Class ii HLA with inter-chain disulfide bond

Abstract

The present disclosure provides modified HLA molecules comprising an α chain and β chain, wherein both chains comprise one or more substituted cysteines which are paired to form a disulfide bond, and wherein the modified HLA molecules are optionally further conjugated at the 3' end to an Fc domain, and methods of using the modified HLA molecules to induce or enhance cytotoxicity.

Description

CLASS II HLA WITH INTER-CHAIN DISULFIDE BONDCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to U.S. Application Ser. No. 63 / 613,880, filed on December 22, 2023. The entire disclosure of the application identified in this paragraph is incorporated herein by reference.SEQUENCE LISTING

[0002] This application contains references to amino acid sequences and / or nucleic acid sequences which have been submitted concurrently herewith as the sequence listing XML file entitled “000002wopoa_SequenceListing.XML”, file size 38,105 bytes, created on 20 December 2024. The aforementioned sequence listing is hereby incorporated by reference in its entirety pursuant to 37 C.F.R. §1 .52(e)(5).FIELD

[0003] The present disclosure generally relates to modified HLA molecules wherein the a and p chains are engineered to comprise substituted cysteine pairs, compositions comprising the modified HLA molecules, compositions of the modified HLA molecules for treating cancer, pathogen-driven diseases, and alloimmune responses, methods for enhancing cytotoxicity, and methods of producing the modifiedHLA molecules.BACKGROUND

[0004] The major histocompatibility complex (MHC) encodes human leukocyte antigen (HLA) genes. The HLA is a critical component of the immune system and HLA is responsible for the regulation of the immune response through the presentation of antigens to T cells. HLA-DQ molecules, which is a subset of class II HLA molecules, are known to drive unwanted alloimmune responses after solid-organ transplants and several autoimmune diseases, including celiac disease. Biologies with HLA molecules as part of the design are emerging therapeutic options for these allo- and autoimmune conditions. However, the a and [3 chains of soluble class II HLA molecules do not dimerize efficiently without their transmembrane domains, which hinders their production.

[0005] Serra et al. describes fusion of class II MHC a and p chains to mutated lgG1 -Fc to forms knob-into-hole structures that increases the stability and production yields of MHC class II molecules. Serra et al. “Increased yields and biological potency of knob-into-hole-based soluble MHC class II molecules.” Nature communications vol. 10,1 4917. 29 Oct. 2019.

[0006] WO 2011 / 101681 A2 describes that a and p chains of a recombinant MHC class II molecules may be stabilized by disulfide bonds located between cysteine residues positioned at Pro 96a2- Ser 119p2(rank 1 ), Ser 95a2- Ser 121p2(rank 2), Arg 94a2- Asn 151 P2(rank 3), Phe 148a2- Gly 152P2(rank 4), Pro 96a2- Thr 101 P2(rank 5), Pro 96a2- Ser 121 P2(rank 6), lie 106a2- Asn 151 P2(rank 7) and Ser 95a2- Asp 122P2(rank 8).

[0007] WO 2024 / 006576 A1 describes that MHC class II constructs may be stabilized by one or more disulfide bonds and one or more amino acid substitutions.SUMMARY

[0008] This section provides a general summary of disclosure and is not a comprehensive disclosure of its full scope or all of its features.

[0009] The present disclosure provides modified HLA molecules comprising a first peptide comprising a wildtype a chain and a signal peptide, wherein at least one amino acids of the wildtype a chain are substituted with cysteine; and a second peptide comprising a wildtype p chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein at least one amino acids of the wildtype p chain are substituted with cysteine, wherein each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the p chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the p chain to form one or more interchain disulfide bonds. The substituted cysteines are located such that, when the modified HLA molecule assumes its quaternary structure, a substituted cysteine on the a chain is in proximity to a substituted cysteine on the p chain and the pair of substituted cysteines form a disulfide bond. In an embodiment, the modified HLA molecule is an HLA-DQ2.5, HLA-DQ7, an HLA-DQ8, or an HLA-DR molecule.

[0010] Also disclosed are the first peptide has an amino acid sequence of SEQ ID No: 1 and the second peptide has an amino acid sequence of SEQ ID No: 2.

[0011] Also disclosed are nucleic acid sequences encoding the modified HLA-DQ molecule.

[0012] Also disclosed are the human lgG1 Fc fragment sequence linked to the 3’ end of the [3 chain has at least one gain-of-function mutation.

[0013] Also disclosed are cells expressing either the modified HLA molecule or comprising a nucleic acid sequence encoding the modified HLA molecule.

[0014] Also disclosed are a composition comprising a modified HLA molecule presented on the surface of a cell or an acellular membrane.

[0015] Also disclosed are the wherein the composition comprising a modified HLA molecule for treatment of cancer, pathogen-driven diseases, and immune responses.

[0016] Also disclosed are methods of enhancing cytotoxicity comprising targeting the disclosed modified HLA molecule, a cell expressing the modified HLA molecule, or a cell comprising a nucleic acid sequence encoding the modified HLA molecule to a target cell.

[0017] Also disclosed are methods of producing a modified HLA molecule, comprising substituting one or more amino acids of a wildtype a chain with cysteine, substituting one or more amino acids of a wildtype [3 chain with cysteine in a manner that the a and p chains comprise the same number of substituted cysteines, and forming a disulfide bond between each substituted cysteine on the a chain and each substituted cysteine on the chain to form one or more interchain disulfide bonds.BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are for illustrative purposes only of selected embodiments and not all possible implementations, and are not intended to limit the scope of the present disclosure.

[0019] FIG. 1A depicts paired amino acid residues in HLA-DQ2 which are juxtaposed to each other and potentially available for disulfide bonds if mutated to cysteine. FIG. 1 B grades the candidate pairs for the potential for disulfide bonds. FIG 1 C measures the expression level of modified HLA molecules via ELISA. FIG. 1 D shows the expression of wildtype and variant pDQ7-Fc proteins by SDS-PAGE.

[0020] FIG. 2A depicts the a and p chains of HLA-DQ with potential amino acid pairs suitable for disulfide bonds.

[0021] FIGs. 2B, 2C, and 2D depict three-dimensional structures of YCDC, ACEC, and ACNC and the pair of mutated amino acid residues in each molecule.

[0022] FIG. 3A depicts the analysis of modified HLA-DQ with different peptides under reducing and non-reducing conditions by SDS-PAGE.

[0023] FIG. 3B depicts the quantification of expression level of the modified HLA molecules with different peptides.

[0024] FIG. 4A depicts the binding profiles of the modified HLA molecule to anti- DQ7 as measured by ELISA.

[0025] FIG. 4B depicts the binding profile of empty YCDC, YCDC carrying CLIP, and YCDC carrying Mem-36 as measured by ELISA.

[0026] FIG. 4C shows a Western blot of the modified HLA molecules.

[0027] FIG. 5A provides a schematic of how a pDQ7-Fc protein can guide the killing of antibody-producing cells with specific B cell receptors.

[0028] FIG. 5B shows binding of pDQ7-Fc(YCDC) to HB144 B cell hybridomas as measured by flow cytometry.

[0029] FIG. 5C shows the percent survival of the hybridoma under increasing concentrations of peptide.

[0030] FIG. 5D shows the normalized percent survival of the hybridoma under increasing concentrations of peptide.

[0031] FIG. 6 depicts the [3 chain and the a chain.

[0032] FIG. 7 depicts SDS-PAGE analysis of modified HLA molecules under reducing conditions.

[0033] FIG. 8 depicts ELISA of pDQ2 expression in CHO cells.

[0034] FIG. 9 depicts expression of variant pDQ8-Fc (YCDC) under a nonreducing condition by SDS-PAGE.

[0035] FIG. 10 depicts grades of eight pairs of candidate positions for disulfide engineering were predicted by the MODIP procedure based on HLA-DR4 models, and expression of variant pDR4-Fc (PCPC) under a non-reducing condition by SDS-PAGE.DETAILED DESCRIPTION

[0036] The following description is merely exemplary in nature and is not intended to limit the present disclosure, application, or uses.

[0037] As used here, an “HLA molecule” is a protein dimer comprising an a chain and a [3 chain, wherein the a and [3 are separate peptides that, when assuming theirrespective three-dimensional shapes, interact to form a protein with a quaternary structure. Human leukocyte antigen (HLA) is the human form of the major histocompatibility complex (MHC) found in all mammals. In various embodiments, HLA molecule is selected from the group consisting of HLA-A, HLA-B, HLA-C, HLA-DM, HLA-DO, DP, HLA-DQ, and HLA-DR. In an embodiment, HLA molecule is HLA-DR. HLA-DQ molecule is a subset of HLA molecules. HLA-DQ molecule includes HLA-DQ2, -DQ3, -DQ4, -DQ5, -DQ6, -DQ7, -DQ8, and -DQ9. In an embodiment, HLA-DQ is HLA- DQ2.5. In an embodiment, HLA-DQ is HLA-DQ7.ln an embodiment, HLA-DQ is HLA- DQ8.

