules. In some embodiments, the transmembrane polypeptide comprises the wild-type transmembrane domain and cytosolic domain of costimulatory ligand molecules.Major histocompatibility complex I and fl-2 microglobulin

[0070] Described herein, in some embodiments, are antigen-presenting vesicles comprising a first fusion protein that comprises a single-chain trimer peptide major histocompatibility complex class (I) (SCT-pMHC-I). Described herein, in some embodiments, are antigen-presenting vesicles comprising a first fusion protein that comprises a peptide major histocompatibility complex class (I) and a second fusion protein. In some embodiments, the SCT-pMHC-I molecule comprises a MHC-I alpha chain. In some embodiments, the MHC-I alpha chain comprises a mammalian MHC- I allele. In some embodiments, the MHC-I alpha chain comprises a mouse MHC-I allele. In some embodiments, the MHC-I alpha chain comprises a human leukocyte antigen (HLA) allele. In some embodiments, the MHC-I alpha chain comprises an amino acid sequence with at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, at least 99.9%, or at least 100% sequence identity to a mouse or human MHC-I allele according to any one of SEQ ID NOs: 1-20. In some embodiments, the MHC-I alpha chain comprises an amino acid sequence with at least 75% sequence identity to a mouse or human MHC- I allele according to any one of SEQ ID NOs: 1-20. In some embodiments, the MHC-I alpha chain comprises an amino acid sequence with at least 80% sequence identity to a mouse or human MHC- I allele according to any one of SEQ ID NOs: 1-20. In some embodiments, the MHC-I alpha chain comprises an amino acid sequence with at least 85% sequence identity to a mouse or human MHC- I allele according to any one of SEQ ID NOs: 1-20. In some embodiments, the MHC-I alpha chain comprises an amino acid sequence with at least 90% sequence identity to a mouse or human MHC- I allele according to any one of SEQ ID NOs: 1-20. In some embodiments, the MHC-I alpha chain comprises an amino acid sequence with at least 95% sequence identity to a mouse or human MHC- I allele according to any one of SEQ ID NOs: 1-20. In some embodiments, the MHC-I alpha chain comprises an amino acid sequence with at least 98% sequence identity to a mouse or human MHC- I allele according to any one of SEQ ID NOs: 1-20. In some embodiments, the MHC-I alpha chainWSGR Docket No. 48295-718.601 comprises an amino acid sequence with at least 90% sequence identity to a mouse or human MHC- I allele according to any one of SEQ ID NOs: 1-20.TABLE 1: Exemplary MHC-I and HLA SequencesWSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601

[0071] In some embodiments, disclosed herein are antigen-presenting vesicles comprising a first fusion protein that comprises SCT-pMHC-I comprise a peptide and a MHC-I alpha chain and a P-2 microglobulin (P2m), and a second fusion protein. In some embodiments, the first fusion protein comprises a human P2m. In some embodiments, the first fusion protein comprises a mouse P2m. In some embodiments, the P2m comprises an amino acid sequence with at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, at least 99.9%, or at least 100% sequence identity to an amino acid sequence of human and mouse P2m according to SEQ ID NO: 21 or 22. In some embodiments, the P2m comprises an amino acid sequence with at least 75% sequence identity to an amino acid sequence of human and mouse P2m according to SEQ ID NO: 21 or 22. In some embodiments, the P2m comprises an amino acid sequence with at least 80% sequence identity to an amino acid sequence of human and mouse P2m according to SEQ ID NO: 21 or 22. In some embodiments, the P2m comprises an amino acid sequence with at least 85% sequence identity to an amino acid sequence of human and mouse P2m according to SEQ ID NO: 21 or 22. In some embodiments, the P2m comprises an amino acid sequence with at least 90% sequence identity to an amino acid sequence of human and mouse P2m according to SEQ ID NO: 21 or 22. In some embodiments, the P2m comprises an amino acid sequence with at least 95% sequence identity to an amino acid sequence of human and mouse P2m according to SEQ ID NO: 21 or 22. In some embodiments, the P2m comprises anWSGR Docket No. 48295-718.601 amino acid sequence with at least 98% sequence identity to an amino acid sequence of human and mouse P2m according to SEQ ID NO: 21 or 22. In some embodiments, the P2m comprises an amino acid sequence with at least 99% sequence identity to an amino acid sequence of human and mouse P2m according to SEQ ID NO: 21 or 22.

[0072] In some embodiments, disclosed herein is an antigen-presenting vesicle comprising a first fusion protein that comprises a single-chain trimer pMHC-I and a second fusion protein. In some embodiments, the first fusion protein comprises Antigenic Peptides

[0073] In some embodiments, antigen-presenting vesicle comprises a first fusion protein comprising a single-chain trimer peptide major histocompatibility complex (SCT-pMHC-I) molecule and a second fusion protein. In some embodiments, the SCT-pMHC-I molecules comprises an antigenic peptide, P-2 microglobulin, and an MHC-I alpha chain. In some embodiments, the antigenic peptide is derived from a disease-associated antigen, a model antigen, a viral antigen, a mutated tumor antigen, a tumor-specific antigen, a differentiated antigen, an overexpressed antigen, or a neoantigen. In some embodiments, the antigenic peptide is derived from a cancer antigen. In some embodiments, the antigen peptide is derived from a cancer antigen as disclosed in TABLE 2. In some embodiments, the antigenic peptide is selected from the group comprising ovalbumin (Ova), gplOO, Hsf2-K72N, an HBV or HCV viral protein, an E6 or E7 of human papillomavirus, a large T and small T protein of Merkel polyomavirus, KRAS, BRAF, BRAC1, BRAC2, NYE-ESO-1, MAGE-A1, MAGE-A3, CEA, gpl00 / Pmell7, PAP, HER2 / neu, GPC3, and WTI. In some embodiments, the antigenic peptide comprises an amino acid according to any one of SEQ ID NOs: 59-66. In some embodiments, the antigenic peptide comprises an amino acid sequence with at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, at least 99.9%, or at least 100% sequence identity to an amino acid sequence according to SEQ ID NOs: 59-66.WSGR Docket No. 48295-718.601TABLE 2: Exemplary Cancer antigensNeoantigen Genomic variants, transcriptomic All tumors variants, proteomic variants, viral-derived neoantigensTABLE 3: Exemplary model antigenic peptidesWSGR Docket No. 48295-718.601

[0074] In some embodiments, the antigenic peptide comprises a viral antigenic peptide. In some embodiments, the antigenic peptide comprises a viral antigen viral listed in TABLE 4. In some embodiments, the antigenic peptide comprises an amino acid sequence according to any one of SEQ ID NOs: 149-218. In some embodiments, the antigenic peptide comprises an amino acid sequence with at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, at least 99.9%, or at least 100% sequence identity to an amino acid sequence according to SEQ ID NOs: 149-218.TABLE 4. Viral antigenic peptidesWSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601

[0075] In some embodiments, the antigenic peptide comprises an antigenic peptide derived from human cancer antigens. In some embodiments, the antigenic peptide derived from human cancer antigens comprises a mutated antigen, a tumor-specific antigen, a differentiated antigen, or an over-expressed antigen. In some embodiments, the antigenic peptide comprises an antigen peptide derived from human cancer antigens listed in TABLE 5. In some embodiments, the antigenic peptide comprises an amino acid sequence according to any one of SEQ ID NOs: 219- 544. In some embodiments, the antigenic peptide comprises an amino acid sequence with at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, at least 99.9%, or at least 100% sequence identity to an amino acid sequence according to SEQ ID NOs: 219-544.TABLE 5. Antigenic peptides derived from human cancer antigensWSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601

[0076] In some embodiments, the oligomerization domain comprises an amino acid sequence disclosed in TABLES 2, 4, and 5, or an amino acid sequence that is substantially identical to an amino acid sequence in TABLES 2, 4, and 5 (e.g. 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% sequence identity). In some instances, the oligomerization domain comprises an amino acid sequence comprising at least a portion having at least or about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130 amino acid sequences of any sequence according to TABLES 2, 4, and 5.

[0077] In some embodiments, the antigenic peptide comprises one or more listed in TABLES 2, 4, and 5. In some embodiments, the antigenic peptide comprises an amino acid sequence according to any one of SEQ ID NOs: 59-66, and 149-544.

[0078] In some embodiments, the antigen-presenting vesicle comprises a first fusion protein that comprises a single-chain trimer peptide major histocompatibility complex class I (SCT-pMHC- I) molecule and a second fusion protein. In some embodiments, the pMHC-I molecule comprises at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least 10, or more than ten separate fusion proteins. In some embodiments, the at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least 10, or more than ten separate fusion proteins comprises a different antigenic peptide.

[0079] In some instances, the antigenic peptide comprises an amino acid sequence comprising at least a portion having at least or about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140,150, 160, 170, 180, 190, 200, 210, 220, 230, 240, 250, 260, 270, 280, 290, 300, 310, 320, 330, 340,350, 360, 370, 380, 390, 400, 410, 420, 430, 440, 450, 460, 470, 480, 490, 500, 510, 520, 530, 540,550, 560, 570, 580, 590, 600, 610, 620, 630, 640, 650, 660, 670, 680, 690, 700, 710, 720, 730, 740,750, 760, 770, 780, 790, 800 or more than 800 amino acids of SEQ ID NOs: 59-66 and 149-544.WSGR Docket No. 48295-718.601Costimulatory ligands

[0080] In some embodiments, disclosed herein is an antigen-presenting vesicle comprising a first fusion protein that comprises a single-chain trimer peptide major histocompatibility complex I (SCT-pMHC-I) molecule and a second fusion protein that comprises a costimulatory ligand molecule. In some embodiments, the costimulatory ligand molecule comprises GITRL (GITR ligand), CD80, CD86, OX40L (0X40 ligand), CD48, CD30L (CD30 ligand), ICOSL (ICOS ligand), CD70, CD40, CD40L (CD40 ligand), 4-1BBL (4- IBB ligand), TLIA, TIM4, SLAM, CD48, CD58, CD155, CD112, LIGHT, or ICAM-1. In some embodiments, the costimulatory ligand molecules comprises an amino acid sequence with at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, at least 99.9%, or at least 100% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58. In some embodiments, the costimulatory ligand molecule comprises an amino acid sequence with at least 75% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58. In some embodiments, the costimulatory ligand molecule comprises an amino acid sequence with at least 80% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58. In some embodiments, the costimulatory ligand molecule comprises an amino acid sequence with at least 90% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58. In some embodiments, the costimulatory ligand molecule comprises an amino acid sequence with at least 95% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58. In some embodiments, the costimulatory ligand molecule comprises an amino acid sequence with at least 98% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58.TABLE 6: Exemplary costimulatory ligand molecule sequencesWSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601

[0081] In some embodiments, the costimulatory ligand comprises an amino acid sequence disclosed in TABLE 6, or an amino acid sequence that is substantially identical to an amino acid sequence in TABLE 6 (e.g., 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% sequence identity). In some instances, the oligomerization domain comprises an amino acid sequence comprising at least a portion having 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180,190, 200, 210, 220, 230, 240, 250, 260, 270, 280, 290, 300, 310, 320, 330, 340, 350, 360, 370, 380,390, 400, 410, 420, 430, 440, 450, 460, 470, 480, 490, 500, 510, 520, 530, 540, 550, 560, 570, 580,590, 600, 610, 620, 630, 640, 650, 660, 670, 680, 690, 700, 710, 720, 730, 740, 750, 760, 770, 780,790, 800 or more than 800 amino acids of any sequence according to TABLE 6.

[0082] In some embodiments, the second fusion protein of displayed costimulatory ligand molecules comprises at least two, three, four, five, six, seven, eight, nine, ten, or more than ten separate fusion proteins. In some embodiments, the second fusion protein comprises at least two, three, four, five, six, seven, eight, nine, ten, or more than ten separate fusion proteins, wherein theWSGR Docket No. 48295-718.601 at least two, three, four, five, six, seven, eight, nine, ten, or more than ten separate fusion proteins each comprises a different costimulatory ligand molecule.

[0083] In some embodiments, the costimulatory ligand molecule comprises at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, or more than ten costimulatory ligand molecules. In some embodiments, the costimulatory ligand molecule comprises at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, or more than ten costimulatory ligand molecules selected from the group comprising GITRL (GITR ligand), CD80, CD86, OX40L(OX40 ligand), CD48, CD30L (CD30 ligand), ICOSL (ICOS ligand), CD70, CD40, CD40L (CD40 ligand), 4-1BBL (4-1BB ligand), TLIA, TIM4, SLAM, CD48, CD58, CD155, CD112, LIGHT, and ICAM. In some embodiments, the costimulatory molecule comprises GITRL (GITR ligand). In some embodiments, the costimulatory ligand molecule comprises GITRL (GITR ligand), CD80, CD86, OX40L, CD48, CD30L, and ICAM-1. In some embodiments, the costimulatory molecule comprises GITRL (GITR ligand) and ICAM-1. In some embodiments, the costimulatory ligand molecule comprises GITRL(GITR ligand), OX40L, and ICAM-1. In some embodiments, the costimulatory ligand molecule comprises GITRL(GITR ligand), CD80, and ICAM-1. In some embodiments, the costimulatory ligand molecule comprises GITRL(GITR ligand), CD86, and ICAM-1.Oligomerization Domains

[0084] In some embodiments, the first or second fusion protein of the antigen-presenting vesicle comprises an oligomerization domain. In some embodiments, the oligomerization domain is a dimerization domain. In some embodiments, the dimerization domain comprises a leucine zipper dimerization domain. In some embodiments, the oligomerization domain is a trimerization domain. In some embodiments, the trimerization domain comprises a post-fusion oligomerization domain of viral surface protein. In some embodiments, the trimerization domain comprises a D4 post-fusion trimerization domain of VSV-G protein. In some embodiments, the trimerization domain comprises a Dengue E protein post-fusion trimerization domain. In some embodiments, the trimerization domain comprises a foldon trimerization domain. In some embodiments, the trimerization domain comprises a coiled coil trimerization domain. In some embodiments, the trimerization domain comprises a LINE1 retrotransposon coiled coil domain (Llcc). In some embodiments, the trimerization domain comprises a collagen XV coiled coil domain (CXVcc). In some embodiments, the trimerization domain comprises a tetranectin coiled coil domain (TNTNcc). In some embodiments, the oligomerization domain is a tetramerization domain. In some embodiments, the tetramerization domain comprises an influenza neuraminidase stem domain. In some embodiments,WSGR Docket No. 48295-718.601 the oligomerization domain is a hexamerization domain. In some embodiments, the hexamerization domain comprises a cc-hex domain. In some embodiments, the oligomerization domain is a heterohexamerization domain. In some embodiments, the heterohexamerization domain comprises a L24D domain. In some embodiments, the heterohexamerization domain comprises a L24H domain. In some embodiments, the oligomerization domain is a heptamerization domain. In some embodiments, the heptamerization domain comprises a cc-hept domain. In some embodiments, the oligomerization domain is an octamerization domain. In some embodiments, the octamerization domain comprises a wazOct domain.TABLE 7. Exemplary Oligomerization Domain SequencesWSGR Docket No. 48295-718.601

[0085] In some embodiments, the oligomerization domain comprises an amino acid sequence disclosed in TABLE 7, or an amino acid sequence that is substantially identical to an amino acid sequence in TABLE 7 (e.g., 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% sequence identity). In some instances, the oligomerization domain comprises an amino acid sequence comprising at least a portion having at least or about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130 amino acid sequences of any sequence according to TABLE 7.WSGR Docket No. 48295-718.601

[0086] In some embodiments, the oligomerization domain comprises an amino acid sequence that has at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, at least 99.9%, or at least 100% sequence identity to an amino acid sequence according to SEQ ID NOs: 87-101. In some embodiments, the oligomerization domain comprises an amino acid sequence that has at least 70% sequence identity to an amino acid sequence according to SEQ ID NOs: 87-101. In some embodiments, the oligomerization domain comprises an amino acid sequence that has at least 80% sequence identity to an amino acid sequence according to SEQ ID NOs: 87-101. In some embodiments, the oligomerization domain comprises an amino acid sequence that has at least 90% sequence identity to an amino acid sequence according to SEQ ID NOs: 87-101. In some embodiments, the oligomerization domain comprises an amino acid sequence that has at least 95% sequence identity to an amino acid sequence according to SEQ ID NOs: 87-101.Transmembrane Polypeptides

[0087] Described herein, in some embodiments, are antigen-presenting vesicles comprising a first or second fusion protein with a transmembrane polypeptide anchoring peptide. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of a Vesicular Stomatitis virus glycoprotein (VSV-G). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of a Vesicular Stomatitis virus glycoprotein (VSV-G). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of a Dengue E protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of a Dengue E protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of influenza Hemagglutinin (HA). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of influenza Hemagglutinin (HA). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of HIV surface glycoprotein GP120 or GP41. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of HIV surface glycoprotein GP120 or GP41. In some embodiments, the transmembrane domain comprises the transmembrane polypeptide of measles virus surface glycoprotein hemagglutinin (H) protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolicWSGR Docket No. 48295-718.601 domain of measles virus surface glycoprotein hemagglutinin (H) protein. In some embodiments, the transmembrane domain comprises the transmembrane polypeptide of measles virus hemagglutinin envelope protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of measles virus envelope glycoprotein fusion protein. In some embodiments, the transmembrane domain comprises the transmembrane polypeptide of measles virus envelope glycoprotein fusion protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of measles virus envelope glycoprotein fusion protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of influenza Neuraminidase (NA). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of influenza Neuraminidase (NA). In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of spike protein SI. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of spike protein SI. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of spike protein S2. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of spike protein S2. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of Sindbis virus envelope (SINDBIS) protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of Sindbis virus envelope protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of lentiviral RD114 envelope protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of lentiviral RD114 envelope protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain of lentiviral BaEV envelope protein. In some embodiments, the transmembrane polypeptide comprises the transmembrane domain and cytosolic domain of lentiviral BaEV envelope protein. In some embodiments, the transmembrane polypeptide comprises the wild-type transmembrane domain of costimulatory ligand molecules. In some embodiments, the transmembrane polypeptide comprises the wild-type transmembrane domain and cytosolic domain of costimulatory ligand molecules.TABLE 8. Exemplary Transmembrane Domain SequencesWSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601

[0088] In some embodiments, the transmembrane domain comprises an amino acid sequence disclosed in TABLE 8, or an amino acid sequence that is substantially identical to an amino acid sequence in TABLE 8 (e.g., 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% sequence identity). In some instances, the transmembrane domain comprises an amino acid sequence comprising at least a portion having at least or about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130 amino acid sequences of any sequence according to TABLE 8.