[0038] As used herein, a “modified HLA-DQ molecule” is an HLA-DQ molecule, wherein the amino acid sequences of the wildtype a chain and wildtype p chain each comprise at least one substitution with cysteine. Each substituted cysteine on the a chain is proximal to a corresponding substituted p chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the p chain to form one or more interchain disulfide bonds.

[0039] As used herein, a “modified HLA-DR molecule” is an HLA-DR molecule, wherein the amino acid sequences of the wildtype a chain and wildtype p chain each comprise at least one substitution with cysteine. Each substituted cysteine on the a chain is proximal to a corresponding substituted p chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the p chain to form one or more interchain disulfide bonds.

[0040] As used herein, “wildtype” refers to amino acid or nucleic acid sequences that are generally found in a naturally occurring population. For example, wildtype achain refers to putative HLA a chain sequences are generally present in the human population. Similarly, wildtype [3 chain refers to putative HLA p chain sequences that are generally present in the human population. The wildtype sequence serves as the basis for cysteine substitutions in the amino acid sequence encoded by nucleic acid sequences. For example, a substitution at position 16 of an amino acid sequence refers to changing the amino acid at position 16 that occurs naturally in the general human population and replacing it with another amino acid that is not commonly present at position 16.

[0041] As used herein, “pharmaceutically acceptable” means suitable for use in pharmaceutical preparations, generally considered as safe for such use, officially approved by a regulatory agency of a national or state government for such use, or being listed in the U.S. Pharmacopoeia or other generally recognized pharmacopoeia for use in animals, and more particularly in humans.

[0042] As used herein, “pharmaceutically acceptable carrier” refers to a diluent, adjuvant, excipient, or carrier, other ingredient, or combination of ingredients that alone or together provide a carrier or vehicle with which a compound or compounds of the invention is formulated and / or administered, and in which every ingredient or the carrier as a whole is pharmaceutically acceptable.

[0043] As used herein, “pharmaceutically acceptable excipient” refers to any substance, other than the active pharmaceutical ingredient (API), that is included in a pharmaceutical formulation to aid in the manufacturing process, enhance stability, improve bioavailability, or facilitate the administration of the drug. These excipients are pharmacologically inactive and are selected based on their compatibility with the APIand their ability to perform specific functions such as binding, filling, disintegration, or preservation.

[0044] As used herein, “cancer” refers to benign or malignant cancer including, but not limited to fibromas, lipomas, adenomas, hemangiomas, leiomyomas, myxomas, chondromas, osteomas, fibrous histiocytomas, hemangiopericytomas, lymphangiomas, plasmacytosis, rhabdomyomas, papillomas, hepatic adenomas, renal tubular adenomas, bile duct adenomas, transitional cell papillomas, hydatidiform moles, meningiomas, schwannomas, neurilemmomas, neurofibromas, carcinomas, sarcomas, leukemias, lymphomas, myelomas, gliomas, glioblastomas, neuroblastomas, medulloblastomas, malignant meningiomas, malignant schwannomas, neurofibrosarcomas, fibrosarcomas, myxosarcomas, liposarcomas, chondrosarcomas, osteosarcomas, chordomas, malignant fibrous histiocytomas, hemangiosarcomas, angiosarcomas, lymphangiosarcomas, mesotheliomas, plasmacytomas, multiple myelomas, Hodgkin lymphomas, non-Hodgkin lymphomas, leiomyosarcomas, rhabdomyosarcomas, squamous cell carcinomas, adenocarcinomas, hepatocellular carcinomas, renal cell carcinomas, cholangiocarcinomas, seminomas, embryonal cell carcinomas, and various types of brain and central nervous system cancers.

[0045] As used herein, “cancer cell” refers to cells of cancers.

[0046] As used herein, “pathogens” refers to pathogens including but not limited to Escherichia coll, Staphylococcus aureus, Mycobacterium tuberculosis, Salmonella enterica, Clostridium botulinum, Bacillus anthracis, Vibrio cholerae, Helicobacter pylori, Influenza virus, HIV, SARS-CoV-2, Hepatitis B virus, Ebola virus, Zika virus, Rabies virus, Herpes simplex virus, Candida albicans, Aspergillus fumigatus, Cryptococcusneoformans, Histoplasma capsulatum, Blastomyces dermatitidis, Coccidioides immitis, Plasmodium falciparum, Giardia lamblia, Toxoplasma gondii, Trypanosoma brucei, Leishmania donovani, Schistosoma mansoni, Trichinella spiralis, Onchocerca volvulus, prions causing Creutzfeldt-Jakob disease, prions causing Kuru, prions causing Fatal Familial Insomnia, Acinetobacter baumannii, Actinomyces israelii, Propionibacterium propionicus, Adenoviridae, Trypanosoma brucei, Entamoeba histolytica, Anaplasma species, Angiostrongylus, Anisakis, Arcanobacterium haemolyticum, Junin virus, Ascaris lumbricoides, Astroviridae species, Babesia species, Bacillus cereus, Borrelia burgdorferi, Brucella species, Campylobacter jejuni, Chlamydia trachomatis, Corynebacterium diphtheriae, Coxiella burnetii, Cryptosporidium parvum, Cyclospora cayetanensis, Dengue virus, Echinococcus granulosus, Enterococcus faecalis, Francisella tularensis, Haemophilus influenzae, Hantavirus, Hendra virus, Human papillomavirus, Klebsiella pneumoniae, Lassa virus, Legionella pneumophila, Leptospira interrogans, Listeria monocytogenes, Marburg virus, Measles virus, Mumps virus, Neisseria gonorrhoeae, Neisseria meningitidis, Norovirus, Parvovirus B19, Pneumocystis jirovecii, Pseudomonas aeruginosa, Respiratory syncytial virus, Rickettsia rickettsii, Rift Valley fever virus, Rotavirus, Rubella virus, Shigella dysenteriae, Streptococcus pneumoniae, Streptococcus pyogenes, Taenia solium, Treponema pallidum, Trichomonas vaginalis, Varicella-zoster virus, Vibrio vulnificus, West Nile virus, Yersinia pestis, and Zygomycetes.

[0047] As used herein, “pathogen-driven disease” refers to a disease caused by a pathogen.

[0048] As used herein, “immune response” refers to autoimmune and alloimmune responses and diseases including but not limited to Rheumatoid Arthritis, Systemic Lupus Erythematosus (Lupus), Myositis, Polymyalgia Rheumatica, Psoriasis, Psoriatic Arthritis, Dermatomyositis, Scleroderma, Vitiligo, Bullous Pemphigoid, Lichen Planus, and Alopecia Areata. Additionally, autoimmune diseases affecting the digestive system include Crohn’s Disease, Celiac Disease, Ulcerative Colitis, and Autoimmune Hepatitis. Endocrine system disorders such as Type 1 Diabetes, Addison’s Disease, Hashimoto’s Thyroiditis, and Graves’ Disease are also included. The nervous system can be impacted by Multiple Sclerosis (MS), Myasthenia Gravis (MG), Guillain-Barre Syndrome, and Chronic Inflammatory Demyelinating Polyneuropathy (CIDP). Blood and blood vessel disorders include Autoimmune Hemolytic Anemia, Vasculitis, and Antiphospholipid Syndrome. Other autoimmune diseases include Sjogren’s Syndrome, Autoimmune Uveitis, Goodpasture’s Syndrome, Wegener’s Granulomatosis, Autoimmune Gastritis, Autoimmune Pancreatitis, Autoimmune Encephalitis, Autoimmune Inner Ear Disease, Autoimmune Lymphoproliferative Syndrome (ALPS), Autoimmune Neutropenia, Autoimmune Thrombocytopenia, Autoimmune Retinopathy, Autoimmune Autonomic Ganglionopathy, Autoimmune Enteropathy, Autoimmune Hepatitis, Autoimmune Myocarditis, Autoimmune Oophoritis, Autoimmune Orchitis, Autoimmune Polyendocrine Syndrome (APS), Autoimmune Progesterone Dermatitis, Autoimmune Pulmonary Alveolar Proteinosis, Autoimmune Thyroiditis, Autoimmune Vasculopathy, Autoimmune Vitiligo, and Autoimmune Vulvovaginitis.

[0049] As used herein, “unwanted cells” refers to cells including but not limited to cancer cells, infected cells, senescent cells, autoimmune cells, damaged cells, and excess cells.

[0050] Modified HLA molecules

[0051] The present disclosure provides a modified HLA molecule wherein the wildtype sequence of the a and p chains are modified such that they comprise a pair of substituted cysteines that form a disulfide bond between the a and p chains. In an embodiment, the a chain comprises an amino acid sequence wherein one or more amino acids of a wildtype a chain are substituted with cysteine and the p chain comprises an amino acid sequence wherein one or more amino acids of a wildtype p chain are substituted with cysteine. The a and p chains comprise the same number of substituted cysteines and the substituted cysteines on each chain are proximal to each other when the modified HLA assumes its quaternary structure and each substituted cysteine on the a chain forms a disulfide bond with a corresponding substituted cysteine on the p chain. The mutated residues in the a and p chains have limited solvent exposure. The modified HLA molecule may be further linked to an additional domain, such as an Fc domain, histidine tag, or streptavidin tag. The additional domain may be to assist in purification or to induce complement-dependent or cell-dependent cytotoxicity such as antibody-dependent cellular cytotoxicity. The Fc domain may be attached at the 3’ end of either the a chain or the p chain. In a particular embodiment, the Fc domain is linked to the 3’ end of the p chain. In a particular embodiment, the Fc domain is an Fc fragment. In a particular embodiment, the Fc domain is an IgG Fc fragment. In a particular embodiment, the Fc domain is a Human IgG 1 Fc fragment. TheFc domain may comprise gain-of-function mutations to enhance potency of complement-dependent or cell-dependent cytotoxicity.

[0052] The modified HLA molecule can be a class I HLA molecule (e.g., HLA-A, HLA-B, HLA-C, HLA-E, HLA-F, HLA-G), or a class II HLA molecule (e.g., HLA-DP, HLA- DQ, or HLA-DR). In various embodiments, the modified HLA molecule is HLA-DQ which is selected from the group consisting of HLA-DQ2, -DQ3, -DQ4, -DQ5, -DQ6, -DQ7, - DQ8, and -DQ9. In a particular embodiment, HLA-DQ is HLA-DQ2.5. In a particular embodiment, HLA-DQ is HLA-DQ7. In a particular embodiment, HLA-DQ is HLA-DQ8. The modified HLA molecule may be either a class I HLA molecule conjugated to Fc or a class II HLA molecule conjugated to Fc. In a specific embodiment, the modified HLA molecule is a class II HLA-Fc conjugate. When the modified HLA molecule is an HLA- DQ molecule, the substituted cysteines can be located on the a chain at position 19, 83, 84, or combinations thereof. When the modified HLA molecule is an HLA-DQ molecule, the substituted cysteines can be located on the [3 chain at position 5, 6, 33, or combinations thereof. When the modified HLA molecule is an HLA-DR molecule, the substituted cysteines can be located on the a chain at position 83 and 84, or combinations thereof. When the modified HLA molecule is an HLA-DR molecule, the substituted cysteines can be located on the [3 chain at position 5 and 33, or combinations thereof.

[0053] In an embodiment, the modified HLA molecule is an HLA-DQ7 comprising a first peptide comprising a wildtype a chain and a signal peptide, wherein at least one amino acids of the wildtype a chain are substituted with cysteine; and a second peptide comprising a wildtype [3 chain, a signal peptide, a placeholder peptide, a linker, and anFc fragment sequence, wherein at least one amino acids of the wildtype |3 chain are substituted with cysteine. Each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the [3 chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the [3 chain to form one or more interchain disulfide bonds. In an embodiment, the signal peptide of the first peptide is derived from azurocidin preprotein. In an embodiment, the placeholder peptide of the second peptide is a placeholder CLIP peptide. In an embodiment, the linker of the second peptide is a polyglycine linker. In an embodiment, the Fc fragment sequence is a Human lgG1 Fc fragment sequence. In an embodiment, the first peptide has the amino acid sequence of SEQ ID No: 1 and the second peptide has the amino acid sequence of SEQ ID No: 2. The substituted cysteine on the a chain can be located at various amino acid positions which include ones selected from the group consisting of 19, 83, 84, and a combination thereof. The substituted cysteine on the [3 chain can be located at various amino acid positions which include ones selected from the group consisting of 6, 5, 33, and a combination thereof. In an embodiment, the substituted cysteine on the a chain is at amino acid position 19 and the substituted cysteine on the |3 chain is at amino acid position 6, wherein the substituted cysteine on the a chain is at amino acid position 83 and the substituted cysteine on the [3 chain is at amino acid position 5, and wherein the substituted cysteine on the a chain is at amino acid position 84 and the substituted cysteine on the [3 chain is at amino acid position 33, or a combination thereof. In another embodiment, the substituted cysteine on the a chain is Y19C and the substituted cysteine on the [3 chain is D6C, wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the [3 chain is E5C,wherein the substituted cysteine on the a chain is A84C and the substituted cysteine on the p chain is N33C, or a combination thereof. In yet another embodiment, the substituted cysteine on the a chain is at amino acid position 19 and the substituted cysteine on the p chain is at amino acid position 6, and wherein the substituted cysteine on the a chain is at amino acid position 83 and the substituted cysteine on the p chain is at amino acid position 5. In yet another embodiment, the substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C, and wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the p chain is E5C.

[0054] In yet another embodiment, the modified HLA molecule is any one depicted in Figure 2. In yet another embodiment, the human lgG1 Fc fragment sequence linked to the 3’ end of the p chain has at least one gain-of-function mutation. In yet another embodiment, the gain-of-function mutation is E345R and / or E430G. In a particular embodiment, the modified HLA-DQ7 molecule comprising: a first peptide having an amino acid sequence of SEQ ID No: 1 comprising a wildtype a chain and a signal peptide, wherein a cysteine substitution on the wildtype a chain is located at an amino acid position of 19,83, and 84; and a second peptide having an amino acid sequence of SEQ ID No: 2 comprising a wildtype p chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein a cysteine substitution on the wildtype p chain is located at an amino acid position of 5, 6, and 33, wherein the substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C, wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the p chain is E5C, wherein the substituted cysteine on the achain is A84C and the substituted cysteine on the p chain is N33C, or a combination thereof, wherein each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the p chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the p chain to form one or more interchain disulfide bonds.

[0055] In a particular embodiment, the modified HLA-DQ7 molecule comprising a first peptide having an amino acid sequence having at least 85% sequence identity to SEQ ID No: 1 comprising a wildtype a chain and a signal peptide, wherein a cysteine substitution on the wildtype a chain is located at an amino acid position of 19, 83, and / or 84; and a second peptide having an amino acid sequence having at least 85% sequence identity to SEQ ID No: 2 comprising a wildtype p chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein a cysteine substitution on the wildtype p chain is located at an amino acid position of 5, 6, and / or 33. The substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C, the substituted cysteine on the a chain is A83C and the substituted cysteine on the p chain is E5C, the substituted cysteine on the a chain is A84C and the substituted cysteine on the p chain is N33C, or a combination thereof. Each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the p chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the p chain to form one or more interchain disulfide bonds.

[0056] In an embodiment, the modified HLA molecule comprising a first peptide having at least 80% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, atleast 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 1. and a second peptide having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 2. In an embodiment, the modified HLA-DQ molecule comprising a first peptide having at least 80% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 1. and a second peptide having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 2. In an embodiment, the modified HLA-DQ7 molecule comprising a first peptide having at least 80% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 1. and a second peptide having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 2.

[0057] In an alternate embodiment, the modified HLA molecule is an HLA-DR molecule. The modified HLA-DR molecule can have the same substitutions asmentioned for the modified HLA-DQ molecule. In a particular embodiment, the HLA-DR molecule comprising a first peptide comprising a wildtype a chain and a signal peptide, wherein at least one amino acids of the wildtype a chain are substituted with cysteine; and a second peptide comprising a wildtype [3 chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein at least one amino acids of the wildtype [3 chain are substituted with cysteine. Each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the [3 chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the [3 chain to form one or more interchain disulfide bonds. In an embodiment, the signal peptide of the first peptide is derived from azurocidin preprotein. In an embodiment, the placeholder peptide of the second peptide is a placeholder CLIP peptide. In an embodiment, the linker of the second peptide is a polyglycine linker. In an embodiment, the Fc fragment sequence is a Human lgG1 Fc fragment sequence. In an embodiment, the first peptide has the amino acid sequence of SEQ ID No: 9 and the second peptide has the amino acid sequence of SEQ ID No: 10. The substituted cysteine on the a chain can be located at various amino acid positions which include ones selected from the group consisting of 83 and 84, and a combination thereof. The substituted cysteine on the [3 chain can be located at various amino acid positions which include ones selected from the group consisting of 5 and 33, and a combination thereof. In an embodiment, the substituted cysteine on the a chain is at amino acid position 83 and the substituted cysteine on the [3 chain is at amino acid position 5, the substituted cysteine on the a chain is at amino acid position 84 and the substituted cysteine on the [3 chain is at amino acid position 33, or a combination thereof. In another embodiment, thesubstituted cysteine on the a chain is P83C and the substituted cysteine on the p chain is P5C, the substituted cysteine on the a chain is I84C and the substituted cysteine on the p chain is H33C, or a combination thereof. In yet another embodiment, the substituted cysteine on the a chain is at amino acid position 83 and the substituted cysteine on the p chain is at amino acid position 5. In yet another embodiment, the substituted cysteine on the a chain is at amino acid position 84 and the substituted cysteine on the p chain is at amino acid position 33. In yet another embodiment, the substituted cysteine on the a chain is P83C and the substituted cysteine on the p chain is P5C. In yet another embodiment, the substituted cysteine on the a chain is at amino acid position I84C and the substituted cysteine on the p chain is at amino acid position H33C.