[0089] In some embodiments, the transmembrane domain comprises an amino acid sequence that has at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, atWSGR Docket No. 48295-718.601 least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, at least 99.9%, or at least 100% sequence identity to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane domain comprises an amino acid sequence that has at least 75% sequence identity to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane domain comprises an amino acid sequence that has at least 80% sequence identity to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane domain comprises an amino acid sequence that has at least 85% sequence identity to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane domain comprises an amino acid sequence that has at least 90% sequence identity to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane domain comprises an amino acid sequence that has at least 95% sequence identity to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane domain comprises an amino acid sequence that has at least 98% sequence identity to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane domain comprises an amino acid sequence that has at least 99% sequence identity to an amino acid sequence according to SEQ ID NOs: 67-86.

[0090] Described herein, in some embodiments, are antigen-presenting vesicles comprising one or two fusion peptides and further comprising a transmembrane polypeptide. In some embodiments, the transmembrane polypeptide anchors the fusion protein to a lipid bilayer of the multivalent particle. In some embodiments, the transmembrane polypeptide comprises a spike glycoprotein, a mammalian membrane protein, an envelope protein, a nucleocapsid protein, or a cellular transmembrane protein. In some embodiments, the transmembrane polypeptide comprises VSVG, spike protein SI, spike protein S2, Sindbis virus envelope (SINDBIS) protein, hemagglutinin envelope protein from measles virus, envelope glycoprotein of measles virus fusion (F) protein, RD114, BaEV, GP41, or GP120. In some embodiments, the transmembrane polypeptide comprises VSVG. In some embodiments, the VSVG comprises full length VSVG or a truncated VSVG. In some embodiments, the VSVG comprises a transmembrane domain and cytoplasmic tail. In some embodiments, the hemagglutinin envelope protein from measles virus is a variant of the hemagglutinin envelope protein from measles virus. In some instances, the variant is HCA18.

[0091] In some embodiments, the transmembrane polypeptide comprises an amino acid sequence of at least 75% sequence homology to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane polypeptide comprises an amino acid sequence of at least 80% sequence homology to an amino acid sequence according to SEQ ID NOs: 67-86. InWSGR Docket No. 48295-718.601 some embodiments, the transmembrane polypeptide comprises an amino acid sequence of at least 85% sequence homology to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane polypeptide comprises an amino acid sequence of at least 90% sequence homology to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane polypeptide comprises an amino acid sequence of at least 95% sequence homology to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane polypeptide comprises an amino acid sequence of at least 98% sequence homology to an amino acid sequence according to SEQ ID NOs: 67-86. In some embodiments, the transmembrane polypeptide comprises an amino acid sequence of at least 99% sequence homology to an amino acid sequence according to SEQ ID NOs: 67-86.

[0092] In some instances, the transmembrane polypeptide comprises an amino acid sequence comprising at least a portion having at least or about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, amino acids of SEQ ID NOs: 67-86.Peptide major histocompatibility complex I molecule and costimulatory ligand molecule compositions

[0093] Described herein, in some embodiments, are antigen-presenting vesicles comprising a first fusion protein that comprises a single-chain trimer peptide major histocompatibility complex class I (SCT-pMHC-I) molecule and a second fusion protein that comprises a costimulatory ligand molecule. In some embodiments, the first fusion protein comprises a single-chain trimer pMHC-I molecule binds specifically to an antigen-specific T cell receptor. In some embodiments, the first fusion peptide comprises a single-chain trimer pMHC-I molecule comprising an antigenic peptide, P-2 microglobulin, and MHC-I alpha chain. In some embodiments, the first fusion protein comprises a SCT-pMHC-I molecule that binds specifically to a cognate antigen-specific T cell receptor, a transmembrane domain, and an oligomerization domain In some embodiments, the first fusion protein comprises a SCT-pMHC-I molecule that binds specifically to a cognate antigenspecific T cell receptor, a transmembrane domain, and an oligomerization domain, wherein the SCT-pMHC-I molecule comprises a fusion protein comprising an antigenic peptide, P2m , an MHC-I alpha chain, an oligomerization domain, and a transmembrane anchoring domain. In some embodiments, the first fusion protein comprises a SCT-pMHC-I molecule that binds specifically to a cognate antigen-specific T cell receptor, a transmembrane domain, and an oligomerization domain, wherein the SCT-pMHC-I molecule comprises a fusion protein comprising an antigenic peptide, P2m , an MHC-I alpha chain, an oligomerization domain, and a transmembrane anchoring domain, wherein the first fusion protein is expressed at a valency of at least about 100 copies on a surface of an antigen-presenting vesicle (APV). In some embodiments, the antigenic peptide bindsWSGR Docket No. 48295-718.601 to the pMHC-I alpha chain and is fused to P2m. In some embodiments, the first fusion protein comprises an amino acid sequence with at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, at least 99.9%, or at least 100% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 1-22, 102, and 110-148. In some embodiments, the first fusion protein comprises an amino acid sequence with at least 75% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 1-22, 102, and 110-148. In some embodiments, the first fusion protein comprises an amino acid sequence with at least 80% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 1-22, 102, and 110-148. In some embodiments, the first fusion protein comprises an amino acid sequence with at least 85% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 1-22, 102, and 110-148. In some embodiments, the first fusion protein comprises an amino acid sequence with at least 90% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 1-22, 102, and 110-148. In some embodiments, the first fusion protein comprises an amino acid sequence with at least 95% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 1-22, 102, and 110-148. In some embodiments, the first fusion protein comprises an amino acid sequence with at least 98% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 1-22, 102, and 110-148.

[0094] Described herein, in some embodiments, are antigen-presenting vesicles comprising a first fusion protein that comprises a single-chain trimer peptide major histocompatibility complex class I (SCT-pMHC-I) molecule and a second fusion protein that comprises a costimulatory ligand molecule. In some embodiments, the second fusion protein comprises a costimulatory ligand molecule that binds specifically to a T cell surface costimulatory receptor, an oligomerization domain, and a transmembrane domain. In some embodiments, the second fusion protein comprises a costimulatory ligand molecule that binds specifically to a T cell surface costimulatory receptor, an oligomerization domain, and a transmembrane domain, wherein the second fusion protein is expressed at a valency of at least about 100 copies on a surface of an APV. In some embodiments, the second fusion protein comprises at least two separate fusion proteins. In some embodiments, the second fusion protein comprises at least two separate fusion proteins, wherein the at least two separate fusion protein each comprise a different costimulatory ligand molecule. In some embodiments, the costimulatory ligand molecule comprises GITRL(GITR ligand), CD80, CD86, OX40L (XO40 ligand), CD48, CD30L (CD30 ligand), ICOSL (ICOS ligand), CD70, CD40,WSGR Docket No. 48295-718.601CD40L, 4-1BBL (4-1BB ligand), TLIA, TIM4, SLAM, CD48, CD58, CD155, CD112, LIGHT, or ICAM1. In some embodiments, the second fusion protein comprises an amino acid sequence with at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, at least 99.5%, at least 99.9%, or at least 100% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58, 103-109 and 545-555. In some embodiments, the second fusion protein comprises an amino acid sequence with at least 75% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58, 103-109 and 545-555. In some embodiments, the second fusion protein comprises an amino acid sequence with at least 80% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58, 103-109 and 545-555. In some embodiments, the second fusion protein comprises an amino acid sequence with at least 85% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58, 103-109 and 545-555. In some embodiments, the second fusion protein comprises an amino acid sequence with at least 90% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58, 103-109 and 545-555. In some embodiments, the second fusion protein comprises an amino acid sequence with at least 95% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58, 103-109 and 545-555. In some embodiments, the second fusion protein comprises an amino acid sequence with at least 98% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58, 103-109 and 545-555.TABLE 9. Exemplary SCT or Costimulatory ligand moleculesWSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601

[0095] In some embodiments, the first or second fusion protein comprises an amino acid sequence disclosed in TABLE 9, or an amino acid sequence that is substantially identical to an amino acid sequence in TABLE 9 (e.g., 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% sequence identity). In some instances, the transmembrane domain comprises an amino acid sequence comprising at least a portion having at least or about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700, 750, or 800 amino acid sequences of any sequence according to TABLE 9.

[0096] In some embodiments, the first fusion protein comprises a SCT-HLA molecule and an oligomerization domain. In some embodiments, the first fusion protein comprises a SCT-HLA molecule and a trimerization oligomerization domain. In some embodiments, the first fusion protein comprises a SCT-HLA molecule and a VSV-G protein D4 trimerization domain. In some embodiments, the first fusion protein comprises a SCT-HLA molecule and a Llcc trimerization domain.TABLE 10: Human SCT-HLA with D4 or Llcc trimeric domainWSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No.48295-718.601- Ill -WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601WSGR Docket No. 48295-718.601

[0097] In some embodiments, the first fusion protein comprises an amino acid sequence disclosed in TABLE Y, or an amino acid sequence that is substantially identical to an amino acid sequence in TABLE Y (e.g., 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% sequence identity). In some instances, the transmembrane domain comprises an amino acid sequence comprising at least a portion having at least or about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700, 750, or 800 amino acid sequences of any sequence according to TABLE Y.

[0098] In some embodiments, disclosed herein is an antigen-presenting vesicle comprising a first fusion protein and a second fusion protein, wherein the first or second fusion protein is expressed at least about 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1250, 1500, 1750, 2000, 2500, 3000, 4000, 5000, or 10000 copies on the surface of the antigen-presenting vesicle. In some embodiments, disclosed herein is an antigen-presenting vesicle comprising a first fusion protein and a second fusion protein, wherein the first or second fusion protein is expressed at least about 100 copies on the surface of the antigen-presenting vesicle.

[0099] In some embodiments, when the first or second fusion protein is expressed on the surface of the antigen-presenting vesicle, the oligomerization domain is outside of the antigen- presenting vesicle. In some embodiments, when the fusion protein is expressed on the surface of the antigen-presenting vesicle, the oligomerization domain is inside of the antigen-presenting vesicle. In some embodiments, when the fusion protein is expressed on the surface of theWSGR Docket No. 48295-718.601 multivalent particle, the oligomerization domain is inside of the multivalent particle and adjacent to the transmembrane domain.

[0100] In some embodiments, domains of the first fusion protein can be arranged from the N- terminus to the C-terminus in the following order: antigenic peptide, P2m, MHC-I oligomerization domain, transmembrane domain, and cytosolic domain. In some embodiments, domains of the second fusion protein can be arranged from the N-terminus to the C-terminus in the following orders: (a) costimulatory ligand molecule, oligomerization domain, transmembrane domain, cytosolic; or (b) transmembrane domain, costimulatory ligand molecule.Compositions for Generation of Antigen-presenting vesicles

[0101] Described herein, in some embodiments, are compositions comprising an antigen- presenting vesicle comprising a first fusion protein that comprises a single-chain trimer pMHC-I molecule and a second fusion protein that comprises a costimulatory ligand molecule. In some embodiments, the compositions comprise a first nucleic acid sequence encoding the antigen- presenting vesicle described herein. In some embodiments, the compositions comprise a first nucleic acid sequence encoding the antigen-presenting vesicle described herein and an excipient. In some embodiments, disclosed herein, is a first nucleic acid sequence encoding the first fusion protein of the antigen-presenting vesicle. In some embodiments, disclosed herein is a second nucleic acid sequence encoding the second fusion protein of the antigen-presenting vesicle.

[0102] Compositions for generating antigen-presenting vesicles, in some embodiments, further comprise a second nucleic acid sequence that encodes one or more viral proteins. In some embodiments, the one or more viral proteins is a lentiviral protein, a retroviral protein, an adenoviral protein, or combinations thereof. In some embodiments, the one or more viral proteins comprises gag, pol, pre, tat, rev, or combinations thereof.

[0103] Compositions for generating antigen-presenting vesicles, in some embodiments, further comprise a third nucleic acid sequence that encodes a replication incompetent viral genome, a reporter, a therapeutic molecule, or combinations thereof. In some embodiments, the viral genome is derived from vesicular stomatitis virus, measles virus, Hepatitis virus, influenza virus, or combinations thereof.

[0104] In some embodiments, the reporter protein is a fluorescent protein or an enzyme. Exemplary reporter genes include, but are not limited to, acetohydroxyacid synthase (AHAS), alkaline phosphatase (AP), beta galactosidase (LacZ), beta glucuronidase (GUS), chloramphenicol acetyltransferase (CAT), green fluorescent protein (GFP), red fluorescent protein (RFP), yellow fluorescent protein (YFP), cyan fluorescent protein (CFP), cerulean fluorescent protein, citrine fluorescent protein, orange fluorescent protein , cherry fluorescent protein, turquoise fluorescentWSGR Docket No. 48295-718.601 protein, blue fluorescent protein, horseradish peroxidase (HRP), luciferase (Luc), nopaline synthase (NOS), octopine synthase (OCS), luciferase, and derivatives thereof. Methods to determine modulation of a reporter gene are well known in the art, and include, but are not limited to, fluorometric methods (e.g., fluorescence spectroscopy, Fluorescence Activated Cell Sorting (FACS), fluorescence microscopy), and antibiotic resistance determination. In some embodiments, the reporter is a fluorescent protein. In some embodiments, the fluorescent protein is green fluorescent protein. In some embodiments, the reporter protein emits green fluorescence, yellow fluorescence, or red fluorescence. In some embodiments, the reporter is an enzyme. In some embodiments, the enzyme is P-galactosidase, alkaline phosphatase, P-lactamase, or luciferase.

[0105] In some embodiments, the therapeutic molecule is an inflammatory cytokine, an innate modulating molecule, a dendritic cell stimulating molecule, or a combination thereof. In some embodiments, the therapeutic molecule is an inflammatory cytokine. In some embodiments, the inflammatory cytokine comprises IL-2, IL-12, interferon alpha, interferon beta, TNF-alpha, or TNF-beta. In some embodiments, the therapeutic molecule is an innate modulating molecule. In some embodiments, the innate modulating molecules comprises Treml / 2, LILRB family proteins, HMGB1, or HSP90. In some embodiments, the therapeutic molecule is a dendritic cell stimulating molecule. In some embodiments, the dendritic cell stimulating molecules comprises Flt3L, CCL21, or CD40L.

[0106] In some embodiments, the first nucleic acid sequence, the second nucleic acid sequence, and the third nucleic acid sequence are within a same vector. In some embodiments, the first nucleic acid sequence, the second nucleic acid sequence, and the third nucleic acid sequence are within different vectors.

[0107] In some embodiments, the nucleic acid of the first nucleic acid sequence, the second nucleic acid sequence, and the third nucleic acid sequence is DNA. In some embodiments, the nucleic acid of the first nucleic acid sequence, the second nucleic acid sequence, and the third nucleic acid sequence is mRNA.

[0108] Various vectors, in some embodiments, are used herein. In some embodiments, the vector is a eukaryotic or prokaryotic vector. In some embodiments, the vector is a viral vector. In some embodiments, the vector is a lentivirus vector, an adenovirus vector, or an adeno-associated virus vector. The viral vectors that may be used to deliver the first, second, and third of the displayed costimulatory ligand molecules into tumor cells include lentivirus, retrovirus, adeno associated virus (AAV), adenoviruses, herpes simplex virus (HSV), parvoviruses, and poxviruses such as vaccinia virus (VACV) and myxoma virus (MYXV).Compositions and Pharmaceutical CompositionsWSGR Docket No. 48295-718.601

[0109] Described herein, in some embodiments, are compositions comprising an antigen- presenting vesicle comprising a first fusion protein that comprises a single-chain trimer pMHC-I molecule and a second fusion protein that comprises a costimulatory ligand molecule. Described herein, in some embodiments, are pharmaceutical compositions comprising an antigen-presenting vesicle comprising a first fusion protein that comprises a first fusion protein that comprises a single-chain trimer pMHC-I molecule and a second fusion protein that comprises a costimulatory ligand molecule. Described herein, in some embodiments, are pharmaceutical compositions comprising an antigen-presenting vesicle comprising a first fusion protein that comprises a first fusion protein that comprises a single-chain trimer pMHC-I molecule and a second fusion protein that comprises a costimulatory ligand molecule and a pharmaceutically-acceptable excipient.

[0110] For administration to a subject, the antigen-presenting vesicles as disclosed herein, may be provided in a pharmaceutical composition together with one or more pharmaceutically acceptable carriers or excipients. In some embodiments, the antigen-presenting vesicles as disclosed herein, may be provided in a composition together with one or more carriers or excipients. The term "pharmaceutically acceptable carrier" includes, but is not limited to, any carrier that does not interfere with the effectiveness of the biological activity of the ingredients and that is not toxic to the patient to whom it is administered. Examples of suitable pharmaceutical carriers are well known in the art and include phosphate buffered saline solutions, water, emulsions, such as oil / water emulsions, various types of wetting agents, sterile solutions etc. Such carriers can be formulated by conventional methods and can be administered to the subject at a suitable dose. Preferably, the compositions are sterile. These compositions may also contain adjuvants such as preservative, emulsifying agents and dispersing agents. Prevention of the action of microorganisms may be ensured by the inclusion of various antibacterial and antifungal agents.

[0111] The pharmaceutical composition may be in any suitable form, (depending upon the desired method of administration). It may be provided in unit dosage form, may be provided in a sealed container, and may be provided as part of a kit. Such a kit may include instructions for use. It may include a plurality of said unit dosage forms.

[0112] The pharmaceutical composition may be adapted for administration by any appropriate route, including a parenteral (e.g., subcutaneous, intramuscular, intravenous, or inhalation) route. Such compositions may be prepared by any method known in the art of pharmacy, for example by mixing the active ingredient with the carrier(s) or excipient(s) under sterile conditions.

[0113] Dosages of the substances of the present disclosure can vary between wide limits, depending upon the disease or disorder to be treated, the age and condition of the individual to be treated, etc. and a physician will ultimately determine appropriate dosages to be used.WSGR Docket No. 48295-718.601Methods of Use

[0114] Antigen-presenting vesicles or compositions described herein, in some embodiments, are used to treat cancer. In some embodiments, the cancer is a hematological malignancy. In some embodiments, the cancer is leukemia or lymphoma. In some embodiments, the lymphoma is B-cell lymphoma. In some embodiments, the cancer is a solid tumor. In some embodiments, the solid tumor is sarcoma, breast cancer, lung cancer, or carcinoma. In some embodiments, the lung cancer is non-small cell lung cancer.