[0058] In an alternate embodiment, the modified HLA molecule is an HLA-DQ2.5 molecule. The modified HLA-DQ2.5 molecule can have the same substitutions as mentioned for the modified HLA-DQ molecule. The modified HLA-DQ2.5 comprising a first peptide comprising a wildtype a chain and a signal peptide, wherein at least one amino acids of the wildtype a chain are substituted with cysteine; and a second peptide comprising a wildtype p chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein at least one amino acids of the wildtype p chain are substituted with cysteine. Each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the p chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the p chain to form one or more interchain disulfide bonds. In an embodiment, the signal peptide of the first peptide is derived from azurocidin preprotein. In an embodiment, the placeholderpeptide of the second peptide is a placeholder glia-a1 a peptide. In an embodiment, the linker of the second peptide is a polyglycine linker. In an embodiment, the Fc fragment sequence is a Human lgG1 Fc fragment sequence. In an embodiment, the first peptide has the amino acid sequence of SEQ ID No: 1 and the second peptide has the amino acid sequence of SEQ ID No: 6. The substituted cysteine on the a chain can be located at various amino acid positions which include ones selected from the group consisting of 19, 83, 84, and a combination thereof. The substituted cysteine on the p chain can be located at various amino acid positions which include ones selected from the group consisting of 6, 5, 33, and a combination thereof. In an embodiment, the substituted cysteine on the a chain is at amino acid position 19 and the substituted cysteine on the P chain is at amino acid position 6, wherein the substituted cysteine on the a chain is at amino acid position 83 and the substituted cysteine on the p chain is at amino acid position 5, and wherein the substituted cysteine on the a chain is at amino acid position 84 and the substituted cysteine on the p chain is at amino acid position 33, or a combination thereof. In another embodiment, the substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C, wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the p chain is E5C, wherein the substituted cysteine on the a chain is A84C and the substituted cysteine on the p chain is N33C, or a combination thereof. In yet another embodiment, the substituted cysteine on the a chain is at amino acid position 19 and the substituted cysteine on the p chain is at amino acid position 6, and wherein the substituted cysteine on the a chain is at amino acid position 83 and the substituted cysteine on the p chain is at amino acid position 5. In yet another embodiment, the substituted cysteine on the achain is Y19C and the substituted cysteine on the p chain is D6C, and wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the [3 chain is E5C.

[0059] In an alternate embodiment, the modified HLA molecule is an HLA-DQ8 molecule. The modified HLA-DQ8 molecule can have the same substitutions as mentioned for the modified HLA-DQ molecule. The modified HLA-DQ8 comprising a first peptide comprising a wildtype a chain and a signal peptide, wherein at least one amino acids of the wildtype a chain are substituted with cysteine; and a second peptide comprising a wildtype p chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein at least one amino acids of the wildtype p chain are substituted with cysteine. Each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the p chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the p chain to form one or more interchain disulfide bonds. In an embodiment, the signal peptide of the first peptide is derived from azurocidin preprotein. In an embodiment, the placeholder peptide of the second peptide is a placeholder CLIP peptide. In an embodiment, the linker of the second peptide is a polyglycine linker. In an embodiment, the Fc fragment sequence is a Human lgG1 Fc fragment sequence. In an embodiment, the first peptide has the amino acid sequence of SEQ ID No: 8 and the second peptide has the amino acid sequence of SEQ ID No: 7. The substituted cysteine on the a chain can be located at various amino acid positions which include ones selected from the group consisting of 19, 83, 84, and a combination thereof. The substituted cysteine on the p chain can be located at various amino acid positions which include ones selected from the groupconsisting of 6, 5, 33, and a combination thereof. In an embodiment, the substituted cysteine on the a chain is at amino acid position 19 and the substituted cysteine on the [3 chain is at amino acid position 6, wherein the substituted cysteine on the a chain is at amino acid position 83 and the substituted cysteine on the p chain is at amino acid position 5, and wherein the substituted cysteine on the a chain is at amino acid position 84 and the substituted cysteine on the p chain is at amino acid position 33, or a combination thereof. In another embodiment, the substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C, wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the p chain is E5C, wherein the substituted cysteine on the a chain is A84C and the substituted cysteine on the p chain is N33C, or a combination thereof. In yet another embodiment, the substituted cysteine on the a chain is at amino acid position 19 and the substituted cysteine on the p chain is at amino acid position 6, and wherein the substituted cysteine on the a chain is at amino acid position 83 and the substituted cysteine on the p chain is at amino acid position 5. In yet another embodiment, the substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C, and wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the p chain is E5C.

[0060] In an embodiment, the first peptide of the modified HLA-DQ7 molecule has a nucleotide sequence having at least 80% sequence identity, having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ IDNO: 11. The second peptide of the modified HLA-DQ7 molecule has a nucleotide sequence having at least 80% sequence identity, having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 12.

[0061] In an embodiment, the second peptide of the modified HLA-DQ2.5 molecule has a nucleotide sequence having at least 80% sequence identity, having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 16.

[0062] In an embodiment, the placeholder sequence is Mem_36 and the HLA molecule has a nucleotide sequence having at least 80% sequence identity, having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 13.

[0063] In an embodiment, the gain-of-function mutation is E345R and the modified HLA molecule has a nucleotide sequence having at least 80% sequence identity, having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 14.

[0064] In an embodiment, the gain-of-function mutation is E430G and the modified HLA molecule has a nucleotide sequence having at least 80% sequence identity, having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 15.

[0065] In an embodiment, the first peptide of the modified HLA-DQ8 molecule has a nucleotide sequence having at least 80% sequence identity, having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 17. The second peptide of the modified HLA-DQ8 molecule has a nucleotide sequence having at least 80% sequence identity, having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 18.

[0066] In an embodiment, the first peptide of the modified HLA-DR molecule has a nucleotide sequence having at least 80% sequence identity, having at least 85% sequence identity, at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 19. The second peptide of the modified HLA-DR molecule has a nucleotide sequence having at least 80% sequence identity, having at least 85% sequence identity,at least 85% sequence identity, at least 90% sequence identity, at least 95% sequence identity, at least 96% sequence identity, at least 97% sequence identity, at least 98% sequence identity, or at least 99% sequence identity to SEQ ID NO: 20.

[0067] In specific embodiments, combinations of the locations of the substituted cysteines on the a and [3 chains are shown in Table 1 .

[0068] Table 1

[0069] Preparation of modified HLA molecules and nucleotide sequences

[0070] The modified HLA molecules described herein can be prepared by conventional means known in the art. The amino acid sequence of the modified HLA molecule can be encoded by a nucleotide sequence (e.g., a I sequence) and provided to a cell (e.g., a hybridoma, bacteria, yeast, mammalian cell, etc.) which translates the nucleotide sequence to the modified HLA molecule. The nucleotide sequences described herein may be incorporated into a vector (e.g., a viral vector, a plasmid, etc.) for insertion into a cell (e.g., transformation, transfection, etc.) for subsequent production of the protein.

[0071] In any embodiment, a nucleotide sequence comprising a I sequence encoding an a chain and / or a [3 chain is provided herein. For example, a nucleotide sequence may encode an a chain comprising substituted cysteines at positions 19, 83,and / or 84, including the specific substitutions of Y19C, A83C, and / or A84C. The nucleotide sequence may further encode a p chain comprising substituted cysteines at positions 5, 6, and / or 33, including the specific substitutions of E5C, D6C, and / or N33C.

[0072] The nucleotide sequence may further encode an Fc domain that is linked to the 3’ end of either the a chain or the p chain. In a particular embodiment, the nucleotide sequence encodes an Fc domain linked to the 3’ end of the p chain. The Fc domain encoded by the nucleotide sequence may include gain-of-function mutations which may enhance the potency of antigen-specific killing.

[0073] Cells, nucleic acids, and compositions

[0074] The present disclosure provides a cell expressing the modified HLA-DQ molecule described herein and nucleic acids encoding the modified HLA-DQ molecule.

[0075] The present disclosure also provides a composition comprising: the modified HLA-DQ molecule presented on the surface of a cell or an acellular membrane and a pharmaceutically acceptable carrier. The composition may further include one or more pharmaceutically acceptable excipient. In an embodiment, the cell is an engineered cell. In a particular embodiment, the engineered cell is selected from the group consisting of a monocyte, leukocyte, macrophage, and antigen presenting cell. In an embodiment, the acellular membrane is an exosome. In various embodiments, the composition is used to treat various diseases and symptoms including, for example, cancer, pathogen-driven diseases, and immune diseases.

[0076] Method of enhancing cytotoxicity

[0077] In various embodiments, the modified HLA molecule is used to enhance cytotoxicity of a target cell. In one embodiment, the modified HLA molecule may be targeted to a target cell, such as a cancer cell (benign or malignant), pathogen, or other unwanted cell. The modified HLA molecule may be administered as part of a cell (e.g., an engineered monocyte, leukocyte, macrophage, antigen presenting cell), or within an acellular membrane (e.g., exosome).

[0078] The engineered cells or acellular membranes expressing the modified HLA molecule are administered to a patient in need of such treatment. The administration can be performed via various routes, including intravenous injection, direct tissue injection, and intraperitoneal injection. The dosage and frequency of administration are determined based on the patient’s condition, the severity of the disease, and the desired therapeutic outcome.

[0079] The modified HLA molecules in engineered cells or acellular membranes can be used in various therapeutic applications, including but not limited to autoimmune diseases, transplantation, cancer immunotherapy, and infectious diseases. In autoimmune diseases, the modified HLA molecule can help modulate the immune response. In transplantation, the engineered cells or acellular membranes can be used to improve graft acceptance and reduce the risk of rejection in organ and tissue transplantation. In cancer immunotherapy, the modified HLA molecule can enhance the presentation of tumor antigens, thereby boosting the immune system’s ability to recognize and destroy cancer cells. In infectious diseases, the engineered cells or acellular membranes can be employed to enhance the immune response against infections.EXAMPLES

[0080] Example 1 : Disulfide engineering of soluble peptide-HLA-DQ7-Fc (pDQ7- Fc) fusion protein.

[0081] To identify paired amino acid residues in HLA-DQ7 to introduce an interchain disulfide bond by mutagenesis, existing structural models of HLA-DQ2 and - DQ8 molecules (PDB ID: 1 S9V and 1 JK8, respective) were analyzed by visual inspection and using the modeling of disulfide bonds in proteins (MODIP) procedure. Many pairs of residues from the a1 / p1 domains of the heterodimer were juxtaposed at the bottom or side of the peptide binding groove and additional locations were also identified between the a2 / [31 and a2 / |32 domains (FIG. 1A). The MODIP procedure graded these candidate sites based on whether the atomic distance would accommodate a disulfide bond and whether any significant steric strain is present. Grade A indicated whether the modeled disulfide bond distance and dihedral angles are within accepted ranges; grade B indicated whether the bond is geometrically suitable but with some steric distortion; grade C indicated sites that were spatially too close for disulfide bond; grade D indicated sites where sulfurs cannot be fixed geometrically, and thus unsuitable for disulfide binding. The grades were unfavorable for most candidate pairs and frequently discrepant between the structural models for HLA-DQ2 and -DQ8 (FIG. 1 B).

[0082] DNA encoding paired a and p chains for the wildtype and variant HLA-DQ7 were transfected into Chinese hamster ovary (CHO) cell lines and the expression of these recombinant proteins in the supernatant was measured. In addition to thepaired mutations introduced for disulfide bond formation, sequences for the signal peptide, a placeholder CLIP peptide (PVSKMRMATPLLMQA), and a flexible polyglycine linker (GGGSG)2 were fused to the 5’ end of the p chain sequence. The 3’ end of the p chain sequence was fused to a human lgG1 Fc fragment sequence (FIG. 6). This design was expected to generate a peptide-HLA-DQ7 (pDQ7) complex that would further dimerize via the Fc fragment to form a bivalent pDQ7-Fc fusion protein within an immunoglobulin-like framework. Minimal expression of the wild-type pDQ7-Fc was observed in the supernatant as measured by ELISA, while the expression levels were significantly increased for YCDC, ACEC, and ACNC mutants, with up to 165-fold increase over the wild type (FIG. 1 C). The rest of the mutants did not exhibit significantly increased expression levels compared with the wild type.

[0083] Wildtype and variant pDQ7-Fc proteins were purified and evaluated by SDS-PAGE analysis (FIG. 1 D). Under the non-reducing condition, variants YCDC, ACEC, and ACNC showed a distinct band of about 150 kDa, consistent with the expected molecular weight of the pDQ7-Fc molecule. While this band was absent for the wildtype and other variants, a smaller band of about 100 kDa was observed, with the same size as the p chain-Fc dimer in the absence of a paired a chain. Under a reducing condition that disrupted interchain and intrachain disulfide bonds, a p chain of about 50 kDa across all variants was observed; the a chain was also observed for some variants at about 25 kDa with variable intensities (FIG. 7). These data together suggested that stable pDQ7-Fc variants can be generated by site-directed disulfide engineering with increased expression levels and production yields.

[0084] Example 2: Characteristics of amino acid residues mutated for successful disulfide engineering

[0085] HLA molecules are extremely polymorphic in the human population, allowing the presentation of diverse foreign peptides or neoantigens to launch adaptive immunity. Among the 337 HLA- DQA1 and 1516 HLA-DQB1 proteins (IPD-IMGT / HLA database version 3.54 released in October 2023) (Barker et al, Nucleic Acids Res, 2023), the three pairs of amino acid residues successfully mutated for disulfide engineering were all conserved. Moreover, cysteines have not been found at these positions in humans (FIG. 2A).

[0086] The solvent accessibility of the mutated residues was visualized. Solvent accessibility may indicate the possibility of new epitopes created after mutations (FIGs. 2B-D). In an HLA-DQ7 model created with the pHLA3D program, all three pairs of mutated residues in YCDC, ACEC, and ACNC were visible at the surface of the model. At side views, the 83A / 5E pair appeared to be more exposed than the Y19 / D6 pair and A84 / N33 pair. All three pairs, especially the latter two pairs, were largely shielded by the a helix of the a chain when viewed from the top of the molecule and thus unlikely to be recognized by T cell receptors (TOR). For possible recognition by B cell receptors (BCR), the three pairs of mutated residues were conserved in humans and not known to be immunogenic by themselves. All 27 antibody-verified HLA-DQ eplets cataloged in the HLA epitope registry were reviewed. None of these eplets, either on the a or |3 chain, were in contact with the three pairs of residues mutated for disulfide engineering. These data indicate possible low immunogenicity of these mutations, which warrants further investigation in an alloimmunization model.

[0087] Example 3: Effect of the peptide on the pDQ7-Fc(YCDC) fusion protein

[0088] Whether the stability of pDQ7-Fc (YCDC) was peptide-dependent was examined. Without CLIP, the DQ7-Fc (YCDC) construct failed to express any detectable protein with the expected size (FIG. 3A), suggesting that endogenous peptides in the cultured cells were insufficient to stabilize the soluble protein. A search of the Immune Epitope Database (IEDB) database (Vita, et al. Nucleic Acids Res., 2019) identified three SARS-CoV-2- derived peptides that bound to DQ7 with a high affinity, as defined as a half maximal inhibitory concentration (ICso) of <1000 nM in a radioligand competition binding assay (Heide, et al., PloS Pathol, 2021 ). Variants were generated by replacing CLIP in the pDQ7-Fc (YCDC) protein with the identified SARS-CoV-2 peptides, Mem_36 (SEQ ID NO:3) (ICso: 48 nM), Mem_34 (ICso: 649 nM), and Ncl_54 (ICso:794 nM). Variants containing the Mem_36 peptide was expressed stronger than the variant with CLIP by PAGE analysis. In contrast, variants containing the other two SARS-CoV-2 peptides were undetectable (FIG. 3A). Quantification of these variant proteins in the supernatant by ELISA showed that the expression level of the Mem_36 variant was 8-fold higher than that of the CLIP variant, while expressions of Mem_34- and Ncl_54-containing variant were minimal (FIG. 3B). These data support the notion that the stable expression of disulfide-engineered pDQ7-Fc is peptide-dependent. The presence of a higher affinity peptide correlates with a higher yield of the fusion protein.

[0089] Example 4: Binding characteristics of the pDQ7-Fc fusion protein

[0090] Whether pDQ7-Fc proteins were folded properly with intact surface epitopes was examined. pDQ7-Fc (knob-in-hole) was produced as a reference (“pDQ7(Ref)”) by co-expressing the a chain linked with Fc(knob) and the [3 chain linked with Fc (hole) in CHO cells (Serra, et al., Nat. Commun., 2019). The heterodimerization of Fc (knob) and Fc(hole), typically used for producing bispecific antibodies, promoted the formation of the pDQ7 complex in this case. Serial diluted YCDC, ACEC, and ACNC variants of pDQ7-Fc demonstrated similar binding profiles to anti-DQ7 (HB144), as measured by ELISA (FIG. 4A). All three variants showed higher affinities than the reference protein. In contrast, the wild type and ICGC variant pDQ7-Fc showed minimal binding. The YCDC variants, carrying CLIP and Mem_36 peptides, respectively, also shared similar binding profiles. Both bound to anti-DQ7 at higher affinities than the variant without peptide (FIG.4B).