[0115] In some embodiments, administration of the antigen-presenting vesicles reduces or eliminates the cancer. In some embodiments, administration of the antigen-presenting vesicles increases cancer cell death, decrease tumor size, decrease cancer metastasis, or combinations thereof. In some embodiments, cell death is increased by about 1-fold to about 2.5-fold, about 1- fold to about 5-fold, about 2-fold to about 10-fold. In some embodiments, cell death is increased by at least 5-fold, at least 10-fold, at least 20-fold, at least 30-fold, at least 40-fold, at least 50-fold, at least 60-fold, at least 70-fold, at least 80-fold, at least 90-fold, at least 95-fold, 100-fold, or greater than 100-fold. In some embodiments, tumor size is decreased by about 1-fold to about 2.5-fold, about 1-fold to about 5-fold, about 2-fold to about 10-fold. In some embodiments, tumor size is decreased by at least 5-fold, at least 10-fold, at least 20-fold, at least 30-fold, at least 40-fold, at least 50-fold, at least 60-fold, at least 70-fold, at least 80-fold, at least 90-fold, at least 95-fold, 100- fold, or greater than 100-fold. In some embodiments, cancer metastasis is decreased by about 1- fold to about 2.5-fold, about 1-fold to about 5-fold, about 2-fold to about 10-fold. In some embodiments, cancer metastasis is decreased by at least 5-fold, at least 10-fold, at least 20-fold, at least 30-fold, at least 40-fold, at least 50-fold, at least 60-fold, at least 70-fold, at least 80-fold, at least 90-fold, at least 95-fold, 100-fold, or greater than 100-fold.

[0116] In some embodiments, administration of the antigen-presenting vesicles reduces or eliminates the cancer as compared to a level prior to administration of the antigen-presenting vesicles in the subject. In some embodiments, administration of the antigen-presenting vesicles reduces or eliminates the cancer as compared to a level if the subject had not received the antigen- presenting vesicles. In some embodiments, administration of the antigen-presenting vesicles reduces or eliminates the cancer as compared to a level if the subject had received a different cancer treatment including but not limited to, radiation, surgery, and chemotherapy.

[0117] Antigen-presenting vesicles described herein, in some embodiments, are used to treat an autoimmune disease. In some embodiments, the autoimmune disease is rheumatoid arthritis, systemic lupus erythematosus, multiple sclerosis, inflammatory bowel diseases, psoriasis, or aplastic anemia.WSGR Docket No. 48295-718.601

[0118] In some embodiments, administration of the antigen-presenting vesicles dampens or inhibit autoimmune responses as compared to a level prior to administration of the antigen- presenting vesicles in the subject. In some embodiments, administration of the antigen-presenting vesicles dampens or inhibit autoimmune responses as compared to a level if the subject had not received the antigen-presenting vesicles. In some embodiments, administration of the antigen- presenting vesicles dampens or inhibit autoimmune responses as compared to a level if the subject had received a different treatment. Antigen-presenting vesicles described herein, in some embodiments, are used to treat a viral infection.

[0119] In some embodiments, disclosed herein is a method of inducing protective immunity against cancer or viral infection in a subject in need thereof comprising administering an antigen- presenting vesicle to the subject. In some embodiments, the method comprises administering an antigen-presenting vesicle and a pharmaceutically-acceptable excipient to the subject. In some embodiments, the method comprises administering a composition comprising a first nucleic acid sequence encoding an antigen-presenting vesicle to the subject. In some embodiments, the method further comprises administering a composition comprising a first nucleic acid sequence encoding an antigen-presenting vesicle and a second nucleic acid sequence encoding one or more viral protein to the subject. In some embodiments, the viral protein is a lentiviral protein, a retroviral protein, an adenoviral protein, or combinations thereof. In some embodiments, the one or more viral proteins comprises gag, pol, pre, tat, rev, or combinations thereof.

[0120] Disclosed herein, in some embodiments, is a method of preventing or treating cancer or viral infection in a subject in need thereof comprising: (a) obtaining T cells from the subject; (b) contacting the T cells ex vivo with an antigen-presenting vesicle (APV) or a composition comprising a nucleic acid sequence encoding an APV to the T cells; and (c) administering the activated and expanded T cells from step (b) to the subject.

[0121] In some embodiments, a method disclosed herein further comprises administering a second therapy. In some embodiments, the second therapy comprises a Toll-like receptor agonist for TLR3 or TLR4 or TLR7 or TLR8, or TLR9 as an adjuvant, and optionally CpG, PolylC, LPS, and STING agonists. In some embodiments, the second therapy comprises an immune checkpoint blocked therapy, and optionally an anti-PDl or anti-PDLl or anti-CTLA-4 antibody. In some embodiments, the second therapy comprises a targeted therapy, and optionally a tyrosine kinase inhibitor or PARP inhibitor or RAS inhibitor. In some embodiments, the second therapy comprises T cell therapy, and optionally a chimeric antigen-receptor T cell therapy or tumor-infiltrated lymphocyte therapy. In some embodiments, the second therapy comprises standard care therapy, and optionally surgery or chemotherapy or radiation therapy.WSGR Docket No. 48295-718.601

[0122] In some instances, the subject is a mammal. In some instances, the subject is a mouse, rabbit, dog, pig, cattle, or human. Subjects treated by methods described herein may be infants, adults, or children. In some embodiments, the antigen-presenting vesicles are administered by inhalation, injection, ingestion, transfusion, implantation, or transplantation. In some embodiments, the antigen-presenting vesicles are administered transarterially, subcutaneously, intradermally, intratumorally, intranodally, intramedullary, intramuscularly, by intravenous (i.v.) injection, or intraperitoneally. In some embodiments, the antigen-presenting vesicles are administered intravenously. In some embodiments, the antigen-presenting vesicles are administered by inhalation. In some embodiments, the antigen-presenting vesicles are administered by an intraperitoneal injection. In some embodiments, the antigen-presenting vesicles are administered by an subcutaneous injection.EXAMPLES

[0123] The following examples are set forth to illustrate more clearly the principle and practice of embodiments disclosed herein to those skilled in the art and are not to be construed as limiting the scope of any claimed embodiments. Unless otherwise stated, all parts and percentages are on a weight basis.

[0124] Here, we describe the development of exosome and VLP -based SCT-APVs for antigenspecific T-cell activation in vitro and in vivo. We designed SCT-APVs to mimic the pMHC and TCR engagement between APCs and T cells using exosome or VLP scaffoldings. The following major sections highlight the design, development, and functional analysis of SCT-APVs in vitro and in vivo. Section 1 describes the design and characterization of oligomerized protein display systems, which enable the display of type I and type II transmembrane proteins in oligomerized forms at high copy numbers on vesicles. Section 2 summarizes the search for the optimal composition of SCT and costimulatory molecules required for SCT-APV function in antigenspecific T-cell stimulation in vitro. Section 3 describes the use of neoantigen-specific SCT-APVs as preventative and therapeutic cancer vaccines, either as monotherapy or as combination therapy with other cancer treatment regimens.High-density and oligomerized protein display on vesiclesDesign of multivalent oligomeric protein display on enveloped vesicles

[0125] Given the relatively small size of exosomes and VLPs (-100 nm), we first optimized the display strategy to maximize the density and oligomerization of the displayed molecules on vesicles. The aim was to display pMHC and costimulatory ligands at high density and in oligomerized format on vesicles. We postulate that high density and oligomerized display of pMHC and costimulatoryWSGR Docket No. 48295-718.601 ligands on these small vesicles may help them to engage a large number of TCR and costimulatory molecules on T cells and lead to productive engagement that mimics the formation of immunological synapses during DC and T cell engagement. To this end, we engineered vesicle display systems by mimicking enveloped viruses, which often display their surface proteins in multivalent oligomeric format and high density on virions. Viruses such as Influenza or Hepatitis B displays hundreds of copies of viral spikes per virion, enabling a single virion to simultaneously interact with multiple copies of host cell receptors and attachment proteins. In the case of coronaviruses, host cell receptors Angiotensin-converting enzyme 2 (ACE2) and dipeptidyl peptidase 4 (DPP4) serve as entry receptors for SARS CoV- 1 / 2 and MERS CoV, respectively. The densely packed spike proteins displayed on virions enable them to interact with multiple copies of entry receptors on the host cell surface with both high local avidity — spike to receptor — and global avidity — virion to host cell membrane. This evolutionarily conserved feature is shared by almost all enveloped viruses and underlies viruses ’ability to efficiently infect target cells by forming multivalent, Velcro-like interactions with cognate entry receptors on target cells. Many cellular membrane proteins, including TCR and costim complexes, are analogously clustered on the cell surface in oligomeric forms. We reasoned that vesicles displaying pMHC and costims in oligomeric display formats could demonstrate both enhanced display density and oligomeric clustering, thus improving their local and global avidity to their corresponding oligomeric binding partners on the T cell membrane.

[0126] To mimic multivalent oligomeric protein expression on the cell surface, we designed “display vectors” to express and efficiently target displayed peptides to the surface of enveloped vesicles, such as VLPs and exosomes (Figure 1). Display vectors are designed to express a fusion protein consisting of a signal peptide, a displayed peptide, an oligomerization domain, and a transmembrane and cytosolic tail for membrane anchoring and targeting. Furthermore, since viral surface proteins are highly efficient in targeting budding viral particles, these proteins may contain unique sequences required for vesicle display (Figure 1A). Based on this rationale, we used the transmembrane and cytosolic region of VSV-G (Vesicular Stomatitis virus glycoprotein) as a vesicle membrane anchoring protein. VSV-G can be used to effectively pseudotype lentivirus. Further, we also selected various well-defined oligomerization domains to test their effects on display efficiency and oligomerization format on vesicles (Figure IB). Finally, we functionally tested and optimized the display strategy by generating the “decoy-antivirus” — multivalent particles (MVPs) displaying oligomeric multivalent virus entry receptors as spike-capturing molecules. Specifically, we displayed viral entry receptors for SARS CoV, such as ACE2 and DDP4, on vesicles. Such design allowed us to quantitatively correlate the copy number andWSGR Docket No. 48295-718.601 oligomerization format of the displayed proteins to biological function in rapid virus neutralization assays. More importantly, the neutralization activity of decoy-antivirus is tightly linked to their binding affinity to virus particles.

[0127] Generation of monomeric multivalent particles (MVPs)

[0128] We first designed a monomeric display vector expressing a fusion protein consisting of a display protein linked to the VSVG transmembrane and intracellular domains (Figure 2). The other structural domains of VSVG, including the D4 post-fusion trimerization domain, were eliminated to facilitate the display of monomeric proteins. By design, multivalent proteins can be displayed as monomers on the surface of viral-like particles (VLPs) and extracellular vesicles (EVs) such as exosomes and ectosomes using this monomeric display vector (Figure 3A-C). To produce monomeric VLP-based MVPs with viral RNA genomes, we co-transfected the monomeric peptide fusion construct with a lentiviral packaging construct expressing essential packaging components, such as Gag-Pol and Rev proteins, and a viral genome transfer vector encoding a GFP / luciferase reporter (Figure 3A). Alternatively, monomeric VLP-MVPs without RNA genome were produced by co-transfecting peptide display vector together with a lentiviral packaging construct but no viral genome transfer vector (Figure 3B). Finally, monomeric EV-based MVPs, including exosome-MVPs and ectosome-MVPs, were produced by transfecting only the monomeric peptide displaying vector in 293T cells (Figure 3C).

[0129] To test this design, we used the monomeric display vector (Figure 2) to produce VLPs displaying Angiotensin-converting enzyme 2 (ACE2), the entry receptor for SARS CoV-2, termed ACE2-VM-MVPS (Figure 3). ACE2-VM-MVPs may be generated as VLPs with viral genome (Figure 3A), VLPs without viral genome (Figure 3B), or EVs (Figure 3C). The purified VLP- based ACE2-VM-MVPS concentration was determined via P24 ELISA. Quantitative western blot analysis revealed that ACE2-VM-MVPs display 1200 ±400 copies of ACE2 per particle (Figure 4A). Notably, few studies have systematically addressed the density and oligomerization of displaying proteins on EVs and VLPs, and it is considered a challenge to display large proteins on EVs and VLPs at high density (???). Furthermore, western blot analysis of ACE2-VM-MVPs comparing fusion proteins under reducing and non-reducing conditions demonstrated no difference in bands, indicating that ACE2-VM fusion proteins are expressed as monomers on the surface of MVPs (Figure 4B). To confirm that ACE2-VM-MVPs display functional ACE2, we then tested them in pseudovirus neutralization assays against SARS CoV-2. ACE2-VM-MVPs neutralized SARS CoV-2 pseudovirus at a potent ICso of 1.3pM (Figure 4C). ACE2-VM-MVPs are more potent than most neutralizing antibodies identified so far for SARS CoV-2, demonstrating that ACE2-VM-MVPS display functional ACE2 that allows for multivalent binding and neutralizationWSGR Docket No. 48295-718.601 of SARS CoV-2 pseudovirus. These results demonstrate that monomeric fusion peptides can effectively display thousands of copies of functional peptides on the surface of VLPs.

[0130] Generation of dimeric MVPs

[0131] A dimeric display construct was designed by fusing the displayed peptide to the GCN4 leucine zipper domain, followed by the transmembrane and intracellular domains of VSV-G (Figure 5). To produce dimeric VLP -based MVPs with viral RNA genomes, we co-transfected cells with a dimeric peptide fusion construct alongside a lentiviral packaging construct expressing essential packaging components, such as Gag-Pol and Rev proteins, and a viral genome transfer vector encoding a GFP / luciferase reporter (Figure 6A). Dimeric fusion peptide constructs utilize the leucine zipper dimerization domain. Alternatively, dimeric VLP-based MVPs without RNA genome were produced by co-transfecting dimeric peptide display vector together with a lentiviral packaging construct but no viral genome transfer vector (Figure 6B). Finally, dimeric EV-based MVPs, including exosome-MVPs and ectosome-MVPs, were produced by transfecting only the dimeric peptide displaying vector in 293T cells (Figure 6C).

[0132] To test this design, we used the dimeric display vector (Figure 5) to produce VLPs displaying Angiotensin-converting enzyme 2 (ACE2), the entry receptor for SARS CoV-2, termed ACE2-LZ MVPs (Figure 6A). ACE2-LZ-MVPs may be generated as VLPs with viral genome (Figure 6A), VLPs without viral genome (Figure 6B), or EVs (Figure 6C). The concentration of purified ACE2-LZ MVP was determined via P24 ELISA. Quantitative western blot analysis revealed that ACE2 / LZ MVPs display approximately 900 ±300 copies of ACE2 per particle, slightly fewer copies per particle than monomeric ACE / VM MVPs (Figure 7A). However, reducing versus non-reducing western blot analysis indicated that ACE2-LZ fusion peptides are expressed in both dimeric and monomeric forms on ACE2-LZ-MVPs, though predominantly monomeric form (Figure 7B). To confirm that ACE2-LZ MVPs display functional ACE2, we then tested them in pseudovirus neutralization assays against SARS CoV-2. ACE2-LZ MVPs neutralized SARS CoV-2 pseudovirus at a potent ICso of 3.7pM (Figure 7C). This decrease in potency as compared to monomeric ACE2 / VM MVPs is likely due to decreased overall ACE2 copy number and insignificant amount of dimeric displayed ACE2. Thus, the GCN4 leucine zipper domain is ineffective in generating a dimerized display of peptides on VLPs or EVs. However, other dimerization domains, such as Fc, basic helix-loop-helix, immunoglobulin domain, forkhead- associated domain, pleckstrin homology domain, and death domain, may be used to create more effective dimers on VLPs and EVs.

[0133] 1.4. Generation of trimeric MVPsWSGR Docket No. 48295-718.601

[0134] Several trimeric display constructs were designed by fusing the displayed peptide to the D4 post-fusion trimerization domain of VSV-G (Figure 8), the Dengue E protein post-fusion trimerization domain (Figure 9), or the T4 Phage Fibritin Foldon domain (Figure 10), followed by the transmembrane and intracellular domains of VSV-G. By design, using a trimeric display vector, multivalent proteins can be displayed as trimers on the surface of the viral-like particles (VLP) and extracellular vesicles (EV), such as exosomes and exosomes. To produce trimeric VLP -based MVPs with viral RNA genomes, we co-transfected cells with a trimeric peptide fusion construct utilizing one of several trimerization domains — D4, DE, or Foldon — alongside a lentiviral packaging construct expressing essential packaging components, such as Gag-Pol and Rev proteins, and a viral genome transfer vector encoding a GFP / luciferase reporter (Figure 11 A). Alternatively, trimeric VLP -based MVPs without RNA genome were produced by co-transfecting display vector together with only a lentiviral packaging construct but not the viral genome transfer vector (Figure 11B). Finally, trimeric EV-based MVPs, including exosome-MVPs or ectosome-MVPs, were produced by transfecting only the trimeric peptide display vector (Figure 11C).