[0091] The binding of pDQ7-Fc variants with three anti-DQ antibodies (HB144, SPV- L3, and 1A3) was examined by Western blot. These antibodies had unique binding specificities: HB144 was specific to the DQ3 parent type, including the DQ7 subtype; SPV-L3 was specific to the DQ a chain; 1A3 was specific to the DQ p chain. The pDQ7 (Ref) and four pDQ7-Fc variants tested (e.g., YCDC (CLIP), ACEC (CLIP), ACNC (CLIP), and YCDC (Mem_36)) were detected by all three antibodies and showed expected size (FIG. 4C). The wild type and ICGC variant proteins were undetectable by any of the antibodies. pDR3-Fc (knob-in-hole) (“pDR3(Ref)”) was produced as another negative control and was not detected by any of the DQ-specific antibodies (FIG. 4C). The above results suggest that multiple disulfide-engineered pDQ7-Fc proteins had intact surface epitopes recognized by antibodies of known specificities.

[0092] Example 5: Complement-dependent cytotoxicity effect of pDQ7-Fc on cognate B cell hybridomas

[0093] The ability of disulfide-engineered pDQ7-Fc protein to mediate antigenspecific killing of cognate antibody-producing cells (e.g., HB144 B cell hybridomas) was examined. The pDQ7 complex could guide the fusion protein to specific B cell receptors; at the same time, the Fc portion could activate the complement cascade and trigger the cytotoxicity effect (FIG. 5A). Indeed, significant binding of pDQ7-Fc (YCDC) to HB144 B cell hybridomas was detected by flow cytometry. In contrast, the wild type and ICGC variant were not detected on the cells (FIG. 5B). Using a flow cytometric complement-dependent cytotoxicity assay (Webber, et al., Blood, 2022), significant killing of HB144 cells by the YCDC and ACEC variants of pDQ7-Fc in a dose-dependent manner was detected. However, the killing effect appeared less potent than class I HLA- Fc proteins reported previously (Id.). To further enhance the complement-dependent killing, gain-of-function mutations to the Fc portion of the fusion protein were introduced. With the E345R (SEQ ID NO:4) and E430G (SEQ ID NO:5) mutations, the half-maximal cytotoxic concentration (CCso) of pDQ7-Fc (YCDC) decreased from 0.003 mg / mL to 0.00048 mg / mL and 0.00071 mg / mL, respectively. These results indicate that the dimeric, disulfide-engineered pDQ7-Fc protein was functional as a targeting agent, with the potential of antigen-specific killing of cognate antibody-producing cells.

[0094] Example 6: Disulfide engineering of soluble peptide-HLA-DQ2-Fc (pDQ2- Fc) fusion protein

[0095] Two DNA constructs encoding the two chains for pDQ2-Fc: (i) the extracellular domains of DQ2a chain (SEQ ID NO: 1 ) including N-terminal signal peptide, and (ii) the extracellular domains of DQ2p chain (SEQ ID NO:6) flanked at the N-terminus by the signal peptide and a gluten peptide glia-a1 a (QLQPFPQPELPY) and at the C-terminus by human lgG1 Fc (Fig. 8).

[0096] DNA encoding paired a and [3 chains for the wildtype and variant HLA- DQ2 were transfected into CHO cell lines and the expression of these recombinant proteins in the supernatant was measured. In addition to the paired mutations introduced for disulfide bond formation, sequences for the signal peptide, a placeholder glia-a1 a (QLQPFPQPELPY) peptide, and a flexible polyglycine linker (GGGSG)2 were fused to the 5’ end of the p chain sequence. The 3’ end of the p chain sequence was fused to a human lgG1 Fc fragment sequence. This design was expected to generate a peptide-HLA-DQ2 (pDQ2) complex that would further dimerize via the Fc fragment to form a bivalent pDQ2-Fc fusion protein within an immunoglobulin-like framework. Minimal expression of the wild-type pDQ2-Fc (data not shown) was observed in the supernatant as measured by ELISA (SPV-L3 antibody was coated at 0.25 ug / ml), while the expression levels were significantly increased for YCDC and ACEC mutants. The rest of the mutants did not exhibit significantly increased expression levels compared with the wild type (FIG. 8).

[0097] Example 7: Disulfide engineering of soluble peptide-HLA-DQ8-Fc (pDQ8- Fc) fusion protein

[0098] Two DNA constructs encoding the variant of two chains for pDQ8-Fc (YCDC): (i) the extracellular domains of DQ8a chain (SEQ ID NO: 8) including N- terminal signal peptide with mutation on Y19C, and (ii) the extracellular domains of DQ8p chain (SEQ ID NO: 7) with mutation on D6C flanked at the N-terminus by the signal peptide and a CLIP peptide (PVSKMRMATPLLMQA) and at the C-terminus by human lgG1 Fc (Fig. 9). The mutant was purified by protein A and evaluated by SDS- PAGE analysis. Under the non-reducing condition, variants YCDC-pDQ8 showed a distinct band of about150 kDa, consistent with the expected molecular weight of the pDQ8-Fc molecule.

[0099] Example 8: Disulfide engineering of soluble peptide-HLA-DR4-Fc (pDR4- Fc) fusion protein

[0100] MODIP procedure was used to predict candidate positions to introduce cysteines for disulfide bond formation in DR4 (SEQ ID NOs: 9-10) based on the crystal structures of DR4 with different peptides, including 5LAX (pep 26), 4IS6 (gl100), 5JLZ (Cit_pep 26), and 2SEB (COL2). Eight pairs of potential mutations that could form disulfide bonds were predicted based on these structures (Fig. 10). Interestingly, the pair of P83C-P5C, which corresponds to the lead pair for pDQ7, showed a high grade across all 4 models. P83C-P5C was mutated to cysteines and tested for protein production as described above for pDQ7 using CLIP as the place holder peptide. The p chain was flanked by a leader and CLIP peptide at the N-terminus and human lgG1 Fc at the C-terminus. A distinct band of about150 kDa (denoted by the star) was seen innon-reducing SDS-PAGE, consistent with the expected molecular weight of the dimeric pDR4-Fc molecule with a yield of 16 mg / L of cell culture.

[0101] The foregoing description of the embodiments has been provided for purposes of illustration and description. It is not intended to be exhaustive or to limit the disclosure. Individual elements or features of a particular embodiment are generally not limited to that particular embodiment, but, where applicable, are interchangeable and can be used in a selected embodiment, even if not specifically shown or described. The same may also be varied in many ways. Such variations are not to be regarded as a departure from the disclosure, and all such modifications are intended to be included within the scope of the disclosure.

[0102] Methods

[0103] Cell culture

[0104] CHO-S cells (Invitrogen) were grown in suspension in CHOgro ExpressionMedium (Mirus) in a shaking incubator at 37 °C until transfection for class II HLA protein production. IVD12 B cell hybridoma cells were obtained from ATCC and cultured in DMEM supplemented with glucose (4.5 g / L), 1 mM sodium pyruvate, 15% heat- inactivated fetal bovine serum, and antibiotics at 37 °C.

[0105] Plasmids

[0106] The extracellular domains of HLA-DQ7 [3 chain (DQB1 *03:01 ) and a chain (DQA1 *05:01 ) were codon-optimized for mammalian cells (GenScript) and cloned into the pcDNA3.4 vector. The natural signal peptide sequence of the a chain (MILNKALMLGALALTTVMSPCGG) was replaced with the signal peptide sequence from azurocidin preproprotein (MTRLTVLALLAGLLASSRA) to enhance proteinsecretion into mammalian culture supernatant. For the [3 chain, the natural signal peptide sequence was retained. Either a placeholder peptide CLIP (PVSKMRMATPLLMQA) or other peptides to be tested and a (GGGSG)2 flexible linker were inserted between the signal peptide and p chain sequences; the latter was followed by a human lgG1 Fc fragment. All sequences encoding the wild type and variant DQ7Fc proteins were synthesized at GenScript.

[0107] Protein expression and purification

[0108] Dimeric DQ7Fc recombinant proteins were first transiently expressed in CHO-S cells (Invitrogen). Healthy CHO-S cells with viability >97% and at a concentration of about 4.0 x 106 cells / ml were transfected with TransIT-PRO (Mirus) in Optimum Growth Flasks (Thompson) according to the manufacturer's recommendations. TransIT-PRO was premixed with plasmids for DQ7Fc p and a chains at a molar ratio of 1 :2, with a final concentration of 1 mg total DNA per liter of culture. The Transfection kit enhancer was added immediately post-transfection at the recommended volumes. Cultures were kept at 32 °C for 7 days before harvest. The supernatant of the cell culture was collected by centrifugation at 5000g for 10 min at 4 °C, followed by filtration through a 0.22 pm regenerated cellulose membrane. The secreted proteins were purified by affinity chromatography using a MabSelect PrismA Protein A column (Cytia). Fc fusion proteins were eluted with 0.1 M sodium citrate (pH 3.0) and neutralized immediately with 1 M Tris-HCI (pH 9.0). Purified proteins were then exchanged to 1 x PBS.

[0109] Ion exchange chromatography was used to further purify ACEC. Q Sepharose fast flow resin (Cytia #17051001 ) was used to separate the major dimericDQ7Fc from the minor aggregate and [3 chain-Fc dimer. Dimeric DQ7Fc was eluted with 250 mM NaCI in 20 mM Phosphate buffer, while aggregates and |3 chain-Fc dimer were eluted with higher NaCI concentration. The purity of eluted fractions was analyzed by non-reducing SDS-PAGE and analytical size exclusion chromatography (SEC; TSKgel G3000SWXL using 1XPBS as the elution buffer).

[0110] The reference protein, DQ7Fc(KIH), was produced based on the work by Serra et al. (31 ). Briefly, DNA sequences encoding the a chain (DQA1 *05:01 ) and p chain (DQB1 *03:01 ) of DQ7 were codon-optimized for mammalian cells (GenScript) and cloned into the pcDNA3.4 vector in frame with downstream sequences for human lgG1 Fc_knob (S350C / T362W) and Fc_hole (Y377C / T394S / L369A / Y435V), respectively. Endotoxin-free plasmids encoding the paired a and p chains were cotransfected at 1 :1 into CHO cells using TransIT-PRO (Mirus), after which cells were cultured at 32 °C for 7 days before harvesting. DQ7FcKIH in the culture supernatant was purified by protein A chromatography and exchanged into phosphate-buffered saline (PBS) as described above.

[0111] ELISA for protein quantification

[0112] Anti-DQ7 antibody purified from IVD12 B cell hybridoma cells was used to pre-coat 96-polystyrene-well microplates (Nunc Maxisorp, Thermo Fisher) at the final concentration of 0.3 pg / ml. The wells were washed and blocked with 5% bovine serum albumin in PBS / T (0.1 %). Day 7 supernatant of transfected CHO-S cell culture or purified proteins were then added and incubated for 1 h at room temperature. The plates were washed again, and goat antibodies against human IgG, conjugated to horseradish peroxidase (Jackson Immuno Research), were added. After incubation foran hour at room temperature and subsequent washing, a chromogen-substrate solution (3,3',5,5'-tetramethylbenzidine and hydrogen peroxide) was added. The reaction was stopped with sulfuric acid, and the optical density was measured at 450 nm using the Molecular Devices VersaMax Microplate reader.

[0113] Protein electrophoresis

[0114] The protein samples (5 pg) were electrophoresed in 4 to 20% SDS-PAGE gels (Bio-Rad). For reducing conditions, samples were diluted 1 :4 with 4x Laemmli sample buffer (Bio-Rad), supplemented with 10% 2-Mercaptoethanol (final concentration 2.5%), and then boiled at 100 °C for 5 min. Non-reducing conditions were assessed by diluting samples 1 :4 with 4x sample buffer (100 mM Tris-HCI, pH 6.8, 40% glycerol, and bromophenol blue) without boiling and loading them directly onto the gel. Gel images were captured using a ChemiDoc MP imaging system (Bio-Rad). Subsequent analysis of gel images was conducted using ImageLab software (Bio-Rad) for quantification and characterization of protein bands.

[0115] Western blot

[0116] For Western blot analysis, proteins were transferred onto nitrocellulose membranes following SDS-PAGE (Bio-Rad). The membranes were blocked with 5% BSA in PBS / T (0.1 %) for 1 h and then incubated with one of the following antibodies: 1 ) 1 :200 diluted IVD12 (purified in-house, 0.3 mg / ml), 2) 1 :250 diluted SPV-L3 (0.2 mg / ml; Novus Biologicals), and 3) 1 :10,000 diluted 1A3 (10 mg / ml, Leinco Technologies). After washing with PBS / T, membranes were incubated for 1 h with 1 :5000 diluted IRDye 800CW goat anti-mouse (Licor). Finally, membranes were washed again using PBS / T and detected using Licor ODYSSEY.

[0117] Bioinformatic analysis

[0118] The structure of HLA-DQ2 (PDB ID: IS9V) and -DQ8 (PDB ID: 1 JK8) was visualized using the Mol* 3D viewer at rcsb.org (60). Sites for insertion of disulfide bridges were selected using the MODIP server (caps.ncbs.res.in / iws / modip.html). For analysis of the mutated positions for disulfide engineering, multisequence alignment files for HLA-DQA1 and -DQB1 protein sequences were downloaded from IPD- IMGT / HLA database (version 3.54 released in October 2023) and processed using an in-house script to generate a text file listing each allele and its protein sequence on each line. Graphical representations of the HLA-DQA1 and -DQB1 alignments were generated using weblogo with default parameters (61 ). Visualization of the surface exposure of mutated residues in HLA-DQ7 was performed using pHLA3D (38).

[0119] FlowPRA assay

[0120] To verify the anti-DQ7 specificity of antibodies produced by IVD12 cells, FlowPRA assay was conducted using the FlowPRA Single Antigen HLA Class II - 4 Antibody Detection Test kit (One Lambda, West Hills, CA. #FL2HD04) on the AttuneNxT instrument (Invitrogen) following the manufacturers' instructions.

[0121] Cell binding assay

[0122] IVD12 hybridoma cells (ATCC) were washed in DPBS and plated at 50,000 cells per well on a V-bottom, 96-well plate. After blocking with TruStain FcX (BioLegend) at 1 :200 in staining buffer (1x PBS with 0.5% BSA, 2 mM EDTA), increasing amounts of DQ7Fc proteins was added to the buffer as indicated, or PBS was added as a negative control. After incubation at 4 °C for 20 min, cells were washed three times in buffer and stained with FITC-conjugated anti-human IgG at 1 :100 each.After a second incubation at 4 °C for 30 min, cells were washed three times, resuspended in 200 pl of buffer, and analyzed by flow cytometry on the AttuneNxT instrument (Invitrogen).

[0123] Flow cytometric complement-dependent cytotoxicity (CDC) assay

[0124] DQ7-specific B cell hybridoma cells (IVD12, ATCC) were washed in DPBS and plated at 50,000 cells per well on a V-bottom 96-well plate. Cells were pelleted by centrifugation at 1400 rpm and 4 °C for 4 min. Freshly thawed rabbit serum (One Lambda, #CDR-50) was diluted at 1 :8 in DPBS. DQ7Fc proteins were added to the diluted rabbit serum at a concentration of 0.01 mg / ml followed by 1 :2 serial dilutions to a titer up to 256. Rabbit serum with and without HLA-Fc was added to pelleted hybridoma cells at 45 pl / well and incubated at room temperature for 3 h in the dark. The cells were washed in 150 pl and then 200 pl staining buffer and then resuspended in 100 pl buffer containing 7-AAD diluted at 1 :50 followed by flow cytometric analysis on the AttuneNxT instrument (Invitrogen).

[0125] Data analysis and statistics

[0126] All data were from at least three independent experiments. For protein expression levels measured by ELISA, interpolated concentrations and 95% confidence intervals from the standard curve were reported from representative experiments. For antibody binding with titrated DQ7Fc proteins, means and standard deviations (SD) of the OD values were plotted against protein concentrations and fitted with the four- parameter logistic (4PL) regression model. For the CDC assay, means and SD of percent survival (7AAD-) of treated cells versus untreated cells were plotted against protein concentrations; CC50 was determined for each wild-type or variant protein bynonlinear regression using a variable slope model. All data analyses were performed using Prism version 9.2.0 (GraphPad Software, LLC).

Claims

WHAT IS CLAIMED IS:1 . A modified HLA-DQ molecule comprising: a first peptide comprising a wildtype a chain and a signal peptide, wherein at least one amino acids of the wildtype a chain are substituted with cysteine; and a second peptide comprising a wildtype [3 chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein at least one amino acids of the wildtype |3 chain are substituted with cysteine, wherein each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the p chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the p chain to form one or more interchain disulfide bonds.

2. The modified HLA-DQ molecule of claim 1 , wherein the HLA-DQ molecule is HLA-DQ2.5, HLA-DQ7, or HLA-DQ8.

3. The modified HLA-DQ molecule of claim 1 , wherein the signal peptide of the first peptide is derived from azurocidin preproprotein.

4. The modified HLA-DQ molecule of claim 1 , wherein the placeholder peptide of the second peptide is a placeholder CLIP peptide or a glia-a1 a peptide.

5. The modified HLA-DQ molecule of claim 1 , wherein the linker of the second peptide is a polyglycine linker.

6. The modified HLA-DQ molecule of claim 1 , wherein the Fc fragment sequence is a Human lgG1 Fc fragment sequence.

7. The modified HLA-DQ molecule of claim 2, wherein the first peptide has an amino acid sequence having at least 85% sequence identity to SEQ ID No: 1 and the second peptide has an amino acid sequence having at least 85% sequence identity to SEQ ID No: 2.