[0135] We generated and characterized trimeric ACE2-MVPs using trimeric fusion constructs derived from several possible trimerization domains, including the D4 post-fusion trimerization domain of VSV-G (Figure 8), the Dengue E post-fusion trimerization domain (Figure 9), or the T4 Phage Foldon domain (Figure 10). ACE2-MVPs displaying each of these trimeric ACE2 fusion constructs were termed ACE2-D4, ACE2-DE, and ACE2-Fold-MVPs, respectively. Purified ACE2-D4, ACE2-DE, and ACE2-Fold-MVPs were quantified via P24 ELISA. Quantitative western blot analysis of each type of trimeric MVP revealed that ACE2-D4, ACE2-DE, and ACE2- Fold-MVPs display 3600 ±1300, 2000 ±600, and 1450 ±150 copies of ACE2 per particle, respectively (Figure 12A), demonstrating that inclusion of trimerization domains in fusion constructs significantly increases display copy numbers as compared to monomeric fusion constructs. Notably, monomeric ACE2-VM-MVPs display only 1200 ±400 copies of ACE2 per particle (Figure 4A). To verify that trimerization constructs indeed displayed ACE2 in trimers, we analyzed monomeric and trimeric ACE2 MVPs in western blots under reducing and non-reducing conditions (Figure 12B). We found that while ACE2-VM-MVPs demonstrate only a single, monomeric band under non-reducing conditions, whereas ACE2 / D4, ACE2 / DE, and ACE2 / Fold- MVPs show larger, higher-order oligomeric bands, including trimeric bands (Figure 12B). Of these three, ACE2 / D4 and ACE2 / DE-MVPs but not the ACE2 / Fold-MVPs demonstrate a larger proportion of oligomerized fusion peptides at much higher copies / particle. Finally, we tested trimeric ACE2-MVPs in pseudovirus neutralization assays against SARS CoV-2 to confirm functional ACE2 display. We found that ACE2-D4, ACE2-DE, and ACE2-Fold-MVPs neutralizeWSGR Docket No. 48295-718.601SARS CoV-2 pseudovirus at ICsos of 0.3, 1.2, and 1.1 pM (Figure 12C), respectively, and are up to 10-fold more potent than monomeric ACE2-VM MVPs (Figure 4C). This increase in neutralizing potency can be attributed to both the increase in overall ACE2 valency on trimeric MVPs, as well as the boosting of local multivalence between viral spike trimers and trimeric ACE2 fusion peptides, allowing trimeric ACE2 MVPs to bind and neutralize virions more effectively. These results demonstrate that ACE2 MVPs displaying trimeric fusion peptides display thousands of copies of functional, trimerized ACE2 per particle. More importantly, trimeric fusion display constructs incorporating trimerization domains such as D4, DE, or Foldon can be used to display thousands of functional trimeric peptides on MVPs. It is important to note that ACE2-D4-MVPs have the highest copies of trimeric / oligomeric ACE2 displayed per particle and are most potent in virus neutralization, demonstrating that D4 oligomerization domain may be most effective in displaying high copies of trimeric peptides on the VLP surface. Importantly, these results also demonstrate that optimized valency and oligomerization format of ACE2 on MVPs are key to developing more potent neutralizing antivirals against SARS CoV-2.

[0136] 1.5. Comparison of decoy-MVPs with distinct oligomerized ACE2 display

[0137] Interestingly, ACE2-MVPs with distinct oligomerized ACE2 display have varied copies ACE2 molecules / particle (TABLE 11, Figure 13A). For example, trimeric ACE2-D4 MVP has the highest level of ACE2 protein displayed at ~ 3600 ±1300 copies / particle, whereas monomeric ACE2-VM MVP and dimeric ACE2-LZ MVP display about -1000 copies ACE2 / particle. Among the ACE2-MVPs with D4, or DE, or Fold trimeric domains, ACE2-D4 and ACE2-DE MVPs utilizing trimeric domains from viral surface proteins display over 2000 copies ACE2 / particle, whereas ACE2-Fold MVPs utilizing a trimeric domain from bacteria phage protein. These results clearly illustrated that ACE2 surface display levels on VLPs can be enhanced by incorporating trimerization domains from viral surface proteins into the display vector, such as D4 of VSVG and DE of Dengue E surface protein (Figure 13A). Moreover, D4 and DE promote more robust trimerization / oligomerization of ACE2 display on VLPs (Figures 12B, 13B), suggesting that these trimeric domains may enhance the surface presentation of displayed peptides. Notably, the trimeric configuration is one of the most common features many viral surface / spike proteins share. Finally, the neutralizing potencies of ACE2-MVPs are also impacted by the oligomerization format of displayed ACE2 (Figure 13C). Notably, ACE2-D4 MVP has the highest neutralizing potency, which correlates with its high ACE2 valency and trimeric / oligomeric configuration. Nevertheless, all ACE2-MVPs have ICsos in the low picomolar range, demonstrating that they are all potent neutralizing molecules against SARS CoV-2 in pseudovirus neutralization assays.WSGR Docket No. 48295-718.601TABLE 11

[0138] / .6. Generation of tetrameric MVPs

[0139] A tetrameric display construct was designed by fusing the stem domain of Influenza Neuraminidase to the desired peptide (Figure 14). By design, multivalent proteins can be displayed as tetramers on the surface of VLPs and EVs, such as exosomes and ectosomes, using a tetrameric display vector. To produce tetrameric, VLP -based MVPs with viral RNA genomes, we cotransfected the tetrameric, NA-based Type II peptide fusion construct with a lentiviral packaging construct expressing essential packaging components, such as Gag-Pol and Rev proteins and a viral genome transfer vector encoding a GFP / luciferase reporter (Figure 15A). Alternatively, tetrameric, VLP -based MVPs without RNA genomes were produced by co-transfecting the tetrameric peptide display vector together with a lentiviral packaging construct but no viral genome transfer vector (Figure 15B). Finally, tetrameric, EV-based MVPs, including exosome-MVPs and ectosome- MVPs, were produced by transfecting only the tetrameric peptide displaying vector in 293T cells (Figure 15C).

[0140] We generated and characterized tetrameric MVPs displaying Dipeptidyl Peptidase 4 (DPP4) — the entry receptor for MERS CoV — termed DPP4-NA MVPs. We tested two variations of our tetrameric fusion peptide derived from the Influenza Neuraminidase stem domain (Figure 16A). The concentration of purified DPP4-NA MVPs was determined via P24 ELISA. Quantitative western blot analysis revealed that DPP4 / NA MVPs displaying version 1 and 2 of the DPP4 / NA display constructs displayed approximately 390 ±220 and 150 ±70 copies of DPP4 per particle, respectively (Figure 16B). In addition, western blot analysis of both versions of DPP4 / NA MVPs under reducing and non-reducing conditions demonstrated that both versions result in multimeric DPP4 display on MVPs (Figure 16B). Finally, to determine whether our fusion peptides effectively display functional DPP4 on MVPs, we tested both versions of DPP4 / NA MVPs against MERSWSGR Docket No. 48295-718.601CoV in pseudovirus neutralization assays. Version 1 and 2 of DPP4 / NA MVPs neutralized MERS CoV pseudovirus at ICsos of 0.91 and 0.87 pM, respectively, more than 5-logs more potent than soluble DPP4 protein (Figure 16C). These results demonstrate that DPP4-NA MVPs display hundreds of copies of multimeric, functional DPP4. Moreover, Neuraminidase stem-based fusion peptides are effective, Type II transmembrane proteins capable of displaying numerous tetrameric peptide copies on MVP surfaces. This vector may be uniquely suited to display type II peptide on VLPs and EVs.

[0141] / . 7. Generation MVPs displaying mixed oligomeric proteins

[0142] With the abovementioned display vectors, we also generated MVPs displaying multiple peptides in mixed oligomeric formats by co-transfecting distinct peptide display constructs with differing oligomerization domains. Such a design can be used to increase the display density of a displayed peptide or to create combinatorial display patterns of distinct display peptides. Again, mixed oligomeric MVPs can be built with VLPs and EVs, such as exosomes and ectosomes, by cotransfecting monomeric and oligomeric display vectors. To produce mixed VLP -based MVPs with viral RNA genomes, we co-transfected the mixed oligomeric peptide fusion constructs with a lentiviral packaging construct expressing essential packaging components, such as Gag-Pol and Rev proteins, and a viral genome transfer vector encoding a GFP / luciferase reporter (Figure 17A). Alternatively, mixed VLP -based MVPs without RNA genome were produced by co-transfecting the mixed oligomeric display vectors together with only a lentiviral packaging construct but not the viral genome transfer vector (Figure 17B). Finally, mixed EV-based MVPs, including mixed exosome-MVPs and ectosome-MVPs, were produced by transfecting the mixed oligomeric fusion constructs alone (Figure 17C). Mixed MVPs can be generated by co-transfecting display vectors utilizing distinct oligomerization domains (Figure 1).

[0143] Here we explored whether the neutralizing potency of MVPs displaying spikerecognizing antibodies can be further enhanced with multi-specificity. For example, we can genetically programmed MVPs to display combinations of distinct scFv antibodies recognizing multiple binding sites on the SARS CoV-2 spike protein. This design, if successful, would further enhance the neutralizing potency of such MVPs against pandemic viruses and mitigate the effects of typical spike escape mutagenesis on neutralizing antibodies. To this end, we designed and generated (aRBD:C18 / VM)(aNTD:CV26 / D4) bi-specific MVPs (bi-MVPs) that co-display two scFv fusion peptides — monomeric aRBD:C18 / VM and trimeric aNTD:CV26 / D4 — recognizing the RBD and the NTD regions of CoV-2 spike protein. The expression level and oligomerization of monomeric aRBD:C 18 / VM and trimeric aNTD:CV26 / D4 on MVPs were confirmed by quantitative Western-blot analyses (Figure 18A). We then tested these bi Ab -antiviruses againstWSGR Docket No. 48295-718.601 their respective target viruses in pseudovirus neutralization assays. We showed that mixed (aRBD:C018 / VM)(aNTD:CV26 / D4) biAb-antiviruses have an IC50 of 3.06 ±0.23 pM in pseudovirus neutralization assays, a more than 2-fold increase in potency as compared to mono- specific Ab-antiviruses displaying either aNTD:CV26 / D4 or aRBD:C018 / VM alone (Figure 18B). Furthermore, these mixed biAb-MVPs suppress pseudovirus infection by approximately 10,000- fold as determined by the decrease in pseudovirus infection luciferase signal, a 100-fold increase in suppression compared to each respective mono-specific Ab-antivirus (Figure 18C). These results illustrate the potential to genetically program MVPs to display distinct peptides to create bi-specific functionalities.

[0144] / .8. Patterned display of proteins of distinct structure and functional classes on MVPs

[0145] The display methods described above were employed to generate patterned display of diverse classes of proteins with distinct structural features on MVPs. We demonstrated that proteins of distinct structural classes, such as extracellular domains of type I, II, and III transmembrane proteins, GPI-anchored proteins, secreted proteins, and multi-pass transmembrane proteins, can be effectively displayed on the surface of VLPs in monomeric or trimeric formats (TABLE 12, Figure 19A-C). Moreover, the display methods were employed to generate patterned display of diverse classes of proteins with distinct functional properties on VLPs and EVs. To this end, we showed that single-chain antibodies, the extracellular domain of viral entry receptors, immune checkpoint, cytokines, mitogens, chemokines, and adhesion factors can also be effectively displayed on the surface of VLPs in various oligomeric formats (TABLE 13, Figure 19A-C).TABLE 12WSGR Docket No. 48295-718.601TABLE 13

[0146] For example, we have generated decoy-MVPs as SARS CoV-2 neutralizing therapeutics displaying thousands of copies of monomeric or trimeric ACE2 extracellular domain, which is a type I transmembrane protein (Figures 4, 7, 12, 13). We demonstrated that both monomeric and trimeric ACE2-MVPs can effectively neutralize SARS-CoV-2 virus in pseudovirus neutralization assays with sub-picomolar ICso s (Figures 4C, 7C, 12C, 13C). Furthermore, we also effectively generated decoy-MVPs as MERS CoV neutralizing therapeutics displaying hundreds of copies of the DPP4 extracellular domain, which is a type II transmembrane protein (Figure 16). We demonstrated that DPP4 decoy MVPs can effectively neutralize MERS in a pseudovirus neutralization assay with sub-picomolar ICsos (Figure 16C). Beyond the viral entry receptors, we can also effectively display immune checkpoints, cytokines, mitogens, chemokines, and adhesion factors on the surface of VLPs at high copy numbers (Figures 19A, 19B). Finally, we designed MVPs displaying monomeric and trimeric single-chain variable fragments (scFv) targeting CD19, a specific marker for B-cells or some types of B-cell leukemias, termed aCD19-VM and aCD19-D4 MVPs, respectively. aCD19-VM and aCD19-D4 MVPs display 6200 ±1200 and 1200 ±300 copies of scFv per particle, respectively (Figure 19C). One can envision creating bi-function MVPs displaying aCD19 and, together with various receptors, or ligands, or cytokines to target the MVPs to normal or leukemia B cells. Collectively, these results demonstrate that MVP display technology can be broadly applied to generate vesicles displaying proteins of distinct structural and functional classes of proteins.

[0147] 1.9. EV-based ACE2-MVPs are highly potent inhibitors against live CoV-2 viruses

[0148] By transfecting only the trimeric ACE2-D4 displaying vector to 293T cells (Figure 20A), we generated EVs displaying multiple copies of oligomerized ACE2, designated ACE2-D4 EVs. The sizes of ACE2-D4-EVs are in the range of 131 ±29 nm as determined by tunableWSGR Docket No. 48295-718.601 resistive pulse sensing analysis (TRPS) using qNano (Figure 20B). Each ACE2-D4 EV particle displays approximately 14000 ±6000 copies of ACE2-D4VG fusion molecules (Figure 20C). Further, we showed that ACE2-D4 EVs displaying trimeric H2A / ACE2, a mutant ACE2 with no enzymatic activity, are highly potent inhibitors, neutralizing CoV-2 pseudovirus at ICsos of 26 ±12 fM (Figure 20D). Thus, oligomerized ACE2-D4-EVs are highly potent against live CoV-2 virus infection (Figure 20D). Notably, the level of displayed ACE2 on EVs is at 14000 ±6000 copies / EV, whereas the level of displayed ACE2 on VLPs is 3600 ± 1300 copies / VLP, demonstrating that the display vector with D4 oligomerization domain can display significantly higher copies of ACE2 on EVs than on VLPs.

[0149] 1.10. Coiled-coil human peptides as oligomerization domains for vesicle display

[0150] To further optimize surface display on vesicles, we investigated several common structural motifs found across oligomeric human proteins. We found that coiled-coil domains, which are a series of identical 2 to 7 u-helices that snake around each other like the strand of rope, may make ideal oligomerization domains for vesicle display. These structurally repetitive motifs are found across numerous human proteins and can induce a range of oligomerization states from dimers to heptamers. We postulated that coiled-coil domains from a variety of human proteins could be used to create various oligomeric patterns of peptide display on MVPs. For example, the LINE-1 (LI) retrotransposon (Figure 21), which accounts for roughly seventeen percent of the human genome, contains two open reading frames (orf), the first of which encodes a highly unique, 500 amino acid protein (orflp) including a LINE-1 retrotransposon orflp coiled-coil trimerization domain (Llcc). The prevalence of the LI retrotransposon in the human genome makes the Llcc an ideal non-immunogenic domain for trimeric peptide display on vesicles. In addition to the Llcc trimerization domain, the coiled-coil domains from human collagen XV, lung surfactant protein D, tetranectin, mannose-binding protein, Factor X, and Fibrinogen can also be used to create oligomerized display patterns on the surface of VLPs and EVs. (TABLE 14). Such designs enable multivalent display of fusion peptides on the surface of VLPs and EVs in varied oligomeric configurations and at copy numbers comparable to or greater than their viral display counterparts. Since these coiled-coil oligomerization peptides originated from highly conserved human proteins, multivalent therapeutics produced using these MVP display constructs are less likely to be immunogenic.WSGR Docket No. 48295-718.601TABLE 14

[0151] To this end, we designed a display vector to present fusion peptides on VLPs or EVs by using Llcc as an oligomerization domain (Figure 21). Specifically, the vector includes a mammalian promoter driving the expression of a fusion peptide, which consists of a linked signal peptide(SP), a displayed peptide, an Llcc trimeric peptide, and the transmembrane and cytoplasmic tail domains of the VSV-G protein or to the transmembrane and cytoplasmic tail domains of the LINE1 orfl protein. MVPs displaying various types of Llcc fusion peptides were produced as VLPs both with and without genomes and as extracellular vesicles (EVs) such as exosomes and ectosomes. To produce MVPs displaying Llcc fusion peptides on VLPs containing RNA genomes, HEK 293T cells were co-transfected with Llcc-trimeric peptide display constructs along with a lentiviral packaging construct expressing essential packaging components, such as Gag-Pol and Rev proteins, and a viral genome transfer vector encoding a GFP / luciferase reporter (Figure 22A). To produce MVPs displaying Llcc fusion peptides on VLPs without genomes, HEK 293T cells were co-transfected with Llcc-trimeric peptide display vectors and a lentiviral packaging construct without a viral genome transfer vector (Figure 22B). Finally, to produce MVPs displaying Llcc fusion peptides on EVs, 293T cells were transfected with Llcc-trimeric peptide display constructs alone (Figure 22C). Concentrations of VLP or EV-based humanized ACE2-MVPs were determined via P24 or tunable resistive pulse sensing (TRPS, qNano), respectively. Based on these concentrations, quantitative western blot analysis determined the copy numbers of displayed humanized fusion peptides on MVPs. The oligomerization patterns of displayed fusion peptides were determined via non-reducing PAGE analysis. In particular, MVPs displaying Llcc-ACE2WSGR Docket No. 48295-718.601 were found to display 7200 ±3600 copies of Llcc-ACE2 per particle, more than twice as many copies as ACE2-D4-MVPs (Figure 23 A). Moreover, a significant fraction of displayed Llcc- ACE2 on VLPs is in the oligomerized format as indicated by western-blot analyses under nonreducing conditions (Figure 23B). These findings demonstrate that Llcc-trimeric domains derived from human LINE1 orfl protein can be used to increase the density and oligomerization of the displayed peptides on vesicles. Finally, we showed that ACE2-Llcc-MVPs were able to neutralize three different SARS CoV-2 spike variants at low or sub-picomolar ICsos in pseudovirus neutralization assays (Figure 23C), further supporting that increased peptide display density and oligomerization on vesicles correlates increased function potency of displayed vesicles.

[0152] 1.11. Summary of high-density oligomerized protein display on VLPs and EVs

[0153] The sizes of EVs and VLPs are in the range of 100 to 200 nm. In comparison, the sizes of DCs or other natural APCs are more than 20 micrometers, and thus are more than 100 times larger than EVs and VLPs in diameters. So, DCs and APCs provide much larger surface area and hence provide more interacting molecules for T cell engagement and the formation of immunological synapses. To overcome the size limitations of EVs and VLPs, we developed efficient display technology to target the high density and oligomerized displayed molecules to the surface of EVs and VLPs. We validated these designs functionally by testing their efficacy in pseudovirus neutralization. We have shown that increase the local and global avidity of the displayed molecules with their interaction partners. These findings laid the foundation to build antigen-presenting vesicles based on EVs and VLPs through targeted oligomerized display of the signaling proteins involved in immunological synapse formation during the engagement of antigen- presenting cells and T cells.The compositions of antigen-presenting vesicles for T cell activation

[0154] 2.1. Design and production of SCT-APVs

[0155] APCs express pMHCs and many activating costimulatory molecules that participate in TCR engagement (Figure 24A). During antigen recognition by T cells, TCR and pMHC engagement leads to the formation of immunological synapses, which include highly condensed aggregates of the TCR subunits and signaling proteins, adhesion molecules, costimulatory receptors, and inhibitor receptors on the T cell membrane. With the display toolboxes described in section 1, we contemplated recreating such condensed multivalent interactions using engineered vesicles displaying high copies of oligomerized molecular machinery required for TCR engagement. We postulated that high-density, oligomerized expression of these TCR-engaging proteins may engage equivalently densities of cognate receptors on T cells and mimic the formationWSGR Docket No. 48295-718.601 of immunological synapses. Furthermore, SCT-APVs may be designed to hardwire pro- inflammatory T-cell stimulation without sensitivity to negative environmental cues. In contrast, the outcomes of natural APC and T cell engagement under physiological conditions depend on the state of APCs and the presence or absence of activating and inhibitory receptors on the APCs, leading to either pro-inflammatory or tolerogenic outcomes.

[0156] To create SCT-APVs, we used our established vesicle-display strategies to target the essential TCR-engaging proteins, including pMHC and various costimulatory molecules, onto VLPs or EVs (Figure 24B). Since the class I pMHCs consist of three polypeptides: antigenic peptide, MHC-I alpha chain, and P2m, we adopted the well-established single-chain trimer (SCT) to link the antigenic peptide, P2m, and MHC-I into a single polypeptide for pMHC-I display. We then connect the SCT to a type I transmembrane protein membrane anchor and specified oligomerization domain. Similarly, the extracellular domains of type I or type II costimulatory molecules are linked to corresponding type I or type II transmembrane protein membrane anchors and oligomerization domains. As demonstrated in Section 1, such constructions can lead to the high-copy surface display of SCT and costimulatory molecules at high copies and in oligomerized format on the surface of VLP and EVs (Figure 24C).

[0157] SCT-APVs may be produced as EVs or VLPs. We first developed S293-HLA null cells by inactivating all three HLA alleles, HLA-A, HLA-B, and HLA-C, from a suspension S293 cell line. This cell line will enable the production of SCT-APVs without any endogenous human class I MHC molecules. Moreover, S293-HLA null cells also enable scale-up production of SCT-APVs in industrial-scale bioreactors. To produce EV-based APVs, we transfected DNA constructs encoding SCT-pMHC-I / and costimulatory molecules (X, Y, Z) into S293-HLA null cells without a lentiviral transfer and packaging vector (Figure 25). Alternatively, to produce VLP -based APVs, we transfected DNA constructs encoding SCT-pMHC-E and costimulatory molecules (X, Y, Z) into S293-HLA null cells together with a lentiviral transfer and packaging vector (Figure 25). EV- or VLP -based APVs are harvested from the supernatant, purified through a multi-step process to remove all protein and nucleic acid contaminants, and reconstituted in a stabilizing buffer for in vitro and in vivo usage.

[0158] 2.2. Define the optimal costimulatory signals required for SCT-APVs

[0159] Some early studies have attempted to build T cell activation particles by displaying SCT-pMHC-I and CD80 on the surface of lentiviral VLPs. However, these T cell activation VLP particles had nominal functions in T cell activation in culture compared to anti-CD3 and anti-CD28 activation beads and were not suitable for in vivo T cell activation. Furthermore, the copies of SCT and CD80 displayed on these VLPs are unknown. There have been no efforts made to increase theWSGR Docket No. 48295-718.601 density and / or oligomerization format of the displayed SCT and CD80 to maximize the effects of avidity during their engagement with TCR and target CD28 costimulatory molecules. Finally, no published studies have optimized the combinatorial code of costimulatory molecules required to reconstitute the T cell activation machinery on VLPs.

[0160] To this end, we first carried out a costim subtraction study to evaluate the contribution of a panel of critical costimulatory molecules on the T cell activation function of SCT-APVs (Figure 26). Specifically, we generated SCT-APVs displaying ova-SCT and a combination of costimulatory molecules, including CD80, CD86, GITRL (GITR ligand), OX40L (XO40 ligand), CD48, CD30L (CD30 ligand), and ICAM-1, designated as all-costim APVs (Figure 26A). We then subtracted one costimulatory molecule from the all-costim APVs to create a list of costim subtracted APVs, including AGITRL, ACD80, ACD86, AOX40L, and ACD48-APVs. We also created SCT-APVs displaying only ova-SCT and GITRL (GITR ligand) only. All SCT-APVs are EV-based and display multivalent ICAM-1, and all Type I displayed molecules incorporate the D4 oligomerization domain. All Type II displayed molecules utilize the neuraminidase stem oligomerization domain and transmembrane anchor. We then examined the function of these SCT- APVs in activating naive OT-1 T cells (Figure 26B, 26C). Naive OT-1 T cells were isolated from mice and stimulated all-costim ova- APVs or various all-costim ova- APVs with one costimulatory molecule subtracted. We found that all-costim ova- APVs can effectively induce the expression of CD69, an early T cell activation marker, in naive OT-1 T cells, as indicated by FACS analysis at day 1 post-T cell activation. About 48% of the T cells stimulated by the all-costim ova-APVs are CD69+, with -3% being CD69 and CD25 double-positive (Figure 26B, 26C). Interestingly, all costimulatory molecules contribute to T cell activation, as subtraction of each of the costimulatory molecules results in a significant decrease of the percent of total CD69+cells at day 1 post-T cell activation by ova-APVs. The contribution by each of the costimulatory molecules is quantitatively distinct. Among the costimulatory molecules tested, GITRL (GITR ligand) has a dominant contribution to the function of all-costim ova-APVs in naive mouse OT-1 T cell activation.Subtracting GITRL (GITR ligand) from all -custom ova-APVs causes a reduction of the percent of total CD69 positive cells from -48% to ~ 9.7%, a nearly 80% decrease. Also important to note, subtracting CD48 results in a reduction from -48% to x~28%. In contrast, subtracting other costimulatory molecules results in a less dramatic reduction in the percentage of total CD69 positive cells from -48% to -38%, -32%, -36%, and -33% for CD80, CD86, OX40L, and CD30L subtraction, respectively.

[0161] To further corroborate the above findings, we carried out a costim addition study to evaluate the contribution of various costimulatory molecules on the T cell activation function ofWSGR Docket No. 48295-718.601APVs (Figure 27). Given the critical function of GITRL (GITR ligand) in T cell activation by SCT-APVs, we created base ova-APVs displaying only ova-SCT, ICAM-1, and GITRL (GITR ligand). We then created “+costim ova-APVs” by adding individual costimulatory molecules, including CD80, CD86, OX40L, and CD48, one at a time (Figure 27A). We then tested the function of +costim ova-APVs in activating splenic OT-1 T cells, including naive and antigenexperience OT-1 T cells, and control Pmel T cells, which do not recognize the ova peptide antigen (Figure 27B, 27C). We found that base ova-APVs modestly increased the percent of CD69 and CD25 double-positive T cells among the activated OT-1 cells compared to that of stimulated Pmel T cells. Interestingly, the addition of CD80, CD86, XO40L, and CD48 increased the percent of CD69 and CD25 double-positive T cells among the activated OT-1 cells to 56%, 55%, 49%, and 44%, respectively. By contrast, Pmel T cells stimulated with these +costim ova-APVs have about -20% CD69 and CD25 double-positive T cells and thus do not respond to the addition of costimulatory molecules.

[0162] The above studies demonstrate that producing highly potent EV-based APVs for antigen-specific T-cell activation is feasible by displaying SCT-MHC-I as pMHC and a codisplaying a combination of T cell costimulatory molecules. Interestingly, we found that the GITRL (GITR ligand) costimulatory signal has a dominant role for SCT-APVs in activating antigen-specific mouse T cells. Nevertheless, we also found that all costimulatory molecules tested, including CD80, CD86, CD48, OX40L, CD30L, and GITRL (GITR ligand), contributed to naive antigen-specific T-cell activation by ova-APVs. Finally, ova-APVs selectively activate OT-1 T cells but not Pmel T cells, suggesting the potential of antigen-specific T cell activation by ova- APVs. It’s important to note that we have only examined the early T cell activation markers in the above analysis. The effects of SCT-APVs and individual costimulatory molecules on T cell proliferation and differentiation in culture and in vivo are not addressed here but may be tested in the future. Also important to note, all-costim ova-APVs are produced through transient transfection. Although the transfection efficiency in S293HLA null cells is generally high (>60%), all-costim ova-APVs produced by this method are likely to be heterogeneous. Nevertheless, we still observed consistent and robust contribution of costimulatory molecules to APV functions in T cell activation, suggesting additional room to further improve SCT-APV function by increasing the displayed molecules' homogeneity.

[0163] 2.3. Defining the optimal oligomerization format for SCT-MHC-I display on APVs

[0164] By evolutionary design, pMHC and TCRs form weak interactions with affinity in the low micromolar range. Interestingly, the formation of immunological synapses, which engage hundreds, if not thousands, copies of pMHC and TCR and costimulatory molecules, may overcomeWSGR Docket No. 48295-718.601 the weak interaction between pMHC and TCR complexes by taking advantage of the avidity effects of multivalent interaction. Because EVs and VLPs are significantly smaller than APCs, we aimed to optimize the density and oligomerization of displayed SCT-MHC-I to maximize their local and global avidity on vesicles, overcoming their size limitations and forming effective engagements with antigen-specific T cells. To this end, we designed SCT-MHC-I display vectors with varied oligomerization domains (Figure 28A). We then generated EV-based APVs displaying CD86 / D4, GITRL / NA, ICAM1 / D4, and ova-SCT-MHC-I without an oligomerization domain (VGTM, monomer) or incorporating a distinct oligomerization domain (Figure 28B). The oligomerization domains tested here include D4 (post-fusion trimeric domain from VSVG), cc-hex (coiled-coil hexamer domain), L24D (coiled-coil heterohexamer domain), L24H (coiled-coil heterohexamer domain), cc-Hept (coiled-coil heptamer domain), and wazOct (coiled-coil octamer domain). We sought to identify the optimal oligomerization format for SCT-MHC-I display while examining how varying oligomerized display of SCT-MHC-I on APVs might impact the strength and specificity of antigen-specific T cell activation, as well as the consequences for T cell proliferation.

[0165] To this end, we mixed naive OT-1 T cells (ova-specific, Ly5.2 positive) with nonspecific naive T cells from Ly5.1 mice (Ly5.1 or CD45.1 positive) and then stimulated mixed T cells with ova-APVs displaying ova-SCT with distinct oligomerization domains (Figure 29A). The effects of ova-APVs on T cell activation were determined by FACS analysis of early T cell activation marker CD69 and CD25 expression, and measurement of T cell proliferation. (Figure 29B, 29C, 29D) Such analyses allowed us to compare the observed effects of ova-APVs on the antigen-specific OT-1 (Ly5.2+) and non-specific T cells (Ly5.1) and helped to discern how oligomerization domains used for ova-SCT-MHC-I display affect antigen-specific T cell recognition and activation. We found that SCT-APVs displaying ova-SCT with various oligomerization domains can effectively stimulate expression of early T cell activation markers CD69 and CD25 in OT-1 T cells but with distinct potency and specificity (Figure 29B, 29C).

[0166] Notably, T cells stimulated with ova-APV / D4 (trimeric domain) induced CD69 expression in ~94% OT-1 T cells and both CD69 and CD25 expression in ~74% of OT-1 cells, 48 hours post-stimulation. By contrast, T cells stimulated with ova-APV with various coiled-coil oligomerization domains (Figure 29B, 29C) yielded significantly lower proportions of CD69+CD25+OT-1 cells, ranging from ~7% to ~20%, and T cells stimulated with ova- APV / VGTM (no SCT display oligomerization domain) resulted in approximately 44% of CD69+CD25+OT-1 cells. These results demonstrate that ova-APV / D4 are not only the most potent ova-APVs, but also more potent than anti-CD3 and CD28 Dynabeads at activating OT-1 T cells. Dynabead stimulation resulted in about 63% of CD69+OT-1 T cells and 52% of CD69+CD25+OT-WSGR Docket No. 48295-718.6011 cells. Moreover, ova-APV / D4 is significantly more specific than ova-APVs with other oligomerization domains, as indicated by the significantly lower percent of total CD69+Ly5.1 cells and of CD69+CD25+Ly5.1 cells after stimulation with ova-APV / D4 (Figure 29B, 29C). Stimulation with Dynabeads resulted in significant expression of CD69 and CD25 in both OT-1 and non-specific Ly5.1 T cells. Notably, OT-1 T cells activated by ova-APV / D4 proliferated over 20-fold, comparable to those stimulated with anti-CD3 and CD28 Dynabeads. In contrast, the nonspecific Ly5.1 T cells activated by ova-APV / D4 did not proliferate (Figure 29D). Finally, OT-1 T cells activated by ova-APVs with other oligomerization domains also proliferated more than 10- fold, whereas the non-specific Ly5.1 T cells did not proliferate. These results demonstrate that ova- APV / D4 is highly potent and specific at activating antigen-specific T cells based on the FACS analyses of activation markers and proliferation. Interestingly, ova-APVs with other oligomerization domains had varied effects on antigen-specific T cell proliferation, despite some having weaker effects on inducing T cell activation marker expression.

[0167] 2.4. Functional and physical properties of optimized base-APVs

[0168] In sections 2.2 and 2.3, we systematically screened for critical costimulatory molecules and optimal oligomerization domains required for building effective antigen-specific SCT-APVs. Based on these results, we determined the composition of optimal base-APVs consisting of displayed type I molecules, SCT / D4, CD86 / D4, ICAM1 / D4, and a type II molecule, GITRL / NA (Figure 30A, 30B). We designated this base-APV as APV(86IG). D4 trimeric domains and VGTM transmembrane anchors are used for the display of type I molecules, whereas NA tetrameric domain and its type II transmembrane anchor were used for the display of type II receptors. The composition of the APV(86IG) should simplify the production but also provide flexibility to add additional costimulatory molecules for new and improved functionality when required. Since CD80 and CD86 bind to CD28, we use displayed CD80 and CD86 interchangeably in producing SCT- APVs. To this end, we produced the optimized APV(86IG) on EVs or VLPs. To produce EV- based APV(86IG), we transfected DNA constructs encoding SCT / D4 and costimulatory molecules (CD80 / D4, ICAM1 / D4, GITRL / NA ) into S293-HLA null cells without lentiviral transfer and packaging vector (Figure 30C). Alternatively, to produce VLP -based APV(86IG), we transfected DNA constructs encoding SCT / D4 and costimulatory molecules (CD80 / D4, ICAM1 / D4, GITRL / NA ) into S293-HLA null cells together with lentiviral transfer and packaging vector (Figure 30C). EV- or VLP-based APV(86IG) are harvested from the supernatant, purified through a multi-step process to remove all protein and nucleic acid contaminants, and reconstituted in a stabilizing buffer for in vitro and in vivo usage.WSGR Docket No. 48295-718.601

[0169] We determined the sizes of EV-APV(86IG) and VLP-APV(86IG) by tunable resistive pulse sensing analysis (TRPS) using qNano (Figure 31 A, 31B). Their sizes are in the range of 189 or 193 nm, respectively. We then characterized the density of the displayed ova-SCT-MHCI molecules on EV-APV(86IG) and VLP-APV(86IG) by quantitative western-blot analyses (Figure 31C). We found that each EV-ova-APV(86IG) or VLP-ova-APV(86IG) particle displays approximately 6200 ± 1400 copies or 6500 ± 1700 copies of ova-SCT / D4VG fusion molecules (Figure 31C). We then characterized the density of the displayed costimulatory molecules on EV- APV(86IG) and VLP-APV(86IG) by quantitative Western-blot analyses (Figure 31D, 31E). We found that displayed costimulatory molecules, including CD80 / D4, ICAM1 / D4, and GITRL / NA are highly expressed on EVs or VLPs. Each EV-ova-APV(86IG) particle displays approximately 1400 ± 1300 copies of CD86 / D4, 2700 ± 1000 copies of ICAM1 / D4, and 1100 ± 500 copies of GITRL / NA. Furthermore, Each VLP-ova-APV(86IG) particle displays approximately 1100 ± 600 copies of CD86 / D4, 3100 ± 1700 copies of ICAM1 / D4, and 500 ± 40 copies of GITRL / NA. Overall, except that VLP-APV(86IG) slightly lower level of GITRL / NA, EV-APV(86IG) and VLP-APV(86IG) display comparable levels of ova-SCT / D4, CD80 / D4, ICAM1 / D4, and GITRL / NA.

[0170] We then examined the potency and specificity of EV- and VLP -based ova-APV(86IG) in activating antigen-specific OT-1 T cells (Figure 32). First, we compared the activation of OT-1 T cells by anti-CD3 / CD28 Dynabead, and EV- and VLP -based ova-APV(86IG) and found that they induced comparable levels of T cell activation at day 1 post T cell activation, as indicated by the levels of total CD69+ cells and CD69+CD25+ cells determined by FACS analyses (Figure 32A, 32B). Furthermore, we stimulated mixed T cells, consisting of naive OT-1 T cells (ova-specific, Ly5.2 positive) and non-specific naive T cells from Ly5.1 mice (Ly5.1 positive), with anti- CD3 / CD28 Dynabead and EV-based ova-APV(86IG). We then examined their effects on T cell activation by FACS analyses of T cell activation markers CD69 and CD25 among OT-1 (Ly5.2) and non-specific Ly5.1 T cells (Figure 32C, 32D, 32E). Such analyses helped to discern how oligomerization domains used for ova-SCT-MHC-I display affect the antigen-specific T cell recognition and activation. We found that over 55% of OT-1 T cells are CD69+CD25+ cells, whereas only ~3% of Anti-CD3 / CD28 Dynabeads. Thus, EV-based ova-APV(86IG) selectively activated OT-1 T cells but not non-specific Ly5.1 T cells. In contrast, anti-CD3 / CD28 Dynabeads are equally potent at activating OT-1 (Ly5.2) and non-specific Ly5.1 T cells (Figure 32C, D). Finally, EV-based ova-APV(86IG) differentially activates OT-1 T cells but not non-specific Ly5.1 T cells at various time points after stimulation (Figure 32E).WSGR Docket No. 48295-718.601

[0171] We then further examined the dose-dependent effects of EV-based ova-APV(86IG) and VLP -based ova-APV(86IG) on antigen-specific T-cell activation (Figures 33, 34). We stimulated mixed OT-1 cells and non-specific Ly5.1 T cells with EV-based ova-APV(86IG) or EV-based ova- APV(86IG) at particle-to-cell ratios of 5E4: 1, 1E4: 1, 1E3: 1, and 1E2: 1. Then we carried out FACS analyses to determine the expression of early T cell activation markers CD69 and CD25 at day 1, 2, and 4 post-stimulation. Notably, EV-based ova-APV(86IG) preferentially activated antigen-specific OT-1 T cells over the non-specific Ly5.1, as indicated by the total percent of CD69+ cells (Figure 33 A, 33B) and the percent of CD69+CD25+ cells (Figure 33C), particularly at the particle-to-cell ratios. In comparison, VLP -based ova-APV(86IG) also preferentially activated antigen-specific OT-1 T cells over the non-specific Ly5.1, as indicated by the total percent of CD69+ cells (Figure 34A, 34B) and the percent of CD69+CD25+ cells (Figure 34C), particularly at the particle-to-cell ratios. These results demonstrate that EV-based ova-APV(86IG) and VLP -based ova-APV(86IG) have similar activities in T-cell activation.