8. The modified HLA-DQ molecule of claim 7, wherein the substituted cysteine on the a chain is located at an amino acid position selected from the group consisting of 19, 83, 84, and a combination thereof.

9. The modified HLA-DQ molecule of claim 8, wherein the substituted cysteine on the p chain is located at an amino acid position selected from the group consisting of 6, 5, 33, and a combination thereof.

10. The modified HLA-DQ molecule of claim 9, wherein the substituted cysteine on the a chain is at amino acid position 19 and the substituted cysteine on the p chain is at amino acid position 6,wherein the substituted cysteine on the a chain is at amino acid position 83 and the substituted cysteine on the p chain is at amino acid position 5, wherein the substituted cysteine on the a chain is at amino acid position 84 and the substituted cysteine on the p chain is at amino acid position 33, or a combination thereof.11 . The modified HLA-DQ molecule of claim 9, wherein the substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C, wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the p chain is E5C, wherein the substituted cysteine on the a chain is A84C and the substituted cysteine on the p chain is N33C, or a combination thereof.

12. The modified HLA-DQ molecule of claim 10, wherein the substituted cysteine on the a chain is at amino acid position 19 and the substituted cysteine on the p chain is at amino acid position 6, and / or wherein the substituted cysteine on the a chain is at amino acid position 83 and the substituted cysteine on the p chain is at amino acid position 5.

13. The modified HLA-DQ molecule of claim 11 , wherein the substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C, and / or wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the p chain is E5C.

14. The modified HLA-DQ molecule of claim 1 , which is any one depicted in Figure 2.

15. The modified HLA-DQ molecule of claim 1 , wherein the Fc fragment sequence linked to the 3’ end of the [3 chain has at least one gain-of-function mutation.

16. The modified HLA-DQ molecule of claim 15, wherein the gain-of-function mutation is E345R and / or E430G.

17. A cell expressing the modified HLA-DQ molecule of any one of claims 1 to 16.

18. A nucleic acid encoding the modified HLA-DQ molecule of any one of claims 1 to 16.

19. A composition comprising the modified HLA-DQ molecule of any one of claims 1 to 16 presented on the surface of a cell or an acellular membrane and a pharmaceutically acceptable carrier.

20. The composition of claim 19, wherein the cell is an engineered cell.21 . The composition of claim 20, wherein the engineered cell is at least a monocyte, leukocyte, macrophage, or antigen presenting cell.

22. The composition of claim 19, wherein the acellular membrane is at least an exosome.

23. The composition of claim 19, wherein the composition treats cancer, pathogen- driven diseases, and immune responses.

24. A method for enhancing cytotoxicity, comprising: targeting the modified HLA-DQ molecule of any one of claims 1 to 16, a cell expressing the modified HLA-DQ molecule, or a cell comprising a nucleic acid sequence encoding the modified HLA-DQ molecule to a target cell.

25. The method of claim 24, wherein the cytotoxicity is complement-dependent cytotoxicity or antibody-dependent cellular cytotoxicity.

26. The method of claim 24, wherein the target cells are cancer cells, pathogen, and unwanted cells.

27. A method of producing a modified HLA-DQ molecule, comprising: substituting one or more amino acids of a wildtype a chain with cysteine; substituting one or more amino acids of a wildtype p chain with cysteine in a manner that the a and |3 chains comprise the same number of substituted cysteines, andforming a disulfide bond between each substituted cysteine on the a chain and each substituted cysteine on the [3 chain to form one or more interchain disulfide bonds.

28. A modified HLA-DQ7 molecule comprising: a first peptide having an amino acid sequence of SEQ ID No: 1 comprising a wildtype a chain and a signal peptide, wherein a cysteine substitution on the wildtype a chain is located at an amino acid position of 19, 83; 84, or a combination thereof, and a second peptide having an amino acid sequence of SEQ ID No: 2 comprising a wildtype [3 chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein a cysteine substitution on the wildtype [3 chain is located at an amino acid position of 6, 5, 33, or a combination thereof, wherein the substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C, wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the [3 chain is E5C, wherein the substituted cysteine on the a chain is A84C and the substituted cysteine on the [3 chain is N33C, or a combination thereof, wherein each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the [3 chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the [3 chain to form one or more interchain disulfide bonds.

29. The modified HLA-DQ7 molecule of claim 28, wherein the signal peptide of the first peptide is derived from azurocidin preproprotein.

30. The modified HLA-DQ7 molecule of claim 28, wherein the placeholder peptide of the second peptide is a placeholder CLIP peptide.31 . The modified HLA-DQ7 molecule of claim 28, wherein the linker of the second peptide is a polyglycine linker.

32. The modified HLA-DQ7 molecule of claim 28, wherein the Fc fragment sequence is a Human lgG1 Fc fragment sequence.

33. The modified HLA-DQ7 molecule of claim 28, wherein the human IgG 1 Fc fragment sequence linked to the 3’ end of the p chain has at least one gain-of-function mutation.

34. The modified HLA-DQ7 molecule of claim 28, wherein the gain-of-function mutation is E345R and / or E430G.

35. A modified HLA-DQ7 molecule comprising: a first peptide having an amino acid sequence having at least 85% sequence identity to SEQ ID No: 1 comprising a wildtype a chain and a signal peptide, wherein acysteine substitution on the wildtype a chain is located at an amino acid position of 19, 83; 84, or a combination thereof, and a second peptide having an amino acid sequence having at least 85% sequence identity to SEQ ID No: 2 comprising a wildtype [3 chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein a cysteine substitution on the wildtype [3 chain is located at an amino acid position of 5, 33, or a combination thereof, wherein the substituted cysteine on the a chain is Y19C and the substituted cysteine on the [3 chain is D6C, wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the [3 chain is E5C, wherein the substituted cysteine on the a chain is A84C and the substituted cysteine on the [3 chain is N33C, or a combination thereof, wherein each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the (3 chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the [3 chain to form one or more interchain disulfide bonds.36 A modified HLA-DR molecule comprising: a first peptide having an amino acid sequence having at least 85% sequence identity to SEQ ID No: 9 comprising a wildtype a chain and a signal peptide, wherein a cysteine substitution on the wildtype a chain is located at an amino acid position of 83 and / or 84; anda second peptide having an amino acid sequence having at least 85% sequence identity to SEQ ID No: 10 comprising a wildtype [3 chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein a cysteine substitution on the wildtype [3 chain is located at an amino acid position of 5 and / or 33, wherein the substituted cysteine on the a chain is P83C and the substituted cysteine on the [3 chain is P5C and / or wherein the substituted cysteine on the a chain is I84C and the substituted cysteine on the [3 chain is H33C, wherein each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the [3 chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the [3 chain to form one or more interchain disulfide bonds.

37. A modified HLA-DQ2.5 molecule comprising: a first peptide having an amino acid sequence having at least 85% sequence identity to SEQ ID No: 1 comprising a wildtype a chain and a signal peptide, wherein a cysteine substitution on the wildtype a chain is located at an amino acid position of 19, 83, 84; of a combination thereof, and a second peptide having an amino acid sequence having at least 85% sequence identity to SEQ ID No: 6 comprising a wildtype [3 chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein a cysteine substitution on the wildtype [3 chain is located at an amino acid position of 6, 5, 33, or a combination thereof,wherein the substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C, wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the p chain is E5C, wherein the substituted cysteine on the a chain is A84C and the substituted cysteine on the p chain is N33C, or a combination thereof, wherein each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the p chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the p chain to form one or more interchain disulfide bonds.

38. A modified HLA-DQ8 molecule comprising: a first peptide having an amino acid sequence having at least 85% sequence identity to SEQ ID No: 8 comprising a wildtype a chain and a signal peptide, wherein a cysteine substitution on the wildtype a chain is located at an amino acid position of 19, 83, 84, or a combination thereof; and a second peptide having an amino acid sequence having at least 85% sequence identity to SEQ ID No: 7 comprising a wildtype p chain, a signal peptide, a placeholder peptide, a linker, and an Fc fragment sequence, wherein a cysteine substitution on the wildtype p chain is located at an amino acid position of 6, 5, 33, or a combination thereof, wherein the substituted cysteine on the a chain is Y19C and the substituted cysteine on the p chain is D6C,wherein the substituted cysteine on the a chain is A83C and the substituted cysteine on the p chain is E5C, wherein the substituted cysteine on the a chain is A84C and the substituted cysteine on the p chain is N33C, or a combination thereof, wherein each substituted cysteine on the a chain is proximal to a corresponding substituted cysteine on the p chain such that each substituted cysteine on the a chain forms a disulfide bond with a substituted cysteine on the p chain to form one or more interchain disulfide bonds.

Images