[0172] 2.5. Function of SCT-APVs displaying ova-peptide variants

[0173] To further characterize how displayed antigenic peptides may impact the function of SCT-APVs, we generated SCT-APVs displaying ova antigenic peptides with varied affinity to the OT-1 TCR. Here is a list of ova-peptide variants that may affect pMHC affinity with OT-1 TCR (Figure 35A). These peptides include (1) ova-wt, a strong agonist; (2) ova-A2, also a strong agonist; (3) ova-El, an antagonist; (4) ova-G4, a weak agonist; and (5) ova-K4, a null peptide. We found that SCT-APVs displaying different ova-peptides had distinct activities in activating OT-1 T cells, as indicated by their effects on the expression of early T cell activation markers CD69 and CD25 (Figure 35B, 35C). Notably, the total percent of CD69+ cells and the percent of CD69+CD25+ cells correlated with the strength and affinity of the ova-peptide variants. SCT- APVs displaying ova-wt peptides are most potent in increasing the total percent of CD69+ cells and the percent of CD69+CD25+ cells. In contrast, OT-1 T cells stimulated with all other SCT-APVs displaying various variants yielded a significantly lower percent of CD69+ cells and CD69+CD25+ cells. Moreover, SCT-APVs displaying ova-null peptide yielded essentially no CD69+CD25+ cells. Finally, the effects of ova-SCT-APVs presenting variant ova peptides on the proliferation of activated naive OT-1 T cells also correlate with the affinity of ova peptide variants to TCR and their ability to activate early T cell activation markers (Figure 35D). Unstimulated OT-1 T cells and T cells stimulated ova-K4-APVs did not proliferate significantly. In contrast, OT-1 T cells stimulated with SCT-APVs displaying strong agonist peptide, ova-wt, and ova-A2, proliferated the most, and OT-1 T cells stimulated with SCT-APVs displaying weak peptide, ova-El (antagonist) and ova-G4 (weak agonist) proliferated modestly. These results showed that the ability of SCT-WSGR Docket No. 48295-718.601APVs in T cell activation and proliferation correlated with the affinity of presented peptides to TCRs.

[0174] 2.6. Summary of the effective compositions of SCT-APVs for T cell activation

[0175] In section 2, we systematically explored the effective molecular compositions of SCT- APVs. All APVs displayed SCT-pMHC-I with an antigenic peptide, ICAM-1, and a varied combination of activating co-stimulatory molecules. For type I membrane proteins, the displayed proteins were fused to the post-fusion D4 trimeric domain from VSVG and then to the VSVG transmembrane and intracellular domain. For type II membrane proteins, the neuraminidase stem and tetrameric domain were fused to the extracellular domain of displayed type II proteins.

[0176] pMHC engagement with cognate TCR is the triggering event of the T cell activation process, representing the signal 1 of T cell activation. It dictates the specificity of antigen recognition by TCRs. Notably, the optimal affinity of pMHC and TCR complexes falls in the range of sub-micromolar, which permits pMHC to rapidly scan through billions of TCRs to find the antigen-specific T cells. However, proper pMHC and TCR engagement leads to the formation of the immunological synapses, which results in clustered and multivalent interaction of pMHC and TCR and is essential for T cell activation. Therefore, it is critical that EV- or VLP -based APVs can provide multivalent pMHC and TCR engagement that functionally mimics immunological synapses. To this end, we systematically explored the oligomerized display of SCT-MHC-1 on SCT-APVs with various oligomerization domains. We found that the D4 trimeric domain is most suitable for SCT-MHC-I display on SCT-APVs among the oligomerization domains tested. Notably, SCT-pMHC-I with the D4 trimeric domain were displayed at over 6000 copies / particles on EVs and VLPs.

[0177] Further, we systematically explore the display of costimulatory molecules on SCT- APVs. We were able to display most of the type I and type II costimulatory molecules at over 1000 copies / particles on EVs and VLPs. We found that displaying GITRL (GITR ligand) on SCT-APVs display is essential to enable robust and specific activation of antigen-specific T cells by SCT- APVs as indicated by FACS analyses of early T cell activation markers or by the stimulation of T cell proliferation. Notably, we also found that all costimulatory molecules tested contributed to some degree to the ability of SCT-SCT-APVs to stimulate T cells. However, we want to emphasize that the in vitro T cell activation assays used in characterizing the function of SCT-APVs are limited and may not reveal the function of the costimulatory molecules in many critical aspects of in vivo T cell biology, including lineage differentiation, migration, and memory formation. Collectively, we have demonstrated that EV- and VLP -based SCT-APVs displaying high-density and oligomerized SCT-MHC-1 and costimulatory molecules, including SCT-MHC-I, CD80 / CD86,WSGR Docket No. 48295-718.601 and GITRL (GITR ligand), provide robust and specific signals for antigen-specific T-cell activation in culture.Off-the-shelf SCT-APVs as preventative and therapeutic cancer vaccines

[0178] 3.1. SCT-APVs as preventative cancer vaccines

[0179] Given that SCT-APVs can activate antigen-specific T cells, we then examine whether SCT-APVs can be used to induce protective immunity against cancer in mice through direct activation of antigen-specific T cells in vivo. To this end, we designed SCT-APVs presenting various antigenic peptides from mouse melanoma antigens, including the gplOO peptides and its variant as examples of peptides derived from tumor-associated antigens, and the Hsf2_K72N peptide, as an example of a neoantigen peptide (Figures 36A). VLP-based gplOO-APVs and Hsf2_K72N-APVs were produced by transfecting S293 HLA null cells with the required SCT- MHC-I and costimulatory plasmids for base-APV (Figure 20). We then immunized mice with mixed SCT-APVs with 5E10 particles for each melanoma SCT-APV and injected them into mice intravenously following the immunization schedule depicted (Figure 36B). Concurrently, 10 ug of high molecular weight polylC (HMW-polylC) is also delivered through intramuscular injection. Both control and immunized mice received polylC (HMW-polylC). Immunized mice were challenged at day 35 post-primary immunization with B16F10 melanoma cells. VLP-based SCT- APVs displaying peptide-H2Kb SCT-pMHC-I, CD86, GITRL (GITR ligand), and ICAM1 were used in the study. 4 of the 5 control mice challenged with B16F10 melanoma developed tumors (Figure 36C). In contrast, 3 of the 10 APV-immunized mice challenged with Bl 6F 10 melanoma developed tumors (Figure 36D). These results demonstrate that immunization with SCT-APVs presenting tumor-associated antigens and neoantigens, which are designed to selectively activate antigen-specific T cells, can induce protective immunity against cancer development.

[0180] 3.2. SCT-APVs as therapeutic cancer vaccines

[0181] We designed studies to examine whether SCT-APVs can be used to treat established tumors (Figures 37, 38, 39). In the first study, we used ova-SCT-APVs to treat mice bearing ova- B16F0 tumors at day 7 or day 14 after tumor implantation (Figure 37). VLP-based APVs displaying ova-H2Kb SCT-pMHC-I, CD86, GITRL (GITR ligand), and ICAM1 were used in the study. SCT-APVs presenting an irrelevant peptide were used as a control. Mice bearing ova-B16F0 tumors were treated with 5E10 control or ova-APV particles through intravenous delivery twice at day 7 and 8 or once at day 14 (Figure 37 A). We found that ova- APVs significantly inhibited tumor growth with day 7 and 8 treatment and day 14 treatment (Figures 37B, 37C). In contrast, control SCT-APVs have no effects on tumor control (Figures 37B, 37C). These results demonstrated that SCT-APVs presenting tumor-specific antigenic peptides might be used to treatWSGR Docket No. 48295-718.601 mice bearing established tumors. It is important to note that the effects of tumor-specific SCT- APVs on tumor control may be further enhanced with TLR agonists, including polylC, CpG, and LPS.

[0182] Further, we tested whether SCT-APVs presenting ova variant peptides, which were shown to have distinct activities in T cell activation (Figure 35), may have different activity in tumor control (Figure 38). Again, VLP -based APVs displaying ova-H2Kb SCT-pMHC-I, CD86, GITRL (GITR ligand), and ICAM1 were used in the study. SCT-APVs presenting the K4_null peptide was used as a control. Mice bearing ova-B16F0 tumors were treated with 5E10 control or ova-APV particles through intravenous delivery twice on days 7 and 8 and on day 21 (Figure 38A). We found that SCT-APVs presenting ova-wt or strong agonist A2 peptide significantly inhibited tumor growth (Figures 38B, 38C). In contrast, SCT-APVs presenting the K4-null peptide have no significant effects on tumor control (Figures 38B, 38C). These results further support that the activity of SCT-APVs depends on the peptide presented on SCT-APVs and correlates well with the activity in T cell activation.

[0183] Finally, we used ova-APVs to treat mice bearing ova-MC38 tumors (Figure 39). VLP- based APVs displaying ova-H2Kb SCT-pMHC-I, CD86, GITRL (GITR ligand), and ICAM1 were used in the study. SCT-APVs presenting an irrelevant peptide were used as a control. Mice with ova-B16F0 tumors were treated with 5E10 control or ova-APV particles through intravenous delivery twice on days 7 and 8 (Figure 39A). We found that ova-APVs significantly inhibited ova- MC38 tumor growth (Figures 39B, 39C). In contrast, control APVs and K4-null ova-APV have no significant effects on tumor control (Figures 39B, 39C). These results demonstrated that SCT- APVs presenting tumor-specific antigenic peptides might be used to treat mice bearing established tumors. It is important to note that the effects of tumor-specific SCT-APVs on tumor control may be further enhanced with TLR agonists, including polylC, CpG, and LPS.

[0184] 3.3. Off-the-shelf SCT-APVs for patients with distinct HLA alleles

[0185] The above results demonstrate that SCT-APVs can activate antigen-specific T cells and induce anti-tumor immunity for cancer prevention and treatment. These findings illustrate the potential use of SCT-APVs presenting viral or cancer antigens as preventative or therapeutic cancer vaccines. To realize these potentials, we will build SCT-APVs to present viral and cancer antigens for human patients (Figure 40). The human leukocyte antigen (HLA) system is the human version of the major histocompatibility complex (MHCI found in many animal. The HLAs, corresponding to MHC class I, present peptides from inside the cells. There are three major and three minor MHC class I genes in HLA, including major MHC class I, HLA- A, HLA-B, and HLA-C, and minor genes, HLA-E, HLA-F, and HLA-G. These genes are highly variable, with thousands of variantWSGR Docket No. 48295-718.601 alleles being detected in humans for HLA-A, HLA-B, and HLA-C. HLAs form a heterodimer with P2m, which presents antigenic peptides from endogenous proteins, foreign viral proteins, or endogenous mutated proteins. The foreign or mutated antigens presented by HLAs engage and activate cytolytic T cells and lead to the killing of target cells. Since each cancer patient has a unique HLA profile and TCRs restricted by the HLAs, SCT-APVs with matching HLA alleles to patients will be used for in vivo and ex vivo activation of antigen-specific T cells. A collection of HLA alleles is selected to cover the majority of the Caucasian, African American, and Chinese populations (Figure 40A). We then designed the SCT-pMHC-I display constructs for the selected HLA alleles, which consist of an antigenic peptide (i.e. NY-ESO-1) linked to P2m, ectodomain of the selected HLA alleles, D4 or Lie trimeric domain, and VSVG transmembrane and cytosolic domain (TABLE 10). We will then produce optimized base-APVs consisting of displayed SCT- HLA, CD86 / CD80, ICAM1, and GITRL / NA on EVs or VLPs (Figure 40B). To produce EV- based APVs, we transfect DNA constructs encoding SCT-HLA and costimulatory molecules (CD80 / 86, ICAM1, and GITRL) into S293-HLA null cells without lentiviral transfer and packaging vector (Figure 40B). Alternatively, to produce VLP -based APVs, we transfect DNA constructs encoding SCT-HLA and costimulatory molecules (CD80 / 86, ICAM1, and GITRL / NA ) into S293- HLA null cells together with lentiviral transfer and packaging vector (Figure 40B). EV- or VLP- based APVs are harvested from the supernatant, purified through a multi-step process to remove all protein and nucleic acid contaminants, and reconstituted in a stabilizing buffer. These SCT-APVs may be used to activate cognate antigen-specific T cells in vitro and in vivo (Figure 40C).

[0186] 3.4. Off-the-shelf SCT-APVs for viral and cancer antigens

[0187] Peptide antigens for many known viral and tumor antigens have been identified and characterized (TABLES, 2, 4, 5). Viral antigens, including those from HBV or HCV, E6 and E7 of human papillomavirus, and large T and small T antigens of Merkel polyomavirus, have been known to cause hepatocellular carcinoma (HCC), anogenital cancer, and head and neck cancer, and skin cancer, respectively (TABLES, 2, 4). Many mutated antigens, including EGFRvIII mutations, KRAS mutations (G12C, G12D, and G12V), and BRAF(V600E) mutations, are known to be cancer drivers that can lead to lung, brain, colorectal, pancreatic, and blood cancers (TABLES 2, 5). Furthermore, many tumor-specific antigens, including NYE-ESO-1, MAGE-A3, MAGE-A1, and many other cancer-testis antigens, are known to be highly specific for a variety of cancers (TABLES 2, 5). Finally, differentiated antigens and over-expressed antigens, such as CEA, GP100, GPC3, and WT1, are also associated with a variety of cancers (TABLES 2, 5). More importantly, the antigenic peptides from these cancer antigens were well -characterized for pMHC presentation. Therefore, it is highly feasible to build a collection of SCT-APVs for these antigenicWSGR Docket No. 48295-718.601 peptides as the potential off-the-shelf SCT-APVs for in vivo and ex vivo activation of antigenspecific T cells. EV- or VLP -based APVs may be produced by transfect DNA constructs encoding SCT-HLA and costimulatory molecules (CD80 / 86, ICAM1, and GITRL) into S293-HLA null cells with or without lentiviral transfer and packaging vector (Figure 40B). These SCT-APVs may be used to activate cognate antigen-specific T cells in vitro and in vivo (Figure 40C). Also important to note, SCT-APVs may be produced to present one single type of antigenic peptide by transfection of an SCT-HLA display vector encoding one antigenic peptide only (Figure 41A) or to present multiple types of antigenic peptides by transfection of a combination of SCT-HLA display vectors each encoding one antigenic peptide (Figure 41B). The combination of peptides may be from one protein antigen or multiple protein antigens from the same tumor type.

[0188] 3.5. Ex vivo and in vivo T cell therapy with Off-the-shelf SCT-APVs

[0189] These SCT-APVs may be used to activate autologous T cells in vitro to generate antitumor T cells for autologous T cell therapy (Figure 42A) or activate autologous T cells in vivo as preventative and therapeutic vaccines to treat cancer and virus infection (Figure 42A). For in vivo T cell therapy with SCT-APVs, antigen-specific SCT-APVs may be combined with TLR agonists, including polylC, CpG, and LPS, and inflammatory cytokines, such as IL-2 and interferon-gamma, to provide the danger signals and inflammatory signals to coordinate with the antigen-specific T cell activation signals from the tumor-specific SCT-APVs.

[0190] 3.6. Combination therapy with SCT-APVs and immune checkpoint blockade

[0191] These SCT-APVs may be used in combination with immune checkpoint blockade therapies, such as anti-PDl and anti-CTLA-4 treatment for cancer. Successful antibody -based treatments have been developed to unblock the inhibitory signals exerted onto previously activated tumor-targeting T cells through the PD-1 receptor interaction with the inhibitory ligand PD-L1 expressed on tumor cells or other suppressor cells (Figure 43A). These therapies are highly successful in clinics and have brought cures to many cancer patients. However, the response rate to anti-PDl / Ll treatment in a variety of cancer types remains limited, ranging from low single digits to as much as 26%. There has been limited success in increasing the response rate and bringing cures to more cancer patients despite more than a decade of attempts to identify additional blockade targets to synergize with anti-PDl / Ll treatment. Interestingly, although tumor cells express a large number of tumor-specific antigens, including viral antigens, mutated antigens, tumor-specific antigens, differentially expressed or over-expressed antigens, and neoantigens (TABLES 2, 4, 5), tumor cells may actively tolerize anti-tumor T cells and render them inactive (Figure 43B). Tolerized T cells may have seen the presented tumor antigens but are tolerized because of lacking costimulatory signals and / or danger signals from the innate immune cells. Thus, an alternativeWSGR Docket No. 48295-718.601 explanation for why we seem to have reached the ceiling of immune checkpoint blockade therapy is that immune checkpoint blockade would have no effect under such a scenario in that, tumorspecific T cells are not activated and cannot be activated by unblocking the inhibitory signals (Figure 43B). To overcome the tolerance or lack of proper activation of anti-tumor T cells in many cancer patients, a library of SCT-APVs with matching HLAs with patients presenting a library of antigenic peptides for tumor antigens, including viral antigens, mutated tumor antigens, tumorspecific antigens, differentiated and overexpressed antigens, and neoantigens, will be injected into the patients with TLR agonists. We will then infuse patients with anti-PDl / Ll or anti-CTLA-4 for immune checkpoint blockade to prevent activated T cells from being suppressed by cancer cells or suppressive tumor microenvironment (Figure 43C). Combined anti-tumor SCT-APVs and anti- PDl / Ll or anti-CTLA-4 treatment will enable curative cancer immunotherapy for the majority of cancer patients who still do not respond to anti-PDl / Ll or anti-CTLA-4 immune checkpoint blockade therapy (Figure 43D).MATERIALS

[0192] 4.1. Experimental Animals

[0193] All mice were bred and housed in a pathogen free vivarium under IACUC animal care requirements. Animals were kept under a 12-hour light and dark cycle at a controlled temperature (20-22°C). 6-8-week-old male or female BL / 6 (C57BL / 6, strain code 027), BALB / c (strain code 028), female Ly5.1 (strain code 494) and athymic nude mice (strain code 490) were purchased from Charles River Laboratories. 6-8-week-old female OT-1 mice (strain code 003831) were purchased from Jackson Laboratory.

[0194] 4.2. Cell Lines

[0195] HEK 293T cells (ATCC CRL-3216), MC38 murine colon carcinoma cells (Sigma) and B16F0 murine melanoma cells (ATCC CRL-6322) were cultured in DMEM supplemented with 10% FBS and 1% PS. EO771 murine breast carcinoma cells (CRL-3461) were cultured in DMEM supplemented with 20mM HEPES (Fisher), 10% FBS and 1%PS. LLC murine lung carcinoma cells (CRL-1642) were cultured in DMEM / F12 (Fisher) supplemented with 10% FBS and 1%PS. Panc02 murine ductal adenocarcinoma cells (gifted), CT26 murine colon carcinoma cells (CRL- 2638) and 4T1 murine metastatic breast cancer cells (CRL-2539) were cultured in RPMI 1640 (Fisher) supplemented with 10% FBS and 1% PS. Renca renal adenocarcinoma cells (CRL-2947) were cultured in RPMI 1640 supplemented with 10% FBS, 1% PS, ImM non-essential amino acids (Fisher), ImM sodium pyruvate (Fisher) and 2mM L-Glutamine (Gibco).METHODSWSGR Docket No. 48295-718.601

[0196] 5.1 SCT-APV displaying constructs

[0197] Codon-optimized sequences for single-chain trimer MHC (SCT), adhesion molecule ICAM-1 and co-stimulatory (costim) molecules GITRL (GITR ligand) and CD86 were synthesized (Twist) and cloned into a display construct to create fusion peptides consisting of the extracellular domain of a SCT or a costim molecule and a display anchoring protein. To generate SCT-APVs displaying monomeric SCT and costim molecules, the extracellular domains of SCT or costim molecules were fused to a synthetic VSV-G sequence encoding the transmembrane and cytoplasmic tail domains. To generate SCT-APVs displaying oligomerized SCT or costim molecules, the extracellular domains of SCT or costim molecules were fused to a synthetic VSV-G sequence encoding the D4 post-fusion trimerization domain or other oligomerization domains and the transmembrane and cytoplasmic tail domains from VSVG.

[0198] 5.2 Costimulatory molecule displaying constructs

[0199] Codon-optimized sequences for co-stimulatory molecules (See TABLE 9) were synthesized (Twist). Expression constructs encoding the full, wild type costimulatory molecule or a fusion display peptide were then designed for each co-stimulatory molecule. Depending on the structure of the costimulatory molecule, the fusion display constructs for each were designed as follows. Monomeric display constructs for type I transmembrane proteins were comprised of the ectodomain of the costimulatory molecule fused to the transmembrane and cytoplasmic tail domains of VSV-G protein. Trimeric display constructs for type I transmembrane protein fusion constructs were comprised of the ectodomain of the costimulatory molecule fused to the D4 postfusion trimerization domain or other oligomerization domains and the transmembrane and cytoplasmic tail domains of VSV-G protein. Oligomeric display constructs for type II transmembrane proteins were comprised of the cytosolic tail, transmembrane regions and short stem of Influenza Neuraminidase protein fused to the ectodomain of the costimulatory molecule.

[0200] 5.3 Production of SCT-APVs based on VLPs or extracellular vesicles

[0201] SCT-APVs based on VLPs, or extracellular vesicles can be produced from transfected 293 or 293T cells with endogenous HLA alleles knocked out. To produce lentiviral-VLP based SCT-APVs with viral genomes, we co-transfected HLA KO 293T cells with an SCT construct, adhesion molecule construct, costim-display constructs, a lentiviral packaging vector (i.e. psPAX2), and lentiviral genome transfer vector. To produce lentiviral-VLP based SCT-APVs without viral genomes, we co-transfected HLA KO 293T cells with SCT, adhesion and costim- displaying constructs along with a lentiviral packaging vector (i.e. psPAX2). Finally, to produce extracellular vesicle-based SCT-APVs, we transfected producer cells with only SCT, adhesion molecule and costim displaying constructs. SCT-APV supernatant collection was performed at 48WSGR Docket No. 48295-718.601 hours. Supernatant collections were then pooled, PEG precipitated and purified by size exclusion chromatography (Sigma Aldrich).

[0202] 5.4 Production of off-the-shelf SCT-APVs presenting mixed antigenic peptides

[0203] To produce lentiviral-VLP based SCT-APVs with viral genomes presenting mixed antigenic peptides, we co-transfected HLA KO 293T cells with multiple SCT constructs encoding distinct antigenic peptides displayed on SCTs, adhesion molecule construct, costim-display constructs, a lentiviral packaging vector (i.e. psPAX2), and lentiviral genome transfer vector. To produce lentiviral-VLP based SCT-APVs without viral genomes, we co-transfected HLA KO 293 T cells with the desired mixture of distinct SCT constructs, adhesion and costim-displaying constructs along with a lentiviral packaging vector (i.e. psPAX2). Finally, to produce extracellular vesiclebased SCT-APVs, we transfected producer cells with only the desired mix of SCT constructs, adhesion molecule and costim displaying constructs.

[0204] 5.5 SCT-APV particle quantification via p24 ELISA and Tunable Resistive PulseSensing.

[0205] P24 concentrations in VLP -based SCT-APV samples were determined using an HIV p24 SimpleStep ELISA kit (Abeam) per the manufacturer’s protocol. Concentrations of lentiviral SCT-APV particles were extrapolated from the assumption that each lentiviral particle contains approximately 2000 molecules of p24, or 1.25 x 104SCT-APVs per picogram of p24 protein.

[0206] EV-based APV concentrations and VLP-based APV concentrations determined via p24 ELISA were corroborated by tunable resistive pulse sensing via Exoid (TRPS, IZON). Purified SCT-APVs were diluted in 0.2pm filtered PBS with 0.03% Tween-20 (Thermo Fisher Scientific) prior to Exoid analysis. Concentration and size distributions of SCT-APV particles were then determined using an NP200 nanopore at a 47.5mm stretch, and applied voltages between 0.5 and 0.7V were used to achieve a stable current of 120nA through the nanopore. Measurements for each SCT-APV sample were taken at pressures of 100, 300 and 500 Pa, and considered valid if at least 500 events were recorded, particle rate was linear, and root mean squared signal noise was maintained below 10 pA. SCT-APV concentrations were then determined by comparison to a standardized multi-pressure calibration using CPC200 (mode diameter: 200nm) (IZON) carboxylated polystyrene beads diluted 1 :200 in 0.2pM filtered PBS from their original concentration of 8.3 x 1011particles per / mL. Measurements were analyzed using IZON Control Suite 3.4 software to determine original sample concentrations.

[0207] 5.6 Western blot analysis of SCT-APVs

[0208] Expression of SCT, adhesion molecules and costim molecules on SCT-APVs was confirmed via western blot analysis of purified particles. Samples of purified SCT-APVs wereWSGR Docket No. 48295-718.601 lysed at 4°C for 10 minutes with cell lysis buffer (Cell Signaling) before being mixed with NuPage LDS sample buffer (Thermo Fisher Scientific) and boiled at 95°C for 5 minutes. Differences in oligomerization were determined by running samples in reducing and non-reducing conditions. Under reducing conditions, 5% 2-Mercaptoethanol (Thermo Fisher Scientific) was added to samples to dissociate oligomerized IC-MVPs. Protein samples were then separated on NuPAGE 4- 12% Bis-Tris gels (Thermo Fisher Scientific) and transferred onto a polyvinylidene fluoride (PVDF) membrane (Life Technologies). PVDF membranes were blocked with TRIS-buffered saline with Tween -20 (TBST) and 5% skim milk (Research Products International) for 1 hour, prior to overnight incubation with primary antibody diluted in 5% milk. Staining for CD86, GITRL (GITR ligand) and ICAM-1 display peptides expressing VSV-G tag was done using an anti-VSV-G epitope tag rabbit polyclonal antibody (BioLegend, Poly29039) at a 1 :2000 dilution. Staining for SCT was done using an anti-HA epitope tag goat polyclonal antibody () at a 1 :2000 dilution. The following day, the PVDF membrane was washed 3 times with IX TBST and stained with a goat- anti-rabbit secondary antibody (IRDye 680) at a 1 :5000 dilution for 60 minutes in 5% milk. Postsecondary antibody staining, the PVDF membrane was again washed 3 times with TBST before imaging on a Licor Odyssey scanner.

[0209] Alternatively, quantitative western blot analyses were performed using an automated Simple Western size-based protein assay (Protein Simple) following the manufacturer’s protocols. Unless otherwise mentioned, all reagents used here are from Protein Simple. Concentrated samples were lysed as described above, before being diluted 1 : 10 in 0. IX sample buffer for loading on capillaries. ICAM-1, CD86 and GITRL (GITR ligand) fusion protein expression levels were identified using identical anti-VSV-G epitope tag rabbit polyclonal primary antibody at a 1 :200 dilution and an HRP conjugated anti-rabbit secondary antibody (Protein Simple). SCT fusion protein expression levels were identified using the same anti-HA epitope tag goat polyclonal primary antibody at a 1 :50 dilution, and an HRP conjugated anti -goat secondary antibody (Protein Simple). A synthetic protein standard of known concentration comprising the same target epitope tags was used to generate a standard curve. Chemiluminescence signal analysis was performed using Compass software (Protein Simple). Standard curve analysis of synthetic protein standards and interpolation of sample chemiluminescence signals to determine exact copy numbers was done using GraphPad Prism (Version 10).

[0210] 5. 7 Preparation of Naive v.s Antigen Experienced Mouse T cells

[0211] Whole spleens were harvested from sacrificed mice and dissociated by mashing tissue over a 70pM cell strainer into a 50mL conical tube using a sterile syringe plunger. Dissociated splenocytes were washed down using cold, sterile FACS buffer. Total splenocytes were countedWSGR Docket No. 48295-718.601 using a hemacytometer, and spun down at 500 x g, 4°C, for 5 minutes. Supernatant was aspirated and cells were resuspended at IxlO8cells / mL. Naive murine T cells were then isolated from total mouse splenocytes using EasySep Pan-Naive T Cell isolation kits (Stemcell), per the manufacturer’s instructions.

[0212] To isolate antigen-experienced murine T cells, pan-mouse T cells were first isolated from total mouse splenocytes using EasySep Mouse T Cell isolation kits (Stemcell). Purified pan T cells were resuspended in cold FACS buffer at a concentration of IxlO7cells / mL and stained with BV650 anti-mouse / human CD44 (BioLegend), APC anti-mouse CD62L (BioLegend) and antimouse CD25 antibodies at a 100-fold dilution. Antigen experienced T cells (CD44+CD25+) and naive T cells (CD44 CD62L+CD25 ) were sorted into separate tubes.

[0213] 5.8 Preparation of Naive v.s Antigen Experienced Human T Cells

[0214] Peripheral blood monocytes (PBMCs) from HLA-type matched donors were recovered from liquid nitrogen storage and cultured in plain X-VIVO15 medium with gentamicin (Lonza) at 37°C, 5% CO2 for 24 hours. T cells were then purified from PBMCs using an EasySep Human T Cell Isolation Kit (Stemcell) as per the manufacturer’s instructions. Stained pan-T cells were then centrifuged again and resuspended at a concentration of 5x106 cells / mL in cold FACS buffer. Antigen experienced (CD45RO+CD25+) and naive (CD45RO CD25 ) human T cells were then sorted into separate tubes via SH800 cell sorter (Sony Biotechnology).

[0215] 5.9 In vitro T cell activation (TCA) of murine T cells stimulated by SCT-APVs

[0216] Naive mice were sacrificed, and spleens were harvested and dissociated by mashing tissue over a 70pM cell strainer into a 50mL conical tube using a sterile syringe plunger.Dissociated splenocytes were washed down using cold, sterile FACS buffer. Total splenocytes were counted using a hemacytometer, and spun down at 500 x g, 4°C, for 5 minutes. Desired naive or pan-T cells were then isolated using EasySep negative selection mouse T cell isolation kits (Stemcell) according to the manufacturer’s protocol. For mixed T cell activation studies, antigenspecific T cells from transgenic mice (OT-1) and non-specific T cells from Ly5.1 mice were then mixed prior to seeding. SCT-APVs were diluted to desired concentrations in plain RPMI. Untreated T cells and T cells mixed with mouse T-Activator CD3 / CD28 Dynabeads (Thermo Fisher) served as negative and positive controls, respectively. Dynabeads were added at a 1 : 1 T cell to bead ratio.

[0217] T cell activation was determined 24 hours and 48 hours post-stimulation. T cells were stained with BV421 anti -mouse TCR-B or anti -mouse CD3 (BioLegend), FITC anti -mouse CD8a (BioLegend), BV570 anti-mouse CD4 (BioLegend), APC / Cyanine7 anti-mouse CD69 (BioLegend), and PerCP anti-mouse CD25 antibodies (BioLegend). When appropriate, Ly5.1 T cells were additionally stained with Pacific Blue anti-mouse CD45.1 antibodies (BioLegend) toWSGR Docket No. 48295-718.601 differentiate them from CD45.2 T cells. Stained cells were analyzed for early T cell activation marker expression via SP6800 Spectral Analyzer.

[0218] 5.10 In vitro T cell activation (TCA) of human T cells by SCT-APVs

[0219] Naive or antigen-experienced human T cells were purified from HLA-typed PBMCs. Human SCT-APVs were diluted to the desired concentrations in Human T cell growth activation medium. Untreated human T cells and T cells mixed with human T-Activator CD3 / CD28 Dynabeads (Thermo Fisher) served as negative and positive controls, respectively. Dynabeads were added at a 1 : 1 T cell to bead ratio. T cell activation then was determined 24 hours and 48 hours post-stimulation. Cells were stained with PE anti-human TCR-a / B (BioLegend), APC anti-human CD4 (BioLegend), Pacific Blue anti-human CD8a (BioLegend), FITC anti-human CD69 (BioLegend), PE-Cyanine7 anti-human CD25 (BioLegend). Stained T cells were analyzed for early T cell activation marker expression via SP6800 Spectral Analyzer.

[0220] 5.11 In vitro T cell proliferation and viability assay

[0221] Human or mouse T cells stimulated with SCT-APVs were transferred to 96-well U- bottom plates and centrifuged at 500 x g, 5 min, 4°C to remove medium. T cells were then washed with 200pL cold FACS buffer before being stained with PE anti-human TCR-a / B (BioLegend) or anti-mouse TCR-B (BioLegend) and propidium iodide at 12.5pg / mL in FACS buffer (Thermo Fisher) at 4°C in the dark for 15 minutes. 50pL of AccuCount Blank beads (Spherotech) were then added to T cells, and T cell viability and cell number was assessed via FACS using an SP6800 spectral analyzer. T cell concentration was determined via AccuCount Blank particle count as per the manufacturer’s instructions.

[0222] 5.11 In vivo prophylactic melanoma tumor vaccination with SCT-APV

[0223] Naive BL / 6 mice were immunized with VLP -based SCT-APVs displaying SCTs with known melanoma neoantigens. A total of 5E10 SCT-APVs displaying each neoantigen were mixed and delivered to each mouse via intravenous retro-orbital injection. lOpg of poly(LC) was simultaneously delivered via intramuscular injection. A second identical dose of VLP -based SCT- APVs and poly(LC) was delivered to each mouse 21 days after the initial vaccination. At day 35 post-first dose, naive mice and mice vaccinated with a cocktail of SCT-APVs were then injected subcutaneously with Bl 6F 10 cells and observed for tumor growth.

[0224] 5.12 In vivo treatment of established ova tumors with ova-SCT-APVs

[0225] To generate tumor recipients with established ova tumors, naive BL / 6 mice were implanted subcutaneously with ova-B16F10 cells or ova-MC38 tumor cells. At day 7-post tumor implantation, mice were randomized and treated with ova-APVs delivered intravenously via retro- orbital injection. A second dose of ova-APVs was delivered on day 8 post-tumor implantation, andWSGR Docket No. 48295-718.601 subsequent doses were given biweekly. Tumor growth and survival proportions in treated and untreated mice was monitored on a daily basis.

[0226] 5.13 Generation of murine tumor-activated T cells (tac-T) using murine SCT-APVs

[0227] Murine Pan-T cells were purified from splenocytes of SCT-APV-vaccinated mice using Easy Sep Mouse T Cell Isolation Kits (Stemcell) as per the manufacturer’s instructions. Purified pan-T cells were then co-cultured with non-specific SCT-APVs or tumor-specific SCT-APVs for 24 hours in T cell Priming Medium. After 24 hours of exposure to SCT-APVs, T cells were then cultured in T cell Expansion medium. At the endpoint of the assay, antigen-specific cytolytic activity of activated, expanded tac-T cells against target tumor cells was assessed as described below. Unstimulated pan-T cells, pan-T cells stimulated via mouse T activator Dynabeads, and pan-T cells exposed to wild-type tumor cells were used as negative controls for tac-T expansion and cytolytic activity.

[0228] 5.14 Generation of human tumor-activated T cells (tac-T) using human SCT-APVs

[0229] Human pan-T cells were isolated from HLA-typed PBMCs using EasySep Human T Cell Isolation Kits (Stemcell) as per the manufacturer’s instructions. Purified human pan-T cells were then co-cultured with non-specific SCT-APVs or tumor-specific SCT-APVs. Activated tac-T cells were expanded, and T cell activation and viability were assessed at day 5 and day 9 post-SCT- APV exposure. Antigen-specific cytolytic activity of expanded human tac-T was assessed as described below. Unstimulated human pan-T cells, pan-T cells stimulated via human T activator Dynabeads, and pan-T cells exposed to wild-type tumor cells were used as negative controls for tac-T expansion and cytolytic activity.

[0230] 5.15 In vitro CTL activity of mouse and human tac-T cells

[0231] Target tumor cells were labelled with CellTrace Violet dye (CTV). CTV-stained cells were then mixed with serial diluted tac-T cells. Unstimulated mouse pan-T cells, pan-T cells stimulated with T Activator Dynabeads, and pan-T cells co-cultured with wild type tumor cells were similarly added to target tumor cells as negative controls. Target cells and T cells were then incubated overnight at 37°C, 5% CO2. Post-incubation, all cells were transferred to 96-well, U bottom plates and centrifuged to remove medium. Cells were then stained with BV421 anti -mouse CD3 (BioLegend). The number of live vs dead target cells (CTV+, PI+ / ) was determined via FACS analysis using an SP6800 spectral analyzer. Cytolytic activity of each T cell population at each ET ratio was determined and plotted using GraphPad Prism.

[0232] 5.16 In vivo tumor control by adoptive transfer of tac-T cells

[0233] Established wild-type tumors were generated in mice, as described previously. Tumorspecific tac-T cells were expanded as described previously. At the specified timepoints post-tumorWSGR Docket No. 48295-718.601 implantation, IxlO6tac-T cells were adoptively transferred to mice via RO injection. Tumor growth and mouse survival were then monitored on a biweekly and daily basis, respectively.

[0234] 5.17 Combination in vivo tumor treatment with SCT-APVs and anti-PD-l / Ll immune checkpoint blockade

[0235] Mice with established tumors were generated as described previously. Tumor-specific SCT-APVs were delivered intravenously as described previously. Anti-PD-1 or anti-PD-Ll immune checkpoint blockade monoclonal antibodies were injected via intraperitoneal (IP) injection. ICB doses were given once weekly, beginning at day 7 post-tumor implantation. Treated and untreated mice were monitored for survival and tumor growth.

Claims

WSGR Docket No. 48295-718.601CLAIMSWHAT IS CLAIMED IS:

1. An antigen-presenting vesicle (APV) comprising a first fusion protein that comprises a Single-chain trimer peptide major histocompatibility complex class I (SCT-pMHC-I) molecule and a second fusion protein that comprises a costimulatory ligand molecule, wherein the first or second fusion protein is expressed at a valency of at least about 100 copies on a surface of the antigen-presenting vesicle (APV).

2. The antigen-presenting vesicle (APV) of claim 1, wherein the SCT-pMHC-I molecule binds specifically to an antigen-specific T cell receptor.

3. The antigen-presenting vesicle (APV) of claim 1 or 2, wherein the SCT-pMHC-I molecule comprises an antigenic peptide, P-2-microglobulin (P2m), and an MHC-I alpha chain.

4. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the antigenic peptide binds to the SCT-pMHC-I alpha chain and is fused to P2m.

5. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the SCT-pMHC-I molecule comprises a single-chain trimer (SCT) peptide MHC-I molecule (SCT- pMHC-I).

6. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the alpha chain of SCT-pMHC-I comprises an amino acid sequence with at least 95% sequence identity to a human leukocyte antigen (HLA) HLA allele or a mouse MHC-I allele according to any one of SEQ ID NOs: 1-20.

7. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the P2m comprises an amino acid sequence with at least 95% sequence identity to an amino acid sequence of human and mouse P2m according to SEQ ID NO: 21 or 22.

8. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the antigenic peptide is derived from a disease-associated antigen, a model antigen, a viral antigen, a mutated tumor antigen, a tumor-specific antigen, a differentiated antigen, an over-expressed antigen, or a neoantigen.

9. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the first fusion protein comprises at least two, three, four, five, six, seven, eight, nine, ten, or more than ten separate fusion proteins, wherein the at least two, three, four, five, six, seven, eight, nine, ten, or more than ten separate fusion proteins each comprises a different antigenic peptide.

10. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the antigenic peptide is selected from the group comprising ovalbumin (Ova), gplOO, Hsf2-K72N, an HB V or HCV viral protein, an E6 or E7 of human papillomavirus, a large T and small T proteinWSGR Docket No. 48295-718.601 of Merkel polyomavirus, KRAS, BRAF, BRAC1, BRAC2, NYE-ESO-1, MAGE-A1, MAGE-A3, CEA, gpl00 / Pmell7, PAP, HER2 / neu, GPC3, and WTI.

11. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the antigenic peptide comprises one or more listed in Tables 2, 4, and 5.

12. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the antigenic peptide comprises an amino acid sequence according to any one of SEQ ID NOs: 59- 66, and 149-544.

13. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the costimulatory ligand molecule binds specifically to a T-cell surface costimulatory receptor.

14. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the second fusion protein comprises at least two, three, four, five, six, seven, eight, nine, ten, or more than ten separate fusion proteins, wherein the at least two, three, four, five, six, seven, eight, nine, ten, or more than ten separate fusion proteins each comprises a different costimulatory ligand molecule.

15. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the costimulatory ligand molecule comprises GITRL (GITR ligand), CD80, CD86, OX40L (0X40 ligand), CD48, CD30L (CD30 ligand), ICOSL (ICOS ligand), CD70, CD40, CD40L, 4-1BBL (4- 1BB ligand), TLIA, TIM4, SLAM, CD48, CD58, CD155, CD112, LIGHT, or ICAM-1.

16. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the costimulatory ligand molecule comprises an amino acid sequence with at least 90% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58.

17. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the first or second fusion protein further comprises a transmembrane domain.

18. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the transmembrane domain comprises the transmembrane domain from VSV-G, Dengue E protein, influenza hemagglutinin, influenza neuraminidase, spike protein SI, spike protein S2, Sindbis virus envelope (SINDBIS) protein, hemagglutinin envelope protein from measles virus, envelope glycoprotein of measles virus fusion (F) protein, RD114, BaEV, GP41, GP120, or the wild-type transmembrane domain of costimulatory ligand molecules.

19. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the VSV-G comprises a transmembrane domain and cytoplasmic tail.

20. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the transmembrane domain of the fusion proteins comprises an amino acid sequence with at least 90% sequence identity to an amino acid sequence of SEQ ID NOs: 67-86.WSGR Docket No. 48295-718.60121. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the first or second fusion protein further comprises an oligomerization domain.

22. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the oligomerization domain of the fusion proteins is a dimerization domain, a trimerization domain, a tetramerization domain, a hexamerization domain, a heterohexamerization domain, a heptamerization domain, or an octamerization domain.

23. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the oligomerization domain of the fusion proteins comprises a leucine zipper dimerization domain, a post-fusion oligomerization domain of viral surface protein, a D4 post-fusion trimerization domain of VSV-G protein, a Dengue E protein post-fusion trimerization domain, a foldon trimerization domain, a coiled coil domain, a LINE1 retrotransposon coiled coil domain (Ll cc), a collagen XV coiled coil domain (CXVcc), a tetranectin coiled coil domain (TNTNcc), a cc-hex domain, a L24D domain, a L24DH domain, a cc-Hept domain, a wazOct domain, or an influenza neuraminidase stem domain.

24. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the oligomerization domain of the fusion proteins comprises an amino acid sequence that has at least 90% sequence identity to an amino acid sequence according to SEQ ID NOs: 87-101.

25. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein the APV is capable of activating or stimulating a cognate antigen-specific T cell.

26. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein when the first or second fusion protein is expressed on the surface of the antigen-presenting vesicle (APV), the oligomerization domain is outside of the antigen-presenting vesicle (APV).

27. The antigen-presenting vesicle (APV) of any one of the preceding claims, wherein when the first or second fusion protein is expressed on the surface of the antigen-presenting vesicle (APV), the oligomerization domain is inside of the antigen-presenting vesicle (APV).

28. The antigen-presenting vesicle (APV) of any one of claims 1-27, wherein the first or second fusion protein is expressed at a valency of about 100 copies on a surface of the antigen- presenting vesicle (APV).

29. The antigen-presenting vesicle (APV) of any one of claims 1-27, wherein the first or second fusion protein is expressed at a valency of about 250 copies on a surface of the antigen- presenting vesicle (APV).

30. The antigen-presenting vesicle (APV) of any one of claims 1-27, wherein the first or second fusion protein is expressed at a valency of about 500 copies on a surface of the antigen- presenting vesicle (APV).WSGR Docket No. 48295-718.60131. The antigen-presenting vesicle (APV) of any one of claims 1-27, wherein the first or second fusion protein is expressed at a valency of about 1000 copies on a surface of the antigen- presenting vesicle (APV).

32. The antigen-presenting vesicle (APV) of any one of claims 1-27, wherein the first or second fusion protein is expressed at a valency of about 2000 copies on a surface of the antigen- presenting vesicle (APV).

33. The antigen-presenting vesicle (APV) of any one of claims 1-27, wherein the first or second fusion protein is expressed at a valency of about 4000 copies on a surface of the antigen- presenting vesicle (APV).

34. The antigen-presenting vesicle (APV) of any one of claims 1-27, wherein the first or second fusion protein is expressed at a valency of about 6000 copies on a surface of the antigen- presenting vesicle (APV).

35. The antigen-presenting vesicle (APV) of any one of claims 1-27, wherein the first or second fusion protein is expressed at a valency of about 8000 copies on a surface of the antigen- presenting vesicle (APV).

36. The antigen-presenting vesicle (APV) of any one of claims 1-35, wherein the first fusion protein comprises an amino acid sequence with at least 95% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 1-22, 102, and 110-148.

37. The antigen-presenting vesicle (APV) of any one of claims 1-36, wherein the second fusion protein comprises an amino acid sequence with at least 90% sequence identity to an amino acid sequence according to any one of SEQ ID NOs: 23-58, 103-109 and 545-555.

38. The antigen-presenting vesicle (APV) of any one of claims 1-37, wherein the antigen-presenting vesicle (APV) does not comprise viral genetic material.

39. The antigen-presenting vesicle (APV) of any one of claims 1-38, wherein the antigen-presenting vesicle (APV) is a viral-like a particle.

40. The antigen-presenting vesicle (APV) of any one of claims 1-39, wherein the antigen-presenting vesicle (APV) is an extracellular vesicle.

41. The antigen-presenting vesicle (APV) of any one of claims 1-40, wherein the antigen-presenting vesicle (APV) is an exosome.

42. The antigen-presenting vesicle (APV) of any one of claims 1-41, wherein the antigen-presenting vesicle (APV) is an ectosome.

43. The antigen-presenting vesicle (APV) of any one of claims 1-42, wherein the costimulatory ligand molecule comprises at least two, three, four, five, six, seven, eight, nine, ten, or more than ten selected from the group comprising GITRL (GITR ligand), CD80, CD86, OX40LWSGR Docket No. 48295-718.601(XO40 ligand), CD48, CD30L (CD30 ligand), ICOSL (ICOS ligand), CD70, CD40, CD40L (CD40 Ligand), 4-1BBL (4-1BB ligand), TLIA, TIM4, SLAM, CD48, CD58, CD155, CD112, LIGHT, and ICAM.

44. The antigen-presenting vesicle (APV) of any one of claims 1-42, wherein the costimulatory ligand molecule comprises GITRL (GITR ligand).

45. The antigen-presenting vesicle (APV) of any one of claims 1-42, wherein the costimulatory ligand molecule comprises GITRL (GITR ligand), CD80, CD86, OX40L, CD48, CD30L, and ICAM-1.

46. The antigen-presenting vesicle (APV) of any one of claims 1-42, wherein the costimulatory ligand molecule comprises GITRL (GITR ligand) and ICAM-1.

47. The antigen-presenting vesicle (APV) of any one of claims 1-42, wherein the costimulatory ligand molecule comprises GITRL (GITR ligand), OX40L (0X40 ligand), and ICAM-148. The antigen-presenting vesicle (APV) of any one of claims 1-42, wherein the costimulatory ligand molecule comprises GITRL (GITR ligand), CD80, and ICAM-1.

49. The antigen-presenting vesicle (APV) of any one of claims 1-42, wherein the costimulatory ligand molecule comprises GITRL (GITR ligand), CD86, and ICAM-1.

50. A composition comprising a first nucleic acid sequence encoding an APV according to any one of the preceding claims, and an excipient.

51. The composition of claim 50, further comprising a second nucleic acid sequence that encodes one or more viral proteins.

52. The composition of claim 51, wherein the one or more viral proteins is a lentiviral protein, a retroviral protein, an adenoviral protein, or combinations thereof.

53. The composition of claim 51, wherein the one or more viral proteins comprises gag, pol, pre, tat, rev, or combinations thereof.

54. A pharmaceutical composition comprising the APV of any one of claims 1-49 and a pharmaceutically acceptable excipient.

55. A method of preventing or treating cancer or viral infection in a subject in need thereof comprising administering the antigen-presenting vesicle (APV) of any one of claims 1-49 or the composition of any one of claims 50-54.

56. A first fusion protein comprising a SCT-pMHC-I molecule that binds specifically to a cognate antigen-specific T cell receptor, a transmembrane domain and an oligomerization domain, wherein the SCT-pMHC-I molecule comprises a fusion protein comprising an antigenic peptide, P2m , an MHC-I alpha chain, an oligomerization domain, and a transmembrane anchoringWSGR Docket No. 48295-718.601 domain, wherein the first fusion protein is expressed at a valency of at least about 100 copies on a surface of an antigen-presenting vesicle (APV).

57. A second fusion protein comprising a costimulatory ligand molecule that binds specifically to a T cell surface costimulatory receptor, an oligomerization domain, and a transmembrane domain, wherein the second fusion protein is expressed at a valency of at least about 100 copies on a surface of an APV.

58. The first fusion protein of claim 56, wherein the antigenic peptide binds to the pMHC-I alpha chain and is fused to P2m.

59. The second fusion protein of claim 57, wherein the second fusion protein comprises at least two separate fusion proteins, wherein the at least two separate fusion proteins each comprises a different costimulatory ligand molecule.

60. The second fusion protein of any one of claims 57-59, wherein the costimulatory ligand molecule comprises GITRL (GITR ligand), CD80, CD86, OX40L (0X40 ligand), CD48, CD30L (CD30 ligand), ICOSL (ICOS ligand), CD70, CD40, CD40L (CD40 ligand), 4-1BBL (4- 1BB ligand), TLIA, TIM4, SLAM, CD48, CD58, CD155, CD112, LIGHT, or ICAM1.

61. An antigen-presenting vesicle comprising the first fusion protein of claim 56 and the second fusion protein of claim 57, wherein the first or second fusion protein is expressed at least about 100 copies on a surface of the antigen-presenting vesicle.

62. A first nucleic acid sequence encoding the first fusion protein of claim 56.

63. A second nucleic acid sequence encoding the second fusion protein of claim 57.

64. A method of inducing protective immunity against cancer or viral infection in a subject in need thereof comprising administering the antigen-presenting vesicle (APV) of any one of claims 1-49 or the composition of any one of claims 50-54.

65. A method of preventing or treating cancer or viral infection in a subject in need thereof comprising:(a) obtaining T cells from the subject;(b) contacting the T cells ex vivo with the antigen-presenting vesicle (APV) of any one of claims 1-48 or the composition of any one of claims 50-54 to activate the T cells; and(c) administering the activated and expanded T cells from step (b) to the subject.

66. The method of any one of claims 55, 64, and 65, further comprising administering the subject a second therapy.

67. The method of claim 66, wherein the second therapy comprises a Toll-like receptor agonist for TLR3 or TLR4 or TLR7 or TLR8, or TLR9 as an adjuvant, and optionally CpG, PolylC, LPS, and STING agonists.WSGR Docket No. 48295-718.60168. The method of claim 66, wherein the second therapy comprises an immune checkpoint blocked therapy, and optionally an anti-PDl or anti-PDLl or anti-CTLA-4 antibody.

69. The method of claim 66, wherein the second therapy comprises a targeted therapy, and optionally a tyrosine kinase inhibitor or PARP inhibitor or RAS inhibitor.

70. The method of claim 66, wherein the second therapy comprises T cell therapy, and optionally a chimeric antigen-receptor T cell therapy or tumor-infiltrated lymphocyte therapy.

71. The method of claim 66, wherein the second therapy comprises standard care therapy, and optionally surgery or chemotherapy or radiation therapy.

72. The composition of any one of the preceding claims, further comprising a third nucleic acid sequence that encodes a replication incompetent viral genome, a reporter, a therapeutic molecule, or combinations thereof.

73. The composition of any one of the preceding claims, wherein the first nucleic acid sequence, the second nucleic acid sequence, and the third nucleic acid sequence are within a same vector.

74. The composition of any one of the preceding claims, wherein the first nucleic acid sequence, the second nucleic acid sequence, and the third nucleic acid sequence are within different vectors.

75. The composition of any one of the preceding claims, wherein the nucleic acid is DNA or mRNA.

76. The composition of any one of the preceding claims, wherein the vector is a lentivirus vector, an adenovirus vector, or an adeno-associated virus vector.

77. A pharmaceutical composition comprising the antigen-presenting vesicles of any one of claims 1-49 and a pharmaceutically acceptable excipient.

78. A method of treating a viral infection, comprising administering the antigen- presenting vesicles of any one of claims 1-49 or the composition of any one of claims 50-78.

79. The method of claim 78, wherein the antigen-presenting vesicle is administered intravenously.

80. The method of claim 78, wherein the antigen-presenting vesicle is administered through inhalation.

81. The method of claim 78, wherein the antigen-presenting vesicle is administered by an intraperitoneal injection.

82. The method of claim 78, wherein the antigen-presenting vesicle is administered by a subcutaneous injection.WSGR Docket No. 48295-718.60183. An antigen-presenting cell (APC) comprising the first fusion protein of claim 56 and the second fusion protein of claim 57, wherein the first or second fusion protein is expressed at least about 100 copies on a surface of the APC, and wherein the APC is derived from an off-the- shelf cell line.

84. The APC of claim 83, wherein the antigenic peptide of the SCT-pMHC-I molecule binds to the pMHC-I alpha chain and is fused to P2m.

85. The APC of claim 83, wherein the second fusion protein comprises at least two separate fusion proteins, wherein the at least two separate fusion proteins each comprises a different costimulatory ligand molecule.

86. The APC of claim 85, wherein the costimulatory ligand molecule comprises GITRL (GITR ligand), CD80, CD86, OX40L (0X40 ligand), CD48, CD30L (CD30 ligand), ICOSL (ICOS ligand), CD70, CD40, CD40L (CD40 ligand), 4-1BBL (4-1BB ligand), TLIA, TIM4, SLAM, CD48, CD58, CD155, CD112, LIGHT, or ICAM1.

Images